Albany Medical College

Orthopaedic Journal
 1367 Washington Avenue
Albany, NY 12206
www.theboneandjointcenter.com

Providing world-class orthopaedic medical and surgical care, all under one roof.

This is an accredited state-of-the-art facility providing
musculoskeletal ambulatory surgical and pain management
services. The Center employs a tremendous team focused
solely on the care of its patients, from pre-operative assessment,
through the surgical procedure, to post-operative recovery.
Visit us on-line at www.crasconline.com
Complex problems like back and neck pain require a team
approach. Capital Region Spine includes two fellowship-
trained spine surgeons and two non-surgical physicians who
focus 100% on spine care.
Visit our on-line spine encyclopedia at:
www.capitalregionspine.com
or call 518-618-1111 for an appointment

In 1999, Capital Region Orthopaedics affiliated with the

Division of Orthopaedics at Albany Medical College. Since
that time, the group has assumed a major role in the training of
Orthopaedic residents at the Albany Medical College. Our 24
Orthopaedic surgeons include internationally known surgeons
who enjoy outstanding reputations for clinical excellence in the
areas of trauma, total joint replacement, sports medicine, spine Thomas Nicolla Physical Therapy
care, hand and upper extremity surgery, foot and ankle surgery,
and arthroscopic surgery. Hand Therapy | Work Assessment
Aquatic Therapy | Occupational Therapy
Above all, the staff members of Capital Region Orthopaedics Pediatric Services | Fitness Programs
never lose sight of the fact that they are caring for individuals
who need treatment on many levels. They are skilled in the Specializing in orthopaedic rehabilitation,
science of medicine and gifted in the art of caring. using the most up-to-date techniques.

For further information call 518-489-2666 Please call 518-438-7926 to schedule an appointment
or visit us on-line at www.caportho.com www.nicollatherapy.com
 Albany Medical College Orthopaedic Journal
Comparison of Computer Monitors in Evaluating Digital Radiographs for the Diagnosis of Scaphoid
Fractures
Justin M Ferrara, MD, Paul Hospodar, MD, Richard L. Uhl, MD, Erin L. Kaufman, MPH ........................................3

The Effects of Non-enzymatic Glycation on Cancellous Bone: A Comparison between Hip Osteoarthritis and
Hip Fractures
Michael Mulligan, MD, Lamya Karim, Joseph Bellapianta, MD, Jared Roberts, MD, Richard L. Uhl, MD,
Deepak Vashishth, Ph.D. ................................................................................................................................................8

How Much Motion is There in the Normal Scapholunate Joint?

Richard L. Uhl, MD, Mitchell Campito, MD, Miguel Perela-Cruz, MD.....................................................................14

Displaced Intra-articular Calcaneal Fractures Comparison of Two Different Surgical Techniques: Modified
Palmer’s Approach and Extensile Lateral Approach: Early Complications, Radiographic Findings and
Outcomes
Camilo Guzman, MD, Paul Hospodar, MD, John DiPreta, MD, Paul Johnson, MD...................................................17

Comparison of Endoscopic and Open Carpal Tunnel Release in the Same Patient
Ali Hashemi, MD, Richard Whipple, MD, Erin L. Kaufman, MPH............................................................................25

Total Shoulder Arthroplasty vs. Hemiarthroplasy for Osteoarthritis: A Retrospective Review of

Clinical Outcomes
Brent Carlson, MD, John Czajka, MD, Toni Hansen, CCRC, Patrick Marinello, Ashar Ata . ....................................28

Adjacent Level Effects: A Novel In Vivo/In, Vitro Active/Passive Protocol for Assessing Redistribution of Mo-
tion of the Entire Thoracolumbar Spine Following Surgical Intervention
Kyle A Elsabee, John D. Wason, James Boyle, MD, Ronniel Nazarian, MD, John T. Wen, Ph.D.,
Allen L. Carl, MD, Eric H. Ledet, Ph.D. .....................................................................................................................32

The Role of Extensor Muscle Fatigue in the Onset of Intervertebral Disc Degeneration: An In Vivo Model
Mary Beth M. Grabowsky, MS, Nicholas A. Pallotta, MS, Matthew W. Connelly, MS, Brett Van Etten, Rebecca A.
MacDonald, MS, Maxwell Alley, Joseph C. Glennon, VMD, Andrew H. Dubin, MD, John W. German, MD,
Richard L. Uhl, MD, Eric H. Ledet, Ph.D....................................................................................................................40

In Vivo Effects of Cyclic Compression on the Intervertebral Disc and Endplates of the Lumbar Spine:
A Pilot Study with a Novel Animal Model
Glenn P. Sanders, Ph.D., Sarah E. Linley, Gwendolyn A. Sowa, MD, Ph.D., Joseph C. Glennon, VMD, Darryl J.
DiRisio, MD, James P. Lawrence, MD, Jacqueline Vanderzanden, MD, Elizabeth Leimer, Jeffrey C. Lotz, Ph.D.,
Eric H. Ledet, Ph.D.......................................................................................................................................................48

The Effect of Anti-Rotation and Anti-Cutout Features in Cephalomedullary Nail Fixation of

Intertrochanteric Femur Fractures: An Acute Biomechanical Study
Matthew W. Connelley, MS, Jared Roberts, MD, Rosemary Buckle, MD, Eric H. Ledet, Ph.D. ...............................60

RPI 2010 Senior Design Showcase.............................................................................................................................70

Albany Medical College, Division of Orthopaedic Surgery Grand Rounds.........................................................72

The Albany Medical College Orthopaedic Journal is an annual publication of the Division of Orthopaedic Surgery at the Albany
Medical College. If you have any questions, or would like to add your name to our mailing list, you are welcome to contact us
at Albany Medical College Orthopaedic Journal, 1367 Washington Ave, Suite 202, Albany, NY 12206. You are also welcome to
contact us via phone at 1-518-453-3074 or fax at 1-518-453-1463.
1
 Albany Medical College Orthopaedic Journal
Richard L. Uhl, MD
Professor
Editor-in-Chief
James P. Lawrence, MD
Assistant Professor
Deputy Editor – Research
Michael Mulligan, MD
Assistant Professor
Deputy Editor – Clinical
Eric H. Ledet, Ph.D
Assistant Professor, Rensselaer Polytechnic Institute
Deputy Editor – Biomedical Engineering

Toni Hansen, CCRC

Research Coordinator
Managing Editor
Kathy Pangburn
Residency Coordinator

Editorial Review Board

Richard H. Alfred, MD John A. DiPreta, MD Jordan M. Lisella, MD
Clinical Associate Professor Clinical Associate Professor Clinical Assistant Professor
Sports Medicine Foot & Ankle Foot & Ankle

R. Maxwell Alley, MD Andrew H. Dubin, MD Jeffrey Lozman, MD

Clinical Assistant Professor
Sports Medicine
Kaushik Bagchi, MD
Assistant Professor
Trauma & Adult Reconstructive
Robert A. Cheney, MD
Clinical Assistant Professor
Spine Surgery
John Czajka, MD
Clinical Associate Professor
Adult Reconstructive
Pediatric Orthopaedics
Shankar P. Das, MD
Clinical Assistant Professor
Sports Medicine
Clinical Associate Professor
Physical Medicine &
Rehabilitation
Marc D. Fuchs, MD
Professor
Adult Reconstructive
Andrew C.Gerdeman, MD
Clinical Assistant Professor
Sports Medicine
Robert J. Hedderman, MD
Clinical Assistant Professor
General Orthopaedics
Paul P. Hospodar, MD
Associate Professor
Trauma & Adult Reconstructive
2
Clinical Professor
Adult Reconstructive
Daniel T. Phelan, MD
Clinical Assistant Professor
Sports Medicine
David E. Quinn, MD
Clinical Assistant Professor
Hand & Upper Extremity
Richard R. Whipple, MD
Clinical Assistant Professor
Hand & Upper Extremity
George Zanaros, MD
Clinical Assistant Professor
Hand & Upper Extremity
 Comparison of Computer Monitors in Evaluating Digital Radiographs for the Diagnosis of
Scaphoid Fractures
Justin M. Ferrara, MD, Paul Hospodar, MD, Richard L. Uhl, MD, Erin L. Kaufman, MPH
Division of Orthopaedic Surgery, Albany Medical College, Albany, NY

Presented at the 2008 AMC Orthopaedic Surgery Annual Resident Thesis Day

Background: The diagnosis of acute scaphoid fractures can be a challenge. Advances in technology have lead to
the increased use of digital radiographs as opposed to actual radiographs in the evaluation of musculoskeletal in-
jury.  There is much variation in the quality of the monitors used to evaluate these studies.  This study compares
personal computer (PC) monitors and picture archiving and communication system (PACS) monitors in the evalu-
ation of non-displaced scaphoid fractures.   The purpose of this study is to retrospectively compare the ability of
orthopaedic surgeons to identify scaphoid fractures on digital radiographs using a PC and a PACS monitor.     
 
Methods: Twenty-five patients who had previously been diagnosed in an outpatient orthopaedic setting with an
acute, non-displaced scaphoid fracture were identified from a review of patient charts.  A control group of twenty-
five subjects with a diagnosis of wrist sprain who had wrist radiographs taken were also identified.  All subjects
had wrist radiographs that included a posterioanterior view, a posterioanterior view with ulnar deviation, and a
lateral view. Two orthopaedic attending surgeons reviewed the radiographs of both groups while being blinded to
the diagnoses.  The reviewers evaluated for the presence or absence of scaphoid fracture. 

Results: The reviewers correctly identified a non-displaced scaphoid fracture in 44 of 50 radiographic images
using the PC monitors. The reviewers correctly identified a negative radiographic image (no fracture) in 46 of 50
radiographic images using the PC monitors. This represents a sensitivity of 0.88 and a specificity of 0.92 for the
PC monitor. Using the PACS monitors, the reviewers correctly identified a non-displaced scaphoid fracture in 39
of 50 radiographic images. The reviewers correctly identified a negative radiographic image (no fracture) in 45 of
50 radiographic images using the PACS monitors. This represents a sensitivity of 0.78 and a specificity of 0.90 for
the PACS monitors.

Conclusions: Orthopaedic surgeons are justified in their use of PC monitors in the evaluation of non-displaced
scaphoid fractures. PACS monitors do not offer the orthopaedic surgeon additional diagnostic power.
Introduction Many orthopaedic offices use personal computer moni-
The diagnosis of acute scaphoid fractures can be a tors as opposed to a picture archiving and communication
challenge.2,8,11  There may not be obvious radiological system (PACS) used by radiologists.  Currently there
evidence of fracture until several weeks after the initial are no formally defined standards or specific guidelines
injury, and ultimately additional radiographic studies may for the type of monitor used in the evaluation of digital
be require to make a diagnosis.1 Non-displaced scaphoid radiological studies.4-6, 8   There are significant differences
fractures can be subtle, even when being evaluated with in quality of monitor used in the orthopaedic outpatient
the highest quality monitors used by radiologists.2,4,8,11 Ac- setting compared to those used by radiologists.4, 6 
cording to Pillai and Jain, the incidence of true scaphoid There are varying opinions as to the optimal
fractures with initially negative radiographs is 6.66%.8  number of views that need to be taken to diagnose a sca-
Conventional x-rays miss these types of fractures on the phoid fracture.7,10   When consulting orthopaedic surgeons,
patient’s initial presentation.7  57.5% preferred four views, while 33.3% preferred three
Advances in technology have lead to the in- views.7,10  The majority of consulted radiologists preferred
creased use of digital radiographs as opposed to actual four views (68.4%) while 15.8% preferred three views.7,10 
radiographs in the evaluation of musculoskeletal injury.4  There is no known literature that evaluates the number of
There is much variation in the quality of the monitors used views preferred in America.  Our study will utilize three
to evaluate these studies.4, 6   Radiologists tend to utilize views: posterioanterior, lateral, and posterioanterior with
higher quality monitors, which optimize brightness, lumi- ulnar deviation. 
nance, and resolution.  Radiologists’ monitors generally The PACS system has several qualities that make
either undergo periodic self-checks or have formal quality it unique.  It is composed of several components which
assurance evaluation performed by the radiology staff.  include a digital device, a network with a bandwidth be-
3
tween 10 and 50 megabytes/study, an archiving device, wrist radiographs taken at their initial visit (ICD-9 code
diagnostic workstations and software that has archiving 842.00).  All subjects had wrist radiographs that included
and routing features. 9    The benefits of having a PACS a posterioanterior view of the wrist, a posterioanterior
system include, cost savings as compared to conventional view of the wrist with ulnar deviation, and a lateral view
radiographs, productivity improvements, superior image of the wrist. 
quality, decreased interpretation time, and an avenue for A fifth year orthopaedic resident selected sub-
rapid consultation with other specialists.9   The PACS jects and radiographic images (JMF).   The patient data-
system can be expanded to include additional worksta- base used to identify the subjects came from the Capital
tions and may utilize the Internet to allow the films to Region Orthopaedic Group billing office (Albany, NY). 
be accessed outside of the facility by other consultants.9  The database included all patients with the diagnosis of a
Several studies have demonstrated no statistical diagnostic scaphoid fracture and wrist sprain in the last four years,
difference between the PACS workstation and personal the beginning of the utilization of digital imaging at this
computer monitors.3,4 The PACS monitors are not routinely facility.  Fractures deemed to be acute, non-displaced,
utilized in the outpatient orthopaedic settings likely due to and without concomitant pathology in skeletally mature
the additional cost associated with such systems.  patients were included.  All radiographs were taken at
An acute scaphoid fracture is a diagnosis that is the Bone and Joint Center.  The Bone and Joint Center/
commonly made in the orthopaedic office setting, without Capital Region Orthopaedic Group is a private, outpatient
the assistance of a radiologist and/or the PACS monitor.  facility closely affiliated with Albany Medical Center Di-
The recent trend is for offices to become “filmless” as vision of Orthopaedics.  Typical radiographs used in the
computer technology improves to the point of digital study are shown in Figure 1A, 1B, and 1C.
radiograph quality rivaling that of the printed radiograph.  Two orthopaedic attending surgeons (RLU, PH)
Prior study has demonstrated that  the benefit of routine reviewed the radiographs of both groups while being
duplicate radiograph interpretation by both an orthopaedist blinded to the subjects’ diagnoses.  Each subject was ran-
and a radiologist does not justify its cost.12 domly assigned a number and the subject’s name and sex
Our study compared two systems (PC and PACS were removed from the radiograph.  The radiographs were
monitors) in the evaluation of non-displaced scaphoid evaluated for the presence or absence of scaphoid fracture. 
fractures.  The null hypothesis was that there was no Both reviewers evaluated all subjects first on the PC moni-
statistically significant difference in the identification of tor and then on the PACS monitor.  The order of the cases
scaphoid fractures in digital radiographs evaluated by at- was random. The reviewers were allowed to magnify the
tending orthopaedic surgeons using either a PC monitor digital images as each monitor’s system allowed. 
or the PACS monitor. The purpose of this study was to The data was calculated for the positive and
retrospectively compare the ability of orthopaedic surgeons negative predicted value, sensitivity, and specificity.  A
to identify non-displaced scaphoid fractures on digital chi-square was used to determine the significance of the
radiographs using PC and PACS monitors.    variables.  Power was recalculated at the end of the study
to determine the true statistical power of the study.
Materials and Methods The study protocol was approved by the Albany
The study population consisted of adults aged 18 and older.  Medical Center’s Institutional Review Board and carried
The experimental group consisted of 25 patients who were out in accordance with the Declaration of Helsinki of the
identified with a diagnosis of an acute non-displaced sca- World Medical Association.  Informed consent was not
phoid fractures at their initial visit (ICD-9 code 814.00). required.
The control group consisted of 25 subjects who ruled out
for scaphoid fractures (wrist sprain diagnosis) but had

Figure 1A, 1B, and 1C.  1A- P/A wrist, 1B-P/A wrist with ulnar deviation, 1C-
lateral wrist in a study subject with scaphoid fracture.     
4
Table 1: Subject Characteristics for the Two Groups
Scaphoid fracture
Variable
group (N=25)
Age (Mean) 33
Sex (M:F) 17:8
Wrist laterality (R:L) 15:10
Results  led to a positive predictive value of 0.89 and a negative
The radiographic images were evaluated by the two
reviewers separately on both the PC and the PACS moni-
tors. There were twenty-five subjects with non-displaced
scaphoid fractures and twenty-five subjects without a
scaphoid fracture (wrist sprain). The mean age for the
subjects in the fracture group is 33 years. The mean age for
the subjects in the non-fractured group is 39 years. There
are 17 male subjects and 8 female subjects in the fracture
group. There are 9 male subjects and 16 female subjects
in the non-fractured group. The subject characteristics
can be seen in Table 1.
The data from the two reviewers were combined
such that for each system evaluation (PC or PACS) there
were data from fifty images demonstrating a scaphoid
fracture and fifty images without a scaphoid fracture.
The reviewers correctly identified a non-displaced
scaphoid fracture in 44 of 50 radiographs using the PC
monitors. The reviewers correctly identified a negative
radiographic image (no fracture) in 46 of 50 radiographs
using the PC monitors. This represents a sensitivity of
0.88 and a specificity of 0.92 for the PC monitor. Using
the PACS monitors, the reviewers correctly identified a
non-displaced scaphoid fracture in 39 of 50 radiographs.
The reviewers correctly identified a negative radiographic
image (no fracture) in 45 of 50 radiographs using the PACS
monitors. This represents a sensitivity of 0.78 and a speci-
ficity of 0.90 for the PACS monitors. This information is
represented in Table 2.
Further extrapolating these data gives a positive
predictive value of 0.92 and a negative predictive value
of 0.88 for the PC monitor. The PACS monitor’s results
Table 2: Results of the Radiographic Image Review Using PC and PACS Monitors
  Scaphoid fracture group Non-fractured group
  Correct Incorrect Correct
Reading Reading Reading
PC 44 (88%) 6 (12%) 46 (92%)
PACS 39 (78%) 11 (22%) 45 (90%)
5
Non-fractured group
p-value
(N=25)
39 0.27
9:16 <0.0001, 0.007
14:11 0.057, 0.271
predictive value of 0.80. The positive predicitive value
is the proportion of patients with positive test results who
are correctly diagnosed. The negative predictive value is
the proportion of patients with negative test results who
are correctly diagnosed.
There were a total of 26 of 100 images that were
evaluated and diagnosed incorrectly. The data collection
sheet contained a question regarding the recommended
treatment based on the diagnosis, either conservative or
surgical management. An incorrect diagnosis resulted in
a change in the treatment recommendation in only one
case. The radiograph was incorrectly read as showing a
scaphoid fracture using the PACS monitor and surgical
intervention was recommended. Surgical intervention
was recommended in only 5 other cases, all of whom were
correctly diagnosed with a scaphoid fracture. Three of
these recommendations were made when evaluating the
images with the PACS monitor and two were made using
the PC monitor.
The data collection sheet used by the reviewers
contained a question regarding the desire for additional
imaging to further clarify the diagnosis. The reviewers
would have liked to have reviewed additional images for
diagnostic purposes in 84 of 100 in the non-fractured group
and in 72 of 100 in the fracture group.
The data collection sheet also asked the review-
ers their opinion of the quality of the radiographic image.
The image was judged as good, fair, or poor. Of the PC
images, 46 were rated as good, 47 as fair, and 7 as poor.
Of the PACS images 87 were rated as good, 12 as fair, and
1 as poor (see Table 3).
 
Incorrect
Sensitivity Specificity
Reading
4 (8%) 0.88 0.92
5 (10%) 0.78 0.90
 Table 3: Image Quality as Rated by the Reviewers
   
Good
Image Quality
Fair
Poor
Discussion
Prior investigation in the radiology literature demonstrated
no difference in the detection of wrist fractures with PC
compared to PACS.4   In that study, Doyle et al. evalu-
ated the ability of radiologists to detect wrist fractures.  occasionally arrive at the wrong diagnosis.  This occurred
However, Liang et al. demonstrated a significant difference
in the ability of radiologists to detect fractures on digital
radiographs using two different monitors.6 There currently
remains no formally defined standards or specifications
for the types of monitors used to read digital radiographic
images.13   Brem et al. demonstrated that low-cost com-
puter monitors could be safely used to detect cervical spine
fractures in the emergency department.3  orthopaedic surgeons are justified in their use of PC moni-
In the orthopaedic clinic setting, it is common
practice for orthopaedic surgeons to evaluate the radio-
graphs taken in their office.  Prior study has demonstrated
that  the benefit of routine duplicate radiograph interpre-
tation by both an orthopaedist and a radiologist does not
justify its cost.12 Routine radiological consultation of
radiographs in the orthopaedic setting is unnecessary.
Major savings in health-care costs can be seen by not
routinely undertaking radiological consultation for elec-
tive orthopaedic problems.14 In many office settings plain
radiographs are no longer printed, but rather are digitally
stored and evaluated via computer by the orthopaedic
surgeon.  results to a broader population.  The focus of this study
This study demonstrates that there was no clear
advantage in using the PACS by orthopaedic surgeons to
evaluate wrist radiographs for scaphoid fractures.  There
was no significant difference in the ability of the review-
ers to determine the presence of a non-displaced scaphoid
fracture using the conventional PC monitor or the PACS
monitor.  Using the PC monitor a sensitivity of 0.88 and a
specificity of 0.92 were demonstrated. The PACS monitors
demonstrated a sensitivity of 0.78 and a specificity of 0.90
in the identification of the scaphoid fractures. While the
References
1.  Adey L, Souer JS, Lozano-Calderon S, Palmer W,
Lee SG, Ring D. Computed tomography of suspected
scaphoid fractures. J Hand Surg Am. 2007; 32(1): 61-
67.
6
PC PACS
46 87
47 12
7 1
sample size was not sufficient enough to give this study
adequate power, there was no meaningful difference be-
tween the two systems evaluated.
Over the course of this study, the reviewers did
in 26 of 100 radiographs. The largest percentage of these
occurred using the PACS monitors in a group of subjects
that had scaphoid fractures, but were diagnosed as hav-
ing no fracture. This inability to diagnose a fracture did
change the treatment that was recommended based on the
radiographic evaluation in one case.     
This study lends evidence to the notion that
tors in the evaluation of digital imaging for the purpose
of identifying fractures.  With a sensitivity of 0.88, the
reviewers were able to diagnose non-displaced scaphoid
fractures using a PC monitor to evaluate digital images
without difficulty.
This study has several limitations. Recall bias
cannot be excluded in the reviewers’ interpretation of the
images even though they were presented in random order
each time.  They may have recognized the images in the
second evaluation, however the reviewer would still have
been blinded to the true diagnosis.  There was a lack of
power in this study making it difficult to generalize the
was on the presence of a scaphoid fracture therefore it
is difficult to generalize to a wider variety of fractures. 
Bias may have been introduced by the fact that one of the
systems to be evaluated by the study (the PC monitor) was
used to make the original diagnosis.
This study represents further justification for
the use of personal computer monitors in the evaluation
of skeletal fractures by the orthopaedic surgeon. PACS
monitors do not offer the orthopaedic surgeon additional
diagnostic power.
2.  Ballas MT, Tytko J, Mannarino F. Commonly missed
orthopedic problems. Am Fam Physician. 1998; 57(2):
267-74.
3.  Brem MH. et al.: Evaluation of low-cost computer
monitors for the detection of cervical spine injuries in
the emergency room: an observer confidence-based
study. 2006; Emerg Med J, 23(11): 850-3.
4.  Doyle AJ, Le Fevre J, Anderson GD. Personal com-
puter versus workstation display: observer performance
in detection of wrist fractures on digital radiographs.
Radiology. 2005; 237(3): 872-7.
5.  Krupinski EA, Kallergi M. Choosing a radiology
workstation: technical and clinical considerations.
Radiology. 2007; 242(3): 671-82.
6.  Liang Z, Li K, Yang X, Du X, Liu J, Zhao X, Qi
X. ROC analysis for diagnostic accuracy of fracture
by using different monitors. J Digit Imaging. 2006;
19(3): 276-8.
7.  Parvizi J, Wayman J, Kelly P, Moran CG. Combin-
ing the clinical signs improves diagnosis of scaphoid
fractures. A prospective study with follow-up.  J Hand
Surg. 1998; [Br], 23(3): 324-7.
8.  Pillai A, Jain M. Management of clinical fractures of
the scaphoid: results of an audit and literature review.
Eur J Emerg Med. 2005;12(2): 47-51.
7
9.  Robertson I. Image dissemination and archiving. Clin
Tech Small Anim Pract. 2007;22(3): 138-44.
10.  Shenoy R, Pillai A, Hadidi M. Scaphoid fractures:
variation in radiographic views - a survey of current
practice in the West of Scotland region. Eur J Emerg
Med. 2007;14(1): 2-5.
11.  Welling RD, Jacobson JA, Jamadar DA, Chong S,
Caoili EM, Jebson PJ. MDCT and radiography of wrist
fractures: radiographic sensitivity and fracture patterns.
AJR Am J Roentgenol. 2008;190(1): 10-6.
12.  Anglen J, Marberry K, Gehrke J. The clinical utility
of duplicate readings for musculoskeletal radiographs.
Orthopedics. 1997;20(11):1015-9.
13.  Andriole K. Display monitors for digital medi-
cal imaging.  American College of Radiology. 2005;
543-6.
14.  Zohman G, Watts H. Is routine radiological
consultation cost-effective for pediatric orthopedic
radiographs?. J Pediatric Orthopaedics. 1998;18(4):
549-51.
 The Effects of Non-enzymatic Glycation on Cancellous Bone:
A Comparison between Hip Osteoarthritis and Hip Fractures
Michael Mulligan, MD*, Lamya Karim°, Joseph Bellapianta, MD*, Jared Roberts, MD*,
Richard Uhl, MD*, Deepak Vashishth, Ph.D°
*Division of Orthopaedic Surgery, Albany Medical College, Albany, NY
°Department of Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, NY

Presented at the 2008 AMC Orthopaedic Surgery Annual Resident Thesis Day

Background: The non-enzymatic glycation (NEG) of tissue has been shown to occur as a normal consequence of
aging.   Whether the accumulation of advanced glycation end products (AGEs) contributes clinically to osteo-
porotic fragility fractures is not well understood.  We hypothesized that cancellous bone from individuals who
sustained a hip fracture would have higher levels of AGEs than patients without fractures, and that these samples
would fail under less biomechanical compression. 
 
Methods: A total of eighteen femoral heads were used.  One group consisted of nine patients with hip osteoar-
thritis undergoing total hip arthroplasty, and a second group contained nine patients who sustained a femoral neck
fracture requiring hip hemiarthroplasty.  Cores of bone were mechanically tested to determine stress/strain proper-
ties.  Non-enzymatic glycation was measured through described methods utilizing photochemistry.
 
Results: The bone from the hip fracture group was an average of 54% weaker than the bone from the osteoar-
thritic group (p<0.05).  The NEG content was nearly 12% higher in the hip fracture group, but this value was not
statistically significant. 
 
Conclusions: This study demonstrates the phenomenon of NEG occurs in the cancellous bone collagenous matrix
of both osteoarthritic and osteoporotic patients and may contribute in different ways to these disease processes. 
Based on previous research, the increase in NEG within fractured bone helps explain that under compression it
behaved in a significantly more brittle manner.  If further data shows a significant increase in AGE levels in osteo-
porotic patients compared to others of similar ages, then we may have another target for the treatment of osteopo-
rosis and the prevention of fragility fractures.
Introduction It has been shown that as a natural consequence
Hip fractures and their associated morbidity are a signifi- of aging, a gradual decline in bone mass occurs on the
cant public health care concern in this country. In older order of 0.3 to 0.5% a year starting by the age of 303. Hu-
patients, hip fractures often occur as a result of increased man bone is in a constant state of remodeling secondary
fragility of the proximal femur due to osteoporosis, de- to a delicate coupling at the cellular level of osteoblast
fined as a “systemic skeletal disorder characterized by a mediated bone formation and osteoclast mediated bone
quantitative deficit of bone mass and micro-architectural resorption. In comparing the main types of human adult
breakdown of bone tissue”1. We are an ever-aging soci- bone, the turnover rate of cancellous bone is approximately
ety, with the elderly population being the fastest growing eight times that of cortical bone. This is because cancel-
segment. It is predicted that worldwide hip fractures will lous bone occupies a larger amount of surface area in the
reach epidemic proportions; rising from about 1.7 million human skeleton, and thus it is this type of bone which is
in 1990 to 6.3 million in the next forty years. It is estimated affected most by osteoporosis. While this loss of bone
1 in 3 women and 1 in 12 men will suffer a hip fracture mass over time occurs macroscopically, it is also well
during their lifetime1. The impact here goes well beyond established that there are corresponding changes within
the health care system’s ability to handle this growing bone at the molecular level. Specifically, the collagen
census of patients, as research indicates that only about matrix undergoes various biochemical reactions as the
50% of these patients are able to regain their pre-fracture body ages. Numerous studies support the concept that
ambulatory status2. An alarming financial burden is also collagen plays a substantial role in the ability of bone to
on the horizon, with costs of over $20,000 per patient for absorb energy, defined as its toughness. Collagen makes
the first year after fracture alone and an estimated $131,500 up 90% of the organic phase of bone and subsequently
million total cost by the year 20501. undergoes a variety of biosynthetic events which alter its

8
properties4. In particular, advanced glycation end products
(AGEs) can be made as posttranslational modifications to
collagen when reducing sugars react non-enzymatically
with free amino groups in lysine, hydroxylysine or arginine
residues. These products can act as adducts to proteins or
even create covalently bonded collagen cross-links5.
Previous research has documented that this
process of non-enzymatic glycation (NEG) occurs in the
extra-cellular matrix of many human tissues, including
bone. Investigators have shown that NEG of collagen, not
only increases with age, but also has a negative impact on
the mechanical properties of both cortical and cancellous
bone7,8. In vitro models of NEG demonstrate a change
in post-yield properties such that the altered bone has a
decreased capacity to accumulate damage and dissipate
energy8. Recent utilization of 3D Cat Scan and finite
element analysis to measure bone geometry and trabecular
connectivity has called into question our current methods
of calculating bone mass and mineral density, citing their
inconsistent predictions of bone strength9. Given that most
fragility fractures secondary to osteoporosis are of cancel-
lous bone, a more complete understanding of cancellous
bone fragility at the molecular level would allow us to
better assess the risk of age-related osteoporotic fractures.
The goal of this study is to examine the NEG process in
vivo as it pertains to the clinical entities of degenerative
hip arthritis and femoral neck fractures. We hypothesize
that, with controlling for age, cancellous bone samples
from individuals who sustain hip fractures will (1) have
higher concentrations of NEG compared to patients with-
out fractures and will (2) fail under less biomechanical
compression than those samples from individuals who
have osteoarthritis.
Materials and Methods
Sample Selection
After Institutional Review Board approval was granted
and informed consent was obtained from each study
participant, the entire femoral head including a portion
of the femoral neck was taken from patients undergoing
both total hip arthroplasty procedures and hip hemiar-
throplasties. All surgeries were completed at Albany
Medical Center (Albany, NY) between the periods
of October 2006 through November 2007. Care was
taken intra-operatively in order to minimize damage to
the cancellous bone within the bone sample as it was
removed from the patient. A section of bone was taken
from the portion of the femoral neck in each case so that
it could be evaluated by the pathology department per
standard protocol. The remaining bone was preserved
in saline soaked gauze and deep frozen at -80°C until
it was time for specimen preparation and mechanical
testing.
9
A total of eighteen femoral heads were used for
this study. The first experimental group consisted of nine
patients with degenerative hip disease having total hip
replacement surgery. The second group was composed
of nine patients who sustained a displaced femoral neck
fracture requiring hip hemiarthroplasty. Of note, all cases
of femoral neck fracture occurred from a fall from stand-
ing height, which is defined by the literature as fragility
fracture1. Patients with rheumatoid arthritis, congenital or
acquired dysplasia, or other inflammatory arthropathy were
excluded from the study. Blood samples were collected
from each patient in accordance with the requirements for
human tissue handling at Rensselaer Polytechnic Institute
(RPI) in Troy, NY. Hemoglobin A1c (HbA1c ) levels, which
are proportional to a patient’s average blood glucose con-
centration over the previous four weeks to three months,
were also drawn.
Mechanical Testing
A drill press with a modified jig was used to obtain mul-
tiple cores of cancellous bone from each femoral head.
The cores measured approximately 8mm in diameter and
20mm in length. They were taken in line with the princi-
pal compressive trabeculae orientation of the bone. This
orientation was pre-determined using plain radiographs of
the femoral heads. The cores were then cut into 8-10mm
sections under constant hydration (ISOMET 11-1180,
Buehler Corp., Lake Bluff, IL). The final core measure-
ments averaged 7.6±0.3mm in diameter and 8.6±0.9mm
in length. Between testing, samples were submerged in
normal saline and stored at -80°C.
Mechanical testing of cores from degenerative
and fractured femoral heads was conducted using an MTS
Bionix 858 hydraulic servo mechanical testing station
(MTS, Eden Prairie, MN). Each specimen was glued to
brass endcaps and loaded to failure in unconfined com-
pression at 833 microstrains/second at 37 degrees Celsius
under constant hydration. Load-displacement data was
collected during testing, and stress-strain curves were
created. The following parameters were then calculated:
(1) elastic modulus; (2) yield stress and strain (determined
using the 2% offset method); (3) ultimate stress and strain
(defined as the point of maximal load); and (4) post-yield
strain (defined as the difference between the ultimate and
yield strain values). Student t-tests were used to evaluate
the differences due to the underlying disease process of
each experimental group, with the level of significance
set at P=0.05.
Determination of NEG
After completing the mechanical testing, the nine cancel-
lous bone specimens from each experimental group were
decalcified with a 20% solution of 90% formic acid diluted
Figure 1. Comparison of age between hip osteoarthritis
and femoral neck fracture groups. No significant differ-
ence was noted. (Student’s t-test, p =0.26).
with deionized water. The solution was changed daily
until the presence of calcium was no longer detected by
a calcium determination assay. The specimens were then
digested with papain collagenase (0.4 mg/ml in 0.1 mM
sodium acetate buffer, pH 6.0, 16 h, 65 degrees Celsius).
NEG content was quantified using fluorescence readings
taken with a Synergy-HT Microplate reader at wavelengths
of 360nm excitation and 460nm emission (BioTek USA,
Winooski, VT) against a quinine sulfate standard. The
amount of collagen in each cancellous bone sample was
estimated based on the level of hydroxyproline8. The
content of hydroxyproline was determined by recording
the absorbance of digested specimens against a standard at
570nm wavelength using a Dynatech MR-600 Microplate
reader (Dynatech Inc., Alexandria, VA).
Results
Patient Demographics
In the degenerative hip disease group, there were five
males and four females, with an average overall age of
78. In the hip fracture group, there were three males and
six females, with an average overall age of 82. The two
experimental groups were not significantly different with
respect to age (Figure 1).
Table 1: Mechanical testing data from unconfined compression of the cancellous bone cores
Bone Elastic Yield
Yield Stress
Type Modulus Strain
[MPa] [MPa] [%]
Arthritis 531±315 5.72±3.4 1.13±0.43
Fracture 344±283 3.98±2.5 1.30±0.71
*Denotes statistical significance (Student’s t-test, p<.05).
10
Figure 2. Comparison of HbA1c levels between groups.
No significant difference was noted. (Student’s t-test, p
= 0.21).
Hemoglobin A1c Levels
The mean HbA1c level from blood samples of the nine
patents undergoing total hip arthroplasty for degenera-
tive joint disease was 5.79. The average level from blood
samples of the nine patients sustaining a femoral neck
fracture was 6.32. Both of these values were within normal
reference ranges for our lab (4.1-6.5), and no statistical
difference between the two groups was found (Figure 2).
Mechanical Testing
In comparison to bone taken from patients who sustained
a hip fracture, there was about a 54% average increase in
elastic modulus (stiffness) for the bone taken from patients
with osteoarthritis; however, this difference was not sta-
tistically significant (p=0.20). In addition, no significant
differences were found between the groups with respect to
yield stress, yield strain, ultimate stress or ultimate strain
(Table 1). The data for post yield strain demonstrated an
average 116% increase within the osteoarthritic bone,
which resulted in a statistically significant difference
between experimental groups (Figure 3).
Ultimate Ultimate Post Yield
Stress Strain Strain*
[MPa] [%] [%]
6.54±3.5 2.02±0.59 0.887±0.41
4.29±2.6 1.71±0.79 0.411±0.12
Figure 3. Comparison of post yield strain between hip osteoarthritis and femoral neck fracture groups. Results are
statistically significant (p<0.05).

AGE Quantification Discussion

Compared to cancellous bone cores from patients with
degenerative hip disease, calculations for the content of
NEG were nearly 12% higher in the hip fracture group
(Figure 4). This difference was not statistically significant.
Upon further analysis, no significant difference was found
between the NEG content in males and females, either
within each group or across groups.
Osteoarthritis (OA) and osteoporosis (OP) are the two most
common musculoskeletal disorders found in the elderly
population. Over the last thirty years, the prevailing theory
in the literature was that these two disease entities were
mutually exclusive. This largely stemmed from work
by Foss and Byers which studied 140 patients with hip
fractures and found only 3 with radiographic evidence of
osteoarthritis10. The prevailing opinion was that the pres-
ence of hip OA was protective against OP and helped to

Figure 4. AGE content was higher in the fracture group than in the osteoarthritis group, but this increase was not
statistically significant (p=0.52).
11
prevent osteoporotic hip fractures11. This inverse relation-
ship has been called into question over the past decade,
with studies suggesting an underappreciated number of
patients with concomitant OA and OP, especially in the
female, post-menopausal population. Glowacki et al.
found that 25% of a cohort of post-menopausal women
with degenerative joint disease (DJD) scheduled for elec-
tive total hip arthroplasty also had occult osteoporosis12. A
more recent study by Mäkinen et al. saw that nearly 73%
of their cohort of female patients with OA undergoing hip
replacement were either osteopenic or osteoporotic based
on bone mineral density13. Our study attempts to better
understand both the biomechanics and biochemistry of
cancellous bone samples taken from patients at the clini-
cal endpoint of these two disorders. The phenomenon of
NEG has already been shown in our lab to increase with
age and results in a decrease in the mechanical proper-
ties of bone8. These resulting advanced glycation end
products have been implicated as a pathogenic factor
in a variety of diseases, including the articular cartilage
degeneration seen in osteoarthritis14. To our knowledge,
this is only the second study to examine the NEG process
in patients with fractured cancellous bone and the first
study to quantify AGE levels in the underlying bone of
osteoarthritic patients.
Other studies have performed mechanical testing
on both osteoporotic and osteoarthritic bone. The values
for elastic modulus in our study revealed a 36% increase
in stiffness for the OA bone. Although the result is not
statistically significant, this trend is consistent with work
done by Li and Aspen, who compared subchondral and
cancellous bone samples in OA and OP hips to age- and
gender- matched controls and demonstrated an increase
in stiffness for the OA group compared to OP15,16. These
results support the idea that the underlying bone indeed
plays some role in the pathogenesis of OA. In a recent
report, Sun et al. compared seven femoral heads taken
from fractured hips with seven femoral heads from total
hip replacements due to degenerative joint disease 17.
Also lacking a control group, they found a significantly
higher elastic modulus and yield stress in the osteoarthritic
bone. Unlike our study, Sun et al. did not examine post-
yield properties, which are clinically more relevant when
determining the toughness of bone. They postulated the
increase in stiffness and yield stress may explain why
several aforementioned epidemiological reports have
shown fractures to be rare in the OA population. The data
in our study reveals a significant reduction in post-yield
strain in the fracture group, supporting our hypothesis
that the fractured bone would undergo less compression
prior to failure.
Saito et al. reported an increase in advanced gly-
cation end products (AGE) within cancellous bone in cases
12
of hip fractures compared to age- and gender- matched
non-fractured controls18. Within the control group of
their study, osteoarthritic cases were excluded. In our
study, quantifiable AGE levels were found in cancellous
bone from both hip fracture and osteoarthritic patients.
While the amount of NEG we measured was not com-
pared against a cadaver control group, it is important to
acknowledge the inherent difficulties of controlling for age
with the possibility of sub-clinical OA and OP acting as a
confounding variable. We have also shown previously in
our lab that increasing age has a significant positive effect
on the accumulation of AGEs in cortical and cancellous
bone7.8. It is thus important that the experimental groups in
this study were of similar ages (p= 0.26) in order to make
a fair assessment of NEG in each separate disease cohort.
Also, in order to eliminate the possibility of any poorly
controlled diabetic patients affecting the AGE content of
the cancellous bone, HbA1c levels were found to be similar
between experimental groups. Overall, there was indeed
a slight trend in our data towards more AGEs in the hip
fracture group by 12%. Although this difference was
not statistically significant, it is possible that an increase
in sample size may yield results to support our original
hypothesis. The increase in NEG also helps to explain the
fact that under compression the fractured bone was found
to behave in a significantly more brittle manner than the
osteoarthritic bone, especially considering our previous
work that has shown NEG to have detrimental effects on
the post-yield properties of cancellous bone8.
At a minimum, it cannot be discounted that
this study demonstrates that the phenomenon of NEG is
occurring in the cancellous bone collagenous matrix of
both osteoarthritic and osteoporotic patients and may be
contributing in different ways to these disease processes.
Whether the mechanical alterations found in arthritic
bone in our study are primary or secondary to degenera-
tive changes in the overlying cartilage, and whether the
accumulation of AGEs, which is known to occur at the
articular surface, also affects the NEG process within
bone warrants future research. If further data collection
shows a significant increase in AGE levels in osteoporotic
patients compared to others of similar ages, then we may
have another target in the treatment of osteoporosis and
the prevention of fragility fractures. Recent advances in
pharmaceutical research proposing “AGE inhibitors” and
“AGE breakers” offer much potential promise as addi-
tional weapons against the increasing health care burden
of osteoporotic hip fractures.
Acknowledgements
This work was supported by NIH grant AG20618. Thanks
to Howie Morris of Albany Medical Center in the Depart-
ment of Radiology for assistance with the x-rays.
 References

1. Johnell O, Kanis J. Epidemiology of osteoporotic 11. Cumming RG, Klineberg RJ. Epidemiological
fractures. Osteoporos Int. 2005; 16: S3-S7. study of the relation between arthritis of the hip and hip
fractures. Annals of Rheumatic Diseases. 1993;52:707-
2. Sernbo I, Johnell O. Consequences of a hip frac-
710.
ture: a prospective study over 1 year. Osteoporos Int.
1993;3:148-153. 12. Glowacki J, Hurwitz S, Thornhill TS, Kelly M,
LeBoff MS. Osteoporosis and vitamin-D deficiency
3. Einhorn TA. Metabolic Bone Disease. In: Einhorn
among postmenopausal women with osteoarthritis
TA, O’Keefe RJ, Buckwalter JA, editors. Orthopaedic
undergoing total hip arthroplasty. J Bone Jt Surg, Am.
Basic Science. Rosemont: American Academy of
2003; 85:2371-2377.
Orthopaedic Surgeons; 2007; p 415-426.
13. Mäkinen TJ, Alm JJ, Laine H, Svedström E, Aro
4. Viguet-Carrin S, Garnero P, Delmas PD. The
HT. The incidence of osteopenia and osteoporosis in
role of collagen in bone strength. Osteoporos Int.
women with hip osteoarthritis scheduled for cementless
2006;17:319-336.
total joint replacement. Bone. 2007;40:1041-1047.
5. DeGroot J. The AGE of the matrix: chemistry, con-
14. Verzijl N, Bank RA, TeKoppele JM, DeGroot J.
sequence and cure. Current Opinion in Pharmacology.
AGEing and osteoarthritis: a different perspective.
2004;4:301-305.
Current Opinion in Rheumatology. 2003;15:616-622.
6. Wang X, Shen X, Li X, Mauli Agrawal C. Age-
15. Li B, Aspden RM. Mechanical and material proper-
related changes in the collagen network and toughness
ties of the subchondral plate from the femoral head of
of bone. Bone. 2002;31:1-7.
patients with osteoarthritis and osteoporosis. Annals of
7. Vashishth D, Gibson GJ, Khoury JI, Schaffler MB, Rheumatic Diseases. 1997;56:247-254.
Kimura J, Fyhrie DP. Influence of nonenzymatic glyca-
16. Li B, Aspden RM. Composition and mechanical
tion on biomechanical properties of cortical bone. Bone.
properties of cancellous bone from the femoral head of
2001; 28:195-201.
patients with osteoporosis or osteoarthritis. Journal of
8. Tang SY, Zeenath U, Vashishth D. Effects of non- Bone and Mineral Research. 1997;12: 641-651.
enzymatic glycation on cancellous bone fragility. Bone.
17. Sun S, Ma H, Liu C, Huang C, Cheng C, Wei H.
2007;40:1144-1151.
Difference in femoral head and neck material proper-
9. Bostrom MPG. Bone metabolism and metabolic ties between osteoarthritis and osteoporosis. J Clin
bone disease. Orthopaedic Knowledge Update 9. Rose- Biomech. 2008.
mont: American Academy of Orthopaedic Surgeons;
18. Saito M, Fujii K, Marumo K. Degree of mineral-
2008:189-196.
ization-related collagen crosslinking in the femoral neck
10. Foss MV, Byers PD. Bone density, osteoathrosis cancellous bone in cases of hip fracture and controls.
of the hip, and fracture of the upper end of the femur. Calcif Tissue Int. 2006;79:160-168.
Ann Rheum Dis. 1972;31:259-264.

13
 How Much Motion is There in the Normal Scapholunate Joint?
Richard L. Uhl, MD*, Mitchell Campito, MD±, Miguel Perela-Cruz, MDß
*
Division of Orthopaedic Surgery, Albany Medical College, Albany, NY
±
Department of Radiology, Albany Medical Center, Albany, NY
ß
Department of Orthopaedics, Wichita Clinic, Wichita, KS

This study was undertaken to evaluate motion which occurs in the normal scapholunate joint, as the wrist moves
from extension to flexion, and to compare that motion to the radiocarpal (radiolunate) and midcarpal (lunocapi-
tate) motion occurring at the same time. We found that there was an average of 32 degrees of motion at the
scapholunate joint. This represented 52% of the midcarpal motion, with 29 degrees of scaphocapitate motion
accounting for the remaining 48% of midcarpal motion. Midcarpal motion averaged 54% of total wrist motion,
thus for each 4 degrees of wrist motion, approximately 2 degrees occurred at the midcarpal joint, with 1 degree of
motion occurring at the scapholunate joint. Motion outside the range described may indicate wrist pathology.
Introduction Materials and Methods
The scaphoid plays a unique role in carpal mechanics, act- Thirty volunteers, without wrist problems, had plain lateral
ing as a linkage to stabilize the midcarpal joint.1,5 Although X-rays taken with their wrists in extension, neutral and
reports of the ratio of radiocarpal to midcarpal motion vary, flexion (Figure 1). Each was asked to hold their wrist in
both joints contribute to total wrist motion. 3,4,6,7 Since the maximum extension and flexion; no attempt to passively
scaphoid is attached by ligaments to both the lunate and bend the wrist further was made. Each wrist was care-
capitate, the scapholunate and scaphocapitate joints each fully positioned with the radius and ulna superimposed to
must contribute to the total midcarpal motion. In this study minimize variation.
we used plain X-rays to evaluate the normal motion of the The X-rays were then digitized, at 150 dots per
scapholunate, scaphocapitate, and lunocapitate joints while inch, 256 shades of gray, using a transilluminated radio-
the wrist moves from extension to flexion. graph scanner (XRS Company, San Jose, CA). During
scanning, the image was adjusted to compensate for
X-ray technique differences. The digitized images were
displayed on a high resolution computer monitor at 8 to
10X zoom. Computer software was used to adjust the
level (brightness) and window (contrast) of the images so
that the outlines of the overlapped carpal bones could be
clearly distinguished on the lateral X-ray (Image Systems,
Troy, NY).
Standard radiographic axes of the scaphoid, lu-
nate, capitate, and radius were plotted over the digitized
X-ray image, using a computer assisted drafting (CAD)
program (Image Meterology, Ballston Spa, NY). 2,4,7.
Angles between each axis were measured using the CAD
program and recorded (Figures 2 and 3). The average
angle and standard deviation were determined for each
wrist position, and the motion and contribution to total
wrist motion for each joint calculated.

Results
Active wrist motion averaged 64 degrees extension and 50
degrees flexion, for a total active motion of 114 degrees
(+/- 22 degrees). The average midcarpal (lunocapitate)
motion was 61 degrees (+/- 18 degrees). This accounted
for 54% of the total wrist motion.
Figure 1. Lateral X-ray films with the wrist in extension
(top), neutral (middle), and flexion (bottom). Note the
change of position of the scaphoid relative to the lunate as
the wrist moves from extension to flexion.
14
Figure 2. Computer screen image with the wrist in ex-
tension, showing the axis of the lunate and axis of the
scaphoid. After the axes are drawn on the screen the
software calculates the angle between them. Additional
axes are plotted for the radius and capitate, and the angles
between each measured.
The scapholunate angle changed from 37 degrees
(+/- 10 degrees) in extension to 69 degrees (+/- 10 degrees)
in flexion. The scaphocapitate angle changed from 72
degrees (+/- 8 degrees) in extension to 43 degrees (+/- 12
degrees in flexion).
The 32 degrees of scapholunate motion accounted
for 52% of the total midcarpal motion. The 29 degrees of
scaphocapitate motion accounted for the remaining 48%
of midcarpal motion.
Fifty-five of the average 64 degrees of extension
occurred at the midcarpal joint, however individual varia-
tion was wide from 17 percent to 76 percent. Likewise,
of the 50 degrees of average flexion, 48% occurred at the
midcarpal joint, but again, individual variation was wide
(14% to 80%).
Discussion
Wrist motion is the result of complex 3-dimensional mo-
tion of the carpal bones as they interact with each other 6.
While the scaphoid stabilizes the midcarpal joint, it also
allows midcarpal motion. This study examines only the
motion of flexion and extension as it projects on the lateral
References
1. Bereger RH, Blair WF, Crowninshield RD, Flatt AE.
The scapholunate ligament. J Hand Surg. 1982;7:87-
91.
15
Figure 3. Computer screen image with the wrist in
flexion showing an increase in the angle between the
scaphoid and lunate, indicating the normal motion in that
joint.
X-ray.4,7 Motion of the scaphoid also involves rotation
under the capitate, which cannot be adequately analyzed
by this simple method.
Based on the data presented, for every 4 degrees
of wrist motion, there is approximately 2 degrees of
midcarpal motion, and approximately 1 degree of motion
at the scapholunate joint. Any efforts to reconstruct the
scapholunate joint must allow a similar degree of motion or
midcarpal stiffness will result.3 Fusion of the scaphocapi-
tate joint will eliminate approximately 50% of midcarpal
and 25% of total wrist motion. Ligament reconstruction
that allows motion at the scapholunate joint may help to
increase the resulting motion.
Motion at the midcarpal joint varied considerably
among the volunteers, which may account for the wide
variation of the contribution of this joint in previously pub-
lished reports.4,7 Motion outside the range described may
indicate pathology, but, for evaluation of the individual
patient, comparison views my be needed to determine
normal for that patient, rather than using a published
range of values.
2. Destouet JM, Gilula LA, Reinus WR. Roentgeno-
graphic diagnosis of wrist pain and instability. In: Licht-
man DM, ed. The wrist and its disorders. Philadelphia:
WB Saunders, 1988;82-95.
3. Gellman H, Kauffman D, Lenihan M, Botte MJ,
Sarmiento A. An in vitro analysis of wrist motion: The
effect of limited intercarpal arthrodesis and the contribu-
tions of the radiocarpal and midcarpal joints. J Hand
Surg. 1988; 13A:390-5.
4. Larsen CF, Mathiesen FK, Lindequist S. Measure-
ments of carpal bone angles on lateral wrist radiographs.
J Hand Surg. 1991; 16A:888-93.
6. Ruby LK, Cooney WP, An KN, Linscheid RL, Chao
EYS. Relative motion of selected carpal bones. A
kinematic analysis of the normal wrist. J Hand Surg.
1988;13A:1-10.
7. Schuhl JF, Leroy B, Comtet JJ. Biodynamics of the
wrist: Radiologic approach to scapholunate instability.
J Hand Surg. 1985;10A:1006-8.

5. Ruby LK, An KN, Linscheid RL, Cooney WP 3d,

Chao EY. The effect of scapholunate ligament section
on scapholunate motion. J Hand Surg. 1987;5A:767-
71.

16
 Displaced Intra-articular Calcaneal Fractures Comparison of Two Different Surgical Techniques:
Modified Palmer’s Approach and Extensile Lateral Approach: Early Complications,
Radiographic Findings and Outcomes
Camilo Guzman, MD, Paul Hospodar, MD, John DiPreta, MD, Paul Johnson, MD
Division of Orthopaedic Surgery, Albany Medical College, Albany, New York

Presented at the 2009 AMC Orthopaedic Surgery Annual Resident Thesis Day

Background: Intra-articular calcaneal fractures remain a challenging group of injuries to manage. Recent studies
have suggested operative intervention provides advantages over non-operative treatment, albeit with an increased
risk of wound complications. Varying surgical approaches have been advocated, and selecting one which minimiz-
es the risk to the soft-tissue envelope would provide a significant advantage in maximizing the patient’s post-op-
erative outcome. This study was undertaken to examine the differences in radiographic and functional outcomes,
as well as in rates of soft-tissue complications associated with intra-articular calcaneal fractures treated surgically
using either an extensile lateral or a modified Palmer’s approach.

Methods: We retrospectively reviewed surgically treated calcaneal fractures treated at a level I trauma center
over a period of 4.5 years. Twenty-two fractures were treated with a modified Palmer’s approach and 52 fractures
were treated with an extensile lateral approach. Functional outcomes were evaluated using the AAOS foot and
ankle questionnaire administered at office visits or by telephone interviews.

Results: Sixty-eight patients with 74 displaced intra-articular fractures, were followed for an average of 20
months. The population skewed predominantly male (76%; p<0.001). Mean age at time of injury was found to be
38.5 years (range, 15 to 70) with no significant differences noted between the extensile lateral and the modified
Palmer groups. The single largest mechanism of injury was fall from height (66.7%) with motor vehicle accident
the second most common (25.3%). Severity of initial injury, as evaluated by Sander’s classification, was not found
to be significantly different between groups. Two patients (9%) treated with the modified Palmer’s approach
developed wound complications that were successfully managed with wound care, while 24 patients (46%) treated
with the extensile lateral approach developed wound complications.

Conclusions: There was no significant difference in outcomes between the two groups. Our data would support
a decrease in the rate of wound complications associated with a modified Palmer’s approach as compared to the
extensile lateral approach without a significant change in the patient’s functional outcome or in the degree of
radiographic improvement.
Introduction popularized by Palmer10 in 1948. However, inability to
Calcaneal fractures account for 60% of fractures of the reproduce Palmer’s results led to a decline in the use of
tarsal bones and 2% of all fractures1,2. These fractures can this method.
be classified broadly into intra-articular and extra-articular During the last decade there has been renewed
types, with the intra-articular variant being more common, enthusiasm for the operative treatment of intra-articular
representing 70-75% of all fractures of the os calcis, fre- calcaneal fractures. Thordarson11, in a prospective, ran-
quently resulting from axial loading with varying degrees domized study comparing operative and nonoperative
of shear force3. Despite their frequency, advances in di- treatment of calcaneal fractures, found that functional
agnostic imaging and methods of internal fixation, their scores were better for fractures treated surgically than for
management remains challenging and controversial8. those managed by nonoperative methods. Buckley7, in
Treatment goals focus on the minimization of another prospective study, demonstrated similar results,
pain and the maximization of functional use of the foot but in a select group of patients. The meta-analysis by
in performance of activities of daily living. Current rec- Randle et al. 12 also reported a tendency for better results
ommendations for treatment range from nonoperative to with operative treatment.
operative intervention4,5,6,7, including primary arthrodesis, There remains, however, no consensus regarding
closed reduction and percutaneous pinning, and open the surgical approach, with many having been described,
reduction with internal fixation (ORIF). Open technique including medial, lateral, combined medial and lateral,
was described by Lenormant and Wilmoth9 in 1932 and extended lateral, and sinus tarsi approaches. Further, the
17
method of fixation remains a point of debate, with various
proponents advocating fixation with pins, screws, or plate
fixation with screws2,6,8,29,30,33,35,36,37,39,40.
While the literature suggests significant benefit
from operative management of these fractures, complica-
tions have been shown to be a common problem in many
studies, with some series reporting up to 33% incidence
of surgical wound breakdown13,14,15,16,17,18,19,20, a finding that
has prompted interest in minimally invasive techniques,
including the use of a modified Palmer’s approach, that
offer the potential to avoid compromising the already
damaged soft tissue envelope.
The purpose of this study was to review differ-
ences in outcomes and early wound complication between
patients with intra-articular calcaneal fractures treated
with a modified Palmer’s approach and the extensile
lateral approach.
Materials and Methods
A retrospective review was done of all calcaneal fractures
treated at a level I trauma center over a period of 4.5 years.
This study was given full approval from the local Institu-
tional Review Board and ethics committee.
Inclusion criteria for the entry into the study
consisted of adult patients older than 17 years of age, with
a diagnosis of displaced intra-articular calcaneal fractures
treated surgically, and with fractures meeting Sanders II
classification or greater, as defined by the findings in the
CT scan. Patient selection was restricted to those treated
by either a trauma surgeon or an orthopaedic foot special-
ist. Patient with concomitant injuries to the ipsilateral foot
were excluded.
Between Sept 2002 and April 2007, 72 patients
with 78 intra-articular calcaneal fractures underwent sur-
gical management. Four patients were excluded because
a primary fusion was done at the time of the initial treat-
ment. This resulted in a study population of 74 displaced
intra-articular fractures of the os calci in 68 patients. The
patients were divided into two groups: Group 1 consisted
of 22 fractures that were treated with the sinus tarsi ap-
proach; Group 2 included 52 fractures treated with the
extended lateral approach. Patients were initially treated
with ice and elevation. Timing for the surgical interven-
tion was delayed until soft tissue swelling had adequately
subsided and skin wrinkling was apparent.
Operative Technique
All the patients in this study underwent open reduction and
internal fixation through either an extensile lateral or sinus
tarsi approach, which are described in detail below.
Modified Palmer Lateral (Sinus Tarsi) Approach20: In this
approach, a straight incision is made on the lateral side of
the foot, approximately 1 to 1.5 cm distal to the tip of the
18
fibula and roughly perpendicular to the long axis of the
fibula. The incision starts approximately 1 cm posterior to
the fibula and continues distally for a total length of 3 to 4
cm. The incision is more transverse than that used by Palm-
er. We have found that this technique facilitates excellent
visualization of the posterior facet. The peroneal tendons
are identified and the distal portion of the retinaculum is
released. The tendons are mobilized and retracted anteri-
orly. The fat pad in the sinus tarsi is mobilized proximal to
distal, and the capsule of the posterior talocalcaneal joint is
opened if the capsule is intact. An elevator is used to align
the posterior facet with the medial sustentaculum fragment,
and the depressed posterior facet is reduced to the talus.
Kirschner wires are used to provisionally fix the posterior
facet to the medial fragment, and one or two lag screws
are passed across the fracture site from lateral to medial.
Once this posterior facet fragment is fixed to the medial
fragment and the joint has been successfully elevated,
correction of the varus deformity of the posterior tuberos-
ity is performed. A Schantz pin is inserted into the lateral
aspect of the calcaneus tuberosity. This pin is then pulled
downward to recreate the calcaneal height and to correct
the varus alignment. Often a laminar spreader is placed
between the tuberosity and posterior facet fragments to
help maintain height. K-wires are then passed from the
heel into the posterior facet fragment to provisionally
stabilize the tuberosity fragment. A 6.5 cannulated screw
is used to complete the fixation. Finally, bone grafting of
the defect below the posterior facet is performed by either
bone grafting with cancellous bone chips or with calcium
pyrophosphate bone cement. The wound is irrigated prior
to closure and closed with deep 2-0 absorbable sutures.
Skin closure is performed with 4-0 nylon sutures
Extensile Lateral Approach: Described by Benirschke et
al.15, a modification of the techniques described by Palmer
and Letournel was used. This is an extensile, L-shaped
lateral incision. The vertical part of the incision paral-
lels the long axis of the limb, the horizontal part roughly
parallels the plantar surface of the foot. A full thickness
flap is raised containing the peroneal tendons, sural nerve
and calcaneofibular ligament. This exposure allows vi-
sualization from the tuberosity to the calcaneo-cuboid
joint. Kirschner wires are placed in the fibula, talar neck
and cuboid. These are bent to minimize retraction on the
wound edges, using a so-called “no touch” technique.
Fracture reduction is initiated with elevation of the ex-
panded lateral wall, disimpaction of the posterior facet
and lag screw fixation of the facet to the sustentacular
segment. The tuberosity malalignment is then corrected,
as is the height. Provisional K wire fixation is utilized.
Fluoroscopic imaging with lateral, axial and Broden’s
projection are analyzed to assess reduction. Plate fixation
is then applied in the tuberosity, beneath the posterior facet
and along the anterior calcaneus. The surgical wound is
closed over an eighth-inch Hemovac drain. Wound closure
is accomplished with inverted 2-0 absorbable sutures to
close the subcutaneous layer and 3-0 nylon horizontal
mattress sutures for the skin.
Post-Operative Management
Postoperatively the management for the two groups was
similar; the patients were placed in a ‘bulky Jones’ dress-
ing with the extremity elevated for 48 hours. Intravenous
antibiotics were continued for 24-48 hours. When the
soft tissue swelling had abated, the patient was allowed
to begin active range of motion of the ankle and subtalar
joints. Walking on the injured foot was not permitted for
twelve weeks. Resumption of weight-bearing was initiated
at that time to patient tolerance. Individuals with unilateral
fractures were instructed to bear weight on the forefoot ini-
tially, with gradual resumption of heel-to-toe ambulation.
For the group treated with the modified Palmer’s approach,
25% weight bearing was allowed at eight weeks, which
was incremented by 25% for every week thereafter, by
twelve weeks the patient was full weight bearing.
Medical records and radiographs were retrospec-
tively reviewed. Wound complications were defined using
clinical criteria, specifically persistent drainage, inflam-
matory signs, events requiring unplanned procedure (i.e.
debridement), need for additional special wound care and
patients requiring treatment with antibiotics beyond two
weeks postoperatively.
Radiographic Data
The preoperative radiograph and computer tomography
(CT) scans were used to evaluate and classify the fracture.
All patients had displacement greater than 2mm from
anatomic position. Sander’s classification8,31, without
subclasses, was utilized in order to describe the fracture.
Lateral radiographs immediately after the surgery and
during the last clinic follow-up visit were also evaluated.
The angles of Gissane and Böhler, as measured on these
three sets of radiographs, were compared.
Functional Outcome
The AAOS foot and ankle questionnaire was used as a
measure of functional outcome. This is a patient-reported,
foot-specific instrument. The questionnaire has twenty
items that, combined, form the Foot and Ankle Core
Scale and an additional five questions that form the Shoe
Comfort Scale. The Core Scale can be further subdi-
vided into four subscales: Pain (nine items), Function
(six items), Stiffness and Swelling (two items) and Giving
Way (three items) 22. Patients with bilateral fracture were
asked to complete the questionnaire for the foot with the
most complaints and functional disability. For purposes
of comparison, the AAOS Foot and Ankle scores were
19
divided into the same groups used for the Maryland Foot
and Ankle score (MFS): a score of 90-100 was graded as
an excellent result; 75-89 as good; 50-74 as fair, and less
than 49 was graded as a failure. Patients were evaluated
during an office visit, or by telephone interview.
Results
Sixty-eight patients with 74 displaced intra-articular cal-
caneal fractures treated surgically were registered in the
study. Patients were followed on average for 20 months.
There were 50 (74%) men and 18 (26%) women (p<0.001).
The age of the patients at the time of injury was a mean of
39 (range, 15 to 70) (Table 1). Twenty-five fractures (33%)
were due to polytrauma and fifty (67%) fractures due to
isolated trauma. The trauma mechanism was a motor
vehicle accident in 19 fractures (25.3%), fall from height
represented 50 fractures (66.7%), and 6 (8%) fractures
were the result of other type of mechanism.
No significant differences between the groups
treated with the modified Palmer’s approach (Group 1)
and extensile lateral approach (Group 2) were detected
with regard to the age at the time of injury, mechanism of
injuries, or type of fracture classifications. Any wound
that did not heal primarily, required dressing changes,
local wound care, further surgery, became infected with
discharge, exhibited any signs of skin necrosis, or that
required antibiotics was recorded as wound complication
(Table 2). Two patients (9%) of Group 1 developed wound
complications, compared to twenty-four patients (46%) in
Group 2.
Further description of this subgroup of patients
was performed, examining these patients for additional
factors such as smoking history and diabetes. In the uni-
variate analysis, there were seven smokers in Group 1, one
diabetic, four worker’s compensation patients, and two
patients who experienced some type of wound problem.
In Group 2, there were nineteen smokers, eight
diabetics, twelve patients with worker’s compensation,
and twenty-four who experienced some type of wound
problem. Of note, all surgical wounds healed and no major
complications, such as deep infections or amputations,
were documented. Incision and drainage was required in
only three patients (5.7%) in Group 2, none in Group 1,
and no patients were noted to have developed osteomy-
elitis. Using the Pearson’s Chi-Square, in Group 2 it was
shown that smoking was significantly related to wound
problems (p=0.025), and diabetes trended towards, but was
not significantly related to, wound problems (p=0.067).
In Group 2 smoking trended towards, but was not signifi-
cantly related to, wound problems (p=0.075).
Four patients, one (4.5%) from Group 1 and three
(5.6%) from Group 2 had hardware removal at the time of
most recent follow up. Removal of the hardware was car-
ried out at the patient request, or if there was found to be
Table 1. Patient Demographics
  Modified Palmer’s Extensile Lateral
Total
Approach (Group 1) Approach (Group 2)
68 Patients
19 Patients 49 Patients
74 Fractures
22 Fractures 52 Fractures
Sex      
•    Male 50 (74%) 15 (79%) 35 (71%)
•    Female 18 (26%) 4 (21%) 14 (29%)
Age/mean (years) 38.5 (19-71) 36 (19-71) 41 (22-65)
Other Factors      
 Smoker 26 (38%) 7 (37%) 19 (39%)
 Diabetes 9 (13%) 1 (5%) 8 (16%)
Worker’s
16 (24%) 4 (21%) 12 (24%)
Compensation Injury
Wound Complications 26 (38%) 2 (11%) 24 (49%)

Table 2. Treatment of Wound Complications

  Modified Palmer’s Extensile Lateral

Approach (Group 1) Approach (Group 2)
  22 fractures 52 fractures
Wound Complications 2 (9%) 24 (46%)
I&D ----- 3 (5%)
Wound Care 2 (9%) 16 (30.7%)
Antibiotics Only ----- 5 (9.6%)

scar tethering and pain at the site of hardware insertion. No Discussion

significant difference was found between the two groups.
Late fusion was performed in one (4.5%) patient from the
sinus tarsi approach and in three (5.6%) from the lateral
extensile approach at the time of final follow-up.
Forty-four fractures (60%) were classified as
Sanders’s II, twenty-six (35%) as Sanders’s III and four
(5%) as Sander’s IV. No significant difference was found
between the groups (Table 3). Information on the Böhler’s
angle was available for fifty-nine cases. There were no
differences between the two groups at the time of last
follow-up (34° Group 1 vs. 31° Group 2) (Table 3).
Six patients had a bilateral injury; three in each
group. Sixty-nine was the maximum possible number of
recorded the AAOS foot and ankle scores. Eighty-four
percent of Group 1 subjects and seventy percent of Group
2 subjects were able to answer the questionnaire. The four
patients who had already undergone arthrodesis did not
complete the questionnaire and were considered to have
a poor result from the primary treatment. The functional
outcomes are presented in Table 4. In Group 1, 69% of
patients had a good to excellent result compared to 66%
in Group 2.
20
Debate on the treatment of calcaneal fractures continues
despite numerous published series. Recent retrospective
and small prospective studies have suggested that operative
care of displaced intra-articular calcaneal fractures had an
advantage over non-operative care6,13,31,32,38,40. The largest
prospective, randomized, multicenter study of Buckley et
al7 showed benefit of surgical over conservative treatment
in a selected number of patients, specifically those who are
younger, female, have a light-to-moderate workload in-
volving the foot, and those who are not receiving Worker’s
Compensation do well with operative care. Another study
suggests that patients treated operatively are 5.5 times less
likely to need a subtalar arthrodesis3.
Open reduction procedures have the main ad-
vantage of an anatomic reduction, restoring the subtalar
joint congruity, arch height, heel width and decreased
the possibility of lateral impingement. Proponents of this
approach emphasize that this anatomic restoration, with
a limited period of immobilization, lead to superior func-
tional results5,6,8,14,29.
Despite the high enthusiasm in recent years for
the operative treatment of intra-articular calcaneal frac-
tures, the soft tissue problems associated with it 15,16,17,18,19
Table 3. Radiographic Evaluation
  Modified Palmer’s Extensile Lateral
Total Cases
(Group 1) (Group 2)
Sander’s Classification      
Sanders II 44 (60%) 13 (59%) 31 (60%)
Sanders III 26 (35%) 6 (27%) 20 (38%)
Sanders IV 4 (5%) 3 (14%) 1 (2%)
Böhler’s Angle      
(mean ± SD)      
Initial   15.5° ± 16.5 12.4° ± 14.8
Immediate   36.8° ± 6 31.1° ± 7.8
Post-operative
Last Follow-up   34.1° ± 6.1 30.9° ± 8.2

has dismayed many authors. Additionally, there remains
poor consensus regarding the surgical approach to be used.
The literature describes a multitude of different approaches
to the calcaneus, including lateral, medial, the combined
lateral and medial approach, sinus tarsi and, currently the
most popular, the extensile lateral approach6,11,13,20,21.
The authors of this study propose that the use of a
modified Palmer’s approach as treatment of intra-articular
calcaneal fractures has the potential advantage of avoiding
compromise of the already damaged soft tissue coverage
on the lateral side. Since wound breakdown has the pos-
sibility of resulting in a more severe complication, such as
osteomyelitis or in some cases the need to perform flaps to
cover the soft tissue defect, and in extreme cases amputa-
tion, it is very important to avoid this complication.
Soft tissue complications in our series were less
frequent for the group of patients treated with a modified
Palmer’s approach (Group 1) versus the extensile lateral
approach (Group 2), (9% vs 46% respectively). These
rates approach those reported by Gupta et al.21 where the
incidence of wound complications was noted to be 3%
in the treatment of 32 fractures treated with the modified
Palmer’ approach.
Table 4. Outcome according to AAOS foot and ankle score. Bilateral injuries were counted once. Patients who had
subtalar fusion after the initial treatment were considered poor outcome.
Wound complications have proven to be a prob-
lem: Stephenson et al. 11 report an incidence of wound
breakdown of 27% while Benirschke et al.15,17 report 11%
rate of wound complications using the lateral approach.
The high incidence of wound complications in our series
may reflect the broad criteria that were used to define them.
Our results should be carefully analyzed as the disparity
in wound complications between the two groups may
somehow be attributed to the difference in the number
of patients (22 vs. 53), the disproportionate number of
ipsilateral injuries (9% vs. 21%) and number of patients
that sustained other, significant musculoskeletal injuries
(18% vs. 40%), (Group 1 modified Palmer’s approach
vs. Group 2 extensile lateral approach, respectively). Our
analysis, however, suggests that there is no significant
difference between patients with isolated injuries and
polytrauma injury.
Aktugle et al. 41, in a study of 36 displaced intra-
articular calcaneal fractures, showed poorer outcomes for
those patients who sustained polytrauma when compared
with patients with isolated injuries, independent of the type
of treatment chosen. This may be accounted for given the
tendency for polytrauma patients to have been subject to

Modified Palmer’s Approach Extensile Lateral Approach

Number of patients evaluated
Group 1 Group 2
16 (84%) 35 (71%)
AAOS Score Average ± SD   82.6 ± 10.9   77.5 ± 14.9
•    Excellent 31% 20%
•    Good 38% 46%
•    Fair 25% 20%
•    Poor 6% 9%
21
higher energy mechanisms of trauma, and the understand-
ing that the additional injury to other systems, in addition
to their significant orthopaedic injuries, leads to a major
catabolic response that has the potential for impairment
of soft tissue healing.
Radiographic findings demonstrated equivalent
improvement for the two groups, improving Böhler’s
angle, restoring measured values to within a normal range.
Because Group 1 subjects received only screw fixation
and bone graft or calcium pyrophosphate, there was a
concern during follow-up that a loss of reduction might
result, due to the poorer biomechanical advantages of this
type of fixation as compared to the use of plate and screw
fixation. Our data demonstrate that at the time of final
radiographic evaluation there was no evidence of further
collapse or change in Böhler’s angle. These findings are
in line with those reported by Gupta et al. 21, where the
use of the modified Palmer lateral approach, demonstrated
significant improvement in mean values of Böhler’s angle,
posterior facet angle, posterior facet depression, heel width
and calcaneal height on CT scans post-operatively.
At last follow-up the American Orthopaedic Foot
and Ankle score was classified as good to excellent in 66%
to 69%, with no difference found between groups. The
rate of late fusion similarly showed no difference and its
incidence was found to be similar to what has described
in the literature. Buckley et al. 3 found that the need for
arthrodesis was 4% in a series of patients treated with
ORIF. Csizy et al. 3 in a review of a prospective randomized
database, found a subtalar fusion rate of 9.8%.
In summary, this retrospective review study
comparing two different techniques for the treatment of
intra-articular calcaneal fractures found similar results
in outcome measures and improvement of radiographic
criteria. However, a lower rate of wound complications
was observed with the group of patients treated with the
modified Palmer’s approach. Differences may be attributed
References
1. Cave EF. Fractures of the os calcis- the problem in
general. Clin. Orthop. 1963;30:64- 644.
2. Eastwood DM, Gregg PJ, Atkins RM. Intra-articular
fractures of the calcaneum. Part I: pathological anatomy
and classification. J. Bone Joint Surg. 1993;7-B:183-
188.
22
to decreased operative time and dissection required. The
modified Palmer’s approach is more distal to the lateral
calcaneal artery and thus appears to avoid injury to this
artery, implicated as a cause of wound complications with
the extended lateral approach. Smoking seems to be the
other only factor associated with wound complications.
However, our results should be evaluated carefully. There
are limitations of our study. First, this is a retrospective
review, and is limited accordingly. Secondly, the number
of patients treated in the extensile lateral approach outnum-
ber those treated with the modified Palmer’s approach by
a ratio of 2:1. This disparity may have limited the ability
of this study to clearly identify significant differences.
Thirdly, we have used broad criteria to define wound
complications. In our series only three patients (5.7%) had
a severe infection that required additional intervention and
no patient required reconstruction with a flap, was diag-
nosed with osteomyelitis, or required amputation. These
findings are closer to what has been published in the recent
literature. Of all the patients treated in our institution with
the diagnosis of intra-articular calcaneal fracture, only one
required reconstruction with a flap, and of note, this patient
had initially been treated at another institution.
We conclude that both techniques are equally suc-
cessful at improving radiographic parameters. Outcomes
were similar for the two groups. Although there was a
difference in the rate of wound complications, there was
no significant difference in the incidence of severe com-
plications. There is a learning curve for both techniques.
In our study the surgical intervention was performed by
either an orthopaedic foot specialist or a trauma specialist,
and at least one study has suggested there is a significant
relationship between the deep infection rate, traumatic
subtalar arthritis and the institutional fracture load42.
Smoking seems to be the only other factor associated with
wound complications.
3. Csizy M, Buckley R, Tough S, Leighton R, Smith
J, McCormack R, Pate G, Petrie D, Galpin R. Dis-
placed intra-articular calcaneal fractures: variables
predicting late subtalar fusion. J Orthop Trauma. 2003
Feb;17(2):106-12.
4. Sanders R, Hansen S, Reynolds I. Fractures of the
calcaneus. In disorders of the foot and ankle. Medical
& Surgical Management. 2nd Ed., M. Jahhs, Ed., Phila-
delphia, W.B. Saunders, 1992; 2326-2354.
5. Lowery RBW, Calhoun JH. Fractures of the calca-
neus. Part I anatomy, injury mechanism, and classifica-
tion. Foot Ankle Int. 1996;17:230-235.
6. Bezes H, Massart P, Delvaux D, et al. The operative
treatment of intra-articular calcaneal fractures. Indica-
tions, technique and results in 237 cases. Clin Orthop.
1993;290:55-59.
7. Buckley R, Tough S, McCormack R, Pate G, Leigh-
ton R, Petrie D, Galpin R. Operative compared with non-
operative treatment of displaced intra-articular calcaneal
fractures: a prospective, randomized, controlled multi-
center trial. J Bone Joint Surg Am. 2002;84:1733-44.
8. Sander R. Current concepts review: displaced intra-
articular fractures of the calcaneus. J. Bone Joint Surg.
2000;82-A:225-250.
9. Lenormant C, Wilmoth P, Lecour P. A propos du
traitement sanglant des fractures du calcaneum. Bull.
Mem. Soc. Nat. Chir. 1928; 54:1353-1355.
10. Palmer I. The mechanism and treatment of fractures
of the calcaneus: open reduction with the use of cancel-
lous graft. J Bone and Joint Surg. 1948; 30-A:2-8.
11. Thordarson DB, Krieger LE. Operative vs. non-
operative treatment of intra-articular fractures of the
calcaneus: a prospective randomized trial. Foot Ankle
Int. 1996;17:2-9.
12. Randle JA, Kreder HJ, Stephen D, et al. Should cal-
caneal fractures be treated surgically? A meta-analysis.
Clin Orthop. 2000;377:217-227.
13. Stephenson JR. Treatment of displaced intra-
articular fractures of the calcaneus using medial and
lateral approaches, internal fixation and early motion.
J Bone Joint Surg Am. 1987;69:115-130.
14. Stephenson JR. Treatment of displaced intra-artic-
ular fractures of the calcaneus: a combined medial and
lateral approach. Clin Orthop. 1993;290:68-75
23
15. Harvey EJ, Grujic L, Early JS, Benirschke SK,
Sangeorzan BJ. Morbidity associated with ORIF of
intra-articular calcaneus fractures using a lateral ap-
proach. Foot Ankle Int. 2001;22:868-73.
16. Al-Mudhaffar M, Prasad CV, Mofidi A. Wound
complications following operative fixation of calcaneal
fractures. Injury. 2000;31:461-4.
17. Benirschke SK, Kramer PA. Wound healing compli-
cations in closed and open calcaneal fractures. J Orthop
Trauma. 2004;18:1-6.
18. Koski A, Kuokkanen H, Tukiainen E. Postoperative
wound complications after internal fixation of closed
calcaneal fractures: a retrospective analysis of 126
consecutive patients with 148 fractures. Scand J Surg.
2005;94:243-5.
19. Howard JL, Buckley R, McCormack G, et al.
Complications following management of displaced
intra-articular calcaneal fractures: A prospective ran-
domized trial comparing open reduction internal fixation
with nonoperative management. J Orthop Trauma. 2003
Apr;17(4):241-9.
20. Hospodar P, Guzman C, Johnson P, Uhl R. Treat-
ment of displaced calcaneus fractures using a mini-
mally invasive sinus tarsi approach. Orthopedics. 2008;
31:11-12.
21. Gupta A, Ghalambor N, Nihal A. The modified
palmer lateral approach for calcaneal fractures: wound
healing and postoperative computed tomographic evalu-
ation of fracture reduction. Foot and Ankle Int. 2003
Oct;24(10):744-53.
22. Yates B. Merriman’s assessment of the lower limb.
Third edition: 40-41.
23. Lowery RBW, Calhoun JH. Fractures of the calca-
neus. Part I anatomy, injury mechanism, and classifica-
tion. Foot Ankle Int. 1996;17:230-235.
24. Paley D, Hall H. Intra-articular fractures of the 34. Shuler FD, Conti SF, Gruen GS, Abidi NA.
calcaneus: a critical analysis of results and prognostic Wound-healing risk factors after open reduction and
factors. J Bone Joint Surg. 1993;75A:342-354. internal fixation of calcaneal fractures: does correction
of Bohler’s angle alter outcomes? Orthop Clin North
25. Buckley R, Tough S, McCormack R, Pate G,
Am. 2001;32:187-92.
Leighton R, Petrie D, Galpin R. Operative compared
with nonoperative treatment of displaced intra-artic- 35. Essex-Lopresti P. The mechanism, reduction
ular calcaneal fractures: a prospective, randomized, technique, and results in fractures of the os calci. Br. J.
controlled multicenter trial. J Bone Joint Surg Am. Surg. 1952;39:395.
2002;84:1733-44.
36. Burdeaux BD. Reduction of calcaneal fractures by
26. Stiegelmar R, McKee MD, Waddell JP, Schemitsch the McReynolds medial approach technique and experi-
EH. Outcome of foot injuries in multiply injured pa- mental basis. Clin Orthop. 1983;177:87-103.
tients. Orthop Clin North Am. 2001;32:193-204.
37. Benirschke SK, Sangeorzan BL. Extensile intra-
27. Zmurko MG, Karges DE. Functional outcome articular fractures of the foot: surgical management of
of patients following open reduction internal fixa- calcaneal fractures. Clin Orthop. 1993;290:76-86.
tion for bilateral calcaneus fractures. Foot Ankle Int.
38. Letournel E. Open treatment of acute calcaneal
2002;23:917-21.
fractures. Clin Orthop. 1993;290:60-67.
28. Buckley RE, Meek RN. Comparison of open versus
39. Ebraheim N, Elgafy H, et al. Sinus tarsi approach
closed reduction of intra-articular calcaneal fractures:
with trans-articular fixation for displaced intra-
a matched cohort in workmen. JJ Orthop Trauma.
articular fractures of the calcaneus. Foot Ankle Int.
1992;6:216-22.
2000;21(2):105-13.
29. Kundel K, Funk E, Brutscher M, Bickel R. Calca-
40. McGarvey WC, Burris MW, Clanton TO, Melissi-
neal fratures: operative versus non operative treatment.
nos EG. Calcaneal fractures: indirect reduction and
J Trauma. 1996;41:839-45.
external fixation. Foot Ankle Int. 2006;27:494-9.
30. Sanders R. Intra-articular fractures of the calcaneus:
41. Aktuglu K, Aydogan U. The functional outcome of
present state of the art. J Orthop Trauma. 1992;6:252-
displaced intra-articular calcaneal fractures: a compari-
65.
son between isolated cases and polytrauma patients.
31. Sanders R, Fortin P, Dipasquale DO, Wailing A.
42. Poeze M, Verbruggen JP, Brink PR. The relation-
Operative treatment in 120 displaced intra-articular
ship between the outcome of operatively treated cal-
calcaneal fractures. Clin. Orthop. 1993;290:87-95.
caneal fractures and institutional fracture load. J Bone
32. Bajammal S, Tornetta P 3 , Sanders D, Bhandari M.
rd
Joint Surg Am. 2008 May;90(5):1013-21.
Displaced intra-articular calcaneal fractures. J Orthop
Trauma. 2005 May-Jun;19(5):360-4.

33. Lawrence SJ, Grau GF. Evaluation and treatment

of open calcaneal fractures: a retrospective analysis.
Orthopedics. 2003;26:621-6.

24
 Comparison of Endoscopic and Open Carpal Tunnel Release in the Same Patient
Ali Hashemi, MD, Richard Whipple, MD, Erin L. Kaufman, MPH
Division of Orthopaedic Surgery, Albany Medical College, Albany, New York

Presented at the 2009 AMC Orthopaedic Surgery Annual Resident Thesis Day

Background: Considerable debate still exists over whether endoscopic carpal tunnel release is superior to con-
ventional open carpal tunnel release. The aim of this study is to retrospectively review a set of patients who have
had bilateral carpal tunnel surgery, open on one hand and endoscopic on the other hand, and look at their overall
function, satisfaction and return to activities.

Methods: Of 550 endoscopic releases done from December 2003 to October 2008, 33 patients had had previous
open carpal tunnel releases performed in the opposite hand prior to the endoscopic release. Of these 33 patients,
22 patients were evaluated for this study.

Results: Seventeen of 22 patients enrolled in this study had the initial (open) procedure performed in the domi-
nant hand. The average return to ADLs (Activities of Daily Living) and work was shorter for the endoscopic pro-
cedure. There were no wound healing problems noted with either procedure. There were no significant differences
in grip strength or pinch strength in the 2 groups. DASH (Disabilities of the Arm, Shoulder and Hand) scores
were also similar in both groups. Twenty of the 22 patients surveyed would recommend endoscopic surgery over
open carpal tunnel surgery to a family member or friend.

Conclusions: Patients who have undergone bilateral carpal tunnel surgery with one side done endoscopically
and one side open greatly favor the endoscopic surgery side from an overall satisfaction standpoint. This study
supports previous prospective studies that demonstrate faster short term return of function in the endoscopic group
compared to the open group.
Introduction
Median nerve compression at the wrist is the most common
compression neuropathy of the upper extremity. Annu-
ally, an estimated 1 million American adults have CTS
requiring carpal tunnel release. The cost of this has been
estimated to be over 2 billion dollars per year.1 Median
nerve compression at the wrist was first described by Sir
James Paget in 1854 following a distal radius fracture.2 In
1880 James Putnam described the first series of patients
with classical carpal tunnel symptoms with pain and
paresthesia in the median nerve distribution of the hand.3
The first carpal tunnel release was reportedly performed by
Herbert Galloway in 1924.4 Endoscopic techniques were
introduced to reduce the length of incision thereby decreas-
ing postoperative pain. There is still considerable debate
over the efficacy of endoscopic carpal tunnel versus open
in regards to postoperative morbidity and complications.
The first endoscopic carpal tunnel release was performed
by Okutsu et al in 1987 in Japan.5 The 2 portal endoscopic
technique was introduced by Chow in 1989 and the one
incision was described by Agee in 1992.6,7 Several studies
have been performed in order to compare endoscopic and
open carpal tunnel release but controversy still exists over
which procedure is superior. One study in the British JBJS
by Ferdinand et al in 2002 was a prospective randomized
study of patients with bilateral carpal tunnel syndrome.8
25
Each patient had undergone an open carpal tunnel re-
lease on one hand and a one incision endoscopic release
on the other hand. The results of this study showed no
advantage of the endoscopic release over open release in
terms of patient satisfaction and return to normal function.
Trumble et al in 2002 did a prospective randomized study
of 192 hands in 147 patients comparing the single incision
endoscopic release and the open release.9 The study found
that good functional outcomes and patient satisfaction are
achieved more quickly in the endoscopic group of patients.
This study also found that there was no significant cost
difference in the two procedures.
The purpose of this study is to evaluate patients
who have had bilateral carpal tunnel surgery with one
side done endoscopically and the other open. Complica-
tions, resolution time of symptoms, DASH scores, grip
strength, pinch strength, length of time to return to ADLs
and work and overall patient satisfaction will be recorded
for these patients.
Materials and Methods
In order to qualify for this study a patient would have
bilateral carpal tunnel surgery, one which was performed
open (by various surgeons) and one which was performed
via a single incision endoscopic technique by a single sur-
geon (R.W). Patients that had any type of revision surgery
were disqualified. Prior to chart review, IRB approval was
obtained for the study. Overall, 550 patients qualified for
chart review. Of these, 33 qualified for the study. Of these
33, only 22 could be included either because of refusal to
enroll in the study or inability to contact patient.
Prior to examining the patient, the examiner was
not told which hand was done endoscopically. Each patient
was examined for: grip strength, tip pinch and key pinch
strength, 2 point discrimination, carpal tunnel provocative
testing (Tinel’s, Phalen’s, Compression), abductor pol-
licis brevis strength, numbness in the palmar cutaneous
branch distribution, and scar length. Grip strength was
measured using a Jamar dynamometer. Pinch strength was
measured using a pinch gauge. Abductor pollicis brevis
strength was determined to be strong or weak based on
exam and relative to the contralateral side. Pillar pain was
also recorded for each patient. When there was a radial
sided pillar pain, thumb CMC testing and extensor ten-
don tenosynovitis testing was performed to rule out these
pathologies. Range of motion deficits for each hand and
wrist were also documented.
Each patient was asked to rank, on a scale of
1-5, their impression of each side. Patients also noted the
time it took to return to work and return to ADLs after
each surgery. The total number of postoperative visits to
physical therapy were recorded for each patient. Patients
were also asked if they had any residual symptoms such
as numbness, weakness, scar tenderness, swelling, and
loss of coordination on each side using a 0-5 scale. The
time to resolution of symptoms such as numbness, palm
tenderness, and hand weakness was also recorded for
each hand. Finally, the patients filled out a DASH form
for each extremity.
Results
Of the 22 open procedures, 17 were done on the dominant
hand. Of the 22 endoscopic procedures, 5 were done on
the dominant hand. There was a mean of 9 years between
open and endoscopic surgeries. The average return to work
for the endoscopic side was 21 days and for the open side it
was 25 days (p value =.195). The average return to ADLs
for the endoscopic side was 15 days, and for the open side
34 days (p value<.001). The average DASH score for the
endoscopic side was 20 and for the open side it was 16 (p
value =.665). Mean grip strength for the endoscopic side
was 21 and for the open side 20 kg (p value =.61).
Seventeen of the 22 patients had comorbid
conditions such as diabetes (3 patients), thyroid disease
(6 patients), hypertension (4 patients), and other medical
comorbidities (4 patients). One patient had a cervical
spine fusion after her carpal tunnel surgeries. She stated
that her median nerve symptoms had resolved after each
hand procedure. Of the patients that went to physical
therapy, the endoscopic group averaged 7 total visits, and
26
the open side averaged 13 visits. The average length of
the endoscopic scar was .94 cm, and the average length of
scar for the open side was 3.3 cm. There were no wound
healing complications in either group.
There were 6 cutaneous nerve complications
as defined by a patch area of numbness over or near the
thenar eminence. Three of these were endoscopic and
the other three were from the open group. Six patients
reported residual paresthesias in their hands, five of these
were in the open group. Of the 4 hands that had weak-
ness in APB testing, 2 were endoscopic and 2 were open.
Of the 4 hands that failed 2-point discrimination testing,
2 were endoscopic and 2 were open. During provocative
testing, 5 of the open procedure hands recreated symptoms
of carpal tunnel syndrome and 4 of the endoscopic hands
exhibited signs of carpal tunnel syndrome. Ten patients
demonstrated symptoms of pillar pain. Of these, 7 hands
were done endoscopically.
When patients were asked which surgery they
would recommend for a family member or friend, 20 of
22 patients recommended the endoscopic surgery. The
most common response for why was that it was a faster
recovery, less pain and a smaller scar.
Discussion
From the results of this study it appears that the single
incision endoscopic carpal tunnel surgery offers a shorter
recovery time and better overall satisfaction for the patient
when compared to conventional open carpal tunnel release.
While the short term results appear to be improved for the
endoscopic group, the data does not appear to demonstrate
that the long term results of these two techniques are any
different as evidenced by similar motor strengths and
DASH scores. This study offers a unique perspective
for the patient and the examiner when comparing both
hands. Furthermore there does not seem to be a significant
increase in complications associated with either surgery
in this particular study.
There are several limitations to this study. This is
not a randomized, prospective study with a large number
of patients which certainly can question the statistical sig-
nificance of the results. While the endoscopic surgery was
done by a single surgeon, the open sides were done by vari-
ous surgeons with different surgical training (Neurosur-
gery, Orthopaedic Surgery, and Plastic surgery). Moreover
there may have been bias in the endoscopic group. These
patients may have selected the endoscopic procedure after
their open procedure because of their level of dissatisfac-
tion with their previous surgeon. Another source of bias
in this study is the fact that more open procedures were
done on the dominant hand of the patient. In this scenario
the patient is more likely to use the dominant hand more
compared to the non-dominant, which could delay the
recovery time compared to the endoscopic side.
 References
1. Palmer DH, Hanrahan LP. Social and economic-
cocosts of carpal tunnel surgery. Instr Course Lect.
1995;44:167-72.
2. Paget J.  Lectures on surgical pathology.  Philadel-
phia, Pa:  Lindsay and Blakiston;  1854.
3. Marie P, Foix C.  Atrophie isolee de l’eminence
thenar d’origine nevritique: role du ligament annulaire
anterieur du carpe dans la pathogenie de la leion.  Revue
Neurology (Paris).  1913;26:647-9.
4. Amadio PC. The first carpal tunnel release? J Hand
Surg Br 1995;20(1):40.
5. Okutsu I, Ninomiya S, Takatori Y, Ugawa Y. En-
doscopic management of carpal tunnel syndrome.
Arthroscopy. 1989;5(1):11-8.
27
6. Chow JC. Endoscopic release of the carpal ligament:
a new technique for carpal tunnel syndrome. Arthros-
copy. 1989;5(1):19-24.
7. Agee JM, McCarroll HR, Jr., Tortosa RD, Berry DA,
Szabo RM, Peimer CA. Endoscopic release of the carpal
tunnel: a randomized prospective multicenter study. J
Hand Surg Am. 1992;17(6):987-95.
8. Ferdinand RD, MacLean JGB. Endoscopic versus
open carpal tunnel release in bilateral carpal tunnel
syndrome: a prospective, randomized, blinded assess-
ment. JBJS Br April 2002.
9. Trumble TE, Diao E, Abrams RA, Gilbert-Anderson
MM. Single-portal endoscopic carpal tunnel release
compared with open release: a prospective, randomized
trial. JBJS July 2002.
 Total Shoulder Arthroplasty vs. Hemiarthroplasty for Osteoarthritis:
A Retrospective Review of Clinical Outcomes
Brent Carlson, MD, John Czajka, MD, Toni Hansen, CCRC, Patrick Marinello, Ashar Ata
Division of Orthopaedic Surgery, Albany Medical College Albany, NY

Presented at the 2010 AMC Orthopaedic Surgery Annual Resident Thesis Day

Background:
Background: Osteoarthritis
Osteoarthritisofofthe
theshoulder
shouldercan
canbebetreated
treatedbybyboth total
either arthroplasty
total arthroplastyandorhemiarthroplasty.
hemiarthroplasty. Several
Sev-
studies comparing
eral studies the two
comparing thetechniques have have
two techniques foundfound
eithereither
slightly betterbetter
slightly resultsresults
for total
for arthroplasty at short-term
total arthroplasty at short-
follow-up or no difference.
term follow-up The purpose
or no difference. of thisof
The purpose study
this was
studytowas
compare the outcomes
to compare of hemiarthroplasty
the outcomes and total
of hemiarthroplasty and
total shoulder
shoulder arthroplasty
arthroplasty in patients
in patients with osteoarthritis
with osteoarthritis and anand an intact
intact rotatorrotator
cuff. cuff.

Methods:and
Materials Thirty-four
Methods: shoulders withshoulders
Thirty-four osteoarthritis
withand an intact rotator
osteoarthritis and ancuff, who
intact had received
rotator cuff , whoa total or hemi-a
had received
arthroplasty,
total were clinically
or hemiarthroplasty wereevaluated
clinicallywith range of
evaluated motion,
with rangestrength, University
of motion, strength,of CaliforniaofatCalifornia
University Los Angelesat Los
(UCLA)(UCLA)
Angeles shouldershoulder
scale and the and
scale Western Ontario Ontario
the Western Osteoarthritis of the Shoulder
Osteoarthritis (WOOS)(WOOS)
of the Shoulder index. index.

Results: No
Results: Nosignificant
significantdifferences
differenceswere
werefound
foundbetween
betweentotal
totalshoulder
shoulderarthroplasty
arthroplastyand
andhemiarthroplasty
hemiarthroplastyregarding
regard-
ing shoulder
shoulder scores,
scores, strength,
strength, flexion,
flexion, abduction
abduction or internal
or internal rotation
rotation at anataverage
an average 39 months
39 months follow-up
follow-up (15-145
(15-145 months).
months). A difference in external rotation was found favoring total shoulder arthroplasty
A difference in external rotation was found favoring total shoulder arthroplasty (14 ; p = 0.03).
0 (140
; p = 0.03).

Conclusions:AtAtananaverage
Conclusion: average3939months
monthsfollow-up,
follow-up,thetheresults
resultsofofprimary
primarytotal
totalshoulder
shoulderarthroplasty
arthroplastyand
andhemiarthro-
hemiar-
throplasty for osteoarthritis with an intact rotator cuff are similar for range of motion, strength, and quality
plasty for osteoarthritis with an intact rotator cuff are similar for range of motion, strength, and quality of life. of life.

Introduction Materials and Methods

Charles Neer II designed the first modern shoulder hemi-
arthroplasty in 1951. Although it was initially designed
for fracture-dislocations and avascular necrosis, it was
also used for arthritis. Problems using the prosthesis in
rotator-cuff deficient shoulders and with glenoid erosions
prompted Neer’s next design, the total shoulder arthro-
plasty, in 1974. Since that time, both general shoulder
design types have been used for glenohumeral arthritis.
Significant controversy still exists, however,
in patients with osteoarthritis and an intact rotator cuff.
Several studies have compared outcomes after total and
hemiarthroplasty for osteoarthritis and have found either
slight favor for total arthroplasty regarding pain relief
and flexion at short-term follow-up or no difference1-6.
Their heterogeneity, inconsistent use of shoulder scores,
occasional use of patients with rotator cuff tears, and
short-term follow-up make it difficult to reach a conclu-
sion on the subject.
Further, failed hemiarthroplasties are often
revised to total shoulders and failed total arthroplasties
often convert to hemiarthroplasties. Whether these revised
shoulders are clinically similar to primary arthroplasties is
also debated. This is important, as some may claim that a
benefit of hemiarthroplasty is the easy conversion to total
arthroplasty at a later date, for example.The purpose of this
study, was to compare the outcomes of hemiarthroplasty
and total shoulder arthroplasty in patients with osteoar-
thritis and an intact rotator cuff.
Patient Selection
The billing records of a large orthopaedic group (Capital
Region Orthopaedics in Albany, NY) were searched for
all cases of shoulder hemiarthroplasty and total shoulder
replacement since 2000. Search criteria were shoulder
arthroplasty for reason other than fracture (CPT 23470,
23472) and surgery performed by an orthopaedist perform-
ing more than one arthroplasty per year. Charts were then
reviewed to identify patients for inclusion.
Inclusion criteria:
• Diagnosis of osteoarthritis in clinical or operative
notes.
• Intact rotator cuff in operative note.
• Surgery performed greater than one year prior to
study.
Exclusion criteria:
• Other arthritic diagnosis such as rheumatoid arthritis,
gouty arthritis, cuff tear arthropathy or avascular
necrosis of the humeral head.
• Diagnosis of rotator cuff tear or lack of rotator cuff
notation in chart.
• Less than one year follow up.
• Revision of primary arthroplasty.
Seventy seven shoulders (33 hemiarthroplasties
and 44 total arthroplasties) performed by 6 surgeons fit the
criteria for inclusion in the study. Thirty-four shoulders

28
Table 1: Range of Motion for Primary Shoulder Arthroplasty

Motion Type Hemi (avg) TSA (avg) Difference (CI) p-value z-score

Flexion 1230 1320 90 (-29 to 11) 0.3641 0.3850

Abduction 1040 1150 110 (-36 to 15) 0.4114 0.5235

Int. Rotation 650 600 50 (-8 to 19) 0.4459 0.3652

Ext. Rotation 330 470 140 (-26 to -1) 0.0300 0.0676

TSA = Total Shoulder Arthroplasty, CI = Confidence Interval

in 33 patients were available for clinical evaluation, and
form the study group.
Clinical Evaluation
Patients were asked to return to the office for a physi-
cal exam, evaluation by the UCLA Shoulder Scale, and
completion of a questionnaire. Range of motion and
strength were tested through flexion, abduction, internal
rotation and external rotation. The questionnaire was the
WOOS Index, a validated scoring system for quality of
life assessment.
Consent
Our study protocol was approved by our facility’s Insti-
tutional Review Board. All patients provided consent for
participation.
Data Analysis
T-tests and ranksum analysis were used to compare the
two groups regarding range of motion, strength, the UCLA
scale and the WOOS index.
Results
Of the 34 shoulders studied, 13 received hemiarthroplasty
and 21 received total arthroplasty. The average follow-
up was 39 months and the median was 31 months (range
14-145 months).
Within the hemiarthroplasty group, the average
age of patients was 63 years. Average follow-up was
48 months and median was 39 months (range 15- 145
months). There were 9 men and 4 women. Three shoul-
ders had undergone a previous procedure. One received
a labral debridement, one a debridement and excision of
loose bodies, and one had excision of loose bodies, acro-
mioplasty and open distal clavicle excision.
Within the total arthroplasty group, the average
age of patients was 68 years. Average follow-up was 32
months and median was 28 months (range 14- 79 months).
There were 15 men and 5 women. Two shoulders had
previously undergone arthroscopic labral debridement.
Range of Motion
There was no significant difference in range of motion
between the two groups for flexion, abduction, or inter-
nal rotation using t-test or ranksum analysis. There was
significantly better external rotation with total arthro-
plasty, 470 vs. 330 (p=0.03) using the t-test. Ranksum
analysis found no significant difference in external rotation
(z=0.0676) (Table 1).
Strength
No differences were found comparing strength of flexion,
abduction, internal rotation or external rotation for the
primary or revision groups using either t-test or ranksum
analysis (Table 2).

Table 2: Strength for Primary Shoulder Arthroplasty

Strength Type Hemi (avg) TSA (avg) Difference (CI) p-value z-score

0.15 (-0.23 to
Flexion 4.76 4.61 0.4347 0.4890
0.54)
0.23 (-0.60 to
Abduction 4.53 4.76 0.2423 0.3152
0.15)
0.05 (-0.35 to
Int. Rotation 4.38 4.33 0.7940 0.8530
0.45)
0.14 (-0.49 to
Ext. Rotation 4.53 4.67 0.4699 0.4614
0.23)
TSA = Total Shoulder Arthroplasty, CI = Confidence Interval

29
Table 3: Shoulder Scores for Primary Shoulder Arthroplasty
Shoulder Score Hemi (avg) TSA (avg)
UCLA 28.1 30.0
WOOS 79 88
TSA = Total Shoulder Arthroplasty, CI = Confidence Interval
Shoulder Scores
No difference was found in either the UCLA scale or
WOOS index between the primary arthroplasty groups
using t-test and ranksum analysis (Table 3).
Discussion
Our study showed little difference between primary total
shoulder arthroplasty and hemiarthroplasty for osteoarthri-
tis with an intact rotator cuff. While no differences were
found in strength or the motions of flexion, abduction or
internal rotation, a 140 difference in external rotation in
favor of total arthroplasty was significant. It is possible
that the exposure of the glenoid in total shoulders may
lead some surgeons to also release the subscapularis off
of the anterior scapula, leading to greater external rotation
postoperatively. The z-score of the ranksum analysis did
not show significance for external rotation, but showed
a trend toward significance, and the confidence interval
for the t-test was large, suggesting a large variability in
data. Flexion and external rotation were also found to be
better in total shoulders at 4 years in another study, but
that study included patients with rotator cuff tears in the
hemiarthroplasty group, introducing bias in favor of the
total shoulder group.2
As with another study comparing WOOS index
scores, we found no difference between the primary ar-
throplasty groups.3 While the previous study found no
difference at 2 years, our findings are at an average of 3.25
years. The WOOS index measures quality of life through
physical symptoms, recreation/work, lifestyle, and emo-
tions. The score is given as a number out of 100, with 100
References
1. Haines JF, et al. The results of arthroplasty in os-
teoarthritis of the shoulder. J Bone Joint Surg Br. 2006;
88:496-501
2. Edwards TB, et al. A comparison of hemiarthroplasty
and total shoulder arthroplasty in the treatment of pri-
mary glenohumeral osteoarthritis. J Shoulder Elbow
Surg. 2003; 12:207-213
30
Difference (CI) p-value z-score
1.9 (-5.3 to 1.5) 0.2711 0.0840
9 (-424 to 89) 0.1937 0.0956
being a completely unaffected quality of life as relates to
the shoulder. This index was developed specifically for
shoulder osteoarthritis and consists of 19 questions.
We also included the UCLA shoulder score
which, unlike the Constant, ASES, SPADI or SST, was
designed to assess outcome after shoulder arthroplasty.
While many studies utilize other shoulder scores or sev-
eral scores for comparison, only the UCLA score assesses
pain, function, ROM, strength and patient satisfaction. We
found no significant difference in UCLA score between
the two groups.
This study included only patients with an intact
rotator cuff noted at the time of surgery. Large rotator cuff
tears in total shoulder arthroplasty are associated with less
postoperative motion, more pain, and loose components7.
This is likely due to a “rocking horse effect” of the head
on the glenoid and, thus, stress on the glenoid component/
glenoid bone interface.8 Despite this knowledge, sev-
eral studies investigating differences between total and
hemiarthroplasty for shoulder osteoarthritis have failed
to exclude patients with rotator cuff tears from the study
population1,2,5,9.
In conclusion, primary total shoulder arthroplasty
and hemiarthroplasty for osteoarthritis with an intact
rotator cuff are similar in range of motion, strength, and
quality of life at an average 3.25 years follow-up. The
statistically significant difference found for external
rotation favoring total shoulder arthroplasty may not be
clinically relevant.
3. Lo IK, et al. Quality of life outcome following
hemiarthroplasty or total shoulder arthroplasty in pa-
tients with osteoarthritis. J Bone Joint Surg Am. 2005;
87:2178-2185
4. Radnay CS, et al. Total shoulder replacement com-
pared with humeral head replacement for the
treatment of primary glenohumeral osteoarthritis. J
Shoulder Elbow Surg. 2007; 16:396-402
5. Bryant D, et al. A comparison of pain, strength, 8. Franklin JL, et al. Glenoid loosening in total shoulder
range of motion and functional outcomes after hemi- arthroplasty: association with rotator cuff arthroplasty.
arthroplasty and total shoulder arthroplasty in patients J Arthroplasty. 1988; 3: 3
with osteoarthritis of the shoulder. J Bone Joint Surg
9. Sperling JW, et al. Minimum 15-year follow-up of
Am. 2005; 87:1947-1956
Neer hemiarthroplasty and total shoulder arthroplasty in
6. Clinton J, et al. Nonprosthetic glenoid arthroplasty patients aged fifty years or younger. J Shoulder Elbow
with humeral hemiarthroplasty and total shoulder ar- Surg. 2004; 604-613
throplasty yield similar self-assessed outcomes in the
management of comparable patients with glenohumeral
arthritis. J Shoulder Elbow Surg. 2007; 16:534-538

7. Edwards TB, et al. The influence of rotator cuff

disease on the results of shoulder arthroplasty for pri-
mary osteoarthritis. J Bone Joint Surg Am. 2002; 84:
2240-2248

31
Adjacent Level Effects: A Novel In Vivo/In Vitro, Active/Passive Protocol for Assessing Redistribu-
tion of Motion of the Entire Thoracolumbar Spine Following Surgical Intervention
1
Kyle A. Elsabee, 2John D. Wason, 3James Boyle, MD, 3Ronniel Nazarian, MD, 2John T. Wen, Ph.D.,
3
Allen L. Carl, MD, 1Eric H. Ledet, Ph.D.
1
Department of Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, NY
2
Center for Automation Technologies and Systems, Rensselaer Polytechnic Institute, Troy, NY
3
Division of Orthopaedic Surgery, The Albany Medical College, Albany, NY

Adjacent segment disease is recognized as a potential complication of spinal fusion. In vitro testing and finite ele-
ment analyses largely indicates that fusion instrumentation placed at one level significantly changes the stress and
motion at adjacent levels. However, advances in dynamic stabilization systems and motion preserving technolo-
gies show promise for minimizing adjacent level effects. The purpose of this study was to use a new in vivo/in
vitro active/passive biomechanical testing method to quantitatively assess the three-dimensional motion redistribu-
tion of the entire thoracolumbar spine following surgical intervention during full dynamic range of motion based
on in vivo kinematics, without the confounding factors inherent in a clinical trial.

Whole bony-ligamentous thoracolumbar spines (C7 through S1) were harvested from cadavers. C7 was attached
to the end-effector of the six degree-of-freedom robot and three-dimensional in vivo kinematic data from nor-
mal healthy volunteers was loaded into the robot controller. The three-dimensional motion of 11 vertebrae were
simultaneously measured during replication of in vivo motion. After completion of the intact motion assessment,
specimens were then subjected to surgical interventions at the L4-L5 level including laminotomy, laminectomy, dis-
cectomy, facetectomy, facet screw fixation, and then pedicle screw fixation, followed by two level pedicle fixation
at L3-L5. The range of motion testing was repeated and recorded between each intervention.

With successive interventions (laminotomy, laminectomy, discectomy, facetectomy), intersegmental motion at

the L4-5 level increased and there was a redistribution of motion at adjacent levels. Pedicle screw fixation at the
L4-5 level resulted in a reduction in motion at L4-5 and a reduction at L3-4 with concomitant increases in motion at
all other levels relative to the intact state. These preliminary data indicate that surgical interventions at the L4-5
level affect motion at all other levels of the spine (from S1 up to at least T7). There is a redistribution of motion
at adjacent levels following surgical intervention that compensates for the changes in motion at the index level.
These results suggest that an intervention at one level does not strongly predispose the immediate super- and sub-
adjacent levels to substantially altered kinematics while sparing other levels, but rather that all levels are subjected
to some redistribution.
Introduction the most common changes include; disc degeneration,
Single and multi-segment spinal fusions have been used listhesis, instability, disc herniation, facet hypertrophy,
to treat spinal instability and chronic pain. Relatively high and osteophyte formation. The ambiguity of the defini-
rates of success have been reported and fusion remains a tion of this degenerative process has made the study of
good option after failure of conservative treatment. Sur- its etiology difficult.
geons and researchers as early as five decades ago began Although many reports have correlated ad-
to describe what has become known as adjacent level jacent level disease to poor functional outcomes and
disease1. Clinical follow-up of fusion patients has led to to the need for further surgical intervention, others
the publication of numerous scientific articles demonstrat- have doubted the cause and effect relationship. Some
ing changes at levels adjacent to fused segments. These report earlier degeneration in patients with fusions2,4-7
largely radiographic changes have been linked to poor while others fail to find differences between ongo-
outcomes and to time-dependant symptoms, thus adjacent ing degenerative processes and problems caused by
segment disease is recognized as a potential complication surgery8,9-11.
of spinal fusion2,3. In vitro testing and finite element analyses
Adjacent segment disease is a term with broad largely indicates that fusion instrumentation placed
meaning. It encompasses any changes that occur in a at one level significantly changes the stress and motion
mobile segment adjacent to a spinal fusion1. Some of at adjacent levels.12,13 Other studies have shown that
32
instrumented fusion causes an increase in intradiscal
pressure14-17. Some reports indicate an increase in adjacent
level motion following fusion18,19 while others report a
decrease20.
There are many reasons for the disparity in
results. Clinical studies are fraught with the difficulty of
discerning the difference between pathology at adjacent
levels whose progression was inherent, versus pathology
that was initiated or exacerbated by fusion at an adjacent
level. In vitro testing and finite element models afford the
luxury of eliminating the confounding effects of natural
disease progression, yet, ex vivo models are generally
based on application of arbitrary loads or motions that
may have limited clinical relevance.
Advanced image analysis techniques have be-
come available for use in quantifying motion of the spine
using plain radiographs. This has allowed quantitative
analysis of range of motion and center of rotation of
patients before and after treatment and over time fol-
lowing treatment11. The comparison of before and after
data allows quantification of spinal motion redistribution.
However, this valuable technology is limited to analysis of
static images and also is limited to analyzing the number
of levels that are captured on a plain film.
In vitro mechanical testing has also been used to
assess adjacent level affects. The most recent techniques
have moved beyond the “flexibility” and “stiffness”
protocols originally proposed by Panjabi21,22. The newer
“hybrid” protocol encompasses the concepts of the flex-
ibility and stiffness protocols but is based on a recently
posed theorem: that the overall range of motion of the
spine does not change (relative to its intact state) as a result
of surgical intervention, even if that surgical intervention
affects motion at a treated level23. Thus, the overall range
of motion of the trunk will remain the same even if a surgi-
cal intervention (such as fusion in the lumbar spine) alters
the kinematics of one level. Likewise, the overall range of
motion of the head and neck will remain the same follow-
ing a cervical spine intervention. The theoretical basis for
this is that muscles, tendons, ligaments, and other intact
tissues are trained to achieve an inherent range of motion
and by replacing one component of the system (such as
replacing an intervertebral disc with an implant) the inher-
ent range of motion does not necessarily change. Although
this theory has not yet explicitly been proven, there is
evidence to support it24,25 and it is now a widely accepted
basis for in vitro for biomechanical testing23,26.
Advances in dynamic stabilization systems
and motion preserving technologies show promise for
minimizing adjacent level effects, however this remains
controversial27-30. Data from in vitro testing and finite
element models are only able to predict clinical adjacent
level effects to the extent that the experimental or simu-
lated mechanical and kinematic environment mimics in
33
vivo conditions. When in vitro testing is conducted on
specimens consisting of three motion segments or fewer,
additional artifact is introduced because the redistribu-
tion of stress and motion is constrained at the super and
subadjacent levels.
The purpose of this study was to use a new in
vivo/in vitro, active/passive biomechanical testing method
to quantitatively assess the three-dimensional motion re-
distribution of the entire thoracolumbar spine following
surgical intervention during full dynamic range of motion
based on in vivo kinematics.
Materials and Methods
We have previously conducted work to develop a hybrid
in vivo to in vitro methodology for quantifying adjacent
level effects. The method is based on whole spine, three
dimensional kinematic data collected from volunteer sub-
jects performing voluntary trunk range of motion activities.
The data were collected using three dimensional motion
tracking sensors (Flock of Birds, Ascension Technology,
Burlington, VT) which were mounted on the skin super-
ficial to C7, T12, and S1. The data were collected at over
100 samples per second throughout range of motion to the
maximum extent achievable by the subjects. Subject mo-
tion was also recorded on high definition video during the
course of data acquisition. Previous radiographic studies
have shown less than 2° of difference between superficial
sensors mounted on the skin in this fashion and actual
vertebral movement31. Data were collected during flexion,
extension, right and left lateral bending, and rotation as
shown in Figure 1.
Following in vivo data collection, each data set
recorded from the volunteers was reduced to command
set to drive a robotic simulator (Puma 560, Rensselaer
Polytechnic Institute, Troy, NY). The robotic system was
used for in vitro testing.
Specimen Preparation
In this pilot study, three whole bony-ligamentous tho-
racolumbar spines (C7 through S1) were harvested from
male cadavers within 24 hours of death. Specimens were
harvested by transection through the C6-C7 disc space
followed by circumferential release of extraneous soft
tissue, osteotomies through each rib approximately 10 cm
lateral to the vertebrocostal joint, and transection through
the sacrum caudal to S1 with a reciprocating saw. Care
was taken not to disrupt the facet capsules, ligaments, or
intervertebral discs of all levels. Once excised, the bony-
ligamentous spines were wrapped in saline soaked gauze
and stored frozen at -20° C in plastic bags until the time
of use.
The skeletal dimensions of the cadaver (head to
toe length) and the C7 to S1 length were measured at the
time of harvest. Prior to in vitro testing, the dimensions of
Figure 1: Three-dimensional data were previously collected from volunteers during flexion, extension, lateral bend-
ing, and torsion.
the cadaver and spine were compared to those in the data-
base of in vivo subjects and the most appropriate in vivo
data set (with the closest spinal and skeletal dimensions
to the cadaveric spine) was retrieved for testing.
Intact Motion Assessment
At the time of testing, each cadaveric spine was thawed
to room temperature and kept moist throughout the ex-
periment by saline spray. The caudal end of the spine
was potted in a cup fixture with Low (47° C) Melting
temperature Alloy (LMA). Screws were placed in the
sacrum to facilitate rigid fixation of the sacrum in the
LMA. Once potted in LMA, the caudal end of the spine
and fixture were fixed to a platform rigidly attached to the
floor. The cranial end of the spine, (C7,) was attached to
the end-effector of the six degree-of-freedom robot using
a custom clamp, as shown in Figure 2.
The three-dimensional kinematic data from the
appropriate in vivo subject was loaded into the robot
controller. The data are used to control the end-effector
relative to the sacrum such that the position and orientation
Figure 2: C7 was fixed to the end-effector of the robot
using a custom clamp.
34
of the C7 vertebra relative to the sacrum precisely matches
that of the position and orientation of the volunteer. The
robot control is maintained throughout all range of motion
maneuvers (flexion/extension, lateral bending, torsion).
Position and orientation sensors (Flock of Birds,
Ascension Technology, Burlington, VT) were mounted
to 11 vertebrae (T7 to L5) using polycarbonate mounting
fixtures and screws (Figure 3). The robot then moved C7
relative to S1 in an identical fashion to the active motion
of the in vivo volunteer in left and right lateral bending.
The three-dimensional motion of the 11 vertebrae were
simultaneously measured at a rate of approximately 60
samples per second during spine maneuvers with the spine
in its intact state as a baseline measurement. The spine was
moved at 50% of the speed of the volunteers to eliminate
any inertial effects. The spine was moved through three
sets of lateral bending motion and the data from the 3rd
set was used for comparison.
Figure 3: Eleven sensors were mounted to thoracic and
lumbar vertebrae (left) to track three-dimensional mo-
tion during lateral bending (right).
Clinical Therapeutic Interventions
After completion of the intact motion assessment, speci-
mens were then subjected to surgical interventions at
the L4-L5 level. Each spine then underwent a successive
laminotomy, laminectomy, discectomy, facetectomy, facet
screw fixation, and then pedicle screw fixation, followed
by two level pedicle fixation at L3-L5. The range of mo-
tion testing was repeated and recorded between each
intervention. The range of motion for each vertebra was
compared to the intact state. Changes in range of motion
at each level were calculated and were used to calculate
the redistribution of motion at adjacent levels.
Motion Redistribution Assessment
For each treatment, at each of the 11 recorded levels of
the spine, the intact state and post-intervention three di-
mensional kinematic data were compared quantitatively.
These data were used to quantify the redistribution of mo-
tion (if any), and thus quantify the adjacent level effects
of each treatment.
Results
Representation of the typical flexion/extension motion
for the volunteers is shown in Figure 4. All previously
collected data sets were successfully converted to com-
mand sets for the robotic motion simulator. Preliminary
data from in vitro testing of the intact spines showed a
Figure 4: The data previously collected from normal healthy volunteers during flexion/extension, lateral bending,
and torsion were converted to a command set for the robotic motion simulator.
35
pattern of increasing intersegmental motion from caudal
to cranial with peaks at T11-12 and T7-8 as shown in Figure
5. The patterns in intersegmental motion are consistent
with previous reports from in vivo studies32.
With successive interventions (laminotomy,
laminectomy, discectomy, facetectomy), intersegmental
motion at the L4-5 level increased and there was a redis-
tribution of motion at adjacent levels. Instrumentation
reduced motion at the index level with a redistribution at
adjacent levels as shown in Figures 6 and 7. Pedicle screw
fixation at the L4-5 level resulted in a reduction in motion
at L4-5 and a reduction at L3-4 with concomitant increases
in motion at all other levels relative to the intact state.
Discussion
In this pilot study, we have introduced a novel methods for
quantitatively evaluating the effects of surgical interven-
tion at one level of the lumbar spine on the kinematics of
the entire thoracolumbar spine. The model is novel in that
in vitro spine motion is based on in vivo active voluntary
motion, not arbitrary motions or moments applied to the
spine. As such, the fidelity with which in vivo motion is
reproduced is higher than other in vitro testing protocols
and intuitively the predictive value of in vivo effects from
in vitro testing is also high.
The intersegmental distribution of motion of
the intact specimens in vitro is consistent with previous
Figure 5: Mean intersegmental motion of intact in vitro specimens increased from caudal to cranial with characteris-
tic peaks at T11-12.

Figure 6: Frontal plane motion was captured during lateral bending.

36
Figure 7: Intersegmental motion during lateral bending intact vs. facetectomy vs. pedicle screw.
in vivo studies32. This validates the model with respect to
reproduction of in vivo motion in vitro. This indicates that
although in vitro motion of the entire thoracolumbar spine
is only controlled by manipulating C7 relative to S1, the
passive intersegmental motion at all levels of the spine dur-
ing in vitro testing is similar to in vivo active motion.
These preliminary data indicate that surgical
interventions at the L4-5 level affect motion at all other
levels of the spine (from S1 up to at least T7). There is a re-
distribution of motion at adjacent levels following surgical
intervention that compensates for the changes in motion
at the index level. Relative to an intact state, fusion at the
L4-5 level reduced motion at L4-5 and L3-4 and increased
motion at all other levels. The extent of redistribution of
motion was small, less than 1° of motion in most cases,
but was consistent at all levels. Importantly, these results
suggest that an intervention at one level does not strongly
predispose the immediate super- and sub-adjacent levels
to substantially altered kinematics while sparing other
levels, but rather that all levels are subjected to some
redistribution.
As with all new models, there are associated
strengths and limitations using this technique. This model
allows each individual vertebral segment to be studied
both before and after surgical interventions to identify any
changes in motion at a given level. The motions during
in vitro testing are based on active in vivo motion under
37
voluntary control of the paraspinal muscles. The motion
of the spine in vitro is representative of active physiologic
range of motion. The effects of the rib cage in stabilizing
the spinal column has also been taken into account at least
in part as rib cage was left intact and the vertebrocostal
joints left undisrupted.
While our technique focuses on reproduction of
the three-dimensional kinematics of the thoracolumbar
spine, the protocol does not necessitate application of an
axial preload or follower load to the in vitro specimens33.
The reported advantage of applying a follower load to
specimens in vitro is to stabilize the spine and foster higher
fidelity reproduction of spine motion when the specimen is
tested in unconstrained motion34. Using the in vivo/in vitro
active/passive protocol, it is not clear that application of a
follower load would add to the fidelity of the kinematics
or mechanics of the specimens in vitro.
The novel method described facilitates com-
prehensive evaluation of the entire thoracolumbar spine
for adjacent level effects following surgical intervention.
Preliminary data indicate that motion is redistributed
throughout the spine and not just at the levels immediately
adjacent to the intervention. Future work will include
evaluations of motion sparing and dynamic stabilization
therapies and the effects of an axial preload on spine
kinematics.
 References
1. Park P, Garton JH, Gala VC, Hoff JT, McGillicuddy
JE. Adjacent segment disease after lumbar or lum-
bosacral fusion: review of the literature. Spine. 2004;
29(17):1938-44.
2. Lee CK. Accelerated degeneration of the segment
adjacent to a lumbar fusion. Spine. 1988; 375-77.
3. Disch AC, Schmoelz W, Matziolis G, Schneider SV,
Knop C, Putzier M. Higher risk of adjacent segment
degeneration after floating fusions: long-term outcome
after low lumbar spine fusions. J Spinal Disord Tech.
2008;21:79–85.
4. Lehmann TR, Spratt KF, Tozzi JE, Weinstein JN,
Reinarz SJ, El-Khoury GY, Colby H. Long-term
follow-up of lower lumbar fusion patients. Spine. 1987;
12(2):97-104.
5. Etebar S, Cahill DW. Risk factors for adjacent-
segment failure folllowing lumbar fixation with rigid
instrumentation for degenerative instability. J Neuro-
surg. 1999; 90(2 Suppl):163-9.
6. Aota Y, Kumano K, Hirabayashi S. Postfusion insta-
bility at the adjacent segments after rigid pedical screw
fixation for degenerative lumbar spinal disorders. J
Spinal Disord. 1995; 8(6):464-73.
7. Yang JY, Lee JK, Song HS. The impact of adjacent
segment degeneration on the clinical outcome after
lumbar spinal fusion. Spine. 2008; 33(5):503–7.
8. Van Horn JR, Bohnen LM. Development of dis-
copathy in lumbar discs adjacent to a lumbar anterior
interbody spondylodesis. A retrospective matched-pair
study with a postoperative follow-up of 16 years. Acta
Orthop Belg. 1992; 58(3):280-6.
9. Seitsalo S, Schlenzka D, Poussa M, Osterman K. Disc
degeneration in young patients with isthmic spondylolis-
thesis treated operatively or conservatively: a long-term
follow-up. Eur Spine J. 1997; 6(6):393-7.
38
10. Hambly MF, Wiltse LL, Raghavan N, Schneiderman
G, Koenig C. The transition zone above a lumbosacral
fusion. Spine. 1998; 23(16):1785-92.
11. Axelsson P, Johnsson R, Stomqvist B. Adjacent
segment hypermobility after lumbar spine fusion. No
association with progressive degeneration of the seg-
ment 5 years after surgery. Acta Orthopaedica. 2007;
78(6): 834–839.
12. Lee CK, Langrana NA. Lumbosacral spinal fusion
– a biomechanical study. Spine. 1984; 9(6):574-81.
13. Dekutoski MB, Schendel MJ, Ogilvie JW, Olsewski
JM, Wallace LJ, Lewis JL. Comparison of in vivo and
in vitro adjacent segment motion after lumbar fusion.
Spine. 1994; 19(15):1746-51.
14. Kim YE, Goel VK, Weinstein JN, Lim TH. Effect
of disc degeneration at one level on the adjacent level
in axial mode. Spine. 1991; 16(3):331-5.
15. Weinhoffer SL, Guyer RD, Herbert M, Griffith
SL. Intradiscal pressure measurements above an in-
strumented fusion. A cadaveric study. Spine. 1995;
20(5):526-31.
16. Cunningham BW, Kotani Y, McNulty PS, Cappuc-
cino A, McAfee PC. The effect of spinal destabilization
and instrumentation on lumbar intradiscal pressure.
An in vitro biomechanical analysis. Spine. 1997;
22(22):2655-63.
17. Eck JC, Humphreys SC, Lim TH, Jeong ST, Kim
JG, Hodge SD. An HS. biomechanical study on the
effect of cervical spine fusion on adjacent-level intra-
discal pressure and segmental motion. Spine. 2002;
27(22):2431-34.
18. Rao RD, David KS, Wang M. Biomechanical
changes at adjacent segments following anterior lum-
bar interbody fusion using tapered cages. Spine. 2005;
30(24):2772-6.
19. Schwab JS, DiAngelo DJ, Foley KT. Motion
compensation associated with single-level cervical
fusion: where does the lost motion go? Spine. 2006;
31(21):2439-48.
20. Axelsson P, Johnsson R, Stromqvist B. The spon-
dylolytic vertebra and its adjacent segment. Mobility
measured before and after posterolateral fusion. Spine.
1997;22(4):414-7.
21. Panjabi MM, Abumi K, Duranceau J, Crisco JJ.
Biomechanical evaluation of spinal fixation devices:
II. Stability provided by eight internal fixation devices.
Spine. 1988; 13:1135-40.
22. Panjabi MM, Kato Y, Hoffman H, Cholewicki J,
Krag M. A study of stiffness protocol as exemplified
by testing of a burst fracture model in sagittal plane.
Spine. 2000; 25(21):2748-54.
23. Panjabi MM. Hybrid multidirectional test method
to evaluate spinal adjacent-level effects. Clinical Bio-
mechanics. 2007;22:257-65.
24. Frymoyer JW, Hanley EN, Howe J, Kuhlmann D,
Matteri RE. A comparison of radiographic findings in
fusion and nonfusion patients ten or more years follow-
ing lumbar disc surgery. Spine. 1979; 4(5):435-40.
25. Chow DHK, Luk KDK, Evand JH, Leong JCY.
Effects of short anterior lumbar interbody fusion on
biomechanics of neighboring unfused segments. Spine.
1996; 21(5):549-55.
26. Goel VK, Grauer JN, Patel TC, Biyani A, Sairyo
K, Vishnubhotla Matyas A, Gowgill I, Shaw M, Long
R, Dick D, Panjabi MM, Serhan H. Effects of charite
artifical disc on the implanted and adjacent spinal seg-
ments mechanics using a hybrid testing protocol. Spine.
2005; 30(24):2755-64.
39
27. Cunningham BW, Gordon JD, Dmitriev AE, Hu
N, McAfee PC. Biomechanical evaluation of total disc
replacement arthroplasty: an in vitro human cadaveric
model. Spine. 2003; 28(20S):S110-7.
28. Auerbach JD, Wills BPD, McIntosh TC, Balderston
RA. Evaluation of spinal kinematics following lumbar
total disc replacement and circumferential fusion using
in vivo fluoroscopy. Spine. 2007; 32(5):527-36.
29. Bothmann M, Kast E, Boldt GJ, Oberle J. Dynesys
fixation for lumbar spine degeneration. Neurosurg Rev.
2008; 31:189–196.
30. Shim CS, Lee SH, Shin HD, Kang HS, Choi WC,
Jung B, Choi G, Ahn Y, Lee S, Lee HY. Charite versus
prodisc. A comparative study of a minimum 3-year
follow-up. Spine. 2007; 32(9):1012–18.
31. Meskers CGM, Fraterman H, Van der Helm HCT,
Vermeulen HM, Rozing PM. Calibration of the flock of
birds electromagnetic tracking device and its application
in shoulder motion studies. Journal of Biomechanics.
1999; 32(6):629-33.
32. White AA, Panjabi MM. Clinical biomechanics of
the spine. 1990. Philadelphia:Lippincott.
33. Patwardhan AG; Havey RM; Meade KP, Lee B,
Dunlap B. A follower load increases the load-carrying
capacity of the lumbar spine in compression. Spine.
1999; 24(10):1003-9.
34. Panjabi MM, Krag MH, White AA, Southwick WO.
Effects of preload on load displacement curves of the
lumbar spine. Orthopaedic Clinics of North America.
1977 January; 8(1):181-192.
 The Role of Extensor Muscle Fatigue in the Onset of Intervertebral Disc Degeneration:
An In Vivo Model
Mary Beth M. Grabowsky, MS1, Nicholas A. Pallotta, MS1, Matthew W. Connelly, MS1, Brett Van Etten2,
Rebecca A. MacDonald, MS1, Maxwell Alley4, Joseph C. Glennon, VMD3, Andrew H. Dubin, MD4,
John W. German, MD4, Richard L. Uhl, MD4, Eric H. Ledet, Ph.D.1
1
Rensselaer Polytechnic Institute, Troy, NY
2
University of Vermont, Burlington, VT
3
Veterinary Medical Specialties, Pattersonville, NY
4
Albany Medical College, Albany, NY

Epidemiologic evidence indicates a correlation between occupation and the development of low back pain (LBP) and
degenerative disc disease (DDD). Specific occupational activities associated with LBP include poor posture, whole
body vibration, and repetitive lifting. These activities have a common link: they result in fatigue of the primary spi-
nal extensor musculature supporting the spine. This fatigue has been shown to cause recruitment of both secondary
agonist, as well as antagonist muscles to maintain support of the spine. However, this fatigue and accompanying
recruitment may lead to increased intervertebral loading – a stimulus for disc degeneration. This association is a vital
connection between everyday physical activities and LBP. However, this link between muscle fatigue and increased
load across the disc space has never been quantified in vivo. The purpose of this study is two-fold: 1) to develop
and test a wireless force-sensing interbody implant, and 2) to create an in vivo large animal model to study primary
extensor muscle fatigue. Combined, these tools allow in vivo spinal forces to be measured prior to and following
muscle fatigue to determine alterations in loading due to fatigue.

Introduction: tigue over time 9,10,11,12. Similarly, whole-body vibration

It is estimated that up to 80% of the general population will
experience at least one significant bout of low back pain
(LBP) in their lifetime1,2. LBP had the highest prevalence
of any reported medical condition in 2005, with up to 40%
of the United States adult population stating that they had
experienced low back pain in the previous 3 months. This
is more than two times the incidence of other common
maladies including pain in other joints such as the hip,
knee, or shoulder (8%, 15%, and 12% respectively)3. The
leading known cause of low back pain is degenerative disc
disease (DDD) 3,4,5.
Over the past decade, mechanical, genetic, and
psychosocial factors have been linked to the onset of
disc degeneration, and it has been further shown that
these risk factors are cumulative6,7,8. The majority of the
well-established risk factors for LBP are mechanical in
nature, and are often related to activities of daily living
and everyday work activities such as poor postural control,
experiencing whole-body vibrations, and frequent lifting.
Despite these correlations, the underlying mechanism by
which DDD and LBP develop remains unclear.
One common link between these activities is
that they can cause prolonged submaximal muscle activ-
ity in the primary spinal extensor muscles. Intuitively,
muscle fatigue may play a role in the development of
DDD and LBP. Poor posture forces the extensor muscles
to work harder to support the spine, and thus causes fa-
40
causes instability of the spine as well as fatigue in the
spinal musculature, especially at its resonant frequency
of 5 Hz.13,14,15,16. Additionally, frequent and heavy lifting,
in itself has been shown to cause fatigue in the primary
spinal extensors17,18.
Primary extensor muscle fatigue initiates recruit-
ment of secondary agonistic muscles to help support the
spine19,20,21,22,23. For example, when the erector spinae
complex is fatigued, it will recruit the multifidus muscles.
However, unlike the primary extensor muscles, these
secondary muscles are normally much shorter and less
optimized for supporting the spine. In addition to these
secondary muscles, the body will also recruit antagonistic
muscles to prevent spine instability. Although normally
categorized as a primary spinal flexor, the activity of the
rectus abdominus muscles have been shown to increase
following erector spinae fatigue24. Appropriately, the cor-
relation between patients with low back pain and an in-
creased recruitment of agonistic and antagonistic muscles
has also been demonstrated25,26,27.
Primary extensor muscle fatigue causes altered
spinal recruitment patterns. The link between altered
muscle activity and spinal degeneration may be the result
of a net increase in force across the disc space28,29. How-
ever, this has not yet been proven experimentally.
Several types of these mathematical models
have been developed and the process of estimating spinal
loads has become increasingly complex through the use of
multiple technologies, such as electromyography (EMG)
and finite element analysis (FEA)30,31,32. However, these
techniques have not been verified by direct in vivo data.
Limited in vivo studies using instrumented inter-
body implants and posterior implants have been conducted
to monitor loads in the spine33,34. While these instrumented
implants have collected data which provide insight into
the changing forces as a function of activities, they have
been limited because they are designed around fusion
devices. These implants inhibit the natural motion at the
level of interest and therefore cause an artifact in the data.
As an alternative, a multi-axial articulating force-sensing
implant overcomes this limitation. Furthermore, current
research with these force-sensing implants has been lim-
ited to brief sessions recording load data during normal
activities, such as sitting and standing, while changes in
disc space forces as a function of primary muscle fatigue
have never before been measured. The purpose of this
study was: 1) to develop and test a wireless force-sensing
interbody implant, and 2) to create an in vivo large animal
model to study primary extensor muscle fatigue. Using
the combination of these tools, changes in multi-axial
forces in the disc space can be measured in vivo prior to
and following muscle fatigue, thereby providing data on
an incompletely understood step in the disc degeneration
process.

Materials and Methods

Implant Design
We developed a novel wireless force-sensing implant simi-
lar to many current total disc replacements. The implant
has distinct superior and inferior articulating components
to allow for vertebral motion and the collection of multi-
axial force data through wireless sensors, as shown in
Figure 1. During articulation of adjacent vertebrae in
flexion, extension, lateral bending, or a combination of
these movements, the superior concave component of
the implant articulates on the three force transfer struts
arranged convexly in the inferior component assembly.
Embedded below the force transfer struts are three capaci-
tive wireless sensors. The orientation and geometry of the
struts ensures that they remain normal to the contacting
surface of the superior component during articulation. A
single strut is centered anteriorly, while the remaining two
are located posterolaterally, equidistant from the centerline
and each 110 degrees from the anterior strut in the axial
plane. These struts allow the entire load seen in the anterior
and middle columns of the spine to be transferred to the
three sensors in a unique distribution dependent upon the
positioning of the adjacent vertebrae.
The distribution of forces measured by the three
sensors facilitates calculation of the multi-axial forces be-
ing applied to the implant, including the axial compressive
41
Figure 1 – (Above) An exploded CAD view of the
multi-axial implant depicts the superior concave compo-
nent which articulates on the inferior convex assembly.
(Below) A prototype of the implant shown in lateral
bending.
force, the anterior/posterior shear force, and the left/right
lateral shear force. An analytical model was developed
wherein once the sensors are experimentally calibrated
such that a voltage-force relationship is determined, the
three forces measured by the sensors can be used, along
with their constant orientation parameters, to determine
the three components of the load being applied as is shown
in Figure 2.
When taking readings from the implant, each
sensor’s force is used in combination with the following
equations to calculate the axial compressive and shear
forces.
(Equation 1)
(Equation 2)
(Equation 3)
Figure 2 – The three wireless sensors are arranged in an array within the inferior component such that their force
transfers each always apply a fraction of the total force along their respective vectors. These constant parameters aid
the calculation of the compressive and lateral shear forces.

Implant Experimental Model
To experimentally validate our implant design, a func-
tional, scaled model of the implant was created into which
three subminiature button load cells were integrated. The
model was assessed by applying both axial and eccentric
loads up to 445 N using a custom-built seven degree of
freedom spine testing machine, as shown in Figure 3 be-
low. The forces measured by the button load cells were
used to back-calculate the applied load via the analytical
model (Equations 1-3). The calculated values were then
compared to the actual applied load.
Implant Computational Model and Finite Element
Analysis
Validation of the implant design was also achieved using
finite element analyses (FEA). A computer generated
model of the implant was created using SolidWorks (Das-
sault Systèmes SolidWorks Corporation, Concord, MA)
and analyzed using CosmosWorks (Dassault Systèmes
SolidWorks Corporation, Concord, MA). This analysis
included loading the implant both axially, as well as in
flexion, lateral bending, and combinations thereof. The
loads transferred to the sensors were again used to cal-
culate the applied loads, and these computed values were
compared to the actual applied loads.
in this study. To develop the fatigue model, the goats
were anesthetized and needle electrical stimulation was
employed bilaterally to repeatedly activate the splenius.
The goats were placed in a sling such that its body was
supported in a prone position with legs suspended, as
shown in Figure 4. Two cables were attached to either
side of a halter and were extended downward to attach in
series with a miniature load cell. Needle electrodes were
placed bilaterally in the splenius to stimulate it while
minimizing co-contraction of adjacent muscle groups.
During stimulation of the splenius, the load cell-halter
system was used to monitor and record the head extensile
force elicited by the stimulation. These data were used to
determine the extent of fatigue.
Results
Implant Experimental Model
Results of the experimental model of our articulating
multi-axial implant demonstrated excellent agreement

Fatigue Animal Model

The goat cervical spine is a well established analog to
the human lumbar spine because it has a similar structure
and loading pattern and is oriented vertically35,36. The
splenius is the most superficial of the intrinsic muscles in
the goat cervical spine and is primarily responsible for
extension of the neck. It is an analog to the human erector
spinae complex which is primarily responsible for trunk
extension in the lumbar spine. Figure 3 – An experimental model of the wireless
Following IACUC approval, four skeletally multi-axial implant is tested in a custom seven degree of
mature male Alpine-Nubian cross-bred goats were used freedom spine testing machine.

42
Figure 4 – A diagram of the goat during fatigue testing illustrates the head halter which is in series with the load cell
(a) and the needle electrodes, a close-up of which is shown in (b).

between the calculated and actual applied loads. The
implant was first loaded in neutral compression with a
maximum load of 445 N. With six trials completed in this
orientation, the mean axial, A/P shear, and lateral shear
forces calculated using the button load cell results and the
analytical model had errors less than ±22 N (or 5% error),
when compared to their respective applied loads. When
subjected to eccentric loading, the mean error in measured
loads was also less than ±22 N.
Implant Computation Model and Finite Element
Analysis (FEA)
Table 1 shows the results of the finite element analysis.
The implant underwent loading in the neutral position
via an applied axial load and in the flexed and laterally
bent positions via off-axis loads. All of the loads were
calculated within 0.3 % error of the applied loads and the
angles within 0.2 degrees.
Fatigue Animal Model
The optimal electrical stimulation parameters were de-
termined iteratively and consisted of a frequency of 1
Hz with a 200 μs duration and an amplitude of 300 volts.
This combination provided the low frequency stimulation
desired to mimic low frequency fatigue in humans, while
also providing a duration and amplitude with enough force
to cause a visible movement of the head with each stimula-
tion, but not cause co-contraction of nearby muscles. The
elapsed time between the onset of stimulation (baseline
reading) and muscle exhaustion was approximately 2
hours. The initial baseline reading illustrated distinct peaks
with consistent magnitudes averaging approximately 37
N. The first sign of fatigue detected was variability in the
force magnitude of each peak, and an increasing standard
deviation within each 10 second segment was observed
as the test progressed. The decrease in average amplitude
over the 2 hours was linear with a correlation coefficient
of 0.81, as shown in Figure 5. The last data set before a
ten minute pause at the end of the study exhibited only
four of the ten expected response peaks with an overall
average magnitude for the set just over 5 N. After the ten
minute pause at the end of the study, the first ten second
data set demonstrated that the mean peak amplitude was
still indistinguishable from noise and that the muscles had
not recovered from the fatigue.

Table 1 - Calculated forces and angles which resulted from finite element analysis and the analytical model were
determined within 0.3% error.
Degrees
Applied Sensor 1 Sensor 2 Sensor 3 Calculated Calculated Percent
Position from
Load (N) (N) (N) (N) Angle (°) Force (N) Error (%)
Vertical
150 Flexion 2 33.59 59.89 59.81 1.82 150.30 0.200
150 Flexion 4 16.06 68.56 68.46 3.98 150.37 0.246
150 Lateral Bending 2 51.06 66.35 35.96 2.00 150.30 0.200
150 Lateral Bending 4 50.98 81.38 20.74 3.99 150.39 0.260
267 Neutral 0 90.98 91.12 90.82 5.80x10-03 267.60 0.200
43
Figure 5 – A plot showing the relationship between the decrease in force response to stimulation and the elapsed
time of the stimulation reveals a linear relationship.
Discussion
We have designed a wireless, articulating, multi-axial
load cell implant. Both scaled functional models and
computational models have been tested for this implant,
and the results of this study demonstrate that our implant
is sufficiently sensitive to measure axial and shear forces.
Initial muscle fatigue testing showed that the use of elec-
trical stimulation was effective in isolating the primary
extensor muscles of the caprine cervical spine and creat-
ing low frequency fatigue. In addition, we showed that
using electrical stimulation at the specified parameters to
stimulate the caprine cervical primary extensors can result
in a 100% decrease in force over a 2 hour time period.
In this study, the novelty of the implant is two-
fold: the implant-sensor design itself and the unique data
it will provide. The multi-axial force sensing implant
allows for the measurement of loads directly imposed
on the vertebral bodies and discs in real-time throughout
a wide range of motion. In this way, the data will more
closely reflect actual in vivo loading relative to previous
estimates. In addition, much of the current data available
on spinal loading focuses on axial compressive forces.
Although the compressive force is the major component
of spinal loading, A/P and lateral shear forces are also
present in the disc space and may play a significant role
in the development of degenerative disc disease. While
44
the concentric layers of the anulus fibrosis provide excel-
lent support with respect to axial forces, the orientation of
the collagen fibers within the anulus fibrosis provide no
ability to resist shear forces.37 Furthermore, there is an
increased risk of disc injury when compressive forces are
combined with shear forces in activities such as flexion
and lateral bending.38 Therefore, an increase in A/P and
lateral shear forces due to fatigue could be significant in
the development of degenerative disc disease.
There are currently no experimental data which
relate fatigue in the primary spinal extensor musculature
to the magnitude of force across the intervertebral disc.
If muscle fatigue does in fact increase forces across the
disc, it is a potential initiator for degenerative disc disease,
as increased loading is a known precursor to low back
pain. We predict that increased muscle recruitment is the
primary mechanism through which fatigue in the primary
spinal extensors causes an increase in interbody forces.
There is an increased recruitment of secondary agonistic
muscles upon fatigue 39,40,41,42,43. However, the secondary
agonists in the back support one or at most a few levels
within the spine. Thus, the insertion and origins of these
muscles are not as optimized as the primary extensors
for supporting large structures. Their recruitment as an
adjunct to support the spine once the primary extensors
fatigue may be causing more harm than good. In addition,
although counterintuitive, there are several examples in
the body other than the spine where fatigue in a primary
agonistic muscle results in recruitment of antagonistic
muscles for support. This antagonistic recruitment may
prevent injury in the short term by stabilizing the joint,
but like the secondary agonists, the antagonists provide
suboptimal support which may further increase the joint
loading.
Epidemiologic data indicate correlations be-
tween activities which cause fatigue in the primary spinal
References
1. Walker BF. The prevalence of low back pain: a
systematic review of the literature from 1966 to
1998. Journal of Spinal Disorders & Techniques.
2000;13(3):205-217.
2. Andersson GBJ. Epidemiological features of chronic
low-back pain. Lancet. 1999;354:581-585.
3. American Academy of Orthopaedic Surgeons and the
United States Bone and Joint Decade. The Burden of
Musculoskeletal Diseases in the United States. Rose-
mont, IL: American Academy of Orthopaedic Surgeons.
2008;21-33.
4. Schwarzer AC, Aprill CN, Derby R, Fortin J, Kine
G, Bogduk N. The relative contributions of the disc
and zygapophyseal joint in chronic low back pain.
Spine.1994;19:801-806.
5. Jones KC, Murray MM. Meniscus and intervertebral
disks. From: Bernstein J (Ed.). Musculoskeletal Medi-
cine. Rosemont, IL: American Academy of Orthopaedic
Surgeons. 2003;41-49.
6. Rubin DI. Epidemiology and risk factors for spine
pain. Neurologic Clinics. 2007;25 353-371.
7. Battié MC, Videman T. Lumbar disc degeneration:
epidemiology and genetics. The Journal of Bone and
Joint Surgery. 2006;88-A(Supplement 2): 3-9.
8. Lis AM, et al. Association between sitting and oc-
cupational LBP. European Spine Journal. 2007;6:283–
298.
45
extensors and the onset of low back pain. The overlying
process through which degenerative disc disease develops
is multi-faceted and not well understood. Therefore, a
better understanding of this link can help to prevent the
onset of disc degeneration, treat patients, and elucidate
the mechanism through which everyday activities cause
low back pain.
9. Adams MA, Hutton WC. The effect of posture on the
role of the apohysial joints in resisting intervertebral
compressive forces. British Journal of Bone and Joint
Surgery. 1980;62-B(3):358-362.
10. Schultz AB, Warwick DN, Berkson MH, Nachemson
AL. Mechanical properties of human lumbar motion
spine segments - Part 1: Responses in flexion, extension,
lateral bending and torsion. Journal of Biomechanical
Engineering. 1979;101:46-52.
11. Dunlop RB, Adams MA, Hutton WC. Disc narrow-
ing and the lumbar facet joints. British Journal of Bone
and Joint Surgery. 1984;66(5):706-710.
12. Adams MA, Hutton WC. The effect of posture on the
lumbar spine. British Journal of Bone and Joint Surgery.
1985;67-B(4):625-629.
13. Seidel H, et al. On human response to prolonged
repeated whole-body vibration. Ergonomics. 1980;
23(3):191-211.
14. Seroussi RE, Wider DG, Pope MH. Trunk muscle
electromyography and whole body vibration. Journal
of Biomechanics. 1989;2(3): 219-229.
15. Hansson T, Magnusson M, Broman H. Back
muscle fatigue and seated whole body vibrations: an
experimental study in man. Clinical Biomechanics.
1991;6(3):173-178.
16. Pope MH, Magnusson M, Wilder DG. Low back
pain and whole body vibration. Clinical Orthopaedics
and Related Research. 1998;354:241-248.
17. Bonato P, Ebenbichler GR, Roy SH, Lehr S, Posch
M, Kollmitzer J, et al. Muscle fatigue and fatigue-related
biomechanical changes during a cyclic lifting task.
Spine. 2003;28:1810–20.
18. Kim SH, Chung MK. Effects of posture, weight and
frequency on trunk muscular activity and fatigue during
repetitive lifting tasks. Ergonomics. 1995;38(5):853-
863.
19. Sparto PJ, Parnianpour M, Marras WS, Granata KP,
Reinsel TE, Simon S. Neuromuscular trunk performance
and spinal loading during a fatiguing isometric trunk
extension with varying torque requirements. Journal of
Spinal Disorders. 1997;10(2) 145-156.
20. O’Brien PR, Potvin JR. Fatigue-related EMG re-
sponses of trunk muscles to a prolonged, isometric twist
exertion. Clinical Biomechanics. 1997;12(5):306-313.
21. Novak GJ, Schipplein OD, Trafimow JH, Anders-
son GBJ. Influence of erector spinae muscle fatigue
on the lumbo-sacral moment during lifting. European
Journal of Experimental Musculoskeletal Research.
1993;2(1):39-44.
22. Marras WS, Granata KP. Changes in trunk dynam-
ics and spine loading during repeated trunk exertions.
Spine.1997;22(21):2564-2570.
23. Marras WS, Parakkat J, Chany AM, Yang G, Burr
D, Lavender SA. Spine loading as a function of lift
frequency, exposure duration, and work experience.
Clinical Biomechanics. 2006;21:345-352.
24. Gardner-Morse MG, Stokes IAF. The effects of
abdominal muscle coactivation on lumbar spine stabil-
ity. Spine. 1998;23(1):86-91.
25. van Dieën JH, Cholewicki J, Radebold A. Trunk
muscle recruitment patterns in patients with low back
pain enhance the stability of the lumbar spine. Spine.
2003;28(8):834-841.
46
26. Ferguson SA, Marras WS, Burr DL, Davis KG,
Gupta P. Differences in motor recruitment and resulting
kinematics between low back pain patients and asymp-
tomatic participants during lifting exertions. Clinical
Biomechanics. 2004;19:992-999.
27. Silfies SP, Squillante D, Maurer P, Westcott S, Kar-
duna AR. Trunk muscle recruitment patterns in specific
chronic low back pain populations. Clinical Biomechan-
ics. 2005;20(5):465-73.
28. Granata KP, Marras WS. The influence of trunk
muscle coactivity on dynamic spinal loads. Spine.
1995;20:913-919.
29. Song YW, Chung MK. Quantitative assessment of
trunk muscle coactivation in sub-maximal isometric
exertion tasks. International Journal of Industrial Er-
gonomics. 2004;34:13-20.
30. Cholewicki J, McGill SM, Norman RW. Lumbar
spine loads during the lifting of extremely heavy
weights. Medicine and Science in Sports and Exercise.
1991;23(10):1179-86.
31. Arjmand N, Shirazi-Adl A. Biomechanics of
changes in lumbar posture in static lifting. Spine. 2005;
30(23):2637-2648.
32. Granata KP, Marras WS. An EMG-assisted model
of trunk loading during free-dynamic lifting. Journal of
Biomechanics. 1995;28(11):1309-1317.
33. Rohlmann A, Bergmann G, Graichen F, Mayer HM.
Telemetrized load measurement using instrumented
spinal internal fixators in a patient with degenerative
instability. Spine. 1995;20(24): 2683-2689.
34. Ledet EH, Tymeson MP, DiRisio DJ, Cohen B, Uhl
RL. Direct real-time measurement of in vivo forces in
the lumbar spine. The Spine Journal. 2005;5:85-94.
35. Zdeblick TA, et al. Anterior cervical discectomy and
fusion. A comparison of techniques in an animal model.
Spine. 1992;17(10 Suppl): S418–S426.
36. Agazzi S, et al. Refinement of interbody implant test-
ing in goats: a surgical and morphometric rationale for
selection of a cervical level. Journal of Neurosurgery:
Spine. 2007;7:549-553.
37. Gupta R, Caiozzo V, Cook SD, Barrack RL, Skinner
HB. Basic science in orthopaedic surgery. From: Skinner
HB (Ed.). Current Diagnosis & Treatment in Orthopae-
dics. New York: McGraw-Hill. 2003; 1-59.
38. Adams MA, Freeman BJ, Morrison HP, Nelson IW,
Dolan P. Mechanical initiation of intervertebral disc
degeneration. Spine. 2000;25(13):1625–1636.
39. Sparto PJ, Parnianpour M, Marras WS, Granata KP,
Reinsel TE, Simon S. Neuromuscular trunk performance
and spinal loading during a fatiguing isometric trunk
extension with varying torque requirements. Journal of
Spinal Disorders. 1997;10(2):145-156.
47
40. O’Brien PR, Potvin JR. Fatigue-related EMG re-
sponses of trunk muscles to a prolonged, isometric twist
exertion. Clinical Biomechanics. 1997;12(5):306-313.
41. Novak GJ, Schipplein OD, Trafimow JH, Anders-
son GBJ. Influence of erector spinae muscle fatigue
on the lumbo-sacral moment during lifting. European
Journal of Experimental Musculoskeletal Research.
1993;2(1):39-44.
42. Marras WS, Granata KP. Changes in trunk dynam-
ics and spine loading during repeated trunk exertions.
Spine. 1997;22(21):2564-2570.
43. Marras WS, Parakkat J, Chany AM, Yang G, Burr
D, Lavender SA. Spine loading as a function of lift
frequency, exposure duration, and work experience.
Clinical Biomechanics. 2006;21:345-352.
 In Vivo Effects of Cyclic Compression on the Intervertebral Disc and Endplates
of the Lumbar Spine: A Pilot Study With a Novel Animal Model
Glenn P. Sanders, Ph.D1, Sarah E. Linley1, Gwendolyn A. Sowa, MD, Ph.D2, Joseph C. Glennon,VMD3,
Darryl J. DiRisio, MD4, James P. Lawrence, MD4, Jacqueline Vanderzanden, MD4, Elizabeth Leimer 1,
Jeffrey C. Lotz, Ph.D5, Eric H. Ledet, Ph.D1

Rensselaer Polytechnic Institute, Troy, NY

1

2
University of Pittsburgh, Pittsburgh, PA
3
Veterinary Medical Specialties, Pattersonville, NY
4
Albany Medical College, Albany, NY
5
University of California San Francisco, San Francisco, CA

Background: Epidemiologic data indicate a correlation between repeated mechanical loading of the lumbar
spine and painful degenerative intervertebral disc disease. Clinically, sclerotic changes of the vertebral endplates
and subchondral bone are often observed concomitant to intervertebral disc degeneration, however the temporal
relationship between endplate sclerosis and disc degeneration is not well characterized. The lumbar spine of the
New Zealand white rabbit is a promising model to study the response of the lumbar spine to dynamic mechanical
loading. Unlike the mouse and rat tail models however, no cyclic loading model exists for the rabbit lumbar spine.

Purpose: The purpose of this study was to characterize the correlation between cyclic axial loading and the initia-
tion and propagation of (a) intervertebral disc degeneration and (b) endplate sclerosis in our model.

Methods: Following surgical placement of transfixing pins and dorsally extending percutaneous posts, the lumbar
spines of 11 New Zealand white rabbits were loaded cyclically via an external apparatus for two hours per day,
five days per week. Loads were applied at 5 x BW at either 0.5 Hz for 10 weeks (n=4), 0.5 Hz for 26 weeks (n=2),
5 Hz for 8 weeks (n=2), or 5 Hz for 30 weeks (n=2). One animal (n=1) was designated as a surgical control animal
and underwent the identical surgical procedure as the other animals but received no daily loading. Serial plain
radiographs were used to monitor changes in the spine. Spines were harvested and diffusion and disc health were
analyzed using post-contrast enhanced MRI imaging. Endplate thickness and bone density were quantified via
microCT analysis. Disc and endplate changes were characterized by gene expression analysis and histopathology.

Results: Results indicate that the disc, endplates, and subchondral bone degenerate in response to cyclic mechani-
cal loading. Endplate cartilage thins and calcifies, resulting in thickened sclerotic subchondral bone. Diffusion to
the disc is reduced proportionally to the extent of sclerosis. The disc becomes fibrotic and disorganized, the extent
of which is proportional to load exposure. The observed trends in gene expression analysis suggest that the slower
accumulation of load cycles in the low frequency group may have a beneficial effect on matrix homeostasis and
inflammation in our model, whereas the rapid accumulation of cycles in our high frequency group may stimulate
matrix synthesis accompanied by pro-inflammatory and catabolic activity.

Conclusions: We have developed an apparatus, a technique, and an animal model to study the long-term effects of
mechanical loading on the intervertebral disc, endplates, and subchondral bone in the lumbar spine in vivo. Physi-
ologic mechanical loading alone will initiate degenerative changes in the intervertebral disc and sclerotic changes
in the endplates accompanied by a reduction in passive diffusion from the subchondral bone to the disc. Rapid
accumulation of cycles stimulated matrix synthesis accompanied by pro-inflammatory and catabolic activity,
whereas the slower accumulation of cycles may have a beneficial effect on matrix homeostasis and inflammation
in our model. Results indicate that mechanisms affecting subchondral bone density also affect disc diffusion and
homeostasis.

Introduction chanical loading of the lumbar spine and painful degenera-

Low back pain is a leading cause of disability, lost wages,
and physician office visits.1-3 The causes of low back pain
are complex and not fully understood, but epidemiologic
data indicate a strong correlation between repeated me-
48
tive intervertebral disc disease.4-8
Clinically, sclerotic changes of the vertebral
endplates and subchondral bone are often observed con-
comitant to intervertebral disc degeneration. However, the
temporal relationship between endplate sclerosis and disc
degeneration is not well characterized. The endplates and
subchondral bone play a critical role in intervertebral disc
homeostasis via regulation of small molecule diffusion to
and from the avascular disc.9-12 There is evidence to sug-
gest that bony endplate sclerosis occurs secondary to disc
degeneration as a result of altered loading distributions
relative to a healthy disc.13 Yet, there is also evidence
indicating that the temporal sequence is reversed, and
that sclerotic endplates initiate disc degeneration via a
reduction in diffusion to and from the disc.14
In diarthrodial joints, subchondral bone remod-
els in response to mechanical loading.15 Increasing bone
density and reduction in porosity correlate to exposure to
cyclic loading. These changes have been indicated as a
possible initiator of osteoarthritis via a reduction in diffu-
sion to the articular cartilage.16 Changes in vertebral sub-
chondral bone are common,17 but the correlation to cyclic
mechanical spinal loading has received little attention.
The response of the intervertebral disc (both anu-
lus fibrosis and nucleus pulposus) to repeated mechanical
loading has been studied previously, primarily through uti-
lization of the rat caudal (tail) vertebrae model.18-21 Using
this in vivo model, the metabolic response of the disc to
loads of varying magnitude, frequency, and duration have
been described. The rat tail model is unique in that it allows
for study of the gradual onset of chronic disc degeneration
resulting from mechanical loading, whereas other com-
mon animal models utilize iatrogenic anular disruption
or chemical alteration to initiate disc degeneration.22-24
While the rat tail model has yielded valuable data for
characterizing the response of the disc to loading, the
effect of loading on the endplates and subchondral bone
have not been studied in this model. Rat vertebral physes
remain open throughout adulthood25,26 and caudal motion
segments lack posterior elements and zygapophyseal
joints, potential confounding factors for elucidating the
critical role of the endplates and subchondral bone in the
degenerative process.
The lumbar spine of the New Zealand white
(NZW) rabbit is a promising alternative model. The
vertebral physes close at skeletal maturity and there re-
main only sparse cells of the notochord phenotype in the
disc.27-29 With respect to dynamic loading, the resonant
frequency of the rabbit spine is 4.5 Hz, similar to that of
the human spine.30,31 Pressures in the human intervertebral
disc have been measured at 2.6 MPa during activities of
daily living at forces that correspond to approximately
6 times body-weight.32,33 Based on the morphology and
cross-sectional area of the skeletally mature NZW rabbit
lumbar intervertebral disc, an axial load of approximately
6 times body weight corresponds to a stress of approxi-
mately 2.6 MPa in the disc.34,35 These properties and the
49
larger physical size of the rabbit lumbar spine relative to
the rat and mouse suggest its utility as a model to study
the response of dynamic mechanical loading. Unlike the
mouse and rat tail models however, no cyclic loading
model exists for the rabbit lumbar spine.
To comprehensively characterize the complex
interplay between the intervertebral disc, vertebral end-
plates, and subchondral bone in response to repeated
mechanical loading, we have developed a novel in vivo
animal model. The purpose of this study was to character-
ize the correlation between cyclic axial loading and the
initiation and propagation of intervertebral disc degenera-
tion and endplate degeneration in our model.
Materials and Methods
We have developed an external apparatus for applying
chronic, cyclic, compressive, axial-only loads to the
lumbar spine of a live animal. The loading apparatus
(“The Degenerator”) is a novel portable external ap-
paratus capable of applying axial compressive forces at
controlled magnitudes and frequencies. The Degenerator
interfaces with the spine to apply loads by engaging four
percutaneous posts that extend dorsally from the vertebral
bodies. During loading, The Degenerator is attached to the
posts, and following loading, it is disengaged. The four
percutaneous posts are attached to two adjacent lumbar
vertebrae via two transfixing pins that extend bilaterally
from the vertebral bodies (one per vertebra). During load-
ing, the animals are placed into the apparatus awake, and
unrestrained.
In this study, we utilized 11 skeletally mature
(>4.5 kg) New Zealand White (NZW) rabbits to test our
apparatus and to characterize the effects of cyclic loading
on the intervertebral disc and endplates.
Apparatus
The Degenerator utilizes a single linear actuator (Mirai
Inter-Technologies, Rochester, NY) which interfaces with
a personal computer and moves under closed-loop control.
Through a series of bearings, the motion of the actuator is
transferred to a clamping mechanism which secures the
Degenerator to the percutaneous posts (and thus the spine).
By aligning the actuator with the long axis of the spine, it
applies an axial-only load (no bending or torsion) to the
spine even if the percutaneous posts are not perpendicular
to the spine or parallel to each other. The Degenerator al-
lows for control of load frequency, duration, magnitude,
and displacement. The apparatus is mounted to a cart and
extends below to the top of a rabbit restrainer system (Plas
Labs, Lansing, MI) mounted directly underneath. The ani-
mal is placed in a restrainer and the clamping mechanism
is attached to the percutaneous posts.
Figure 1: Transfixing pins are used to attach stainless steel posts to adjacent vertebrae (left). The posts extend percu-
taneously to attach to the external loading apparatus (right).
Apparatus-Spine Interface
For interfacing the loading system with the lumbar spine,
we chose to modify the technique described by Kroeber et
al42 where transfixing pins are placed transversely through
adjacent vertebral bodies and attached to dorsally extend-
ing percutaneous posts. After IACUC approval, using
general anesthesia standard aseptic techniques, sharp and
blunt dissection were used to expose the lateral aspect
of the lumbar spine via a single 10 cm lateral incision to
facilitate placement of the transfixing pins. The anatomic
landmark for caudal pin insertion is 3 mm from the caudal
endplate of the caudal vertebral body at the interface of
the transverse process and vertebral body. The landmark
on the cranial vertebra is approximately 3 mm from the
cranial endplate and 1 mm ventral to the ridge where the
transverse process transitions into the vertebral body, as
shown in Figure 1.
A 1 mm k-wire was drilled through the vertebral
body at each location. The k-wires were then removed
and a 2 mm transfixing pin with 2.5 mm central threaded
region (IMEX Veterinary, Inc., Longview, TX) was placed
into the pre-drilled hole. The threaded region of each pin
was coated with nanophase hydroxyapatite (Spire, Corp.,
Bedford, MA) to facilitate rapid osseointegration.
Four stab incisions were then made dorsal to the
spine to facilitate placement of each post. Incisions were
made just cranial to the L5-6 facet joints and just caudal to
the L3-4 facet joints. The 6 mm stainless steel posts were
placed blindly through the two left incisions and guided
onto the left lateral aspect of the caudal and cranial trans-
fixing pins. The right side of the transfixing pins were then
trimmed using pin cutters through the lateral incision and
the right posts were placed percutaneously onto the pins
through the two right stab incisions. The lateral incision
was then closed in layers. Sutures were used as necessary
to close the dorsal incisions around the posts. The posts
were cleaned daily with betadine and antibiotic ointment
for at least two weeks following surgery. Animals were
allowed a four week quiescent period following surgery to
50
allow for bony ingrowth into the pins during which their
acclimation to the apparatuses was continued.
Loading Regimen
One animal (n=1) was designated as a surgical control
animal and underwent the identical surgical procedure
as the other animals but received no daily loading. At ap-
proximately 4 weeks post-operatively, the other animals
began their daily loading regimen for 2 hours, five days
per week. Animals were loaded cyclically from no load to
220 N at either 0.5 Hz for 10 weeks (n=4), 0.5 Hz for 26
weeks (n=2), 5 Hz for 8 weeks (n=2), or 5 Hz for 30 weeks
(n=2). A 220 N load is approximately 5 x BW for the 4.5 kg
animals. This is within the range of estimated physiologic
loading in the lumbar spine of humans during activities
of daily living36-38 and similar to force levels used in other
models.39-43 Based on our previous work measuring the
morphometric properties of the rabbit intervertebral disc,
the mean cross-sectional area of the disc is approximately
95 mm2 resulting in a disc space stress of approximately
2.3 MPa.34,35 Five times body weight applied to a typical
human lumbar intervertebral disc of cross-sectional area
1,800 mm2 results in a stress of approximately 2 MPa.44
This is consistent with intradiscal pressure measure-
ments taken from human discs during lifting (2.3 MPa)
and sitting while holding a weight (2.6 MPa).33,45 Based
on mean stiffness of the NZW lumbar motion segment,
application of 220 N results in compression of the disc
of approximately 25%. By comparison, at 5.7 X BW,
the human disc space narrows by approximately 25%.46
Loading frequencies were chosen to mimic exposure in
occupations which correlate to disc degeneration such as
truck drivers, operators of heavy machinery (5 Hz)47-49 and
manual laborers (0.5 Hz).
Serial Plain Radiographs
Disc space narrowing, osteophyte formation, and endplate
sclerosis are clinical indicators of disc degeneration. Dur-
ing the time course of loading, plain x-rays were used
to measure these parameters. Lateral plain radiographs Micro CT Analysis
were taken every two weeks and digitized for analysis. Osteophytes and subchondral sclerosis are features com-
Standardized exposure times and intensities (64 kVp 2 mon to disc degeneration. Subchondral bone volume and
MA) were used. Disc Height Index (DHI)50 was measured bone mineral content were quantified to assess ectopic
at each time point and changes in loaded levels were mineralization and subchondral sclerosis. A μCT analy-
compared to unloaded adjacent levels. The presence or sis (SCANCO, vivaCT 40, Bassersdorf, Switzerland)
absence of osteophytes and changes in lucency of the was performed on loaded and adjacent unloaded levels.
endplates were graded by a single blinded investigator. Following reconstruction of 19 μm sagittal sections, a
Time-point radiographs were compared to baseline films volume-of-interest (VOI) was established which included
to assess changes. a 2 mm deep region of subchondral bone at the central 1/3
of the endplate as shown in Figure 2. VOIs were created
Administration of Contrast Agent/Euthanasia for both the cranial and caudal endplates. Bone density of
Forty-five minutes prior to euthanizing each animal the VOI was quantified at the loaded level and compared
at its designated time point, gadodiamide (Omniscan, to the adjacent unloaded levels. Mid-sagittal slices were
Amersham Health, Princeton, NJ) was administered (0.3 also used to measure endplate thickness and intervertebral
ml/kg BW, IV) to assess diffusion into the intervertebral disc height. Values from loaded levels were compared to
disc. Forty-five minutes after administration of the ga- non-loaded adjacent levels and aged matched controls.
dodiamide, animals were euthanized (Pentobarbital 100
mg/kg BW IV). The entire lumbar spine was harvested Gene Expression Analysis
en bloc, and the pins and posts carefully removed. The Immediately following post-mortem imaging, specimens
spine was sectioned into individual motion segments were either designated for gene expression analysis or
and temporarily stored in Fomblin (Sovay Solexis, West histopathology. Experimental and adjacent unloaded
Deptford, NJ) to prevent dehydration and facilitate high motion segments designated for gene expression analy-
quality MRI images. The super-adjacent, loaded level, and sis were transected through the intervertebral disc using
sub-adjacent motion segments were harvested as a basis standard aseptic technique. Samples of nucleus pulposus
for intra-subject normalization. (NP) and anulus fibrosis (AF) were harvested via sharp
dissection then snap frozen and stored at -80° C. At the
Post Contrast Enhanced T1 and T2 Mapping time of analysis, mRNA was isolated separately from AF
Diminished diffusion through the vertebral endplates
is associated with degenerative disc disease.51-53 In this
study, measurement of gadodiamide in the disc via post
contrast enhanced MRI was the primary means by which
diffusion was quantified. A series of T1 and T2-weighted
images were taken of the lumbar spine using a 7T MRI
scanner (Bruker Biospin, Billerica, MA) equipped with a
9 cm actively shielded gradient set capable of generating
magnetic field strengths of 300 mT/m and a 62 mm cir-
cularly polarized quadrature volume coil. Sagittal images
for T1 relaxation mapping were acquired using a RARE
sequence. MSME sequence scans were used for obtaining
T2 relaxation values. Regions-of-interest (ROIs) were
created in the nucleus pulposus, anterior and posterior
anulus, superior and inferior vertebral bodies, and supe-
rior and inferior endplates. Data for measuring effective
T1 and T2 relaxation constants were calculated from a
series of images acquired with a minimum of 8 relaxation
delays. In each ROI, T1 and T2 values were computed as
quantitative measures of small molecule (gadodiamide)
diffusion through the endplates into the intervertebral disc
and water content, respectively.54,55 Values from loaded
discs were compared to non-loaded adjacent levels and
aged matched controls. Figure 2: Mid-sagittal μCT scans were used to measure
disc space height, endplate thickness, and subchondral
bone density in all specimens.
51
Table 1: Transverse axial MRI images were used to measure intervertebral disc cross-sectional area. Based on slight
variations in disc size, mean applied stress ranged from 2.56 to 2.95 MPa.
Treatment Group Applied Load (N) Mean Disc Area (cm2) Calculated Stress (MPa)
Sx Control 0 0.86 Not Loaded
0.5 Hz 10 wk 220 0.86 2.56
0.5 Hz 26 wk 220 0.75 2.95
5.0 Hz 8 wk 220 0.83 2.65
5.0 Hz 30 wk 220 0.79 2.74

and NP tissue using trizol extraction followed by Qiagen Serial Plain Radiographs
mini-prep kit utilizing a DNAse step to remove genomic Plain radiographs proved to have limited sensitivity for
DNA. Gene expression analysis (real time PCR) was subtle changes in disc space height. There were no statisti-
performed on the NP and AF using custom designed, cally significant changes (p>0.05) in DHI discernable from
validated primers to compare expression of inflammatory plain x-rays in either the high or low frequency loaded
genes (TNF-α, and IL-1β) and genes involved in matrix discs at any time point relative to baseline. In all loaded
homeostasis (Col I, Col II, aggrecan, versican, TIMP-1 levels, the mean change in disc height up to 30 weeks was
and MMP-3). The ∆∆Ct method was utilized to calculate less than 10%. X-rays did indicate osteophyte formation
relative gene expression after normalizing to the house- adjacent to the endplates of some animals subjected to
keeping gene GAPDH. long term (30 weeks) high frequency (5.0 Hz) loading as
shown in Figure 3.
Histopathology
Histologic analyses were used to assess cellular and tissue Post Contrast Enhanced T1 and T2 Mapping
level indicators of disc health and degeneration. Experi- T1 constants were higher (indicative of less gadodiamide
mental and adjacent unloaded motion segments were fixed diffusion) in loaded intervertebral discs relative to adja-
in 10% neutral buffered formalin and decalcified using a cent control levels. Data indicate increased T1 constants
7 day protocol (Formical 2000, Decal Chemical Corp., in the nucleus and anterior anulus of high frequency long
Tallman, NY). Decalcified specimens were routinely
processed, paraffin embedded, and sectioned serially in
the mid-sagittal and lateral para-sagittal planes. Speci-
mens were stained with either hematoxylin and eosin or
safranin-O and fast green. Discs were assessed qualita-
tively for lamellar organization and changes in structural
properties, proteoglycan content, cell morphology, cell
distribution, and fibrosis. The cartilaginous endplates were
assessed for thinning, ossification, disruption, and fracture.
Chondrocyte disorganization, clumping, and hypertrophy
were also assessed.

Results
Of the 11 animals in the study, all animals tolerated the
surgical procedure well. All were sufficiently acclimated
to the Degenerator so that none showed signs of distress
during loading and none required restraining. One poorly
placed transfixing pin failed at approximately 750,000
cycles (4 weeks) in the high frequency 8 week group and
the data from the animal were not included in subsequent
analyses. There were no other study related complications.
The mean disc cross-sectional area and corresponding disc
space stress for each group are shown in Table 1. Figure 3: Axial loading resulted in substantial disc
height loss in some animals.
52
Figure 4: Post contrast enhanced MRI indicated
diminished diffusion in loaded levels (lower) relative
to unloaded adjacent levels (upper).
term animals, but no differences were detected in the
posterior annulus. More subtle differences were detected
in low frequency long term animals. Analysis of endplates
indicated a 10.5% decrease in T1 constants (increase in
gadodiamide) in the caudal endplate relative to the non-
loaded surgical control, as shown in Figure 4. The high
frequency long term group also showed trends of decreased
T2 constant in the nucleus, indicative of a reduction in
water as shown in Figure 5.
Micro CT Analysis
CT analysis revealed that cyclic loading resulted in an
Figure 6: There was a 12% increase in caudal endplate subchondral bone density resulting from 10 weeks of low
frequency cyclic loading (left). Bone density increased by 21% in motion segments exposed to high frequency
loading for 30 weeks (right).
Figure 5: Post-contrast enhanced MRI imaging normal-
ized to the sub-adjacent level indicated an increase in the
T1 constant (decrease in gadodiamide) following 30 weeks
of high frequency loading (top). The corresponding T2
constants were reduced (bottom).
increase in subchondral bone density at loaded levels
proposal to duration and frequency of exposure. After
10 weeks of low frequency loading, subchondral bone
density increased by a mean of 11.9% and by 30 weeks
of high frequency loading, bone density increased by
20.6% compared to the sub-adjacent unloaded level as
shown in Figure 6.
Gene Expression Analysis
Data from a limited sample of NPs indicate a frequency ef-
fect and duration effect with early (10 week) low frequency
(0.5 Hz) samples illustrating a reduction in expression
53
Figure 7: Mid-sagittal sections of surgical control (upper left), 26 week low frequency (upper right), 30 week high
frequency (lower left), and 48 week static load (lower right) were stained with safranin-O and fast green to assess
disc morphology and matrix organization. All images are 2X.
of versican and an increase in Col II accompanied by a
dramatic reduction in Col I, TNF-α, IL-1β, MMP-3 and
TIMP-1 relative to unloaded levels. Early (8 week) high
frequency (5.0 Hz) samples illustrated a slight increase in
expression of IL-1β and TIMP-1, with a slight reduction in
COX-2, ADAMTS5, and MMP-3. Late (26 week) low fre-
quency samples indicate an increase in versican and Col II.
In contrast, late (30 week) high frequency (5.0 Hz) samples
express an increase in aggrecan and Col II accompanied
by an increase in TNF-α, MMP-3, and TIMP-1.
Data from the low frequency AF samples indi-
cate an initial pro-inflammatory response with dramatic
increases in iNOS and IL-1β expression at 10 weeks, fol-
lowed by apparent remodeling at 26 weeks as indicated
by a decrease in all pro-inflammatory markers and a
sharp increase in Col II expression. Relative to unloaded
adjacent controls, high frequency AF samples indicated
an early reduction in expression of all inflammatory and
homeostatic genes followed by an upregulation in expres-
sion of all measured in the 30 week high frequency loaded
discs with the most substantial increase in expression of
Col II.
Histopathology
The NP of unloaded age-matched surgical control sections
were sparsely cellular with a gelatinous extracellular ma-
trix. Twenty-six week loaded low frequency sections indi-
cated fibrosis and increased cell density in the NP. Thirty
week loaded high frequency sections showed advanced
fibrosis, cell clustering, significant matrix disorganization,
and disc space narrowing.
As shown in Figure 7, relative to an age-matched
unloaded surgical control disc, 30 weeks of high frequency
loading has a deleterious effect on the disc. Loaded discs
exhibited substantial fibrosis and increased cell number
54
and clustering in the nucleus, accompanied by matrix
disorganization and a loss in disc height.
Analysis of endplate cartilage and subchondral
bone indicated that unloaded surgical control endplates
have a distinct border with the adjacent subchondral bone.
There are resting chondrocytes near the nucleus side and
relatively sparse hypertrophic chondrocytes on the bone
side. With loading, there is evidence of chondrocyte
cloning and hypertrophy. The cartilaginous endplates
are thinned while the subchondral bone is thickened. As
shown in Figure 8, the bony side of the cartilage is more
diffusely integrated with bone and there are more numer-
ous hypertrophic chondrocytes present.
Discussion
We have developed an apparatus, atechnique, and an ani-
mal model to study the long-term effects of mechanical
loading on the intervertebral disc, endplates, and subchon-
dral bone in the lumbar spine in vivo. The model allows us
to apply physiologic loading to an individual motion seg-
ment of the NZW rabbit and can be used to characterize the
effects of load magnitude, frequency, and duration on the
health and degeneration of the spine. For the current study,
we utilized the NZW rabbit, but the technique is potentially
scalable to other species using a modified apparatus. The
model is unique in that it facilitates investigation of the
earliest effects of mechanical loading at the molecular,
cellular, tissue, and global scale of both the intervertebral
disc and endplates in the lumbar spine.
Unlike anular puncture models and chemonucle-
olysis models which presuppose that degeneration is
initiated by anular failure or disc matrix changes,22,23 re-
spectively, the model utilized in the current study is based
on mechanical loading alone with no assumptions on the
mechanism of degeneration. Likewise, previous models
Figure 8: In unloaded levels (left), the disc side of the cartilaginous endplate is comprised primarily of resting
articular chondrocytes (RC). In loaded levels (right), there is evidence of clustered hypertrophic chondrocytes (HC)
and endochondral bone formation with no distinct tidemark. 20X.

for studying the role of endplate diffusion in disc degenera-
tion are based on iatrogenic alteration of the endplates or
subchondral bone.56,57 In the current study, no disruptions
were made to the loaded disc, endplates, or subchondral
bone, which uniquely facilitated study of the early effects
of mechanical loading on the lumbar spine.
Like the rat caudal vertebrae model, our results
indicate that physiologic mechanical loading alone
will initiate degenerative changes in the intervertebral
disc.18-21 Unique to our model, we have also shown that
the endplates respond to loading with increasing bone
density and endplate thickness. Because of the critical
role of the endplates in disc homeostasis, characterizing
the complex interrelationship between disc and endplate
pathology is critical in defining the mechanism of spinal
degeneration.
In this pilot study, the significance of the results
is hampered by the small sample size. In cyclically loaded
specimens, we measured an increase in subchondral bone
density and an increase in uptake of MRI contrast agent in
the subchondral bone, indicative of sclerotic changes to the
endplates in response to loading. In the same specimens,
there was a decrease in contrast agent into the disc. This
suggests a reduction in passive diffusion from the subchon-
dral capillaries to the disc. Endplate histology corroborated
these findings, indicating thickened mineralized endplates
in specimens exposed to prolonged loading. As shown in
Figure 9, the correlation suggests that mechanisms affect-
ing subchondral bone density also potentially affect disc
diffusion and homeostasis. This has been demonstrated
previously in an unstable cervical spine injury model.58
Gene analyses demonstrated substantial changes
in expression of loaded levels compared to adjacent
unloaded controls in the NP and AF. Although the exact
interpretation of these results is hampered by the small
sample size, the observed trends suggest that the slower

Figure 9: Among all surgical levels, there is a correlation between T1 constant and subchondral bone density.
55
accumulation of load cycles in the low frequency group
may have a beneficial effect on matrix homeostasis and
inflammation in our model, whereas the rapid accumula-
tion of cycles in our high frequency group may stimulate
matrix synthesis accompanied by pro-inflammatory and
catabolic activity. Loading at specific frequencies and
magnitudes may have a protective effect, upregulating
anabolic activity and initiating remodeling rather than
chronic degeneration.41,43,59
Based on our preliminary data, the mechanism
of disc degeneration and the relationship between cyclic
References
1. Kumar S, Panjabi MM. In vivo axial rotations and
neutral zones of the thoracolumbar spine. J Spin Disord.
1995; 8(4):253-63.
2. Pope MH, Novotny JE. Spinal biomechanics. J
Biomech Eng. 1983; 115:569-574.
3. Martin BI, Deyo RA, Mirza SK, Turner JA, Comstock
BA, Hollingworth W, Sullivan SD. Expenditures and
health status among adults with back and neck problems.
Journal of the American Medical Association. 2008;
299(6):656-64.
4. A critical review of epidemiologic evidence for work-
related musculoskeletal disorders of the neck, upper
extremity, and low back. Second Printing. July, 1997;
DHHS (NIOSH) Publication No. 97-141.
5. Andersson GBJ. Functional evaluation of the back.
From: Biomechanics of Human Movement: Applica-
tions in Rehabilitation Sports and Ergonomics. Edited
by Berme N, Cappozzo A. Worthington:Bertec Cor-
poration.
6. Hutton WC, Ganey TM, Elmer WA, Kozlowska E,
Ugbo JL, Doh ES, Whitesides TE. Does long-term
compressive loading on the intervertebral disc cause
degeneration? Spine. 2000; 25(23):2993-3004.
7. Setton LA, Chen J. Cell mechanics and mecha-
nobiology in the intervertebral disc. Spine. 2004;
29(23);2710-23.
56
loading, diffusion, and disc degeneration remains unclear.
But our preliminary data support the hypothesis that load-
ing results in subchondral sclerosis, decreased diffusion,
and ultimately disc degeneration. Characterization of the
temporal sequence of these events, the mechanism un-
derlying degeneration, and the dose response of the spine
to mechanical loading may provide insights into future
strategies for novel preventative measures, diagnostic
assessments, and therapies targeting low back pain. The
novel we have developed can uniquely facilitate future
work in these critical areas.
8. Zhang YG, Zhengming S, Zhang Z, Liu J, Guo X.
Risk factors for lumbar intervertebral disc herniation
in Chinese population. A case-control study. Spine.
2009;34(25):E918-22.
9. Jung YJ, Isaacs JS, Lee S, Trepel J, Neckers L. IL-
1β-mediated up-regulation of HIF-1α via an NFκB/
COX-2 pathway identifies HIF-1 as a critical link be-
tween inflammation and oncogenesis. FASEB Journal.
2003;17:2115-7.
10. Urban JPG, Holm S, Maroudas A, Nachemson
A. Nutrition of the intervertebral disc – effect of fluid
flow on solute transport. Clin Orthop Rel Res. 1982;
170:296-302.
11. Sandover J. Dynamic loading as a possible source
of low back disorders. Spine. 1983; 8(6):852-8.
12. Holm S, Maroudas A, Urban JPG, Selstam G, Nach-
emson A. Nutrition of the intervertebral disc: solute
transport and metabolism. Connective Tissue Research.
1981; 8:101-19.
13. Haldeman SD, Kirkaldy-Willis WH, Bernard TN.
An atlas of back pain. London:Parthenon Publishing,
2002.
14. Lusins JO, Cicoria AD, Glodsmith SJ. SPECT and
lumbar MRI in back pain with emphasis on changes in
end plates in association with disc degeneration. Journal
of Neuroimaging. 1998; 8(2):78-82.
15. Farkas T, Boyd RD, Schaffler MB, Radin EL, Burr
DB. Early vascular changes in rabbit subchondral bone
after repetitive impulsive loading. Clinical Orthopae-
dics and Related Research. 1987; 219:259-67.
16. Burr DB, Radin EL. MicroFractures and microc-
racks in subchondral bone: are they relevant to osteoar-
throsis? Rheum Dis Clin N Am. 2003; 29:675-85.
17. Modic MT, Masaryk TJ, Ross JS, Carter JR. Imaging
of degenerative disk disease. Radiology. 1988;168:177-
86.
18. Iatridis JC, MacLean JJ, Roughley PJ, Alini M.
Effects of mechanical loading on intervertebral disc
metabolism in vivo. J Bone Joint Surg Am. 2006;
88A(Suppl. 2):41-46.
19. MacLean JJ, Lee CR, Alini M, Iatridis JC. Anabolic
and catabolic mRNA levels of the intervertebral disc
vary with magnitude and frequency of in vivo dynamic
compression. J Orthop Res. 2004;22:1193-1200.
20. MacLean JJ, Lee CR, Alini M, Iatridis JC. The
effect of short term load duration on the anabolic and
catabolic gene expression in the rat tail intervertebral
disc. J Orthop Res. 2005; 23:1120-7.
21. Lotz JC, Colliou OK, Chin JR, Duncan NA, Lie-
benberg E. 1998 Volvo award winner in biomechani-
cal studies: compression-induced degeneration of the
intervertebral disc: an in vivo mouse model and finite-
element study. Spine. 1998;23(23):2493-506.
22. Kim KS, Yoon ST, Li J, Park JS, Hutton WC. Disc
degeneration in the rabbit: a biochemical and radiologi-
cal comparison between four disc injury models. Spine.
2004; 30(1):33-7.
23. Takaishi H, Nemoto O, Shiota M, et al. Type-II
collagen gene expression is transiently upregulated in
experimentally induced degeneration of rabbit interver-
tebral disc. J Orthop Res. 1997;15:528-38.
57
24. Lipson SJ, Muir H. Proteoglycans in experimental
intervertebral disc degeneration. Spine. 1981;6:194-
210.
25. Farkas T, Boyd RD, Schaffler MB, Radin EL, Burr
DB. Early vascular changes in rabbit subchondral bone
after repetitive impulsive loading. Clinical Orthopae-
dics and Related Research. 1987; 219:259-67.
26. Kennedy J, Baris C, Hoyland JA, Selby PL, Free-
mont AJ, Braidman IP. Immunofluorescent localization
of estrogen receptor-alpha in growth plates of rabbits,
but not in rats, at sexual maturity. Bone. 1999;24(1):9-
16.
27. Hunter CJ, Matyas JR, Duncan NA. Cytomorphol-
ogy of notochordal and chondrocytic cells from the
nucleus pulposus: a species comparison. J Anat. 2004;
205:357-62.
28. Scott NA, Harris PF, Bagnall KM. A morphological
and histological study of the postnatal development of
intervertebral discs in the lumbar spine of the rabbit. J.
Anat. 1980; (130):75–81.
29. Miyazaki T, Kobayashi S, Takeno K, Meir A, Urban
J, Baba H. A phenotypic comparison of proteoglycan
production of intervertebral disc cells isolated from
rats, rabbits, and bovine tails; which animal model is
most suitable to study tissue engineering and biologi-
cal repair of human disc disorders? Tissue Eng Part A.
2009; 15(12):3835-46.
30. Panjabi MM, Andersson GB, Jorneus L, Hult E,
and Mattson L. In-vivo Measurement of spinal column
vibrations. Proceedings of American Society of Biome-
chanics. Rochester, MN. Sept. 28-30, 1983.
31. Weinstein J, Pope M, Schmidt R, and Seroussi R.
Neuropharmacologic effects of vibration on the dorsal
root ganglion. Spine. 1988; 13(5):521-5.
32. Nachemson A and Morris JM. In vivo measure- 41. Walsh AJL, Lotz JC. Biological response of the
ments of intradiscal pressure: discometry, a method intervertebral disc to dynamic loading. J Biomech.
for the determination of pressure in the lower lumbar 2004; 37:329-37.
discs. Journal of Bone and Joint Surgery. 1964; 46-
42. Kroeber MW, Unglaub F, Wang H, Schmid C,
A(5):1077-1092
Thomsen M, Nerlich A, Wiltrud R. New in vivo animal
33. Wilke HJ, Neef P, Caimi M, Hoogland T, and model to create intervertebral disc degeneration and to
Claes LE. New in vivo measurements of pressures investigate the effects of therapeutic strategies to stimu-
in the intervertebral disc in daily life. Spine. 1999; late disc regeneration. Spine. 2002; 27(23):2684-90.
24(8):755-762.
43. Handa T, Ishihara H, Ohshima H, Osada R, Tsuji
34. Ledet EH, McVoy JL, Sanders GP, Tymeson MP. H, Obata K. Effects of hydrostatic pressure on matrix
Normal mechanical and physical properties of com- synthesis and matrix metalloproteinase production in
monly used animal models for study of the lumbar the human lumbar intervertebral disc. Spine. 1997;
spine. Proceedings of the 51st Annual Meeting of 22(10):1085-91.
the Orthopaedic Research Society. Washington, DC.
44. Berry JL, Moran JM, Berg WS, Steffee AD. A
February 19-23, 2005. 274.
morphometric study of human lumbar and selected
35. Jackson JM, Tymeson MP, Ledet EH. A comprehen- thoracic vertebrae. Spine. 1987; 12(4):362-7.
sive interspecies comparison of the mammalian lumbar
45. Polga DJ, Beaubien BP, Kallemeier PM, Schellhas
spine – Part II: Morphometrics. The Spine Journal.
KP, Lew WD, Buttermann GR, Wood KB. Measure-
2007; 7(5S):160S.
ments of in vivo intradiscal pressure in healthy thoracic
36. Ledet EH, Sachs BL, Brunski JB, Gatto CE, Donzelli intervertebral discs. Spine. 2004; 29(12):1320-4.
PS. Real time in vivo loading in the lumbar spine. Part
46. Markolf KL, Morris JM. The structural com-
1: Interbody implant – load cell design and preliminary
ponents of the intervertebral disc. J Bone Joint Surg.
results. Spine. 2000;25(20): 2595-600.
1974;56-A(4):675-87.
37. Ledet EH, Tymeson MP, DiRisio DJ, Cohen B, Uhl
47. Huston DR, Johnson CC, Zhao XD. A human ana-
RL. Direct real time measurement of in vivo forces in the
log for testing vibration attenuating seating. Journal of
lumbar spine. The Spine Journal. 2005; 5(1):85-94.
Sound and Vibration. 1998; 214(1):195-200.
38. Schultz A, Andersson G, Ortengren R, Haderspeck
48. Wasserman DE, Wilder DJ, Pope MH, Magnusson
K, Nachemson A. Loads on the lumbar spine. Journal of
M, Aleksiev AR, Wasserman JF. Whole-body vibration
Bone and Joint Surgery. June 1982;64-A(5):713-720.
exposure and occupational workhardening. J Occup
39. Kroeber M, Unglaub F, Guegring T, Nerlich A, Hadi Environ Med. 1997; 39(5):403-7.
T, Lotz J, Carstens C. Effects of controlled dynamic disc
49. Wilder DG, Frymoyer JW, Pope MH. The effect of
distraction on degenerated intervertebral discs. Spine.
vibration on the spine of the seated individual. Auto-
2005; 30(2):181-7.
medica. 1985; 6(2):5-35.
40. Unglaub F, Guehring T, Lorenz H, Carstens C,
Kroeber MW. Effects of unisegmental disc compression
on adjacent segments: an in vivo animal model. Eur
Spine J. 2005; 14:949-55.

58
50. Masuda K, Aota Y, Muehleman C, Imai Y, Okuma
M, Thona EJ, Andersson GB, An HS. A novel rabbit
model of mild, reproducible disc degeneration by an
anulus needle puncture: correlation between the degree
of disc injury and radiological and histological appear-
ances of disc degeneration. Spine. 2004; 30(1): 5–14.
51. Boos N, Weissbach S, Rohrhach H, Weiler C,
Spratt KF, Nerlich AG. Classification of age-related
changes in lumbar intervertebral discs. Spine. 2002;
27(23):2631-44.
52. Rajasekaran S, Babu JN, Arun R, Armstrong BRW,
Shetty AP, Murugan S. A study of diffusion in human
lumbar discs: a serial magnetic resonance imaging
study documenting the influence of the endplate on
diffusion in normal and degenerate discs. Spine. 2004;
29(23):2654-67.
53. Katz MM, Hargens AR, Garfin SR. Intervertebral
disc nutrition – diffusion versus convection. Clin Orthop
Rel Res. 1986; 210:243-5.
54. Arun R, Freeman BJC, Scammell BE, McNally
DS, Cox E, Gowland P. 2009 ISSLS Prize Winner:
What influence does sustained mechanical load have
on diffusion in the human intervertebral disc? Spine.
2009;34(21);2324-37.
59
55. Niinimaki J, Korkiakoski A, Parviainen O, Haapea
M, Kuisma M, Ojala RO, Karpppinen J, Korpelainen R,
Tervonen O, Nieminen M. Association of lumbar artery
narrowing, degenerative changes in disc and endplate
and apparent diffusion in disc on postcontrast enhance-
ment of lumbar intervertebral disc. Magn Reson Mater
Phy. 2009; 22:101-9.
56. Cinotti G, Rocca CD, Romeo S, Vittur F, Toffanin
R, Trasimeni G. Degenerative changes of porcine in-
tervertebral disc induced by vertebral endplate injuries.
Spine. 2005; 30(2):174-80.
57. Benneker LM, Heini PF, Alini M, Anderson SE,
Ito K. Vertebral endplate marrow contact channel oc-
clusions and intervertebral disc degeneration. Spine.
2005; 30(2):167-73.
58. Peng B, Hou S, Shi Q, Jia L. The relationship
between cartilage end-plate calcification and disc de-
generation – an experimental study. Chinese Medical
Journal. 2001; 114(3):308-12.
59. Wang DL, Jiang SD, Dai LY. Biologic response of
the intervertebral disc to static and dynamic compres-
sion in vitro. Spine. 2007; 32(23):2521–8.
 The Effect of Anti-Rotation and Anti-Cutout Features in Cephalomedullary Nail Fixation of
Intertrochanteric Femur Fractures: An Acute Biomechanical Study
*Matthew W. Connelly, MS, **Jared Roberts, MD, †Rosemary Buckle, MD, *Eric H. Ledet, Ph.D.
* Department of Biomedical Engineering, Rensselaer Polytechnic Institute, Troy, NY.
** Division of Orthopaedic Surgery, The Albany Medical College, Albany, NY.
†
Orthopaedic Associates, LLP, Bellaire, TX.

Background: Extra-capsular hip fractures are common in the elderly population. Cutout, loss of fixation, and
proximal fragment rotation remain as possible complications of treatment with cephalomedullary nails. The pur-
pose of this study was to biomechanically compare two cephalomedullary nails that incorporate anti-rotation and/
or anti-cutout features to a previous generation nail with no such features.

Methods: The Gamma 3 Nail with standard lag screw was used as the benchmark for comparison to the Gamma
3 RC Clip nail and the Trigen InterTAN nail, both of which incorporate anti-cutout and anti-rotation features. An
intertrochanteric fracture was created in composite bone models. Each specimen was loaded to simulating single
leg stance 1,000 times while stiffness, migration (cutout), compressive force across the fracture site, and distal
fragment rotation were monitored.

Results: Maximum lag force during compression was very dynamic, highly variable, and decreased rapidly. Stiff-
ness was highest in the Trigen InterTAN group and mean migration of the proximal fragment was maximum in the
InterTAN group and least in the Gamma 3 group. Proximal fragment rotation was small for all constructs.

Conclusions: In the dense bone of the composite specimens, the enhanced designs of the Gamma 3 RC and the
InterTAN yielded little improvement relative to the Gamma 3. The stiffer InterTAN resulted in higher migra-
tion likely because a stiffer construct will cause compaction of bone around the lag screw. Additional research
in osteopenic bone is necessary to comprehensively characterize the effects of the design enhancements of these
implants.

Introduction: Cephalomedullary devices initially gained popularity for

Extracapsular hip fractures are common in the elderly
population. In 2004, an estimated 756,000 proximal femur
fractures occurred in the United States.1 That number
is increasing, and estimates suggest that four out of ten
women over age 50 in the United States are likely to ex-
perience a hip, spine, or wrist fracture sometime during
the remainder of their lives.2 Hip fractures in the elderly
population are associated with a three times greater risk
of mortality during the first three months after fracture
compared to non-fractured cohorts.3,4 Nearly one in five hip
fracture patients ends up in a nursing home and mortality
after hip fracture is around 30% at one year.5 Prevention
and treatment of hip fractures has thus become a national
priority.6
Despite significant improvements in both surgery
and rehabilitation in recent decades, hip fracture treatment
remains a challenge. Treatment options for extracapsular
hip fractures include extramedullary devices such as com-
pression hip screws or cephalomedullary devices. Until
recently, the compression hip screw was the gold standard
for intertrochanteric fractures, both stable and unstable.
treatment of unstable fractures but have now become a
suitable option for even the stable intertrochanteric frac-
ture. Mechanically, implant strains are decreased with
intramedullary fixation of intertrochanteric fractures
relative to dynamic side plates and sliding screw fixation.7
However, cutout, loss of fixation, and proximal fragment
rotation remain as potential complications of treatment
with cephalomedullary nails.8-11 The latest generation of
nails are designed to minimize cutout and incorporate
anti-rotation features in the design. These design features
purport to minimize loss of fixation and result in better
outcomes. Several intramedullary products are currently
available to the orthopaedic surgeon, however, the relative
performance characteristics of these implants have not
been reported and to date there have been no comprehen-
sive reports on the biomechanics of the latest generation
nails compared to previous designs.
The purpose of this study was to compare two
cephalomedullary nails that incorporate anti-rotation and/
or anti-cutout features to a previous generation nail with
no such features with respect to fracture compressive lag

Acknowledgements: This study was funded by an institutional grant from Stryker.

60
force, bending stiffness, migration, cutout, and proximal
fragment rotation using biomechanical testing in an inter-
trochanteric in vitro fracture model.
Materials and Methods
The Gamma 3 Nail with standard lag screw (Stryker,
Mahwah, NJ) was used as the benchmark for comparison
to the Gamma 3 RC Clip nail (Stryker, Mahwah, NJ) and
the Trigen InterTAN nail with compression screw (Smith
& Nephew, Memphis, TN). As shown in Figure 1, both
the Gamma 3 RC Clip nail and the Trigen InterTAN nail
incorporate anti-rotation and/or anti-cutout features not
found in previous nail designs.
To evaluate the biomechanical performance of
the nails, Evans Type I stable intertrochanteric fractures
were created in fourth-generation composite Sawbone
left femurs (Pacific Research Laboratories, Vashon, WA).
The composite bones have similar properties to healthy
human bone.12-14
All specimens were prepared in identical fashion.
Using a jig for repeatability, a bandsaw was used to create a
partial intertrochanteric osteotomy centered proximally in
the greater trochanter and extending distally to the center
of the lesser trochanter.
Before completing the osteotomy, the femurs
Figure 1: The lag screws used with the Gamma 3 RC Clip nail (center) and the Trigen InterTAN nail (right) incorpo-
rate anti-rotation and/or anti-cutout features not found in previous nail designs such as the Gamma 3 nail (left).
61
were randomly assigned to receive either the 125° Gamma
3 Nail (180 mm long with 10.5 mm ɸ x 95 mm standard
lag screw) in the “G3S” group, the 125° Gamma 3 RC Clip
Nail (180 mm long with 10.5 mm ɸ x 95 mm lag screw
and clip) in the “G3RC” group, or the 125° InterTAN Nail
with compression screw (180 mm long with 11 mm ɸ x 95
mm lag screw and 7 mm ɸ x 90 mm compression screw)
in the “ITI” group. A sample size of six (n=6) was used
for each treatment group for a total of 18 specimens.
Implants were placed according to each manu-
facturer’s recommended technique with every attempt
made to position the lag screws so that the final tip to
apex distance was less than 25 mm as recommended by
Baumgaertner.9 Pilot holes for placement of the nails and
the lag screws were drilled using a drill press and custom
jigs to minimize variability in placement technique. Sub-
sequently, implants were placed using standard techniques
with each manufacturer’s instruments. The lag screws
were placed across the partial osteotomy extending sev-
eral turns into the femoral head but were not advanced
completely into the proximal fragment. Using a bandsaw,
the osteotomies were then completed without the blade
touching the lag screw. The proximal fragment was then
removed by carefully unscrewing it from the lag screw.
A second osteotomy was cut from the proximal fragment

Figure 2: All specimens underwent an identical proce-
dure to create a 5 mm osteotomy.
parallel to the first resulting in a 5 mm thick section of
intertrochanteric bone being removed. The proximal frag-
ment was then carefully placed back onto the lag screw
in a position identical to when it was removed. This fa-
cilitated anatomic reduction of each fracture with a 5 mm
osteotomy and minimized variation between specimens
as shown in Figure 2.
To measure compressive lag force across the
osteotomy site, an array load cells was placed in the os-
teotomy gap. Three 3.5 mm thick button load cells (Futek
Advanced Sensor Technology, Inc., Irvine, CA) were fixed
between two aluminum plates so that the overall thick-
ness of the sensor array was 5 mm. The sensor array was
placed into the osteotomy gap around the lag screw so
that the three sensors were located proximal to the screw,
anterodistal to the screw, and posterodistal to the screw.
The sensor array allowed free passage of the lag screw
into the femoral head, as shown in Figure 3.
Before tightening the lag screws, the distal third
of each femur was cut off, the nails were locked distally,
and the specimen was potted vertically using quick setting
epoxy (Bondo, Atlanta, GA) in an aluminum cup. The
cup was mounted to a fixture and placed in a mechanical
testing machine (MTS Systems, Carry, NC) so that it was
oriented at 25° adduction in the coronal plane and neutral
in the sagittal plane.15-17
Once mounted in the mechanical testing ma-
chine, the lag screws were inserted completely into the
distal fragment, extending through the sensor array into
the femoral head. The lag screws were then compressed
2.5 mm one half screw turn at a time while compressive
force data were recorded from the sensor array at a rate
of at least 10 samples per second.
62
Figure 3: Posterior view images were used to gauge
movement of the proximal fragment relative the femoral
shaft. Compression forces across the fracture site (in-
dicated by arrows) were measured via a load cell array
placed in the osteotomy gap.
After a one minute quiescent period, a high
resolution digital photograph was taken of the proximal
femur from a posterior view with a scale marker positioned
within the image for reference, as shown in Figure 3. Ad-
ditional “flag” pins were placed in the proximal and distal
fragments extending posteriorly to monitor any rotation of
the proximal fragment during loading. A high resolution
digital photograph was taken of the femur from a lateral
view with the markers positioned within the image for
reference, as shown in Figure 4.
After baseline (no load) images were taken, the
crosshead of the mechanical testing machine was lowered
and zeroed at the point of contact with the femoral head.
A flat loading platen attached to the crosshead was used
to apply a vertical compressive load to the femoral head.
The platen was attached to a roller plate to allow uncon-
strained motion of the femoral head under axial loading.
The specimen was then loaded to 750 N simulating single
leg stance17-19 and unloaded back to zero at a rate of 25
mm per minute while fracture compression force data,
mechanical testing machine force data, and crosshead
displacement data were collected. At peak load, posterior
and lateral view photographs were taken for later assess-
ment of deformation/relative motion. After unloading, the
specimen was again photographed from both posterior and
lateral directions. Each specimen was then cycled 1,000
times from 50 to 750 N. At 10, 100, 250, 500, 750, and
1,000 cycles, data were recorded and each specimen was
photographed at peak load and after unloading to monitor
stiffness, changes in femoral head displacement (migra-
tion), compressive force across the fracture site, and distal
fragment rotation.
Figure 4: Image analysis and two pins were used to
determine the amount of proximal fragment rotation
relative to the fixed distal component.
Migration, defined as extent of the residual
collapse of the proximal fragment after removal of load
at each cycle, was measured using the crosshead of the
mechanical testing machine. After completion of each
measured loading cycle, the difference between the
baseline location of the crosshead and the femoral head
was measured. Residual femoral head displacement (mi-
gration) was compared between treatment groups at all
loading cycles.
After completion of cyclic testing, the sensor
array was removed and the proximal fragment was un-
screwed from the lag screw. The proximal fragment was
sectioned using a bandsaw and the tip to apex distance
was measured using a caliper, as shown in Figure 5.
Photographs were analyzed digitally (ImageJ, NIH) to
determine proximal fragment rotation about the lag screw.
Compressive lag force across the osteotomy was analyzed
at cycles 1 through 100. Stiffness of each construct was
calculated from the linear region of the force-displacement
curves from the mechanical testing machine. Quantitative
data were compared between the three treatment groups
using an ANOVA and Games-Howell post-hoc testing with
values considered statistically significant at p<0.05.
Results
All of the implants were placed in a similar fashion us-
ing the custom jigs. Due to the density of the composite
bone models (simulating healthy bone), compression of
the lag screws was very difficult in all specimens and
hand tools (pliers) were necessary to rotate the lagging
apparatus. Compression beyond 2.5 mm would likely
have been impossible in this model. One specimen (in
the G3S group) cracked during testing. This specimen
was the first tested and an initial attempt had been made
to compress it beyond 2.5 mm. The specimen was likely
damaged during compression and subsequently it was not
used in the analysis after it cracked. All other specimens
63
Figure 5: Tip to apex distance (arrow) was measured
from the proximal fragment using the technique pro-
posed by Baumgaertner9 after sectioning with a bandsaw.
were compressed to 2.5 mm and completed the testing
protocol with no device related complications.
Compressive Lag Force
Maximum lag force during compression was very dynamic
and highly variable. As shown in Figure 6, the force
was highly time dependent and decreased rapidly. The
mean maximum compressive force during lagging was
highest in the G3S group followed by the G3RC group
followed the ITI group, as shown in Figure 7. There was
no significant difference in mean maximum force between
G3S and G3RC groups after compression, but both were
significantly higher than the ITI group (p<0.002).
After a single loading cycle, there was a substan-
tial loss in compressive force in all treatment groups. The
compressive force across the osteotomy gap dropped to
less than 10% of the peak value by completion of the first
cycle. The ITI and G3RC groups maintained the highest
forces after the first cycle, with the G3S group dropping
to near zero by cycle 100. There were no statistically
significant differences between treatment groups at any
time after the initial lag likely due to the high variability
in the readings, as shown in Figure 7. Continued cyclic
loading did affect the magnitude of the compressive force
across the osteotomy with some cycles resulting in a slight
increase in force.
Construct Stiffness
Construct stiffness was calculated from the linear portion
of the load/crosshead displacement curve. The mean stiff-
ness was highest in the ITI treatment group at all cycles,
as shown in Figure 8. The mean stiffness of all constructs
increased with cycles. The differences in stiffness between
treatment groups were not significant at the first cycle of
testing, but the difference between the ITI stiffness and the
other two treatments approached significance by the 10th
cycle (p<0.09). Mean ITI stiffness became significantly
Figure 6: Compressive force was measured as the lag screw was used to compress the osteotomy.

Figure 7: The mean maximum compressive force during lagging was significantly higher in the G3S and G3RC
groups compared to the ITI group, but the force diminished substantially for all groups by completion of the first
cycle of loading. Error bars indicate on standard deviation.
64
Figure 8: The mean stiffness was highest in the ITI treatment group at all cycles and was significantly higher than
the G3S and G3RC groups at cycles 100 through 1000. Error bars indicate on standard deviation.
higher from the 100th to 1,000th cycle (p<0.038). There
were no significant differences between the G3S and
G3RC groups at any cycle.
Proximal Fragment Rotation
The proximal fragment rotation was small for all con-
structs. The maximum mean rotation of any construct was
1.8°, measured in the G3S group after cycle 250. Although
the rotation of the G3S group was at least three times
higher than the other groups at every measured cycle, the
differences between groups were not statistically signifi-
cantly different at any cycle number, likely due to the high
variation in specimens, as shown in Table 1.
Migration
There was no cutout in any specimen and migration of
the lag screw within the femoral head was minimal for
all treatment groups. At every measured cycle, mean
migration was maximum in the ITI group and least in
the G3S group. Mean migration ranged from 1.2 to 2.3
mm as shown in Figure 9. There were no statistically
significant differences in cutout between any treatments
at any measured cycle.
Tip to Apex Distance
Mean tip to apex distance for each treatment group is
shown in Table 1. The tip to apex distance was statistically
significantly (p<0.002) lower in the ITI group compared to
65
the other two treatment groups, but there were no signifi-
cant differences between the G3S and G3RC groups.
Discussion
The Gamma 3 RC nail incorporates a clip that slides over
the lag screw and flares out at the medial end of the screw
after it has been placed. The Trigen InterTAN nail incor-
porates a second compression screw that interlocks with
the lag screw. The addition of these features is intended
to reduce proximal fragment rotation and cutout. Both
designs add to the complexity and cost of the systems
relative to more traditional nails that do not incorporate
such features. In this study, we evaluated the effects of
the enhanced design features of the latest generation of
cephalomedullary nails relative to the more traditional
design of the Gamma 3 Nail. To evaluate the effects on
the biomechanical properties, we used a physiologically
relevant model of hip biomechanics to simulate the acute
post-operative phase of loading.
Our results indicate that all of the systems are
capable of generating substantial compressive force to
lag a fracture intra-operatively. The G3S and G3RC
nails generated over 800 N of compressive force in the
dense bone of the composite models. This force is ample
for reduction and compression of most intertrochanteric
fractures.20 However, the compressive force was imme-
diately lost and dropped to less than 10% of its peak after
a single cycle of loading. This indicates that while the
Table 1: Proximal Fragment Rotation and Tip to Apex Distance

G3S ITI G3RC

Mean Stdev Mean Stdev Mean Stdev
Rotation Cyc 1 [°] 1.2 1.3 0.4 0.2 0.4 0.2
Rotation Cyc 10 [°] 1.2 1.5 0.4 0.3 0.3 0.4
Rotation Cyc 100 [°] 1.6 2.1 0.4 0.3 0.4 0.3
Rotation Cyc 250 [°] 1.8 2.5 0.4 0.3 0.2 0.3
Rotation Cyc 500 [°] 1.8 2.4 0.5 0.3 0.3 0.3
Rotation Cyc 750 [°] 1.6 1.9 0.4 0.2 0.4 0.3
Rotation Cyc 1,000 [°] 1.5 1.8 0.5 0.3 0.4 0.3
TAD [mm] 34.1 1.4 16.9 2.2 33.4 6.1
TAD, tip to apex distance.

implants are effective at lagging the proximal fragment
intra-operatively, none are capable of maintaining the
compression under the loads associated with single leg
stance. The G3S system was able to generate the highest
mean compressive force, but resulted in the lowest com-
pressive force following one cycle of loading and dropped
to zero compressive force by cycle 100. The other two
systems were not able to generate as much compression,
but were able to maintain some compression through the
first 100 cycles of loading.
Like the sliding hip screw, the cephalomedullary
nail is designed to allow shortening of the fracture under
load to allow bone impaction and facilitate increased
stability at the fracture site.21 All three implants used in
this study have set screws which hold the lag screw in
the nail. While the set screws appear to be effective at
preventing rotation of the lag screw relative to the nail, all
three systems allow dynamic sliding of the screw when the
femur is loaded, even when the set screws are tightened.
Our results indicate that the set screws are sufficient to
maintain the relative position of the lag screw while the
proximal fragment is lagged into place, but the set screws

Figure 9: Mean migration in all treatment groups was less than 2.5 mm and reached a plateau by 750 cycles.

66
do not facilitate “locking” of the lag screw relative to the
nail. Our results also indicate that the ability of each sys-
tem to lag the proximal fragment is not dependent on the
novel design features (the clip or interlocking compression
screw), however the new designs do appear to facilitate
some retention of compression under load. It is not clear
from this study what feature in the new designs enhances
maintenance of fracture compression.
Proximal fragment rotation was small for all
constructs. The amount of friction between the lag screw
and the proximal fragment was high for all specimens
based subjectively on the amount of torque necessary
to advance and compress the lag screw. Mean proximal
fragment rotation was highest in the Gamma 3 Nail group.
However, the maximum rotation in this group was 1.8°
and is still within an acceptable range for clinical practice.
The enhanced designs of the Gamma 3 RC and the Trigen
InterTAN resulted in negligible rotation in all specimens
(mean less than 0.5°), a substantial reduction in rotation
relative to the traditional design. In osteopenic bone, it is
likely that proximal fragment rotation would be greater in
all cases and that the advantages of the enhanced designs
would be more prominent.17
Rigid fixation and ample implant stiffness are
critical for minimizing interfragmentary motion and main-
taining fracture reduction under load.22 In this study, the
InterTAN nail constructs had the highest stiffness of the
three implants. The addition of the interlocking compres-
sion screw inferior to the lag screw increases the moment
of inertia of the construct and subsequently enhances its
bending stiffness.23 There was little difference in the stiff-
ness between the Gamma 3 and the Gamma 3 RC nails
indicating that the addition of the clip to the nail does not
significantly enhance bending stiffness. In all treatment
groups, the mean stiffness increased with loading cycles,
likely due to settling and compaction of the bone around
the tip of the lag screw.
The trends in migration of the three treatment
groups mirror the trends in stiffness with highest migra-
tion in the InterTAN group. This is likely because a stiffer
construct will cause compaction of bone around the lag
screw. A large mismatch in stiffness between the bone and
the implant will result in the bone deforming more than
67
the screw under load and local compaction of the bone
around the screw. Compaction of the bone results in a void
adjacent to the screw and settling.11 The more stiff the lag
screw is relative to the adjacent bone, the more relative
deformation will occur and the more compaction and
migration will result. As the bone becomes compacted,
its local stiffness increases and migration plateaus with
increasing cycles. In osteopenic bone, the effects of a mis-
match in stiffness will be increased and further migration
or cutout is more likely.
Among other factors, the likelihood of cutout in
the treatment of peritrochanteric fractures is dependent
on surgical technique.8,9 Tip to apex distance is strongly
correlated to incidence of cutout with 25 mm being a
threshold value above which cutout is more likely.9 In
this study, every attempt was made to size and place the
implants to achieve a tip to apex distance of less than 25
mm. However, a post-hoc measurement revealed that the
mean tip to apex distance in two of the three treatment
groups (G3S and G3RC) was greater than 25 mm while
the third group (ITI) was less than 25 mm. This inherently
predisposes two treatment groups to a higher likelihood
of cutout and thus is a confounding factor in this study.
However, as shown in Figure 10, a regression analysis of
all specimens correlating tip to apex distance and migration
indicated only a very weak correlation (R2=0.10) indicat-
ing that either the composite model bones or the implant
systems were insensitive to tip to apex distance within the
range of our testing.
In this study, we used composite models to
simulate bone of normal density. In osteopenic bone,
the likelihood of cutout and proximal fragment rotation
are higher and the benefit of the design features of the
nails warrants further assessment. The addition of the
interlocking compression screw in the Trigen InterTAN
implant appears to primarily increase the stiffness of the
device, whereas the addition of the clip in the Gamma 3
RC implant appears to primarily reduce proximal fragment
rotation. In this study, both implants performed well with
clinically acceptable levels of cutout and rotation. How-
ever, additional research in osteopenic bone is necessary
to comprehensively characterize the effects of the design
enhancements of these implants.
Figure 10: A regression analysis of all specimens indicates only a weak correlation (R2=0.10) between tip to apex
distance and migration.

References

1. American Academy of Orthopaedic Surgeons. The
Burden of Musculoskeletal Diseases in the United
States. AAOS: Rosemont. 2008.
2. Cummings SR, Melton LJ. Epidemiology
and outcomes of osteoporotic fractures.Lancet.
2002;359(9319):1761-7.
3. Leibson CL, Tosteson AN, Gabriel SE, Ransom JE,
Melton LJ. Mortality, disability, and nursing home use
for persons with and without hip fracture: a population-
based study. J Am Geriatr Soc. 2002;50(10):1644-50.
4. Richmond J, Aharonoff GB, Zuckerman JD, Koval
KJ. Mortality risk after hip fracture. J Orthop Trauma.
2003; 17(1):53-6.
5. Salkeld G, Cameron ID, Cumming RG, et al. Quality
of life related to fear of falling and hip fracture in older
women: a time trade off study. British Medical Journal.
2000;320(7231):341-6.
6. United States Department of Health and Human
Services. Bone health and osteoporosis: a report of
the surgeon general. Rockville, MD: U.S. Department
of Health and Human Services, Office of the Surgeon
General, 2004.
7. Rosenblum SF, Zuckerman JD, Kummer FJ, Tam
BS. A biomechanical evaluation of the gamma nail. J
Bone Joint Surg Br. 1992;74(3):352-7.

68
8. Thomas AP. Dynamic hip screws that fail. Injury.
1991;22(1):45-6.
9. Baumgaertner MR, Curtin SL, Lindskog DM, Keggi
JM. The value of tip-apex distance in predicting failure
of fixation of peritrochanteric fractures of the hip. J Bone
Joint Surg Am. 1995; 77:1058-64.
10. Haynes RC, Poll RG, Miles AW, Weston RB. Fail-
ure of femoral head fixation: a cadaveric analysis of
lag screw cut-out with the gamma locking nail and AO
dynamic hip screw. Injury. 1997; 28(5-6):337-41.
11. Sommers MB, Roth C, Hall H, et al. A laboratory
model to evaluate cutout resistance of implants for
pertrochanteric fracture fixation. J Orthop Trauma.
2004;18:361-8.
12. Heiner A, Brown T. Structural properties of a new
design of composite replicate femurs and tibias. J
Biomech. 2001;34:773-82.
13. Chong ACM, Miller F, Buxton M, Friis EA.
Fracture toughness and fatigue crack propagation rate
of short fiber reinforced epoxy composites for analogue
cortical bone. J Biomech Eng. 2007;129:487-93.
14. Chong ACM, Friis EA, Ballard GP, Czuwala PJ,
Cooke FW. Fatigue performance of composite analogue
femur constructs under high activity loading. Ann
Biomed Eng. 2007; 35(7):1196–1205.
15. Kummer FJ, Olsson O, Pearlman CA, Ceder L,
Larsson S, Koval K. Intramedullary versus extramedul-
lary fixation of subtrochanteric fractures. A biomechani-
cal study. Acta Orthop Scand. 1998;69(6):580-4.
69
16. Kubiak EN, Bong M, Park SS, Kummer F,
Egol K, Koval K. Intramedullary fixation of unstable
intertrochanteric hip fractures. One or two lag screws.
J Orthop Trauma. 2004;18:12-17.
17. Strauss E, Frank J, Lee J, Kummer FJ, Tejwani.
Helical blade versus sliding hip screw for treatment of
unstable intertrochanteric hip fractures: a biomechanical
evaluation. Injury. 2006;37:984-9.
18. Taylor SJG, Perry JS, Meswania JM, Donaldson
N, Walker PS, Cannon SR. Telemetry of forces from
proximal femoral replacements and relevance to fixa-
tion. J Biomech. 1997;30(3):225-34.
19. Davy DT, Kotzar GM, Brown, RH, et al. Telemet-
ric Force measurements across the hip after total arthro-
plasty. J Bone Joint Surg Am. 1988;70(1):45-50.
20. Graham AE, Xie SQ, Aw KC, Mukherjee S, Xu
WL. Bone–muscle interaction of the fractured femur.
J Orthop Res. 2008;26:1159–1165.
21. Kyle RF, Wright TM, Burstein AH. Biomechanical
analysis of the sliding characteristics of compression hip
screws. J Bone Joint Surg Am. 1980;62(8):1308-14.
22. Kraemer WJ, Hearn TC, Powell JN, Mohamed
N. Fixation of segmental subtrochanteric frac-
tures: a biomechanical study. Clin Orthop Rel Res.
1996;332:71-9.
23. Gere JM, Timoshenko SP. Mechanics of Materials,
3rd Edition. PWS Publishing: Boston. 1990.
 A New ACL Guide for Accurate Tunnel Placement
Jason Wong, Rachel Michos, Ryan Harting, Whitney Naslund
Advisors: R. Maxwell Alley, MD, Eric H. Ledet, Ph.D,

2010 Senior Design Showcase Finalist, Rensselaer Polytechnic Institute

The Problem: Design alternatives:

The accepted method of tunnel placement for ACL recon- • Hinge design
struction involves drilling a guide pin after locating the • Nitinol (shape memory metal)
exit point on the tibial plateau. If the original tibial guide
pin was placed too vertically, there will be a misalignment Final design concept:
as the pin is drilled through the femur. Slider design
• Marks the femoral and tibial ACL footprint simul-
Design Goals: taneously
Design a guide that allows the position of the tibial tunnel • Sliding track mechanism
to be determined using landmarks on both the tibia and the • Lever system
femur simultaneously prior to choosing the path through • Angle adjustment preserves linearity
for the guide through the tibia.

Figure 1: Concept drawing of the ACL Figure 2: Prototype of the hinged ACL guide.
guide.

Figure 3: Placement of the guide on a saw bones model.

70
 Minimally Invasive Reusable Carpal Tunnel Release System
Jonathan Chung, Matthew Connelly, Courtney Dumont, Alyssa Kowcz, Jason Taylor,
Jeffrey Teixeira, Samantha Wozniak
Advisors: David E. Quinn, MD, Richard R. Whipple, MD, Eric H. Ledet, Ph.D

2010 Senior Design Showcase Finalist, Rensselaer Polytechnic Institute

The Problem: Design alternatives:

Current endoscopic carpal tunnel release systems exhibit Three tip designs considered before the final design are
limitations including having a limited range of view, re- presented in Figure 1.
duced visualization due to fogging and difficulty separat-
ing the tenosynovium. Final Design Concept:
• Replaceable blade
Design Goals: • Enhanced field of view
• Improve the visualization throughout the procedure • Improved tenosynovium stripping capabilities
• Maintain alignment with the ring finger axis (Figures 2 and 3)
• Minimize the number of non-reusable parts
• Reduce the number of procedure steps

Figure 1: Tip design evolution prior to the final design choice.

Figure 2: The final design choice showing the enhanced tenosynovium stripping edge.

Figure 3: Profile of the final design choice.

71
 Albany Medical College
Division of Orthopaedic Surgery
Grand Rounds Presentations
Grand Rounds for the Division of Orthopaedic Surgery are presented every Wednesday (except the first Wednes-
day of the month), 7 AM to 8 AM EST (EDT), from the auditorium at the Bone and Joint Center, 1367 Washington
Avenue, Albany, NY.

Grand Rounds speakers include Division of Orthopaedic Surgery faculty and residents, and guest lecturers. Topics
are chosen to provide a broad coverage of current orthopaedic treatment as well as relevant clinical and basic science
topics.

Grand Rounds Schedule for 2010

DATE SPEAKER TOPIC

1/13/10 Brent Carlson, MD Shoulder Arthroplasty: Various Options & Indications

1/20/10 William Robb III, MD Advances in Prevention of VTE in Orthopaedics

1/27/10 Jared Roberts, MD Blounts Disease

2/17/10 Jeffrey Klauser, MD Acromioclavicular Joint Injuries

2/24/10 Ronniel Nazarian, MD History of Disc Replacement

3/10/10 Ricardo Valle, MD Treatment of 5th Metatarsal Fractures in Athletes

3/17/10 Jacqueline Vanderzanden, MD Flexor Tendon Repairs

3/24/10 Samuel Dellenbaugh, MD Upper Extremity Injuries in Little Leagers

3/31/10 Reginald Knight, MD Historic Overview & Recent Evidence for MIS Spine Surgery

4/14/10 Adam Suslak, MD Pediatric Knee Injuries

4/21/10 Christopher Palestro, MD SPECT/CT Applications in Orthopaedics

4/28/10 Paul Johnson, MD Health Care Reform

5/12/10 Scott Brotherton, MD The History of Orthopaedics

72
DATE SPEAKER TOPIC

5/19/10 Michael Flaherty, MD Ewing Sarcoma

5/26/10 Andrew Gerdeman, MD Alphabet Shoulder Soup

6/9/10 Matthew DiCaprio, MD Which Bone & Soft Tissue Tumors Can be Treated by

6/16/10 Joel Horning, MD Subtalar Dislocations

6/23/10 William Lavelle, MD Adult Lumbar Scoliosis

6/30/10 Bruce White, MD Ethical Dilemmas in Orthopaedics

73
 Extraordinary Products Matter

As pioneers in spine surgery, our

comprehensive range of products has

been the most trusted and respected in

the world for years. Yet our experience

shows that what matters goes far beyond

the technologies we produce.

DePuy Spine is more than a spine implant

company. We’re here to help you provide

the best care possible.

www.depuyspine.com

74

comprehensive surgical solutions reimbursement surgeon education patient access clinical evidence
r
NOTES
NOTES
6901C Trauma Ad_Fullpg 2/1/07 3:32 PM Page 1

Synthes is the leader in orthopaedic

internal fixation devices.

Synthes develops, manufactures, and markets the AO ASIF system of

orthopaedic implants and instruments. Our goal is to provide the most
advanced implants, biomaterials, instruments, and technologies that
meet or exceed the highest expectations in safety and quality. Our
products are designed to ensure reliable operating procedures, rapid
recovery, and a pain-free life after surgery.

Synthes has committed significant resources to assist medical schools

and the AO ASIF organization in conducting the most comprehensive
postgraduate training programs available for the orthopaedic surgeon.
With emphasis on practical work in a laboratory setting, these courses
enable the surgeon to learn how to use AO ASIF implants and
instruments for a wide range of fracture types and surgical techniques.
 At Stryker, we believe that results
speak louder than words.
Since 1941, that philosophy has made us one of the world’s leading medical
technology companies. Today, we have the most broadly based range of products
in orthopaedics and a significant presence in other medical specialties.

We measure our success by our ability to partner with respected medical

professionals — to help millions of people around the world lead more
active and more satisfying lives.

Stryker Corporation www.stryker.com