ORIGINAL ARTICLE
A Clinical Analysis of Shoulder and Hip Joint Infections
in Children
Mohan V. Belthur, MD,* Debra L. Palazzi, MD,w Jerry A. Miller, MS, PhD,z
William A. Phillips, MD,* and Jacob Weinberg, MD*
Background: Septic arthritis of the shoulder is a rare infection
in healthy children. This infection requires prompt surgical
drainage and antibiotic treatment. A delay in surgical interven-
tion can result in damage to the articular surface of the
glenohumeral joint, adjacent osteomyelitis, and possible growth
disturbance. The clinical course of septic arthritis of the
shoulder was compared with that of septic arthritis of the hip,
a more common disease in children.
Methods: We identified 9 children with infections of the
glenohumeral joint who presented to our pediatric hospital
between 2001 and 2007. The average age at presentation was 7
years (range: 7 mo to 12 y). These patients were compared with
14 selected patients treated for septic arthritis of the hip (mean
age 7 y, range: 1 to 12 y). Surgical drainage was performed by
open arthrotomy in each case. A retrospective review and
analysis of the medical records, laboratory tests, and radio-
graphs of these patients were performed.
Results: Children with shoulder infections differed significantly
(P<0.05) from patients with hip infections with regard to
temperature, white blood cell count, and erythrocyte sedimenta-
tion rate at the time of admission. The average time from the
onset of symptoms to presentation was notably longer in the
shoulder group compared with the hip group (P = 0.012).
Adjacent osteomyelitis was found in 67% of the shoulders and
36% of the hips (P = 0.214). Children suffering from septic
arthritis of the shoulder showed higher rates of repeat surgical
drainage (P = 0.056) and extended hospitalizations (P = 0.028).
The total duration of antibiotics was longer in the shoulder
group (P = 0.059).
Conclusions: Septic arthritis of the shoulder in the pediatric
population often has a delayed presentation with a more
complicated disease course than an infection of the hip. Children
with shoulder infections require a longer duration of treatment
and may experience a higher likelihood of skeletal complications.
Level of Evidence: Level III, retrospective comparative study.
From the *Pediatric Orthopaedics and Scoliosis Surgery, Texas
Children’s Hospital; wPediatric Infectious Diseases, Texas Children’s
Hospital; and zShriners Hospital for Children, Houston, TX.
None of the authors received financial support for this study.
Study conducted at Texas Children’s Hospital, Baylor College of
Medicine, Houston, TX.
Reprints: Jacob Weinberg, MD, Pediatric Orthopaedics and Scoliosis
Surgery, Texas Children’s Hospital 6701 Fannin Street, CC 670.01
Houston, TX 77030. E-mail: jweinb2000@yahoo.com.
Copyright r 2009 by Lippincott Williams & Wilkins
828 | www.pedorthopaedics.com
Key Words: septic arthritis, musculoskeletal infection, shoulder,
hip
(J Pediatr Orthop 2009;29:828–833)
S eptic arthritis most commonly affects the hip and the
knee in the pediatric population.1–6 A delay in
diagnosis and treatment can lead to local destruction of
articular cartilage, as well as adjacent osteomyelitis and
growth arrest.7–9 Long-term effects of untreated septic
arthritis of the hip include bony deformity of the femoral
head and neck, pseudarthrosis of the femoral neck, femoral
head dislocation, and limb length discrepancy.10–12 Early
diagnosis and appropriate treatment with antibiotics and
surgical drainage are important.1–3,13 With the availability
of a validated clinical prediction rule for septic arthritis of
the hip, most infections of the hip are diagnosed early and
many of the sequelae are avoided.13,14
In comparison with the hip, the glenohumeral joint is
a less commonly involved site of infection in healthy
children. The shoulder represents 4% of all joint infec-
tions.9,15,16 There are few reports in the orthopaedic
literature describing septic arthritis of the shoulder and
the results of surgical intervention.8,9,12,14–18 Septic arthritis
of the shoulder can result in the destruction of the humeral
head, adjacent osteomyelitis, deformity of the proximal
humerus, and growth arrest, leading to shortening of the
upper extremity.9,17,19,20 The purpose of this study was to
compare the clinical course of septic arthritis of the
shoulder with septic arthritis of the hip in children.
METHODS
We retrospectively reviewed our experience of
treating children with septic arthritis of the shoulder
and septic arthritis of the hip in a tertiary care pediatric
hospital between 2001 and 2007. This study was approved
by our medical school institutional review board. Inclu-
sion criteria included age greater than 6 months and the
absence of underlying medical conditions. Culture-nega-
tive or culture-positive staphylococcal and streptococcal
infections were included in this study.
Nine children with septic arthritis of the shoulder
and 14 with septic arthritis of the hip were identified. Age
and sex distributions of the 2 groups were not signifi-
cantly different. There were 5 boys and 4 girls in the
shoulder group (Table 1). There were 9 boys and 5 girls in
J Pediatr Orthop  Volume 29, Number 7, October/November 2009
J Pediatr Orthop  Volume 29, Number 7, October/November 2009 Analysis of Pediatric Shoulder and Hip Infections

TABLE 1. Clinical Characteristics of Children With Septic Arthritis of the Shoulder

Shoulder Duration of
Age Time to Drainage Days in Antibiotics,
Pt (yr+mo) M/F Presentation (d) Organism Procedures Hospital (d) (IV+Oral, wk) Comments
1 12+5 F 21 MSSA 2 45 6+52 Bilateral shoulder involvement,
septic shock, pericarditis,
pleural effusion, pathologic
fracture
2 0+8 M 8 Negative 1 13 6+26
3 9+6 F 5 MRSA 2 18 4+12
4 12+6 M 5 MRSA 2 37 8+26 Septic shock, pericarditis, pleural
effusion
5 1+1 F 8 Streptococcus 1 9 4+6
pneumonia
6 6+3 M 15 MSSA 1 8 4+4
7 11+5 F 13 MSSA 1 8 6+6
8 7+11 M 5 MRSA 4 41 6+12 Pathologic fracture
9 0+7 M 7 Negative 1 8 4+4
IV indicates intravenous; MRSA, methicillin-resistant Staphyloccocus aureus; MSSA, methicillin-sensitive Staphyloccocus aureus; Negative, no organism identified on
culture.

the hip group. The average age at presentation was
7 years (range: 1 to 12 y) in the shoulder group and 7
years (range: 7 mo to 12 y) in the hip group. Two patients
in the shoulder group had isolated right-sided joint
infection, 6 had isolated left-sided involvement, and
1 child had bilateral disease (patient 1). In the hip group,
all joint infections were unilateral: 8 children had right-
sided hip disease and 6 had left-sided disease.
Medical records were reviewed to obtain informa-
tion on patient demographics, time to presentation, and
clinical presentation. Laboratory investigations included
a review of serum inflammatory markers (white cell
count, erythrocyte sedimentation rate, C reactive protein)
and microbiology of cultures from the blood and from
operative sites were recorded. The details of treatment,
number of surgical procedures needed, duration of
hospitalization, and duration of antibiotics were evalu-
ated. Final analysis of each patient included clinical
outcomes such as pain, range of motion, and joint
stability at the last documented medical encounter.
Radiologic outcomes included evidence of osteomyelitis,
joint destruction, and pathologic fractures. The data were
collated on a Microsoft Excel spreadsheet (Microsoft,
Corporation, Redmond, Washington) and statistical
analysis was performed using SPSS version 16.0.1 (SPSS
Inc, Chicago, IL). Student t, w2, and Fisher exact tests
were used to compare the clinical course of septic arthritis
of the shoulder and the hip.
RESULTS
Findings at Presentation
The time to presentation in the shoulder group was
9.7 days and in the hip group was 3.7 days (P = 0.012)
(Table 2). The presenting features in the shoulder group
were fever, painful glenohumeral motion, and inability to
use the upper extremity. The presenting features in the hip
group were fever, pain, limp, and difficulty in bearing
weight on the affected lower limb. The mean temperature
at presentation in the shoulder group was 103.051F and in

TABLE 2. Comparison of Mean Values for Patients With Septic Arthritis of the Shoulder and Septic Arthritis of the Hip
Characteristic Shoulder Hip P
Total number of patients 9 14 —
Mean age in years at presentation (SD) 7 (3.8) 7 (5.0) 0.968
Gender (males:females) 5:4 9:5 0.505
Days to presentation 9.7 3.7 0.012
Temperature on admission (1F) 103.05 101.8 0.001
White cell count on admission (  103 cells/L) 16.04 11.79 0.041
Eryrothrocyte sedimentation rate on admission (mm/hr) 97.1 68.8 0.003
C reactive protein on admission (mg/dL) 10.2 10.9 0.306
Associated osteomyelitis (%) 67 36 0.214
Surgical procedures (range) 1.67 (1-4) 1.07 (1-2) 0.056
Hospitalization (d) (range) 20.78 (8-45) 6.78 (3-13) 0.028
Duration in weeks of oral+intravenous antibiotics 20.67 9.11 0.059
Pathologic fracture 2/9 (22%) 0/14 (0%) 0.142
Normal range: white blood cell count: 5 to 14.5  103 cells/L.
Erythrocyte sedimentation rate: 0 to 20 mm/hr.
C reactive protein: <1.0 mg/dL.

r 2009 Lippincott Williams & Wilkins www.pedorthopaedics.com | 829

Belthur et al
the hip group was 101.81F (P = 0.001). The average white
cell count at presentation was 16.04  103 cells/L in the
shoulder group and 11.79  103 cells/L in the hip group
(P = 0.041, normal: 5 to 14.5  103 cells/L). The mean
erythrocyte sedimentation rate at presentation in the
shoulder group was 97.1 mm/h (range: 67 to 115 mm/hr,
normal: 0 to 20 mm/hr) and in the hip group was
68.8 mm/hr (range: 47 to 107 mm/hr). This difference
was statistically significant (P = 0.003). There was no
significant difference between the C-reactive protein at
presentation between the groups.
Septic hips were diagnosed with hip ultrasound or
magnetic resonance imaging (MRI) in each case. All
patients with hip ultrasounds were aspirated before
arthrotomy. The shoulder infections were diagnosed with
MRI or computed tomography with contrast.
Osteomyelitis was associated in 6 of 9 shoulder
infection patients compared with 5 of 14 hips of the hip
patients. In the shoulder group, 5 children developed
osteomyelitis of the proximal humerus and 1 had an
infection of the scapula. In comparison, the hip series
showed adjacent osteomyelitis in the proximal femur in 3
patients and in the pelvis in 2 patients. MRI showed
signal changes in the adjacent bone at presentation in
patients treated for associated osteomyelitis, indicating
coexisting infection early in the treatment of the disease
for both the hip and shoulder infections.
In the shoulder group, 2 patients (22%) had
multiple joints involved, compared with single joint
infection in all children treated for septic arthritis of the
hip. Septic shock was a unique presentation in 2 patients
with shoulder infection and none in the hip group.
Similarly, 4 of 9 patients in the shoulder group presented
with simultaneous infections in other organ systems: 1
with pneumonia, 1 with otitis media, 1 with furuncles,
and 1 with a soft tissue abscess. None of the patients in
the hip group experienced other infections.
Microbiology
Definitive organism identification and sensitivities
were determined from cultures of the blood and operative
specimen in 7 patients in the shoulder group (78%) and 7
patients in the hip group (50%). In the shoulder group,
there were 3 cases of methicillin-sensitive Staphylococcus
aureus, 3 cases of methicillin-resistant S. aureus (MRSA),
and 1 case of Streptococcus pneumoniae. In the hip group,
4 patients presented with methicillin-sensitive S. aureus
infection, and 3 with MRSA infection.
Surgical Procedures
All patients in the shoulder group underwent open
drainage through an anterior deltopectoral approach.
Patients with septic arthritis of the hip underwent open
drainage through an anterior Smith-Peterson approach.
Repeat incision and drainage were required when the
patient’s condition or inflammatory markers did not
improve, fever persisted, or when significant drainage was
noted postoperatively. The shoulder patients required an
average of 1.67 surgical drainage procedures (range: 1 to
830 | www.pedorthopaedics.com
J Pediatr Orthop  Volume 29, Number 7, October/November 2009
4 procedures) to control their infections, in comparison
with the average of 1.07 procedures in the hip patients
(range: 1 to 2 procedures) (P = 0.056). Patient 8 required
4 surgical procedures to his shoulder to contain persistent
drainage (Figs. 1A, B). His incision healed with the aid of
a vacuum-assisted closure device placed superficially to
his closed wound.
Additional surgical procedures included arthro-
tomies of other joints that were concurrently infected in
2 children in the shoulder group. In this group, 1 patient
required an intercostal drain insertion for a massive
pleural effusion and a pericardiotomy for pericardial
effusion. None of the patients in the septic hip group
required additional procedures.
Antibiotics
In both groups, the pediatric infectious disease
department directed the choice and duration of anti-
biotics in patients with septic arthritis of the shoulder and
septic arthritis of the hip. Septic arthritis is typically
treated at our institution with 3 weeks of antibiotics. The
transition from intravenous to oral antibiotics is depen-
dent on the individual pediatric infectious disease
physician. Therapy is concluded, generally after around
3 weeks of therapy, when the patient is clinically
improved and the inflammatory markers are normal.
Acute osteomyelitis generally is treated with 4 to 6
weeks of antibiotic therapy with the same variation in
clinical practice and criteria for stopping therapy as
described above. Chronic osteomyelitis is treated with 4
to 6 weeks of intravenous or oral antibiotics, followed by
long-term oral antibiotics. The decision to discontinue
antibiotic treatment is based on improvement in the
patient’s clinical picture, radiographs, and normalization
of inflammatory markers.
In the shoulder group, the combined average
duration of intravenous and oral antibiotics was 20.67
weeks (range: 8 to 58 wk) versus 9.11 weeks (range: 4 to
24 wk) in the hip group (P = 0.059). A peripherally
inserted central catheter (PICC line) was used to deliver
intravenous antibiotics in each patient after discharge
from the hospital.
Hospital Stay
The average duration of hospitalization in the
shoulder group was significantly longer in the shoulder
group. Two patients in the shoulder group (patients 1 and
4) were admitted to the pediatric intensive care unit for
treatment of septic shock and multiorgan disease. Both
patients had pericarditis. Patient 4 suffered from pyelone-
phritis associated with MRSA septicemia.
Clinical Course
The average duration of follow-up for the shoulder
group was 61.7 months (range: 33 to 85 mo) and 32.5
months (range: 20 to 65 mo) for the hip group. Two
patients in the shoulder group (patients 1 and 8)
developed pathologic fractures of the proximal humerus
r 2009 Lippincott Williams & Wilkins
J Pediatr Orthop  Volume 29, Number 7, October/November 2009 Analysis of Pediatric Shoulder and Hip Infections

(Figs. 2A, B). None of the patients in the hip group

developed pathologic fractures.
At final follow-up, patients in both groups had no
evidence of active infection, as indicated by normalization
of their inflammatory markers. In the shoulder group, 8

FIGURE 2. A, Radiograph of the shoulder demonstrates a

pathological fracture through the proximal humerus physis in
a 12 years and 5 months old girl with adjacent glenohumeral
joint infection (patient 1). B, Fracture healing is evident 6
months later.

of 9 patients (89%) had good clinical and radiologic

outcomes. In the hip group all patients had good clinical
and radiologic outcomes. Patient 1 developed proximal
humeral varus with limited abduction and forward
flexion. Humeral shortening was noted at follow-up.

FIGURE 1. A, T1-weighted oblique coronal postintravenous

gadolinium fat suppressed magnetic resonance imaging
demonstrates fluid distention of the shoulder joint in a 7
years and 11 months old boy (patient 8) with extensive
adjacent soft tissue enhancement and heterogeneous intraoss-
eous enhancement of the proximal humerus. B, Postintrave-
nous gadolinium fat suppressed T1-weighted oblique coronal
magnetic resonance imaging of the shoulder performed 1
month later reveals progressive soft tissue enhancement and
worsening proximal humeral enhancement with distal exten-
sion into the diaphysis.

r 2009 Lippincott Williams & Wilkins www.pedorthopaedics.com | 831

Belthur et al J Pediatr Orthop
DISCUSSION
The incidence of septic arthritis is between 5.5 and
12 cases per 100,000 children.1,2 Boys are affected
approximately twice as often as girls and children less
than 3 years are the most frequently involved age
group.2 The most commonly affected joints are those of
the lower extremities, including hips, knees, and ankles,
which account for up to 80% of cases.1–6 Patients with
diabetes mellitus, sickle cell disease, and immunodefi-
ciency are more susceptible to septic arthritis than the
general population.2 S. aureus is the most common
causative organism.2 Other frequently identified organ-
isms include group A b-Hemolytic Streptococcus and
S. pneumoniae.1,2 Forty percent of infections are culture
negative.2
The risk factors for poor outcomes after septic
arthritis include delayed diagnosis, onset earlier than 3
months of age, metaphyseal osteomyelitis, and infection
caused by a penicillinase-producing organism.1,2,7 With
the availability of a validated clinical prediction rule for
septic arthritis of the hip and advanced imaging
modalities such as MRI scan, most infections of the hip
are diagnosed early.2,13,19 In the past, 10 to 40% of
patients with hip involvement have been reported to
suffer significant sequelae, such as growth plate damage
and loss of hip function.1,2 However, more recent studies
have reported the incidence of long-term sequelae of
appropriately treated septic arthritis of the hip to be
around 3% to 5%.21
In developed countries, septic arthritis of the upper
extremities occurs less commonly than in the lower limbs,
with the elbow being the most commonly involved
joint.1–6 The shoulder joint is involved in about 4% of
cases.9 Risk factors for poor outcomes after septic
arthritis of the shoulder include younger age at onset of
infection, delayed diagnosis, nonoperative management,
and increased virulence of the organism.9 The incidence
of long-term severe sequelae has been reported to be
around 30% in the presence of the above risk factors.9
The proximal humeral growth plate accounts for 80% of
the length of the upper limb and hence growth plate arrest
can occur and result in humeral head deformity, short-
ening, and inferior subluxation.9
This is the first study comparing the clinical
characteristics of children with septic arthritis of the
shoulder with those of children with septic arthritis of the
hip, a more commonly treated condition. In our study,
children with septic arthritis of the shoulder present later
than those with septic arthritis of the hip. This may be
related to poor awareness of this condition in caregivers
and medical personnel who rarely see this infection. In
addition, the existence of a validated clinical practice
guideline for septic arthritis of the hip may account for a
higher index of suspicion and the earlier diagnosis of hip
joint infections. In addition, as children are active and
ambulate most of the day, loss of function in the lower
limb is more readily noticeable by a parent or caregiver
than upper extremity pathology, especially in the younger
patient.
832 | www.pedorthopaedics.com
 Volume 29, Number 7, October/November 2009
We found that shoulder joint infection was asso-
ciated with septic shock in 2 of our 9 patients and none of
the patients in the hip joint group. This may be related to
the delay in the diagnosis of the shoulder infection or
the presence of more virulent strains of S. aureus. The
presence of other organ system involvement in these 2
patients also could be related to the above factors.
In the shoulder group, 3 of 9 patients were infected
with MRSA. In the hip group, 3 of 14 patients had
MRSA infections. All 3 patients of the shoulder group
with MRSA required repeat surgery. These patients could
have harbored a more virulent strain of S. aureus
responsible for a more aggressive infection.22 MRSA
osteomyelitis is associated with more severe infection and
higher inflammatory markers.23 Nonetheless, our num-
bers are too small to make any conclusions regarding
MRSA septic arthritis of the shoulder.
The 2 pathologic fractures in our shoulder series
also indicates more severe infection. This complication
has not been previously reported before in association
with septic arthritis of the shoulder.
Children with septic arthritis of the shoulder re-
quired a longer duration of antibiotics and had a longer
duration of hospitalization than those with septic arthritis
of the hip. Chronic osteomyelitis was more common
in the shoulder group, requiring a longer duration of
antibiotics. Simultaneous infections in 4 of 9 patients in
the shoulder group may have explained in part the higher
fevers and inflammatory markers. These factors suggest
more severe infection, possibly related to a delay in
diagnosis, resulting in a more well-established infection,
or to the possible presence of more virulent organisms in
the shoulder group.
The primary limitation of our study is its small
sample size, which may not represent a complete
description of the disease in the general population. The
retrospective nature of our study may have affected our
data collection. Further studies with a larger number of
patients and multicenter involvement are recommended.
A prospective study design with a long follow-up period
would be desirable to assess long-term sequlae, such as
limb length discrepancy and deformity related to growth
plate arrest, factors that may not be clinically obvious at
short-term follow-up.
Clinicians should have a high index of suspicion
for diagnosing infections of the shoulder joint. Clinical
and laboratory criteria for diagnosing shoulder and hip
joint infections and a high index of suspicion can lead to
an earlier diagnosis of a shoulder joint infection and
possibly prevent skeletal complications and long-term
sequelae.
ACKNOWLEDGMENT
The authors thank Sherri B. Birchansky, MD
(Department of Radiology, Texas Children’s Hospital,
Houston, TX) for her assistance in the preparation of the
figures.
r 2009 Lippincott Williams & Wilkins
J Pediatr Orthop  Volume 29, Number 7, October/November 2009
REFERENCES
1. Frank G, Mahoney HM, Eppes SC. Musculoskeletal infections in
children. Pediatr Clin North Am. 2005;52:1083–1106.
2. Gutierrez K. Bone and joint infections in children. Pediatr Clin
North Am. 2005;52:779–794.
3. McCarthy JJ, Dormans JP, Kozin SH, et al. Musculoskeletal
infections in children: basic treatment principles and recent advance-
ments. Instr Course Lect. 2005;54:515–528.
4. Nade S. Acute septic arthritis in infancy and childhood. J Bone Joint
Surg Br. 1983;65:234–241.
5. Shaw BA, Kasser JR. Acute septic arthritis in infancy and
childhood. Clin Orthop. 1990;257:212–225.
6. Wilson NIL, DiPaola M. Acute septic arthritis in infancy and
childhood: 10 years’ experience. J Bone Joint Surg Br. 1986;68:
584–587.
7. Fabry G, Meire E. Septic arthritis of the hip in children: poor results
after late and inadequate treatment. J Pediatr Orthop. 1983;3:
461–465.
8. Kawashima A, Toyama Y, Mikasa M. Septic arthritis of the
shoulder in infants and children: a long-term follow-up. Shoulder
Joint. 1987;11:174–178.
9. Saisu T, Kawashima A, Kamegaya M, et al. Humeral shortening
and inferior subluxation as sequelae of septic arthritis of the
shoulder in neonates and infants. J Bone Joint Surg Am. 2007;89:
1784–1793.
10. Choi IH, Pizzutillo PD, Bowen JR, et al. Sequelae and reconstruc-
tion after septic arthritis of the hip in infants. J Bone Joint Surg Am.
1990;72:1150–1165.
11. Choi IH, Shin YW, Chung CY, et al. Surgical treatment of the
severe sequelae of infantile septic arthritis of the hip. Clin Orthop
Relat Res. 2005;434:102–109.
12. Dick HM, Tietjen R. Humeral lengthening for septic neonatal
growth arrest: case report. J Bone Joint Surg Am. 1978;60:
1138–1139.
r 2009 Lippincott Williams & Wilkins
Analysis of Pediatric Shoulder and Hip Infections
13. Kocher MS, Lee B, Dolan M, et al. Pediatric orthopedic infections:
early detection and treatment. Pediatr Ann. 2006;35:112–122.
14. Lossos IS, Yossepowitch O, Kandel L, et al. Septic arthritis of the
glenohumeral joint. A report of 11 cases and review of the literature.
Medicine (Baltimore). 1998;77:177–187.
15. Lejman T, Strong M, Michno P, et al. Septic arthritis of the shoulder
during the first 18 of life. J Pediatr Orthop. 1995;15:172–175.
16. Schmidt D, Mubarak S, Gelberman R. Septic shoulders in children.
J Pediatr Orthop. 1981;1:67–72.
17. Bos CF, Mol LJ, Obermann WR, et al. Late sequelae of neonatal
septic arthritis of the shoulder. J Bone Joint Surg Br. 1998;80:
645–650.
18. Smith SP, Thyoka M, Lavy CB, et al. Septic arthritis of the shoulder
in children in Malawi. A randomized, prospective study of
aspiration versus arthrotomy and washout. J Bone Joint Surg Br.
2002;84:1167–1172.
19. Kocher MS, Mandiga R, Murphy JM, et al. A clinical practice
guideline for treatment of septic arthritis in children: efficacy in
improving process of care and effect on outcome of septic arthritis of
the hip. J Bone Joint Surg Am. 2003;85:994–999.
20. Ugwonali OF, Bae DS, Waters PM. Corrective osteotomy for
humeral varus. J Pediatr Orthop. 2007;27:529–532.
21. Christiansen P, Frederiksen B, Glazowski J, et al. Epidemiologic,
bacteriologic, and long-term follow-up data of children with acute
hematogenous osteomyelitis and septic arthritis: a ten-year review.
J Pediatr Orthop B. 1999;8:302–305.
22. Martı́nez-Aguilar G, Avalos-Mishaan A, Hulten K, et al. Commu-
nity-acquired, methicillin-resistant and methicillin-susceptible Sta-
phylococcus aureus musculoskeletal infections in children. Pediatr
Infect Dis J. 2004;23:701–706.
23. Bocchini CE, Hulten KG, Mason EO Jr, et al. Panton-Valentine
leukocidin genes are associated with enhanced inflammatory
response and local disease in acute hematogenous Staphylococcus
aureus osteomyelitis in children. Pediatrics. 2006;117:433–440.
www.pedorthopaedics.com | 833