See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/7288971

Using Prealbumin as an Inflammatory Marker

for Patients With Deep Space Infections of
Odontogenic Origin

Article in Journal of Oral and Maxillofacial Surgery April 2006

Impact Factor: 1.43 DOI: 10.1016/j.joms.2005.11.008 Source: PubMed

CITATIONS READS

6 106

3 authors, including:

Larry L Cunningham Matthew J Madsen

University of Kentucky University of Louisville
59 PUBLICATIONS 529 CITATIONS 15 PUBLICATIONS 157 CITATIONS

SEE PROFILE SEE PROFILE

Available from: Larry L Cunningham

Retrieved on: 14 May 2016
 BASIC AND PATIENT-ORIENTED RESEARCH
J Oral Maxillofac Surg
64:375-378, 2006

Using Prealbumin as an Inflammatory

Marker for Patients With Deep Space
Infections of Odontogenic Origin
Larry L. Cunningham, Jr, MD, DDS,* Matthew J. Madsen, BS,
and Joseph E. Van Sickels, DDS

Background: The general health of the host may contribute to the severity of odontogenic infections.
An example is the link between protein malnutrition and host defense mechanisms, most notably
cell-mediated and mucosal immune responses. The plasma concentration of prealbumin, a protein
synthesized in the liver with a half-life of 1.8 days, is an indicator of malnutrition.
Materials and Methods: We retrospectively reviewed the medical records of all patients who were
admitted to our hospital with a diagnosis of odontogenic infection in 2002 and 2003. We recorded
prealbumin concentrations, number of infected spaces (ie, severity of infection), hematocrit, blood urea
nitrogen (BUN) concentration, and the results of general blood studies.
Results: The median age of the 22 patients included in this study was 33 years (range, 17 to 57 years);
59% were men. The median prealbumin concentration was 12.85 (range, 5.7 to 27.5); the accepted
normal level is 19 or higher (81.8% of concentrations were below normal). A Mann-Whitney rank sum
test found a statistically significant relationship between prealbumin concentrations and length of
hospital stay (P .038). Logistic regression analysis showed that length of hospital stay was positively
related to the number of infected spaces and the BUN concentration and negatively related to prealbumin
concentrations. These factors could accurately predict the length of hospital stay for 77% of patients
(Fisher exact test; P .026).
Conclusions: Prealbumin levels are significantly lower than normal among patients admitted for
treatment of severe odontogenic infection. There appears to be a relationship between length of stay and
lower prealbumin concentrations.
2006 American Association of Oral and Maxillofacial Surgeons
J Oral Maxillofac Surg 64:375-378, 2006

Low plasma concentrations of proteins, such as albu-
min, transferrin, C-reactive protein, and retinol-binding
proteins, act as nutritional indicators during a period of
infection and after recovery from the infection.1,2 Albu-
min concentrations are traditionally measured as an es-
timate of nutritional status; however, the disadvantage
Received from Division of Oral and Maxillofacial Surgery, Univer-
sity of Kentucky College of Dentistry, Lexington, KY.
*Assistant Professor.
Dental Student.
Professor and Chair.
Address correspondence and reprint requests to Dr Cunning-
ham: Division of Oral and Maxillofacial Surgery, University of Ken-
tucky College of Dentistry, 800 Rose Street, D-508, Lexington, KY
40536-0297; e-mail: llcunn2@email.uky.edu
2006 American Association of Oral and Maxillofacial Surgeons
0278-2391/06/6403-0004$32.00/0
doi:10.1016/j.joms.2005.11.008
of this measure is the long half-life of albumin (18 to 20
days).2 Prealbumin, also known as transthyretin, is a
54-kD protein synthesized primarily by the liver. Its main
function is the transport of thyroxine and, indirectly,
vitamin A. Because the half-life of prealbumin is much
shorter (1.8 days) than that of albumin (20 days) and
transferrin (8 days), it may provide a more timely indi-
cation of plasma protein levels.3 Plasma prealbumin con-
centrations in adults generally range from 14 to 36 mg/
dL.4 Concentrations of 16 to 25 mg/dL indicate
borderline malnutrition, whereas concentrations below
16 mg/dL indicate definite malnutrition.4,5 Prealbumin
concentrations are lower among patients with infection
and among those with burns.6 Because of the short
half-life of prealbumin, its concentrations can fluctuate
as the result of changes in diet and can serve as an
indicator of malnutrition or fasting. Protein malnutrition
is linked to impaired host defense mechanisms, most
notably cell-mediated immunity.

375
376 PREALBUMIN AS MARKER FOR INFECTIONS OF ODONTOGENIC ORIGIN
Early recognition of protein-energy malnutrition
(PEM) is useful in preventing sepsis, morbidity, and
mortality; prealbumin has been used as an indicator of
PEM.7 Severe PEM has been shown to lengthen healing
time after infection, decrease immunocompetence, and
increase postoperative complications.8 Mullen et al9 re-
ported that normal nutritional and immunologic indices
are found among only 3% of patients undergoing sur-
gery. Using prealbumin as an indicator of PEM is very
useful because prealbumin is the first protein known to
exhibit alterations in plasma concentrations among pa-
tients who are acutely malnourished.10 Because of its
short half-life, prealbumin has been shown to be the
most sensitive indicator of malnutrition; its plasma con-
centration will return to normal as soon as nutritional
reposition occurs.11
Materials and Methods
In 2002, our unit began following prealbumin lev-
els on patients admitted with odontogenic infections.
We retrospectively reviewed the medical records of
all patients admitted to our hospital in 2002 and 2003
with a diagnosis of odontogenic infection. Informa-
tion collected from the charts included prealbumin
concentration, number of infected spaces, hemato-
crit, blood urea nitrogen (BUN) concentration, and
results of a complete metabolic panel (CMP). Surgical
treatment of patients admitted to the hospital in-
cluded incision and drainage and tooth extraction.
These data were available for all patients.
Results
The study included the records of 22 consecutive
patients with a median age of 33 years (range, 17 to
57 years); 59% were men. Three causes of infection
were determined. For 20 patients, oral examination
detected gross carious decay; radiographs indicated
advanced periapical infections of 1 or more teeth.
One patient had a unilateral submandibular infection
as the result of a fracture of the left mandibular body.
The other patient had unilateral infections of the left
submandibular, buccal, and submental spaces after
tooth extractions.
The severity of infection was determined by the
number of fascial spaces involved. The patients had
an average of 2.4 infected spaces, including lateral
pharyngeal, submandibular, temporal, buccal, sub-
mental, masseteric, masticator, and sublingual spaces
(Table 1).
The following blood values were recorded: concen-
trations of prealbumin, glucose, blood urea nitrogen
(BUN), sodium, potassium, chloride, carbon dioxide,
and calcium; anion gap; and white blood cell (WBC)
count. The median WBC for all patients at the time of
Table 1. INFECTED DEEP FASCIAL SPACES
OF STUDY SUBJECTS
Patient Spaces Infected
1 Lateral pharyngeal, peritonsillar
2 Lateral pharyngeal, submandibular, submental,
masticator
3 Left submandibular
4 Right temporal
5 Left buccal, submandibular, submental
6 Left masseteric
7 Right submandibular, lateral pharyngeal
8 Right submandibular, submental, buccal
9 Right masticator, submandibular
10 Left lateral pharyngeal
11 Right lateral pharyngeal, submandibular
12 Bilateral submandibular, sublingual,
submental, right lateral pharyngeal
13 Right submandibular
14 Left submandibular, pterygomasseteric
15 Left medial pterygoid, masticator
16 Right submental, submandibular, lateral
pharyngeal
17 Left submandibular, lateral pharyngeal
18 Left masticator, left submandibular
19 Right submandibular, sublingual
20 Right submandibular, sublingual, submental
21 Right submandibular, sublingual, submental,
retropharyngeal
22 Right buccal, submandibular, lateral
pharyngeal
Cunningham, Madsen, and Van Sickels. Prealbumin as Marker
for Infections of Odontogenic Origin. J Oral Maxillofac Surg
2006.
hospital admission was 15.5; the mean WBC for the 7
female patients was 17.4, whereas the mean for the 15
male patients was 14.7 (Table 2). There was no statisti-
cally significant relationship between length of hospital
stay and concentrations of glucose, sodium, potassium,
chloride, carbon dioxide, or the anion gap.
As shown in Table 2, the median prealbumin con-
centration was 12.85 mg/dL (range, 5.727.5 mg/dL).
The accepted normal concentration is 19 mg/dL or
higher (81.8% of results were below normal). In the
hematology laboratory of the University of Kentucky
Chandler Medical Center, the accepted normal con-
centration is 19 to 42 mg/dL. A Mann-Whitney rank
sum test found a statistically significant relationship
between length of hospitalization and prealbumin
concentrations (P .038). For this study, a stay of 5
days or less was considered normal; a stay of more
than 5 days was considered extended, and the reason
for the extension was presumed to be related to
difficulty in managing the disease and stabilizing the
patients condition. A logistic regression analysis
found that the number of infected spaces and the
BUN concentration were positively related to length
of stay and that prealbumin concentrations were neg-
CUNNINGHAM, MADSEN, AND VAN SICKELS 377

Table 2. RESULTS OF LABORATORY ANALYSIS IN STUDY SUBJECTS

Prealbumin No. of Infected

Patient Concentration (mg/dL) Spaces Hematocrit BUN (mg/dL) WBC (k/L)

1 9.6 2 38.9 4 16
2 11.2 4 36.7 5 14.2
3 13.1 1 43.7 7 11.8
4 23.0 1 39.4 19 10.8
5 14.9 3 39.6 14 11
6 13.9 1 38.2 17 14.2
7 12.2 2 31 19.5
8 11.2 3 31.6 10 24
9 9.9 2 41.8 8 18.4
10 11.1 1 38.3 4 12.7
11 8.0 2 42 6 5.7
12 5.7 7 30.3 1 21
13 21.9 1 42.7 11 19.3
14 22.4 2 34.3 8 20.5
15 10.5 2 44.3 10 20.2
16 12.6 3 41.2 9 17.2
17 18.9 2 47.5 7 21.4
18 17.6 2 7.7
19 13.4 2 44.8 14 16.9
20 11.5 3 33.8 15 8.2
21 27.5 4 13 14.9
22 17.3 3 43.7 4
Cunningham, Madsen, and Van Sickels. Prealbumin as Marker for Infections of Odontogenic Origin. J Oral Maxillofac Surg 2006.

atively related to length of stay. The levels of these
three factors could accurately predict the length of
hospital stay for 77% of patients (the Fisher exact test;
P .026).
Discussion
In 2001, the American Association of Oral and Max-
illofacial Surgeons published Parameters and Path-
ways: Clinical Practice Guidelines for Oral and Max-
illofacial Surgery.12 This document contained a
section devoted to odontogenic infection; the risk
factors associated with such infections are listed in
Table 3. In this study, each patient hospitalized for
surgical incision and drainage followed by antibiotic
therapy had 1 or more of these risk factors. The
patients history and the review of systems indicated
that 7 of the 22 patients had a systemic illness such as
diabetes. Statistical analysis showed no correlation
between prealbumin levels and the presence of sys-
temic illness in patients with a severe odontogenic
infection (Table 4).
Plasma prealbumin concentration, BUN concentra-
tion (BUN is a breakdown product of protein metabo-
lism), and the number of fascial planes involved in the
infection were significantly related to length of hospital
stay. In this study, the prealbumin concentrations of
patients admitted for treatment of severe odontogenic
infection were significantly lower than the concentra-
tions considered to be clinically normal. Low concentra-
tions of prealbumin seemed to result from the anorexia
and malnutrition associated with pain and trismus, in
addition to the systemic effects of infection.
BUN concentrations higher than normal in these
patients may be explained by increased protein me-
tabolism as the result of a hypermetabolic state. In
addition, dehydration and starvation can lead to ele-
vated BUN concentrations, which would indicate that
infections are more severe for these patients than for
Table 3. RISK FACTORS ASSOCIATED WITH
ODONTOGENIC INFECTION*
Presence of coexisting major systemic disease (eg, disease
that increases a patients ASA classification)
Extent of infection (eg, localized, diffuse)
Direction and/or rate of extension of infection
Presence of impending airway obstruction
Susceptibility of organism to antibiotics
Virulence of organism
Presence of generalized periodontitis
Presence of inadequate oral hygiene
Presence of dental crowding or malocclusion
Proximity to contiguous structures
Presence of foreign bodies or implanted materials
Dental management objectives that are altered and/or
adversely affected by therapy
*Adapted from Parameters and Pathways: Clinical Practice
Guidelines for Oral and Maxillofacial Surgery (AAOMS, 2001).12
Cunningham, Madsen, and Van Sickels. Prealbumin as Marker
for Infections of Odontogenic Origin. J Oral Maxillofac Surg
2006.
378 PREALBUMIN AS MARKER FOR INFECTIONS OF ODONTOGENIC ORIGIN

Table 4. RELATIONSHIP BETWEEN PREALBUMIN LEVELS AND LENGTH OF HOSPITAL STAY

Patient Prealbumin Concentration (mg/dL) Days Hospitalized Medical History

1 9.6 6 NIDM, pancreatitis

2 11.2 10 Noncontributory
3 13.1 3 Noncontributory
4 23.0 6 IDDM, HTN, MI
5 14.9 4 NIDM
6 13.9 3 Noncontributory
7 12.2 9 Advanced MS, IDDM
8 11.2 8 Noncontributory
9 9.9 3 Noncontributory
10 11.1 6 Noncontributory
11 8.0 7 Obesity
12 5.7 12 IDDM, hyperglycemia
13 21.9 3 Noncontributory
14 22.4 5 Noncontributory
15 10.5 6 Noncontributory
16 12.6 6 Noncontributory
17 18.9 4 Noncontributory
18 17.6 3 Obesity
19 13.4 10 Noncontributory
20 11.5 7 Noncontributory
21 27.5 21 COPD, CHF, HTN, IDDM, obesity, OSA
22 17.3 3 Obesity, osteoarthritis
Abbreviations: NIDM, noninsulin diabetes mellitus; IDDM, insulin-dependent diabetes mellitus; HTN, hypertension; MI, myocardial
infarction; COPD, chronic obstructive pulmonary disease; CHF, coronary heart failure; OSA, obstructive sleep apnea.
Cunningham, Madsen, and Van Sickels. Prealbumin as Marker for Infections of Odontogenic Origin. J Oral Maxillofac Surg 2006.

others because of the length of time during which
they have been unable to eat or drink. A high BUN
concentration may also indicate how long patients
have waited before seeking treatment, but this study
was not designed to detect such an indication.
The number of fascial spaces involved in an odon-
togenic infection is an indication of the severity of
infection. In general, the more fascial planes involved,
the longer the patients hospital stay will be (Table 4).
The findings of this retrospective analysis indicate
that prealbumin, a measure of nutritional status, in com-
bination with the BUN concentration and the number of
infected spaces, can predict length of hospital stay for
patients with odontogenic infections. Future studies will
attempt to determine whether providing additional nu-
trition to these patients may speed recovery time and
thereby shorten their hospital stay.
Acknowledgments
This research was funded in part by resources and support from
the General Clinical Research Center (GCRC), University of Ken-
tucky, NIH/NCRR grant M01 RR02602. The authors thank Jeff
Ebersole, PhD, for assistance with statistical analysis and Flo Witte
for editorial assistance.
References
1. Ziccardi VB, Ochs MW, Braun TW: Indications for enteric tube
feedings in oral and maxillofacial surgery. J Oral Maxillofac
Surg 51:1250, 1993
2. Hassanein el-S A, Assem HM, Rezk MM, et al: Study of plasma
albumin, transferrin, and fibronectin in children with mild to
moderate protein-energy malnutrition. J Trop Pediatr 44:362,
1998
3. Malave I, Vethencourt MA, Pirela M, et al: Serum levels of
thyroxine-binding prealbumin, C-reactive protein and interleu-
kin-6 in protein-energy undernourished children and normal
controls without or with associated clinical infections. J Trop
Pediatr 44:256, 1998
4. Sanchez-Lastres J, Eiris-Punal J, Otero-Cepeda JL, et al: Nutri-
tional status of mentally retarded children in northwest Spain.
II. Biochemical indicators. Acta Paediatr 92:928, 2003
5. dos Santos Junqueira JC, Cotrim Soares E, Rodrigues Correa
Filho H, et al: Nutritional risk factors for postoperative compli-
cations in Brazilian elderly patients undergoing major elective
surgery. Nutrition 19:321, 2003
6. Gottschlich MM, Baumer T, Jenkins M, et al: The prognostic
value of nutritional and inflammatory indices in patients with
burns. J Burn Care Rehabil 13:105, 1992
7. Chertow GM, Ackert K, Lew NL, et al: Prealbumin is as impor-
tant as albumin in the nutritional assessment of hemodialysis
patients. Kidney Int 58:2512, 2000
8. Haupt W, Hohenberger W, Mueller R, et al: Association be-
tween preoperative acute phase response and postoperative
complications. Eur J Surg 163:39, 1997
9. Mullen JL, Buzby GP, Waldman MT, et al: Prediction of opera-
tive morbidity and mortality by preoperative nutritional assess-
ment. Surg Forum 30:80, 1979
10. Bernstein LH: Relationship of nutritional markers to length of
hospital stay. Nutrition 11:205, 1995 (suppl 2)
11. Sachs E, Bernstein LH: Protein markers of nutrition status as
related to sex and age. Clin Chem 32:339, 1986
12. AAOMS. Parameters and Pathways: Clinical Practice Guidelines
for Oral and Maxillofacial Surgery (AAOMS ParPath 01), Ver-
sion 3.0. J Oral Maxillofac Surg 59: (suppl) 2001