ORIGINAL ARTICLES

Acute Phase Reactants Predict the Risk of Amputation in

Diabetic Foot Infection
Baris Akinci, MD*
Serkan Yener, MD*
Sena Yesil, MD*
Nur Yapar, MD†
Yasin Kucukyavas, MD‡
Firat Bayraktar, MD*

Background: Prediction of amputation would aid clinicians in the management of

diabetic foot infections. We aimed to assess the predictive value of baseline and post-
treatment levels of acute phase reactants in the outcome of patients with diabetic foot
infections.
Methods: We collected data prospectively during minimum follow-up of 6 months in
patients with infected diabetic foot ulcers hospitalized in Dokuz Eylul University Hospital
between January 1, 2003, and January 1, 2008. After excluding patients who did not
attend the hospital for follow-up visits regularly (n = 36), we analyzed data from 165 foot
ulcer episodes.
Results: Limb ischemia and osteomyelitis were much more frequent in patients who
underwent amputation. Wagner grade, which assesses ulcer depth and the presence of
osteomyelitis or gangrene, was higher in patients who needed amputation. Ulcer size
was slightly larger in the amputation group. Baseline and post-treatment C-reactive
protein levels, erythrocyte sedimentation rates, white blood cell counts, and platelet
counts were significantly elevated in patients who underwent amputation. Albumin
levels were significantly suppressed in the amputation group. Univariate analysis
showed that a 1-SD increase in baseline and post-treatment C-reactive protein levels,
erythrocyte sedimentation rates, and white blood cell counts and a 1-SD decrease in
post-treatment albumin levels were significantly associated with increased risk of
amputation. Post-treatment C-reactive protein level was strongly associated with
amputation risk.
Conclusions: Circulating levels of acute phase reactants were associated with
amputation risk in diabetic foot infections. (J Am Podiatr Med Assoc 101(1): 1-6, 2011)

Diabetic patients with long-term, inadequately
controlled blood glucose levels are at significant
risk for diabetic foot ulcers, a major reason for
lower-extremity amputations.1 Standard manage-
*Division of Endocrinology and Metabolism, Department of
Internal Medicine, Dokuz Eylul University Medical School,
Izmir, Turkey.
†Department of Infection Diseases, Dokuz Eylul University
Medical School, Izmir, Turkey.
‡Department of General Internal Medicine, Dokuz Eylul
University Medical School, Izmir, Turkey.
Corresponding author: Baris Akinci, MD, Division of
Endocrinology and Metabolism, Department of Internal
Medicine, Dokuz Eylul University Medical School, Inciralti,
Izmir, Turkey 35340. (E-mail: baris.akinci@deu.edu.tr)
ment of diabetic foot ulcers includes evaluation of
vascular status, identification of infection and
osteomyelitis, antibiotic therapy, surgical debride-
ment, and metabolic control of diabetes.2, 3 Patients
whose ulcers fail to heal after standard treatment
may undergo amputation. Despite well-defined risk
factors for diabetic foot ulcer development, little is
known about which factors predict amputation in a
diabetic foot ulcer episode. Previous studies4-8 have
shown that limb ischemia, ulcer depth, and osteo-
myelitis are important predictors of amputation.
Ulcer classification by several systems was also
found to predict the risk of amputation.9-11 Addi-
tional factors that have been proposed to be

Journal of the American Podiatric Medical Association  Vol 101  No 1  January/February 2011 1
associated with amputation risk include older age
and macrovascular and microvascular comorbidi-
ties.1, 8, 9, 12, 13
Levels of acute phase reactants alter in response
to infection, tissue injury, and inflammation.14 Acute
phase reactants, primarily erythrocyte sedimenta-
tion rate, C-reactive protein level, and white blood
cell count, are commonly used in routine clinical
practice when there is a suspicion of infection.14, 15
However, these measures are not specific to
infection, and the values may be elevated owing to
noninfectious conditions such as ischemia.16 These
measures should be considered markers of inflam-
mation that rise in the presence of systemic
inflammation.17 Altered levels of acute phase
reactants have been proposed to be useful in
indicating disease activity in patients with inflam-
matory disorders and may be predictive of either
functional outcome or mortality.18 The aim of the
present study was to assess the predictive value of
baseline and post-treatment levels of acute phase
reactants in the outcome of patients with diabetic
foot infections.
Materials and Methods
The study population was composed of patients
with infected diabetic foot ulcers hospitalized in
Dokuz Eylul University Hospital between January 1,
2003, and January 1, 2008. Data were collected
prospectively during minimum follow-up of 6
months. After patients who did not attend the
hospital for follow-up visits regularly (n = 36) were
excluded, data from 165 foot ulcer episodes were
analyzed. The procedures were approved by the
institutional review board of Dokuz Eylul University.
Characteristics of patients, including diabetic
complications, smoking habits, and physical exam-
ination findings, were recorded. At baseline, the
ulcer was photographed. The site and the largest
diameter of the ulcer were noted. The depth of the
ulcer was determined by inspection, with additional
use of a sterile probe if indicated. Foot lesions were
classified according to the Wagner classification as
follows: grade 0, risk of foot ulcer; grade 1,
ulcerated skin and subcutaneous tissue; grade 2,
deeper lesions may penetrate to tendon, bone, or
joint capsule, without abscess or osteomyelitis;
grade 3, deep tissues are involved, and abscess,
osteitis, or osteomyelitis is present; grade 4, local
gangrene; and grade 5, diffuse gangrene.
Standard radiographs were taken. Magnetic res-
onance imaging of the extremity was performed
according to consensus in weekly diabetic foot
2 January/February 2011  Vol 101  No 1  Journal of the American Podiatric Medical Association
team meetings. Baseline hemoglobin A1c level was
recorded. Arterial circulation was evaluated by
palpation of the peripheral pulses and ankle
brachial index with a handheld Doppler. Patients
with absent or reduced pedal pulses or an ankle
brachial index less than 0.9 underwent conventional
Doppler examination. Patients with vascular insuf-
ficiency were evaluated by the vascular surgeon,
and a revascularization procedure was performed if
indicated. Conventional or magnetic resonance
angiography was performed in selected patients.
Symptoms of neuropathy were questioned. All of
the patients were tested for neuropathy using the
10-g monofilament test. Loss of vibration perception
was evaluated with a biothesiometer on the pulp of
the hallux. Further neurologic assessments were
performed when required.
Standard treatment included wound care, bed
rest, proper off-loading, parenteral antibiotics, and
debridement. Wound debridement was performed
routinely to remove extensive callus and necrotic
tissue. Infected diabetic foot ulcer was defined
according to the Infectious Diseases Society of
America guidelines as the presence of purulent
wound drainage or at least three designated
systemic or local inflammatory findings. Samples
were obtained for culture by deep-needle aspiration,
bone biopsy, or curettage of the ulcer. In patients
with infected diabetic foot ulcers, antibiotics were
given according to the decision of the infectious
diseases specialist. After obtaining culture speci-
mens, empirical parenteral treatment was started;
change in the antimicrobial regimen was guided by
culture results and clinical follow-up. Parenteral
treatment was followed by prolonged oral therapy.
Levels of acute phase reactants were obtained
first at admission and then 1 week after standard
treatment. Erythrocyte sedimentation rate was
analyzed with the Sedimatic 100 method. White
blood cell count and platelet count were measured
with an automatic analyzer (LH 780; Beckman
Coulter, Krefeld, Germany). Serum albumin level
was measured spectrophotometrically with the
Abbott Architect c16000 system (Abbott Diagnos-
tics, Wiesbaden-Delkenheim, Germany). Serum
highly sensitive C-reactive protein level was mea-
sured by an autoanalyzer, using a particle-enhanced
turbidimetric assay (Cobas Integra 400; Roche
Diagnostics, Indianapolis, Indiana). The sensitivity
of C-reactive protein was 0.11 mg/L. The intra-assay
and interassay coefficients of variation were 1.34
and 5.70, respectively.
Logistic regression was used to estimate the
independent effect of each selected variable on the
outcome. The association between prognostic var- Table 1. Comparison of Baseline Characteristics of
iables and amputation rate was evaluated by Patients Who Underwent Amputation and Those Who Did
calculating the odds ratios and their corresponding Not Require Amputationa
95% confidence intervals. The t test for independent Amputation No Amputation
samples, after correction for equality of variance, (n = 70) (n = 95)
was used to compare patient variables. Differences
Age (y)b 62.76 6 9.98 58.32 6 11.51
in proportions were compared with the v2 test.
Male sex (No. [%]) 46 (65.7) 63 (66.3)
Receiver operating characteristic curves were gen-
Type 2 diabetes (No. [%]) 68 (97.1) 90 (94.7)
erated to determine the predictability of levels of
acute phase reactants for amputations. Sensitivity, Diabetes duration (y) 15.44 6 9.34 14.67 6 8.65
specificity, and positive and negative predictive Previous insulin use (No. [%]) 48 (68.6) 67 (70.5)
values for different cutoff levels of C-reactive Smoking (No. [%]) 24 (34.3) 32 (33.7)
protein were calculated. Analyses were conducted BMI 25.84 6 3.65 26.78 6 4.42
with statistical software (SPSS version 11.0; SPSS Retinopathy (No. [%]) 41 (58.6) 62 (65.3)
Inc, Chicago, Illinois). Values are given as mean 6 Nephropathy (No. [%]) 34 (48.6) 50 (52.6)
SD. Tests of significance were 2-tailed. A P , .05 Neuropathy (No. [%]) 53 (75.7) 83 (87.4)
was considered statistically significant. Limb ischemia (No. [%])b 54 (77.1) 36 (37.9)
Osteomyelitis (No. [%])b 51 (72.9) 38 (40.0)
Results Ulcer size (cm) 5.81 6 4.03 5.26 6 3.88
Site of ulcer (No. [%])b
Seventy patients underwent amputation (20 toe Toe 38 (54.3) 36 (37.9)
amputations, 21 ray amputations, ten transmetatar- Forefoot 24 (34.3) 20 (21.1)
sal amputations, one Syme’s amputation, 17 below- Midfoot 3 (4.3) 15 (15.8)
the-knee amputations, and one above-the-knee Hindfoot 5 (7.1) 15 (15.8)
amputation). Patients who underwent amputation Leg 0 9 (9.5)
were older. There was no significant difference b
Wagner score (No. [%])
between patients who underwent amputation and
Grade 1 0 6 (6.3)
those who did not in terms of sex, type of diabetes,
Grade 2 3 (4.3) 43 (45.3)
diabetes duration, previous insulin use, smoking,
Grade 3 29 (41.4) 39 (41.1)
body mass index, and microvascular complications
of diabetes. Baseline hemoglobin A1c levels were Grade 4 34 (48.6) 6 (6.3)
similar. More people had ischemia and osteomyeli- Grade 5 4 (5.7) 1 (1.1)
tis in the amputation group. Patients who under- Hemoglobin A1c (%) 9.68 6 2.78 9.36 6 2.44
went amputation had a slightly increased ulcer size; Baseline CRP (mg/dL)b 127.99 6 86.92 58.26 6 75.87
however, it was not statistically significant. Site of Post-treatment CRP 95.8 6 83.61 28.95 6 42.64
ulcers and Wagner scores are given in Table 1. (mg/dL)b
Baseline and post-treatment C-reactive protein Baseline ESR (mm/h)b 71.06 6 27.04 56.56 6 28.21
levels, erythrocyte sedimentation rates, white blood Post-treatment ESR 70.68 6 29.12 55.34 6 29.39
cell counts, and platelet counts were significantly (mm/h)b
elevated in patients who underwent amputation. Baseline WBC (cells/lL)b 13.55 6 4.91 10.38 6 3.66
Albumin levels were significantly suppressed in the Post-treatment WBC 11.89 6 4.18 8.97 6 2.61
amputation group (Table 1). (cells/lL)b
Clinical and laboratory predictors of amputation Baseline PLT (cells/lL)b 369.64 6 107.22 316.3 6 127.01
were evaluated with univariate analysis (Table 2). Post-treatment PLT 392.11 6 138.99 326.74 6 143.9
Limb ischemia; osteomyelitis; presence of gangrene; (cells/lL)b
ulcer depth; a 1-SD increase in baseline and post- Baseline albumin (g/dL)b 3.54 6 0.57 3.88 6 0.64
treatment C-reactive protein levels, erythrocyte Post-treatment albumin 3.29 6 0.6 3.76 6 0.55
sedimentation rates, and white blood cell counts; (g/dL)b
and a 1-SD decrease in levels of post-treatment Abbreviations: BMI, body mass index (calculated as weight
albumin were found to be significantly associated in kilograms divided by the square of the height in meters); CRP,
with increased risk of amputations. C-reactive protein; ESR, erythrocyte sedimentation rate; PLT,
platelet count; WBC, white blood cell count.
Receiver operating characteristic curves were a
Data are given as mean 6 SD except where indicated
generated to evaluate the relationship between otherwise.
levels of acute phase reactants and amputations. b
Amputation versus no amputation, P , .05.

Journal of the American Podiatric Medical Association  Vol 101  No 1  January/February 2011 3
Table 2. Clinical and Laboratory Factors Predicting Table 3. Baseline and Post-treatment Levels of Acute
Amputation Phase Reactants in the Prediction of Amputationa
OR (95% CI)a P Value Area 95% CI

Age 2.318 (0.972–5.528) .058 AUCBaseline CRP 0.754 0.678–0.830

Smoking 1.027 (0.535–1.971) .936 AUCPost-treatment CRP 0.809 0.744–0.874
Limb ischemia 5.531 (2.760–11.083) ,.001 AUCBaseline ESR 0.641 0.557–0.726
Osteomyelitis 4.026 (2.065–7.851) ,.001 AUCPost-treatment ESR 0.649 0.563–0.735
Ulcer diameter 1.833 (0.710–4.732) .210 AUCBaseline WBC 0.690 0.605–0.774
Gangrene (Wagner 14.924 (6.056–36.778) ,.001 AUCPost-treatment WBC 0.713 0.632–0.794
grades 4 and 5) AUCBaseline PLT 0.646 0.562–0.729
Ulcer depth (Wagner 12.137 (3.441–42.812) ,.001 AUCPost-treatment PLT 0.662 0.577–0.746
grade 3 versus AUCBaseline b
0.661 0.577–0.745
albumin
grades 1 and 2) b
AUCPost-treatment albumin 0.724 0.641–0.807
Baseline CRP 3.428 (1.485–7.916) .004
Post-treatment CRP 5.933 (2.236–15.744) ,.001 Abbreviations: AUC, area under the curve; CI, confidence
interval; CRP, C-reactive protein; ESR, erythrocyte sedimen-
Baseline ESR 2.760 (1.268–6.008) .011
tation rate; PLT, platelet count; WBC, white blood cell count.
Post-treatment ESR 2.300 (1.099–4.815) .027 a
Data are expressed as AUC of the corresponding receiver
Baseline WBC 4.676 (2.001–10.926) ,.001 operating characteristic curve.
b
Post-treatment WBC 8.599 (2.781–26.581) ,.001 The receiver operating characteristic curve is generated
regarding suppressed albumin levels.
Baseline PLT 1.424 (0.579–3.500) .441
Post-treatment PLT 1.333 (0.522–3.407) .548
Baseline albumin 1.924 (0.835–4.419)b .124
Post-treatment albumin 4.343 (1.683–11.203)b .002

Abbreviations: CI, confidence interval; CRP, C-reactive

protein; ESR, erythrocyte sedimentation rate; OR, odds ratio;
PLT, platelet count; WBC, white blood cell count.
a
For continuous parameters, the ORs were standardized to
express the risk associated with a 1-SD increase.
b
The OR for serum albumin level was standardized to
express the risk associated with a 1-SD decrease.

Post-treatment levels of acute phase reactants were

more closely associated with amputations accord-
ing to area under the curve values, which were
obtained from receiver operating characteristic
curves (Table 3). There was a strong relationship
between post-treatment C-reactive protein level and
amputation (area under the curve, 0.809; 95%
confidence interval, 0.744–0.874). Multivariate anal-
ysis showed that post-treatment C-reactive protein
level was an independent predictor of amputation
Figure 1. Receiver operating characteristic curves
when the data were controlled for age, sex, showing serum levels of post-treatment C-reactive
presence of ischemia, and osteomyelitis (a 1-SD protein in the prediction of amputations.
increase in post-treatment C-reactive protein level;
model r2, 0.269; odds ratio, 4.445; 95% confidence
Discussion
interval, 1.532–12.9; P = .006). Potential cutoff
values of post-treatment C-reactive protein were These results suggest that levels of acute phase
determined for prediction of amputations (Fig. 1). reactants, which were obtained first at admission
Sensitivity, specificity, and positive and negative and then 1 week after management of the diabetic
predictive values for different cutoff levels of post- foot infection, were associated with amputation
treatment C-reactive protein are given in Table 4. risk. According to receiver operating characteristic

4 January/February 2011  Vol 101  No 1  Journal of the American Podiatric Medical Association
Table 4. Analysis of Different Cutoff Values of Post-treatment CRP in the Prediction of Amputation
Sensitivity (%)
Post-treatment CRP 30 mg/dL 68.57
Post-treatment CRP 50 mg/dL 58.57
Post-treatment CRP 90 mg/dL 41.42
Abbreviations: CRP, C-reactive protein; NPV, negative predictive value; PPV, positive predictive value.
curves, post-treatment levels of acute phase reac-
tants were more closely associated with outcome.
Post-treatment C-reactive protein levels were
strongly related to amputation risk.
Circulating levels of acute phase reactants are
affected by the presence of infection, tissue injury,
and inflammation.17 Levels of acute phase reactants
in diabetic foot ulcers mostly alter in response to
superficial and deep tissue infections, osteomyelitis,
and limb ischemia.14, 16 Several studies have report-
ed that baseline levels of acute phase reactants are
associated with the outcome of the diabetic foot
ulcer. In one study,19 elevated C-reactive protein
levels were found to be strongly predictive of major
amputation in long-standing diabetic patients with
ischemic foot lesions. A prospective trial conducted
by Lipsky et al20 showed that baseline white blood
cell counts, C-reactive protein levels, erythrocyte
sedimentation rates, and albumin levels were
related to clinical treatment failure in diabetic foot
infections treated with broad spectrum antibiotics.
Low serum albumin level was also reported to be
associated with increased amputation risk.6 In-
creased baseline white blood cell counts were
reported to be associated with worse clinical
outcomes in diabetic foot ulcers.20, 21 A baseline
white blood cell count greater than 12.0 cells/lL has
been proposed to be associated with increased risk
of amputation.22 Pittet et al23 showed that neutro-
phil count was an independent predictor of treat-
ment failure.
On the other hand, Armstrong et al24 found that
elevated white blood cell count was a poor indicator
of acute osteomyelitis, although there was a
significant relationship between osteomyelitis and
elevated erythrocyte sedimentation rate. Elevated
erythrocyte sedimentation rate has been proposed
to be useful in the diagnosis of osteomyelitis when
combined with clinical data. An elevated erythro-
cyte sedimentation rate of more than 70 mm/h has
been reported to increase the probability of
osteomyelitis 11 times.25 It has been found that an
elevated erythrocyte sedimentation rate of more
than 70 mm/h predicted the presence of osteomy-
elitis with a sensitivity of 89.5% and a specificity of
Journal of the American Podiatric Medical Association  Vol 101  No 1  January/February 2011
Specificity (%) PPV (%) NPV (%)
72.63 64.86 75.82
82.10 70.68 72.89
93.68 82.85 68.46
100%.26 However, increased C-reactive protein
levels were reported in hematogenous osteomyelitis
in children, and these levels decreased faster than
erythrocyte sedimentation rates after appropriate
treatment, reflecting the effectiveness of the therapy
more sensitively than erythrocyte sedimentation
rate.27 The present results also suggest that C-
reactive protein levels obtained early after starting
standard treatment for the infected diabetic foot
ulcer are strongly correlated with the outcome.
Although univariate analysis revealed a more
elevated odds ratio of a 1-SD increase in post-
treatment white blood cell count for predicting
amputation risk, receiver operating characteristic
curve analysis suggested that post-treatment C-
reactive protein level was a better indicator of
amputation risk. On the other hand, erythrocyte
sedimentation rates obtained early after treatment
were similar to those taken at admission, probably
owing to its relatively long halftime.
In conclusion, we showed that circulating levels
of acute phase reactants were associated with
amputation risk in diabetic foot infections. Promi-
nent acute phase response after treatment seemed
more likely to be associated with amputation than
did baseline levels of acute phase reactants. Post-
treatment C-reactive protein level was a strong
predictor of treatment failure and amputation risk in
patients with infected diabetic foot ulcers. We
suggest that increased circulating levels of acute
phase reactants reflect the presence of inflammation
that occurs in response to tissue injury, superficial
and deep tissue infections, osteomyelitis, limb
ischemia, and gangrene, and they should be
considered a marker for the underlying abnormality
causing amputation.
Financial Disclosure: None reported.
Conflict of Interest: None reported.
References
1. JEFFCOATE WJ, HARDING KG: Diabetic foot ulcers. Lancet
361: 1545, 2003.
2. AMERICAN DIABETES ASSOCIATION. Consensus Development
Conference on Diabetic Foot Wound Care: 7–8 April
5
 1999, Boston, Massachusetts. Diabetes Care 22: 1354,
1999.
3. LIPSKY BA, BERENDT AR, DEERY HG, ET AL: Diagnosis and
treatment of diabetic foot infections. Clin Infect Dis 39:
885, 2004.
4. REIBER GE, PECORARO RE, KOEPSELL TD: Risk factors for
amputation in patients with diabetes mellitus: a case-
control study. Ann Intern Med 117: 97, 1992.
5. MAYFIELD JA, REIBER GE, NELSON RG, ET AL: A foot risk
classification system to predict diabetic amputation in
Pima Indians. Diabetes Care 19: 704, 1996.
6. FLORES RIVERA AR: Risk factors for amputation in
diabetic patients: a case-control study. Arch Med Res
29: 179, 1998.
7. TREECE KA, MACFARLANE RM, POUND N, ET AL: Validation of
a system of foot ulcer classification in diabetes mellitus.
Diabet Med 21: 987, 2004.
8. WINKLEY K, STAHL D, CHALDER T, ET AL: Risk factors
associated with adverse outcomes in a population-based
prospective cohort study of people with their first
diabetic foot ulcer. J Diabetes Complications 21: 341,
2007.
9. OYIBO SO, JUDE EB, TARAWNEH I, ET AL: A comparison of
two diabetic foot ulcer classification systems: the
Wagner and the University of Texas wound classifica-
tion systems. Diabetes Care 24: 84, 2001.
10. CALHOUN JH, CANTRELL J, COBOS J, ET AL: Treatment of
diabetic foot infections: Wagner classification, therapy,
and outcome. Foot Ankle 9: 101, 1988.
11. ARMSTRONG DG, LAVERY LA, HARKLESS LB: Validation of a
diabetic wound classification system: the contribution
of depth, infection, and ischemia to risk of amputation.
Diabetes Care 21: 855, 1998.
12. MOST RS, SINNOCK P: The epidemiology of lower
extremity amputations in diabetic individuals. Diabetes
Care 6: 87, 1983.
13. FAGLIA E, FAVALES F, MORABITO A: New ulceration, new
major amputation, and survival rates in diabetic
subjects hospitalized for foot ulceration from 1990 to
1993: a 6.5-year follow-up. Diabetes Care 24: 78, 2001.
14. JOHNSON HL, CHIOU CC, CHO CT: Applications of acute
phase reactants in infectious diseases. J Microbiol
Immunol Infect 32: 73, 1999.
6 January/February 2011  Vol 101  No 1  Journal of the American Podiatric Medical Association
15. BLACK S, KUSHNER I, SAMOLS D: C-reactive protein. J Biol
Chem 279: 48487, 2004.
16. CASSAR K, BACHOO P, FORD I, ET AL: Markers of coagulation
activation, endothelial stimulation and inflammation in
patients with peripheral arterial disease. Eur J Vasc
Endovasc Surg 29: 171, 2005.
17. COLTEN HR: Tissue-specific regulation of inflammation. J
Appl Physiol 72: 1, 1992.
18. THOMPSON D, WHICHER JT, BANKS RE: Acute phase
reactants in predicting disease outcome. Baillieres Clin
Rheumatol 6: 393, 1992.
19. VOLACO A, CHANTELAU E, RICHTER B, ET AL: Outcome of
critical foot ischaemia in longstanding diabetic patients:
a retrospective cohort study in a specialised tertiary
care centre. Vasa 33: 36, 2004.
20. LIPSKY BA, SHEEHAN P, ARMSTRONG DG, ET AL: Clinical
predictors of treatment failure for diabetic foot infec-
tions: data from a prospective trial. Int Wound J 4: 30,
2007.
21. AKANJI AO, FAMUYIWA OO, ADETUYIBI A: Factors influencing
the outcome of treatment of foot lesions in Nigerian
patients with diabetes mellitus. Q J Med 73: 1005, 1989.
22. ENEROTH M, APELQVIST J, STENSTROM A: Clinical character-
istics and outcome in 223 diabetic patients with deep
foot infections. Foot Ankle Int 18: 716, 1997.
23. PITTET D, WYSSA B, HERTER-CLAVEL C, ET AL: Outcome of
diabetic foot infections treated conservatively: a retro-
spective cohort study with long-term follow-up. Arch
Intern Med 159: 851, 1999.
24. ARMSTRONG DG, LAVERY LA, SARIAYA M, ET AL: Leukocytosis
is a poor indicator of acute osteomyelitis of the foot in
diabetes mellitus. J Foot Ankle Surg 35: 280, 1996.
25. BUTALIA S, PALDA VA, SARGEANT RJ, ET AL: Does this patient
with diabetes have osteomyelitis of the lower extrem-
ity? JAMA 299: 806, 2008.
26. KALETA JL, FLEISCHLI JW, REILLY CH: The diagnosis of
osteomyelitis in diabetes using erythrocyte sedimenta-
tion rate: a pilot study. JAPMA 91: 445, 2001.
27. ROINE I, FAINGEZICHT I, ARGUEDAS A, ET AL: Serial serum C-
reactive protein to monitor recovery from acute
hematogenous osteomyelitis in children. Pediatr Infect
Dis J 14: 40, 1995.