LAUSR

Background Rapid diagnosis and treatment of diabetic foot osteomyelitis (DFO) could reduce the risk of amputation and death in patients with diabetic foot infection (DFI). Erythrocyte sedimentation rate (ESR) is considered the most useful serum inflammatory marker for the diagnosis of DFO. However, whether severe renal impairment (SRI) affects its diagnostic accuracy has not been reported previously. Objective To investigate the accuracy of ESR in diagnosing DFO in DFI patients with and without SRI. Methods This was a retrospective cross-sectional study. From the inpatient electronic medical record system, the investigators extracted demographic information, diagnostic information, and laboratory test results of patients with DFI who had been hospitalized in Longhua Hospital from January 1, 2016 to September 30, 2021. Logistic regression was performed to analyze the interaction between ESR and SRI with adjustment for potential confounders. The area under the curve (AUC), cutoff point, sensitivity, specificity, prevalence, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+), and negative likelihood ratio (LR−) were analyzed by receiver operating characteristic (ROC) curve analysis and VassarStats. Results A total of 364 DFI patients were included in the analysis. The logistic regression analysis results showed that elevated ESR increased the probability of diagnosing DFO (adjusted odds ratio [OR], 2.40; 95% confidence interval [CI], 1.75–3.28; adjusted P < 0.001); SRI was not associated with the diagnosis of DFO (adjusted OR, 3.20; 95% CI, 0.40–25.32; adjusted P = 0.271), but it had an obstructive effect on the diagnosis of DFO by ESR (adjusted OR, 0.48; 95% CI, 0.23–0.99; adjusted P = 0.048). ROC analysis in DFI patients without SRI revealed that the AUC of ESR to diagnose DFO was 0.76 (95% CI, 0.71–0.81), with the cutoff value of 45 mm/h (sensitivity, 67.8%; specificity, 78.0%; prevalence, 44.7%; PPV, 71.3%; NPV, 75.0%; LR+, 3.08; LR−, 0.41). In contrast, in patients with SRI, the AUC of ESR to diagnose DFO was 0.57 (95% CI, 0.40–0.75), with the cutoff value of 42 mm/h (sensitivity, 95.0%; specificity, 29.2%; prevalence, 45.5%; PPV, 52.8%; NPV, 87.5%; LR+, 1.34; LR−, 0.17). Conclusions The accuracy of ESR in diagnosing DFO in DFI patients with SRI is reduced, and it may not have clinical diagnostic value in these patients.