Performance of the probe-to-bone test in a population suspected of having osteomyelitis of the foot in diabetes

Abstract

Background: We investigated the validity of probe-to-bone testing in the diagnosis of osteomyelitis in a selected subgroup of patients clinically suspected of having diabetic foot osteomyelitis.

Methods: Between January 1, 2007, and December 31, 2008, inpatients and outpatients with a diabetic foot ulcer were prospectively evaluated, and those having a clinical diagnosis of foot infection and at least one of the osteomyelitis clinical suspicion criteria were consecutively included in this study.

Results: Sixty-five patients met the inclusion criteria and were prospectively enrolled in the study. Forty-nine patients (75.4%) were hospitalized, and the remaining 16 (24.6%) were followed as outpatients. Osteomyelitis was diagnosed in 39 patients (60.0%). Probe-to-bone test results were positive in 30 patients (46.1%). The positive predictive value for the probe-to-bone test was fairly high (87%), but the negative predictive value was only 62%. The sensitivity and specificity of the test were 66% and 84%, respectively. White blood cell counts and mean C-reactive protein levels did not statistically significantly differ between groups. However, erythrocyte sedimentation rates greater than 70 mm/h reached statistical significance between groups. Wound area and depth were not found to be statistically significantly different between groups.

Conclusions: Positive probe-to-bone test results and erythrocyte sedimentation rates greater than 70 mm/h provide some support for the diagnosis of diabetic foot osteomyelitis, but it is not strong; magnetic resonance imaging or bone biopsy will probably be required in cases of doubt.