Concordance Between Bone Pathology and Bone Culture for the Diagnosis of Osteomyelitis in the Presence of Charcot Neuro-Osteoarthropathy

• Published in: The Journal of foot and ankle surgery 2018-09, Vol.57 (5), p.919-923
• Main Authors: Elmarsafi, Tammer, Kumar, Anagha, Cooper, Paul S., Steinberg, John S., Evans, Karen K., Attinger, Christopher E., Kim, Paul J.
• Format: Article
• Language: English
• Subjects: Adult; Aged; Aged, 80 and over; Arthropathy, Neurogenic - complications; Arthropathy, Neurogenic - diagnosis; Bone and Bones - microbiology; Bone and Bones - pathology; bone culture; bone pathology; Charcot; concordance; diabetes; Diabetes Complications - complications; Diabetes Complications - microbiology; Diabetes Complications - pathology; Female; histopathology; Humans; Male; Middle Aged; osteomyelitis; Osteomyelitis - complications; Osteomyelitis - diagnosis; Retrospective Studies; Sensitivity and Specificity
• ISSN: 1067-2516; 1542-2224
• DOI: 10.1053/j.jfas.2018.03.016

The diagnosis of osteomyelitis (OM) is a challenging but critical pathology to uncover in patients with concomitant Charcot neuro-osteoarthropathy (CN). The reference standard to diagnose OM is bone biopsy for histopathologic and microbiologic examination. The presence of CN, however, can have a negative effect on the accuracy of either method to identify OM. The aim of the present study was to examine the concordance between bone pathology and bone cultures in the presence of CN in the diagnosis of OM. A total of 286 patients with diabetes mellitus (DM) and CN were identified retrospectively, with 48 patients identified with OM. OM was confirmed by radiographs, magnetic resonance imaging, erythrocyte sedimentation rate, and C-reactive protein, positive probe-to-bone test results, and intraoperative inspection. Seventy matched pairs of bone pathology and cultures with complete data were compared and analyzed. Statistical analysis included concordance, positive predictive value, negative predictive value, sensitivity, specificity, and kappa coefficient. Concordance between bone pathology and bone culture was 41.4%, with agreement in 29 of 70 paired specimens. The diagnostic test accuracy of histopathologic examination to diagnose OM in CN bone in our study was 51.4%. The diagnostic test accuracy of microbiologic examination to diagnose OM in CN bone was 50%. The positive predictive value was 72.2%. The negative predictive value was 44.1%. The sensitivity was 57.8%. The specificity was 60.0%. The kappa coefficient was 0.165. The reference standard method of histopathologic and microbiologic examination of bone specimens has little concordance and can lead to inaccurate or inconclusive information. The low sensitivity and specificity demonstrated in the present study does not support the use of the current reference standard method of bone biopsy for histologic and microbiologic diagnosis of OM when CN is present. Thus, a diagnosis of OM in patients with CN should only be considered in the presence of strong clinical, laboratory, and imaging correlates.