Diagnostic Performance of an Abbreviated Screening Protocol for Osteomyelitis and Soft Tissue Infection of the Foot: A Cross Sectional Analytic Study

Document Type : Original Article

Authors

The Departments of Diagnostic & Interventional Radiology* and General Surgery**, Faculty of Medicine, Kasr Al-Ainy, Cairo University

Abstract

Abstract Background: Treatment of a foot infection can vary significantly depending on the presence of osteomyelitis or an abscess. Rapid and efficient evaluation of patients with diabetic foot ulcer is of great importance. While MRI is known to be the imaging modality of choice in these cases, it is also known to be a costly and lengthy exam. Aim of Study: The aim of our study is to compare the diagnostic performance of a rapid protocol to a full protocol in the evaluation of presence osteomyelitis and abscesses of the foot and to detect patients that profit on performing a standard protocol. Patients and Methods: This cross sectional analytic study involved 41 patients with diabetic foot ulcer (29 male, 12 female with age range from 41 to 77 years with mean ±  SD = 57.29±10.7). Clinical reports of the full protocol were used as gold standard, combined with clinical data. Information regarding the utilization of intravenous contrast, the existence of osteomyelitis , reactive osteitis, and abscessesof soft tissues were extracted from clinical reports. By the utilization of one T1 non fat-saturated and one STIR sequences, two radiologists with experience evaluated all patients for the existence of osteomyelitis, reactive osteitis, or a soft tissue abscess. Inter-rater agreement was calculated, also sensitivity, specificity and accuracy of both radiologists were tested. Results: There was an almost perfect inter-observer agree-ment for diagnosis of osteomyelitis with a kappa value of 0.951. A moderate inter-rater agreement was noted for the presence of soft tissue abscesses. Both observers showed a high sensitivity for the diagnosis of osteomyelitis (100% and 95.2% for observer 1 and 2 respectively). They both were less sensitive for diagnosing soft tissue infection (sensitivity of 80% for both observers). Conclusions: This rapid two sequenced protocol has a high reliability and significant diagnostic performance in detection of osteomyelitis, and its high sensitivity, specificity and accuracy make it beneficial in screening for cases that can profit after a full standard protocol.

Keywords