Interobserver and intraobserver reproducibility of plain X-rays in the diagnosis of diabetic foot osteomyelitis

The purpose of this study was to analyze the interobserver and intraobserver variability in plain radiography in the diagnosis of diabetic foot osteomyelitis. A prospective observational study was performed from October 1, 2009, to July 31, 2011, on patients with diabetic foot ulcers, with clinically suspected osteomyelitis who were admitted to the Diabetic Foot Unit of the Complutense University of Madrid. Two professional groups examined 123 plain X-rays, each group comprising 3 different levels of clinical experience. To analyze intraobserver variability, 2 months later plain X-rays were reanalyzed by one of the clinical groups. When using only plain radiography for the diagnosis of osteomyelitis in the diabetic foot, low concordance rates were observed for clinicians with a similar level of experience: experienced clinicians (K(11AB) = .35, P < .001), moderately experienced clinicians (K(22AB) = .39, P < .001), and inexperienced clinicians (K(33AB) = .40, P < .001). Intraobserver agreement was highest in experienced clinicians (K(11A) = .75, P < .001), followed by moderately experienced clinicians (K(22A) = .61, P < .001) and inexperienced clinicians (K(33A) = .57, P < .001). Plain radiography for the diagnosis of diabetic foot osteomyelitis is operator dependent and shows low association strength, even among experienced clinicians, when interpreted in isolation without knowing the clinical characteristics of the lesion.