Validation of the diagnostic criteria for segmental vitiligo.

Abstract

Segmental vitiligo has a different clinical course and prognosis as compared to nonsegmental vitiligo, which necessitates its correct diagnosis. It may be difficult to distinguish segmental vitiligo from the limited or focal types of nonsegmental vitiligo. To validate the previously proposed diagnostic criteria for segmental vitiligo. This was a cross-sectional validation study involving patients with limited vitiligo. The diagnostic criteria were used to classify vitiligo lesions as segmental or nonsegmental, and was compared with the experts' diagnosis, which was considered as the "gold standard". The study included 200 patients with 225 vitiligo lesions. As per the diagnostic criteria, 146 vitiligo lesions were classified as segmental and 79 as nonsegmental. The experts classified 147 vitiligo lesions as segmental and 39 as nonsegmental, while the diagnosis either was labeled "unsure" or could not be agreed upon for 39 lesions. As compared with the experts' opinions ("for sure" cases, n = 186), the sensitivity and specificity of the diagnostic criteria was 91.8% (95% confidence interval [CI]: 86.2%-95.7%) and 100% (95% CI: 91%-100%), respectively. The positive predictive value was 100% (95% CI: 97.3-100%), while the negative predictive value was 76.5% (95% CI: 62.5%-87.2%). There was a 93.5% agreement between the clinical criteria and experts' opinions (k = 0.83, P < 0.001). The diagnostic criteria were compared with the experts' opinion in the absence of an established diagnostic "gold standard". The proposed diagnostic criteria for segmental vitiligo performed well, and can be used in clinical practice, as well as in research settings.