INTRODUCTION: Post-kala-azar dermal leishmaniasis (PKDL) is a cutaneous complication appearing after treatment of visceral leishmaniasis, and PKDL patients are considered infectious to sand flies and may therefore play a role in the transmission of VL. We estimated the risk and risk factors of PKDL in patients with past VL treatment in south-eastern Nepal. METHODS: Between February and May 2010 we traced all patients who had received VL treatment during 2000-2009 in five high-endemic districts and screened them for PKDL-like skin lesions. Suspected cases were referred to a tertiary care hospital for confirmation by parasitology (slit skin smear (SSS)) and/or histopathology. We calculated the risk of PKDL using Kaplan-Meier survival curves and exact logistic regression for risk factors. RESULTS: Out of 680 past-treated VL patients, 37(5.4%) presented active skin lesions suspect of PKDL during the survey. Thirty-three of them underwent dermatological assessment, and 16 (2.4%) were ascertained as probable (2) or confirmed (14) PKDL. Survival analysis showed a 1.4% risk of PKDL within 2 years of VL treatment. All 16 had been previously treated with sodium stibogluconate (SSG) for their VL. In 5, treatment had not been completed (</=21 injections). Skin lesions developed after a median time interval of 23 months [interquartile range (IQR) 16-40]. We found a higher PKDL rate (29.4%) in those inadequately treated compared to those who received a full SSG course (2.0%). In the logistic regression model, unsupervised treatment [odds ratio (OR) = 8.58, 95% CI 1.21-374.77], and inadequate SSG treatment for VL in the past (OR = 11.68, 95% CI 2.71-45.47) were significantly associated with PKDL. CONCLUSION: The occurrence of PKDL after VL treatment in Nepal is low compared to neighboring countries. Supervised and adequate treatment of VL seems essential to reduce the risk of PKDL development and active surveillance for PKDL is needed.