The primary objective of this study was to determine the prognostic significance of computed tomography (CT)-determined sarcopenia in small-cell lung cancer (SCLC) patients. This retrospective study consisted of a total of 149 consecutive SCLC patients. The cross-sectional area of muscle at the level of the third lumbar vertebra (L3) was measured using baseline CT images. Sarcopenia was defined as a L3 muscle index of less than 55 cm/m for men and of less than 39 cm/m for women as proposed by international consensus of cancer cachexia. In addition, Korean-specific cutoffs for sarcopenia was also applied (49 cm/m for men and 31 cm/m for women). Overall survival (OS) and clinical characteristics of patients with or without sarcopenia were compared. Mean patient age was 68.6 ± 9.5 years. Most were male (85.3%) and 67.8% had extensive disease at time of diagnosis. Sarcopenia was present in 118 patients (79.2%) and was significantly related to an advanced age (p = 0.028), male sex (p < 0.001), lower body mass index (p < 0.001), and poor performance status (p = 0.049). Sarcopenic patients had shorter OS than nonsarcopenic patients (median: 8.6 months versus 16.8 months; p = 0.031). Multivariable analysis revealed that sarcopenia was an independent prognostic factor of poor survival (hazards ratio: 1.68; 95% confidence interval: 1.04-2.72; p = 0.034), along with extensive stage (p < 0.001), supportive care only (p < 0.001), and an elevated lactate dehydrogenase level (p = 0.020). Using Korean sarcopenia cutoffs, sarcopenic patients were also found to have poorer OS than nonsarcopenic patients, however, the survival difference was not statistically significant (median: 8.4 months versus 12.7 months; p = 0.144 by the log-rank test). Sarcopenia as determined by CT could be used to predict prognosis in patients with SCLC. Optimum reference values to predict cancer-specific outcomes should be tailored by further studies.