The aim of this retrospective study was to assess the efficacy of percutaneous cholecystostomy (PC) for patients with acute cholecystitis (AC) according to severity. A total of 325 patients who underwent cholecystectomy between January 2008 and October 2010 were enrolled. Patients were classified into three groups based on severity grade according to the Tokyo guidelines for AC: grade I (mild), grade II (moderate), and grade III (severe). These groups were further classified into two subgroups based on whether or not they underwent preoperative PC. A total of 184 patients were classified into the grade I group (57%), 135 patients were classified into the grade II group (42%), and five patients were classified into the grade III group (1%). In the grade I and II groups, the mean length of hospital stay was significantly shorter in the patients who did not undergo PC than in those who received PC (10.7 ± 4.4 vs. 13.7 ± 5.8, p < 0.001; 11.8 ± 6.5 vs. 16.9 ± 12.5, p = 0.003, respectively). The mean length of preoperative hospital stay was significantly shorter in the patients without PC than in those with PC in the grade I and II groups (5.8 ± 3.3 vs. 8.2 ± 4.6, p = 0.001; 6.0 ± 4.4 vs. 8.8 ± 5.2, p = 0.002). In addition, the operative time was shorter in patients without PC, especially in the grade I group (94.6 ± 36.4 vs. 107.3 ± 33.5, p = 0.034). Preoperative PC should be reserved for only selected patients with mild or moderate AC. No significant benefit of preoperative PC was identified with respect to clinical outcome or complications.