Background Laparoscopic gastrectomy (LG) is an established minimally invasive procedure for gastric cancer. However, it is controversial whether LG is useful for patients with poor physical status classified into higher classes of the American Society of Anesthesiologists physical status (ASA-PS) classification. The aim of this study was to determine the feasibility of LG in patients with ASA-PS class ≥ 3. Methods We extracted data for a total of 28,160 patients with an ASA-PS class ≥ 3 who underwent distal or total gastrectomy for gastric cancer between January 2013 and December 2017 from the National Clinical Database Japan society for gastroenterological surgery registry. We developed a propensity score model from baseline demographics and comorbidities and matched patients undergoing LG to those undergoing open gastrectomy (OG) using a 1:1 ratio. Mortality and morbidities (within 30 days and in-hospital) were compared between the 6998 matched patient pairs. Results In-hospital mortality was significantly lower in patients undergoing LG than in those undergoing OG (2.3% vs. 3.0%, p = 0.01), while the 30-day mortality was similar (1.6% vs. 1.5%). The length of hospital stay was significantly shorter in the LG group (median, 14 days vs. 17 days, p < 0.001). The LG group had a significantly lower incidence of postoperative complications in patients with any grade complication (20.3% vs. 22.5%, p = 0.002) as well as those with ≥ grade 3 complications (8.7% vs. 9.8%, p = 0.03). Conclusion LG was associated with decreased in-hospital mortality and a lower incidence of several postoperative complications when compared to OG among patients with poor physical condition.