A total of 1109 patients were included in this study, with 179 patients in the SIDG group and 930 patients in the MLDG group. After 1:2 PSM, the SIDG group consisted of 179 patients, and the MLDG group consisted of 358 patients. The standardized mean difference for all matching variables was less than 0.1 after PSM, indicating successful matching (Fig. 2).

The covariate balance before and after 1:2 propensity score matching between single-incision distal gastrectomy (SIDG) and multiport laparoscopic distal gastrectomy (MLDG) groups

Table 1 shows the clinicopathologic characteristics of the SIDG and MLDG groups before and after 1:2 PSM. Before PSM, the MLDG group had a higher proportion of male patients, greater height and weight, and higher BMIs than the SIDG group. However, after 1:2 PSM, there was no difference in age, sex, height, weight, BMI, or ASA classification between the two groups. Before matching, the year of surgery was statistically significantly different between the two groups (P < 0.001). In the case of SIDG, which was introduced relatively recently, the number of cases inevitably increased over time. Therefore, the year of surgery was included as a matching variable to mitigate bias from changes in the surgical technique and perioperative management. After PSM, the year of surgery was well balanced between the two groups. (P = 0.96).

No significant differences in pathological outcomes were found between the two groups before and after PSM, including tumor size and proximal and distal resection margins. The number of retrieved lymph nodes did not significantly differ between the SIDG and MLDG groups, either before PSM (SIDG: 52.8 ± 19.3 vs. MLDG: 54.6 ± 21.9, P = 0.26) or after (SIDG: 52.8 ± 19.3 vs. MLDG: 53.9 ± 21.0, P = 0.56). Similarly, the number of positive lymph nodes was not significantly different between the SIDG and MLDG groups before (SIDG: 0.4 ± 1.2 vs. MLDG: 0.3 ± 1.1, P = 0.19) or after PSM (SIDG: 0.4 ± 1.2 vs. MLDG: 0.4 ± 1.2, P = 0.98). No significant differences in lymphatic invasion, venous invasion, and perineural invasion were found between the SIDG and MLDG groups before or after PSM. Before PSM, the SIDG group had more cases of pathologic stage II (P = 0.003) due to a higher proportion of advanced N stages than the MLDG group. However, there were no significant differences in pathologic stages between the groups after PSM (P = 0.45).

Table 2 presents the surgical outcomes and postoperative course of the patients in the SIDG and MLDG groups before and after 1:2 PSM. The mean operation time was significantly shorter in the SIDG group compared to the MLDG group both before (170.8 ± 60.0 min vs. 190.6 ± 53.9 min, P < 0.001) and after PSM (170.8 ± 60.0 min vs. 186.1 ± 52.6 min, P = 0.004). In the subgroup analysis (Table 3, Supplementary Table 2), a significant difference in operation time was observed between SIDG and MLDG groups within the BMI 25–30 kg/m2 category, with SIDG demonstrating a shorter operation time (BMI 25–27.5 kg/m2 group: SIDG, 168.6 ± 64.8 vs. MLDG, 182.6 ± 53.6, P = 0.047; BMI 27.5–30 kg/m2 group: SIDG, 171.9 ± 52.5 vs. MLDG, 195.4 ± 52.0, P = 0.008). However, no significant difference in operation time was found in the subgroup with a BMI of 30 kg/m2 or higher between the SIDG and MLDG groups before (SIDG group: 191.1 ± 39.4, MLDG group: 192.3 ± 53.1, P = 0.95) or after PSM (SIDG group: 191.1 ± 39.4, MLDG group: 183.4 ± 43.8, P = 0.64). No significant differences in the type of anastomosis were found between the SIDG and MLDG groups. Roux-en-Y anastomosis was the most frequently used method in both the SIDG (54.7%) and MLDG groups (53.6%), followed by Billroth-I and Billroth-II. Before matching, the SIDG group had a shorter postoperative hospital stay compared to the MLDG group (SIDG: 5.9 ± 3.4 days vs. MLDG: 6.6 ± 5.2 days, P = 0.03). However, after PSM, a trend toward shorter postoperative hospital stays in the SIDG group compared to the MLDG group was seen, but there was no statistical difference (SIDG: 5.9 ± 3.4 days vs. MLDG: 6.3 ± 5.1 days, P = 0.23). This trend was maintained in the BMI subgroup analysis (Supplementary Table 2).

As shown in Table 4, there was no significant difference in the early postoperative complication rate between the two groups before (SIDG group: 13.4%, MLDG group: 11.8%, P = 0.53) or after PSM (SIDG group: 13.4%, MLDG group: 12.8%, P = 0.89). Similarly, no significant differences were seen in the CCI between the two groups before (SIDG group: 2.6 ± 7.7 vs. MLDG group: 2.3 ± 7.0, P = 0.61) or after PSM (SIDG group: 2.6 ± 7.7 vs. MLDG group: 2.5 ± 7.2, P = 0.88). The distribution of individuals with Clavien-Dindo classification grade IIIa or higher did not differ between the two groups (SIDG group: 3.9% vs. MLDG group: 3.0%, P = 0.49). No cases of early postoperative mortality occurred in either group, and the incidence of local complications was not different after PSM (SIDG group: 6.7%, MLDG group: 6.1%, P = 0.74). The most common local complication was motility disorder in both groups (SIDG group: 2.2%, MLDG group: 2.2%). In the SIDG group, the most common local complication was anastomosis stricture in the BMI 25–27.5 kg/m2 subgroup (2.7%), motility disorder in the BMI 27.5–30 kg/m2 subgroup (3.4%), and fluid collection in the BMI ≥ 30 kg/m2 subgroup (11.1%) (Supplementary Table 3). No differences were seen in systemic complications between the two groups (SIDG group: 6.1%, MLDG group: 6.7%, P = 0.86). Pulmonary complications were the most common systemic complication in the two groups. (SIDG group: 6.1%, MLDG group: 5.6%). This trend was maintained in the subgroup analysis according to BMI (Supplementary Table 3).

The median costs of different categories were compared between the SIDG and MLDG groups. For total hospital cost, the SIDG group had a median cost of $7556 (IQR 6879, 8457), while the MLDG group had a median cost of $7601 (IQR 6988, 8320) (P = 0.46). In terms of operation and procedure cost, the SIDG group had a median cost of $1754 (IQR 1708, 1809), while the MLDG group had a median cost of $1777 (IQR 1700, 1820) (P = 0.08). Regarding treatment material cost, the SIDG group had a median cost of $2621 (IQR 2295, 2891), while the MLDG group had a significantly higher median cost of $2774 (IQR 2439, 3209) (P < 0.001) (Fig. 3). Data from the Korean Health Insurance Review and Assessment Service indicate that the indirect cost for the assistant surgeon’s workload for distal gastrectomy with LND is approximately 279,280 KRW (202 USD).

Comparison of operation and procedure cost, and treatment material cost between single-incision distal gastrectomy (SIDG) and multiport laparoscopic distal gastrectomy (MLDG) groups

In the subgroup analysis (Supplementary Table 3) of individuals with a BMI between 25 and 27.5 kg/m2, a difference in the occurrence of anastomosis strictures was found between the SIDG group (three cases, 2.7%) and the MLDG group (0 cases, 0.0%) after 1:2 PSM (P = 0.03). Specifically, within the SIDG group, three cases of anastomosis strictures occurred: a 78-year-old female, who underwent Billroth I anastomosis, developed an anastomosis stricture resulting in aspiration pneumonia requiring intensive care unit management (Clavien-Dindo grade IVa). This patient underwent conversion surgery to gastrojejunostomy after conservative management failed. A 60-year-old male, who had Billroth I anastomosis, experienced an anastomosis stricture and was treated with balloon dilatation (Clavien-Dindo grade IIIa). Lastly, a 47-year-old female, who underwent Roux-en-Y gastrojejunostomy, was discharged following conservative management for the anastomosis stricture (Clavien-Dindo grade II).

The results of the subgroup analysis after 1:2 PSM by sex are shown in Supplementary Table 4. The operation time was shorter in the SIDG group than in the MLDG group in both males and females (males; SIDG group: 176.4 ± 63.5 vs. MLDG group: 191.4 ± 51.8, P = 0.049; females: SIDG group: 164.7 ± 55.6 vs. MLDG group: 179.8 ± 53.1, P = 0.04) (Supplementary Table 5). The hospital length of stay after surgery tended to be shorter in the SIDG group than in the MLDG group in both males and females, but there was no statistical difference (males: SIDG group: 5.7 ± 2.0 vs. MLDG group: 6.4 ± 5.0, P = 0.12; females: SIDG group: 6.1 ± 4.5 vs. MLDG group: 6.3 ± 5.3, P = 0.73). No differences were seen in complication rates, CCI values, local complication rates, and systemic complication rates between males and females in the SIDG and MLDG groups (Supplementary Table 6).