Colorectal cancer (CRC) is one of the most common cancer-causing mortalities in the world [12]. The current prevailing treatment of CRC is radical surgery followed by adjuvant therapy [3]. Historically, open laparotomy was the standard surgical procedure in CRC surgery; however, with the introduction of laparoscopy, the proportion of laparoscopic colorectal surgeries has increased [4]. Some prospective randomized studies of laparoscopic surgical treatment for colon cancer have shown superior short-term outcomes such as reduced hospital stay, better cosmetic effect, and less pain, as well as comparable long-term outcomes to those of open surgical treatment [5678]. Consequently, laparoscopic colon cancer surgery has been established as a standard surgical treatment [91011].

Recently, along with advancements in surgical skills and devices, several surgeons have reported the feasibility of single-port laparoscopic surgery (SPS) in colon cancer [1213141516]. SPS has a cosmetic benefit compared with conventional multiport surgery (MPS) [1213141516]; however, single-port laparoscopic colectomy requires a longer operation time, highly experienced surgeons, and advanced surgical techniques owing to its technical challenges, including collision of laparoscopic instruments and the limitation of triangular tissue traction [1216].

Reduced-port laparoscopic surgery (RPS), which involves the insertion of an additional port in SPS, was introduced to overcome these challenges. The additional port in RPS may provide several advantages over SPS, including the avoidance of internal and external collisions between instruments and the ability to achieve efficient traction [17]. Although several previous studies have reported the short-term outcomes of RPS for colon cancer [1718192021], the feasibility of RPS for colon cancer remains unclear.

Therefore, we designed this study to evaluate the feasibility of RPS for colon cancer by comparing the short-term outcomes of RPS and MPS.

This study was approved by the Institutional Review Board of our institution. A waiver of informed consent was requested, and approval was obtained.

A total of 184 patients who underwent AR and 118 patients who underwent RHC were included. Seventy-three patients (39.7%) in the AR group and 23 (19.5%) in the RHC group underwent RPS. The clinicopathologic characteristics of the included patients are shown in Table 1. In general, baseline characteristics were similar between the RPS and MPS groups; however, patients who underwent RPS had earlier T stage in the AR group (P = 0.038), and less previous abdominal operation history in the RHC group (P = 0.034).

To reduce selection bias, we performed propensity score matching analysis, and the clinicopathologic characteristics of the matched cohorts are shown in Table 2. After matching, no significant differences between RPS and MPS in both the AR and RHC groups were found.

Table 3 presents the short-term outcomes according to surgical procedures after propensity score matching. In both the AR and RHC groups, there was no difference between the number of harvested lymph nodes (AR: 33.3 ± 14.5 vs. 31.1 ± 17.0, P = 0.409; RHC: 50.8 ± 11.9 vs. 48.7 ± 17.5, P = 0.638) between RPS and MPS. However, the operation time of RPS was shorter than that of MPS in both the AR (114.4 ± 28.7 minutes vs. 126.7 ± 34.5 minutes, P = 0.021) and RHC (112.61 ± 26.0 minutes vs. 146.52 ± 31.2 minutes, P < 0.001) groups. In addition, less intraoperative blood loss occurred during RPS than during MPS in both the AR (47.2 ± 57.2 mL vs. 69.4 ± 102.8 mL, P = 0.133) and RHC (60.0 ± 75.5 mL vs. 103.65 ± 81.7 mL, P = 0.066) groups, though it was not statistically significant. None of the patients experienced conversion to open surgery in either RPS or MPS, and no case in the RPS group was converted to MPS.

Regarding postoperative outcomes, length of hospital stay (AR: 7.3 ± 1.5 days vs. 7.3 ± 1.7 days, P = 0.958; RHC: 7.9 ± 3.1 days vs. 7.6 ± 2.5 days, P = 0.755) was similar between RPS and MPS (Table 3). No differences in postoperative complications such as pneumonia, urinary retention, prolonged postoperative ileus, or wound infection were observed between the 2 groups (AR: 9.7% vs. 9.7%, P = 0.999; RHC: 17.4% vs. 17.4%, P = 0.999) (Table 3). Most complications were grade I or II by the Clavien-Dindo classification, and only 2 patients (2.8%) who underwent AR by MPS experienced grade III complication (Table 3). However, time to gas passage was longer in RPS in both the AR (3.6 ± 1.7 days vs. 2.6 ± 1.5 days, P < 0.001) and RHC (3.8 ± 1.3 days vs. 2.7 ± 1.1 days, P = 0.004) groups (Table 3). No 30-day mortality occurred in either group.

In this study, we compared the short-term outcomes of RPS and MPS in colon cancer after propensity score matching. RPS was associated with a shorter operation time and a longer time to gas passage compared with MPS in both the AR and RHC groups. Intraoperative blood loss tended to be less in the RPS group than in the MPS group. In general, the short-term outcomes of RPS in colon cancer were acceptable to those of MPS.

In light of the trend toward minimally invasive surgery, single-port laparoscopic colectomy in colon cancer has become increasingly popular worldwide, and many previous studies have reported the feasibility of SPS in colon cancer [1213141516]. However, SPS has the inherent weakness of technical difficulties owing to the potential collision of laparoscopic instruments and the limitation of triangular tissue traction [1216]; consequently, SPS is an infrequently utilized colectomy procedure worldwide [17]. In contrast, enhanced by the addition of another small-size working port, RPS may partially overcome these drawbacks [17], which may increase the generalizability of the procedure. Moreover, RPS has other advantages, such as the convenience of an intracorporeal suture and stable drain placement via the additional port [17]. While several previous studies have demonstrated the feasibility of RPS in colon cancer [1718192021], these studies had certain limitations such as the absence of a control group [1819], the use of 4 ports [20], and a small number of patients who underwent RPS [17182021]. Furthermore, due to their retrospective study

Interestingly, the operation time was shorter in the RPS group than in the MPS group (Table 3), which is similar to the results of previous studies [1721]. It seems paradoxical that RPS, which could be considered more technically difficult than MPS, resulted in a shorter operation time. In previous studies, selection bias was identified as a possible explanation for this observation, suggesting that a greater number of less-complicated cases were included in the RPS group, thereby leading to shorter operation times [1721]. However, after compensating for potential confounders such as abdominal operation history, TNM stage, and tumor size by propensity score matching, the results of our study also demonstrated decreased operation times in RPS. Another proposed explanation was that the shorter operation time of RPS could be attributed to the surgeon's high level of expertise [21]. In the present study cohort, RPS was generally performed more recently during the study period; therefore, the surgeon's advanced surgical techniques might have enabled shorter operation times in RPS. Lastly, the time of wound closure may also be a factor in the paradoxical difference. In our institution, all of the incision wounds were generally closed using a layer-by-layer method and knot-bearing skin suture by less-skilled junior residents, the timing of which might have been prolonged. Therefore, the fewer number and smaller size of the wounds might have led to decreased time of wound closure, resulting in shorter overall operation times in RPS.

Comparing postoperative outcomes, there was no significant difference in hospital stay and postoperative complications between the 2 groups. However, time to first flatus in the RPS group was longer than that in the MPS group (Table 3). This may be related to bowel manipulation during the surgical procedure. It is more difficult to establish the operative field in RPS than in MPS; therefore, increased manipulation of the bowels is inevitable in RPS, which may result in delayed gas passage. However, in the present study, no difference in the incidence of prolonged postoperative ileus and hospital stay between the 2 groups was found. Thus, delayed gas passage did not compromise the short-term outcomes of RPS in colon cancer.

In the present study, we were unable to compare long-term outcomes, due to the short follow-up period. Instead, we compared the total number of harvested lymph nodes, which has a reported association with prognosis in colon cancer [232425]. No difference in the number of harvested lymph nodes was found between RPS and MPS (Table 3). Since both RPS and MPS were performed by a single surgeon in a similar fashion, similar numbers of harvested lymph nodes might represent the overall oncologic safety of RPS. Further studies are needed to evaluate the long-term outcome of RPS in colon cancer.

The present study has several limitations. First, although we utilized propensity score matching analysis, hidden confounders and bias may still exist due to the retrospective nature of the study. To overcome this limitation, a large-scale randomized controlled trial is necessary. Second, the number of patients who underwent RHC by RPS is relatively small (n = 23). Last, although RPS has been reported to have cosmetic benefits compared with MPS [1821], we could not show the total incision length or the satisfaction of patients due to the retrospective study design. Nevertheless, this study is one of the few to investigate the feasibility of RPS in colon cancer, and the minimization of bias by propensity score matching is a definite strength.

In conclusion, RPS is a feasible procedure in colon cancer surgery in light of the comparable short-term outcomes of RPS and MPS after adjusting for confounding factors via propensity score matching. The long-term oncologic outcomes of RPS should be evaluated in further studies.