Approximately 8–34% of patients with colorectal cancer have obstructive symptoms [1,2,3]. Obstructive colorectal cancer (OCRC) is considered a life-threatening condition that requires immediate intervention. However, emergency one-stage resection for OCRC is associated with significantly higher mortality and morbidity rates than elective surgery [4, 5]. In the last two decades, elective surgery following self-expandable metallic stent (SEMS) placement, known as a bridge to surgery (BTS), has been introduced and is widely accepted as an alternative treatment strategy to emergency surgery [6,7,8,9,10,11,12,13]. BTS for OCRC is now recommended in the 2020 European Society of Gastrointestinal Endoscopy guidelines [14]. Although the usefulness and safety of laparoscopic surgery is well established in patients with colorectal cancer [15,16,17,18], there have been no comprehensive studies in OCRC. Recent retrospective cohort studies have demonstrated better results for laparoscopic surgery than for open surgery following SEMS placement [7, 19,20,21,22,23,24,25,26,27,28,29,30,31,32]. However, it is quite difficult to perform randomized controlled trials (RCTs) with large enough sample sizes to reach a conclusion, partly because of the rarity of OCRC. Therefore, we performed this systematic review and meta-analysis of the relevant published studies, which included a total of 883 patients, to determine the efficacy and safety of laparoscopic surgery following SEMS placement for OCRC.

Methods

The meta-analysis was performed in accordance with the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [33].

Literature retrieval and study selection

The literature in the MEDLINE (PubMed), Google Scholar, and Cochrane Library databases was systematically searched through to December 2022 to identify relevant studies (Fig. 1). The search was limited to human studies published in English or Japanese. The search terms used were (“colon cancer” OR “colonic obstruction” OR “malignant obstruction”) AND (“laparoscopic surgery” OR “minimally invasive surgery” OR “open surgery” OR “laparotomy”) AND (“stent” OR “endoscopic decompression”). The related articles function was used to broaden the search. The reference lists of all relevant publications were searched manually for additional studies that may have been initially overlooked using our search strategy. The quality of the included studies was assessed using the Newcastle–Ottawa scale (NOS) for observational studies [34]. Studies were considered to be of high quality if they had an NOS score of ≥ 7. The MINORS (Methodological Index for Non-Randomized Studies) tool was used to assess the risk of bias for individual studies [35]. GRADE (Grading of Recommendations Assessment, Development and Evaluation) methodology was used to assess the quality of evidence and reported in the results with the help of GRADE Pro software (McMaster University and Evidence Prime Inc., Ontario, CA; https://www.gradepro.org/) [36].

Fig. 1
figure 1

Flowchart showing the study selection process according to the PRISMA guidelines

Full size image

Inclusion and exclusion criteria

The inclusion and exclusion criteria were defined a priori. Studies were included if they compared postoperative complications between patients with acute OCRC who underwent laparoscopic surgery and those who underwent open surgery following SEMS placement. Surgery was defined as primary tumor resection with or without primary anastomosis. Duplicated study reports and studies for which predefined outcomes were not reported or it was impossible to extract the number of outcome events were excluded.

Extraction of data

The full-text version of each eligible study was evaluated by two investigators (S.K., A.M.) working independently. The following data were extracted: name of the primary author, year of publication, country in which the study was performed, number of participating institutions, design and duration of the study, number of study participants and their characteristics, including age, sex, and tumor-related variables, interval between SEMS placement and surgery, and all available information on short-term and long-term outcomes.

Data synthesis and statistical analysis

Short-term outcomes (primary anastomosis, stoma construction, overall morbidity, surgical site infection, anastomotic leakage, and postoperative ileus) and long-term outcomes (3-year recurrence-free survival [RFS] and 3-year overall survival [OS]) were compared between patients who underwent laparoscopic surgery (the LS group) and those who underwent open surgery (the OpS group). Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. An OR of < 1 favored the BTS group, and the point estimate of the OR was considered statistically significant at P < 0.05 if the 95% CI did not include the value 1. The pooled OR was calculated using a Mantel–Haenszel fixed-effect model or a DerSimonian–Laird random-effects model to combine ORs for outcomes of interest. The meta-analysis was performed using Review Manager (Version 5.1) for Windows (Nordic Cochrane Center, Cochrane Collaboration, Copenhagen, Denmark; http://www.cc-ims.net/RevMan). The Cochran’s chi-square-based Q statistic test was used to assess between-study heterogeneity. The I2 value was used to test for heterogeneity among the included studies. Study heterogeneity was measured using the χ2 and I2 statistics, with a χ2 P value of < 0.05 and an I2 value of ≥ 50% indicating heterogeneity [37]. A fixed-effect model was used to estimate the overall effect if the OR was homogeneous; if the OR was not homogeneous, a random-effects model was used [38]. Publication bias was assessed by visual examination and statistical analysis of a funnel plot, with asymmetry formally assessed by use of Egger’s linear regression test and the rank correlation (Begg’s) test using WINPEPI software (available at http://www.brixtonhealth.com/pepi4windows.html) [39, 40].

Results

Literature review and included studies

In total, 485 potentially relevant citations were identified during the initial screening. After reviewing the titles and abstracts, 359 studies were excluded. One hundred and eleven further studies were excluded after full-text evaluation, leaving 15 studies published between 2004 and 2022 [7, 19,20,21,22,23,24,25,26,27,28,29,30,31,32] for inclusion in the meta-analysis (Fig. 1). The background characteristics of the included studies are shown in Table 1. Three studies [19,20,21] originated from Europe and 12 [7, 22,23,24,25,26,27,28,29,30,31,32] from Asia. All the studies had a retrospective observational design. In total, 467 (52.9%) of the 883 patients included in the meta-analysis underwent laparoscopic surgery and 416 (47.1%) underwent open surgery. The risk of bias was assessed independently using the NOS score (Table 2). The NOS score was ≤ 6 in three studies [19, 20, 24] and ≥ 7 in 12 [7, 21,22,23, 25,26,27,28,29,30,31,32]. The included studies had a mean MINORS score (± standard deviation) of 12.93 ± 4.15, indicating that the quality of evidence for non-randomized studies was fair. The MINORS results for the included studies are shown in Table 3. According to the GRADE criteria, the overall quality of evidence was very low for ileus, stoma construction, and primary anastomosis, low for mortality, anastomotic leak, wound infection, and 3-year OS and RFS, and moderate for postoperative complications (Table 4).

Table 1 Basic characteristics of the fifteen included studies
Full size table
Table 2 NOS score of included studies
Full size table
Table 3 MINORS assessment of included studies
Full size table
Table 4 GRADE evidence profile
Full size table

Short-term outcomes

Postoperative complications

All 15 studies reported postoperative complications. Only four studies [7, 19, 31, 32] reported the severity of these complications and eight [7, 19, 23, 25, 29,30,31,32] reported postoperative complications occurring within 30 days after surgery. Therefore, postoperative complications were defined as the overall morbidities listed in the included studies [7, 19,20,21,22,23,24,25,26,27,28,29,30,31,32]. The postoperative complication rate was 13.3% (62/467) in the LS group and 23.8% (99/416) in the OpS group. The heterogeneity test indicated a χ2 value of 19.06 and an I2 value of 27%, demonstrating homogeneity. Therefore, a fixed-effect model was adopted (OR 0.47, 95% CI 0.32–0.67, P < 0.001) (Fig. 2). This meta-analysis demonstrated that laparoscopic surgery contributed to a significant reduction in postoperative complications compared to open surgery. We found no significant publication bias by visual inspection of the funnel plot (Supplementary Fig. 1) or on Egger’s test (P = 0.772) or Begg’s test (P = 0.255).

Fig. 2
figure 2

Meta-analysis of postoperative complication rates using a fixed-effect Mantel–Haenszel model. Odds ratios are shown with 95% CIs. CI confidence interval, LS laparoscopic surgery, OpS open surgery

Full size image

Mortality

Mortality was reported in eleven studies. The postoperative mortality rates in the LS and OpS groups were 0.5% (2/408) and 0.3% (1/329), respectively. The heterogeneity test indicated a χ2 value of 0.63 and an I2 value of 0%, indicating homogeneity. Therefore, the fixed-effect model was adopted (OR 1.01, 95% CI 0.18–5.55, P = 0.99) (Fig. 3). There was no significant difference in mortality between the two groups. We found no significant publication bias during visual inspection of the funnel plot (Supplementary Fig. 2) or on Egger’s test (P = 0.975) or Begg’s test (P = 0.602).

Fig. 3
figure 3

Meta-analysis of postoperative mortality rates using a fixed-effect Mantel–Haenszel model. Odds ratios are shown with 95% confidence intervals. CI confidence interval, LS laparoscopic surgery, OpS open surgery

Full size image

Other outcomes

The surgical outcomes are shown in Table 5 and in Supplementary Figs. 3, 4, and 5. The operation time was shorter in the OpS group than in the LS group, and the postoperative hospital stay was shorter in the LS group than in the OpS group. The between-group differences in both outcomes were statistically significant (P < 0.01).

Table 5 Meta-analysis of other short-term outcomes
Full size table

The other short-term outcomes are shown in Table 6 and in Supplementary Figs. 6, 7, 8, 9, and 10. The primary anastomosis rate was favored the LS group over OpS group (97.9% [190/194] vs. 91.2% [145/159]) and as was the stoma construction rate (5.5% [10/182] vs. 15.1% [24/159]). The between-group differences for both these outcomes were statistically significant (OR 0.23, 95% CI 0.08–0.66, P = 0.006 for primary anastomosis; OR 0.28, 95% CI 0.13–0.62, P = 0.002 for stoma construction) and without between-study heterogeneity (χ2 = 1.86, I2 = 0%, P = 0.87 and χ2 = 2.90, I2 = 0%, P = 0.41, respectively).

Table 6 Meta-analysis of other short-term outcomes
Full size table

Postoperative anastomotic leakage, wound infection, and ileus were analyzed. The meta-analyses of wound infection demonstrated significantly favorable results in the LS group over OpS group (OR 0.42, 95% CI 0.21–0.84, P = 0.02) without between-study heterogeneity (χ2 = 3.51, I2 = 0%, P = 0.94, respectively). There was no significant between-group difference in the anastomotic leak rate (OR 0.67, 95% CI 0.31–1.45, P = 0.31) or in the frequency of ileus (OR 0.64, 95% CI 0.34–1.21, P = 0.17 for ileus). There was no between-study heterogeneity (χ2 = 2.18, I2 = 0%, P = 0.98, and χ2 = 7.26, I2 = 0%, P = 0.61, respectively).

Long-term outcomes

The long-term outcomes are demonstrated in Figs. 4 and 5. Four studies [23, 30,31,32] reported the 3-year RFS rate and six [20, 23, 29,30,31,32] reported the 3-year OS rate. Data on 3-year RFS were available for 381 patients and data on 3-year OS for 486 patients. There was no significant difference in 3-year RFS (OR 0.78, 95% CI 0.50–1.24, P = 0.30) or 3-year OS (OR 0.68, 95% CI 0.41–1.12, P = 0.13) between the LS and OpS groups or any between-study heterogeneity (χ2 = 1.67, I2 = 0%, P = 0.30 and χ2 = 1.08, I2 = 0%, P = 0.90, respectively). We found no significant publication bias in terms of either outcome by visual inspection of the funnel plot (Supplementary Figs. 11 and 12) or on Egger’s test (P = 0.509) or Begg’s test (P = 1.00) and Egger’s test (P = 0.299) or Begg’s test (P = 0.142).

Fig. 4
figure 4

Meta-analysis of 3-year recurrence-free survival using a fixed-effect Mantel–Haenszel model. Odds ratios are shown with 95% confidence intervals. CI confidence interval, LS laparoscopic surgery, OpS open surgery

Full size image
Fig. 5
figure 5

Meta-analysis of 3-year overall survival using a fixed-effect Mantel–Haenszel model. Odds ratios are shown with 95% confidence intervals. CI confidence interval, LS laparoscopic surgery, OpS open surgery

Full size image

Discussion

This meta-analysis was performed to obtain an overview of the recent literature on the outcomes of laparoscopic surgery following SEMS placement for OCRC. Although several retrospective studies of laparoscopic surgery following SEMS placement have been published, there have been no RCTs to date, possibly because of the rarity of OCRC, the urgent situation at the time of initial diagnosis, and the oncological safety of BTS itself not having been established [41, 42]. Based on our present findings, an RCT comparing short-term and long-term outcomes between laparoscopic surgery and open surgery following SEMS placement for OCRC would be expected to have a 3-year RFS rate of 30% for open surgery and 20% for laparoscopic surgery. Therefore, a sample size of at least 260 would be required in each group for a statistical power of 80% and a significance level of 0.05 [15, 16]. It would be very difficult to collect such a large number of cases for an RCT. This systematic review and meta-analysis was performed to determine if laparoscopic surgery following SEMS placement is safe and effective for OCRC and drew on as much evidence as possible from previous reports in a sample of adequate size (n = 883).

Large-scale RCTs in patients with colon cancer have established that laparoscopic surgery decreases surgical trauma and perioperative complications, allows more rapid recovery, and has a non-inferior oncological prognosis [15,16,17,18]. However, in patients with OCRC, laparoscopic surgery following SEMS placement may be contraindicated because of the limited surgical field as a result of the distended bowel and the peculiarities of tumor size and depth. Morino et al. [13] were the first to describe use of the laparoscopic approach following SEMS placement for OCRC and concluded that the colonic segment was bulkier and more technically difficult to resect by laparoscopy. Thirteen of the studies in this meta-analysis [7, 19, 21,22,23,24,25,26,27, 29,30,31,32] compared baseline characteristics between an LS group and an OpS group. Although only one of the studies reported a significantly higher proportion of men in its LS group [25], there was no significant difference in background characteristics. Twelve studies analyzed data on pathological stage and found no significant between-group difference [7, 20,21,22,23, 25,26,27, 29,30,31,32]. However, one study excluded patients with suspected invasion of other organs from its LS group [26] and another study reported a significantly greater number of patients with pathological T4b disease in its OpS group [31]. Therefore, differences in patient characteristics in the individual studies would not be expected to have much statistical impact on our results.

Many large-scale trials have demonstrated the feasibility and safety of laparoscopic colorectal surgery, particularly a reduction in postoperative complications [15,16,17,18, 43, 44]. Our present meta-analysis found that the risk of postoperative complications was significantly lower in patients who underwent laparoscopic surgery than in those who underwent open surgery (OR 0.47, 95% CI 0.32–0.67, P < 0.001) after BTS for OCRC. Postoperative complications have been widely reported to have a negative oncological impact after digestive cancer surgery [45,46,47,48], and the same finding has been reported for colorectal cancer surgery. Several explanations for this finding have been suggested, including local and systemic activation of proinflammatory cytokines and mediators, delayed or canceled adjuvant chemotherapy, and abdominal implantation of intraluminal cancer cells in patients with anastomotic leakage [49,50,51,52,53]. A significant correlation of postoperative complications with a worse prognosis was also reported in a BTS cohort [54]. Therefore, efforts to minimize postoperative complications are important in BTS for OCRC and choice of laparoscopic surgery may be useful.

This study had several limitations. First, as with all systematic reviews, the strength of our conclusions depends on the quality of the primary studies. Unfortunately, the design and quality of the studies included in this review were not high. Furthermore, no relevant RCTs or prospective studies were available for analysis. RCTs investigating the safety and usefulness of the BTS strategy with an adequate sample size are difficult to perform because of (1) lack of evidence for its use (an RCT is currently underway in Japan) and (2) the fact that a proportion of patients with OCRC have metastasis to other organs, which makes laparoscopic surgery difficult and inevitably introduces significant bias stemming from the surgeon’s level of skill and judgment. Second, there was heterogeneity between studies because of differences in sample size, pathological staging, study design, and follow-up. There was also heterogeneity in the definitions of morbidity and mortality. Moreover, there was considerable heterogeneity in long-term outcomes in terms of pathological staging and postoperative adjuvant therapy. This heterogeneity had a marked effect on our results.

In conclusion, laparoscopic surgery following placement of a SEMS for OCRC significantly reduces the postoperative complication rate. Furthermore, there were no significant differences in long-term outcomes between the two procedures. Our findings suggest that laparoscopic surgery following SEMS placement is a safe and effective treatment option for OCRC.

Disclosures

Shintaro Kanaka, Takeshi Yamada, Akihisa Matsuda, Kay Uehara, Seiichi Shinji, Yasuyuki Yokoyama, Goro Takahashi, Takuma Iwai, Kohki Takeda, Sho Kuriyama, Toshimitsu Miyasaka, and Hiroshi Yoshida have no conflicts of interest or financial ties to disclose.