Enhanced recovery after surgery in emergency resection for obstructive colorectal cancer: a systematic review and meta-analysis

Enhanced recovery after surgery (ERAS) improves outcomes after elective colorectal operations. Whether it is beneficial for emergency colorectal surgery is unclear. This study aimed to systematically review and summarize evidence from all studies comparing ERAS versus conventional care in patients having emergency colectomy and/or proctectomy for obstructive colorectal cancer. EMBASE, MEDLINE, and PUBMED from 1981 to December 2019 were systematically searched. Any studies comparing our primary outcome of interest (length of hospitalization) among patients having emergency resection for obstructive colorectal cancer who received ERAS versus conventional care were selected. Primary outcome was length of hospitalization. Secondary outcomes were gastrointestinal recovery, postoperative complication, 30-day readmission and mortality, and time to start adjuvant therapy. Three cohort studies with 818 participants (418 received ERAS and 400 received conventional care) were included. Length of hospitalization (mean reduction 3.07 days; 95% CI, − 3.91 to − 2.23) and risk of overall complication (risk ratio 0.78; 95% CI, 0.63 to 0.97) were significantly lower in ERAS than in conventional care. ERAS was also associated with quicker time to gastrointestinal recovery, a lower incidence of ileus, and a shorter interval between operation and commence of adjuvant chemotherapy. There was no significant difference in the rates of anastomotic leakage, surgical site infection, reoperation, readmission, and mortality within 30 days after surgery between groups. ERAS had advantages over conventional care in patients undergoing emergency resection for obstructive colorectal cancer—including a shorter length of hospitalization, a lower incidence of overall complication, and a quicker gastrointestinal recovery.


Background
Emergency colectomy and/or proctectomy are among the top three most common emergency laparotomies performed in the UK [1] and the USA [2]-with a high rate of 30-day postoperative morbidity (ranging from 26.8 to 69.2%) [2] and mortality (ranging from 8.8 to 31.6%) [1]. Acute colonic obstruction is a common manifestation of colorectal cancer (CRC) that is responsible for 15-30% of initial clinical presentation [3]. Emergency resection of the tumor with or without anastomosis remains the mainstay treatment for obstructive CRC-especially right-sided colon cancer, left-sided colonic obstruction that is not eligible for colonic stenting, and CRC with suspected bowel perforation [3]. However, postoperative morbidities and mortalities following surgical treatment for Electronic supplementary material The online version of this article (https://doi.org/10.1007/s00384-020-03652-5) contains supplementary material, which is available to authorized users. obstructive CRC are high [4]-even performed by colorectal surgeons [5].
Enhanced recovery after surgery (ERAS) minimizes patient's surgical stress responses, optimizes their function, and enables rapid recovery [6]. Fundamentally, ERAS has 3 phases of intervention: preoperative optimization of patient's medical condition, intraoperative care, and postoperative multimodal rehabilitation. Several meta-analyses of randomized controlled trials of elective colorectal operations have consistently found the advantages of ERAS over perioperative conventional care-including shorter hospitalization and fewer complications [7][8][9]. However, only few studies examining the benefits of ERAS in emergency colorectal surgery are available and their results were inconsistent [10][11][12]. Notably, the ERAS protocol for elective colorectal operations has to be adapted to emergency colorectal surgery because not all ERAS interventions can be used in an emergency setting, e.g., preoperative nutritional support and carbohydrate loading. Also, currently, there are no published ERAS Society guidelines or recommendations for emergency colorectal surgery.
To our knowledge, there is no meta-analysis evaluating the benefits of ERAS protocol for colorectal surgery in an emergency setting. Our study hypothesized that ERAS could facilitate patient's discharge and improves surgical outcomes after emergency colectomy and/or proctectomy for obstructive CRC.

Materials and methods
This systematic review and meta-analysis were performed in accordance to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [13].

Search strategy
Literature published in English language and indexed in EMBASE, MEDLINE, and PUBMED from 1981 to 31 December 2019 was independently searched by three investigators. The following search terms were used, in all possible combination: enhanced recovery, fast track, multimodal pathway, accelerated program, surgery, resection, colon, sigmoid, colorectum, rectum, neoplasms, obstructive, intestinal obstruction, emergency, and urgency. Synonyms of each of the terms were also used in the search. Detailed searching strategies for EMBASE, MEDLINE, and PUBMED are shown in Supplement file 1. References of the included studies were further reviewed to identify any additional suitable studies that may be missed by the aforementioned search strategy.

Inclusion criteria
Studies that were eligible for this meta-analysis must be cohort studies that enrolled patients having emergency colectomy and/or proctectomy for obstructive CRC. One group of patients must follow ERAS pathway while the other group must receive conventional care. The eligible studies must report the primary outcome, which is the average length of hospital stay of both groups along with standard deviation (SD). Secondary outcomes, including time to pass stool, time to resume regular diet, time to start adjuvant chemotherapy, rate of postoperative ileus, anastomotic leakage, surgical site infection, overall complications, reoperation, readmission, and mortality within 30 days after an operation, were extracted and analyzed but were not part of the inclusion criteria. Study eligibility was independently evaluated by the three investigators with differing opinions resolved by group discussion. The Newcastle-Ottawa quality assessment scale for cohort studies was used to determine the quality of each study [14].

Data extraction
The three investigators independently extracted data using a standardized form containing the primary outcome, secondary outcomes and characteristics of the study which included, country where the study was conducted, study period, year of publication, number of cases, age and gender of participants, location of obstructive CRC, and types of surgery performed in that study. Any discrepancies were reviewed and corrected according to the original reports.

Statistical analysis
Mean length of hospital stay, time to first bowel movement, time to tolerate solid food, and time to start adjuvant chemotherapy along with SD of participants in both groups were extracted from each study and the mean difference (MD) was calculated. Pooled MD was then calculated by combining MDs of each study using random effects model. If the study provided median and interquartile range, median would be used as an estimate for mean and SD would be estimated from interquartile range divided by 1.35.
Relative risk (RR) of postoperative ileus, leakage, surgical site infection, overall complications, reoperation, readmission, and mortality were calculated using rate of that event in each group. RRs from all studies included were added together to calculate pooled RR using the DerSimonian and Laird random effects model [15]. To confirm the robustness of the pooled results, any outcomes with more than two eligible studies would undergo sensitivity analysis by excluding one study at a time from the complete analysis to see if the pooled results would be significantly altered.
The heterogeneity of the MDs and RRs across studies was calculated using the Q statistic (I 2 statistics) where I 2 value of 0-25%, 26-50%, 51-75%, and 76-100% means insignificant, low, moderate, and high heterogeneity, respectively [16]. Visual inspection of funnel plots would be used to determine publication bias if enough eligible studies were identified. Statistical analysis was done using Review Manager 5.3 software from the Cochrane Collaboration (London, UK).

Results
A total of 704 potentially relevant articles (399 from EMBASE, 109 from MEDLINE, and 196 from PUBMED) were identified. After removal of 125 duplicated articles, 579 articles underwent title and abstract review. At this stage, 573 articles were excluded because they did not meet our inclusion criteria-leaving the remaining 6 articles retrieved for a complete review. After reviewing full-length articles, 3 articles were excluded because of conference abstract (n = 2) and mixed small bowel and colon diseases (n = 1). Finally, three cohort studies [10][11][12] with 818 participants (418 underwent ERAS and 400 underwent conventional care) fulfilled the inclusion criteria and were included into the meta-analysis. Study flow is shown in Fig. 1. Characteristics of the included studies are summarized in Table 1.

Length of hospital stay
Length of hospital stay was significantly shorter in patients receiving ERAS than that of those receiving conventional care with the pooled MD of − 3.07 days (95% CI, − 3.91 to − 2.23).

Overall complications
The rate of postoperative complication was significantly lower in patients receiving ERAS than that of those receiving conventional care with the pooled RR of 0.78 (95% CI, 0.63 to 0.97). Heterogeneity among studies was negligible (I 2 of 0%) (Fig. 2b).

Thirty-day readmission and mortality
Rates of 30-day readmission and mortality were comparable between the two groups-with the pooled RR of 0.91 (95% CI, 0.54 to 1.51) and 0.79 (95% CI, 0.11 to 5.38), respectively. Heterogeneity among studies was negligible (I 2 of 0% and 24%, respectively) ( Fig. 2 c and d).

Time to gastrointestinal recovery
Time to pass flatus was significantly shorter in patients receiving ERAS than that of those receiving conventional care with the pooled MD of − 1.39 days (95% CI, − 1.52 to − 1.25). The between-study heterogeneity was negligible (I 2 of 0%) (Fig.  3a). Time to first defecation was also significantly shorter in patients receiving ERAS-with the pooled MD of − 0.73 days (95% CI, − 1.40 to − 0.06), but heterogeneity between studies was moderate (I 2 of 73%) (Fig. 3b). A significantly shorter time to resume regular diet was found in the ERAS group with the pooled MD of − 2.39 days (95% CI, − 2.62 to − 2.15), and the between-study heterogeneity was negligible (I 2 of 0%) (Fig. 3c).  Table 1 Main characteristics of the cohort studies included in the meta-analysis Lohsiriwat [10] Shida et al. [11] Shang et al. [12] Country

Postoperative ileus
The rate of postoperative ileus was significantly lower in patients receiving ERAS than that of those receiving conventional care with the pooled RR of 0.55 (95% CI, 0.33 to 0.91). Heterogeneity among studies was negligible (I 2 of 0%) (Fig. 4a).

Anastomotic leakage
Rate of anastomotic leakage was comparable between the two groups-with the pooled RR of 0.96 (95% CI, 0.25 to 3.65). The between-study heterogeneity was negligible (I 2 of 0%) (Fig. 4b).

Surgical site infection
Rate of surgical site infection was comparable between the two groups-with the pooled RR of 0.78 (95% CI, 0.47 to 1.28). The between-study heterogeneity was negligible (I 2 of 1%) (Fig. 4c).

Reoperation
Rate of 30-day reoperation was comparable between the two groups-with the pooled RR of 0.78 (95% CI, 0.32 to 1.92). The between-study heterogeneity was negligible (I 2 of 0%) (Fig. 4d).

Time to start adjuvant chemotherapy
Patient receiving ERAS had a significantly shorter time to start adjuvant chemotherapy, with the pooled MD of − 12.05 days (95% CI, − 14.76 to − 9.35) and negligible betweenstudy heterogeneity (I 2 of 0%) (Fig. 5).

Sensitivity analysis
A total of three outcomes (length of hospital stay, overall complications, and postoperative ileus) were eligible for sensitivity analysis. For the length of hospital stay, exclusion of the study by Lohsiriwat [10] resulted in the new pooled MD of − 3.00 days (95% CI, − 3.92 to − 2.08; I 2 0%; Supplementary Figure 1a). Exclusion of the study by Shida et al. [11] resulted in the new pooled MD of − 3.22 days (95% CI, − 4.73 to − 1.71; I 2 0%; Supplementary Figure 1b). Exclusion of the study by Shang et al. [12] resulted in the new pooled MD of − 3.08 days (95% CI, − 3.98 to − 2.18; I 2 0%; Supplementary Figure 1c). For the rate of overall complications, exclusion of the study by Lohsiriwat [10] resulted in the new pooled RR of 0.79 (95% CI, 0.64 to 0.98; I 2 0%; Supplementary  Figure 2a). Exclusion of the study by Shida et al. [11] resulted in the new pooled RR of 0.79 (95% CI, 0.63 to 0.98; I 2 0%; Supplementary Figure 2b). Exclusion of the study by Shang et al. [12] resulted in the new pooled RR of 0.63 (95% CI, 0.28 to 1.43; I 2 0%; Supplementary Figure 2c).
For the rate of postoperative ileus, exclusion of the study by Lohsiriwat [10]

Evaluation for publication bias
Due to a limited number of eligible studies, funnel plot was not created for evaluating publication bias.

Discussion
This systematic review and meta-analysis of ERAS in emergency resection for obstructive CRC incorporated 3 cohort studies that included 418 cases with ERAS and 400 cases with conventional care. This meta-analysis indicated that patients receiving ERAS had a significantly shorter length of hospital stay, quicker time to gastrointestinal recovery, lower incidence of postoperative ileus and overall complications, and shorter time to initiate adjuvant chemotherapy.
The approximately 3-day reduction in hospital stay and a 22% decreased risk of overall complications observed in this study are comparable with the benefit of ERAS in the setting of elective colorectal operation. A systematic review and meta-analysis of 13 randomized controlled trials including 1910 patients undergoing elective major colorectal operations showed that ERAS decreased length of hospitalization by 2.4 days and decreased risk of overall complications by 29% compared with conventional care [8]. Another systematic review and meta-analysis of 25 trials including 3787 patients undergoing elective laparoscopic or open surgery for malignant and benign colorectal diseases [9] found a 2.6 days reduction of hospital stay and a 34% decrease in risk of perioperative complications compared with conventional care. All of these results have suggested that benefit of ERAS is seen across different surgical approaches and indications. Notably, our sensitivity analyses confirmed the positive effect of ERAS on reducing the length of hospitalization in emergency colorectal surgery with a high degree of certainty-but a lesser degree of robustness of its benefits in diminishing the rate of overall complications.
Nevertheless, according to our meta-analysis, the implementation of ERAS significantly decreased the rate of overall complications but not that of surgical site infection-which is one of the most common complications following emergency colorectal surgery [17]. These findings were supported by a larger meta-analysis of ERAS in the setting of elective colorectal surgery-in which the investigators found that ERAS significantly reduced total complications and medical complications, but not surgical complications including surgical site infection [8].
There was also no significant difference in the rate of anastomotic leakage and reoperation. Therefore, the benefit of early resumption of solid food and accelerated gastrointestinal recovery by ERAS does not come at the expense of increased anastomotic leakage. In fact, it has been shown that implementation of ERAS [18], early consumption of solid food [19], and prophylactic administration of prokinetic drug [20] can facilitate the return of bowel function and minimize the rate of prolonged ileus following elective colorectal operation.
The current study demonstrated that patients in ERAS pathway had an approximately 2 weeks shorter time interval between operation and starting date of adjuvant chemotherapy (which is usually indicated in patients with obstructive CRC [21]). Time to adjuvant therapy is a potential mid-term outcome measure that could be used to determine patient's overall recovery and performance [22], as it has been shown that delayed commencement of adjuvant chemotherapy (especially > 8 weeks) is linked to a worse overall survival in patients with CRC. One possible explanation of this is that early initiation of postoperative chemotherapy may prevent tumor growth or recurrence [23]. Since postoperative complications are one of the major reasons for delayed adjuvant chemotherapy [24], the benefits of ERAS, therefore, may go beyond just the immediate postoperative period.
It is worth noting that all three cohort studies included in this meta-analysis used their "modified" ERAS pathway [10][11][12] as there is still no standard ERAS Society guideline or recommendation for emergency colorectal surgery. Thus, it remains unknown which interventions of ERAS pathway are beneficial and should be recommended in clinical practice. Obviously, some interventions, such as preoperative nutritional support and carbohydrate loading, are not advisable for patients with acute malignant colonic obstruction. However, most ERAS interventions recommended for elective colorectal operation appear to be applicable to emergency surgery as well [25].
It should also be noted that patients with obstructive CRC with colonic perforation were not included in any of the included studies. In fact, a recent study of ERAS in emergency colorectal resection for various colorectal conditions, including obstructive CRC and fecal peritonitis, suggested that patients with bowel perforation had a lower rate of ERAS compliance and a higher rate of complications than those without Fig. 5 Forest plots of the comparisons of time to start adjuvant chemotherapy between ERAS and conventional care. ERAS, enhanced recovery after surgery [17]. Further investigations are still needed to determine whether ERAS could provide similar benefits to these patients.
A major strength of this study is the unique advantage of the systematic review and meta-analysis technique that offers comprehensive evaluation of all available evidence. The study was conducted in accordance to the PRISMA guideline to ensure the highest standard of the analyses and data on several outcomes were analyzed. Nonetheless, there are some limitations that should be acknowledged.
First, only three cohort studies were included and they were all retrospective in nature. This may increase the risk for both performance and measurement bias-although a matching analysis was performed in two of the three studies [10,12] and one study was a relatively large multicenter trial [12]. It is noteworthy that, to date, there is no published or registered randomized controlled trial that compares outcomes between ERAS versus conventional care in emergency setting for malignant colonic obstruction (http://www.clinicaltrials.gov/ ClinicalTrials.gov; accessed on 28 April 2020). It may be difficult, or even ethical [25], to conduct such randomized controlled trials because ERAS has increasingly become a standard of surgical care [6]. As a result, the present study might be the best available evidence to support that ERAS can be applied effectively and safely in the setting of emergency resection for obstructive CRC-similar to elective colorectal operations [7][8][9].
Second, the role of compliance on the benefit of ERAS is not known as none of the included studies investigated the impact of compliance on surgical outcomes although a large international registry of patients undergoing elective CRC resection [26] and a small cohort of patients undergoing emergency surgery for obstructive CRC and colonic perforation [17] have demonstrated that higher compliance to ERAS was associated with better outcomes. Similarly, the impact of laparoscopy is not known because only one study in this meta-analysis included patients who underwent laparoscopic surgery in their "modified" ERAS protocol (and only 4 patients were included) [11]. Although studies of elective colorectal surgery found a significant benefit of laparoscopy combined with ERAS with markedly improved surgical outcomes and patient recovery [26,27], the benefit of laparoscopy may not be translated into emergency setting as the conversion rate is still relatively high [25].

Conclusions
This systematic review and meta-analysis of 3 nonrandomized observational studies incorporating 818 patients undergoing emergency resection for obstructive CRC found that, in comparison with conventional care, ERAS pathway was associated with a shorter length of hospital stay, quicker time to bowel function recovery, a lower incidence of postoperative ileus and overall complications, and a shorter time to start adjuvant chemotherapy. Anastomotic leakage, surgical site infection, reoperation, 30-day readmission, and mortality were comparable between the groups.