Comparative efficacy of bowel preparation strategies in colorectal surgery: A network meta-analysis of randomized controlled trials

Abstract

BACKGROUND

Colorectal surgeries are commonly performed procedures, yet they are associated with a considerable risk of complications such as surgical site infections (SSIs) and anastomotic leaks. Preoperative approaches, including mechanical bowel preparation (MBP) and oral antibiotics (OAB), aim to reduce these risks.

AIM

To comprehensively evaluate and compare the efficacy of various bowel preparation approaches in reducing postoperative complications following colorectal surgery.

METHODS

A thorough literature search was conducted in PubMed, Cochrane Library, Web of Science, and Scopus up to April 28, 2025. A frequentist network meta-analysis with a random-effects model was used to calculate relative risks with 95% confidence intervals (CIs), and interventions were ranked using P scores. Confidence in Network Meta-Analysis assessment was used to evaluate the confidence in the evidence. Primary outcomes included incisional SSI and organ space SSI. Secondary outcomes encompassed anastomotic leakage, postoperative ileus, and Clavien-Dindo I-II complications.

RESULTS

Our analysis, conducted on 38 studies, showed that the combination of MBP and OAB considerably reduced incisional SSI with a relative risk of 0.53 (95%CI: 0.29-0.99, P = 0.04) and organ-space SSI with a relative risk of 0.62 (95%CI: 0.39-0.99, P = 0.04), demonstrating a statistically significant effect. In the secondary outcomes, OAB-containing regimens showed superiority to the other treatments, and they also had the highest P score.

CONCLUSION

Even though the combination of MBP and OAB is strongly supported as the preferred strategy for preventing infectious postoperative complications, these findings are not only in line with current recommendations but also provide a foundation for additional research that will investigate their relevance to a larger variety of surgical procedures.

Key Words: Surgical site infections; Preoperative cleansing; Antimicrobial prophylaxis; Colorectal surgery; Meta-analysis

Core Tip: This network meta-analysis provides a comprehensive comparison of bowel preparation strategies in colorectal surgery using data from 38 randomized controlled trials. Mechanical bowel preparation combined with oral antibiotics (OAB) significantly reduced incisional surgical site infection (relative risk 0.53, 95% confidence interval: 0.29-0.99) and organ-space surgical site infection (relative risk 0.62, 95% confidence interval: 0.39-0.99) compared with other regimens. OAB-containing approaches ranked highest for preventing postoperative infectious complications. These findings reinforce the superiority of combined mechanical bowel preparation + OAB as the optimal strategy and support current clinical recommendations while highlighting areas for further research.

INTRODUCTION

Over 600000 colorectal surgeries are performed annually in the United States, making it one of the most common procedures worldwide. For the 2024-2026 period, the American Cancer Society projected up to 158850 new colorectal cancer (CRC) cases annually, with surgical resection remaining the primary curative treatment[1]. It includes both urgent surgeries for issues like diverticulitis, bowel perforation, or lower gastrointestinal bleeding, as well as planned resections for CRC and inflammatory bowel disease[2,3].

Since colorectal surgeries involve entering a bacterially colonized portion of the gastrointestinal tract, exposure to gut bacteria during the operation is nearly unavoidable. This makes preventing infections a key focus of perioperative management. Guidelines from organizations like the American Society of Colon and Rectal Surgeons recommend a combination of mechanical bowel preparation (MBP) and oral antibiotics (OAB) for elective colorectal cases before surgery to help reduce the bacterial load in the colon[4]. However, the effectiveness of MBP alone is increasingly being questioned. Many patients find it hard to tolerate due to dehydration, electrolyte imbalances, and discomfort, and recent studies have shown that it may not significantly reduce postoperative infections or leaks when used by itself[5,6].

Even with significant improvements in surgical techniques and care before and after surgery, complications after the procedure still pose a serious challenge in both open and laparoscopic approaches, with an incidence of 15% and 20%, respectively[3]. Factors like anaesthesia effects, bowel swelling, and postoperative ileus can all hinder healing and raise the risk of infections[7]. As a result, infectious complications, especially surgical site infections (SSIs), continue to be a major concern in colorectal surgery, with rates reported between 5.4% and 23.2%[8]. SSIs can involve both superficial and deep incisional infections, as well as infections within internal organs. They often lead to longer hospital stays, increased healthcare costs, and greater risks of illness and death[9]. Among these complications, anastomotic leak (AL) occurs in about 2%-10% of cases and is closely linked to serious adverse events such as sepsis, reoperation, and higher mortality[9,10]. The median postoperative hospital stay for patients with AL is approximately 18 days, reflecting the severity and resource-intensive nature of this complication[9].

A growing amount of evidence, especially from large studies like the American College of Surgeons National Surgical Quality Improvement Program, indicates that combining OAB and MBP can significantly lower the risk of SSI, AL, and postoperative ileus compared to using either method alone or no preparation at all[11-13]. These insights have led major surgical organizations, such as the American Society of Colon and Rectal Surgeons, the American College of Surgeons, and the Surgical Infection Society, to update their guidelines to recommend the combined use of OAB and isosmotic MBP before elective colorectal surgeries[4,14]. Still, the actual implementation of these guidelines is lacking. A recent international audit across Europe found that only 16.8% of patients received the recommended combined treatment, while 52.9% underwent MBP alone, showcasing a notable gap between evidence and practice[15].

Despite numerous meta-analyses that have looked into bowel preparation strategies, the findings remain inconsistent. Traditional pairwise meta-analyses fall short because they can’t simultaneously compare multiple interventions or incorporate both direct and indirect evidence. With various regimens available, ranging from no bowel preparation (NBP) to MBP alone, OAB alone, and the combination of MBP and OAB, there’s still uncertainty about which strategy is most effective for different postoperative outcomes.

To address the literature gap, we conducted a comprehensive network meta-analysis (NMA) that brings together data from randomized studies comparing different bowel preparation strategies in elective colorectal surgery. This method allows us to evaluate and rank all competing regimens at once, creating a cohesive framework for evidence-based decision-making. Our main goal is to figure out which bowel preparation strategy best reduces the risk of SSIs, with secondary outcomes including AL, mortality, Clavien-Dindo classification (a standardized grading system for postoperative complications based on severity), postoperative ileus, urinary tract infection (UTI), sepsis, and how long patients stay in the hospital. By presenting an updated and methodologically sound comparison, this study aims to shed light on the relative effectiveness of each approach to better inform clinical practices and guideline development in the field of colorectal surgery.

MATERIALS AND METHODS

This systematic review and NMA adhered to the PRISMA guidelines[16]. PROSPERO registration number, No. CRD420251041181.

Selection criteria

This systematic review and NMA included randomized controlled trials (RCTs) involving adult patients (≥ 18 years) undergoing elective colorectal surgery. Studies were required to compare at least two of the following bowel preparation strategies: Combined MBP and OAB, MBP alone, OAB alone, or NBP. Studies were excluded if they: (1) Involved emergency colorectal surgery; (2) Included pediatric patients; (3) Focused on non-colorectal abdominal surgeries; (4) Were observational studies, non-randomized trials, conference abstracts, or unpublished data; (5) Evaluated intravenous antibiotics without an OAB component as the primary intervention; or (6) Had major differences in Enhanced Recovery After Surgery (ERAS) protocols that could confound the comparison of bowel preparation strategies.

Search strategy and data extraction

A comprehensive search was conducted across PubMed, Cochrane Library, Web of Science, and Scopus using keywords related to colorectal surgery, bowel preparation, and OAB. The search was completed on April 28, 2025. The full search strategy is provided in Supplementary material.

Two authors independently conducted the screening of titles/abstracts and full texts, and conflicts were resolved through discussion with a third author. At least two reviewers independently extracted data from all selected trials using Microsoft Excel. Extracted information included study identifiers (first author, year, country, center, and study period), patient characteristics (sample size, age, gender distribution, and comorbidities), surgical details (approach and type of surgery), and intervention characteristics (type of bowel preparation, substances used, timing, dosage, and any prior treatments).

PRISMA flow chart

The study selection process is summarized in the PRISMA flow chart (Figure 1) and followed the PRISMA 2020 checklist.

Figure 1  PRISMA flow diagram for search and screening processes.

Outcome measures

The primary outcomes were the rates of incisional and organ/space infections following elective colorectal surgery. The secondary outcomes included the incidence of AL, Clavien-Dindo grade I-II (CD I-II) complications, postoperative ileus, sepsis, UTI, length of stay (LOS), and 30-day mortality, where reported. These outcomes were selected to comprehensively assess the comparative safety and efficacy of bowel preparation strategies.

Bias assessment and certainty of evidence

The Cochrane Risk of Bias Tool (RoB 2.0) was used to assess methodological quality of the included RCTs[17]. Two reviewers independently evaluated each domain, and discrepancies were resolved by discussion with a third coauthor. Overall confidence in the network estimates was assessed with the Confidence in NMA[18].

Publication bias

Comparison-adjusted funnel plots were generated to assess small-study effects and potential publication bias. Where 10 or more studies contributed to a comparison, Egger’s regression test was used to assess asymmetry[19].

Statistical analysis

A frequentist NMA was conducted using a random-effects model to account for between-study variability. The NMA was performed using R software version 4.3.2, R Foundation for Statistical Computing, Vienna, Austria, with Netmeta packages[20]. For each comparison, we estimated relative risks (RRs) and 95% confidence intervals (CIs). A lower RR was considered favorable for undesirable outcomes such as SSI and AL. Hence, treatments with smaller effect sizes were ranked higher. To rank the interventions, we derived P-scores using the net rank function, where higher scores reflected more favorable outcomes[21,22].

Heterogeneity was assessed using the I², Tau², and Cochran’s Q test. We assessed network consistency using node-splitting models to compare direct and indirect evidence. Network geometry plots were used to visualize the treatment comparisons and data distribution across interventions. All analyses adhered to the PRISMA-NMA guidelines for transparent reporting of network meta-analyses[23]. Transitivity was assessed by comparing baseline covariates across interventions comparisons using ANOVA and boxplot[24]. Sensitivity analyses involved removing studies judged to be at high risk of bias, excluding trials with small sample sizes, and performing leave-one-out analyses to assess the impact of each individual study on the pooled estimates (Supplementary Figures 1-8; Supplementary Table 1).

RESULTS

Study characteristics

Our analysis, conducted on 38 studies originating from various countries, exhibited diverse characteristics. These studies spanned various study periods and employed different surgical approaches (open, laparoscopic, or both). They encompassed a range of colorectal surgical procedures, including hemicolectomy, colectomy, sigmoid resection, rectal resections, and proctocolectomies. The mechanical preparation substances varied, commonly including polyethylene glycol, sodium phosphate, macrogol, and magnesium preparations. For OAB, a wide array of agents was utilized, including metronidazole, kanamycin, neomycin, erythromycin, levofloxacin, ciprofloxacin, rifampicin, streptomycin, tobramycin, colistin, and tetracycline, which were often administered in two or three single doses on the day before surgery or over two to three days. Patient comorbidities were also documented across studies (Supplementary Tables 2 and 3). The majority of included studies were assessed as having a low risk of bias. However, a subset of trials revealed some concerns, particularly in the randomization process and selection of the reported results. Importantly, no study was judged as having a high risk of bias across any domain. These findings generally indicate a robust and reliable body of evidence, supporting the validity of the results obtained from this NMA. The quality assessment results are summarized in Figure 2. The certainty of evidence was generally low due to concerns of reporting bias, imprecision, heterogeneity, and within-study bias (Supplementary Table 4). Publication bias was generally rated as low concern (Supplementary Figures 9-12).

Figure 2 Quality assessment results. A: Risk of bias assessment for each included study; B: Overall risk of bias summary across the included studies.

Primary outcomes

Incisional SSIs: In a random effects NMA with 7405 participants, the combination of MBP and OAB had significantly lower risk of incisional SSI compared with NBP (RR 0.53, 95%CI: 0.29-0.99, P = 0.04). While OAB alone did not show a significant difference in risk of incisional SSI (RR 0.59, 95%CI: 0.24-1.43, P = 0.24) and MBP alone also had no effect on this outcome (RR 0.99, 95%CI: 0.57-1.76, P = 0.99) (Supplementary Figures 13 and 14). There was some heterogeneity (I² = 61.8%). In particular, heterogeneity within designs was statistically significant (Q = 52.38, df = 20, P < 0.0001). The transitivity assumption was fulfilled and there was no important inconsistency between treatment comparisons. Ranking of the interventions based on P-scores was as follows: The combination of MBP and OAB (0.85), OAB alone (0.72), while both MBP and NBP were equally rated (0.21). The overall certainty of the evidence was very low to low (Supplementary Table 4).

The meta-regression for incisional SSI did not find any significant moderator effects (Q = 9.05, P = 0.25). None of the covariates (male proportion, age, surgical approach, tumor type, or center) was significantly associated with SSI risk (all P > 0.05). The adjusted estimates remained similar to the unadjusted estimate. Sensitivity analysis including leave-one-out, excluding studies judged by some concerns-risk-of-bias, and excluding studies with small sample size was performed, to assess the robustness of our NMA. Across the sensitivity models, combined MBP and OAB consistently showed the greatest efficacy in reducing incisional SSI.

Organ space SSI: In the random effects model of 7535 patients assessing organ space SSI, combined MBP and OAB showed a significant reduction in the risk of organ space SSI with an RR of 0.62 (95%CI: 0.39-0.99, P = 0.04) in comparison to NBP. OAB didn’t show a significant effect in incisional SSI risk, with a RR of 1.00 (95%CI: 0.36-2.80, P = 0.99). Additionally, MBP showed a non-significant trend toward increased organ space SSI risk with a RR of 1.1 (95%CI: 0.72-1.69, P = 0.65) (Supplementary Figures 13 and 14). No significant heterogeneity was detected by the heterogeneity analysis, with an I² of 4.2%. Within-design heterogeneity tests did not reach statistical significance (Q = 21.91, df = 21, P = 0.40 supporting consistency among the included study findings. The transitivity assumption held, with no major inconsistency detected between treatment comparisons. P-scores, which order treatments according to relative effectiveness (higher values reflecting better outcomes), were greatest for combined MBP and OAB (0.93), followed by OAB (0.42), NBP (0.40), and MBP (0.25). The certainty of evidence was generally high, moderate (Supplementary Table 4).

The meta-regression for incisional SSI did not find any significant moderator effects (Q = 7.00, P = 0.87). None of the covariates (male proportion, age, surgical approach, tumor type, or center) was significantly associated with SSI risk (all P > 0.05). The adjusted estimates remained similar to the unadjusted estimate. Sensitivity analysis was performed. Across all models, combined MBP and OAB consistently showed the greatest efficacy in reducing organ space SSI.

Secondary outcomes

AL: In the random effects model assessing AL, OAB (RR = 0.42, 95%CI: 0.18-0.94, P = 0.04) showed a statistically significant decrease in the risk of AL. However, combined MBP and OAB (RR = 0.63, 95%CI: 0.38-1.07, P = 0.09) showed a non-significant decrease in AL risk, and MBP with a RR of 1.08 (95%CI: 0.67-1.74, P = 0.74) showed a non-significant increase in AL (Supplementary Figures 13 and 14). No significant heterogeneity was detected by the heterogeneity analysis, with an I² value of 13.8%. Tests for heterogeneity within designs didn’t reach significance (Q = 29, df = 25, P = 0.26), indicating consistent results. The transitivity assumption held, with no major inconsistency detected between treatment comparisons. The P scores were highest for OAB (0.94), then combined MBP and OAB (0.70), NBP (0.23), and MBP (0.13). The certainty of evidence was generally moderate (Supplementary Table 4).

Mortality: In the random effects model assessing mortality, combined MBP and OAB (RR = 0.79, 95%CI: 0.20-3.11, P = 0.73), OAB (RR = 0.81, 95%CI: 0.16-3.99, P = 0.79), and MBP (RR = 0.59, 95%CI: 0.15-2.35, P = 0.45) showed a non-significant decrease in mortality compared to NBP (Supplementary Figures 13 and 14). The heterogeneity analysis revealed no significant heterogeneity, with an I² value of 0.0%. Tests for heterogeneity within designs were not significant (Q = 2.44, df = 8, P = 0.96), indicating consistency among the study results. The transitivity assumption held, with no major inconsistency detected across treatment comparisons. The P scores were highest for MBP (0.74), then OAB (0.47), combined MBP and OAB (0.46), and NBP (0.33). The certainty of evidence was generally low (Supplementary Table 4).

CD I-II: In the random effects model assessing CD grade I-II, OAB showed a significant decrease in the risk of CD I-II events with an RR of 0.56 (95%CI: 0.35-0.88, P = 0.01) compared to NBP. Combined MBP and OAB and MBP alone showed a non-significant trend toward increased CD I-II risk, with a RR of 1.24 (95%CI: 0.96-1.61, P = 0.09), (RR = 1.31, 95%CI: 0.88-1.96, P = 0.19), respectively (Supplementary Figures 13 and 14). No significant heterogeneity was detected, with an I² value of 15%. Tests for heterogeneity within designs didn’t reach statistical significance (Q = 4.71, df = 4, P = 0.32), indicating consistent results. The transitivity assumption held, with no major inconsistency detected between treatment comparisons. The P scores were highest for OAB (0.99), then NBP (0.62), combined MBP and OAB (0.23), and MBP (0.16). The certainty of evidence was generally moderate (Supplementary Table 4).

Postoperative ileus: In the random effects model assessing postoperative ileus, OAB (RR = 0.71, 95%CI: 0.45-1.11, P = 0.14), combined MBP and OAB (RR = 0.95, 95%CI: 0.66-1.37, P = 0.79), and MBP (RR = 1.10, 95%CI: 0.66-1.85, P = 0.71) showed a non-significant effect on postoperative ileus (Supplementary Figures 13 and 14). No significant heterogeneity, with an I² value of 0.0%. Heterogeneity within designs were not significant (Q = 7.74, df = 10, P = 0.65), indicating consistent study results. The transitivity assumption held, with no major inconsistency detected across treatment comparisons. The P scores were highest for OAB (0.93), then combined MBP and OAB (0.48), NBP (0.37), and MBP (0.22). The certainty of evidence was generally high (Supplementary Table 4).

UTI: In the random effects model assessing UTI, combined MBP and OAB (RR = 2.20, 95%CI: 0.09-54.73, P = 0.63), OAB (RR = 1.56, 95%CI: 0.03-73.06, P = 0.82), and MBP (RR = 2.00, 95%CI: 0.11-37.34, P = 0.64) showed a non-significant increase in UTI risk compared to NBP (Supplementary Figures 13 and 14). The heterogeneity analysis was high, with I² = 61.1%. The heterogeneity within designs were not significant (Q = 5.14, df = 2, P = 0.08). The transitivity assumption held, with no major inconsistency detected between comparisons. The P scores were highest for NBP (0.65), followed by OAB (0.54), MBP (0.43), and combined MBP and OAB (0.38). The certainty of evidence was generally very low (Supplementary Table 4).

LOS: In the random effects model assessing LOS, combined MBP and OAB (mean difference = -0.19, 95%CI: -1.81 to 1.44, P = 0.82) showed a non-significant reduction in LOS compared to MBP (Supplementary Figures 13 and 14). No significant heterogeneity was detected, with an I² of 0.0%. The heterogeneity within designs didn’t reach significance (Q = 0.82, df = 2, P = 0.66). The transitivity assumption held, with no major inconsistency detected between comparisons. The P scores were highest for combined MBP and OAB (0.59), then MBP (0.41). The certainty of evidence was generally low (Supplementary Table 4).

Sepsis: In the random effects model assessing sepsis risk, combined MBP and OAB (RR = 1.22, 95%CI: 0.44-3.42, P = 0.7) showed a non-significant increase in sepsis risk compared to MBP (Supplementary Figures 13 and 14). No significant heterogeneity was detected, with an I² of 0.0%. The heterogeneity within designs didn’t reach significance (Q = 1.65, df = 3, P = 0.65). The transitivity assumption held, with no major inconsistency detected between comparisons. The P scores were highest for MBP (0.65), then combined MBP and OAB (0.35). The certainty of evidence was generally moderate (Supplementary Table 4).

DISCUSSION

This NMA provides comprehensive evidence on the effectiveness of various bowel preparation strategies in reducing postoperative complications in CRC. These strategies include MBP, OAB, NBP, and a combined MBP and OAB approach. By comparing these regimens in a network framework, we were able to evaluate both direct and indirect evidence and rank the interventions according to their relative performance across several clinically important endpoints.

The main findings of this NMA indicate that the combination of MBP and OAB is the most effective strategy for reducing both incisional and organ-space SSIs. OAB-containing regimens demonstrated the strongest performance for AL and CD I-II complications. No strategy showed a clear advantage for postoperative ileus, UTI, LOS, or sepsis, while mortality rates were too low to detect meaningful differences.

For incisional SSI, the combination of MBP and OAB significantly decreased events, demonstrating a statistically significant effect. Additionally, it has the highest P-score, reflecting its high probability of being among the best treatments. OAB alone had the second-highest P-score and a non-significant effect. These findings suggest that regimens incorporating OAB when combined with mechanical cleansing are likely to be the most effective in preventing incisional wound infections. Notably, MBP alone showed virtually no benefit (RR = 0.99). This finding underscores that MBP in isolation may confer little to no protective effect against incisional infections and exposes patients to unnecessary discomfort, fluid shifts, and electrolyte disturbances without clear benefit.

The only intervention that was significantly associated with organ-space SSI was the combination of MBP and OAB. It also had the highest P-score. Organ-space SSI was not significantly impacted by OAB alone or MBP. This implies that the synergistic effect of mechanical cleansing and antibiotic decontamination extends beyond superficial surgical wounds and contributes to reducing deeper, organ-space infections, which are associated with higher morbidity and resource utilization.

In terms of AL, the leakage rates were significantly associated with OAB, with the highest P-score, followed by the combination approach. Similarly, other studies revealed that a combination of MBP and OAB is more efficient in lowering the rate of AL in comparison to MBP alone, which aligns with Tan et al[25]. Our findings reveal that MBP had the lowest ranking, suggesting a potential increase in risk. One explanation is that MBP may induce transient bowel wall edema, altered mucosal integrity, or microbiome dysbiosis that could partially offset the protective antimicrobial effects of OAB at the anastomotic site[26]. These results favor OAB-containing regimens. The strong performance in reducing AL with OAB alone contrasts with the lack of significant effect on organ-space SSI. This may reflect the need for physical clearance of fecal material to fully mitigate deeper contamination and micro-abscess formation in the peritoneal or pelvic spaces, whereas AL is more directly influenced by intraluminal bacterial burden at the suture line, which OAB effectively targets.

Regarding mortality, none of the studied treatments exhibited a statistically significant impact on postoperative death. MBP had the highest P-score for mortality, suggesting it may be linked to a lower risk; however, the effect cannot be accurately assessed due to the large CIs. Mortality after elective colorectal surgery is relatively infrequent and is usually the result of a cascade of complications rather than any single event.

For CD I-II complications, OAB by itself significantly decreased events. It was also the most effective medication for lowering CD I-II complications, with the highest P-score. There were no discernible decreases for the combination of MBP and OAB therapy for this endpoint. This suggests that OAB alone may exert a broader effect on low-grade postoperative morbidity, including milder infectious and functional disturbances that still matter for patient recovery and resource use.

None of the interventions showed a statistically significant effect on postoperative ileus, UTI, LOS, or sepsis. For postoperative ileus, OAB showed a favorable trend, but the effect was not significant. For UTIs, the results were imprecise with very low certainty. Similarly, the mean difference in LOS for the combination of MBP and OAB was not statistically significant, and for sepsis, NBP showed a clear benefit, with MBP having a slightly higher P-score than the combination therapy. This result should be interpreted with great caution. The CIs were extremely wide, and no strategy showed a statistically significant effect. These outcomes are likely influenced by a multitude of factors beyond bowel preparation, such as opioid use, fluid management, and surgical technique, which may diminish any potential impact from the preparation strategy alone[27,28].

Although some of our analyses did not show statistical significance for certain endpoints, the combination of MBP with OAB and OAB alone frequently showed a reduced risk of postoperative infectious complications. This is in line with other meta-analyses and current research showing that the best way to lower SSI rates following elective colorectal surgery is to combine MBP and OAB[29-32]. For instance, the combination of OAB and MBP dramatically lowers the risk of SSI, according to a 2023 study by Koo et al[29], which attributes this advantage to the OAB sufficient aerobic and anaerobic coverage. A number of reasons could be responsible for the lack of full statistical significance in some of our primary analyses, such as heterogeneity among included studies, changes in sample sizes, different types of SSI, or variations in surgical techniques. Even though some estimates fall short of conventional statistical thresholds in this particular dataset, the high P-scores for the combination of OAB and MBP and OAB alone highlight their relative efficacy and point to a significant potential for benefit. Koskenvuo et al[31] found in their trial that the combination should be regarded as a standard regimen for patients undergoing elective rectal resection, as it lowers overall postoperative complications as well as rates of incisional SSI, organ space SSI, and AL.

Other NMAs and studies supported that the combination of MBP and OAB is associated with the lowest risk of incisional SSI[29,30,32]. For instance, the combination of MBP and OAB was linked to the lowest incidence of incisional and organ space SSI, according to the 2018 NMA, with OAB coming in second[30]. This significant impact is explained by the fact that incisional SSIs are frequently brought on by enteric bacterial contamination of the surgical site. The combined action of MBP and OAB is mostly responsible for this strong effect. OAB significantly reduces the bacterial burden, whereas MBP physically reduces the fecal mass inside the intestines. One of the main causes of incisional SSI, bacterial contamination of the surgical wound during incision and closure, is reduced by this dual process[33]. Furthermore, only the combination of MBP and OAB exhibited a statistically significant decrease in organ space SSI in our analysis, indicating that the preventive impact covers internal organs and goes beyond superficial wound infections to deeper infections. The idea is supported by the MECCA trial, which shows that MBP eliminates fecal bulk and OAB targets gut flora and that their combination significantly lowers organ space SSI rates, even though it does not specifically address this mechanism[34]. This combination’s high P-score emphasizes how effective it is in this area.

For AL, OAB alone showed a statistically significant reduction in our NMA, with the highest P-score. However, the combination of MBP and OAB showed non-significant results. Beyond bowel preparation and antibiotics, anastomotic dehiscences are severe and complicated consequences that depend on several other factors, including surgical technique, patient health, and tissue status[35]. Other studies have confirmed the superiority of the OAB alone and combination of MBP and OAB in comparison to MBP alone[36], as well as in comparison to another standard care, such as MBP with intravenous antibiotics[37]. The evidence remains somewhat inconsistent across the literature. For instance, a systematic review by Rollins et al[27] noted varying results regarding the impact of MBP and OAB on AL rates[38]. Our findings, therefore, fit within a broader body of heterogeneous data but still support the view that OAB-containing regimens are likely beneficial for anastomotic integrity.

Mortality was not significantly affected by any of the interventions. Given that postoperative complications are frequently only contributing factors rather than the only cause of postoperative death, which is the consequence of a complex interaction of patient, surgical, and perioperative factors, this is to some extent predicted. On the other hand, lowering long-term mortality and improving overall patient outcomes can be indirectly achieved by lowering complications like SSIs and AL[31].

Although MBP’s high P-score for mortality points to a possible favorable trend, the large CI highlights the need for more study. OAB’s essential role in lowering general complications was shown by the statistically significant decrease in CD I-II problems. This implies that the effectiveness of MBP and OAB as interventions in preventing particular kinds of problems varies. While OAB may be more successful in lowering difficulties caused by the total bacterial load, MBP may be more effective in reducing complications directly caused by fecal volume or higher mechanical contamination[30].

Our findings strongly support the current clinical guidelines that increasingly advocate for the routine use of combined MBP and OAB before elective colorectal surgery to minimize the risk of postoperative complications[28,39]. This combination strategy is advised by prominent surgical associations and organizations, including the American Society of Colon and Rectal Surgeons[40] and the ERAS Society[41]. To lower the incidence of SSIs, recent guidelines increasingly advise using MBP and OAB together before elective colorectal surgery[32,42]. For instance, MBP should be taken in conjunction with OAB during colon surgery, according to the ERAS Society guidelines[41]. Before elective colorectal surgery, American guidelines also advise routinely using OAB in conjunction with an isosmotic MBP[40]. The numerous advantages of OAB and MBP in lowering surgical complications are highlighted by this NMA. These results validate existing recommendations and serve as a foundation for further study to completely comprehend the underlying processes of these impacts. Future studies should focus on: (1) Head-to-head RCTs comparing MBP + OAB vs OAB alone, particularly for anastomotic integrity; (2) The impact of different OAB regimens and MBP agents; (3) Implementation strategies to increase adherence to recommended bowel preparation; and (4) Evaluation of these approaches in emergency colorectal surgery and diverse patient populations.

Limitations

This study has several limitations that warrant consideration. First, there was substantial heterogeneity in antibiotic protocols, including variations in the type, dosage, and timing of OAB administration, which may have affected the precision of efficacy estimates for both OAB and MBP strategies. Second, inconsistent definitions and reporting of SSIs across studies may have introduced misclassification bias, despite our efforts to standardize outcome categories. Third, some treatment comparisons, particularly OAB vs NBP and OAB vs combined MBP and OAB for organ-space SSIs, and secondary outcomes, were based on limited data, resulting in sparse networks and wide CIs. Fourth, the included trials did not report compliance rates or completion of the preparation in detail. Consequently, the pooled estimates may implicitly assume optimal adherence, which may not reflect real-world practice. This limitation may affect the generalizability of our findings. Lastly, although we assessed for publication bias, the potential for selective reporting and small-study effects cannot be fully excluded.

There are also a number of NMA-specific limitations. The observed effects might have been influenced by clinical heterogeneity (e.g., patient characteristics, intervention protocols), methodological heterogeneity (e.g., study design and quality), and statistical heterogeneity among the included studies, particularly for some endpoints where positive trends did not reach statistical significance. Second, variations in patient demographics or surgical techniques among studies may contradict the transitivity assumption, which is crucial for indirect comparisons in NMA and may bias comparisons. Third, because studies with non-significant results are less likely to be published, publication bias may have contributed to the overestimation of treatment effects. Furthermore, some comparisons (like mortality) have small sample sizes and a limited number of studies, which raises uncertainty and decreases precision. Consistency may also be impacted by differences in how outcomes, such as leakage, are defined and measured among studies. Furthermore, the particular patient groups and surgical techniques that are covered may not accurately represent larger clinical contexts, which limits generalizability. The strength of the evidence in some areas of the network is further constrained by the absence of direct head-to-head comparisons for every treatment pair. Finally, unmeasured confounding, like variations in organizational procedures, surgical skill, or adherence to improved recovery protocols, may have affected results but was not always disclosed or controlled for in different trials.

CONCLUSION

This NMA supports the current evidence that OAB-containing regimens offer clinically meaningful advantages. Our findings highlight that the combination of MBP and OAB may be useful for eligible patients undergoing elective surgery; individualized approaches may be needed. For preventing SSI, especially incisional and organ-space SSI. Furthermore, both OAB alone and the combination approach significantly reduce AL, while OAB alone is effective for reducing minor CD I-II complications. These complicated findings are not only in line with current recommendations but also provide a foundation for additional research that will completely clarify the complex mechanisms underlying these advantageous benefits and investigate their relevance to a larger variety of surgical procedures. To improve postoperative recovery and patient safety, these evidence-based therapies must be continuously incorporated into clinical practice.