Total mesorectal excision (TME) is the standard surgical treatment of rectal cancer, which can be performed using open, laparoscopic (L-TME), robot-assisted (R-TME), and transanal TME (TaTME) [1]. All three minimally invasive techniques are considered standard care and offer comparable intraoperative, postoperative, or oncological outcomes when performed by experienced surgeons [2, ]. With comparable oncological outcomes, patients often prioritize postoperative recovery and quality of life (QoL) in their decision making. As survival rates improve, patients may have to cope with reduced functional outcomes and lower QoL for an extended period [3, ]. Therefore, it is essential to assess the evidence regarding functional outcomes and QoL of these techniques.

Research has mostly focused on the oncological results of minimally invasive techniques for rectal cancer treatment, leaving limited evidence concerning their functional outcomes and QoL [4]. The available evidence is contradictory, with some studies suggesting that R-TME and TaTME techniques increase the nerve-sparing dissection rate and improve functional outcomes by optimizing the visualization of adjacent structures and enhancing precision in dissection around the distal rectum [5, 6]. However, other studies have found no evidence for the advantages of R-TME [4, 7] and some have even suggested unfavourable outcomes of TaTME due to anal canal dilatation and the risk of nerve injury during the learning curve [8, 9]. The existing systematic reviews [7, 10,11,12,13] have not included the most recently published studies and only compare two minimally invasive techniques. Furthermore, these reviews also rely on retrospectively gathered data about functional outcomes and QoL.

To address this gap, an updated, comprehensive systematic review and meta-analysis of prospectively collected data were conducted to examine the functional outcomes and QoL of open, L-TME, R-TME, and TaTME techniques for rectal cancer treatment.

Methods

This systematic review and meta-analysis were conducted following Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines and in line with the protocol agreed by all authors. The study protocol was written a priori following the PRISMA-P statement and registered in the International prospective register of systematic reviews (Prospero #240,851) [14].

Data selection

In accordance with the population, intervention, comparison, outcomes, and study design (PICOS) framework, the study eligibility criteria were selected. Population: adult patients with rectal cancer undergoing TME. Interventions: open TME, L-TME, R-TME, and TaTME. Comparisons: studies were deemed eligible independently of head-to-head comparisons. Outcomes: functional outcomes, measured by any validated scale, defined as the concrete consequences of TME surgery in rectal cancer patients. These include urinary, sexual, and fecal function. QoL, measured by any validated scale, is defined as the personal perception of the impact of illness or treatments on physical, psychological, and social well-being [15]. The various domains of QoL were analyzed separately. These include physical, social, and psychological QoL. However, due to variations in questionnaires used per study and definitions used, functional outcomes and QoL were reported as described in individual studies, and different reporting outcomes were evaluated for inclusion on a case-by-case basis. Study design: randomized-controlled trials (RCTs), prospective studies, and retrospective studies reporting from a prospective database. Excluded were studies that met at least one of the following criteria: (a) not published in English, French, German, or Spanish, (b) less than six months follow-up, (c) case reports, case series, letters, editorials, conference abstracts, commentaries, and reviews, and (d) full-text unavailable. In case of overlap** populations between studies, the study with the largest sample size and longest follow-up were included.

The systematic search was supported by a librarian experienced in assisting systematic reviews and peer-reviewed by a second independent librarian. The search dates were from January 1st, 2000, to September 1st, 2023. The pre-defined complete search strategy prioritized sensitivity above specificity and is available in Supplementary File 1. Studies were assessed for eligibility through searches of the PubMed, Medline, Embase, Scopus, Web of Science, and Cochrane Library databases. In addition, databases of ongoing (unpublished) trials (i.e., World Health Organisation (WHO) Registry Network (including ClinicalTrials.gov) and PROSPERO were searched. Two independent reviewers (RG, CM) identified potentially eligible studies through title and abstract screening using Rayyan QCRI, a web-based software management program for systematic reviews. Disagreements were solved through discussion, in which two additional authors were involved (AR and EC). For literature saturation, reference lists of included studies were hand-searched for additional relevant studies using systematic “snowball” procedure guidelines [16].

Data extraction

Two independent reviewers (RG, CM) extracted data from the included studies using a standardized data extraction form. Prior to data extraction, training was provided for using the form. The tabulated data were used for evidence synthesis and quality assessment. Disagreements were solved through discussion, in which two additional authors were involved (AR and EC). For each eligible study, the following information was collected: study characteristics (first author, journal and year of publication, country, study design, aim, primary/secondary outcomes, in/exclusion criteria, study period, type of surgery, number of patients, length of follow-up), patient demographics (gender, age, body mass index, tumour location), preoperative data (disease stage, neoadjuvant treatment, American Society of Anaesthesiologists (ASA) score), intraoperative data (number of surgeons, surgeon experience, operation time, type of procedure, stoma rate, conversion rate), postoperative data (circumferential resection margin involvement and Clavien-Dindo classification of postoperative complications), and functional outcomes and QoL. In case of missing relevant data, the study's corresponding authors were requested for additional data. A reminder email was sent up to three times.

Quality assessment

The quality of included studies was independently appraised by two reviewers (RG, CM) using a Modified Methodological Items for Non-Randomized Studies (MINORS) tool [17] (Supplementary File 2 and 3). The adjusted MINORS tool consists of 12 questions for comparative studies (scale: 0-1-2, max score: 24). Studies with scores ≤ 18 were considered of low quality. In the MINORS tool item assessing whether the follow-up period was appropriate to the aim of the study, the cut-off was a priori set at 12 months postoperatively. This follow-up period was chosen as most events influencing morbidity or mortality will likely occur within one year postoperatively. Criteria 9–12 of the MINORS tool were assessed in the case of comparative studies. The control and study group were considered contemporary if these were managed no more than five years apart. Disagreements were resolved through discussion, in which two additional authors were involved (AR and EC). The Grading of Recommendations Assessment, Development and Evaluation working group approach (GRADE) was used to assess the quality of evidence for the functional outcomes and QoL [18]. The quality of evidence was reported as high, moderate, low, or very low.

Data analysis

A meta-analysis for functional outcomes was performed if ≥ 3 comparative studies reported functional outcomes of urinary, sexual, or fecal function. A meta-analysis for QoL was performed if ≥ 3 comparative studies reported QoL data based on validated questionnaires such as the European Organization for Research and Treatment of Cancer (EORTC) questionnaires. Data from studies was only included in the meta-analyses if the number of patients that participated in the questionnaires was available. The Cochrane Handbook 6 was used as a guideline for these analyses [19]. Standardized mean differences with 95% confidence intervals (CIs) were calculated for continuous variables. As we anticipated considerable between-study heterogeneity, a random-effects model was used to pool effect sizes and checked using a fixed-effect model. Given its robust performance in continuous outcome data, the restricted maximum likelihood estimator was used to calculate the heterogeneity variance τ2 [20]. Knapp-Hartung adjustments were used to calculate the confidence interval around the pooled effect [21]. If the requested data were unavailable, mean(s.d.) values were calculated for the overall analysis, if possible [22]. Outcome reporting bias was assessed by comparing outcomes reported in study protocols to the final published article. As subgroup analyses depend on the statistical power, no subgroup analyses were performed since the number of studies is small (k =  < 10). Following Cochrane guidelines, we did not investigate publication bias as our search considered less than ten studies for each data comparison.

A qualitative synthesis of functional outcomes and QoL was performed to present the findings of studies that were excluded from the meta-analysis, following the European Social Research Council Guidance on the Conduct of Narrative Synthesis in Systematic Reviews [23]. Categorical data were summarized using numbers and percentages. Continuous variables were summarized as means and standard deviations (SDs) or median and interquartile ranges (IQRs). Functional outcomes and QoL were compared between techniques. Studies comparing techniques head-to-head were prioritized.

Results

Study selection

After the removal of duplicates, a total of 5939 citations were identified (Fig. 1). Following the abstract and full-text screening, 19 studies were included for qualitative analyses [2, 24,25,7, 48,49,50]. Additionally, in most included studies, urinary dysfunction was not a primary endpoint, rendering them underpowered to detect statistically significant differences. Notably, in contrast to the other studies included in the meta-analysis, the study by Machakova et al. [35] reported urinary outcomes favouring L-TME instead of R-TME. Due to the limited number of studies included in the meta-analysis, this small sample study may have significantly influenced the outcomes of the meta-analysis regarding urinary and sexual function.

Regarding sexual function, quantitative analyses did not reveal any significant differences between the techniques. These results concur with the qualitative analyses of studies excluded from the meta-analyses. However, contrasting evidence from prior systematic reviews and meta-analyses suggested significantly better R-TME outcomes than L-TME at three and six months postoperatively [42,43,44,45]. Despite this, most studies in our review found no significant differences in sexual function among the techniques. This could be attributed to low response rates and the use of different questionnaires for men and women, leading to small sample sizes and underpowered studies. Nonetheless, certain studies did report better sexual function after R-TME and TaTME compared to conventional laparoscopic surgery. As discussed previously, if there even are any differences favoring novel TME techniques, this may be because of technical differences between the techniques, with R-TME and TaTME potentially reducing the occurrence of autonomic nerve injury [51]. Discrepancies in sexual outcomes may also be due to variations in anastomosis rates between the techniques. R-TME and TaTME could facilitate the construction of a low anastomosis in patients who would otherwise require an abdominoperineal resection (APR) [52]. APR includes extralevator perineal excision, associated with a higher risk of neurovascular bundle injuries and worse sexual outcomes. Patients undergoing APR also receive a permanent stoma, negatively affecting sexual function [53]. Notably, some studies observed improved sexual function after R-TME and TaTME exclusively in male patients, hinting at potential advantages in specific patient subgroups, such as those with a narrow pelvic region.

In terms of fecal function, qualitative analyses failed to uncover any significant differences among the techniques. Nonetheless, certain studies reported significant differences in fecal function after R-TME and TaTME compared to conventional laparoscopic surgery. Van der Heijden et al. included 55 patients receiving TaTME and L-TME and reported worse LARS scores for TaTME at 12 months postoperatively. This contradicts prior studies suggesting worse anorectal outcomes for TaTME only during the initial six months of recovery, attributed to anal stretch and dilatation [54, 55]. The differences in fecal function outcomes after R-TME and TaTME could potentially be attributed to patient selection bias. Surgeons may have chosen R-TME or TaTME for patients with a very low tumour height, avoiding intersphincteric APR. Consequently, these groups may consist of more patients receiving a low anastomosis, which, due to reduced rectal reservoir capacity, could lead to major LARS [56].

Regarding QoL, the qualitative analyses did not discern any differences between the techniques. Nevertheless, some studies did report significant differences in QoL between TME techniques. For instance, Ng et al. compared 25 patients receiving open surgery with 49 patients receiving L-TME and found significantly better outcomes for L-TME at four and eight months postoperatively. These findings were postulated to reflect the superior short-term clinical outcomes of minimally invasive surgery versus open surgery [57,58,59,60,61]. Additionally, Mei et al. reported improved outcomes for TaTME compared to laparoscopic surgery at 12 months postoperatively, although this study juxtaposed TaTME with L-APR. However, as previously mentioned, differences between type of procedure performed could have a notable impact on patients’ QoL. The absence of significant differences in QoL might be due to the inherent complexity of this subjective outcome. Compared to urinary and sexual function, QoL is influenced by even more multifaceted factors, including co** abilities and social environment, and the technique of surgery performed may only have a small effect on global QoL.

The absence of significant differences unearthed by this review could be due to the poor quality of available evidence. Most of the literature consisted of small non-randomized trials that assessed functional outcomes or QoL as secondary endpoints, resulting in a substantial lack of power, particularly regarding open and TaTME. Acknowledging the limitations in the number of patients receiving open and transanal TME in the studies included in this review, this study aimed to provide an overview of the currently available evidence regarding all TME techniques and focused exclusively on studies with prospective data collection for functional outcomes or QoL. Incorporating studies with retrospective data collection could have increased the overall number of patients in our analysis, thereby enhancing statistical power. However, retrospective data collection methods are susceptible to recall bias and may not provide a comprehensive and accurate representation of the patients' true QoL and functional outcomes. Thus, prioritizing prospective data collection enhances the reliability and validity of our findings. Nevertheless, to comprehensively assess functional outcomes and QoL, future studies should be meticulously designed, prospective, and sufficiently powered. Moreover, patient-related factors such as co** skills, social context, and psychological well-being largely influence subjective functional outcomes and QoL, highlighting the necessity of considering these confounders when evaluating these parameters. Beyond patient-related factors, the choice of surgical procedure and clinical outcomes, including postoperative morbidity, can significantly impact reported outcomes, making it difficult to isolate the exclusive effect of surgical technique. Notably, whilst the presence of a temporary or permanent ostomy may significantly affect patients’ functional outcomes and QoL, the studies included in this review often did not report the numbers of patients receiving an ostomy, nor were these factors considered when evaluating patient reported outcomes. Thus, future studies are strongly recommended to account for these confounding factors when assessing functional outcomes and QoL. Additionally, several studies in our analyses compared experienced laparoscopic surgeons to surgeons’ initial experience with robot-assisted surgery, necessitating adjustment for the learning curve's effects [26, 30]. A fair evaluation of techniques should involve comparing surgeons with similar experience levels. Consequently, we eagerly await the forthcoming results of the Vantage trial [62], which will provide prospective insights into the functional and QoL outcomes of surgeons who have surpassed the learning curve. Lastly, functional outcomes and QoL assessments often lacked clear definitions and employed questionnaires not validated for rectal cancer patients, resulting in considerable heterogeneity between groups and rendering meaningful statistical comparisons difficult.

Conclusion

In summary, this systematic review and meta-analysis revealed a significant gap in the literature concerning the evaluation of functional outcomes and QoL after TME for rectal cancer treatment. Existing studies are limited in number and frequently reported functional outcomes and QoL as secondary endpoints. Acknowledging the limited strength of the evidence, this systematic review and meta-analysis found no significant differences in urinary, sexual, and anorectal function, as well as QoL, across various TME techniques. While innovative techniques like R-TME and TaTME have shown oncological safety, it is imperative to undertake high-quality, prospective trials with sufficient statistical power to assess functional outcomes and QoL comprehensively. Such studies should also account for the impact of the learning curve and baseline patient characteristics, ensuring a more robust evaluation of these critical parameters in rectal cancer surgery.