Endoscopic versus Open Carpal Tunnel Release: A Systematic Review and Meta-analysis Authors: Mahgol Sadat Hassan Zadeh Tabatabaei, Ahmad Pour-Rashidi, Alireza Noori, Vali Baigi, Melika Alaeddini, Zahra Ramezani, Rasha Atlasi, Forough Yazdanian, Mohammad Mirahmadi Eraghi, Vafa Rahimi-Movaghar Published: 2026 ABSTRACT Carpal tunnel syndrome accounts for the most prevalent type of peripheral neuropathy encountered by surgeons in clinics worldwide. Despite several studies investigating endoscopic carpal tunnel release (ECTR) in comparison to open carpal tunnel release (OCTR), the best option is yet to be identified. In this systematic review and meta-analysis, we sought to evaluate and compare the surgery outcomes and complications associated with ECTR and OCTR by analyzing evidence from randomized controlled trials (RCTs) to determine if one technique offers superior results compared with the other. In this study, the systematic search was performed using MEDLINE and Embase search engines (via OVID SP), Scopus, Web of Science, and Cochrane Library until April 2024. We hand-searched references of the included studies as well. All RCTs that compared ECTR and OCTR approaches were included. Data on key clinical outcomes and complications, including return to work, revision surgery, nerve injury, operation time, and overall complications, were extracted and analyzed. A meta-analysis was performed if at least three studies reported the prognostic factor. The random-effect models were used. Thirty-three RCTs were included in our study, comprising 1,568 OCTR and 1,447 ECTR. ECTR was associated with shorter time to return to work in comparison with OCTR (mean difference [MD]: −8.08 days; 95% confidence interval [CI]: −12.71 to −3.46). Furthermore, ECTR had lower scar-related complication rates compared with OCTR (odds ratio [OR]: 0.18; 95% CI: 0.08 to 0.37). The meta-analysis showed no significant differences between ECTR and OCTR in revision surgery (OR, 0.78; 95% CI: 0.34 to 1.78), permanent nerve injury (OR: 1.62; 95% CI: 0.08 to 31.81), transient nerve injury (OR: 2.29; 95% CI: 0.73 to 7.20), operation time (MD, −3.22; 95% CI: −9.53 to 3.08), and overall complications (OR: 0.88; 95% CI: 0.49 to 1.61). Based on randomized evidence, ECTR was associated with a shorter time to return to work and fewer scar-related complications compared with OCTR. The assessment of revision surgery, permanent nerve injury, transient nerve injury, operation time, and overall complications showed no significant advantage for any of the techniques. Introduction Carpal tunnel syndrome (CTS) accounts for the most common peripheral neuropathy that surgeons face in their clinics. 1 It has different stages, from mild distal upper-limb impaired sensation and discomfort to severe sensory disturbances along with muscular atrophy and decreased hand dexterity. 2 3 CTS affects 2.3% of the working population annually, resulting in around 600,000 surgeries in the United States each year. It not only impacts individuals but also increases sick leave and strains health care systems. 3 4 5 6 In parallel with the rising surgical volume, contemporary practice has increasingly shifted toward outpatient-based and minimally invasive workflows, emphasizing faster recovery and optimized perioperative care. 7 The management of CTS depends on the severity of the symptoms identified in the examination and other diagnostic tools, some of which include ultrasonography and electrodiagnostic studies. 2 8 Conservative treatment, which is recommended in the early phases of the disease, can slow down the disease progression. However, symptoms progress even after conservative management and also this approach might not be as useful in the advanced stages. 1 9 10 Accordingly, various surgical techniques emerged since 1933 when Dr. Paget suggested an operation to release the median nerve. 11 The endoscopic technique was not introduced until 1987 when Dr. Okutsu reported the use of an endoscope to divide the transverse carpal ligament (TCL) in a patient with CTS. 12 With the advancement of surgical techniques and technology, other surgical methods have also been introduced to lower the complications and improve the outcomes. Recent advances have focused on minimizing surgical trauma while maintaining safety and effectiveness, leading to ongoing refinement of both open and endoscopic techniques. The two surgical techniques that are commonly used in practice are open and endoscopic carpal tunnel release (OCTR and ECTR). 13 14 These two techniques and their respective advantages and disadvantages have been compared in relation to patient satisfaction, complications, costs, and long-term outcomes. 15 16 Despite numerous studies evaluating OCTR versus ECTR, the optimal approach has yet to be determined. Previous systematic reviews have reported conflicting results, particularly regarding key aspects of open and endoscopic techniques, such as iatrogenic nerve injuries. 17 18 19 Recent studies have provided updated comparative data on minimally invasive hand surgery techniques, highlighting that technical variations within both open and endoscopic approaches may influence clinical outcomes. 20 In this systematic review and meta-analysis, we aimed to compare clinical outcomes and complication profiles of ECTR and OCTR using evidence exclusively derived from randomized controlled trials (RCTs). Materials and Methods Literature Search This study was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. 19 In this study, the search was performed using MEDLINE and Embase search engines (via OVID SP), Scopus, Web of Science, and Cochrane Library. Reference lists of included studies were also hand-searched to ensure comprehensive coverage. The search strategy, including the Medical Subject Headings (MeSH) terms, Excerpta Medica Tree (Emtree) terms, and keywords, is shown in Supplementary Material S1 (available in the online version). All relevant terms and keywords were investigated within the titles, abstracts, or keywords of literature entries. There were no restrictions on the time of publication and language, and the final search was performed on April 20, 2024. After eliminating duplicates, two reviewers independently screened titles and abstracts, followed by full-text assessment of potentially eligible studies. Discrepancies were resolved through discussion with a third reviewer. Study Selection All RCTs comparing the ECTR and OCTR techniques were eligible for inclusion in this study. Case–control studies, case reports, case series, letters to the editor, congress abstracts, review articles, non-human studies, and technique studies were excluded. Quality Assessment Risk of bias was assessed using the RoB 2 tool in accordance with the Cochrane Handbook for Systematic Reviews of Interventions. 20 Furthermore, the risk of bias visualization tool (robvis) was utilized to depict the risk of bias assessments for included studies. 21 Two separate reviewers evaluated the potential risk of bias, and any disagreements were addressed through discussion until a consensus was achieved. Data Extraction Four investigators in two groups thoroughly read the full text of the drafted articles and gathered specific information, including: first author, the year study was published, country where the study took place, number of cases, number of cases in both groups, age, types of surgery (i.e., OCTR or ECTR), the anesthetic method used, the side of the operation, time to return to work, scar-related complications (i.e., scar hypertrophy, scar pain, or/and scar hyperesthesia), revision operation, operation time, permanent nerve injury, transient nerve injury (i.e., neurapraxia and numbness), and overall complications (i.e., damage to nerves, blood vessels, and tendons, and wound infection). If the two reviewers had differing opinions on their assessments, a third reviewer was brought in to evaluate the situation and make the final call on the matter. The extracted data were recorded in a preformatted spreadsheet. Statistical Analysis Meta-analysis was performed by STATA version 14 (Stata Corp., College Station, Texas, United States). Results were summarized as weighted mean differences (MDs) with 95% confidence intervals (CIs) for continuous outcomes and as odds ratios (ORs) with 95% CIs for dichotomous outcomes. Meta-analyses were performed when at least three studies reported the same outcome. The presence of statistical heterogeneity in the effect sizes across the studies was assessed using the I 2 statistic, which is derived from the Cochrane Q statistic and also the τ statistic. I 2 statistic indicates the extent of observed variation in study estimates that can be attributed to heterogeneity. A 0% value of I 2 statistic indicates no observed heterogeneity, while higher percentages suggest rising levels of heterogeneity. Tau reflects an estimate of the variance between studies. 2 The random-effects models were used for all studies. Studies were excluded from meta-analysis if the necessary data could not be extracted from the manuscript. Before excluding these studies, the corresponding authors were contacted if they could provide us with the data needed for meta-analysis. Results Search Results The search results are illustrated in Fig. 1 . A total of 5,070 records were identified. After removing 2,105 duplicates and 2,820 irrelevant studies, 145 articles were assessed for eligibility. Ultimately, 33 articles met our inclusion criteria. The characteristics of the 33 included studies are outlined in Table 1 . Flow diagram of the study. Table 1Characteristics of included studiesAuthor (first author)YearCountryNo. of casesAge (years)Surgical techniqueSideAnesthesiaOpenEndoOpenEndoOpenEndo Agee 25 1992 United States6582NSNSConventional1 portalB, UGA, L Brown 36 1993 United States8277NSNS3.5–4.5 cm2 portalsB, UNS Erdmann 26 1994 United Kingdom5253NSNS2 portalsB, UGA Dumontier 40 1995 France405650.753.43–4 cm2 portalsNSNS Sennwald 29 1995 Switzerland2225NSNSNS1 portalNSL Jacobsen 22 1996 Sweden1616NSNSNS2 portalsB, UL Mackenzie 70 2000 United States1422NSNSShort1 portalB, UGA, L Ferdinand 27 2002 Australia202142.748.1Mini 1.5 cm1 portalBL Trumble 32 2002 United States959753 (16)45 (15)3–4 cm1 portalB, UL MacDermid 33 2003 Canada329150 (15)54 (15)Traditional2 portalsNSNS Saw 23 2003 United Kingdom7674NSNS2 cm1 portalB, UL Wong 39 2003 Hong Kong3030NSNSShort2 portalsBL Atroshi 13 2006 Sweden656344444 cm2 portalsUNS Rab 71 2006 Austria1010NSNS2 incisions 5 cm2 portalsBNB Malhotra 30 2007 India313045.345.6Short1 portalB, UGA, L Nascimento 72 2007 Brazil273033.9 (Range: 26–59)35.6 (Range: 24–76)NSNSUNS Atroshi 62 2009 Sweden656344.2 (8.8)44.6 (8.5)4 cm2 portalsNSL Aslani 1 2012 Iran6432NSNSRegular, mini palmar2 portalsB, UNS Ejiri 37 2012 Japan505058 (Range: 28–78)59 (Range: 41–82)Palmar1 portalB, UL Kang 16 2013 South Korea525255 (10)55 (10)Mini2 portalsBGA Michelotti 73 2014 United States303030 (54)30 (54)3 cm1 portalBL Atroshi 34 2015 Sweden6163Conventional2 portalsNSNS Gumustas 74 2015 Türkiye202142.7 (20) (Range: 30–59)48.1 (21) (Range: 29–64)One incision2 incisionsNSNS Zhang 75 2015 China1386946 (Range: 33–65)48 (Range: 34–65)Double small, standard2 portalsUNB Orak 76 2016 Türkiye282249 (8.21)45.1 (8.54)3–4cm2 portalsUL Zhang 35 2016 China1605347 (Range: 29–67)44 (Range: 34–65)With sub-neural reconstruction2 portalsUNB Oh 77 2017 South Korea323551.8 (10.4)59.2 (10.6)1.5 cm miniNSUL, GA Fernandes 78 2018 Brazil1515NSNS4 cm1 portalBNS Martínez-Catasús 41 2019 Spain523557.75 (14.28)50.4 (15.62)1 cm short1 portalNSL Gaspar 79 2019 United States303049.1 (7.1)49.7 (9.0)2 cm2 portalsUNS Michelotti 80 2020 United States303054543 cm1 portalBL Chen 24 2021 China464848.0 (5.33)50.8 (7.97)1 cm1 portalUNS Schwarm 31 2022 Germany1822Median 54 (IQR: 37.0–76.3)Median 63 (IQR: 45.0–73.0)3–4 cm1 portalULAbbreviations: B, bilateral; C, conventional surgery; GA, general anesthesia; IQR, interquartile range; L, local; NB, nerve block; NS, not stated; R, regional; S, small incision surgery; U, unilateral.Note: Age is reported as mean (SD) when possible. Characteristics of included studies Abbreviations: B, bilateral; C, conventional surgery; GA, general anesthesia; IQR, interquartile range; L, local; NB, nerve block; NS, not stated; R, regional; S, small incision surgery; U, unilateral. Note: Age is reported as mean (SD) when possible. Risk-of-Bias Assessment As Fig. 2 shows, most of the RCTs (27 studies) were at low risk of bias regarding selective reporting. Also, 20 included RCTs had low risks of the randomization process. In the majority of the studies (31/33), there were some concerns about the deviations from the intended interventions, and only 2 studies had a high risk of deviations. Eight RCTs were at high risk regarding the missing outcome data, and only five studies had low risks of measurement of the outcome. Overall, 11 RCTs were judged to have a high risk of bias. Risk of bias assessment of RCTs included in this study. RCT, randomized controlled trial. Results of Meta-analysis Time to Return to Work Four studies provided data on the time to return to work based on the type of operation. 1 22 23 24 Meta-analysis showed a significantly shorter time to return to work in the ECTR group compared with the OCTR group (MD: −8.08; 95% CI: −12.71 to −3.46; random-effects model, I 2 = 92.41%; τ 2 = 18.70, p < 0.01; Fig. 3 ). Substantial statistical heterogeneity was observed (I 2 = 92.41%). A leave-one-out sensitivity analysis was performed for the return-to-work outcome. The pooled mean difference remained statistically significant after sequential omission of each study, yielding estimates of −8.76 (95% CI: −15.88 to −1.65) after excluding Aslani et al, −9.44 (95% CI: −15.09 to −3.78) after excluding Jacobsen and Rahme, −7.73 (95% CI: −15.22 to −0.24) after excluding Saw et al, and −5.62 (95% CI: −7.46 to −3.78) after excluding Chen et al, indicating robustness of the primary findings ( Supplementary Fig. S1 [available in the online version]). Comparison of time to return to work between ECTR and OCTR groups. CI, confidence interval; ECTR, endoscopic carpal tunnel release; OCTR, open carpal tunnel release; SD, standard deviation. Scar-Related Complications The scar-related complications were reported in nine studies. 22 23 25 26 27 28 29 30 31 Pooled analysis showed that the rate of scar-related complications was significantly higher in the OCTR group compared with the ECTR group (OR: 0.18; 95% CI: 0.08 to 0.37; random-effects model, I 2 = 0; τ 2 = 0, p = 0.91; Fig. 4 ). Comparison of scar-related complications between ECTR and OCTR groups. CI, confidence interval; ECTR, endoscopic carpal tunnel release; OCTR, open carpal tunnel release; OR, odds ratio. Revision Surgery No significant difference in revision surgery rates was observed between the OCTR and ECTR groups (OR: 0.78; 95% CI: 0.34 to 1.78; random-effects model, I 2 = 0; τ 2 = 0, p = 0.80; Fig. 5 ). 13 22 23 29 30 32 33 34 35 Comparison of the rate of revision surgery between ECTR and OCTR groups. CI, confidence interval; ECTR, endoscopic carpal tunnel release; OCTR, open carpal tunnel release; OR, odds ratio. Permanent Nerve Injury No significant difference in permanent nerve injury was observed when comparing the ECTR group to the OCTR group (OR: 1.62; 95% CI: 0.08 to 31.81; random-effects model, I 2 = 4.69; τ 2 = 0.12, p = 0.24; Fig. 6 ). 25 27 36 37 Comparison of the permanent nerve injury between ECTR and OCTR groups. CI, confidence interval; ECTR, endoscopic carpal tunnel release; OCTR, open carpal tunnel release; OR, odds ratio. Transient Nerve Injury There was no significant difference in transient nerve injury between the ECTR and OCTR techniques (OR: 2.29; 95% CI: 0.73 to 7.20; random-effects model, I 2 = 6.83; τ 2 = 0.15, p = 0.57; Fig. 7 ). 22 23 25 26 29 30 38 Comparison of the transient nerve injury between ECTR and OCTR groups. CI, confidence interval; ECTR, endoscopic carpal tunnel release; OCTR, open carpal tunnel release; OR, odds ratio. Operation Time Six RCTs assessed operation time. 16 23 24 27 31 39 Pooled data extracted from these studies showed no statistically significant differences between the ECTR and OCTR groups (MD: −3.22; 95% CI: −9.53 to 3.08; random-effects model, I 2 = 98.98%; τ 2 = 60.14, p < 0.01; Fig. 8 ). Comparison of operation time between ECTR and OCTR groups. CI, confidence interval; ECTR, endoscopic carpal tunnel release; OCTR, open carpal tunnel release; SD, standard deviation. Complications Among the 15 studies reporting complications, no significant difference in overall complication rates was observed between the two groups (OR: 0.88; 95% CI: 0.49 to 1.61; random-effects model, I 2 = 0; τ 2 =0, p = 0.32; Fig. 9 ). 1 13 22 23 25 26 27 28 29 32 36 37 38 40 41 Comparison of all complications between ECTR and OCTR groups. CI, confidence interval; ECTR, endoscopic carpal tunnel release; OCTR, open carpal tunnel release; OR, odds ratio. Discussion Carpal tunnel release is a prevalent operation that is performed in 5% of the general population annually. 42 This surgery was started via an open approach with a relatively large mid-palmar incision, then various surgical methods were recommended to help improve the outcome. 43 The literature contains numerous studies that discussed and compared all the existing approaches, accompanied by their short-term and long-term outcomes. 38 44 Although several systematic reviews and meta-analyses have been published so far comparing the ECTR and OCTR techniques, these studies have had small sample sizes, leading to an increasing risk of bias, or did not assess some important aspects of the carpal tunnel release outcome. Some studies provided central tendency values but did not include the standard deviation. Additionally, some studies failed to consider crucial factors such as the pre-surgery values related to outcome measurement for meta-analysis. 17 18 19 45 46 47 48 49 50 Thus, we aimed to perform an updated systematic review and meta-analysis of RCTs with a larger sample size, addressing outcomes that were underreported in previous studies and comparing ECTR and OCTR techniques. Our study found that ECTR was associated with a shorter time to return to work and fewer scar-related complications. However, there were no significant variations in the operation time, revision surgery, permanent nerve injury, transient nerve injury, and overall complications. Our study found no significant difference between the operation times of OCTR and ECTR, which is in line with the findings of Zuo et al 46 and Li et al. 47 Sayegh and Strauch showed significantly shorter operation times in the ECTR group when compared with the OCTR group. 18 In contrast, Hu et al reported that ECTR necessitated a longer operation duration than OCTR. 17 Given that conflict, it should be noted that our study included more articles than Hu et al and Sayegh and Strauch. One study has been added to our analysis that was not taken into account by Hu et al. 17 Additionally, our study incorporated two more recent studies. We included one study that Sayegh and Strauch 18 overlooked. In contrast, Sayegh and Strauch included four studies that we did not. One of these studies lacked accessible full text despite our attempts to contact the corresponding author via email, and two studies failed to report the standard deviation for operation times in both ECTR and OCTR. Upon reviewing one study, we noted that they did not report the surgery times (Ref). Sayegh and Strauch mentioned using data imputation to address the unreported standard deviations, which requires making assumptions that may not always be accurate. This approach can introduce uncertainty. As Sayegh and Strauch's study highlights in its limitation part, there is a trade-off between potential inaccuracies in these assumptions and the benefits of having a larger dataset for analysis. 18 Moreover, in RTCs included in existing systematic reviews, there were no reports of surgeons' experiences, such as the surgery duration or the number of operations in endoscopic surgery, a crucial factor affecting the operation time. Additionally, small differences in operation time between ECTR and OCTR may not be clinically meaningful for patients. The substantial heterogeneity observed for operation time may reflect variations in surgical expertise, learning curve effects, anesthesia protocols, and perioperative workflow differences across trials. Contemporary evidence also suggests that modern ambulatory hand surgery approaches, including WALANT techniques, may influence perioperative efficiency and complication profiles, further complicating comparisons across surgical methods. 7 Time to return to work is a crucial outcome to consider when choosing a surgical approach for CTS, as it directly impacts both personal and health care costs. It may significantly influence a patient's choice of surgery, especially if they work for themselves or have financial concerns due to sick leave. 49 Our findings represented a significantly lower time to return to work in ECTR. This result is in line with the findings of Li et al, 47 Sayegh and Strauch, 18 Vasiliadis et al, 45 and Kohanzadeh et al. 48 Similarly, Paryavi and his colleagues showed that ECTR may allow for a slightly quicker return to work compared with OCTR. However, this advantage in recovery time was typically short-term. The study found no significant difference in long-term outcomes, including full recovery and the resolution of symptoms, between endoscopic and open surgical methods. 51 Also, a Cochrane study suggested that endoscopic carpal surgery might enable a shorter time to return to work. 52 Sanati et al delineated that minimally invasive surgery techniques, which consisted of mini-open surgery and ECTR, allow for a shorter time to return to work than OCTR. Nevertheless, Sanati et al found significant differences in how various RCTs measured the outcome “return to work.” 49 Zuo et al 46 showed that there were no statistically significant differences observed between ECTR and OCTR groups in favor of time to return to work. Gerritsen et al found mixed results on whether ECTR leads to a faster resumption of their professional responsibilities and regular everyday tasks. 53 Thoma et al found the data inconclusive regarding this outcome. 50 This inconsistency may be influenced by varying definitions and multiple factors affecting the time to return to work, including job accessibility, medicolegal affairs, the rate of rehabilitation progress, patient incentives, patient's social, economic, and psychological conditions, job-related factors, including different job roles, exposure to hand-intensive work, workers' compensation, and preoperative work absences, persistence and severity of the symptoms, and the type of surgical approach. Notably, work demands are considered a crucial predictor of the ability to return to work. 52 54 55 56 Nevertheless, when focusing exclusively on RCTs, the impact of this inconsistency is relatively minor. Moreover, the variations in health insurance policies, labor regulations, and other influencing factors across countries would similarly affect both surgery groups. The high heterogeneity is likely attributable to differences in health care systems, occupational demands, rehabilitation protocols, and definitions of return to work across trials rather than a single influential study. Our study showed that the time to return to work was lower in ECTR than in OCTR (MD: −8.08; 95% CI: −12.71 to −3.46). Given the substantial heterogeneity observed for this outcome, a leave-one-out sensitivity analysis was performed. The results demonstrated that exclusion of individual studies did not materially alter the pooled effect estimate, indicating robustness of the observed association. Moreover, it is important to consider that the difference detected may not justify greater cost-effectiveness for the endoscopic method in this respect. 13 A prior study conducting decision analysis indicated that the endoscopic technique would become more expensive if the average time it takes to return to work after surgery differs from the open release by less than 21 days. 57 Our meta-analysis showed a significantly higher rate of scar-related complications post-open surgery, such as scar hypertrophy, scar pain, and hyperesthesia. This finding is consistent with Li et al. 47 However, interpretation should be cautious as definitions of scar-related complications varied across trials, including differences in reporting scar pain, hypersensitivity, hypertrophy, and cosmetic outcomes. The small incision of the TCL during ECTR does not typically damage the underlying subcutaneous nerves of the palm. This likely accounts for the reduced occurrence of painful scars associated with endoscopic techniques. Additionally, the use of the hand and repetitive trauma to the extended scar may contribute to a greater likelihood of hypertrophic or hypersensitive scars following OCTR. 45 Zhang and colleagues showed that received carpal tunnel release through double small incisions has better functional outcomes and scar appearance compared with conventional OCTR. 35 Recent evidence from studies on minimally invasive hand surgery further supports these findings, demonstrating reduced scar-related morbidity without an increase in major complications. 20 Our study illustrated no significant differences in revision surgery between the ECTR and OCTR groups. This is in line with Vasiliadis et al 45 and Hu et al. 17 There is a theory that ECTR may not directly identify the distal edge of the TCL, which could lead to an inadequate release of the carpal tunnel, a risk of recurrence of symptoms, and a need for revision surgery. 58 59 Yet, there is debate about how important it is to fully divide the distal end of the TCL for effective treatment. 60 A study indicated that partial release, meaning dividing the last 4 mm of the TCL, did not significantly affect the widening of the carpal arch, suggesting that incomplete release might not be detrimental. 61 The rationale behind a higher rate of revision surgery in OCTR may relate to differences in surgical technique or incomplete release, warranting further investigation. Moreover, the learning curve associated with ECTR could be a factor in the reduced rates of recurrence and reoperation observed in studies. 45 Our study detected no significant differences in nerve injury between the ECTR and OCTR groups. This important finding contradicts the results of most previous meta-analyses 17 18 19 50 and challenges the belief in possible risks of iatrogenic damage to nerves or blood vessels in the ECTR. 62 63 Most nerve injuries reported in the included studies were reversible. The transient nerve injury typically resolves in a few weeks and is likely to result from neuropraxia from the surgical instrumentation. 64 The nerve itself is not directly visible during the ECTR. Expanding the already compressed canal during surgery could increase pressure on the nerve. Since the initial steps of inserting and expanding the dilators are done without visual guidance, there is a risk of accidentally injuring the nerve or placing the dilators incorrectly. 65 66 ECTR is more challenging to master, requiring substantial training and experience. 67 To perform it safely, surgeons must undergo meticulous training, including courses and cadaver practice, to reduce the risk of complications. ECTR presents specific technical difficulties, such as limited space in the carpal tunnel that can put the median nerve at risk and difficulties in clearly visualizing the TCL, which can increase the likelihood of complications. As surgeons gain more experience and as ECTR techniques are refined, the rates of re-operation have decreased. 19 68 The retractor-integrated endoscope was designed to address the limitations of previous techniques. This device is inserted into a subcutaneous pocket above the nerve, providing a comprehensive view of both the nerve and any structures compressing it. It prevents the temporary nerve injuries that can happen with other ECTR methods. 69 Our study had several limitations. One of the limitations of our study was the lack of accessibility to some articles in languages other than English, even after frequent contact with the corresponding author. Despite their analysis in previous studies, we did not include certain endpoints in our meta-analysis for several reasons. The variable “Grip Strength” could not be adequately analyzed due to significant variations in follow-up periods across the studies, which included assessments conducted immediately after a surgical procedure, at 3 months, 6 months, and 12 months post-operation. This inconsistency hindered subgroup analysis, as we had only one or two articles within each subgroup. Additionally, the heterogeneity among the studies was notably high, likely stemming from these differences in follow-up times. Furthermore, we were unable to conduct a meta-analysis on clinical indices such as pinch strength, the Boston Carpal Tunnel Questionnaire Symptom Severity Scale (BCTQ-S), the Boston Carpal Tunnel Questionnaire Functional Status Scale (BCTQ-F), the Two-point Discrimination test, the Semmes–Weinstein monofilament test, and the Visual Analog Scale (VAS). This limitation was primarily due to insufficient studies—specifically, at least three—reporting baseline and postoperative or follow-up values. We did not conduct subgroup analyses on the different ECTR or OCTR techniques for surgery outcomes and complications, and the follow-up durations were primarily due to insufficient participants in the study and limited data. Specifically, subgroup analyses comparing mini-open versus standard OCTR or single- versus dual-portal ECTR techniques were not feasible due to insufficient reporting and limited sample sizes across trials. Moreover, we excluded patient satisfaction outcomes from the meta-analysis due to the absence of validated instruments for assessment. We also faced challenges in obtaining standard deviations for some variables. Despite reaching out to the corresponding authors of studies that did not report standard deviations, we unfortunately did not receive any responses. As a result, we could not include these studies in our meta-analysis. Some included RCTs in our study had methodological issues, such as unblinded outcome assessments. Moreover, the small sample sizes in some RCTs may have reduced the power to detect subtle differences, and the potential for publication bias should also be considered, as negative results may be underreported. Furthermore, substantial heterogeneity observed in selected outcomes may limit the interpretability of pooled estimates despite the use of random-effects modeling and sensitivity analyses. We recommend that future studies involve larger sample sizes and standardized reporting of outcomes. Conclusion Our study suggests that the endoscopic approach may offer advantages in terms of shorter time to return to work and reduced scar complications; however, these findings should be interpreted with caution due to heterogeneity across trials.