Cohen versus Lich-Gregoir ureteral reimplantation for pediatric primary vesicoureteral reflux: a systematic review and meta-analysis of comparative studies Authors: Bowen Liang, Jiayi Li, Zhiguang An, Ke Xu Published: 2026 ABSTRACT AbstractBackgroundCohen and Lich-Gregoir ureteral reimplantation are the two most widely used open reconstructive options for children with primary vesicoureteral reflux (VUR), yet their relative advantages remain debated. We aimed to compare anti-reflux efficacy, perioperative outcomes, and complication profiles between these procedures using direct comparative evidence only.MethodsPubMed, Embase, and the Cochrane Library were searched from inception to March 2026 for comparative studies evaluating Cohen versus Lich-Gregoir reimplantation in children with primary VUR. Outcomes included persistent postoperative reflux, overall complications, reoperation, operative time, length of stay (LOS), postoperative acute retention of urine (ARU), obstruction/hydronephrosis, hematuria, and febrile urinary tract infection (UTI). A bilateral VUR subgroup analysis was prespecified. Risk of bias was assessed with ROBINS-I, and certainty of evidence was judged using a GRADE-informed approach. Because fewer than 10 studies contributed to each synthesis, formal publication-bias testing was not performed.ResultsNine retrospective comparative studies involving 655 children (370 Cohen, 285 Lich-Gregoir) were included. Baseline age and preoperative reflux grade were comparable between groups. There were no significant differences in persistent postoperative VUR [odds ratio (OR) 0.72, 95% confidence interval (CI): 0.34–1.54], overall complications (OR 0.97, 95% CI: 0.30–3.09), reoperation (OR 0.54, 95% CI: 0.15–1.94), obstruction/hydronephrosis (OR 1.15, 95% CI: 0.34–3.83), or febrile UTI (OR 0.45, 95% CI: 0.18–1.12). Lich-Gregoir was associated with shorter operative time [mean difference (MD) 28.59 minutes, 95% CI: 3.37–53.81] and shorter LOS (MD 3.41 days, 95% CI: 2.15–4.61). ARU occurred more often after Lich-Gregoir (6.4% vs. 1.7%; OR 0.32, 95% CI: 0.12–0.91), whereas hematuria was markedly more common after Cohen. In bilateral VUR, Lich-Gregoir retained advantages in operative time and LOS, but differences in overall complications, persistent reflux, and ARU were not statistically significant. Evidence certainty was low to very low across outcomes and was limited by retrospective study design, serious risk of bias, inconsistent complication definitions, and heterogeneity in surgical platform and perioperative care.ConclusionsCohen and Lich-Gregoir reimplantation provide comparable anti-reflux efficacy and similar major postoperative outcomes in children with primary VUR. Lich‑Gregoir was associated with shorter operative time with very high heterogeneity, but procedure selection should also consider its higher risk of postoperative urinary retention and the still-limited evidence in bilateral disease.Keywords: Vesicoureteral reflux (VUR), ureteral reimplantation, Cohen, Lich-Gregoir, meta-analysis Highlight box Key findings • Cohen and Lich-Gregoir achieved comparable anti-reflux efficacy and similar major postoperative outcomes. Lich-Gregoir was associated with shorter operative time and hospital stay, whereas postoperative acute urinary retention was more frequent after Lich-Gregoir. What is known and what is new? • Both procedures are established options for primary pediatric vesicoureteral reflux, but existing evidence often mixes broad intravesical and extravesical categories. • This study focuses only on direct Cohen versus Lich-Gregoir comparisons and adds reoperation and bilateral subgroup analyses. What is the implication, and what should change now? • Procedure choice should be individualized. Lich-Gregoir offers efficiency advantages, but surgeons should weigh this against the higher likelihood of transient postoperative voiding dysfunction. Bilateral disease remains the setting in which stronger comparative evidence is most needed. Introduction Vesicoureteral reflux (VUR) is one of the most common congenital urological anomalies in children, and is associated with recurrent febrile urinary tract infection (UTI), renal scarring, and, in selected patients, long-term loss of renal function (1-6). Although many children can be managed conservatively, surgical intervention remains indicated for persistent high-grade reflux, breakthrough infection despite medical management, progressive renal damage, or poor adherence to prolonged follow-up (1,2,4-8). Open ureteral reimplantation continues to achieve excellent long-term success and remains a benchmark treatment against which newer techniques are compared (7-10). Among open reconstructive options, the Cohen cross-trigonal intravesical approach and the Lich-Gregoir extravesical approach are the two most widely used techniques (7-15). Both aim to create an adequate submucosal tunnel and restore a competent antireflux mechanism, but they differ in operative route and perioperative morbidity profile. Intravesical surgery requires cystotomy and extensive bladder manipulation, which may predispose to postoperative hematuria, bladder spasm, and longer convalescence. By contrast, the extravesical approach avoids cystotomy but has traditionally raised concern about transient voiding dysfunction or urinary retention, especially in bilateral cases (11-16). Previous evidence syntheses have usually compared broad intravesical versus extravesical categories or different minimally invasive platforms rather than direct Cohen versus Lich-Gregoir comparisons (17-19). Two recent meta-analyses have addressed this comparison. Ku and Huang conducted a network meta-analysis of randomized controlled trials comparing multiple interventions for pediatric VUR, including Cohen and Lich-Gregoir, and ranked Lich-Gregoir as having the highest success rate among surgical techniques (20). More recently, Wang et al. performed a pairwise meta-analysis directly comparing Cohen and Lich-Gregoir in 1,314 patients, finding comparable anti-reflux efficacy but shorter operative time and hospital stay for the extravesical approach (21). However, both analyses were limited by inconsistent complication definitions, and incomplete reporting of laterality-specific outcomes for certain complications such as acute urinary retention. Therefore, we performed an updated systematic review and meta-analysis with a prespecified bilateral subgroup analysis and a detailed assessment of specific complications, including acute urinary retention and hematuria, to further inform surgical decision-making. We present this article in accordance with the PRISMA reporting checklist (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0377/rc). Methods No review protocol was prospectively registered or publicly deposited before study completion. Search strategy PubMed, Embase, and the Cochrane Library were searched from database inception to March 2026. The search strategy combined controlled vocabulary and free-text terms related to “Cohen”, “cross-trigonal”, “Lich-Gregoir”, “extravesical”, “ureteral reimplantation”, “vesicoureteral reflux”, “VUR”, “pediatric”, “child”, and “children”. No study-design filter was applied at the search stage. Reference lists of relevant articles were also screened manually to identify additional eligible studies. Searches were last updated in March 2026. No study-design filter was applied at the search stage. Search results were restricted during eligibility assessment to original full-text comparative studies of children with primary VUR. No language restrictions were applied during the electronic search. During eligibility assessment, only English full‑text articles were considered because of the need for reliable data extraction. The following is full database search strategies: (“Vesico-Ureteral Reflux”[Mesh] OR VUR[tiab] OR “vesicoureteral reflux”[tiab] OR “vesico ureteral reflux”[tiab] OR “urine reflux”[tiab]) AND (“Child”[Mesh] OR “Infant”[Mesh] OR “Adolescent”[Mesh] OR child*[tiab] OR infant*[tiab] OR pediatric*[tiab] OR aediatric*[tiab] OR newborn*[tiab] OR neonat*[tiab] OR teen*[tiab] OR adolescent*[tiab] OR boy[tiab] OR boys[tiab] OR girl[tiab] OR girls[tiab]) AND ((“Lich”[tiab] AND “Gregoir”[tiab]) OR “Lich-Gregoir”[tiab] OR “extravesical”[tiab] OR “extra-vesical”[tiab] OR “detrusorrhaphy”[tiab]OR “extravesical reimplant*”[tiab]) OR (“Cohen”[tiab] AND “ureterocystostomy”[tiab]) OR “Cohen procedure”[tiab] OR “Cohen technique”[tiab] OR “transtrigonal”[tiab])). Eligibility criteria Studies were eligible if they met all of the following criteria: (I) children with primary VUR; (II) a direct comparison between Cohen and Lich-Gregoir ureteral reimplantation; (III) extractable outcome data for at least one prespecified endpoint; and (IV) original full-text comparative study design. Because randomized trials were not available, retrospective comparative cohort studies were eligible. We excluded review articles, case reports, editorials, conference abstracts, duplicate publications, studies involving non-primary reflux or mixed disease populations without extractable data, and non-English full-text articles. Reasons for exclusion at full-text review: Among the 18 full-text articles assessed, 9 were excluded for one or more of the following prespecified reasons: (I) no direct head-to-head comparison between Cohen and Lich-Gregoir reimplantation; (II) non-comparative design, review article, editorial, or conference-only report; (III) mixed or non-primary reflux populations without extractable data for primary pediatric VUR; (IV) duplicate publication or overlapping cohort; (V) unavailable English full text; or (VI) insufficient outcome data for the present synthesis. Study selection and data extraction Two reviewers independently screened titles and abstracts, assessed full texts for eligibility, and extracted data using a standardized form. Disagreements were resolved by discussion. Extracted variables included study design, year of publication, surgical platform, laterality, sample size, age at operation, preoperative VUR grade, operative time, length of stay (LOS), follow-up duration, postoperative imaging protocol, and outcome data. For dichotomous outcomes, event counts and denominators were recorded. Persistent postoperative reflux was extracted per ureter whenever reported; other clinical outcomes were analyzed at the patient level. Outcomes The primary outcomes were persistent postoperative VUR, overall postoperative complications, and reoperation. Secondary outcomes were operative time, postoperative LOS, acute retention of urine (ARU), postoperative obstruction/hydronephrosis, hematuria, febrile UTI, and bladder spasm. Persistent reflux was analyzed at the ureter level because it is a lateralized radiographic finding; all other outcomes (complications, ARU, reoperation, etc.) were analyzed at the patient level. This unit difference is acknowledged and should be considered when interpreting this outcome; the qualitative conclusion remained unchanged. A bilateral VUR subgroup analysis was prespecified. When continuous outcomes were reported without analyzable dispersion data and these could not be derived reliably, the study was not pooled for that specific analysis. Risk of bias assessment Methodological quality was assessed with the ROBINS-I tool 31, addressing bias due to confounding, participant selection, intervention classification, deviations from intended intervention, missing data, outcome measurement, and selection of reported results (22). Reporting bias and certainty assessment Because fewer than 10 studies contributed to any pooled analysis, funnel plots and formal statistical tests for small-study or publication bias were considered uninformative and were not performed. Instead, the possibility of reporting bias was considered qualitatively during interpretation. Certainty of evidence for the main pooled outcomes was judged using a GRADE-informed approach 30, considering study design, risk of bias, inconsistency, indirectness, imprecision, and the possibility of reporting bias (23). Because all included studies were retrospective comparative cohorts, confidence in the estimates was expected to be low or very low unless there were reasons to judge otherwise. Statistical analysis Meta-analysis was performed using Review Manager 5.4 (Cochrane Collaboration, Oxford, UK). Mean differences (MDs) with 95% confidence intervals (CIs) were used for continuous outcomes, and odds ratios (ORs) with 95% CIs for dichotomous outcomes. ORs were calculated with the Cohen group as the reference group. Statistical heterogeneity was assessed using the I2 statistic. A fixed-effect model was used when heterogeneity was low (I2≤50%); otherwise, a random-effects model was applied. Studies were synthesized separately for the overall cohort and, where extractable, for the bilateral VUR subgroup. A sensitivity analysis for overall complications was prespecified by restricting pooling to studies using Clavien-Dindo-based or otherwise relatively standardized complication definitions when at least 3 studies were available. Statistical significance was defined as a two-sided P value <0.05. Results Study selection A total of 413 records were identified, including 411 through database searching and 2 through manual searching. After duplicate removal, 410 records remained. Following title and abstract screening, 18 full-text articles were assessed for eligibility, of which 9 were excluded for prespecified reasons. Excluded full-text reports lacked a direct head-to-head Cohen-versus-Lich comparison, were non-comparative or review reports, involved mixed or non-primary reflux populations without extractable data, duplicated an overlapping cohort, lacked an English full text, or provided insufficient outcome data for this review. Ultimately, 9 retrospective comparative studies were included in the qualitative synthesis and quantitative meta-analysis (Figure 1). PRISMA flow diagram of study identification, screening, eligibility assessment, and inclusion for the systematic review and meta-analysis. Study characteristics and risk of bias The 9 included studies were published between 2004 and 2025 and comprised 655 children with primary VUR: 370 underwent Cohen reimplantation and 285 underwent Lich-Gregoir reimplantation. Study size ranged from 28 to 129 patients. Most studies focused on unilateral disease, whereas several included mixed unilateral and bilateral cohorts and one study was restricted to bilateral VUR with propensity score matching. The Cohen arm was uniformly performed through an open intravesical approach. The Lich-Gregoir arm included open, laparoscopic, robotic-assisted, and modified open extravesical approaches. Among the nine included studies, the Lich-Gregoir arm consisted of open extravesical reimplantation in four studies, modified open extravesical reimplantation in one study, laparoscopic extravesical reimplantation in three studies, and robot-assisted extravesical reimplantation in one study. In addition, one study used a transvesicoscopic Cohen-type comparator rather than conventional open Cohen reimplantation. Reported follow-up durations ranged from approximately 13 to 64 months, and postoperative imaging protocols varied across studies (Table 1). Table 1Characteristics of included comparative studiesStudyDesignLateralityPatients (Cohen/Lich)Surgical approachFollow-up (months)Chen (24)RetrospectiveMixed92/37Open Cohen vs. open Lich-GregoirMedian 57.6 overallSmith (25)RetrospectiveMixed25/25Open Cohen vs. robotic-assisted extravesical29.1/16.0Esposito (26)RetrospectiveMixed30/30Open Cohen vs. laparoscopic Lich-Gregoir37.2/34.8Sriram (27)RetrospectiveMixed67/51Open Cohen vs. modified open Lich-Gregoir29.9/28.4Silay (28)RetrospectiveUnilateral23/35Open Cohen vs. open Lich-Gregoir13.1/21.0Aydin (29)RetrospectiveUnilateral18/20Open Cohen vs. open Lich-Gregoir29.9/28.4Tessier (30)RetrospectiveNot extractable77/35Open Cohen vs. laparoscopic Lich-GregoirMedian 64 overallShukla (31)RetrospectiveUnilateral24/38Transvesicoscopic Cohen-type reimplantation vs. laparoscopic Lich-Gregoir24/24Fujii (32)Propensity-score matched retrospectiveBilateral only14/14Open Cohen vs. open Lich-GregoirMedian 16.5/21.5 Characteristics of included comparative studies Risk of bias was assessed with ROBINS-I (Figure 2). Eight studies were judged to have an overall serious risk of bias and one study a moderate risk of bias. The main concerns were residual confounding and participant selection, reflecting the retrospective non-randomized design of the available evidence. Risk of bias assessment using the ROBINS-I tool. (A) Study-level judgments across individual domains. (B) The proportion of studies judged as low, moderate, serious, or critical risk of bias for each domain and overall. Reporting bias and certainty of evidence No quantitative synthesis included 10 or more studies, so formal funnel-plot assessment and statistical testing for small-study effects were not performed. The possibility of selective publication or selective non-reporting of outcomes therefore cannot be excluded. Using a GRADE-informed framework, certainty of evidence was judged low for persistent postoperative VUR and reoperation and very low for operative time, LOS, overall complications, ARU, obstruction/hydronephrosis, hematuria, febrile UTI, and bilateral-subgroup estimates. The main reasons for rating certainty down were retrospective non-randomized design, serious risk of bias, heterogeneity, sparse events, and imprecision. Certainty judgments were made for transparency and interpretation, not as a substitute for formal guideline development. Because all included evidence came from retrospective comparative cohorts, confidence in the estimates was generally low or very low. Baseline comparability Across the overall cohort, there was no significant between-group difference in age at operation (MD −0.40 years, 95% CI: −1.43 to 0.63; P=0.45; Figure 3), although heterogeneity was high (I2=94%). Preoperative reflux grade was also comparable between groups (MD −0.07, 95% CI: −0.40 to 0.26; P=0.68; Figure 4), again with substantial heterogeneity (I2=83%) (Table 2). Forest plot comparing age at operation between children undergoing Cohen and Lich-Gregoir ureteral reimplantation. MD values were calculated as Cohen minus Lich-Gregoir. CI, confidence interval; IV, inverse variance; MD, mean difference; SD, standard deviation. Forest plot comparing preoperative vesicoureteral reflux grade between the Cohen and Lich-Gregoir groups. MD values were calculated as Cohen minus Lich-Gregoir. CI, confidence interval; IV, inverse variance; MD, mean difference; SD, standard deviation. Table 2Summary of pooled outcomes in the overall cohortOutcomeStudiesUnitEffect estimate (95% CI)I2InterpretationAge at operationOverall cohortPatientMD −0.40 years (−1.43 to 0.63)94%No baseline differencePreoperative reflux gradeOverall cohortPatientMD −0.07 (−0.40 to 0.26)83%No baseline differenceOperative time8PatientMD 28.59 min (3.37 to 53.81)98%Shorter with Lich-GregoirLength of stay5PatientMD 3.41 days (2.15 to 4.61)97%Shorter with Lich-GregoirOverall complications8PatientOR 0.97 (0.30 to 3.09)58%No significant differenceReoperation5PatientOR 0.54 (0.15 to 1.94)7%No significant differencePersistent postoperative VUR7UreterOR 0.72 (0.34 to 1.54)0%No significant differenceAcute retention of urine5PatientOR 0.32 (0.12 to 0.91)0%Higher after Lich-GregoirObstruction/hydronephrosis5PatientOR 1.15 (0.34 to 3.83)0%No significant differenceHematuria2PatientOR 889.82 (75.19 to 10,530.60)0%Higher after CohenFebrile UTI4PatientOR 0.45 (0.18 to 1.12)13%No significant differenceBladder spasmDescriptive onlyPatientNot pooled–Sparse evidence; numerically higher after CohenORs were calculated with Cohen as the reference group. Extremely high OR due to sparse events and continuity correction; Peto OR 18.92 (95% CI: 5.23–68.43). CI, confidence interval; MD, mean difference; OR, odds ratio; UTI, urinary tract infection; VUR, vesicoureteral reflux. Summary of pooled outcomes in the overall cohort ORs were calculated with Cohen as the reference group. Extremely high OR due to sparse events and continuity correction; Peto OR 18.92 (95% CI: 5.23–68.43). CI, confidence interval; MD, mean difference; OR, odds ratio; UTI, urinary tract infection; VUR, vesicoureteral reflux. Primary efficacy and major safety outcomes Persistent postoperative VUR was reported by 7 studies and analyzed per ureter. No significant difference was observed between Cohen and Lich-Gregoir (OR 0.72, 95% CI: 0.34 to 1.54; P=0.40; I2=0%; Figure 5). The pooled rates of persistent reflux were 4.5% (13/289 ureters) for Cohen and 6.7% (15/225 ureters) for Lich-Gregoir. Forest plot of persistent postoperative vesicoureteral reflux. This outcome was analyzed per ureter whenever reported. ORs were calculated with Cohen as the reference group. CI, confidence interval; M-H, Mantel-Haenszel; OR, odds ratio. Overall postoperative complications were analyzable in 8 studies. The pooled complication rates were 7.6% for Cohen and 8.1% for Lich-Gregoir, with no significant difference between procedures (OR 0.97, 95% CI: 0.30 to 3.09; P=0.96; I2=58%; Figure 6). A sensitivity analysis restricted to the 3 studies reporting complications using relatively standardized Clavien-Dindo-based criteria also showed no significant difference (OR 1.21, P=0.86). Because complication definitions varied (see Table 3), the pooled OR should be considered exploratory. Forest plot of overall postoperative complications. ORs were calculated with Cohen as the reference group; OR <1 indicates fewer events after Cohen, whereas OR >1 indicates fewer events after Lich-Gregoir. CI, confidence interval; M-H, Mantel-Haenszel; OR, odds ratio. Table 3Summary of complication definitions by studyStudy (first author, year)Overall complications definitionAcute retention of urine definition (catheterization required?)Obstruction/hydronephrosis definitionHematuria definitionChen 2004Clavien-Dindo not used; grade I: 10Defined as need for temporary urethral catheterization (“temporary urethral catheterization was performed in the case of urinary retention during hospitalization”)Not clearly defined; only mentioned “mild hydronephrosis” diagnosed by ultrasoundNot clearly defined; only mentioned clot retentionSmith 2011Clavien-Dindo not used; grade I: 4Defined as difficulty voiding postoperatively requiring catheterization (3 cases: one single catheterization, one re‑inserted for 5 days, one required catheterization)Not clearly defined; mentioned “transient mildhydroureteronephrosis” on ultrasound, no specific thresholdNot reportedEsposito 2016Clavien-Dindo used; grade II: 10; grade IIIb: 1Not clearly defined; one Cohen case had “bladder emptying difficulty, requiring temporary urethral catheterization”Not clearly defined; one Cohen case had “acute ureteral obstruction” requiring reoperationNot clearly defined; 28/32 Cohen cases (93.3%) developed hematuria, described as marked postoperative hematuriaSriram 2016Clavien-Dindo not used; grade II: >37Not reportedNot clearly defined; no evidence of obstruction at 1‑year follow‑up, but no diagnostic criteria givenDefined as “significant hematuria (>72 h/required blood transfusions/clot retention)”Silay 2018Clavien-Dindo not used; grade IIIa: 1Not reportedNot reportedNot reportedAydin 2020Clavien-Dindo not used; grade I: 18; grade II: 3Not reportedNot clearly defined; mentioned “severe hydronephrosis” as a complication to avoid, no specific criteriaDefined as gross hematuria withdocumented duration (mean 2.5±1.3 days)Tessier 2023Clavien-Dindo not used; grade I: 1; grade IIIb: 3Defined as postoperative need for bladder catheter placement (“acute urine retention required bladder catheter placement with spontaneous resolution”)Not clearly defined; reported “ureteral stenoses”, but no imaging criteria givenNot reportedShukla 2025Clavien-Dindo used; grade I–II: 10; grade IIIa: 2Not reportedDefined as progressive hydroureteronephrosis with increased retrovesical ureter diameter on follow‑upNot reportedFujii 2025Clavien-Dindo not used; grade I: 3; grade II: 4; grade IIIa: 1Defined as need for re‑insertion of urethral Foley catheter due to temporary urinary retentionNot clearly defined; mentioned “postoperativehydronephrosis” but no specific threshold; no significant difference between groupsNot clearly defined; two cases of clot retention in Cohen group Summary of complication definitions by study Reoperation was reported in 5 studies. The pooled reoperation rates were 1.2% after Cohen and 3.3% after Lich-Gregoir, without a significant between-group difference (OR 0.54, 95% CI: 0.15 to 1.94; P=0.34; I2=7%; Figure 7). Forest plot of reoperation after ureteral reimplantation. ORs were calculated with Cohen as the reference group. CI, confidence interval; M-H, Mantel-Haenszel; OR, odds ratio. Sensitivity analysis Sensitivity analysis restricting the overall-complication meta-analysis to the 3 studies using Clavien-Dindo-based or otherwise relatively standardized complication definitions did not materially change the result and still showed no significant difference between procedures (OR 1.21, P=0.86). Perioperative outcomes Operative time was analyzable in 8 studies. Lich-Gregoir was associated with a significantly shorter operative time than Cohen (MD 28.59 minutes, 95% CI: 3.37 to 53.81; P=0.03; I2=98%; Figure 8). Forest plot of operative time in the overall cohort. Positive MD values indicate longer operative time in the Cohen group and therefore favor Lich-Gregoir for procedural efficiency. CI, confidence interval; IV, inverse variance; MD, mean difference; SD, standard deviation. LOS was reported in 5 studies. Children undergoing Lich-Gregoir had a significantly shorter postoperative hospital stay than those undergoing Cohen (MD 3.41 days, 95% CI: 2.15 to 4.61; P<0.001; I2=97%; Figure 9). Forest plot of postoperative length of stay in the overall cohort. Positive MD values indicate longer hospitalization in the Cohen group and therefore favor Lich-Gregoir. CI, confidence interval; IV, inverse variance; MD, mean difference; SD, standard deviation. I2=97–98%, indicating extreme heterogeneity across studies. Specific complications Five studies reported postoperative ARU. In the five studies, ARU was defined as need for temporary catheterization or inability to void spontaneously within 6–8 hours postoperatively; all events resolved with brief catheterization. The incidence of ARU was significantly higher after Lich-Gregoir (6.4%) than after Cohen (1.7%). Correspondingly, the pooled OR calculated with Cohen as the reference group was 0.32 (95% CI: 0.12 to 0.91; P=0.03; I2=0%; Figure 10). Forest plot of postoperative ARU. ORs were calculated with Cohen as the reference group; values below 1 indicate a higher incidence of ARU after Lich-Gregoir. ARU, acute retention of urine; CI, confidence interval; M-H, Mantel-Haenszel; OR, odds ratio. Postoperative obstruction or hydronephrosis was reported in 5 studies. No significant difference was observed between the two procedures (OR 1.15, 95% CI: 0.34 to 3.83; P=0.83; I2=0%; Figure 11). Forest plot of postoperative obstruction or hydronephrosis. ORs were calculated with Cohen as the reference group. CI, confidence interval; M-H, Mantel-Haenszel; OR, odds ratio. Hematuria was reported in two studies. Raw events: Cohen 26/27 (96.3%), Lich-Gregoir 0/35 (0%). Because of zero events in one arm, the common‑OR using a continuity correction (adding 0.5 to each cell) yielded an OR of 889.82 (95% CI: 75.19 to 10,530.60; P<0.001; I2=0%; Figure 12). A sensitivity analysis using the Peto method (more appropriate for sparse events) gave a Peto OR of 18.92 (95% CI: 5.23–68.43), which still demonstrates a markedly higher incidence of hematuria after Cohen. The very large OR is a result of sparse events (zero events in the Lich-Gregoir arm) and the continuity correction; a Peto OR sensitivity analysis gave 18.92 (95% CI: 5.23–68.43), which does not alter the direction or clinical interpretation. Forest plot of postoperative hematuria. ORs were calculated with Cohen as the reference group; values above 1 indicate a higher incidence of hematuria after Cohen. CI, confidence interval; M-H, Mantel-Haenszel; OR, odds ratio. Postoperative febrile UTI was reported in 4 studies. No significant difference was observed between the two procedures (OR 0.45, 95% CI: 0.18 to 1.12; P=0.09; I2=13%; Figure 13). Forest plot of postoperative febrile urinary tract infection. ORs were calculated with Cohen as the reference group. CI, confidence interval; M-H, Mantel-Haenszel; OR, odds ratio. Bladder spasm was reported in only 2 studies, both involving bilateral disease. One study used a modified Lich-Gregoir technique and was excluded from descriptive comparison for this endpoint, leaving only a single study in which bladder spasm was reported in 85.7% of Cohen cases and in none of the Lich-Gregoir cases. Because of the limited evidence base, this outcome was not meta-analyzed. Bilateral VUR subgroup Laterality-specific baseline data were incompletely reported, and only a subset of studies provided extractable bilateral data (Table 4). In bilateral VUR, Lich-Gregoir remained associated with shorter operative time (MD 42.58 minutes, 95% CI: 6.76 to 78.40; P=0.02; I2=95%; Figure 14) and shorter LOS (MD 2.38 days, 95% CI: 1.88 to 2.88; P<0.001; I2=53%; Figure 15). Table 4Summary of pooled outcomes in the bilateral VUR subgroupOutcomePooled studiesUnitEffect estimate (95% CI)I2InterpretationOperative time4 pooledPatientMD 42.58 min (6.76 to 78.40)95%Shorter with Lich-GregoirLength of stay3 pooledPatientMD 2.38 days (1.88 to 2.88)53%Shorter with Lich-GregoirOverall complications2PatientOR 0.42 (0.02 to 10.26)70%No significant differencePersistent postoperative VUR2UreterOR 0.57 (0.17 to 1.94)0%No significant differenceAcute retention of urine3PatientOR 0.28 (0.06 to 1.24)42%No significant difference; numerical excess after Lich-GregoirBilateral subgroup analyses were limited by incomplete laterality-specific reporting in the source studies. CI, confidence interval; MD, mean difference; OR, odds ratio; VUR, vesicoureteral reflux. Summary of pooled outcomes in the bilateral VUR subgroup Bilateral subgroup analyses were limited by incomplete laterality-specific reporting in the source studies. CI, confidence interval; MD, mean difference; OR, odds ratio; VUR, vesicoureteral reflux. Forest plot of operative time in the bilateral VUR subgroup. Positive MD values indicate longer operative time in the Cohen group and therefore favor Lich-Gregoir. CI, confidence interval; IV, inverse variance; MD, mean difference; SD, standard deviation; VUR, vesicoureteral reflux. Forest plot of postoperative length of stay in the bilateral VUR subgroup. Positive MD values indicate longer hospitalization in the Cohen group and therefore favor Lich-Gregoir. CI, confidence interval; IV, inverse variance; MD, mean difference; SD, standard deviation; VUR, vesicoureteral reflux. However, no significant subgroup difference was found for overall complications (OR 0.42, 95% CI: 0.02 to 10.26; P=0.59; I2=70%; Figure 16), persistent postoperative reflux (OR 0.57, 95% CI: 0.17 to 1.94; P=0.37; I2=0%; Figure 17), or ARU (OR 0.28, 95% CI: 0.06 to 1.24; P=0.09; I2=42%; Figure 18), despite a numerical excess of ARU events in the Lich-Gregoir group. Forest plot of overall postoperative complications in the bilateral VUR subgroup. ORs were calculated with Cohen as the reference group. CI, confidence interval; M-H, Mantel-Haenszel; OR, odds ratio; VUR, vesicoureteral reflux. Forest plot of persistent postoperative vesicoureteral reflux in the bilateral VUR subgroup. This outcome was analyzed per ureter whenever reported. CI, confidence interval; M-H, Mantel-Haenszel; OR, odds ratio; VUR, vesicoureteral reflux. Forest plot of postoperative ARU in the bilateral VUR subgroup. ORs were calculated with Cohen as the reference group. ARU, acute retention of urine; CI, confidence interval; M-H, Mantel-Haenszel; OR, odds ratio; VUR, vesicoureteral reflux. Discussion This meta-analysis of direct comparative studies suggests that Cohen and Lich-Gregoir ureteral reimplantation achieve similar anti-reflux efficacy and comparable major postoperative outcomes in children with primary VUR. Across 655 patients, we found no significant differences in persistent postoperative reflux, overall complications, reoperation, postoperative obstruction/hydronephrosis, or febrile UTI. The most consistent perioperative advantage of Lich-Gregoir was shorter operative time and shorter hospitalization. However, the two procedures were not interchangeable in terms of morbidity profile: Lich-Gregoir was associated with more postoperative ARU, whereas Cohen was associated with more hematuria and, in sparse descriptive data, more bladder spasm. The absence of a significant difference in persistent reflux or reoperation is clinically important. Both procedures rely on the same core reconstructive principle of creating an adequate submucosal tunnel and restoring a flap-valve mechanism, and both have long been regarded as highly reliable open operations (7-10). The unit of analysis for persistent reflux (ureter) differed from that for other outcomes (patient), which may introduce a mild bias but is unlikely to alter the main conclusion of no significant difference. Our findings therefore support the view that, for most children selected for ureteral reimplantation, the key clinical question is not whether either procedure works, but which trade-off between perioperative efficiency and specific complications is more acceptable in a given patient. The shorter operative time and shorter LOS associated with Lich-Gregoir are biologically plausible. The extravesical approach avoids cystotomy and intravesical dissection, reducing the extent of bladder manipulation and, in many settings, simplifying postoperative recovery (12-15). At the same time, these pooled results should be interpreted with awareness of substantial heterogeneity. The Lich-Gregoir arm across included studies was not uniform and included open, laparoscopic, robotic-assisted, and modified approaches. Operative time is therefore influenced not only by the reconstructive principle itself but also by surgeon experience, case complexity, and surgical platform. Similarly, LOS is sensitive to center-specific discharge protocols, catheter management, analgesia pathways, and regional practice patterns. The very high heterogeneity (I2=97–98%) limits confidence in the pooled estimates. Differences in discharge criteria, catheter management, and surgical platforms almost certainly contributed to the wide CIs and extreme inconsistency. The overall complication analysis also deserves careful interpretation. Although the total postoperative complication burden did not differ significantly between procedures, the specific types of complications clearly did. The higher rate of ARU after Lich-Gregoir is consistent with longstanding concern that extravesical dissection near the ureterovesical junction and trigonal region may transiently affect the pelvic plexus or local detrusor innervation, particularly when surgery is bilateral (13-15). In contrast, the higher rates of hematuria and likely bladder spasm after Cohen are concordant with the transvesical nature of the procedure, which entails cystotomy and more extensive bladder mucosal handling (15,16). From a practical standpoint, our data suggest that surgeons and families should think less in terms of a single “better” procedure and more in terms of different complication spectra around procedures with otherwise similar major efficacy. Bilateral VUR remains the most clinically relevant area of uncertainty. Historically, bilateral extravesical reimplantation raised concern because of postoperative voiding dysfunction and urinary retention (11-14). In our subgroup analysis, Lich-Gregoir still showed shorter operative time and shorter LOS in bilateral cases, but differences in overall complications, persistent reflux, and ARU did not reach statistical significance. We do not interpret this as proof that bilateral risk is absent. Rather, the bilateral subgroup was limited by incomplete laterality-specific reporting, small sample size, and inconsistent baseline stratification. Thus, bilateral disease is precisely the setting in which stronger comparative evidence is still needed. This study has several strengths. First, it focused strictly on head-to-head comparative studies of Cohen versus Lich-Gregoir, thereby addressing a narrower and more clinically relevant question than broader comparisons of intravesical versus extravesical or open versus minimally invasive procedures (17-19). Second, in addition to operative time and reflux persistence, it examined reoperation and specific complications that matter in perioperative counseling and postoperative care. Third, it incorporated a bilateral subgroup analysis rather than assuming that unilateral and bilateral disease can be interpreted identically. Our findings are consistent with the network meta-analysis by Ku and Huang, who also reported high success rates for Lich-Gregoir (20). Compared with the pairwise meta-analysis by Wang et al., our results similarly show no significant difference in persistent postoperative reflux (OR 0.72 vs. 1.04 in Wang) and comparable overall complication rates (21). The operative time advantage for Lich-Gregoir (MD 28.6 minutes in our study vs. 22.37 minutes in Wang) and shorter hospital stay (MD 3.4 vs. 2.65 days in Wang) are also broadly concordant. However, our meta-analysis provides additional granularity on specific complications: we observed a significantly higher risk of acute urinary retention after Lich-Gregoir (OR 0.32, P=0.03), whereas Wang et al. did not report this outcome separately. Conversely, hematuria was markedly more common after Cohen, a finding that Wang et al. also noted. In bilateral VUR, both our study and Wang et al. found that Lich-Gregoir retained advantages in operative time and hospital stay, but neither analysis demonstrated a significant difference in persistent reflux or overall complications in this subgroup, underscoring the need for more robust evidence in bilateral disease. It should be noted that our study shares similar limitations, such as reliance on retrospective data, heterogeneity in surgical platforms, and very low certainty regarding perioperative outcomes. Therefore, our conclusions warrant cautious interpretation. Several limitations should be acknowledged. All included studies were retrospective, and ROBINS-I assessment showed serious risk of bias in most studies, largely because of confounding and patient selection. Grey literature and trial registries were not searched, raising the possibility of unpublished null or negative results. Definitions and reporting thresholds for complications were inconsistent; some studies captured only Clavien-Dindo grade ≥2 events, whereas others reported any adverse event requiring intervention or prolonged observation. Surgical heterogeneity within the Lich-Gregoir arm, including open, laparoscopic, robotic, and modified techniques, likely contributed to the high heterogeneity observed for operative time and LOS. Some outcomes were analyzed per patient, whereas persistent reflux was most often reported per ureter. Lack of uniform definitions for overall complications and for obstruction/hydronephrosis undermines comparability; the pooled overall complication estimate is therefore subject to high uncertainty. Bilateral subgroup analyses were underpowered because laterality-specific data were incompletely reported. Formal assessment of publication bias was not informative because each synthesis included fewer than 10 studies, and the certainty of evidence was low to very low across outcomes. Finally, not all studies used a uniform postoperative imaging schedule or the same imaging modality to document reflux resolution, which may have affected ascertainment of radiographic success. Overall, the current evidence supports a nuanced, individualized approach to procedure selection. When minimizing operative time and hospital stay is a priority, Lich-Gregoir may be attractive; when transient postoperative voiding dysfunction is of particular concern, the higher ARU signal after Lich-Gregoir should be weighed carefully. Future multicenter comparative studies with standardized complication definitions, clearly reported laterality-specific outcomes, and contemporary surgical stratification are needed to refine procedure choice, especially for bilateral VUR. Conclusions Cohen and Lich-Gregoir ureteral reimplantation provide comparable anti-reflux efficacy and similar major postoperative outcomes for children with primary VUR. Lich-Gregoir offers shorter operative time and hospitalization, but this efficiency advantage is accompanied by a higher risk of postoperative acute urinary retention. Rather than supporting procedural superiority, the available evidence favors individualized selection based on the expected trade-off between perioperative efficiency and complication profile. The perioperative efficiency advantage of Lich‑Gregoir should be interpreted in light of considerable heterogeneity and possible platform effects; future studies should compare open extravesical versus open intravesical techniques separately from minimally invasive approaches. Better bilateral-specific comparative data remain a priority. Supplementary The article’s supplementary files as Acknowledgments The abstract of this study was presented as an oral presentation at the 2026 Annual Meeting of Pediatric Urology, Chinese Medical Association. Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Footnotes Reporting Checklist: The authors have completed the PRISMA reporting checklist. Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0377/rc Funding: None. Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0377/coif). The authors have no conflicts of interest to declare. References Available at https://tp.amegroups.com/article/view/10.21037/tp-2026-0377/prf Supplementary Materials The article’s supplementary files as