BACKGROUND: In the treatment of urothelial carcinoma of the bladder, we are currently uncertain of the benefits and harms of standard pelvic lymph node dissection (PLND) compared to extended PLND. OBJECTIVES: To assess the effects of extended versus standard PLND in patients undergoing cystectomy to treat muscle-invasive (cT2 and cT3) and treatment-refractory, non-muscle-invasive (cT1 with or without carcinoma in situ) urothelial carcinoma of the bladder. SEARCH METHODS: We performed a comprehensive literature search using multiple databases (PubMed, Embase, Cochrane Controlled Trials, Web of Science, and LILACS), trial registries, and conference proceedings published up to April 29, 2019, with no restrictions on the language or status of publication. SELECTION CRITERIA: We included randomized controlled trials in which participants underwent radical cystectomy (RC) for muscle-invasive or therapy-refractory non-muscle-invasive urothelial carcinoma of the bladder with either an extended PLND with a superior extent reaching as far cranially as the inferior mesenteric vein, or a standard PLND with a superior extent of the bifurcation of the internal and external iliac artery, with otherwise the same anatomical boundaries. DATA COLLECTION AND ANALYSIS: Two review authors independently assessed the included studies and extracted data from them for the primary outcomes: time to death from any cause, time to death from bladder cancer and Clavien-Dindo classification of surgical complications grade III-V, and the secondary outcomes: time to recurrence, Clavien-Dindo I-II complications and disease-specific quality of life.We performed statistical analyses using a random-effects model and rated the certainty of evidence according to the GRADE approach. MAIN RESULTS: The search identified one multicenter trial based in Germany that enrolled 401 participants with histologically confirmed T1 grade 3 or muscle-invasive urothelial carcinoma. The median age was 67 years (range: 59 to 74) and the majority of participants were male (78.3%). No participant received neoadjuvant chemotherapy; a small subset received adjuvant chemotherapy (14.5%).Primary outcomesOur results indicate that extended PLND may reduce the risk of death from any cause over time as compared to standard PLND, but the confidence interval includes the possibility of no effect (hazard ratio [HR]: 0.78, 95% confidence interval [CI]: 0.57 to 1.07, 401 participants, low-certainty evidence). After five years of follow-up, this may result in 83 fewer deaths (95% CI: 174 fewer to 24 more overall deaths) per 1000 participants: 420 deaths for extended PLND compared to 503 deaths per 1000 for standard PLND. We downgraded the certainty of evidence by two levels due to study limitations and imprecision.Our results indicate that extended PLND may reduce the risk of death from bladder cancer over time as compared to standard PLND but, again, the confidence interval includes the possibility of no effect (HR: 0.70, 95% CI: 0.45 to 1.07, participants = 401, low-certainty evidence). After five years of follow-up, this corresponds to 91 fewer deaths per 1000 participants (95% CI: 176 fewer to 19 more bladder cancer deaths): 264 deaths for extended PLND compared to 355 deaths per 1000 for standard PLND. We downgraded the certainty of evidence by two levels due to study limitations and imprecision.Based on follow-up of up to 30 days, we are uncertain whether extended PLND leads to more grade III-V complications as compared to standard PLND, because of study limitations and imprecision (risk ratio [RR]: 1.13, 95% CI: 0.84 to 1.52, participants = 401, very low-certainty evidence).Secondary outcomesWe are uncertain whether extended PLND reduces the risk of recurrence over time as compared to standard PLND, because of study limitations and imprecision (HR: 0.84, 95% CI: 0.58 to 1.22, participants = 401, very low-certainty evidence).Based on follow-up of up to 30 days, we are uncertain whether extended PLND leads to similar grade I-II complications as compared to standard PLND because of study limitations and imprecision (RR: 0.94, 95% CI: 0.74 to 1.19, participants = 401, very low-certainty evidence).We found no trials that reported on disease-specific quality of life. AUTHORS' CONCLUSIONS: Results from a single trial indicate that extended PLND in patients undergoing radical cystectomy for invasive urothelial carcinoma of the bladder may reduce death from any cause and death from bladder cancer over time; however, the results include the possibility of no effect. We are uncertain whether the risk of serious complications up to 30 days may be increased. We are also uncertain as to whether the risk of recurrence over time or the risk of minor complications up to 30 days changes. We were unable to conduct any of the preplanned subgroup analyses, in particular, analyses based on extended lymph node dissection templates, clinical tumor stage, and use of neoadjuvant chemotherapy that may be important effect modifiers. Important additional data is expected from a larger, ongoing trial that will also consider the role of neoadjuvant chemotherapy. Inclusion of this trial in the meta-analysis may help address the issue of imprecision which was a common reason for downgrading the certainty of the evidence.