Bladder cancer remains a commonly diagnosed malignancy worldwide, bringing huge economic burden and high morbidity for patients. Assessment of prognostic significance of lymphovascular invasion (LVI) is a critical issue in the surgical management of bladder cancer after transurethral resection or radical cystectomy. A systematic search of PubMed, Embase and Cochrane Library was performed up to Oct 10, 2014 to identify eligible studies. Outcomes of interest were collected from studies comparing overall survival (OS), cancer specific survival (CSS) and recurrence free survival (RFS) in patients with the LVI. Results of studies were pooled, and combined hazard ratios (HRs) with corresponding 95% confidence intervals (CIs) for survival were used as the effect size estimation. Funnel plots were done to show the publication bias, while the forest plots and subgroup analyses were used to limit the heterogeneity. A total of 20 studies (10 663 patients) met the eligibility criteria and were included for this meta-analysis. Our pooled results showed that there were significant differences in OS (pooled HR, 1.71; 95%CI, 1.52-1.92; P < 0.00001), CSS (pooled HR, 2.25; 95% CI, 1.80-2.81; P < 0.00001) and RFS (pooled HR, 1.91; 95% CI, 1.57-2.32; P < 0.00001) between the patients with LVI and the patients without LVI. There were significant heterogeneities observed in the studies concerning the relationship between LVI and CSS, RFS. There was no clear evidence of publication bias. When tumor stage was beyond T3, LVI lost its predictive value for CSS and RFS. For the patients who had negative lymph nodes, LVI was still an adverse predictor. Our pooled results demonstrate that LVI indicates poor prognosis of patients with bladder cancer after surgical procedures, and it can be of particular importance in clinical practice. However, these results need to be further confirmed by more adequately designed prospective studies.