Presentation Authors: Michael Gross, Mark Alshak, Peter Y Cai*, Michelle Demetres, Elizabeth Mauer, Jim Hu, New York, NY
Introduction: The risk of prostate needle biopsy (PNBx) infections and sepsis is increasing, with recent estimates of 7% and 3%, respectively. Each post-PNBx infection costs an estimated $5,800 per event. With 165,000 incident prostate cancers annually in the US and greater use of active surveillance, there is greater need for strategies to decrease post-PNBx infections. We performed a meta-analysis with contemporary evidence to evaluate transrectal (TR) vs. transperineal (TP) PNBx infections to determine the importance of biopsy approach.
Methods: A comprehensive literature search was conducted in July 2018 in Ovid MEDLINE, Ovid EMBASE and Cochrane Library using predefined inclusion/exclusion criteria. Independent reviewers (MG and MA) screened a total of 2898 records and reviewed 49 full texts. Eight studies were included in our final meta-analysis. All double-arm comparative studies, including randomized-controlled trials and cohort studies, reporting the infectious complication rates following prostate biopsy were included. All TP biopsies were performed under general anesthesia. Meta-analysis was performed using the Mantel-Haenszel Method and Dersimonian-Laird estimator for Ï„2. Heterogeneity was analyzed using the Cochrane Q and I2 tests.
Results: TR PNBx was associated with higher odds for infection (Figure 1) in both the fixed effect model (odds ratio [OR] 3.40, 95% confidence interval [CI] 1.90â€“6.10, p < 0.01) and random effects model (OR 3.21, 95% CI 1.75â€“5.86, p < 0.01). Heterogeneity was analyzed using the Cochrane Q (4.15, p=0.76) and I2 (45.4%) tests. Funnel plot analysis shows minimal publication bias (Figure 2).
Conclusions: The meta-analysis demonstrates that TR versus TP PNBx was associated with more than three times higher odds ratio for infectious complications. Our findings may guide the adoption of targeted biopsy and in-office TP approaches, which currently lacks comparative evidence.