UCSF

Objective

To systematically aggregate and summarize existing data on fistula prevalence among patients with a history of pelvic radiotherapy for prostate cancer.

Materials and methods

We queried PubMed, Embase, and Web of Science on October 7, 2020 for peer-reviewed publications pertaining to radiation-induced fistulas in the pelvis. For meta-analysis, we used the random-effects model. We used the I² statistic to quantify heterogeneity and the Newcastle-Ottawa Scale to assess risk of bias.

Results

Our final meta-analysis included 6 cohort studies with a total of 7665 patients exposed to pelvic radiotherapy between 1967 and 2013. Median follow-up time was 35.5 months (IQR 33.5-57.5). Pooled prevalence of radiation-induced fistula across all 6 cohort studies was 0.2% (95% CI: 0.1-0.4, I² = 0.000%, P < .608). In subgroup analysis, we did not detect significant heterogeneity in fistula prevalence in patients who were re-irradiated (0.3%, 95% CI: 0.1-0.4; P = .762) or patients on concurrent chemotherapy (0.4%, 95% CI: -0.3 -1.2; P = .664) compared to those receiving their first course of radiotherapy alone. No randomized controlled trials met inclusion criteria due to ambiguous and inconsistent reporting language for fistula occurrence.

Conclusion

There is limited published literature reporting fistula as an adverse event of prostate cancer radiotherapy, especially in the medium and long-term period. Patients undergoing pelvic radiotherapy for prostate cancer appear at low short-term risk for developing fistulas. Adverse event reporting in randomized controlled trials merits greater granularity where fistulas should be reported with specificity rather than aggregating into broad categories of genitourinary or gastrointestinal adverse events.