Long-Term Surveillance of Complex Cystic Renal Masses and Heterogeneity of Bosniak 3 Lesions

Purpose: We sought to better characterize the frequency of Bosniak cyst class changes, identify predictors of change and progression, determine the average growth rate of cysts and validate the safety of active surveillance. Materials and Methods: Consecutive patients referred for the management of complex cysts (Bosniak 2F or greater) were included in analysis. All imaging studies were reevaluated and any stage change was reviewed with a blinded genitourinary radiologist. Bosniak 3 cysts were subclassified as 3s—septated enhancing Bosniak cysts and 3n—cysts with wall or septation-only nodularity. A multivariate Cox proportional hazards model was used to identify predictors of the Bosniak classification change. Kaplan-Meier curves were applied to analyze Bosniak cyst progression and regression. Results: A total 140 lesions were included in study, necessitating review of a total of 1,011 abdominal scans in 111 patients. Median followup was 46 months (IQR 23–65.5) and patients underwent a mean of 6.8 diagnostic scans. On multivariate analysis nodular cysts progressed (HR 6.16, 95% CI 2.58–14.72, p <0.00004) and entirely endophytic cysts were less likely to progress (HR 0.21, 95% CI 0.05–0.85, p = 0.028). On Kaplan-Meier analysis Bosniak 3s cysts were more likely to regress while Bosniak 3n cysts were more likely to progress than 3s cysts (p = 0.0178 and 0.0002, respectively). The growth rate of 3n and 2F cysts was 0.19 and 0.11 cm per year (p = 0.0493 and 0.0327, respectively). Locally advanced or metastatic disease did not develop in any patient. Conclusions: A diagnostic change in Bosniak 3s and 2F cysts is common and Bosniak 3n cysts behave more like Bosniak 4 cysts. Most complex kidney cysts can be safely monitored without intervention and the interval between serial imaging procedures should be increased.