Abstract No. 461 - Prophylactic arterial embolization of renal tumors to reduce hemorrhagic complications following cryoablation

Abstract

The incidence of hemorrhagic complications following percutaneous cryoablation of renal tumors is non-trivial, occurring in up to 4% of cases. Patients with tumors exceeding 5 cm are at greatest risk for bleeding, and in our practice we routinely embolize large tumors prior to ablation in order to mitigate this risk. Recently, some authors have disputed the utility of pre-embolization in this setting, although data on this subject are sparse. To that end, we aimed to determine the effect of prophylactic tumor embolization on hemorrhagic complications following renal cryoablation. In this single-institution retrospective study, 837 patients underwent renal cryoablation from 2003-2016. A total of 45 patients treated with pre-embolization were identified and matched 1:1 on the basis of primary tumor size to patients who were not pre-embolized. Clinically significant (CTCAE grade 3 + ) hemorrhagic complications were identified through a review of the electronic medical record and relevant imaging. Conditional logistic regression was used to test for a difference in the complication rates while controlling for the matched pair design. 45 pre-embolized patients (mean age 72 ± 12 years; 71% male) were matched to 45 non-embolized controls (71 ± 13 years, 80% male). Mean tumor size was 5.2 ± 1.7 cm (range: 1.7-9.7 cm). Overall, 71% of pairs were matched identically on the basis of tumor size, with 95% matched within 5 mm; the largest difference in size was 9 mm. Three hemorrhagic complications (6.7%) occurred in the pre-embolization group and six (13.3%) in the control group. The odds ratios for developing a hemorrhagic complication were 60% lower in the pre-embolization group; however, the effect was not statistically significant in this sample (OR = 0.40; 95% CI: 0.08–2.06; p = 0.27). While the OR did not reach statistical significance in this sample due to limited power, the magnitude of the effect warrants further investigation given the potential clinical significance of any reduction in hemorrhagic complications. A prospective study, which would benefit from randomization to balance the sample and account for putative confounding variables, is supported by these data.