Laparoscopic cryoablation versus partial nephrectomy for the treatment of small renal masses: systematic review and cumulative analysis of observational studies.

CONTEXT: For small renal masses (SRMs), partial nephrectomy (PN) represents the therapeutic standard of care. Laparoscopic cryoablation (LCA) could be regarded as an alternative to surgical excision in selected patients, if perioperative complication rates and oncologic results are comparable. OBJECTIVE: To perform a cumulative analysis of observational studies regarding oncologic outcomes and perioperative complications of both procedures. EVIDENCE ACQUISITION: Medline, Embase, and Web of Science searches were performed for clinically localized sporadic SRMs that were treated with PN or LCA. A total of 6785 lesions were analyzed for local and metastatic tumor progression and 10 906 procedures for perioperative complications. EVIDENCE SYNTHESIS: Patients undergoing LCA were significantly older, mean tumor sizes were lower, and mean follow-up duration was shorter (each p<0.001). Following LCA and PN, 8.5% and 1.9% developed local tumor progression, respectively (p<0.001). In multivariable analysis, the relative risk for local tumor progression of LCA versus PN was 5.24-fold increased (p<0.001); the risk of metastatic progression was similar. The overall complication rate was higher following PN (23.5% vs 17.0%; p<0.001), especially the rate of major complications (19.2% vs 10.2%; p<0.001). In multivariable analysis, the total risk for complications and major complications for PN versus LCA was 4.6-fold (p=0.004) and 9.71-fold (p<0.001) increased, respectively. Limitations of this analysis include follow-up and selection bias, and lack of standardization reporting complications and outcomes. CONCLUSIONS: Both PN and LCA are viable options for the management of SRMs. Compared with PN, LCA results in a higher risk of local tumor progression. The risk of perioperative complications appears to be lower following LCA; however, this difference is strongly influenced by selection bias, and thus limited conclusions can be made regarding true differences in complications. Therefore, PN is the gold standard for SRMs, but LCA may be indicated in selected patients with significant comorbidity.