Does Laparoscopy Beget Underuse of Partial Nephrectomy for T1 Renal Masses? Competing Treatment Decision Pathways May Influence Utilization

Abstract

It has been suggested that renal laparoscopy has resulted in an underuse of partial nephrectomy (PN) for small renal masses in the U.S. In the absence of evidence-based medicine (EBM) guide-lines, multiple-perspective reasoning is required where complete v partial nephrectomy and the laparoscopic v the open surgical approach must be considered. We report on the PN rate in a contemporary laparoscopicera series of patients with T(1) renal masses and examine the potential influence of the management decision tree on the PN rate. An actively managed database of referred patients with T(1) renal masses was utilized retrospectively. All patients were evaluated by a single fellowship-trained urologic oncologist with formal laparoscopic training. Patients were presented with a management decision tree in which PN v total nephrectomy (TN) was the first decision node, laparoscopy v open surgery was the second decision node, and the actual PN rate was reported. We then constructed a hypothetical decision tree in which the first and second decision nodes were reversed and the criteria for performing laparoscopic nephrectomy remained constant. Seventy consecutive patients were entered during a 36-month period (July 2002-June 2005). The actual PN rate was 60%: 91% for lesions <2.0 cm, 68% for lesions 2.1 to 4.0 cm, and 33% for lesions 4.1 to 7.0 cm; and 62% of patients were treated laparoscopically. When the first and second decision nodes were reversed and this hypothetical model was applied to the study cohort, the projected PN rate was 23%, and 96% of the patients were treated laparoscopically. In the hypothetical model, the PN rate fell when patients who chose laparoscopy at the first decision node were excluded from PN at the second decision node if the criteria for laparoscopic PN were not met. Laparoscopy did not appear to result in underuse of PN. We explain this by suggesting that the PN rate may be influenced by variation in the decision tree itself. Such variation is inherent in complex clinical decision making where EBM guidelines are lacking.