Variation in use of lymph node dissection during radical cystectomy for bladder cancer.

摘要

OBJECTIVES: To better inform avenues for improving the quality of bladder cancer care, we evaluated whether the variation in pelvic lymph node dissection during radical cystectomy was primarily due to the patient or the surgeon. In the clinical guidelines, pelvic lymph node dissection has been recommended as an adjunct to radical cystectomy. However, its use and extent have varied across providers and regions. METHODS: Using the national Surveillance, Epidemiology, and End Results-Medicare linked data for 1992-2005, we identified 4472 patients who had undergone radical cystectomy for bladder cancer. Generalized linear multilevel models were fit to assess the relationships between patient and surgeon characteristics and the use and extent (>/=10 nodes) of lymphadenectomy. Using a similar modeling framework, we partitioned the variation between the patient and surgeon levels. RESULTS: Of the 4472 patients who underwent radical cystectomy, 3124 (69.9%) had undergone concurrent lymph node dissection. Of those undergoing lymphadenectomy, only 22% had >/=10 nodes removed. The use of node dissection was primarily determined by the surgeon, which explained 57% of the variation, compared with the patient and disease, which explained only 4.5% of the variability. In contrast, patient level factors explained most of the variation in whether a patient had >/=10 nodes removed. CONCLUSIONS: Pelvic lymph node dissection is relatively common during radical cystectomy, although nearly 1 in 3 patients do not undergo the procedure. Our results also showed that the physician a patient sees for their bladder cancer matters more than the disease severity in terms of the patient receiving recommended care.