SOLITARY METASTATIC BRAIN tumors are the most common intracranial neoplasms encountered by neurosurgeons. Surgical resection of brain metastasis with whole brain radiotherapy (WBR) significantly increases survival in comparison with WBR alone. Stereotactic radiosurgery (SR) seems to provide results that are similar to those of surgical resection. To analyze the economic efficiency of these different treatments, we compared the results of surgical resection and SR as reported in the medical literature between 1974 and 1994. We included studies in which: 1) at least 75% of patients received WBR; 2) study dates were in the computed tomography era (after 1975); 3) operative morbidity, mortality, and median survival were reported; 4) study dates were not included in a more recent update or review; 5) tumor histologies were reported; and 6) the cobalt-60 gamma unit was used for SR. Three surgical resection studies and one SR study met all entry requirements. The WBR baseline was developed from two prospective, randomized trials and used for incremental cost effectiveness analysis. We developed a model of typical resource usage for uncomplicated procedures, reported complications, and subsequent craniotomies (for recurrent tumor or radiation necrosis) for both treatment options. Costs were estimated from the societal viewpoint using the 1992 Medicare Provider Analysis and Review database with average cost:charge ratios for surgery and WBR. A survey of capital and operating costs from five sites was used for radiosurgery. Our analysis revealed that radiosurgery had a lower uncomplicated procedure cost ($20,209 versus $27,587), a lower average complication cost per case ($2,534 versus $2,874), and a lower total cost per procedure ($22,743 versus $30,461), was more cost effective ($24,811 versus $32,149 per life year), and had a better incremental cost effectiveness ($40,648 versus $52,384 per life year) than surgical resection. A sensitivity analysis revealed that large changes in key assumptions would be required to change the analysis outcome. Equalization of the incremental cost effectiveness of the two treatments would require one of the following: 1) a 38.7% reduction in SR annual case volume, 2) a 34.7% increase in SR procedure cost, 3) a 18.8% reduction in surgical resection procedure cost, 4) a 240.5% increase in SR morbidity cost, 5) a 12.7% reduction in SR median survival, 6) a 16.8% increase in surgical resection median survival. Elimination of all surgical resection morbidity cost would still result in superior incremental cost effectiveness for SR. These results indicate the need for prospective clinical trials that examine both the clinical efficacy and the cost effectiveness of surgical resection and SR in the management of solitary metastatic brain tumors.
Departments of Neurological Surgery (MJR, LDL, DK), Radiology (LDL), and Radiation Oncology (LDL, DK), University of Pittsburgh School of Medicine, and The Center for Image Guided Neurosurgery (MJR, LDL, DK), University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania; Health Technology Associates, Inc. (MJS, VK), Washington, District of Columbia; and M. Green Associates (MG), Reston, Virginia
Reprint requests: Michael J. Rutigliano, M.D., M.B.A., Department of Neurological Surgery, University of Pittsburgh School of Medicine, B400 Presbyterian-University Hospital, 200 Lothrop Street, Pittsburgh, PA 15213-0988.
Received, February 13, 1995. Accepted, March 20, 1995.