|
|
Video-Assisted Lung Cancer Resection Among Medicare Beneficiaries
F. Farjah, D. E. Wood*, M. S. Mulligan*, B. Krishnadasan, P. J. Heagerty, R. G. Symons, D. R. Flum. University of Washington, Seattle, WA,
BACKGROUND: Pulmonary resection may be performed via video-assisted thoracoscopy (VATS). The extent to which VATS is being used to manage lung cancer in the community at large is unknown, as is the effectiveness of VATS relative to conventional resection. Prior evaluations of VATS outcomes have been limited to highly selected patients cared for by surgeons and at centers with considerable experience.
METHODS: The Surveillance, Epidemiology, and End-Results (SEER)-Medicare database was used to conduct a cohort study (1994-2002) describing the use and outcomes of VATS among patients who underwent segmentectomy/lobectomy for lung cancer. Follow-up was available through 2005. A non-inferiority based analysis was conducted to explore whether overall survival rates were equivalent for VATS and conventional resection. Equivalence was defined a priori as a difference no greater than 5% for 5-year overall survival. Based on this difference, the confidence interval defining equivalency for overall survival was 0.89-1.11.
RESULTS: Among 12,958 patients who underwent segmentectomy/lobectomy (mean age 74 ± 5 years, 50% male), 6% received VATS. This proportion increased from 1% in 1994 to 9% in 2002 (p-trend <0.001). Compared to patients who underwent conventional resection, those who received VATS had higher unadjusted 5-year survival rates (48% vs 44%, p<0.02), but more frequently had tumors ≤ 3cm (29% vs 41%, p<0.001) and stage I disease (65% vs 61%, p=0.03) and more commonly underwent lymphadenectomy (45% vs 26%, p<0.001). After adjustment for patient, disease, and management characteristics, survival after VATS was not inferior to that following conventional resection (HR 0.98, 95% CI 0.86-1.10). However, VATS resections were more frequently performed by higher volume surgeons (7% vs 2%, p<0.001) and at higher volume (6% vs 3%, p<0.001) and/or teaching (7% vs 2%, p<0.001) hospitals. The hazard of death associated with VATS relative to conventional resection varied by surgeon volume (low volume HR 1.66, 95% CI 1.06-2.60; high volume HR 0.95, 95% CI 0.83-1.08), hospital volume (low volume HR 1.15, 95% CI 0.82-1.62; high volume HR 0.97, 95% CI 0.85-1.11), and hospital teaching status (non-teaching HR 1.16, 95% CI 0.90-1.50; teaching HR 0.97, 95% CI 0.84-1.12). A relatively small number of procedures performed at low volume, non-teaching hospitals precluded a formal analysis of the influence of provider factors on VATS outcomes.
CONCLUSIONS: While unadjusted survival rates associated with VATS were better than that of conventional resection, patient selection and cancer staging accounted for this difference. After adjusting for these variables, survival with VATS was at least as good as conventional resection. Most VATS procedures were performed by higher volume surgeons and at higher volume hospitals, and the inferiority of a VATS procedure was not ruled out at lower volume and non-teaching institutions. VATS actually appeared to be inferior to conventional resection when performed by lower volume surgeons, but only 8% of VATS were done by these providers. Future research should identify the contribution of surgeon and hospital factors to VATS outcomes.
Back to 2008 Annual Meeting
Back to Program Outline
|
