BOSTON – In sarcoma as in other cancers, experience counts.
That’s the conclusion of investigators who found that patients with extra-abdominal sarcomas who were treated in high-volume hospitals had half the 30-day mortality rate, higher likelihood of negative surgical margins, and better overall survival, compared with patients treated in low-volume hospitals.
“It appears there is a direct association between hospital volume and short-term as well as long-term outcomes for soft-tissue sarcomas outside the abdomen,” said Dr. Sanjay P. Bagaria of the Mayo Clinic in Jacksonville, Fla.
The findings support centralization of services at the national level in centers specializing in the management of sarcomas, he said at the annual Society of Surgical Oncology Cancer Symposium.
Previous studies have shown that patients treated in high-volume centers for cancers of the esophagus, pancreas, and lung have better outcomes than patients treated in low-volume centers, he noted.
Given the rarity of sarcomas, with an incidence of approximately 12,000 in the U.S. annually, their complexity, with more than 60 histologic subtypes, and the multimodality approach required for them, it seemed likely that a positive association between volume and outcomes could be found, Dr. Bagaria said.
He and colleagues queried the U.S. National Cancer Database, a hospital registry of data from more than 1,500 facilities accredited by the Commission on Cancer.
They drew records on all patients diagnosed with extra-abdominal sarcomas from 2003 through 2007 who underwent surgery at the reporting hospitals, and divided the cases into terciles as either low volume (3 or fewer surgical cases per year), medium volume (3.2-11.6 per year), or high volume (12 or more cases per year).
One third (33%) of all cases were concentrated in just 44 high-volume hospitals, which comprised just 4% of the total hospital sample of 1,163. An additional third (34%) of cases were managed among 196 medium-volume hospitals (17% of the hospital sample), and the remaining third (33%) were spread among 923 low-volume hospitals (79%).
The 30-day mortality rates for low-, medium-, and high-volume hospitals, respectively, were 1.7%, 1.1%, and 0.6% (P less than .0001).
Similarly, the rates of negative margins (R0 resections) were 73.%, 78.2%, and 84.2% (P less than .0001).
Five-year overall survival was identical for low- and medium-volume centers (65% each), but was significantly better for patients treated at high-volume centers (69%, P less than .001)
Compared with low-volume centers, patients treated at high-volume centers had an adjusted odds ratio (OR) for 30-day mortality of 0.46 (P = .01), an adjusted OR for R0 margins of 1.87 (P less than .001), and OR for overall mortality of 0.92 (P = .04).
Dr. Bagaria noted that the study was limited by missing data about disease-specific survival and by possible selection bias associated with the choice of Commission on Cancer-accredited institutions, which account for 70% of cancer cases nationwide but comprise one-third of all hospitals.