Epidemiology of surgically treated gastric cancer in the United States, 1988-2000.

The incidence of gastric cancer and the need for subsequent surgery has been decreasing in the United States. However, very few population-based studies on the magnitude of these changes are available. The objective of the present study was to characterize temporal trends in the use of gastric resection in the treatment of gastric cancer. Patients with a primary diagnosis code for gastric cancer (N=105,887) and a procedure code for gastric resection (N=23,690) in the Nationwide Inpatient Sample for 1988-2000 were included. The Nationwide Inpatient Sample represents a 20% stratified random sample representative of all United States hospitals. Outcome variables included the overall incidence, in-hospital mortality rate, and length of stay. Rates of surgery are shown as the number of cases per 100,000 hospital discharges. Hospital volume was defined as follows: low volume (1 to 4 cases per year), medium volume (5 to 8 cases per year), and high volume (9 or more cases per year). Rates of gastric resection have shown a 20% decline from 30 cases per 100,000 (1988-1989) to 24 cases per 100,000 (1999-2000) (P=0.001). In-hospital mortality has not changed over the 13-year period and remains at 7.4%. There was significant variation in mortality across hospitals, with very low-volume centers having an 8.9% mortality rate, whereas very high-volume centers had a 6.4% mortality rate (P<0.001). The market share of gastric resections performed at high-volume centers increased a small amount from 43% (1988-1989) to 48% (1999-2000) (P=0.023). Over the 13-year period, length of stay decreased from 15 days (interquartile range [IQR] 11-23) in 1988 to 11 days (interquartile range [IQR] 8-16) in 2000 (P<0.001). Rates of gastric resection for cancer have shown a modest decline over the past 13 years in the United States. Although the length of stay for these patients has decreased, no significant changes to in-hospital mortality have occurred. Given the declining rates of gastric cancer surgery, and the superior outcomes at high-volume centers, regionalization of care may improve mortality rates for this high-risk surgical procedure.