News Release

Large-volume medical centers produce best clinical and economic results for complex gastrointestinal surgery

Peer-Reviewed Publication

Johns Hopkins Medicine

A handful of complicated, high-risk gastrointestinal surgeries are safer and easier on patients -- and pocketbooks -- when performed at medical centers that do the most of them, according to results of a Johns Hopkins study published in the July 1999 issue of the Journal of the American College of Surgeons.

For the study, investigators checked key clinical and economic outcomes for 4,561 consecutive patients who underwent one of five gastrointestinal operations in Maryland from July 1989 through June 1997. Specifically, they measured in-hospital deaths, average length of hospital stay, and average total hospital charges for patients undergoing excision of the esophagus, total gastrectomy (removal of the stomach), total abdominal colectomy (removal of the colon), hepatic lobectomy (removal of a liver lobe), biliary tract anastomosis (reconnecting parts of the bile transport system, or radical pancreaticoduodenectomy (removal of the pancreas head along with an encircling loop of the duodenum), also known as the Whipple procedure.

To assess the role of patient volume and surgical experience on outcomes, the researchers grouped the 51 hospitals studied into four categories based on the number of surgeries performed: minimal (10 or less surgeries per year), low (11-20), medium (21-50), and high-volume (201 or more per year).

Only one hospital, Johns Hopkins, met all the criteria for a high-volume provider, accounting for 1,711, of the 4,561 operations. In contrast, the medium-volume group, consisting of four hospitals, performed 762 procedures. There were seven low-volume hospitals which performed 750 procedures, while the minimal-volume group, consisting of 39 hospitals, performed 1,338.

The statewide in-hospital death rate totaled 8.7 percent, compared to the high-volume center's rate of 2.9 percent. The medium-, low-, and minimal-volume provider's rates were 8.4 percent, 12.7 percent, and 14.2 percent, respectively. After the rates were adjusted for age, admission status, and other diseases, patients who underwent the operations at the medium-, low, and minimal-volume providers had a 2.1, 3.3, and 3.2 times greater risk of dying in the hospital following surgery, respectively, than patients at the high-volume hospital.

Differences in average length of patient hospital stay after surgery among the hospital groups were similarly compared. The average length of stay was 14 days at the high-volume hospital compared with 16.1, 15.7, and 15.5 days at the medium-, low-, and minimal-volume centers, respectively.

Equally significant were the differences in average total hospital charges among the provider groups, says Toby A. Gordon, Sc.D., one of the study's authors and Hopkins vice president. Adjusted charges at the high-volume hospital were $21,393, about $3,000 less than the other groups. "We estimate that 64 in-hospital deaths could be avoided and approximately $2.5 million could be saved for every 1,000 patients referred from a minimal-volume provider to a high-volume provider," says Gordon.

Gordon attributes the more favorable outcomes by the high-volume hospital to a variety of factors, including greater surgical experience among high-volume physicians, more availability of specialized staff, such as nurse clinical specialists and board-certified intensivists, and greater use of state-of-the-art equipment and instruments in the operating rooms. "In addition," Gordon says, "critical pathways (clinical guidelines) and detailed care management plans may be more likely to be used at centers with larger numbers of patients."

###

The study's other authors were Helen M. Bowman; Eric B. Bass, M.D.; Keith D. Lillemoe, M.D.; Charles J. Yeo, M.D.; Richard F. Heitmiller, M.D.;, Michael A. Choti, M.D.; Gregg P. Burleyson; Ginny Hseih, M.P.H.; and John L. Cameron, M.D.

Media Contact: Gary Stephenson 410-955-5384 Email: gstephenson@jhmi.edu



Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.