Science & Technology/Pitt-led Research Calls for Testing Hospital Water Supplies for Legionnaires’ Disease Bacteria

Issue Date: 
September 10, 2007

Pitt’s Yu, Stout are lead authors of 20-hospital study

water-faucet.jpg

A new study spearheaded by Pitt’s School of Medicine has determined that environmental monitoring of institutional water systems can help to predict the risk of hospital-acquired Legionella pneumonia, better known as Legionnaires’ disease.

Reported recently in the journal Infection Control and Hospital Epidemiology, the 20-hospital study also calls for reconsideration of the current national infection-control policy to include routine testing of hospital water systems for Legionella, the bacterial group associated with Legionnaires’.

“Only those hospitals that had high levels of Legionella bacteria in their water systems had patients who contracted Legionnaires’ disease,” senior author and Pitt Professor of Medicine Victor L. Yu said of the study, which involved hospitals in 14 states. “Proactive monitoring of the hospital water supply alerted physicians to the hidden risk of Legionnaires’ disease for their patients.”

Legionella bacteria first were identified as causing pneumonia in 1976 following an outbreak among attendees of an American Legion convention at a Philadelphia hotel, resulting in the name Legionnaires’ disease. There are an estimated 20,000 cases of Legionnaires’ in the United States annually, many of them hospital-acquired, with an average fatality rate of 28 percent.

Currently, the U.S. Centers for Disease Control and Prevention recommends that hospitals and other health care institutions monitor patients for pneumonia incidence before doing environmental surveillance of water systems that can harbor the bacteria.

“Based in part on our work, and in collaboration with the Allegheny County Health Department and the Three Rivers Association for Professionals in Infection Control, the development of proactive guidelines for hospital-acquired Legionnaires’ disease prevention has led to the virtual disappearance of this infection in Pittsburgh,” said study first author Janet Stout, a research assistant professor in the Pitt School of Engineering’s Department of Civil and Environmental Engineering. “We first reported the connection between hospital water supply and these infections in 1982.”

dr-victor-l-yu.jpg

                 Victor L. Yu 

For this investigation, Yu, Stout, and colleagues evaluated samples of hospital-system water at 20 facilities across the country from 2000 to 2002. Water samples were retrieved from at least 10 separate sites at each hospital on multiple occasions over the two-year period. When cases of Legionnaires’ were identified, patient urine and sputum samples from 12 of the hospitals were tested to determine classification of Legionella, which has at least 48 strains.

The researchers found that 14 (70 percent) of hospital water systems tested positive for Legionella species, and that six (43 percent) positive-testing hospitals had high-level colonization. Legionnaires’ cases were found among the 633 patients with hospital-acquired pneumonia whose urine or sputum samples were tested for Legionella bacteria. All were traced to hospitals with high-level colonization.

“Our study provides much-needed evidence to support a national policy change to include routine environmental surveillance of health care facility water systems along with stringent clinical monitoring of patients,” said Stout, who estimates that 39,000 people have died of Legionnaires’ since 1982. “We think this long-overdue approach should be adopted by infection-control and infectious-disease practitioners nationwide.”

This study was based on the Pittsburgh methodology of routine testing of hospital water systems that also has been adopted by New York, Maryland, France, Germany, Spain, the Netherlands, and Italy.

Other authors and members of the Legionella Study Group included researchers at the VA Pittsburgh Healthcare System; William Beaumont Hospital, Royal Oak, Mich.; the VA Medical Center, Omaha, Neb.; Southern Arizona Healthcare System, Tucson, Ariz.; the VA Medical Center, Wilmington, Del.; the Louis Stokes VA Medical Center, Cleveland, Ohio; the VA Medical Center, Dayton, Ohio; Stratton VA Medical Center, Albany, N.Y.; the VA Medical Center, Butler, Pa.; VA Medical Center, Iowa City, Iowa; the VA Medical Center, Gainesville, Fla.; the VA Palo Alto Health Care System, Palo Alto, Calif.; and the VA Medical Center, Long Beach, Calif.

The study was funded by a Department of Veterans Affairs Merit Review grant.