Several years ago, I joined colleagues from surgery, infection control, and epidemiology at Jackson Hole, Wyoming, for a 3-day boondoggle. We were hosted by a multinational corporation. Fly-fishing, horseback riding, hiking, and shooting rapids on the Snake River were interspersed with multiple sessions of “visioning,” “thinktanking,” and “sharing” regarding the future of surgical infection control. In these activities we were joined by corporate scientists and marketing mavens. The corporation had retained an unctuous “facilitator” to run the show, and, at the compulsory opening night welcome party held in a rustic lodge, she asked that each doctor stand before the group to recite his or her credentials. Ice tinkled in cocktail glasses, and a fire roared in the giant hearth as we took our turns. A friend of mine, who happens to be one of the best cardiac surgeons in the United States, brought down the house. Raucous laughter followed his comment: “I don’t know why you invited me and my wife, but it looks like we’re gonna have a good time. Let me just say I’m a cardiac surgeon, and I’m opposed to infections of all kinds. Thank you very much.” Postoperative infections of all kinds can ruin recovery from the most meticulously performed coronary artery bypass operations. In recent years, there has been growing support for organized infection surveillance of cardiacsurgery patients, coupled with thoughtful processimprovement initiatives as indicated. A recent article in this journal will be cited in the future as evidence that surgical-site infection (SSI) surveillance and control programs can add tangible value in cardiac-surgery practices. McConkey and coworkers at Barnes Hospital in St Louis, Missouri, tried to determine whether a newly activated surveillance and control program favorably influenced SSI risk.1 A clever strategy was used and, had the paper been a Lassie movie, regression analysis would certainly have played the starring role of the heroic dog who saves the day. Given the essential impossibility of testing mixed infection-prevention methods in a controlled trial (including the effect of having a surveillance function), the authors used regression models to allow disqualification of factors that might be adduced by skeptics like me to explain away any beneficial secular impact (ie, decreased infection rates) of surveillance and control activity. The article is not perfect, but it is important and its details are rich. It is not just another recital of one hospital’s postoperative infections, and surveillance enthusiasts should dissect it critically in a quiet room. A sharp number 2 pencil and note pad should be kept handy. Both will be needed. Surveillance was performed after 2,230 coronary artery bypass operations during a baseline year and 3 consecutive follow-up years ending with 1994. Annual infection rates were not merely compared to see what happened after numerous prevention modalities were structured and activated. Instead, the rates were “cleaned up” using several maneuvers that come from the analytical epidemiology toolbox. Final adjusted risk models showed that there had been no significant improvement relative to the index year for any of three annual rates for “deep chest infections,” an unorthodox term that included mediastinitis, the much feared and most expensive nosocomial infection in modern cardiac bypass surgery. The significant improvement discovered by these authors, after removing the influences of several confounding variables, was a striking decrease in the annual incidence rate for leg-vein harvest-site infections. This represents genuine improvement. It can be argued that the improvement touched all four healthcare outcome components (clinical, economic, patient satisfaction, and patient functional status). Readers should be
Read full abstract
Jan 27, 2023
Amnuay Kleebayoon + 1
Open Access
