Nosocomial pneumonia in patients undergoing heart surgery

To determine the risk factors related to the presence of postsurgical nosocomial pneumonia (NP) in patients who had undergone cardiac surgery.

A case-control study.

Postcardiac surgical intensive care unit at a university center.

A total of 45 patients with NP and 90 control patients collected during a 4-yr period.

Pre-, intra-, and postoperative factors were collected and compared between two groups of patients (cases vs. controls) to determine their influence on the development of NP. The diagnosis of NP was always microbiologically confirmed as pulmonary specimen brush culture of > or =10(3) colony-forming units/mL or positive blood culture/pleural fluid culture by the growth of identical microorganisms isolated at the lung. For each patient diagnosed with NP, we selected control cases at a ratio of 1:2.

The incidence of NP was 6.5%. Multivariate analysis found a probable association of the following variables with a greater risk for the development of NP: reintubation (adjusted odds ratio [AOR], 62.5; 95% confidence interval [CI], 8.1-480; p = .01); nasogastric tube (AOR, 19.7; 95% CI, 3.5-109; p = .01), transfusion of > or =4 units of blood derivatives (AOR, 12.8; 95% CI, 2-82; p = .01) and empirical treatment with broad-spectrum antibiotics (AOR, 6.6; 95% CI, 1.2-36.8; p = .02). Culture results showed 13.3% of the NP to be of polymicrobial origin, whereas 77.3% of the microorganisms isolated were Gram-negative bacteria. The mortality (51 vs. 6.7%, p < .01) and the length of stay in the intensive care unit (25+/-14.8 days vs. 5+/-5 days, p < .01) were both greater in patients with NP.

We conclude that the surgical risk factors, except the transfusion of blood derivatives, have little effect on the development of NP. Reintubation, nasogastric tubing, previous therapy with broad-spectrum antibiotics, and blood transfusion are factors most likely associated with NP acquisition.