Jy. Fagon et al., NOSOCOMIAL PNEUMONIA AND MORTALITY AMONG PATIENTS IN INTENSIVE-CARE UNITS, JAMA, the journal of the American Medical Association, 275(11), 1996, pp. 866-869
Objective.-To evaluate the role that nosocomial pneumonia plays in the
outcome of intensive care unit (ICU) patients. Design.-Cohort study.
Setting.-Medical ICU, Hospital Bichat, Paris, France, an academic tert
iary care center. Patients.-A total of 1978 consecutive patients admit
ted to the ICU for at least 48 hours. Main Outcome Measures.-Various p
arameters known to be strongly associated with death of ICU patients w
ere recorded: age, location before admission to the ICU, diagnostic ca
tegories, Acute Physiology and Chronic Health Evaluation (APACHE) II s
core, Simplified Acute Physiologic Score, McCabe score, number and typ
e of dysfunctional organs, and the development of nosocomial bacteremi
a and nosocomial urinary tract infection. These variables and the pres
ence or absence of nosocomial pneumonia were compared between survivor
s and nonsurvivors and entered into a stepwise logistic regression mod
el to evaluate their independent prognostic roles. Results.-Nosocomial
pneumonia developed in 328 patients (16.6%) whose mortality rate was
52.4% compared with 22.4% for patients without ICU-acquired pneumonia
(P<.001). APACHE II score (odds ratio [OR]=1.08; 95% confidence interv
al [CI], 1.06 to 1.10; P<.001), number of dysfunctional organs (OR=1.5
4; 95% CI, 1.36 to 1.74; P<.001), nosocomial pneumonia (OR=2.08; 95% C
I, 1.55 to 2.80; P<.001), nosocomial bacteremia (OR=2.51; 95% CI, 1.78
to 3.55; P<.001), ultimately or rapidly fatal underlying disease (OR=
1.76; 95% CI, 1.38 to 2.25; P<.001), and admission from another ICU (O
R=1.30; 95% CI, 1.01 to 1.68; P=.04) were significantly associated wit
h mortality. Conclusion.-These data suggest that, in addition to the s
everity of underlying medical conditions and nosocomial bacteremia, no
socomial pneumonia independently contributes to ICU patient mortality.