Objective: To determine the 1-day prevalence of community-acquired,hospital
-acquired, or intensive care unit (ICU)-acquired infections in Mexican ICUs
. To identify associated risk factors, predominant infecting organisms, and
mortality rates.
Design: A 1-day point-prevalence study.
Setting: A total of 254 adult ICUs in Mexico.
Patients: Adult patients hospitalized in the participating ICUs.
Results: A total of 895 patients were studied, of whom 521 patients (58.2%)
were infected. community-acquired infection occurred in 214 patients (23.9
%), non-ICU nosocomial infection occurred in 99 patients (11.1%), and 208 p
atients had at least one ICU-acquired infection (23.2%; 1.45 episodes/patie
nt). The most frequently reported ICU-acquired infections were pneumonia (3
9.7%), urinary tract infections (20.5%), wound infection (13.3%), and bacte
remia (7.3%). The mortality rate for the ICU-acquired infections after 6 wk
s of follow-up was 25.5%. Multivariate regression analysis showed the follo
wing risk factors for ICU-acquired infections: neurologic failure as a prim
ary cause of admission (odds ratio [OR], 1.697; 95% confidence interval [GI
], 1.001-2.839); length of stay in ICU (OR, 1.119; 95% CI, 1.091-1.151); nu
mber of therapeutic and/or diagnostic interventions during the preceding we
ek (OR, 1.118; 95% CI, 1.016-1.231); peripherally administered infusion of
hyperosmolar solutions (OR, 6.93; 95% CI, 2.452-21.661); sedative usage in
the preceding week (OR, 1.751; 95% CI, 1.183-2.602); history of an emergenc
y surgery in the preceding month (OR, 1.875; 95% CI, 1.251-2.813). The admi
nistration of antimicrobial treatment if there was an infection decreased t
he risk of death (OR, 0.406; 95% CI, 0.204-0.755).
Conclusions: Evidence of a high frequency of nosocomial infections was foun
d, and potential risk factors for acquiring infections and mortality were i
dentified. Mortality rates according to the hierarchy of the systemic infla
mmatory response syndrome in Latin American ICUs are reported.