M. Elebiary et al., PROGNOSTIC FACTORS OF SEVERE LEGIONELLA-PNEUMONIA REQUIRING ADMISSIONTO ICU, American journal of respiratory and critical care medicine, 156(5), 1997, pp. 1467-1472
Citations number
27
Categorie Soggetti
Emergency Medicine & Critical Care","Respiratory System
Despite the fact that the epidemiology of community-acquired pneumonia
and nosocomial Legionella infection is well known, there are no speci
fic reports dealing with severe cases of Legionella pneumophila pneumo
nia admitted to intensive care units. We undertook a prospective study
upon 84 patients with a reliable diagnosis of L. pneumophila pneumoni
a that required ICU admission. The study assessed the prognostic facto
rs, clinical, radiological and outcome variables of both nosocomial (n
= 33) and community-acquired (n = 51) cases of L. pneumophila pneumon
ia. The following variables were more common in nosocomial acquired as
compared to community-acquired Legionella pneumonia: Chronic obstruct
ive pulmonary disease (COPD) (64 versus 41%), cardiac disease (39 vers
us 10%), chronic renal failure (21 versus 4%), alcoholism (54 versus 1
8%), septic shock (33 versus 16%), and unilateral chest X-ray involvem
ent (61 versus 39%). The crude mortality rate in this study was 30% (2
5 of 84) with no differences when comparing mortality between nosocomi
al (9, 27%) to community-acquired (16, 31%) types. The univariate anal
ysis showed that cardiac disease, diabetes mellitus, creatinine greate
r than or equal to 1.8 mg/dl, septic shock, chest X-ray extension, mec
hanical ventilation, hyponatremia less than or equal to 136 mEq/L, PA(
CO2)/FIO2 < 130, and blood urea levels greater than or equal to 30 mg/
dl were factors related to poor outcome. On the other hand, the follow
ing two variables were related to better outcome: adequate treatment f
or Legionella and pneumonia improvement. The logistic regression analy
sis demonstrated that APACHE II score > 15 at admission (RR: 11.5; 95%
CI 1.75 to 76.1; p = 0.025), and serum Na levels less than or equal t
o 136 (RR: 21.3; 95% CI 1.11 to 408; p = 0.023), were the only indepen
dent factors related to death. On the other hand, improving pneumonia
is associated with better outcome in Legionnaires' disease than for pa
tients not having improving pneumonia (RR: 0.019; 95% CI: 0.036 to 0.1
06; p < 0.0001). A better understanding of the prognostic factors in c
ases of severe Legionella pneumonia will optimize our therapeutic appr
oach in this disease and help to decrease both its mortality and morbi
dity rates.