Severe Acute Respiratory Infections (SARIs) are defined by the World Health Organisation as acute respiratory infections of recent onset (within 7 days) that include fever (≥ 38 °C), cough, and dyspnoea requiring overnight hospitalisation. SARI in ICU patients represents a major concern for physicians because of the high mortality and morbidity rate attributable to these episodes. There are many factors that influence SARI development. Microbiological infective type strictly depends upon the environment in which pneumonia is contracted and the immune condition of the patients. However, despite the wide availability of invasive diagnostic procedures in ICU and new molecular techniques which make microbiological diagnosis faster and accurate, little is known about the epidemiological, clinical and prognostic impact of such serious events in ICU patients.
The authors of this paper performed a prospective, observational, inception cohort study with the aim to investigate the epidemiological, microbiological and clinical profile of patients with ICU-SARI. Participation of the centres to the study (organised by the ESICM Trials Group) was entirely voluntary without any financial support. Data collection was performed during two separate weeks (November 2013 and January 2014) in order to identify the vast majority of respiratory pathogens. The primary end-point was 60-days hospital mortality; secondary outcomes included ICU-deaths, length of stay and the degree of organ failures. A three-level statistical analysing method was used to identify risk factors associated with in-hospital outcome.
A total of 5550 patients from 206 hospitals were screened and 663 patients with SARI were included in the study. The majority of admissions were community-acquired (68.8%) and the most common comorbidities were hypertension and chronic obstructive pulmonary disease. Predominant symptoms were fever, cough and dyspnoea, and almost 90% of the patients received empirical antibiotics. First antimicrobial regimen was changed in one third of the cases in light of the microbiological results of diagnostic procedures (i.e. blood cultures and respiratory samples). More than half of isolated microorganisms were bacteria (29.6% Gram-positive, 26.2% Gram-negative), viruses (mostly Influenza A) were detected in about 10% of the cases. Gram-negative bacteria and fungi were most frequently isolated in hospital-acquired SARI. Organ failure (mostly cardiovascular, renal and respiratory) was present in 64.4% of the patients: ICU and hospital mortality rates were 20.2% and 27.2%, respectively. Multivariate analysis revealed that severity score on ICU admission, haematological malignancy and liver disease were independently associated with an increased risk of in-hospital death; on the other hand influenza vaccination prior to ICU admission and adequate antimicrobial empirical therapy were the only predictors of hospital survival. Given the above described interesting results, some limitations should be acknowledged, in particular reported data do not take into account the possible regional/seasonal differences and may be less accurate in differentiating respiratory colonisation from true infections.
Take Home message
Severe Acute Respiratory Infections in ICU are a common cause of in-hospital morbidity and mortality. The severity of the disease and specific comorbidities are main determinants of poor outcome, whereas a prompt appropriate antimicrobial treatment and previous influenza vaccination represent the only valid measures to reduce the clinical detrimental impact associated with SARI.
Article review was submitted by Gennaro De Pascale on behalf of the ESICM NEXT Committee.
Yasser Sakr, Ricard Ferrer, Konrad Reinhart, Richard Beale, Andrew Rhodes, Rui Moreno, Jean Francois Timsit, Laurent Brochard, B. Taylor Thompson, Ederlon Rezende, Jean Daniel Chiche. The Intensive Care Global Study on Severe Acute Respiratory Infection (IC-GLOSSARI): a multicenter, multinational, 14-day inception cohort study. Intensive Care Med. Seven-Day Profile Publication, Volume 42, Issue 5 / May, 2016; Pages 817 – 828