[Efficiency of prognostic scores for the assessment of patients with severe influenza pneumonias].

AIM

To compare the efficiency of some prognostic scores in patients with severe influenza pneumonias.

MATERIALS AND METHODS

The study was performed on a cohort of 75 cases of 2009 AH1N1 influenza associated pneumonias. Clinical and laboratory features at admission were used to calculate retrospectively the following prognostic scores: SCAP (Severe Community Acquired Pneumonia), CAP-PIRO (Community Acquired Pneumonia--Predisposition Infection Reaction, Organ failure), SMRT-CO (Systolic blood pressure, Multilobar infiltrates, Respiration rate, Tachycardia, Confusion, Oxygen), IDSA/ATS (Infectious Diseases Society of America/American Thoracic Society). The scores were used to assess two different outcomes--death and need for invasive mechanical ventilation (IMV). The performance of the prognostic tools were assessed using the area under receiver operating characteristic (AUC), and the sensitivity and specificity for identifying high risk patients for severe course of pneumonia.

RESULTS

IMV was applied in 29 (38.7%) of studied cases, in 15 (20%) the diseases had a fatal outcome. Despite the fact that all scores had a very good discriminatory power in predicting both outcomes (AUC > 0,8), some of them have a very low sensitivity, in classes corresponding to sever pneumonias, in predicting mortality (IDSA/ATS-0%; 95% CI, 0-21.8%), as well as the need for IMV (IDSA/ATS-0%; 95% CI, 0-11.9%); SCAP-58.6% (95% CI, 38.9-76.5%); CAP-PIRO-58,6% (95% CI, 38.9-76.5%).

CONCLUSIONS

The CAP-PIRO and SMRT-CO scores were found to have the best performances to predict death from influenza associated severe pneumonias and the last, also in predicting the need for IVM. Other analyzed scores underestimate the risk of occurrence of both assessed outcomes.