Evaluation of the Binax NOW Streptococcus pneumoniae urinary antigen assay in intensive care patients hospitalized for pneumonia.

OBJECTIVE: To evaluate the diagnostic performance of Binax Now S. pneumoniae urinary antigen rapid immunochromatographic membrane test (ICT) for patients with suspicion of community-acquired pneumonia hospitalized in intensive care and to assess the impact of prior antibiotics on its diagnostic performance. DESIGN AND SETTING: Retrospective study in a medicosurgical ICU in a 700-bed general hospital. PATIENTS: Charts of patients with ICT performed (result blinded) were reviewed between May 2002 and July 2004. ICT has been performed in 140 of the 1,006 patients hospitalized in the unit; two-thirds had received antibiotics prior to admission. MEASUREMENTS AND RESULTS: Diagnosis of pneumonia was made according to usual criteria. All patients had at least one microbiological test. Pneumonia diagnosis confirmed in 108 patients including 32 pneumococcal. ICT was positive in 23 of 32 patients with pneumococcal pneumonia, in 11 of 108 without, and in none of the 32 patients without pneumonia, resulting in sensitivity, specificity, and positive and negative predictive values of, respectively: 72%, 90%, 68% and 92%. Positive likelihood ratios were comparable among patients with or without prior antibiotics (respectively 6 and 12 for 7 in the overall population). CONCLUSIONS: This first clinical evaluation of ICT in intensive care patients hospitalized for suspicion of community-acquired pneumonia, demonstrated performance in accordance with published data even in the case of prior antibiotics. Its clinical interest and impact on antibiotics policy remain to be refined.

Pneumococcal meningitis in the intensive care unit: prognostic factors of clinical outcome in a series of 80 cases.

We have undertaken this retrospective study to determine factors associated with in-hospital mortality and morbidity in 80 adult patients with severe Streptococcus pneumoniae meningitis. Clinical characteristics at admission of patients infected with susceptible (n = 54) and nonsusceptible (n = 17) strains to penicillin G were similar: age: 51 +/- 19 versus 58 +/- 15 yr (p = 0.16); Simplified Acute Severity Score (SAPS II): 39 +/- 14 versus 41 +/- 11 (p = 0.68); and Glasgow Coma Score: 8 +/- 3 versus 9.5 +/- 3 (p = 0.21), respectively. In-hospital mortality was 25% (20/80), with one death among the 17 patients (6%) infected with a nonsusceptible strain (p = 0.03). High-dose dexamethasone was used in 22 cases. By multivariate analysis, three factors were independently associated with death: platelet count < 100 G/L (adjusted odds ratio [aOR] = 32.7; 95% CI = 3.2 to 332.5; p = 0.0032), arterial pH > 7.47 (aOR = 33.1; 95% CI = 3.4 to 319.7; p = 0.0025), and mechanical ventilation (aOR = 48.8; 95% CI = 2.6 to 901.5; p = 0.009). When adjusting for the identified prognostic factors, corticosteroids significantly reduced the risk of death (aOR = 0.069; 95% CI = 0.005 to 0.9; p = 0.048). Only SAPS II was predictive of adverse outcome (death or neurologic deficit). We conclude that in intubated patients with S. pneumoniae meningitis, hyperventilation should be used with caution. Nonsusceptibility to penicillin G is not associated with a worse outcome. High-dose corticosteroids may be beneficial in the most severely ill patients.