PURPOSE: The purpose of this study was to investigate whether a change in C-Reactive Protein (CRP) levels from 0 to 48 hours after admission can be used to predict adverse outcome in critically ill patients with severe pneumonia.
METHODS: A prospective cohort study of 25 consecutive patients with Severe Community or Nursing Home Acquired Pneumonia admitted to the Intensive Care Unit (ICU) from December 2005 to March 2006. The level of CRP was measured on admission and in 48 hours. Clinical Pulmonary Infection Score (CPIS) was calculated on admission and at 48 hours. Patients were followed during the hospital stay and their outcome was recorded.
RESULTS: A total of 25 patients were included. All patients met American Thoracic Society criteria for Severe Pneumonia. The mortality rate was 52%. Results were analyzed in two groups: survivors (n=12) and non-survivors (n=13). Mean age (years) was 78 in survivor group and 82.3 in non-survivor group. Fourty two percent of survivors versus 100% of non-survivors had mulitlobar pneumonia on chest radiograph (p=0.0016). Both groups had median CPIS of 7 on admission, which decreased to 5 in survivors and 6 in non-survivors at 48 hours. Ninety two percent of survivors and 100% of non-survivors required mechanical venitilation. Median CRP levels on admission were 24.85 mg/dl in survivors and 18.5 mg/dl in non-survivors. Eighty five percent (n=11) of non-survivors had an increase, while 83% (n=10) of survivors had a decrease in CRP levels. Mean Delta CRP from 0 to 48 hours was -4.7 in the survivor group versus +5.55 in the non-survivor group (p=0.0091). The sensitivity and specificity of Delta CRP>0 as a diagnostic test to predict mortality in our cohort was 77% and 88% respectively (p=0.0048).
CONCLUSION: A positive Delta C-Reactive Protein at 48 hours can strongly predict poor outcome in patients with Severe Pneumonia admitted to the ICU.
CLINICAL IMPLICATIONS: An early rise in CRP levels can be used to identify patients with worse prognosis who may benefit from more aggressive management.
DISCLOSURE: Lusine Melik-Adamyan, None.