PURPOSE: Although the clinical pulmonary infection score (CPIS) was created as a diagnostic tool for ventilator-associated pneumonia (VAP), changes in the score over time are associated with clinical outcomes in VAP patients. It is unknown whether changes in the CPIS are similarly associated with clinical outcomes in healthcare associated pneumonia (HCAP).
METHODS: In this single-center study we retrospectively reviewed the charts of all patients hospitalized for pneumonia in 2008 at an academic medical center. Data were abstracted for all patients with HCAP in order to calculate the CPIS at the time of admission and after 72 hours of therapy. Changes in the CPIS score were then correlated with clinical outcomes.
RESULTS: The study cohort included 280 patients with HCAP: 39 patients (13.9%) died during hospitalization. Baseline CPIS values were not significantly different between survivors and non-survivors (4.94 vs 5.38, P=0.13). However, the 72-hour values were significantly lower in survivors than in non-survivors (3.13 vs 6.05, P<0.001). The cohort of patients who survived had a clear improvement in the CPIS from baseline to 72 hours while those who died had no change in the CPIS (-1.81 vs. 0.67, P<0.001). Multivariate logistic regression modeling confirmed that the change in CPIS is an independent predictor of mortality. Variables included in the final multivariate model included the change in CPIS (AOR 1.89, P<0.001) and the initial APACHE II score (AOR 1.12, P<0.001).
CONCLUSIONS: Changes in the CPIS values are distinctly different between HCAP survivors and non-survivors during the first 72-hours of treatment.
CLINICAL IMPLICATIONS: Longitudinal changes in CPIS values may be useful in predicting clinical outcomes for patients with HCAP.
DISCLOSURE: The following authors have nothing to disclose: Lee Morrow, Cole Deutz, Kunal Desai, Joshua Holweger, Doug Moore, Mark Malesker
No Product/Research Disclosure Information