Use of CRB-65 and quick Sepsis-related Organ Failure Assessment to predict site of care and mortality in pneumonia patients in the emergency department: a retrospective study.

BACKGROUND: The quick Sepsis-related Organ Failure Assessment (qSOFA) is a new screening system for sepsis that has prognostic performance equal to the full SOFA for patients with suspected infection outside the intensive care unit (ICU). The predictive value of qSOFA for mortality and site of care in patients with pneumonia is not clear. The present study was designed to investigate the predictive performance of qSOFA, CRB-65 (confusion, respiratory rate ≥30/minute, systolic blood pressure <90 mmHg or diastolic blood pressure ≤60 mmHg, age ≥65 years) and CRB (confusion, respiratory rate ≥30/minute, systolic blood pressure <90 mmHg or diastolic blood pressure ≤60 mmHg) for mortality, hospitalisation and ICU admission in patients with pneumonia in the emergency department (ED).

METHODS: Retrospective analyses of published data on adult patients with pneumonia presenting between January 2012 and May 2014 were undertaken. The prevalence of 28-day mortality, hospitalisation and ICU admission were compared with regard to qSOFA, CRB and CRB-65 scores. The performance of these three systems for predicting outcomes was compared.

RESULTS: Of 1641 patients, 861 (53 %) were hospitalised (38 % in a general ward, 15 % in the ICU), and the remaining 780 (47 %) were treated as outpatients or were observed in the ED. Within 28 days, 547 (33 %) of 1641 patients died. CRB-65, CRB and qSOFA scores of patients who died, were hospitalised and admitted to the ICU than those who survived and were not hospitalised or admitted to the ICU (P < 0.001). AUC values of qSOFA for prediction of 28-day mortality, hospitalisation and ICU admission were similar to those for CRB-65 and CRB. Patients with qSOFA scores of 0, 1, 2 and 3 were associated with, respectively, mortality of 16.3 %, 24.4 %, 48.2 % and 68.4 %; prevalence of hospitalisation of 37.2 %, 47.4 %, 61.6 % and 73.7 %; and prevalence of ICU admission of 9.3 %, 9.1 %, 22.4 % and 45.3 %. Patients with qSOFA scores of 2 and 3 had a significantly higher prevalence of mortality and ICU admission than patients with identical CRB-65 scores.

CONCLUSIONS: qSOFA is better than CRB-65 for identification of a high risk of mortality and requirement of ICU admission.