Severity of illness assessment in community-acquired pneumonia

Excerpt

Sir: We welcome Rello and Rodriguez's advocacy of pulse oximetry and pneumonia severity scoring in the emergency department [1]. However, we would like to clarify their interpretation of our findings comparing a physiological–social score and CURB-65 in a community-acquired pneumonia cohort [2]. As a predictor of mortality, CURB-65 achieved an AUC of only 0.79. As a predictor of admission to hospital and to level 2 or 3 (HDU or ITU) care, it achieved an AUC of 0.88 and 0.64 respectively. In comparison, our physiological score (PMEWS) achieved an AUC of 0.94 and 0.83 as a predictor of admission to hospital and to level 2 or 3 care, respectively. We believe this highlights the limitations of a score such as CURB-65 as a track-and-trigger tool and supports the use of a physiological score such as PMEWS which incorporates co-morbid factors and pulse oximetry as suggested by Rello and Rodriguez [1]. …