Prediction tool for mortality after respiratory compromise


Background: Scoring systems exist to predict outcomes following cardiac arrest. There is currently no reliable model to predict outcome of patients who have survived acute respiratory compromise (ARC).

Study Design: A retrospective cohort study.

Setting: Get with the Guidelines Resuscitation (GWTG-R) is an online medical registry that tracks ARC data from more than 300 hospitals.

Synopsis: Using the GWTG-R database of ARC, researchers identified 13,193 cases of ARC to study the variables affecting prognosis. They randomized the group into derivation (75% of patients) and validation (25% of patients) cohorts and used c-statistics to create the prognostic scoring system. The greatest predictors of in-hospital mortality were age greater than 80 years, hypotension in the four hours preceding the ARC event, and the need for intubation.

This scoring system did not take into account any comorbidities (such as organ failure) that occurred shortly after the ARC event, although these likely affect mortality.

Bottom Line: Predicting in-hospital mortality for survivors of ARC events may help clinical prognostication. Such tools could also facilitate comparisons between hospitals and guide quality improvement projects.

Citation: Moskowitz A, Anderson LW, Karlsson M, et. al. Predicting in-hospital mortality for initial survivors of acute respiratory compromise (ARC) events: Development and validation of the ARC score. Resuscitation. 2017 Jun;115:5-10.

Dr. Suman is clinical instructor of medicine in the University of Kentucky division of hospital medicine.

   Comments ()

Next Article: