Premium
The Assessment of Risk in Cardiothoracic Intensive Care ( ARC t IC ): prediction of hospital mortality after admission to cardiothoracic critical care
Author(s) -
Shahin J.,
FerrandoVivas P.,
Power G. S.,
Biswas S.,
Webb S. T.,
Rowan K. M.,
Harrison D. A.
Publication year - 2016
Publication title -
anaesthesia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.839
H-Index - 117
eISSN - 1365-2044
pISSN - 0003-2409
DOI - 10.1111/anae.13624
Subject(s) - medicine , emergency medicine , glasgow coma scale , creatinine , cardiothoracic surgery , intensive care , intensive care unit , intensive care medicine , surgery
Summary The models used to predict outcome after adult general critical care may not be applicable to cardiothoracic critical care. Therefore, we analysed data from the Case Mix Programme to identify variables associated with hospital mortality after admission to cardiothoracic critical care units and to develop a risk‐prediction model. We derived predictive models for hospital mortality from variables measured in 17,002 patients within 24 h of admission to five cardiothoracic critical care units. The final model included 10 variables: creatinine; white blood count; mean arterial blood pressure; functional dependency; platelet count; arterial pH ; age; Glasgow Coma Score; arterial lactate; and route of admission. We included additional interaction terms between creatinine, lactate, platelet count and cardiac surgery as the admitting diagnosis. We validated this model against 10,238 other admissions, for which the c index (95% CI ) was 0.904 (0.89–0.92) and the Brier score was 0.055, while the slope and intercept of the calibration plot were 0.961 and −0.183, respectively. The discrimination and calibration of our model suggest that it might be used to predict hospital mortality after admission to cardiothoracic critical care units.