Minimum Mean Arterial Pressure and Associated Mortality Outcomes in the Cardiac Intensive Care Unit.

BACKGROUND

Hypotension is associated with increased mortality in critical care settings; limited data exist on the minimum mean arterial pressure (minMAP) in cardiac critical illness.

OBJECTIVES

The objective of the study was to investigate the prognostic value of the minMAP within the first 24 hours of admission to cardiac intensive care unit (CICU).

METHODS

This retrospective, single-center study included adult CICU patients (2007-2018). The minMAP within the first 24 hours was the exposure of interest. Patients were categorized into four minMAP groups using a classification and regression tree model. Primary outcome was in-hospital mortality; secondary, 30-day mortality. Multivariable logistic and Cox regression models were adjusted for confounders.

RESULTS

Of 11,930 patients, the median minMAP was 54 (47, 62) mm Hg, distributed as follows: ≥57 (42.4%, n = 5,053); 48 to 57 (30.5%, n = 3,635); 37 to 48 (20.1%, n = 2,392); and <37 (7.1%, n = 850). In-hospital mortality was 9.1% (n = 1,080), and 11.7% (n = 1,364) died within 30-days of CICU admission (30-day mortality 11.7% [11.1% to 12.3%] by the Kaplan-Meier method). The lower minMAP was incrementally associated with higher mortality in a continuous, reverse J-shaped manner, with the lowest mortality at 71 to 75 mm Hg. Patients with minMAP <48 mm Hg had substantially higher in-hospital (adjusted OR: 1.60 [1.36-1.88]; P < 0.001) and 30-day (adjusted HR: 1.47 [1.31-1.65]; P < 0.001) mortality. The minMAP <37 mm Hg had the highest risk in-hospital (adjusted OR: 2.19 [1.67-2.86]; P < 0.001) and 30-day (adjusted HR: 1.95 [1.61-2.36], P < 0.001) mortality. Lower minMAP was associated with higher mortality across admission diagnoses and regardless of vasoactive drugs administration.

CONCLUSIONS

We observed a graded association between lower minMAP and increased mortality, re-emphasizing severe hypotension as a critical physiological marker of patient vulnerability.