Association between blood pressure during vasopressor weaning and in-hospital survival: an analysis of vasopressor targets

Objectives. To analyze the association between blood pressure during vasopressor weaning and in-hospital mortality in patients admitted to an intensive care unit (ICU). Methods. Observational retrospective single-center study including patient data registered in the Medical Information Mart for Intensive Care, version 4. The outcome was in-hospital mortality. We used restricted cubic spline (RCS) functions to analyze the associations between mortality and systolic and diastolic blood pressures and mean arterial pressure (SBP, DBP, and MAP, respectively) during weaning from vasopressors. The data was stratified according to SBP, DBP, and MAP, and sensitivity was assessed with Cox regression analysis. Results. Data for 8294 patients were analyzed. The RCS functions showed that SBP, DBP, and MAP values had nonlinear U-shaped associations with in-hospital mortality. Patients were classified into the following subgroups according to points of intersection of SBP, DBP, and MAP reference values: SBP <110, 110-150, or >150 mmHg; DBP <60, 60-85, or >85 mmHg; and MAP <75, 75-110, or >110 mmHg. In the lowest blood pressure group the hazard ratio was 0.59 (95% CI, 0.52-0.66) for SBP in the 110-150 mmHg range; 0.62 (95% CI, 0.55-0.70) for DBP in the 60.85 mmHg range; and 0.64 (95% CI, 0.57-0.72) for MAP in the 75-110 mmHg range during weaning. The analysis of subgroups also indicated that blood pressures during weaning interacted with cerebral vascular disease and chronic obstructive pulmonary disease. Conclusions. Higher blood pressures during vasopressor weaning are associated with longer in-hospital survival in ICU patients. The optimum pressure ranges are SBP, 110-150 mmHg; DBP, 60-85 mmHg; and MAP, 75-110 mmHg. Blood pressures may behave differently according to diagnosis.