Traditional systolic blood pressure targets underestimate hypotension-induced secondary brain injury

Abstract

Background: Vital signs, particularly blood pressure, are often manipulated to maximize perfusion and optimize recovery from severe traumatic brain injury (sTBI). We investigated the utility of automated continuously recorded vital signs to predict outcomes after sTBI.

Methods: Sixty patients with head Abbreviated Injury Scale score ≥ 3, age >14 years, "isolated" TBI, and need for intracranial pressure monitoring were prospectively enrolled at a single, large urban tertiary care facility. Outcome was measured by mortality and extended Glasgow Outcome Scale (GOSE) at 12 months. Continuous, automated, digital data were collected every 6 seconds for 72 hours after admission, and 5-minute means of systolic blood pressure (SBP) were recorded. We calculated SBP as pressure × time dose (PTD) to describe the cumulative amplitude and duration of episodes above and below clinical thresholds. The extent and duration of the insults were calculated as percent time (%time), PTD, and PTD per day (PTD/D) of defined thresholds (SBP: <90 mm Hg, <100 mm Hg, <110 mm Hg, and <120 mm Hg; mean arterial pressure: <60 mm Hg and <70 mm Hg; heart rate: >100 bpm and >120 bpm; and SpO(2): <88% and <92%) for the first 12 hours, 24 hours, and 48 hours of intensive care unit admission. We analyzed their ability to predict mortality and GOSE by receiver operator characteristics.

Results: Mean age was 33.9 (range, 16-83) years, mean admission Glasgow Coma Scale score 6.4 ± 3, and mean head Abbreviated Injury Scale score 4.2 ± 0.72. The 30-day mortality rate was 13.3%. Of the 45 patients in whom GOSE at 12 months was available, 28 (62%) had good neurologic outcomes (GOSE score >4). Traditional markers of poor outcome (admission SBP, admission Glasgow Coma Scale, and Marshall score) were not different between groups with good or poor outcome. PTD, PTD/D, and %time SBP <110 mm Hg and SBP <120 mm Hg predicted mortality at 12 hours, 24 hours, and 48 hours (p < 0.04). Percent time SBP <110 mm Hg in the first 24 hours was predictive of 12-month GOSE (p = 0.02). PTD/D SBP <120 mm Hg in the first 24 hours and PTD and PTD/D in the first 48 hours were also predictive of 12-month GOSE (p < 0.05).

Conclusions: Within the first 48 hours of intensive care unit admission, hypotension was found to be predictive of mortality and functional outcomes at higher thresholds than traditionally defined. Systemic blood pressure targets closer to 120 mm Hg may be more efficacious in minimizing secondary insults and particularly useful in settings without invasive intracranial monitoring capabilities.

Level of evidence: III, prognostic study.