Rewarming Rate During Cardiopulmonary Bypass Is Associated With Release of Glial Fibrillary Acidic Protein

Abstract

Background: Rewarming from hypothermia during cardiopulmonary bypass (CPB) may compromise cerebral oxygen balance, potentially resulting in cerebral ischemia. The purpose of this study was to evaluate whether CPB rewarming rate is associated with cerebral ischemia assessed by the release of the brain injury biomarker glial fibrillary acidic protein (GFAP).

Methods: Blood samples were collected from 152 patients after anesthesia induction and after CPB for the measurement of plasma GFAP levels. Nasal temperatures were recorded every 15 min. A multivariate estimation model for postoperative plasma GFAP level was determined that included the baseline GFAP levels, rewarming rate, CPB duration, and patient age.

Results: The mean rewarming rate during CPB was 0.21° ± 0.11°C/min; the maximal temperature was 36.5° ± 1.0°C (range, 33.1°C to 38.0°C). Plasma GFAP levels increased after compared with before CPB (median, 0.022 ng/mL versus 0.035 ng/mL; p < 0.001). Rewarming rate (p = 0.001), but not maximal temperature (p = 0.77), was associated with higher plasma GFAP levels after CPB. In the adjusted estimation model, rewarming rate was positively associated with postoperative plasma log GFAP levels (coefficient, 0.261; 95% confidence intervals, 0.132 to 0.390; p < 0.001). Six patients (3.9%) experienced a postoperative stroke. Rewarming rate was higher (0.3° ± 0.09°C/min versus 0.2° ± 0.11°C/min; p = 0.049) in the patients with stroke compared with those without a stroke.

Conclusions: Rewarming rate during CPB was correlated with evidence of brain cellular injury documented with plasma GFAP levels. Modifying current practices of patient rewarming might provide a strategy to reduce the frequency of neurologic complications after cardiac surgery.

Figure 1

Box and whiskers plots showing the changes in plasma glial fibrillary acidic protein (GFAP) levels from the measurements obtained before and after cardiopulmonary bypass (CPB). The horizontal line in the shaded box represents the median value, and the shaded box represents the interquartile range. The error bars below and above the shaded area represents ±1.5× the interquartile range; points beyond the error bar are outliers.

Figure 2

Scatter plot of glial fibrillary acidic protein (GFAP) with rewarming rate (A) and maximum temperature (B). The long dash line represent 95% confidence interval for the fitted line. There was a significant correlation between log GFAP and rewarming rate (r=0.26, 95%CI, 0.11 to 0.41, p=0.001). There were no significant correlation between log GFAP and maximum temperature (r=0.02; 95%CI, −0.14 to 0.18; p=0.77).

Figure 3

Bar graph (A) and box and whisker plot (B) showing the relationship between rewarming rate (p=0.048) and plasma glial fibrillary acidic protein (GFAP) levels for patients with and without postoperative stroke. For the latter, the horizontal line in the shaded box for the box and whisker plot represents the median value, and the shaded box represents the interquartile range. The error bars below and above the shaded area represents ±1.5× the interquartile range; points beyond the error bar are outliers.