Concomitant assessment of depth of sedation by changes in bispectral index and changes in autonomic variables (heart rate and/or BP) in pediatric critically ill patients receiving neuromuscular blockade.

ABSTRACT

OBJECTIVE: We evaluated whether or not changes in bispectral index (BIS) are associated with concomitant changes in autonomic variables that are in agreement with the different level of sedation that the changes in BIS indicate. DESIGN: A retrospective chart review. SETTING: A pediatric ICU of a children's hospital. METHODS AND MAIN RESULTS: Charts of patients who were receiving mechanical ventilation and IV sedation, neuromuscular blockade, and continuous BIS monitoring were enrolled in the study. Changes in BIS values > or = 30% from previous readings were evaluated to determine whether or not concomitant changes of > or = 10% in autonomic variables, in the same direction, coexisted. Forty-seven patients (35 male and 12 female) were enrolled in our study; ages ranged from 10 days to 18 years (mean, 4.2 +/- 6.2 years [+/- SD]). Twenty-five patients were < 1 year of age (53%). All patients were sedated and pharmacologically paralyzed. Overall, 387 BIS readings (15%) showed a > or = 30% change from the previously documented BIS number. These BIS changes were in agreement with heart rate (HR) changes, mean arterial pressure (MAP) changes, and both HR and MAP changes in 10.6%, 23.8%, and 5.7% of the time, respectively. The same analysis of agreement was done for patients < or = 1 year old, and results were no different from those of older patients. Among 21 patients who were not receiving any vasoactive drugs (alpha- and/or beta-adrenergic agonists) during the study period, 157 BIS recordings (15%) showed a > or = 30% change from the previously documented BIS number. The percents of agreement with HR, MAP, and HR and MAP for these patients were 14.6%, 17.2%, and 7.6%, respectively. In 26 patients who were receiving vasoactive medications during the study, 230 BIS recordings (15%) showed a > or = 30% change from the previously documented BIS number. For these patients, the percentages of agreement were 7.8%, 28.3%, and 4.3%, respectively. Agreement with MAP was significantly better than with HR for this group of patients (p < 0.05; Fisher Exact Test). SUMMARY: While significant changes in BIS are thought to reflect significant changes in depth of sedation, they have a very low rate of agreement with changes in vital signs. In the absence of BIS, the level of sedation of chemically paralyzed pediatric patients can be better guided by changes in MAP than in HR, particularly in patients receiving vasoactive drug treatment.