Functional Outcome Trajectories After Out-of-Hospital Pediatric Cardiac Arrest

Abstract

Objectives: To analyze functional performance measures collected prospectively during the conduct of a clinical trial that enrolled children (up to age 18 yr old), resuscitated after out-of-hospital cardiac arrest, who were at high risk of poor outcomes.

Design: Children with Glasgow Motor Scale score less than 5, within 6 hours of resuscitation, were enrolled in a clinical trial that compared two targeted temperature management interventions (THAPCA-OH, NCT00878644). The primary outcome, 12-month survival with Vineland Adaptive Behavior Scale, second edition, score greater or equal to 70, did not differ between groups.

Setting: Thirty-eight North American PICUs.

Participants: Two hundred ninety-five children were enrolled; 270 of 295 had baseline Vineland Adaptive Behavior Scale, second edition, scores greater or equal to 70; 87 of 270 survived 1 year.

Interventions: Targeted temperatures were 33.0°C and 36.8°C for hypothermia and normothermia groups.

Measurements and main results: Baseline measures included Vineland Adaptive Behavior Scale, second edition, Pediatric Cerebral Performance Category, and Pediatric Overall Performance Category. Pediatric Cerebral Performance Category and Pediatric Overall Performance Category were rescored at hospital discharges; all three were scored at 3 and 12 months. In survivors with baseline Vineland Adaptive Behavior Scale, second edition scores greater or equal to 70, we evaluated relationships of hospital discharge Pediatric Cerebral Performance Category with 3- and 12-month scores and between 3- and 12-month Vineland Adaptive Behavior Scale, second edition, scores. Hospital discharge Pediatric Cerebral Performance Category scores strongly predicted 3- and 12-month Pediatric Cerebral Performance Category (r = 0.82 and 0.79; p < 0.0001) and Vineland Adaptive Behavior Scale, second edition, scores (r = -0.81 and -0.77; p < 0.0001). Three-month Vineland Adaptive Behavior Scale, second edition, scores strongly predicted 12-month performance (r = 0.95; p < 0.0001). Hypothermia treatment did not alter these relationships.

Conclusions: In comatose children, with Glasgow Motor Scale score less than 5 in the initial hours after out-of-hospital cardiac arrest resuscitation, function scores at hospital discharge and at 3 months predicted 12-month performance well in the majority of survivors.

Figure 1. Relationship of VABS-II and PCPC scores

A–C illustrate the VABS-II scores that corresponded with each PCPC score category in all THAPCA-OH enrolled subjects with available data at baseline (n= 269) (A), at 3 months (n=99) (B), and at 12 months (n=94) (C). PCPC category definitions are: 1= Normal; 2 = Mild disability; 3 = Moderate Disability; 4 = Severe disability; 5 = Coma or vegetative. Data for all cases (left bar), and separately for hypothermia (middle bar) and normothermia (right bar) groups, are presented as box and whisker plots. Each shaded box represents the interquartile range (IQR)(middle one-half of the data), the horizontal line within the box is the median, and symbols (◇) are means; upper and lower whiskers represent the maximum and minimum values excluding outliers (defined as data points more than 1.5 IQRs from the box). The relationships between VABS-II scores and PCPC categories are all significant (p<0.001): (Pearson correlations: A. r = −0.61; B–C. r = −0.9). There were no differences in the relationships between VABS-II scores and PCPC categories between normothermia and hypothermia groups.

Figure 2. Hospital discharge PCPC score and 3 and 12 month PCPC scores

These graphs illustrate the distributions of 3 month (A) and 12 month (B) PCPC scores that corresponded with each category of hospital discharge PCPC score in THAPCA-OH primary outcome group survivors. The y axis labels are hospital discharge PCPC scores; numbers of survivors per PCPC category with available data at 3 (A) and 12 months (B) respectively were: PCPC 1, n= 22 and 22; PCPC 2, n=10 and 9; PCPC 3, n=8 and 8; PCPC 4, n=21 and 23; PCPC 5, n=23 and 23. Each horizontal bar illustrates the percentage of cases in each PCPC category at follow-up (numbers within each section of the bars are the assigned follow-up scores). Hospital discharge PCPC scores strongly influenced subsequent scores (Pearson correlations: r=0.82, 0.79; both p<0.0001). Treatment group had no effects (not shown).

Figure 3. Hospital discharge PCPC score and 3 and 12 month VABS-II scores

Panels A–B illustrate 3 month (A) and 12 month (B) VABS-II scores that correspond with each category of hospital discharge PCPC score in THAPCA-OH survivors eligible for the primary outcome. Data are presented for all cases (left bar) and for hypothermia (middle bar) and normothermia (right bar) groups, as box and whisker plots. Each shaded box represents the interquartile range (IQR) (middle one-half of the data), the horizontal line within the box is the median, and symbols (◇) are means, upper and lower whiskers represent the maximum and minimum values, excluding outliers (defined as data points more than 1.5 IQRs from the box). Hospital discharge PCPC scores strongly influenced 3 month (A) and 12 month (B) VABS-II scores (Pearson correlations: r= −0.81, −0.77, respectively, p<0.0001; no effect of treatment group). C. Illustrates corresponding 3 month and 12 month VABS-II scores; scores were tightly linked in both treatment groups (r=0.95, p<0001).