Abstract
Aim:
Out-of-hospital cardiac arrest (OHCA) in pediatric patients is associated with high rates of mortality and neurologic injury, with no definitive evidence-based method to predict outcomes available. A prognostic scoring tool for adults, The Brain Death After Cardiac Arrest (BDCA) score, was recently developed and validated. We aimed to validate this score in pediatric patients.
Methods:
Retrospective cohort study of pediatric patients admitted to 5 PICUs after OHCA between 2011 and 2021. We extracted BDCA score elements for those who survived at least 24 hours but died as a result of their OHCA. We assessed score discrimination for the definitive outcome of brain death. Subgroup analysis was performed for infants < 12mo versus children ≥ 12mo, those who likely had brain death but had withdrawal of life sustaining therapy (WLST) prior to declaration, and by etiology and duration of arrest.
Results:
389 subjects were identified across 5 institutions, with 282 meeting inclusion criteria. 169 (59.9%) were formally declared brain dead; 58 (20.6%) had findings consistent with brain death but had withdrawal of life sustaining therapies prior to completion of formal declaration. Area under the receiver operating characteristic curve for the age ≥ 12mo cohort was 0.82 [95% CI 0.75, 0.90], which mirrored the adult subject AUCs of 0.82 [0.77, 0.86] and 0.81 [0.76, 0.86] in the development and validation cohorts. Scores demonstrated worse discrimination in the infant cohort (AUC = 0.61).
Conclusions:
The BDCA score shows promise in children ≥ 12mo following OHCA and may be considered in conjunction with existing multimodal prognostication approaches.
Keywords: Brain Death, Out-of-hospital cardiac arrest, OHCA, Outcomes, Pediatric Critical Care, Pediatric OHCA, Post-Arrest Prognostication, Post-Resuscitation Care, Prognostication
Introduction
Despite recent focus on post-arrest management, out-of-hospital cardiac arrest (OHCA) in pediatric patients continues to have poor outcomes.1–7 In those who attain return of spontaneous circulation (ROSC) and survive to PICU admission, progression to brain death (BD) and withdrawal of life-sustaining therapies (WLST) are important causes of mortality from OHCA; in recent studies, BD accounted for 29–47% of deaths during admission, and WLST due to expected poor neurological outcome accounted for 34–67%.1,4,8–10
However, prognostication of outcomes following OHCA in pediatric patients remains an obstacle. Although a multimodal approach including serial exams, electroencephalography and imaging is often employed,11–13 there remains no validated, objective tool for early (<48 hours after arrest) prediction of outcomes in this population. Effective early prognostication may guide conversations with families and better inform medical decision-making to establish goals of care.
A prognostic scoring tool for adults (The Brain Death After Cardiac Arrest, or BDCA, score), was recently developed and validated.14 The BDCA score is easy to calculate, only requires 7 pieces of information available early in admission, and demonstrated good discrimination and calibration in adults admitted to an ICU who survived at least 24 hours after OHCA. We aimed to validate this score for use in pediatric patients.
Methods
Retrospective chart review of pediatric patients admitted to 5 PICUs after OHCA between 2011 and 2021 (Fig. 1). Inclusion criteria: age 0–18 years admitted to the PICU following OHCA, survived at least 24 hours post-arrest, and died during admission as a result of their OHCA. Among patients who met criteria (n = 282), the elements of the BDCA score were obtained: gender, initial rhythm, cardiac or neurologic cause of arrest, serum sodium level collected closest to 24 hours after arrest, and requirements for vasoactive infusions at admission and/or at 24 hours post-arrest (Fig. 2). These factors had been found to be independently associated with BD in the adult study.14
Fig. 1 –
Inclusion details and brief description of participating PICUs, 1UFHealth/Shands Children’s Hospital: Gainesville, Florida. 24 bed PICU; was a mixed unit with cardiac patients for 4 of the 10 study years. 2Wolfson Children’s Hospital: Jacksonville, Florida. 20 bed PICU; was a mixed unit with cardiac patients for 8 of the 10 study years. 3Children’s Hospital Los Angeles: Los Angeles, California. 24 bed PICU. 4Ann & Robert H. Lurie Children’s Hospital of Chicago: Chicago, Illinois. 60 bed PICU; was a mixed unit with cardiac patients for 4 of the 10 study years. 5University of Maryland Children’s Hospital: Baltimore, Maryland, 19 bed mixed PICU.
Fig. 2 –
Covariates and score calculation details.
This study was approved by the Institutional Review Board (IRB) of each participating institution. Study data were housed using RED-Cap® electronic data capture tools hosted at The University of Florida. All data management and analyses were conducted using SAS 9.4® (Cary, NC). A predetermined level of α = 0.05 was used to evaluate statistical significance.
Descriptive statistics (mean ± STD, and frequencies/percentages) for demographic characteristics, arrest duration, and score components were computed for patients stratified by outcomes in our overall cohort, in infants (<12mo), and children (≥12mo). Arrest duration and score components were compared between those with and without BD. Differences were determined via Chi-Square and Cochran Mantel Haenszel (CMH) testing methods for categorical data, and specific ordinal data, respectively. t-Tests were used to compare the two groups with respect to continuous data; the Satterthwaite approximation for degrees of freedom was used in the instance of unequal variances. Logistic regression was used to characterize independent associations between arrest duration, specific etiologies of arrest, and individual BDCA score components with the eventual outcome of brain death.
Receiver Operating Characteristic (ROC) analysis was conducted to elucidate the utility of the BDCA score in pediatric patients with the primary outcome considered declaration of BD, as defined by currently accepted national standards and recommended examination methods.15,16 Additional ROC analysis compared the test discrimination ability of the BDCA score among those who did receive a formal declaration regarding BD, in addition to those who had at least one exam consistent with BD but had WLST prior to formal declaration. Subgroup analysis was performed to assess results in infants and children. Results for the overall cohort, infants, and children were plotted utilizing ROC. Subgroup analysis was completed for each of these cohorts to evaluate test discrimination in infants and children, including only those who received a formal declaration. The area under the curve (AUC) was calculated for each group to compare with results for the adult development and validation cohorts.14
Youdin’s J statistics were computed across the range of cutoff values to determine the ideal score cutoff value predictive of BD for the overall cohort, infants, and children. Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) were computed to illustrate the clinical utility of various cutoff levels.
Results
389 pediatric patients were identified (Fig. 1). 282 met inclusion criteria; 101 were excluded due to death within 24 hours of arrest, 1 excluded due to record error (arrest occurred in-hospital), and 5 excluded due to >24 hr stays at outside ICUs prior to transfer to the participating institutions.
Among our total cohort (n = 282), 169 (59.9%) were formally declared brain dead, and 58 (20.6%) had exam(s) consistent with BD but had WLST prior to declaration (Table 1). 90 patients (31.9%) were in the infant cohort, among whom 53 (58.9%) were declared brain dead; 19 (21%) of infants had exam(s) consistent but had WLST before declaration. 192 patients (68.1%) were in the children cohort, with 116 (60.4%) being declared brain dead; a further 39 (20.3%) of this cohort had exam(s) consistent but had WLST prior to declaration. The most common etiologic categories of arrest by age group are presented in Table 1 (expanded in).
Table 1 –
Demographic details including etiology of arrest, and outcomes.
| Overall | Brain Death: No | Brain Death: Yes | Brain Death: Yes, no formal declaration | |
|---|---|---|---|---|
|
| ||||
| (n) Total Sample | 282 | 55 (19.5) | 169 (59.9) | 58 (20.6) |
| Sex: Female | 132 (46.8) | 24 (43.6) | 84 (49.7) | 24 (58.6) |
| Race and Ethnicity: | ||||
| White | 92 (32.6) | 16 (29.1) | 58 (34.3) | 18 (31.0) |
| Black | 68 (24.1) | 10 (18.2) | 48 (28.4) | 10 (17.2) |
| Asian | 9 (3.2) | 2 (3.6) | 5 (3.0) | 2 (3.5) |
| Unknown | 72 (25.5) | 12 (21.8) | 43 (25.4) | 17 (29.3) |
| 1 + Race | 1 (0.4) | – | – | 1 (1.7) |
| Hispanic | 40 (14.2) | 15 (27.3) | 15 (8.9) | 10 (17.2) |
| Age < 12 Months | ||||
| (n) | 90 | 18 (20.0) | 53 (58.9) | 19 (21.1) |
| Sex: Female | 45 (50.0) | 9 (50.0) | 25 (47.2) | 11 (57.9) |
| Race and Ethnicity: | ||||
| White | 28 (31.1) | 2 (11.1) | 20 (37.7) | 6 (31.6) |
| Black | 21 (23.3) | 3 (16.7) | 12 (22.6) | 6 (31.6) |
| Asian | 2 (2.2) | – | 2 (3.8) | – |
| Unknown | 24 (26.7) | 6 (33.3) | 13 (24.5) | 5 (26.3) |
| 1 + Race | – | – | – | – |
| Hispanic | 15 (16.7) | 7 (38.9) | 6 (11.3) | 2 (10.5) |
| Etiology | ||||
| Abusive Head Trauma | 7 (7.8) | 1 (5.6) | 4 (7.6) | 2 (10.5) |
| Aspiration | 6 (6.7) | 2 (11.1) | 4 (7.6) | – |
| Co-Sleeping | 10 (11.1) | 3 (16.7) | 5 (9.4) | 2 (10.5) |
| Other Asphyxia | 8 (8.9) | 1 (5.6) | 5 (9.4) | 2 (10.5) |
| SIDS | 31 (34.4) | 4 (22.2) | 20 (37.7) | 7 (36.8) |
| TBI, not abusive HT | 5 (5.6) | – | 5 (9.4) | – |
| Other | 23 (25.6) | 7 (38.9) | 10 (18.9) | 6 (31.6) |
| Age ≥ 12 Months | ||||
| (n) | 192 | 37 (19.3) | 116 (60.4) | 39 (20.3) |
| Sex: Female | 105 (54.7) | 15 (40.5) | 59 (50.9) | 13 (33.3) |
| Race and Ethnicity: | ||||
| White | 64 (33.3) | 14 (37.8) | 38 (32.8) | 12 (30.8) |
| Black | 47 (24.5) | 7 (18.9) | 36 (31.0) | 4 (10.3) |
| Asian | 7 (3.7) | 2 (5.4) | 3 (2.6) | 2 (5.1) |
| Unknown | 48 (25.0) | 6 (16.2) | 30 (25.9) | 12 (30.8) |
| 1 + Race | 1 (0.5) | – | – | 1 (2.6) |
| Hispanic | 25 (13.0) | 8 (21.6) | 9 (7.8) | 8 (20.5) |
| Etiology | ||||
| Asphyxia (hanging or choking) | 17 (8.9) | 4 (10.8) | 12 (10.3) | 1 (2.6) |
| Asthma | 8 (4.2) | – | 7 (6.0) | 1 (2.6) |
| Drowning | 53 (27.6) | 13 (35.1) | 28 (24.1) | 12 (30.8) |
| Respiratory Arrest | 11 (5.7) | 1 (2.7) | 4 (3.5) | 6 (15.4) |
| TBI, not abusive HT | 16 (8.3) | 2 (5.4) | 13 (11.2) | 1 (2.6) |
| Unknown | 16 (8.3) | 5 (13.5) | 10 (8.6) | 1 (2.6) |
| Other | 71 (37.0) | 12 (32.4) | 42 (36.2) | 17 (43.6) |
Data displayed as n (%); column percentages are reported overall and within groups.
CHISQ used for binary outcome Sex, CMH General Association for multi-level categorical variables.
Among children, neurologic cause of arrest, natremia closest to 24 hours post-arrest, vasoactive medications at admission, vasoactive medications at 24 hours, and overall BDCA were all significantly different (p < 0.05) between those with and without BD; these components remained significant even when those who had exam(s) consistent with BD but were not formally declared were included with the formally declared group (Table 2). Among infants, only natremia at 24 hours was found to be significantly different between those with (whether or not formally declared) and those without BD. Duration of arrest, in those patients for whom the information was available, was found to be statistically significant and predictive of BD if lasting > 30 minutes for all children (n = 176), as well as for children who received a formal declaration (n = 138;). There was a significant difference in the number of infants who progressed to BD when compared by duration of arrest (Table 2). However, after adjusting for significantly different risk factors, arrest duration was not independently predictive of BD for infants, for whom the information was available (n = 81;).
Table 2 –
Duration of Arrest & BDCA Score Components Stratified by Age Groups.
| Brain Death: No | Brain Death: Yes or Yes NF1 | p-value | |
|---|---|---|---|
|
| |||
| Age < 12 Months (n) | 18 | 72 | |
|
| |||
| Duration of Arrest: | CMH: 0.0356 | ||
| <15 minutes | 2 (11.1) | 9 (12.5) | |
| 15–30 minutes | 8 (44.4) | 11 (15.3) | |
| >30 minutes | 8 (44.4) | 43 (59.7) | |
| Unknown | – | 9 (12.5) | |
| Sex: Female [+4 points] | 9 (50.0) | 36 (50.0) | CHISQ: 1.000 |
| Non-shockable Rhythm [+24 points] | 17 (94.4) | 68 (94.4) | CHISQ: 1.000 |
| Cardiac Cause of Arrest [−6 points] | – | – | – |
| Neurological Cause of Arrest [+45 points] | 2 (11.1) | 13 (18.1) | CHISQ: 0.4795 |
| Sodium Level closest to 24 hours after Arrest [calculated as +(natremia − 140) points] | 145.0 ± 7.7 | 149.7 ± 8.0 | t-test: 0.0261 |
| Vasoactive Medication Required at time of Admission [+4 points] | 9 (50.0) | 45 (62.5) | CHISQ: 0.3329 |
| Vasoactive Medication Required 24 hours after Admission [+ 6 points] | 12 (66.7) | 47 (65.3) | CHISQ: 0.9117 |
| BDCA Score [Calculated: sum of above] | 42.7 ± 21.6 | 49.7 ± 20.4 | t-test: 0.1995 |
| Brain Death: No | Brain Death: Yes or Yes NF 1 | p-value | |
|
| |||
| Age ≥ 12 Months (n) | 37 | 155 | |
|
| |||
| Duration of Arrest: | CMH: 0.0053 | ||
| <15 minutes | 11 (29.7) | 16 (10.3) | |
| 15–30 minutes | 11 (29.7) | 34 (21.9) | |
| >30 minutes | 14 (37.8) | 90 (58.1) | |
| Unknown | 1 (2.7) | 15 (9.7) | |
| Sex: Female [+4 points] | 15 (40.5) | 72 (46.5) | CHISQ: 0.5164 |
| Non-shockable Rhythm [+24 points] | 36 (97.3) | 150 (96.8) | CHISQ: 0.8695 |
| Cardiac Cause of Arrest [−6 points] | 3 (8.1) | 5 (3.2) | CHISQ: 0.1818 |
| Neurological Cause of Arrest [+45 points] | 1 (2.7) | 30 (19.4) | CHISQ: 0.0134 |
| Sodium Level closest to 24 hours after Arrest [calculated as +(natremia − 140) points] | 145.4 ± 5.5 | 151.0 ± 9.2 | t-test 3 : <0.0001 |
| Vasoactive Medication Required at time of Admission [+4 points] | 10 (27.0) | 112 (72.3) | CHISQ: <0.0001 |
| Vasoactive Medication Required 24 hours after Admission [+ 6 points] | 17 (46.0) | 130 (83.9) | CHISQ: <0.0001 |
| BDCA Score [Calculated: sum of above] | 35.2 ± 13.8 | 53.3 ± 21.7 | t-test 3 : <0.0001 |
| Brain Death: No | Brain Death: Yes, FD 2 | p-value | |
|
| |||
| Age ≥ 12 Months (n) | 37 | 116 | |
|
| |||
| Duration of Arrest: | CMH: 0.0116 | ||
| <15 minutes | 11 (29.7) | 13 (11.2) | |
| 15–30 minutes | 11 (29.7) | 25 (21.6) | |
| >30 minutes | 14 (37.8) | 64 (55.2) | |
| Unknown | 1 (2.7) | 14 (12.1) | |
| Sex: Female [+4 points] | 15 (40.5) | 59 (50.9) | CHISQ: 0.2740 |
| Non-shockable Rhythm [+24 points] | 36 (97.3) | 112 (96.6) | CHISQ: 0.8242 |
| Cardiac Cause of Arrest [−6 points] | 3 (8.1) | 4 (3.5) | CHISQ: 0.2375 |
| Neurological Cause of Arrest [+45 points] | 1 (2.7) | 24 (20.7) | CHISQ: 0.0100 |
| Sodium Level closest to 24 hours after Arrest [calculated as +(natremia − 140) points] | 145.4 ± 5.5 | 151.0 ± 8.7 | t-test 3 : <0.0001 |
| Vasoactive Medication Required at time of Admission [+4 points] | 10 (27.0) | 85 (73.3) | CHISQ: <0.0001 |
| Vasoactive Medication Required 24 hours after Admission [+6 points] | 17 (46.0) | 99 (85.3) | CHISQ: <0.0001 |
| BDCA Score [Calculated: sum of above] | 35.2 ± 13.8 | 54.1 ± 22.3 | t-test 3 : <0.0001 |
Categorical Data: n (%); column percentages reported within groups. Continuous Data: Mean ± STD within groups.
NF – Not Formally Declared Patients with Brain Death.
FD – Formally Declared Brain Death Only; CMH – Cochran-Mantel-Haenszel test for ordinal outcomes; CHISQ: Chi-Square test.
Satterthwaite for unequal variances.
Among those with formal declarations, there was no significant difference in rate of progression to BD between age groups: 53/71 (74.7%) infants, and 116/169 (75.8%) children (p-value: 0.85). Results remained consistent when including those who had exam (s) consistent with BD but underwent WLST prior to formal declaration (p-value: 0.89). There was no significant difference in the prevalence of pre-declaration WLST between infants (19/90, 21.1%) and children (39/192, 20.3%), p-value: 0.88.
Area under the receiver operating characteristic (AUROC) curve for our total cohort of patients who received a formal declaration regarding BD (n = 224) was 0.76 [95% CI 0.69, 0.84] (Fig. 3). Notably, the AUROC was similar when this analysis also included the 58 patients who has exam(s) consistent with BD but had WLST prior to declaration (total n = 282; AUC = 0.75 [0.68, 0.83]). In subgroup analysis, the AUROC improved to 0.82 [0.75, 0.90] in the children who received a definitive declaration regarding BD or not (n = 153); again, results were similar when this analysis included the 39 children who had consistent exam(s) but had WLST prior to definitive declaration (total n = 192; AUC = 0.82 [0.74, 0.89]). These findings were noted to mirror the adult subject AUC of 0.82 [0.77, 0.86] in the development cohort and 0.81[0.76, 0.86] in the validation cohort. Score discrimination was poor among infants who received a definitive declaration of BD or not (AUROC = 0.64 [0.48, 0.79]; n = 71), as well as in the total infant cohort, which included those who had WLST before declaration (AUROC = 0.61 [0.46, 0.75]; n = 90). Inclusion of infants who had WLST before declaration did not have a substantial impact on the AUROC.
Fig. 3 –
ROC curves for all participants, BDCA score and brain death.
Youdin’s J statistic (Table 3) for children indicated a cutoff score of 38 as having optimal sensitivity and specificity, compared to the recommended adult cutoff score of 50. This cutoff value also had superior PPV compared to the adult recommendation. For infants, a score of 45 was found to be most specific, though sensitivity and NPV remained low.
Table 3 –
Pbdca score cutoffs: compared to established adult cutoffs & ideally derived cutoffs.
| Established Score Cutoffs: | Overall | Brain Death | No Brain Death | Sensitivity | Specificity | PPV | NPV |
|---|---|---|---|---|---|---|---|
|
| |||||||
| All (n) | 282 | 227 | 55 | ||||
| >35 points | 212(75.2) | 187(82.4) | 25(45.5) | 82.4%(77.4, 87.3) | 54.6%(41.4, 67.7) | 88.2%(83.9, 92.6) | 42.9%(31.3, 54.5) |
| >50 points | 88(31.2) | 83(36.6) | 5(9.1) | 36.6%(30.3, 42.8) | 90.9%(83.3, 98.5) | 94.3%(89.5, 99.2) | 25.8%(19.6, 31.9) |
| Age < 12 m (n) | 90 | 72 | 18 | ||||
| >35 points | 68(75.6) | 56(77.8) | 12(66.7) | 77.8%(68.2, 87.4) | 33.3%(11.6, 55.1) | 82.4%(73.3, 91.4) | 27.3%(8.7, 45.9) |
| >50 points | 27(30.0) | 24(33.3) | 3(16.7) | 33.3%(22.4, 44.2) | 83.3%(66.1, 100.0) | 88.9%(77.0, 100.0) | 23.8%(13.3, 34.3) |
| Age ≥ 12 m (n) | 192 | 155 | 37 | ||||
| >35 points | 144(75.0) | 131(84.5) | 13(35.1) | 84.5%(78.8, 90.2) | 64.9%(49.5, 80.3) | 91.0%(86.3, 95.7) | 50.0%(35.9, 64.2) |
| >50 points | 61(31.8) | 59(38.1) | 2(5.4) | 38.1%(30.4, 45.7) | 94.6%(87.3, 100.0) | 96.7%(92.3, 100.0) | 26.7%(19.1, 34.3) |
| Using Youdin’s J Cutoffs | |||||||
| Age <12 m | 90 | 72 | 18 | ||||
| >45 points | 38(42.2) | 35(48.6) | 3(16.7) | 48.6%(37.1, 60.2) | 83.3%(66.1, 100.0) | 92.1%(83.5, 100.0) | 28.9%(16.5, 41.2) |
| Age ≥ 12 m | 192 | 155 | 37 | ||||
| >38 points | 132(68.8) | 123(79.4) | 9(24.3) | 79.4%(73.0, 85.7) | 75.7%(61.9, 89.5) | 93.2%(88.9, 97.5) | 46.7%(34.0, 59.3) |
Data displayed as: n (%); column percentages reported overall and within groups.
Sensitivity - Specificity - PPV - NPV: values displayed with associated 95% CI in parentheses.
Discussion
Despite recent focus on and advances in post-arrest management, out-of-hospital cardiac arrest (OHCA) is estimated to be experienced by over 23,000 pediatric patients annually in the United States and continues to be associated with poor outcomes.17,18 The BCDA score demonstrated potential in adults for unifying multiple predictors of eventual BD following OHCA into one easy-to-calculate, objective tool available early in a patient’s course following survival to ICU admission.14 Our findings demonstrate similar potential in children (≥12mo), with an AUC nearly identical to the adult development and validation cohorts. The overall BDCA score and many individual components (neurological cause of arrest, natremia closest to 24 hours post-arrest, vasoactive medications at admission and 24 hours after) were found to have statistically significant differences between children with and without BD in this age group. Non-shockable rhythms were not statistically significant, despite being the strongest predictor of BD in adults and having been associated with poor outcomes in pediatrics.19–22 This is likely due to the differing etiologies of OHCA commonly noted in adults vs paediatrics,23 as well as our general PICU-based study population capturing a small number of shockable rhythms.
The BDCA score was less predictive of BD in infants (<12mo); further, only natremia was found to be individually significant in predicting BD in this cohort. These findings are likely multifactorial. First, the etiologies of arrest in this population are vastly different than those in children and adults. Second, it is likely that the score components are less individually associated with progression to BD in infants than in children or adults, or may have a different degree of association (i.e., may need a different scoring weight in future models). Finally, as the incidence of progression to BD was similar in infants and children in our study, our total number of infants (n = 90) may represent an inadequate sample size to assess the reliability of the overall score in infants. Prior studies have suggested that infants have worse prognostic outcomes and higher mortality rates following OHCA relative to children, and our findings further highlight that more study of outcome prediction in this population is needed.19,20,23,24
Early WLST based on expected poor neurological outcome is a prominent cause of death in OHCA patients in the PICU. Many recent studies focused on describing mortality rates and timing after pediatric OHCA note concerns that WLST may be being performed too early in some patients due to the lack of objective, evidence-based prognostic tools and guidelines.9,10,25.26 In this context, it is significant that the score performed equally in subjects of any age formally declared brain dead vs the 58 children (20.6% of our entire cohort) that were likely brain dead but had WLST without formal declaration. Along with other recent studies that have demonstrated the utility of imaging and other ancillary studies in predicting outcomes following OHCA,13 our findings suggest that the BDCA score (and/or several of its individual components) can offer early, objective information that may impact conversations and decisions regarding WLST. Collectively, this lays the groundwork for the development of universal, validated, multimodal prognostic tools in future studies.
What has not been well described to-date are any age-related differences in rates of progression to BD that may exist amongst the pediatric populations. In this 10-year sample from five U.S. children’s hospitals, the age-related rates of progression did not differ, opening the door for infants to be included in the development of potential prognostication tools or guidelines.
Overall, the potential impact of this score lies in both the clinical and research realms. The potential clinical utility is as an objective tool in conjunction with the current subjective multimodal prognostication approach to inform conversations about goals of care. One potential research utility lies in potentially identifying subjects enrolled in clinical trials aimed at improving outcomes post-OHCA who have likelihood of progression to BD, weighing their impact on results.
This study has limitations. Study inclusion relied on data extracted from medical records. It is possible this data is incomplete; this was mitigated by manual review of all deaths in these PICUs over the 10-year time frame for inclusion when possible. While the sample size was adequate to detect the needed effects in children, it may have been too small for infants. Gestational age details were not collected. The major limitation of this study was inclusion of decedents only, rather than also including survivors. This was done by design, as the potential clinical application of this score is at the critical time point of considering early WLST. Finally, as a PICU-based study, there was limited capture of subjects with underlying cardiac conditions and shockable rhythms; future work could focus on this cohort.
Conclusions
Objective, evidence based prognostic tools and guidelines are needed for children post-OHCA. The BDCA score shows promise in children, performing similarly to adult subjects. The score discrimination performed worse in infants and may require different predictor variables to attain acceptable accuracy in this age group. Addition of the BDCA score to the current multimodal approach of prognosticating outcomes may inform decision-making regarding goals of care in patients who will likely meet criteria for brain death.
Supplementary Material
Acknowledgements
Research reported in this publication was supported by the University of Florida Clinical and Translational Science Institute, which is supported in part by the NIH National Center for Advancing Translational Sciences under award number UL1TR000064. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. We thank the PALISI scientific review committee for their critical appraisal and feedback of our manuscript.
Footnotes
CRediT authorship contribution statement
Matthew Henry: Writing – original draft, Writing – review & editing, Investigation. Stephanie L. Filipp: Formal analysis, Data curation, Resources, Writing – original draft, Writing – review & editing. Elber Yuksel Aydin: Investigation, Writing – review & editing. Nicolas Chiriboga: Investigation, Writing – review & editing. Kailea Zelinka: Investigation, Writing – review & editing. Lorena Espinosa Smith: Investigation, Writing – review & editing. Matthew J. Gurka: Formal analysis, Data curation, Resources, Writing – original draft, Writing – review & editing. Jose Irazuzta: Investigation, Writing – review & editing. Yudy Fonseca: Investigation, Writing – review & editing. Meredith C. Winter: Writing – review & editing. Charlene Pringle: Conceptualization, Project administration, Methodology, Writing – original draft, Writing – review & editing, Supervision, Investigation, Funding acquisition.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Appendix A. Supplementary material
Supplementary material to this article can be found online at https://doi.org/10.1016/j.resuscitation.2023.109727.
REFERENCES
- 1.Lee J, Yang WC, Lee EP, et al. Clinical survey and predictors of outcomes of pediatric out-of-hospital cardiac arrest admitted to the emergency department. Sci Rep 2019;9:7032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Tham LP, Wah W, Phillips R, et al. Epidemiology and outcome of paediatric out-of-hospital cardiac arrests: A paediatric sub-study of the Pan-Asian resuscitation outcomes study (PAROS). Resuscitation 2018;125:111–7. [DOI] [PubMed] [Google Scholar]
- 3.Meert KL, Telford R, Holubkov R, et al. Pediatric out-of-hospital cardiac arrest characteristics and their association with survival and neurobehavioral outcome. Pediatr Crit Care Med 2016;17:e543–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Matsui S, Sobue T, Irisawa T, et al. Poor long-term survival of out-of-hospital cardiac arrest in children. Int Heart J 2020;61:254–62. [DOI] [PubMed] [Google Scholar]
- 5.Fink EL, Prince DK, Kaltman JR, et al. Unchanged pediatric out-of-hospital cardiac arrest incidence and survival rates with regional variation in North America. Resuscitation 2016;107:121–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Gässler H, Fischer M, Wnent J, Seewald S, Helm M. Outcome after pre-hospital cardiac arrest in accordance with underlying cause. Resuscitation 2019;138:36–41. [DOI] [PubMed] [Google Scholar]
- 7.Slomine BS, Silverstein FS, Christensen JR, et al. Neurobehavioral outcomes in children after out-of-hospital cardiac arrest. Pediatrics 2016;137:e20153412. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hansen ML, Lin A, Eriksson C, et al. A comparison of pediatric airway management techniques during out-of-hospital cardiac arrest using the CARES database. Resuscitation 2017;120:51–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hunfeld M, Nadkarni VM, Topjian A, et al. Timing and cause of death in children following return of circulation after out-of-hospital cardiac arrest: A single-center retrospective cohort study. Pediatr Crit Care Med 2021;22:101–13. [DOI] [PubMed] [Google Scholar]
- 10.Du Pont-Thibodeau G, Fry M, Kirschen M, et al. Timing and modes of death after pediatric out-of-hospital cardiac arrest resuscitation. Resuscitation 2018;133:160–6. [DOI] [PubMed] [Google Scholar]
- 11.Dangayach NS, Wright BJ. Taking a SLANT on Postcardiac Arrest Prognostication. Crit Care Med 2022;50:519–22. [DOI] [PubMed] [Google Scholar]
- 12.Kirschen MP, Licht DJ, Faerber J, et al. Association of MRI brain injury with outcome after pediatric out-of-hospital cardiac arrest. Neurology 2021;96:e719–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Smith AE, Ganninger AP, Mian AY, Friess SH, Guerriero RM, Guilliams KP. Magnetic resonance imaging adds prognostic value to EEG after pediatric cardiac arrest. Resuscitation 2022;173:91–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Madelaine T, Cour M, Roy P, et al. Prediction of brain death after out-of-hospital cardiac arrest: Development and validation of the brain death after cardiac arrest score. Chest 2021;160:139–47. [DOI] [PubMed] [Google Scholar]
- 15.Nakagawa TA, Ashwal S, Mathur M, Mysore M, Committee for Determination of Brain Death in Infants Children. Guidelines for the determination of brain death in infants and children: an update of the 1987 task force recommendations-executive summary. Ann Neurol 2012;71:573–85. [DOI] [PubMed] [Google Scholar]
- 16.Martin SD, Porter MB. Performing the brain death examination and the declaration of pediatric brain death. J Pediatr Intensive Care 2017;6:229–33. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Virani SS, Alonso A, Benjamin EJ, et al. American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statistics-2020 Update: A Report from the American Heart Association. Circulation 2020;141:e139–596. [DOI] [PubMed] [Google Scholar]
- 18.Okubo M, Chan HK, Callaway CW, Mann NC, Wang HE. Characteristics of paediatric out-of-hospital cardiac arrest in the United States. Resuscitation 2020;153:227–33. [DOI] [PubMed] [Google Scholar]
- 19.Irvine R, Doan T, Bosley E, Colbeck M, Bowles KA. Pediatric out-of-hospital cardiac arrests: An epidemiological study. Prehosp Emerg Care 2022;22:1–10. [DOI] [PubMed] [Google Scholar]
- 20.Naim MY, Griffis HM, Berg RA, et al. Compression-only versus rescue-breathing cardiopulmonary resuscitation after pediatric out-of-hospital cardiac arrest. J Am Coll Cardiol. 2021;78:1042–52. [DOI] [PubMed] [Google Scholar]
- 21.Nehme Z, Namachivayam S, Forrest A, Butt W, Bernard S, Smith K. Trends in the incidence and outcome of paediatric out-of-hospital cardiac arrest: A 17-year observational study. Resuscitation 2018;128:43–50. [DOI] [PubMed] [Google Scholar]
- 22.Hara M, Hayashi K, Kitamura T. Outcomes differ by first documented rhythm after witnessed out-of-hospital cardiac arrest in children: an observational study with prospective nationwide population-based cohort database in Japan. Eur Heart J Qual Care Clin Outcomes 2017;3:83–92. [DOI] [PubMed] [Google Scholar]
- 23.Chen CY, Lin YR, Zhao LL, et al. Epidemiology and outcome analysis of children with traumatic out-of-hospital cardiac arrest compared to nontraumatic cardiac arrest. Pediatr Surg Int 2013;29:471–7. [DOI] [PubMed] [Google Scholar]
- 24.Jayaram N, McNally B, Tang F, Chan PS. Survival after out-of-hospital cardiac arrest in children. J Am Heart Assoc 2015;4: e002122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Elmer J, Torres C, Aufderheide TP, et al. Resuscitation Outcomes Consortium. Association of early withdrawal of life-sustaining therapy for perceived neurological prognosis with mortality after cardiac arrest. Resuscitation 2016;102:127–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Alqudah Z, Nehme Z, Alrawashdeh A, Williams B, Oteir A, Smith K. Paediatric traumatic out-of-hospital cardiac arrest: A systematic review and meta-analysis. Resuscitation 2020;149:65–73. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.