The PROGNOSE study (Pediatric Resuscitation Neuroprognostication and Outcomes Registry) is a prospective, multicenter, observational patient registry established to improve understanding of short- and long-term outcomes following pediatric cardiac arrest (CA) in the Netherlands. The study focuses on both out-of-hospital (OHCA) and in-hospital cardiac arrest (IHCA) in children under 18 years of age.
Given the rarity yet severity of pediatric CA, the registry aims to collect a large dataset to identify patterns in care, outcomes, and recovery trajectories to ultimately improve treatment strategies and prognosis. The study is considered non-WMO (not subject to the Dutch Medical Research Involving Human Subjects Act), as it involves no additional interventions beyond standard clinical care.
Registry Objectives and Scope The primary objective is to evaluate diagnostic practices and long-term functional and neuropsychological outcomes in pediatric patients post-cardiac arrest. Key secondary objectives include determining the incidence and etiology of pediatric CA in the Netherlands, analyzing survival to discharge, and characterizing post-Return of Circulation (ROC) care.
Data will be collected longitudinally during standard outpatient follow-up visits at specified intervals (3, 12, and 24 months post-arrest; and at the ages of 5, 8, 12, and 17 years, depending on the age at CA event). No study-specific interventions will be performed.
Data Collection and Source Verification
All data collected in this registry will be sourced from routine clinical care and medical records from seven Dutch academic (pediatric) hospitals. Data will be abstracted from:
Emergency services records
Inpatient hospital and ICU documentation
Follow-up outpatient clinic assessments
Neuropsychological and functional evaluations using validated tools (e.g., PCPC, POPC, FSS)
The data will be pseudo-anonymized using unique study IDs. Only site-specific investigators will have access to the link between study ID and patient identity.
Quality Assurance and Monitoring
A robust quality assurance framework has been put in place to ensure data integrity, including:
Data validation procedures: Data entries will be checked against predefined rules for logic, consistency, and range validation in the electronic data capture (EDC) system (Castor EDC).
Source data verification (SDV): Periodic audits by local site PIs will compare registry entries against source data (e.g., EHRs) for completeness and accuracy.
Site monitoring: The coordinating center (Erasmus MC) will provide oversight and perform routine cross-site reviews to ensure harmonized data collection.
Data dictionary: A detailed data dictionary defines all collected variables, including their source, coding schemes (e.g., ICD, MedDRA where applicable), and interpretation ranges.
Standard Operating Procedures (SOPs)
Standard operating procedures have been established to guide the entire registry lifecycle, including:
Site initiation and training
Patient screening and enrollment processes
Data entry and validation procedures
Procedures for obtaining (delayed) informed consent at follow-up
Change management and version control of registry forms
Procedures for data sharing and publication
Statistical Analysis Plan Given its quality improvement nature, no formal sample size calculation was performed. Instead, the registry aims to enroll all eligible pediatric CA cases nationwide, allowing the creation of a comprehensive, hypothesis-generating dataset.
Descriptive statistics will summarize demographics, event characteristics, interventions, and outcomes.
Longitudinal analysis of neuropsychological function and functional status will be performed using repeated-measures ANOVA or mixed-effects modeling.
Comparative statistics (e.g., logistic regression) will be used to identify predictors of survival and favorable neurological outcome. Covariates will be selected based on literature and tested for collinearity. Multivariate models will report odds ratios with 95% confidence intervals.
Missing data will be addressed using appropriate imputation techniques, depending on the mechanism of missingness (e.g., multiple imputation for random missing data).
Plan for Missing Data
To manage incomplete records and potential attrition in follow-up:
All missing values will be coded according to reason (e.g., not applicable, patient deceased, declined follow-up).
Patterns of missingness will be explored. If data are missing at random, multiple imputation may be used in analysis.
Data collection tools prompt required fields and flag missing values to reduce omissions during entry.
Informed Consent and Ethics Due to the emergent nature of CA and the high mortality rate, an exception from consent procedure is employed for initial data collection. For survivors, informed consent is sought at follow-up outpatient visits, as part of standard care. Data are de-identified and used solely for research purposes. No images or human materials are collected.
The study complies with the General Data Protection Regulation (GDPR), the Dutch "Code Goed Gedrag," and the Declaration of Helsinki.
Data Storage and Access Pseudo-anonymized data are stored securely in Castor EDC, managed by Erasmus MC.
Each center has access only to its own data.
A Data Transfer Agreement (DTA) governs anonymized data sharing for multicenter publications.
Data will not be shared outside the EU.
No end date is set for the registry; it will remain open for ongoing data entry and hypothesis generation.
Dissemination Plan Findings from the registry will be disseminated through peer-reviewed publications, with a goal of generating multicenter analyses and contributing to national and international guideline development. A first multicenter manuscript is expected within approximately five years from the start of the registry.
