Is noxious stimulus-evoked electroencephalography response a reliable, valid, and interpretable outcome measure to assess analgesic efficacy in neonates? A systematic review and individual participant data (IPD) meta-analysis protocol

Background: There are several major challenges limiting our ability to test analgesic efficacy for treatment of neonatal pain, and progress in analgesic drug studies in neonates has stalled. One significant issue is the reliance of clinical pain assessments on traditional behavioural and vital signs-based measures and the exclusion of novel brain-based biomarkers. In this review protocol, we outline our strategy to assess the reliability, validity, and interpretability of an electroencephalography (EEG)-based response biomarker for assessment of acute somatic nociceptive pain in neonates. Methods: To standardise EEG analysis and generate the outcome of interest, we will perform an individual participant data (IPD) meta-analysis using data from neonates aged 34-44 weeks postmenstrual age that have had EEG recorded during acute somatic nociceptive skin-breaking procedures. Relevant data from both published and grey literature will be identified by searching six databases (MEDLINE, Embase, CINAHL, Web of Science, Scopus, Google Scholar), two clinical trial registry platforms (ClinicalTrials.gov, WHO ICTRP), and by consulting expert opinion. We will assess availability bias, data accuracy, and data quality by cross-referencing provided data with data descriptions in the literature, identifying duplicates and non-sensical values, and extracting quality control metrics. Data will be synthesised via a two-stage IPD meta-analysis using a random effects modelling approach grouped by site. Reliability (inter- and intra-rater) outcomes will be measured as Gwet's AC1 coefficient. Validity (known-groups and known-stimuli) outcomes will be measured as EEG response magnitude differences between clinically meaningfully different stimuli. Interpretability will be addressed by providing normative values, in both original and standardised units. Discussion: The purpose of this study is to establish the reliability, validity, and interpretability of a specific EEG-based response biomarker for assessing acute somatic nociceptive pain in neonates. It will provide an overview of available data and how EEG is being used globally to assess acute neonatal pain. If sufficient IPD are made available and the outcome is reliable, valid, and interpretable, this work will support the use of EEG-based outcome measures as primary endpoints in clinical trials assessing analgesic efficacy in neonates. Systematic review registration: The protocol was registered with PROSPERO on 14-July-2023: CRD42023444809.