The link between maltreatment history and respiratory sinus arrhythmia in children and adolescents: A meta-analysis

• In Event: The Neurophysiological Embedding of Child Maltreatment: Current State of the Literature and Future Directions for Research

Abstract

Child maltreatment has been found to be associated with negative effects on cognitive, emotional, and physical outcomes over the course of development. One theorized pathway for this connection is via the retuning of brain-body threat response systems that are optimized to respond to danger while increasing risk for psychosocial difficulties and physical symptoms (Kolacz et al., 2019). A common marker of threat response states is the reduction of parasympathetic activity of the autonomic nervous system, which can be non-invasively indexed using respiratory sinus arrhythmia (RSA; high-frequency heart rate variability), the amplitude of beat-to-beat cardiac variation associated with respiratory rhythm (Lewis et al., 2012). Low RSA is associated with poorer performance on executive function measures, anxiety, depression, and somatic symptoms (Thayer et al., 2009; Chalmers et al., 2014; Koenig et al., 2016; Kolacz et al., 2019). Over the past decade, a body of studies have examined whether RSA is altered in children and adolescents who have experienced maltreatment. A recent systematic review found mixed evidence for decreased RSA following maltreatment, with several studies showing null findings (Holochwost et al., 2020). This meta-analysis attempts to reconcile those results to describe the range of effect sizes and study features that give rise to variability in observed effects.

The analysis proposal was pre-registered with Prospero, an international database of systematic reviews. Studies comparing RSA differences between children who were and were not exposed to abuse and/or neglect were compiled. To be considered, studies were required to assess RSA during a resting period using any commonly used quantification methods (e.g., frequency domain, peak-to-trough, Porges-Bohrer method). Comparisons between maltreated and non-maltreated children were converted to standardized mean differences and analysis was conducted using the R “meta" package.

A preliminary search was conducted in June 2020 using PsychInfo and PubMed databases, yielding 190 unique articles. Based on a review of abstracts, 27 of these articles were identified as meeting criteria for full review. Following full review, 9 studies were found to be adequate for analysis, ranging from infant to adolescent samples (total n = 1182; 52% maltreated). Only 2 out of the 9 studies were collected from data outside the U.S. We found a standardized mean difference of .01 [CI: -.01, .04] between maltreated and non-maltreated groups, suggesting there were no consistent overall differences when pooling across studies. Measures of effect heterogeneity supported large variation in effect sizes (Q = 240.62, df = 9, p < .001; tau^2 = .030 [CI: .013; .105], I^2 = 96.3% [CI: 94.6%; 97.4%]; Figure 1).
The preliminary results suggest substantial variability in results that may be due to study methodologies. To examine this question, an additional database search will be conducted in late 2020 and the reference sections of papers will be examined for additional studies. Prior to the SRCD conference, these new articles will be included in analysis and subgroups will be examined to assess the effects that may be due to sample age, maltreatment and RSA measurement methods, and study quality as using guidelines for heart rate variability reporting.

Author

• Jacek Kolacz, Indiana University - Bloomington
  Presenting Author