ABSTRACT
Objective
To evaluate heart rate variability (HRV) as a biomarker of autonomic dysfunction in anorexia nervosa (AN), examining differences across clinical stages and the influence of HRV recording duration.
Method
Quantitative studies comparing HRV parameters between individuals with AN and healthy controls were identified through a systematic search of six databases following PRISMA guidelines. Random-effects meta-analyses and meta-regressions were conducted, with HRV recording duration (5-min vs. 24-h recordings) examined as a moderator. Meta-regression by clinical stage was not feasible due to an insufficient number of studies. Risk of bias was assessed using the Joanna Briggs Institute tool, and certainty of evidence was rated with GRADE.
Results
Nineteen studies were included, with 16 contributing to the meta-analysis (805 participants). Acute AN was associated with significantly lower heart rate (SMD = −1.807; 95% CI: [−3.36, −0.26]; p = 0.022) and reduced LF/HF ratio (SMD = −0.74; 95% CI: [−1.179, −0.301]; p < 0.001). These results indicate parasympathetic predominance. Time-domain indices (SDNN and RMSSD) tended to be higher in AN but were not statistically significant and were supported by low-certainty evidence. Meta-regression analyses revealed that short-term (5-min) HRV recordings yielded significant differences for SDNN (SMD = 0.539; p = 0.008) and HF (SMD = 0.679; p = 0.037).
Conclusions
Overall, HRV reflects autonomic alterations in AN, particularly in the acute phase characterised by parasympathetic predominance. HRV does not consistently normalise during recovery. These findings are strongly moderated by recording duration. Short-term recordings show more consistent effects.