98. How Low Can You Go? The Significance of Bradycardia For Acute Clinical Outcomes In Hospitalised Adolescents With Anorexia Nervosa

Abstract

Bradycardia is one of the most prevalent physical symptoms in adolescents with anorexia nervosa (AN) and has been used as an indicator of physical severity that signals the need for hospital admission. Different clinical guidelines use different heart rate (HR) cut-points to guide admission, ranging from <50 to <40 beats per minute (bpm). These cut-points have been derived from expert consensus rather than evidence. This study aimed to understand the relationship between the severity of bradycardia and acute clinical course in adolescents with AN, with the broader goal of informing clinical practice. A retrospective cohort study was undertaken of adolescents with restrictive eating disorders admitted to a tertiary hospital specialist eating disorder program inpatient ward (May 2014 - Dec 2017). Adolescents with HR <50 bpm at admission were identified and categorised into 3 groups according to severity (mild, 45-49 bpm; moderate, 40-44 bpm; severe, <40 bpm). Clinical outcomes (eg metabolic, cardiovascular, length of stay) were analysed by severity group. Of the 393 admissions during the study period, 106 (27%) had HR <50 bpm at admission (mean [SD] age 15.3 [1.8] years, 83% female). Of these, the number of admissions with mild, moderate and severe bradycardia were 54, 42 and 10. Lower admission HR was associated with significantly longer duration of other measures of HR instability during admission. For example, the mean [SD] number of days with daytime HR<50 for those with mild, moderate and severe bradycardia was 4.22 [4.04], 5.38 [4.70] and 10.30 [5.07], respectively (p<0.001). More severe bradycardia was associated with longer hospital admission. For example, the mean length of stay for those with mild, moderate and moderate bradycardia was 14.4, 15.7 and 17.3 days. Admission HR can be used as an indicator of the severity and duration of HR instability during the course of admission. Applying different severity cut-points did not reveal significant differences in the frequency of other features of medical instability during admission, suggesting admission for bradycardia less than 50 bpm is appropriate. In this service, were the HR cut point for admission to be lowered, the number of admissions could be greatly reduced, but with a subsequent increased length of admission.