Objectives The aim was to compare measures of heart rate variability (HRV) in patients who presented with non-cardiac vascular episodes with age- and gender-matched control patients. Methods One hundred and fifty patients, randomly selected from a cohort of 522 subjects, were enrolled in a screening study. Of these, 256 were identified to have had a stroke or transient ischaemic attack (TIA), or to have peripheral vascular disease (PVD) at the first presentation to Ninewells Hospital, Dundee, Scotland. Only 114 patients remained in the study (100 cases and 14 controls). Multiple regression analysis was used to assess the association between HRV parameters and measures of mean heart rate and ejection fraction. Results Heart rate and HRV indices were significantly inversely correlated with both normal left ventricular (LV) function [r = 0.2–0.5; p = 0.037–0.0001] and left ventricular systolic dysfunction (LVSD) [r = 0.3–0.5; p = 0.07–0.01] in the patients. HRV did not predict LVSD in this cohort of patients. Multiple regression analysis showed only ischaemic heart disease (IHD) and cigarette smoking had an independent relation to HRV parameters. Cigarette smoking (p = 0.008), IHD (p = 0.02) and diabetes (p = 0.03) were significant predictors of reduced HRV (standard deviation of the normal-to-normal interval: SDNN), independent of LVSD. Discussion There were no significant differences in HRV indices between non-cardiac vascular patients (TIA, stroke, PVD) and their age- and gender-matched controls. HRV had no diagnostic value as a pre-screening test to identify suspected LVSD in these patients. Conclusion HRV cannot be used as a screening test to identify hidden LVSD. Further studies will be needed to assess the possibilities that HRV is a convenient marker of endothelial dysfunction.