Background: In geriatric age, cognitive impairment and cardiovascular disorders are frequent comorbidities. Age-related anatomical and functional cardiac changes, including the autonomic system, could interfere with the control of different cognitive domains. Therefore, we assess the relationship between long-term heart rate variability (HRV), as a measure of autonomic nervous system (ANS) functioning, and cognitive performance in elderly patients representative of outpatients in a real-life setting. Methods: Of 155 elderly outpatients (aged >65 years) screened, 117 enrolled patients underwent anthropometric evaluation, cardiac assessment by 12-lead electrocardiogram, 24-h ECG recording, and blood pressure (BP) measurement, as well as global cognitive evaluation by a standardized multidimensional assessment, including the Mini-Mental State Examination (MMSE) and Montreal Cognitive Assessment test (MoCA). HRV analysis was performed on 24-h ECG recordings focusing on time-domain indices [Standard deviation of the NN intervals (SDNN), standard deviations of 5-min mean values of the NN intervals for each 5-min interval (SDANN), and root mean squares of successive differences of the NN intervals (RMSSD)] and on frequency-domain measurements [heart rate (HR), low frequency (LF), high frequency (HF), and LF/HF]. Multivariate linear analysis was used to explore the influence of the HRV significant variables on MMSE and MoCA test values. Results: The MMSE and MoCA scores were both significantly and positively correlated with the sympathetic system parameters (SDNN, SDANN, LF, and LF/HF ratio), but not with the parasympathetic system parameters (RMSSD and HF). Multivariate analysis confirms this relationship. Conclusions: Our results show that, in a representative real-life community elderly population, an increased sympathetic activity, but not decreased vagal activity, is