A high sympathetic and/or a low cardiovagal activity in patients with major depression (MD) may contribute to the higher cardiac morbidity and mortality of MD patients. Standardized tests of heart rate variability (HRV) allow a quantitative estimation of autonomic nervous system function. However, previous studies on the relationship between HRV and MD have revealed conflicting results. Our study compared time and frequency domain HRV indices (5-min resting study, deep breathing test, Valsalva test) between 32 patients with MD (DSM-III-R) and 64 non-depressed controls. The severity of depressive symptoms was assessed by the Hamilton Depression Scale (HAM-D); patients were divided into subgroups with moderate (M-HAM-D<25) or severe depressive symptoms (S-HAM-D>or=25). After controlling for age, gender and smoking, S-HAM-D patients showed a higher heart rate and a significantly lower modulation of cardiovagal activity compared to controls. Although some of the HRV indices of the M-HAM-D group did not differ significantly from controls, they were in the expected direction. There was a significantly negative correlation between the HAM-D scores and the vagal HRV indices, suggesting a direct association between the severity of depressive symptoms and the modulation of cardiovagal activity. Clinical consequences arising from these findings and possible implications for treatment are discussed.