Background:
Dysautonomy has been suggested to be a mechanism contributing to the well-established association between high depressive symptoms (HDS) and cardiovascular disease (CVD). So far however, at the population level, this question has been addressed using parameters of complex interpretation that are rarely used in the clinical setting. The aim of this study was to quantify the respective association of HDS and antidepressants (ATD) on different heart rate parameters of simple investigation.
Methods:
The Paris Prospective Study III included subjects recruited in a large health preventive centre in Paris (France) between 2008 and 2012. Four heart rate parameters were measured including resting heart rate (RHR), heart rate immediately after moderate effort (EHR), differences between EHR and RHR, and RR interval. A total score ≥ 7 on a 13-item standardized questionnaire defined the presence of HDS. Information on classes of ATD was obtained on a face-to-face interview with a medical doctor. The association between HDS or ATD and each of the 4 heart rate parameters were quantified in separate linear regression analysis adjusted for age, sex, body mass index, physical activity, personal history of CVD, smoking, diabetes, high blood pressure and beta blockers. To minimize indication bias, the analysis on ATD was adjusted for the propensity score of receiving ATD.
Results:
The mean age of the 8430 participants was 59.6 years and 39% were women. HDS were noted in 473 subjects and 383 participants were on ATD: 58 on Tricyclics, 232 on Serotonin Specific Reuptake Inhibitor (SSRI) and 80 on Serotonin Norepinephrine Reuptake Inhibitors (NSRI) respectively. Beta-blockers were used by 4.2% of the participants and only 2.1% had a personal history of cardiovascular disease. HDS were associated with slightly higher RHR (+0.9 beats per minute (bpm), p<0.05) and lower RR interval (-21.1 ms, p<0.05). ATD of any class were related to lower RR interval (-22.3 ms, p=0.007), higher EHR (+1.7 bpm, p=0.02) and higher difference between EHR and RHR (+1.5 bpm, p=0.009). While tricyclics and NSRI were both associated with significantly higher RHR (+7.0 and +2.64 bpm respectively, p<0.05), higher EHR (+6.79 and +6.94 bpm respectively, p<0.001), higher difference between EHR and RHR (+2.74, p=0.06 and +5.70, p<0.001 respectively) and lower RR interval (-98.50 ms and -50.63 ms respectively, p<0.001), SSRI was related to lower RHR (-1.81 bpm, p<0.05) and almost significantly higher RR interval (+18.61 ms, p=0.08). Neither sex nor beta-blockers modified these associations. Consistent findings were observed when other propensity score methods were used.
Conclusions:
Our study suggests that ATD more than HDS are associated with dysautonomy as evaluated by simple and routinely used heart rate parameters.
LONG-TERM EFFECT OF ANTIHYPERTENSIVE AGENTS ON THE NEURAL BAROREFLEX SENSITIVITY-A SUBSTUDY OF THE PARIS PROSPECTIVE STUDY III