Background: Although previous studies demonstrated an association between depressive symptoms and cardiac mortality after acute myocardial infarction (AMI) little is known about the possible mechanisms of this association. The aim of this study was to determine whether depressed patients present a cardiac autonomic dysfunction and whether this could represent the mediator of the influence of depression on their prognosis.
Methods: One hundred consecutive patients with AMI were recruited between January and December 1999. Major Depressive Disorder (MDD) was diagnosed by structured clinical interview and the presence of symptoms of depression was evaluated with self-administered Beck Depression Inventory (BDI). The influence of depression on autonomic nervous system was investigated measuring heart rate variability (HRV) and heart rate (HR) during 24-hour electrocardiographic monitoring. The end-points of the study were all-cause mortality, recurrent-AMI, revascularization and a composite end-point of all the previous. Potential confounders for depression status and events were entered into a multivariate regression model.
Results: Fifteen patients met the criteria for MDD and 35 patients showed mild-to-moderate symptoms of depression; women had a higher prevalence of depression than men (35% vs 9%; p<0.01). Depression was not related to the severity of ischaemic disease or to other clinical and demographic variables. Patients with MDD showed lower HRV (76+/-25 SD vs 99+/-33 SD ms; p<0.01) and higher HR (77+/-12 SD vs 68+/-9 SD bpm; p<0.01) than patients without MDD; moreover mild to moderate symptoms of depression (BDI score > or = 10) were associated with lower HRV (84+/-25 SD vs 102+/-35 SD ms; p=0.01) but not with significantly higher HR. After a mean follow-up of 60 months MDD was associated with an increase of all-cause mortality (OR 12; 95% CI 2.6-56; p<0.01) and of composite end-point (OR 2; 95% CI 1.2-3.6; p=0.01) but not with re-AMI and revascularization. In a simple regression model HRV values were predictors of mortality (p<0.01). However when added in the multiple regression model HRV did not have an independent correlation with the end-points considered and did not modify the correlation between depression and mortality.
Conclusions: Patients with post-AMI depression have a cardiac autonomic dysfunction as reflected by decreased HRV and increased HR. This autonomic dysfunction seems not to be an independent mediator of the increased mortality observed in depressed patients during a 5-year follow-up.