Atrial fibrillation: possible influences of rate/rhytm control strategy on cognitive performance

Growing evidence suggests that cognitive impairment and the severity of dementia are amplified in patients with atrial fibrillation (AF), independent of clinical stroke and other co-morbid conditions (1). This link is probably multifactorial with emboli, micro-bleedings, decreased cerebral perfusion (secondary to reduced cardiac output and heart rate variability), and inflammation potentially being involved (2). Rhythm control [through electrical cardioversion (3) or ablate-and-pace strategy (4)] seems to improve cardiac output, cerebral blood flow (3), and cognitive function (4). At the same time, heart rate control strategy can also exert protective effects on cognition. In fact, the reduction of the heart rate variability and the achievement of an optimal heart rate may guarantee an improved cerebral perfusion. It has been demonstrated that both low and high ventricular response rates are predictive of dementia (5). We hypothesize that cognition might be differently affected with the therapeutic choice of controlling rate versus rhythm in older persons with AF. Thus, in the present study, we analyse whether rate or rhythm control strategy may have an impact on cognition in older patients with AF, taking advantage of a large cohort of hospitalized persons. Retrospective analysis of the REPOSI database (6). People aged 65 years or more affected by AF prior to hospital admission to Italian internal medicine and geriatric wards in the years 2010, 2012, 2014 were enrolled. Patients with AF and/or atrial flutter were selected according to the diagnosis or indication for anti-arrhythmic drugs at their hospital admission. Patients diagnosed for the first time with AF during the hospitalization were excluded from the analyses. Patients with AF were classified into 3 groups: rate control, rhythm control (including patients on rhythm control therapy and those receiving drugs for both rhythm and rate control), and no therapy of interest (i.e., patients taking neither rate nor rhythm control drugs). Cognitive performance was evaluated with short blessed test (SBT). The patient socio-demographic characteristics were compared using univariate analysis by means of Chi-squared tests for categorical variables, and ANOVA for continuous variables. Logistic regression models adjusted for age, sex, education, anticoagulant/antiplatelet therapy, and comorbidities were used. RISULTATI: The study sample included 1,082 patients. Figure 1 shows the main characteristics of the study population. Less than half of the patients were treated with an oral anticoagulant both in the total sample (n=436, 40.3%) as well as in the different rhythm/rate control strategy groups (rhythm control n=86, 31.7%; rate control n=162, 48.9%; no therapy of interest n=188, 39.2%). Rhythm control strategy was protective versus the development of cognitive impairment (OR 0.56, 95%CI 0.40–0.79, p=0.001). Older age was associated with an increased risk of presenting cognitive impairment (adjusted OR 2.87, 95%CI 1.93-4.27; p<0.001), whereas education was protective (adjusted OR 0.50, 95%CI 0.35–0.62; p<0.001). Anticoagulant and/or anti-platelet therapy was associated to a non-statistically significant reduction in the risk of cognitive impairment. Results were similar using both SBT cut points. Our study shows that rhythm control strategy is associated with a lower risk of cognitive impairment. To date, only a sub-study of the AFFIRM trial (7) has explored the differences between pharmacological rhythm and rate control strategy on cognition, without founding significant results between the two groups. However, the population of the AFFIRM trial was younger (69.8±8.8 years vs 80.6±7 years) and with a better cognitive performance (MMSE score: 28.3±2.2 in the rate control group and 27.3±2.6 in the rhythm control group) whereas in our sample patients in all groups had a mean SBT score consistent with at least a questionable cognitive impairment. Moreover, on the contrary of what found in our study, anticoagulation was elevated in the AFFIRM trial, thus limiting the possible deleterious effects of microembolizations and strokes on cognition. Differently from what previously reported (5), rate control did not seem to play a protective effect on cognition in our sample. However, whereas in the AFFIRM trial participants in the rate control group had to achieve specific heart rate targets, we are unable to speculate on heart frequency of our sample. Inadequate heart frequency (both too high or too low) may explain the worse cognitive function detected in the rate control compared to other groups. The possible protective effect of rhythm control on cognition is even more relevant if it is taken into account the fact that the rhythm control group in our study was the one with the lowest prevalence of anticoagulant use.