Blood pressure in atrial fibrillation

Abstract

Introduction: Hypertension is a leading risk factor for cardiovascular morbidity and premature death. Prevalence of hypertension in the adult population in Sweden has been estimated to 27%. Atrial fibrillation (AF) is the most prevalent sustained arrhythmia of clinical relevance with an estimated prevalence of at least 2.9% among adults in Sweden. Similarly to hypertension, AF is independently associated with an increased risk for cardiovascular morbidity and with a two-fold increased risk of death. The underlying mechanisms responsible for this association however, are not fully known. Both conditions may impose a heavy burden upon affected patients as well as on the health care system. AF and hypertension are closely intertwined and often coexist. Hypertension is the major risk factor for AF development and conversely, AF affects blood pressure (BP). The irregular heart rhythm in AF is one factor influencing BP, but also other factors may play a part. Furthermore, the presence of AF has implications for conventional BP measurement. AF-related effects on BP are studied to a very limited extent. Possibly, AF-induced BP effects may have pathophysiological consequences and may also influence BP measurement accuracy. Consequently, these factors may negatively influence risk assessment and prognosis in patients with AF.

The aims of this thesis were 1) to systematically quantify beat-to-beat BP variability in patients with AF compared to sinus rhythm (SR); 2) to study how BP, as measured with different techniques, is affected by the presence of AF; 3) to investigate the relationship between peripheral and central intra-arterial BP, in patients with AF compared to SR; 4) to evaluate the accuracy of conventional BP measurement in relation to peripheral and central intra-arterial BP, in patients with AF and compared to SR.

Methods and results: In the prospective study I, patients scheduled for a coronary angiography were recruited. Participants included 21 patients in AF and 12 patients with SR. Intra-arterial BP was recorded from the radial and brachial artery and from the ascending aorta. The primary outcome measure was beat-to-beat BP variability, defined as average systolic and diastolic BP difference between consecutive beats, at each site of measurement. A significant difference (p<0.001) in BP variability, in AF compared to SR, was observed for all locations of measurement. Systolic BP variability was roughly doubled in patients with AF (4.9 vs 2.4 mmHg), whereas diastolic BP variability was approximately six times as high (7.5 vs 1.2 mmHg) in patients with AF compared to SR. Study II was a retrospective registry analysis based on data from electronic medical records. 487 patients, treated with electrical cardioversion (ECV) for persistent AF, were included in the study. Information regarding auscultatory sphygmomanometric BP and rhythm, on the day before and 7 days after ECV, was obtained. The primary outcome measure was BP change in patients with restored SR after ECV. In this group with restored SR, systolic BP increased by 9 mmHg (p<0.01), whereas diastolic BP decreased by 3 mmHg (p<0.01). Furthermore, the proportion of patients with a hypertensive BP-level (≥140/90) increased by 40% in this group. In study III, 98 patients with persistent AF undergoing ECV were prospectively recruited. BP was evaluated with 24-h ambulatory BP monitoring before and approximately one week after ECV. The primary outcome measure was BP change in patients with restored SR after ECV. Among 60 patients maintaining SR, mean systolic 24-h ambulatory BP increased by 5.6 mmHg (p<0.001) and mean diastolic 24-h ambulatory BP decreased by 4.7 mmHg (p<0.001). Accordingly, a 10.4 mmHg (25%) increase in pulse pressure was observed among patients with restored SR. Study IV comprised the same individuals as study I. Conventional BP (auscultatory sphygmomanometric and automated oscillometric) and intra-arterial BP was measured simultaneously. The first aim was to investigate how intra-arterial BP changes throughout the arterial tree in patients with AF in comparison to patients in SR. The second aim was to evaluate the accuracy of conventional BP measurement in patients with AF in comparison to central and peripheral intra-arterial BP, and in comparison to patients in SR. BP changes throughout the arterial tree was similar in patients with AF compared to SR. Conventional BP was in general very accurate in comparison to diastolic intra-arterial BP, both in AF and SR. In patients with AF, oscillometric blood pressure overestimated systolic intra-arterial brachial (4.1 mmHg, p=0.07) and central (5.0 mmHg, p=0.04) BP. With measurement bias in SR taken into account, oscillometric BP over-estimated systolic intra-arterial brachial BP by 14.1 mmHg (p<0.01) and central BP by 9.0 mmHg (p=0.01) in patients with AF.

Conclusions: Beat-to-beat BP variability is increased in patients with AF compared to SR. According to the results from studies in this thesis, systolic BP is lower and diastolic BP is higher in AF compared to SR, as measured by auscultatory sphyghmomanometry or by oscillometric 24-h ambulatory BP monitoring. As a consequence, pulse pressure is markedly lower in AF compared to SR. Intra-arterial BP change throughout the arterial tree is similar in patients with AF and SR. Conventional BP measurement was accurate in relation to diastolic intra- arterial BP, but oscillometric BP measurement overestimated intra-arterial brachial and central systolic BP in patients with AF, in particular when compared to patients in SR. The presence of AF affects BP. This may have implications for the accuracy of conventional BP measurement and may possibly also have pathophysiological consequences. Suboptimal understanding, measurement and treatment of BP may negatively influence prognosis in patients with AF.