Incidence and Implications of Atrial Fibrillation/Flutter in Hypertension: Insights from the SPRINT Trial

Abstract

We evaluated the impact of intensive blood pressure (BP) control on the incidence of new-onset atrial fibrillation/flutter (AF) and the prognostic implications of pre-existing and new-onset AF in Systolic Blood Pressure Intervention Trial (SPRINT) participants. New-onset AF was defined as occurrence of AF in 12-lead electrocardiograms after randomization in participants free of AF at baseline. Poisson regression modeling was used to calculate incident rates of new-onset AF. Multivariable-adjusted Cox proportional hazard models were used to evaluate the risk of adverse cardiovascular events (composite of myocardial infarction (MI), non-MI acute coronary syndrome, stroke, heart failure, or cardiovascular death). In 9,327 participants, 8.45% had pre-existing AF, and 1.65% had new-onset AF. The incidence of new-onset AF was 4.53 per 1000-person years, with similar rates in the standard and intensive treatment arms (4.95 vs. 4.11 per 1000-person years; adjusted p=0.14). Participants with pre-existing AF [adjusted HR: 1.83 (95% CI: 1.46–2.31); p<0.001] and new-onset AF [adjusted HR: 2.45 (95% CI: 1.58–3.80); p<0.001] had a greater risk for development of adverse cardiovascular events compared to those with no AF. Participants with pre-existing AF who achieved BP <120/80 mmHg at three months continued have a poor prognosis [adjusted HR: 1.88 (95% CI: 1.32–2.70); p=0.001] compared to those with no AF. Intensive BP control does not diminish the incidence of new-onset AF in an older, high-risk, non-diabetic population. Both, pre-existing and new-onset AF have adverse prognostic implications. In participants with pre-existing AF, residual cardiovascular risk is evident even with on-treatment BP <120/80 mm Hg.

Graphical Abstract

Introduction

Hypertension is one of the major risk factors associated with the development of atrial fibrillation/flutter (AF).^1–4 Nearly 60% of patients with AF have hypertension, and elevated blood pressure is associated with an increased incidence of AF.^1–6 Atrial fibrillation and hypertension have similar risk factors and the cardiovascular hemodynamic changes, cardiac structural modifications, and renal dysfunction consequent of hypertension, are recognized contributors to the development of AF and its associated complications.^1–6 Adequate blood pressure management is crucial for AF prevention and the associated adverse cardiovascular outcomes. The cardiovascular morphological changes and subclinical pathophysiological changes resulting from elevated blood pressure serve as a substrate for the development and persistence of AF and the eventual occurrence of poor outcomes.^1–7

Despite the role of elevated blood pressure in the etiopathogenesis of AF,^1–9 there is limited data from randomized controlled trials on the impact of intensive blood pressure control on the incidence of AF.^{6, 10} Prior observational data has been conflicting, suggesting that both standard (<140 mm Hg) and intensive (<120 mm Hg) blood pressure control may reduce incident AF.^{5, 6, 9–11} The Systolic Blood Pressure Intervention Trial (SPRINT)¹² helped to establish the benefit of intensive blood pressure management for reducing the risk of adverse cardiovascular outcomes. Thus, the SPRINT study population provides a unique opportunity to evaluate the role of intensive blood pressure control in reducing the incidence of AF.

We hypothesized that a) intensive blood pressure control reduces the risk for development of new-onset AF when compared to standard blood pressure control, b) both new-onset and pre-existing AF in hypertensive participants portend poor prognostic implications, and c) attainment of systolic blood pressure <120/80 mm Hg in individuals with pre-existing AF will attenuate the risk of adverse cardiovascular outcomes.

Methods

SPRINT trial data for this study are publicly available at NHLBI BioLINCC data repository and can be accessed at https://biolincc.nhlbi.nih.gov/home/.

Study Population and Design

The SPRINT trial was a multicenter, open-label, clinical trial that randomized 9,361 hypertensive participants enrolled between November 2010 and March 2013 to intensive and standard blood pressure management.^{12, 13} The local Institutional Review Board at the respective trial sites approved the trial. All participants provided written informed consent, and the trial complied with principles in the Declaration of Helsinki. The inclusion and exclusion criteria, and the results of the SPRINT trial have been previously described (Supplementary Methods).^{12, 13} The enrolled participants were randomized to either standard (systolic blood pressure <140 mmHg) or intensive (systolic blood pressure <120 mmHg) blood pressure treatment targets.

Electrocardiogram (ECG) Ascertainment

Participants with AF present on baseline ECG (within the first 7 days) or having a history of AF were defined as having pre-existing AF. Detection of AF on ECGs after randomization was defined as new-onset AF. The GE MAC 1200 electrocardiograph (GE, Milwaukee, Wisconsin) was used to obtain the digital ECG data with a 10 mm/mV calibration and a speed of 25 mm/s. Scheduled ECGs took place at baseline, year 2, year 4 and closeout visits.¹³ This was in addition to unscheduled ECGs obtained during suspected cardiovascular events and by the study investigators evaluating participants for safety. All ECG readings were adjudicated centrally at the Epidemiological Cardiology Research Center (EPICARE), Wake Forest School of Medicine (Winston-Salem, N.C.).^{13, 14} All ECGs underwent visual inspection for poor quality and technical errors prior to automatic processing by GE 12-SL Marquette version 2001 (GE, Milwaukee, Wisconsin).^{12, 14}

Outcomes

The primary study outcome was the incidence of new-onset AF in those free of pre-existing AF, stratified by treatment strategy. Additional study outcome was the risk of development of adverse cardiovascular events [composite of myocardial infarction (MI), non-MI acute coronary syndrome, stroke, heart failure, or cardiovascular cause of death] by the status of AF and by achieved blood pressure at three months among those with pre-existing AF. Secondary outcomes included individual components of the composite outcome separately.¹³

Statistical Analyses

Baseline characteristics were compared by AF status using descriptive statistics. Continuous variables were summarized as median and interquartile range (IQR), and compared using Wilcoxon rank-sum test. Categorical variables were summarized as counts and percentages and compared using χ² test. Poisson regression was used to assess the incidence rates of new-onset AF in overall population, and by treatment strategy. The difference in risk of new-onset AF by treatment strategy was assessed using multivariable-adjusted poisson modeling. The model was adjusted for age, sex, race, body mass index, baseline systolic blood pressure, baseline heart rate, left ventricular hypertrophy on ECG (defined using Cornell voltage criteria), clinical or subclinical cardiovascular disease, chronic kidney disease, number of ECGs, total cholesterol levels, aspirin use, statin use, angiotensin converting enzyme inhibitors/angiotensin receptor blockers (ACE/ARBs) use, smoking status, thyroid disease, alcohol abuse and treatment strategy.¹⁵ Kaplan-Meier curves were generated for cumulative incidence of new-onset AF by treatment strategy, and compared using log-rank test. Relative strength of association between risk of new-onset AF and various clinical and demographic factors was evaluated using multivariable-adjusted Cox models, and compared using likelihood ratio test, as decribed previously.^{16, 17}

Multivariable-adjusted Cox proportional hazard models were used to evaluate the hazard of adverse cardiovascular event and the secondary outcomes, by the status of AF and by achieved blood pressure among those with pre-existing AF. The models were adjusted for the abovementioned covariates. Potential interaction by treatment strategy was assessed using multiplicative interaction term. All analyses were performed using STATA SE version 16.0 (StataCorp, College Station, TX), with a two-tailed p-value of <0.05 considered statistically significant.

Results

Among 9,327 participants with available data, pre-existing AF was present in 8.45% of the study population. The incidence of new-onset AF was 4.53 per 1000-person years (95%: 3.84–5.34 per 1000-person years).

In comparison to those free of AF, participants with pre-existing AF were older [74 (IQR: 66–79) years] and more likely to be white (76.9%, p<0.001) and male (71.2%, p<0.001). Patients with pre-existing AF were also more likely to have clinical and subclinical cardiovascular disease (36.1% vs. 18.3%) and greater aspirin use (62.4% vs. 49.8%) (p<0.001 for all) (Table 1). Participants who developed new-onset AF were older [75 (IQR: 67–80) years vs. 66 (IQR: 60–75) years; p<0.001], white (76.6% vs. 55.6%; p<0.001), males (71.6% vs. 63.7%; p=0.04), with higher cardiovascular risk (Framingham Risk Score ≥15% in 75.9% vs. 60.4%; p<0.001) and a higher prevalence of chronic kidney disease (28.4% vs. 15.0%; p<0.001), compared to those free of AF in the study period. The body mass index was ~29 kg/m² across the three groups. The median follow-up period for those free of AF, with new-onset AF and pre-existing AF was 3.2 (IQR: 2.8–3.8), 3.1 (IQR: 2.5–3.7), and 3.2 (IQR: 2.6–3.7), respectively. The number of ECGs for each participant in those with no AF, pre-existing AF, and new-onset AF were 3 (IQR: 2–3), 3 (IQR: 2–3), and 3 (IQR: 3–3), respectively. There was a similar rate of new-onset AF in standard [4.95 (95% CI: 3.96–6.19) per 1000-person years] and intensive [4.11 (95% CI: 3.22–5.25) per 1000-person years] treatment arms. The multivariable-adjusted incidence rate ratio for new-onset AF between the two arms was 0.78 (95% CI: 0.56–1.10; p=0.15) (Figure 1).

Table 1.

Baseline Characteristics of the Study Participants

Characteristics	Free of AF N=8,408	Pre-Existing AF N=778	New-Onset AF N=141	p-value (Free of AF vs. Pre-existing)	p-value (Free of AF vs. New-Onset)
Criteria for Increased CV Risk
Age ≥75	2,193 (26.1)	361 (46.4)	71 (50.4)	<0.001	<0.001
CKD	1,261 (15.0)	253 (32.5)	40 (28.4)	<0.001	0.006
CVD	1,541 (18.3)	281 (36.1)	44 (31.2)	<0.001	<0.001
Clinical	1,261 (15.0)	253 (32.5)	40 (28.4)	<0.001	<0.001
Subclinical	430 (5.1)	53 (6.8)	5 (3.6)	0.04	0.40
Framingham 10-yr CV risk ≥15%	5,079 (60.4)	538 (69.2)	107 (75.9)	<0.001	<0.001
Males	5,354 (63.7)	554 (71.2)	101 (71.6)	<0.001	0.04
Age (years)
Overall	66 (60–75)	74 (66–79)	75 (67–80)	<0.001	<0.001
≥75 years	79 (77–82)	79 (77–83)	80 (78–83)	0.006	0.02
Race				<0.001	<0.001
Non-Hispanic Whites	4,678 (55.6)	598 (76.9)	108 (76.6)
Non-Hispanic Blacks	2,640 (31.4)	127 (16.3)	20 (14.2)
Hispanics	935 (11.1)	36 (4.6)	11 (7.8)
Others	155 (1.8)	17 (2.2)	2 (1.4)
Black Race	2,782 (33.1)	129 (16.6)	21 (14.9)	<0.001	<0.001
Baseline Blood Pressure (mm Hg)
Systolic Blood Pressure	143 (136–152)	142 (135–151)	145 (135–154)	0.03	0.49
Diastolic Blood Pressure	81 (73–88)	78 (69–85)	74 (66–82)	<0.001	<0.001
Serum Creatinine (mg/dL)	1.0 (0.9–1.2)	1.1 (0.9–1.3)	1.1 (0.9–1.2)	<0.001	0.01
Estimated GFR (mL/min/1.73m2)	71.9 (58.7–85.2)	65.9 (52.4–79.0)	65.6 (55.6–77.3)	<0.001	<0.001
Ur. Albumin/Creatinine Ratio (mg/gm)a	9.24 (5.6–20.3)	13.5 (6.9–36.4)	12.4 (6.8–31.1)	<0.001	0.003
Fasting Total Cholesterol (mg/dL)	188 (163–216)	173 (149–202)	173 (151–213)	<0.001	0.002
Fasting HDL Cholesterol (mg/dL)	50 (43–60)	50 (43–60)	50 (43–59)	0.68	0.43
Fasting Total TG (mg/dL)	107 (77–151)	99 (74–137)	106 (80–141)	<0.001	0.82
Fasting Plasma Glucose (mg/dL)	97 (91–105)	98 (91–105)	98 (92–106)	0.26	0.13
Statin Useb	3,543 (42.4)	436 (56.6)	68 (49.3)	<0.001	0.10
Aspirin Usec	4,181 (49.8)	485 (62.4)	82 (58.2)	<0.001	0.04
Smoking				<0.001	<0.001
Never	3,763 (44.8)	302 (38.8)	49 (34.8)
Former	3,469 (41.3)	420 (54.0)	78 (55.3)
Current	1,166 (13.9)	56 (7.2)	14 (9.9)
Framingham 10-yr CV Risk Score	17.4 (11.8–25.5)	19.8 (13.3–27.3)	22.8 (15.6–30.8)	<0.001	<0.001
BMI (kg/m2)	29.0 (25.9–32.9)	28.7 (25.7–32.6)	29.8 (26.1–34.6)	0.11	0.30
History of Alcohol Abuse	339 (4.0)	30 (3.9)	5 (3.6)	0.81	0.77
Thyroid Disease	910 (10.8)	113 (14.5)	13 (9.2)	0.002	0.54
CHA2DS2VASc Score	2 (1–3)	2 (2–3)	2 (2–3)	<0.001	<0.001
Antihypertensive agents-no./patient	2 (1–2.5)	2 (1–3)	2 (2–3)	<0.001	<0.001
Medication
ACE Inhibitors/ARBs	4,824 (57.4)	479 (61.6)	89 (63.1)	0.02	0.17
Calcium Channel Blockers	2,873 (34.2)	287 (36.9)	58 (41.1)	0.13	0.08
Beta-Blockers	2,849 (33.9)	427 (54.9)	71 (50.4)	<0.001	<0.001
Diuretics	3,753 (44.6)	318 (40.9)	68 (48.2)	0.04	0.40
Other Medications	675 (8.03)	81 (10.4)	18 (12.8)	0.03	0.04

Categorical variables are represented as counts with proportions and continuous variables are represented as medians with interquartile ranges.

Abbreviations: ACE: Angiotensin Converting Enzyme; AF: Atrial Fibrillation/Flutter; ARB: Angiotensin Receptor Blocker; BMI: Body Mass Index; CKD: Chronic Kidney Disease; CVD: Cardiovascular Disease; CV: Cardiovascular; GFR: Glomerular Filtration Rate; HDL: High-Density Lipoprotein; TG: Triglycerides; Ur: Urinary.

^ain 8,897 participants;

^bin 9,311 participants;

^cin 9,266 participants.

Figure 1. Incidence of New-Onset Atrial Fibrillation/Flutter: Stratified by Treatment Strategy.

Panel A: Cumulative incidence of new-onset atrial fibrillation/flutter stratified by treatment strategy. Panel B: Incidence rate of AF in participants randomized to intensive (blue) and standard (green) blood pressure management. IRR: Incidence rate ratio.

Ranking of the Relative Strength of Association of Risk Factors with New-Onset AF

Clinical and demographic correlates of incident AF in the study population are shown in Figure 2. The factors were ranked by chi-squared value with age being the strongest correlate for new-onset AF followed by number of ECGs, body mass index, baseline heart rate, race, smoking status, presence of clinical or subclinical cardiovascular disease, and left ventricular hypertrophy. Baseline blood pressure and blood pressure treatment strategy had a lower strength of association for incident AF. Total cholesterol, chronic kidney disease, alcohol abuse, and use of ACE/ARBs, statin and aspirin were weak correlates for incident AF.

Figure 2. Ranking of Strength of Association Between New-Onset Atrial Fibrillation/Flutter and Clinical and Demographic Factors.

This figure demonstrates the relative strength of association of respective covariates, ranked according to chi-squared values from the multivariate-adjusted model. Chi-square values were corrected for degrees of freedom allocated to respective covariates in the model ensuring comparison on same scale.

Prognostic Implications of New-Onset and Pre-Existing AF

Adverse cardiovascular events occurred in 5.2% participants free of AF. Adverse cardiovascular events were seen in 12.6% and 16.3% participants with pre-existing AF and new-onset AF, respectively (Table S1). The proportion of participants in each group who developed stroke, MI, non-MI acute coronary syndrome, heart failure, and death due to cardiovascular causes is noted in Table S1. Participants who had pre-existing AF [1.83 (95% CI: 1.46–2.31); p<0.001] and new-onset AF [HR: 2.45 (95% CI: 1.58–3.80); p<0.001] had a greater hazard for the development of adverse cardiovascular events compared to those free of AF (Figure 3). No interaction by treatment strategy was seen (p>0.10). In those with new-onset AF on intensive treatment, the hazard for development of adverse cardiovascular event was 0.84 (95% CI: 0.56–1.26; p=0.39) compared to standard therapy. The hazard for stroke in those with pre-existing AF was 2.20 (95% CI: 1.40–3.48; p=0.001) and for those with new-onset AF was 2.92 (95% CI: 1.16–7.32; p=0.02). Those with pre-existing AF [HR: 3.58 (95% CI: 2.46–5.23); p<0.001] and new-onset AF [HR: 5.32 (95% CI: 2.93–9.67); p<0.001], also had a greater risk of developing heart failure compared to those free of AF. The hazard for cardiovascular death was 2.69 (95% CI: 1.64–4.40; p<0.001) for pre-existing AF and 1.25 (95% CI: 0.17–9.22; p=0.82) for those with new-onset AF.

Figure 3. Prognostic Implications of New-Onset and Pre-Existing Atrial Fibrillation/Flutter.

Adjusted Kaplan-Meier curves for the risk of development of adverse cardiovascular events. The curve represents participants with new-onset (green), pre-existing (red), and no atrial fibrillation/flutter (blue).

Residual Cardiovascular Risk in Participants with Pre-Existing AF and Intensive Blood Pressure Control

Approximately 35% (n=281) of participants with pre-existing AF achieved the blood pressure level of <120/80 mm Hg at three months. Those who achieved the blood pressure level of <120/80 mmHg were clinically and demographically similar to those who did not, except for being older [75 (IQR: 67–80) years vs. 73 (IQR: 65–79) years; p=0.01] and having worse renal function [estimated glomerular filtration rate 63.2 (IQR: 50.6–77.8) vs 68.1 (IQR: 54.3–81.9); p=0.007]. Participants with pre-existing AF and blood pressure of <120/80 mmHg at three months, had a greater risk for adverse cardiovascular events compared to those free of AF [HR: 1.88 (95% CI: 1.32–2.70); p=0.001] (Figure 4). Among individuals with pre-existing AF and blood pressure of <120/80 mmHg the hazard was 2.43 (95% CI: 1.24–4.76; p=0.01) for stroke, 1.74 (95% CI: 0.98–3.11; p=0.058) for MI and 3.17 (95% CI: 1.76–5.69; p<0.001) for heart failure, compared to those free of AF.

Figure 4. Residual Cardiovascular Risk in Participants with Pre-Existing Atrial Fibrillation/Flutter.

Adjusted Kaplan-Meier curves for the risk of development of adverse cardiovascular events. The blue curve represents participants free of atrial fibrillation/flutter (AF). Those with pre-existing AF and having attained blood pressure <120/80 mmHg are represented in red and those who did not are represented in green.

Discussion

In the large, well-phenotyped cohort of relatively older, high-risk, non-diabetic hypertensive participants from the SPRINT trial, we observed a similar incidence of new-onset AF among individuals in the intensive and standard blood pressure control arms. New-onset AF was driven primarily by age, body mass index, race, smoking status, and presence of clinical or subclinical cardiovascular disease. We observed that both pre-existing and new-onset AF increased the hazard of developing adverse cardiovascular events compared to those free of AF. Aggressive control of blood pressure to <120/80 mm Hg was not associated with decreased risk of adverse cardiovascular events in those with prevalent AF. This suggests that residual cardiovascular risk needs to be addressed after stringent blood pressure control in this population.

Data from multiple prospective studies have established the association of hypertension with incident AF, with empirical evidence from observational studies and smaller trials specifying the role of blood pressure reduction in decreasing the incidence of AF.^{1–6, 8–11, 18–20} Prior works have focused on evaluating the baseline or on-treatment blood pressure with progression of AF or development of new-onset AF.^{5, 6, 8, 9, 11, 18–20} Yet, it has been unclear whether intense blood pressure reduction will further decrease the incidence of atrial fibrillation. Our work provides knowledge about the lack of incremental benefit of intensive blood pressure control to reduce new-onset AF in this high-risk, older non-diabetic cohort. Structural remodeling, autonomic dysregulation, increased oxidative stress, interstitial fibrosis and activation of the renin-angiotensin-aldosterone system have been proposed as probable pathophysiological pathways for the development of AF in hypertension.^{1, 2, 21–24} Recent data from the Atherosclerosis Risk in Communities study suggests that the retrospective application of the 2017 high blood pressure guidelines does not correspond to to a higher incidence of AF attributed to hypertension.²⁵ The similar incidence of AF in both intensive and standard blood pressure control arms of the SPRINT cohort suggests that underlying pathophysiology contributing to the development of AF may not be completely mitigated with more intensive blood pressure control. Additionally, left ventricular hypertrophy resulting from prolonged high blood pressure is associated with the occurrence and progression of AF in hypertensive individuals.^26–28 We observed that baseline left ventricular hypertrophy had a strong to modest association with new-onset AF. Strict blood pressure control may partially reverse the architectural changes in the heart, but the effects take time to establish and ongoing subclinical organ damage from various pathological pathways may persist. This may contribute to the residual risk and, eventually, a higher adverse cardiovascular event risk despite optimal blood pressure control. High blood pressure may confer a higher risk for incident AF, but due to its association with various comorbidities, its tight control may not attenuate the competing risk of AF arising out of other uncontrolled residual factors.⁸ Thus, while blood pressure control has been convincingly demonstrated reduce AF burden and incident AF, there may be diminishing returns beyond a moderately strict threshold.^{9, 15}

SPRINT enrolled a relatively older (mean age 67.9 years) population with a higher baseline cardiovascular disease risk (mean Framingham Risk Score of 24.8%).¹² This suggested that the population was predisposed to AF regardless of blood pressure control due to ongoing underlying clinical and subclinical pathology. A younger, lower-risk population with less comorbidity burden may be ideal to demonstrate a more significant preventive role for aggressive blood pressure control strategy in reducing the incidence of AF. Moreover, SPRINT excluded individuals with diabetes mellitus, which is a known pre-disposing factor for new-onset AF.²⁹ Intensive blood pressure control in diabetic individuals has been associated with reduced incidence for the composite of AF or intermediary phenotype (P-wave indices), but not for AF alone.³⁰ Hence, presence of diabetes may be an important effect modifier for the relationship of intensive blood pressure control with new-onset AF, which requires prospective testing in a population inclusive of diabetics. Prior data from epidemiological studies in younger, lower risk populations, have shown that poor renal function is associated with incident AF, which is contrasted by chronic kidney disease being a poor clinical correlate for incident AF in the older, non-diabetic, hypertensive population with large multimorbidity burden.^31–33 SPRINT enrolled participants with systolic blood pressure ≤180 mm Hg, which excluded a population which stands to benefit the most from the intensive BP control for prevention of new-onset AF. Additionally, the short follow-up period (median follow-up 3.3 years) may have restricted our ability to observe the long-term effects of intensive blood pressure control in reducing incident AF. The residual risk burden, a non-linear relationship of incident AF with blood pressure,⁹ and the older, high-risk population substrate with a short follow-up period may explain the lack of benefit of intensive blood pressure control for new-onset AF observed in our study.

AF and high blood pressure are independently predictive of adverse cardiovascular events such as stroke, MI, heart failure and cardiovascular death.^{1, 2} Our work suggests that having both AF and hypertension significantly increases the risk for adverse cardiovascular events irrespective of blood pressure control strategy. Previous studies suggest similar outcomes in AF with and without hypertension, which is in line with our finding of a similar risk of adverse cardiovascular event in participants with pre-existing and new-onset AF.³⁴ Residual cardiovascular risk exists even after the strict control of blood pressure in participants with AF. Once AF develops, intensive blood pressure control may not reduce adverse cardiovascular outcomes. A U-shaped relationship between blood pressure and the risk of adverse cardiovascular outcomes in those with AF has been reported in prior studies^{15, 18} suggesting that a strict blood pressure regulation in hypertensive AF individuals does not attenuate the cardiovascular risk.¹⁵ The competing risk from coexisting comorbidities alongside hypertension may have a significant role to play in this complex relationship.

AF may act as a surrogate marker for clinical and subclinical pathology in hypertensive individuals and an indicator for increased risk of adverse cardiovascular events. This underlines the importance of utilizing AF for risk stratification and prognostication, and promotes clinician vigilance for AF among hypertensive participants. An aggressive lifestyle and risk factor modification strategy has shown to be effective in reduction of AF burden.^35–37 A comprehensive risk reduction strategy alongside blood pressure management may help reduce the risk of incident AF and the associated adverse cardiovascular outcomes. Prospective studies evaluating the impact of lower blood pressure in younger participants, including those with diabetes, with longer follow-up are needed to determine the long term effect of intensive blood pressure control on reducing the AF burden. Identification of optimal treatment targets for blood pressure will help in reducing the incidence of new-onset AF and the overall AF burden in hypertensive individuals. Further research is needed to understand the implications of a heterogeneous population substrate and pathophysiology of AF in hypertension, with a focus on gaining mechanistic insights into the role of aggressive blood pressure control in secondary prevention of AF and associated adverse cardiovascular events.

Study Limitations

Our study has some important limitations. Our findings are limited by the short follow-up duration and infrequent ECG recording to detect incident AF. The ECGs were recorded during study visits or during suspected events, and therefore may not have been able to capture asymptomatic paroxysmal AF. Power analysis using two-sample proportion test based on current data suggests that if a difference in event rate of 0.7% (currently 0.3%) existed between the two treatment arms, we would have 80% power to detect the incremental benefit of intensive blood pressure control for the reduction of risk for incident AF (α=0.05). The study population was older, non-diabetic and had a high comorbidity burden, which may have affected our ability to detect a substantial effect of blood pressure control on incident AF. The generalizability of our results are reduced by the exclusion of individuals with diabetes mellitus, which is a major co-existing comorbidity in hypertensive patients. Our study is also limited by the lack of data on the cardiac imaging parameters of the SPRINT participants. Notably, SPRINT did not include participants with diabetes mellitus, which is a known risk factor for AF and adverse cardiovascular outcomes. Despite these limitations, the robust study methodology followed by the SPRINT investigators enables us to gain insights into the role of high blood pressure management in the incidence and implications of AF.

Perspectives

Intensive blood pressure control in participants with hypertension may not result in a lower incidence of new-onset AF compared to standard blood pressure control in a relatively older, non-diabetic, and high risk population of SPRINT. Pre-existing and new-onset AF both elevated the risk of adverse cardiovascular events to a similar degree in SPRINT participants. These observations suggest that the risk of adverse cardiovascular events is incompletely attenuated by stringent blood pressure control in high risk patients with pre-existing AF, signifying the importance of residual cardiovascular risk after optimal blood pressure control.

Novelty and Significance.

What is New?

• The impact of intensive blood pressure control on the incidence of new-onset atrial fibrillation/flutter (AF) is not known.

• Intensive blood pressure control may not provide benefit for reducing the burden of new-onset AF.

What is Relevant?

• Clinicians caring for hypertensive patients must look out for atrial AF, as both new-onset and pre-existing atrial AF portend increased risk of adverse cardiovascular outcomes.

• The residual cardiovascular risk from coexisting comorbidities in patients of AF with adequately controlled blood pressure may drive the associated risk for adverse cardiovascular outcomes.

Summary

• Intensive blood pressure control may not further diminish the incidence of new-onset AF over standard control, in an older, high-risk, non-diabetic cohort.

• Both, pre-existing and new-onset AF have adverse prognostic implications.

• Residual cardiovascular risk exists in individuals with pre-existing AF and optimally controlled BP <120/80 mm Hg.