Abstract
Aims
Knowledge of long‐term outcomes in patients with atrial fibrillation (AF) remains limited. We sought to evaluate the risk of new‐onset heart failure (HF) in patients with AF and a low cardiovascular risk profile.
Methods and results
Data from the Swedish National Patient Register were used to identify all patients with a first‐time diagnosis of AF without underlying cardiovascular disease at baseline between 1987 and 2018. Each patient was compared with two controls without AF from the National Total Population Register. In total, 227 811 patients and 452 712 controls were included. During a mean follow‐up of 9.1 (standard deviation 7.0) years, the hazard ratio (HR) for new‐onset HF was 3.55 [95% confidence interval (CI) 3.51–3.60] in patients compared with controls. Women with AF (18–34 years) had HR for HF onset 24.6 (95% CI 7.59–80.0) and men HR 9.86 (95% CI 6.81–14.27). The highest risk was within 1 year in patients 18–34 years, HR 103.9 (95% CI 46.3–233.1). The incidence rate within 1 year increased from 6.2 (95% CI 4.5–8.6) per 1000 person‐years in young patients (18–34 years) to 142.8 (95% CI 139.4–146.3) per 1000 person‐years among older patients (>80 years).
Conclusions
Patients studied had a three‐fold higher risk of developing HF compared with controls. Young patients, particularly women, carry up to 100‐fold increased risk to develop HF within 1 year after AF. Further studies in patients with AF and low cardiovascular risk profile are needed to prevent serious complications such as HF.
Keywords: Atrial fibrillation, Heart failure, Case–control study, Registry‐based study
Introduction
Atrial fibrillation (AF) is the most common arrhythmia, and the risk for it increases with age. 1 The prevalence of AF in adults is estimated to be between 2% and 4%, 2 which is expected to increase further with the increased ageing of the general population. 3 AF and heart failure (HF) frequently coexist and affect each other negatively in terms of long‐term prognosis. 4
A range of pathophysiological mechanisms participate in the interplay between AF and HF, regardless HF subtypes. 5 , 6 AF may be linked to development of HF with reduced ejection fraction (HFrEF) through rapid ventricular rates or irregularity. Further, diffuse fibrosis, tachycardia, loss of atrial systole, and irregularity may take part in onset of HF with preserved ejection fraction (HFpEF) in patients with AF. Equally, through elevated atrial pressures, activation of renin‐angiotensin and sympathetic systems HF may lead to AF. 7
Previous reports have highlighted that HF is an important comorbidity associated with AF with a prevalence of approximately 20%. 1 , 8 Among patients with AF, there is also an important interplay of other comorbidities on prognosis; in clinical practice, these are evaluated on a daily basis, with the CHA2DS2‐VASc score being the most commonly used clinical prediction score for stroke. 9 Preventive measures focusing on the use of anticoagulant treatment to prevent embolic stroke in patients with AF are well established and include HF into the score models. 1 However, limited attention has been directed to the prevention of HF in patients with AF, despite its high prevalence and unfavourable impact on outcomes in these patients. 1 , 4 Moreover, similar cardiovascular and non‐cardiovascular comorbidities often occur in both AF and HF and, to a great extent, are also risk factors for both conditions. The burden of risk factors is also known to be associated with the risk for onset of AF and HF. Additionally, HF is a risk factor for AF onset but is also described as an outcome in patients with AF. 1 , 10 However, data about the excess risk of HF in patients with AF without coexisting cardiovascular comorbidities are lacking, as are data about risk models regarding patients with AF in contemporary populations.
Furthermore, age at diagnosis of AF is well defined in risk prediction models for prevention of stroke and is particularly important after the age of ≥65 years, 9 whereas knowledge on how age at time of diagnosis of AF relates to excess risk for HF is lacking. Similarly, sex is an important factor in risk evaluation for stroke prevention 9 ; however, data on sex differences in risk for HF among AF patients remain limited, especially in current populations.
The present study therefore aims to investigate the risk of HF in patients with AF and low cardiovascular risk profile compared with controls without AF from the general population and to evaluate the effect of sex and age on the risk for HF onset over a period of more than 30 years.
Methods
Registers
Since 1947, all residents in Sweden have been assigned a 12‐digit personal identification number (PIN) that enables the linkage of different registries. 11 Demographic data about all residents in Sweden have been registered in the Total Population Register since the beginning of 1960s. 12 The Swedish National Patient Register (NPR), which became mandatory on the national level and is almost complete since 1987, contains data on all hospitalized patients, including codes for principal and contributory diagnoses and procedure codes. Since 2001, these data from specialized outpatient settings have also been included. Currently, the coverage of the registry is close to 100% and has a high validity in general. 13 Information on cause of death and time of death was obtained from the Swedish Cause of Death Register. 14
Study population and definitions
We identified all patients in the NPR with a first recorded primary or secondary diagnosis of AF at ≥18 years of age. Patients with AF were identified using International Classification of Disease (ICD) codes; AF was defined as ICD‐9 code 427D up to 1997 and in subsequent years with ICD‐10 code I48. The observation period was from 1 January 1987 to 31 December 2018, with a minimum observation time for all subjects of 1 year. As an outcome event, HF was defined as ICD‐9 codes 428A, 428B, or 428X and ICD‐10 code I50 (Supporting Information, Table S1 ).
Data on coexisting medical conditions were obtained from the NPR using ICD‐9 and ICD‐10 codes, as both primary and secondary diagnoses (Supporting Information, Tables S1 and S2 ). Selection procedures of the study population are presented in Supporting Information, Figure S1 . Patients with coexisting diagnostic codes for hypertension, diabetes mellitus, coronary or peripheral artery disease, stroke, transient ischaemic attack (TIA), cardiomyopathy, pulmonary arterial hypertension (PAH), congenital heart disease (CHD), valve disease, or renal failure (prior or at baseline) were excluded from the study population (Supporting Information, Tables S1 and S2 and Figure S1 ).
With the use of PINs for each patient with AF, two individuals from the general population registered in the Total Population Register were selected by equivalent age, sex, and county, here labelled as controls. All controls were free of a registered AF before the date of AF for the matching case. However, a control could later be diagnosed with AF and then be registered as a case. Thus, the same individual could be registered first as a control and later as a case.
This study was approved by the Regional Medical Research Ethics Committee of Gothenburg (Dnr: Gbg104‐15 and 893‐18) and confirms to the principles defined by the Declaration of Helsinki.
Statistical analysis
All patients were followed from the date on which the diagnostic code for AF occurred for the first time in the NPR, defined as the index registration or baseline, until the first diagnosed HF, death, or the end of the study. A diagnosis of HF during follow‐up was defined as an outcome event. Baseline characteristics are presented as numbers (percentages) for dichotomous and categorical data and as means [standard deviations (SDs)] and median values for numeric variables. The cohort was stratified by sex and by the following age groups: 18–34, 35–49, 50–59, 60–69, 70–79, and ≥80 years of age. All analyses were performed per sex and age group. The significance level was set at alpha < 0.05.
HF rates and corresponding 95% confidence intervals (CIs) per 1000 person‐years were calculated. The relative risk for occurrence of HF between patients and controls was assessed by Cox proportional hazard regression. The impact of death as a competing risk was evaluated by Fine and Gray's sub‐distribution hazard model. Results are shown as Model 1 representing the crude risk estimation; Model 2 adjusted for age and sex; Model 3 adjusted for age, sex, chronic obstructive pulmonary disease (COPD), and cancer; and Model 4 with estimation of the competing risk. Time to event during the total observation time and within 1 and 5 years is presented with cumulative incidence functions.
All statistical analyses and data management were performed with the statistical software SPSS Windows Version 18.0 (IBM Corp., Armonk, NY, USA) or SAS Version 9.4 (SAS Institute, Cary, NC, USA), and all graphs were created using R Version 3.1.3 (http://R‐project.org).
Results
Baseline characteristics
In total, we identified 860 814 patients with a first‐time diagnosis of AF in the NPR. Of these, 633 003 patients met at least one of the exclusion criteria, and as a result, the final population comprised 227 811 patients. From 1 981 198 controls, using the same exclusion criteria as for patients with a first‐time diagnosis of AF, we identified 452 712 controls. Mean follow‐up time of cases and controls was 9.1 (SD ± 7.0) years.
Men with AF were significantly younger compared with women with AF in both cases and controls, with mean ages of 65.5 years (SD ± 15.0 years) and 72.7 years (SD ± 13.0 years), respectively. Non‐cardiovascular comorbidities, such as COPD, asthma, obesity, thyrotoxicosis, depression, dementia, and cancer, were significantly higher among patients with AF than among controls (Table 1 ). The prevalence of AF increased with age, with the highest prevalence in the 70–79 years age group (Supporting Information, Figure S2 ).
Table 1.
Baseline characteristics of the study population
| Men | Women | Total | |||||||
|---|---|---|---|---|---|---|---|---|---|
| AF patients n = 126 370 | Controls n = 251 033 | P‐value | AF patients n = 101 441 | Controls n = 201 679 | P‐value | AF patients n = 227 811 | Controls n = 452 712 | P‐value | |
| Age, mean (SD) | 65.5 (15.0) | 65.3 (14.9) | 0.0281 | 72.7 (13.0) | 72.7 (12.9) | 0.1424 | 68.7 (14.6) | 68.6 (14.5) | 0.012 |
| Comorbidities | |||||||||
| COPD, n (%) | 5264 (4.2) | 3883 (1.6) | <0.0001 | 4239 (4.2) | 3102 (1.5) | <0.0001 | 9503 (4.2) | 6985 (1.5) | <0.0001 |
| Asthma, n (%) | 3443 (2.7) | 3624 (1.4) | <0.0001 | 3800 (3.8) | 3638 (1.8) | <0.0001 | 7243 (3.2) | 7081 (1.6) | <0.0001 |
| Obesity, n (%) | 598 (0.5) | 394 (0.2) | <0.0001 | 769 (0.8) | 646 (0.32) | <0.0001 | 1367 (0.6) | 1040 (0.2) | <0.0001 |
| Thyreotoxicosis, n (%) | 789 (0.6) | 389 (0.2) | <0.0001 | 2532 (2.5) | 1513 (0.8) | <0.0001 | 3321 (0.7) | 1902 (0.4) | <0.0001 |
| Depression, n (%) | 2660 (2.1) | 4754 (1.9) | <0.0001 | 3065 (3.0) | 5551 (2.8) | <0.0001 | 5725 (2.5) | 10 305 (2.3) | <0.0001 |
| Dementia, n (%) | 6597 (5.2) | 8991 (3.6) | <0.0001 | 5709 (5.6) | 9698 (4.8) | <0.0001 | 12 306 (5.4) | 18 698 (4.1) | <0.0001 |
| Cancer, n (%) | 11 773 (9.3) | 15 946 (6.4) | <0.0001 | 11 397 (11.2) | 16 774 (8.3) | <0.0001 | 23 170 (10.2) | 32 720 (7.2) | <0.0001 |
Abbreviations: AF, atrial fibrillation; COPD, chronic obstructive pulmonary disease; SD, standard deviation.
Note: Values are expressed as number (percentage) if not otherwise stated.
Incidence rate of HF
The incidence rate of HF per 1000 person‐years at risk during the total follow‐up period was 35.6 (27.8–45.5) for all AF patients and 10.0 (7.3–13.5) for controls. Time to HF diagnosis after index registration for AF in the study population is presented in Figure 1 . The corresponding figures within 1 year of AF diagnosis were lowest among patients in the youngest age category and increased with increasing age (Table 2 and Figure 2 ). The incidence rate of HF within 5 years of AF diagnosis was lower than for incidence rates within 1 year in all age categories, among patients with AF, but also among controls.
Figure 1.
Time to heart failure diagnosis after index registration for atrial fibrillation (AF) in the study population. Data from the National Patient Register and the Total Population Register in Sweden. Note: Controls had equivalent age, sex, and county.
Table 2.
Risk for incident heart failure stratified by age and sex
| Heart failure, n (%) | Incidence rate per 1000 person‐years | HR (95% CI) | HR (95% CI) | HR (95% CI) | HR (95% CI) | |||
|---|---|---|---|---|---|---|---|---|
| Patients | Controls | Patients | Controls | Crude Model 1 | Adjusted Model 2 | Adjusted Model 3 | Adjusted Model 4 | |
| All | 61 490 (27) | 39 528 (8.7) | 35.6 (27.8–45.5) | 10.0 (7.3–13.5) | 3.55 (3.52–3.60) | 3.88 (3.83–3.93) | 3.81 (3.77–3.86) | 3.28 (3.23–3.33) |
| Men | 31 539 (25.0) | 20 207 (8.0) | 31.0 (31.7–31.4) | 9.0 (8.8–9.1) | 3.45 (3.39–3.51) | 3.70 (3.63–3.76) | 3.63 (3.57–3.70) |
3.24 (3.16–3.32) |
| Women | 29 951 (29.5) | 19 321 (9.6) | 42.1 (41.6–42.6) | 11.3 (11.1–11.4) | 3.72 (3.65–3.79) | 4.09 (4.02–4.17) | 4.02 (3.95–4.10) |
3.33 (3.27–3.39) |
| Age groups | ||||||||
| 18–34 | 199 (3.2) | 37 (0.3) | 2.4 (2.1–2.8) | 0.2 (0.2–0.3) | 11.0 (7.70–15.53) | 11.06 (7.78–15.7) | 11.04 (7.77–15.69) |
10.9 (7.68–15.50) |
| 35–49 | 1695 (9.4) | 534 (1.5) | 7.5 (7.2–7.9) | 1.1 (1.0–1.2) | 6.76 (6.14–7.45) | 6.80 (6.17–7.49) | 6.76 (6.13–7.45) |
6.63 (6.02–7.31) |
| 50–59 | 5380 (17.8) | 2058 (3.4) | 16.0 (15.6–16.5) | 2.8 (2.7–2.9) | 5.84 (5.56–6.15) | 5.86 (5.57–6.16) | 5.78 (5.49–6.08) |
5.57 (5.30–5.86) |
| 60–69 | 13 630 (26.1) | 7002 (6.7) | 29.2 (28.7–29.7) | 6.6 (6.5–6.8) | 4.57 (4.44–4.71) | 4.58 (4.45–4.72) | 4.50 (4.37–4.63) |
4.19 (4.07–4.31) |
| 70–79 | 22 353 (34.5) | 14 760 (11.4) | 53.1 (52.4–53.8) | 14.6 (14.4–14.8) | 3.75 (3.67–3.83) | 3.78 (3.70–3.86) | 3.70 (3.62–3.78) |
3.26 (3.20–3.33) |
| >80 | 18 233 (32.2) | 15 137 (13.6) | 91.9 (90.5–93.2) | 28.6 (28.2–29.1) | 3.19 (3.12–3.26) | 3.22 (3.15–3.29) | 3.17 (3.10–3.24) |
2.54 (2.48–2.59) |
Abbreviations: CI, confidence interval; HR, hazard ratio; n, numbers.
Note: Values are expressed as number (percentage) if not otherwise stated. Adjusted Model 2: adjusted for age and sex. Adjusted Model 3: adjusted for age, sex, chronic obstructive pulmonary disease, and cancer. Adjusted Model 4: competing risk by Fine and Gray's sub‐distribution hazard model.
Figure 2.
Time to heart failure diagnosis after index registration for atrial fibrillation (AF) within 5 years in patients and controls. Data from the National Patient Register and the Total Population Register in Sweden. Note: Controls had equivalent age, sex, and county.
Risk of HF in patients with atrial fibrillation compared with controls
During the total follow‐up period, patients with AF had 3.55 times higher risk (95% CI 3.51–3.60) for HF onset compared with controls. This was also consistent in different multivariate regression models, as well as in the competing risk analysis (Table 2 ). When compared with controls from the general population, we found that AF patients younger than 50 years had the highest risk for HF onset [hazard ratio (HR) 103.93, 95% CI 46.34–233.11]. After adjustment for age, sex, COPD, and cancer, the risk for HF remained significantly higher in patients with AF than in controls (HR 3.81, 95% CI 3.77–3.86) (Table 2 ).
Sex and age‐related differences
The total incidence rate of HF was slightly higher among women than among men: 42.1 (95% CI 41.6–42.6) per 1000 patient‐years at risk for women vs. 31.0 (95% CI 30.7–31.4) among men (Table 3 and Supporting Information, Figure S4 ). Women also had a higher HR when compared with controls, whereas the incidence rates per age group were slightly higher among men. However, when compared with controls, women with AF had a more increased risk of developing HF than did men with AF. The largest difference in HR was seen among young women of 18–34 years of age: HR 24.6 (95% CI 7.59–80.0) in young women with AF vs. HR 9.86 (95% CI 6.81–14.27) in young men with AF.
Table 3.
Risk for heart failure in male and female patients diagnosed with atrial fibrillation without other cardiovascular comorbidities or risk factors compared with controls from the general population in Sweden during the observation period from 1 January 1987 to 31 December 2018
| Total follow‐up 30 years | ||||||
|---|---|---|---|---|---|---|
| Heart failure, n (%) | Incidence rate per 1000 person‐years | HR (95% CI) | P‐value | |||
| Patients | Controls | Patients | Controls | Crude | ||
| Male | ||||||
| N (%) | 31 539 (24.96) | 20 207 (8.05) | 31.0 (30.7–31.4) | 9.0 (8.8–9.1) | 3.63 (3.57–3.70) | <0.0001 |
| Age groups | ||||||
| 18–34 | 162 (3.36) | 34 (0.35) | 2.5 (2.1–2.9) | 0.3 (0.2–0.4) | 9.86 (6.81–14.27) | <0.0001 |
| 35–49 | 1387 (10.01) | 452 (1.63) | 8.0 (7.5–8.4) | 1.2 (1.1–1.3) | 6.52 (5.87–7.25) | <0.0001 |
| 50–59 | 3862 (18.51) | 1575 (3.77) | 16.8 (16.2–17.3) | 3.2 (3.0–3.3) | 5.33 (5.03–5.66) | <0.0001 |
| 60–69 | 7997 (15.38) | 4391 (6.98) | 29.3 (28.6–29.9) | 7.3 (7.1–7.5) | 4.02 (3.87–4.17) | <0.0001 |
| 70–79 | 10 885 (33.29) | 7475 (11.49) | 54.5 (53.5–55.6) | 16.1 (15.8–16.5) | 3.38 (3.28–3.48) | <0.0001 |
| >80 | 7246 (32.04) | 6280 (14.26) | 98.1 (95.9–100.4) | 32.6 (31.8–33.4) | 2.97 (2.87–3.07) | <0.0001 |
| Female | ||||||
| N (%) | 29 951 (29.53) | 19 321 (9.58) | 42.1 (41.6–42.6) | 11.3 (11.1–11.4) | 4.02 (3.95–4.10) | <0.0001 |
| Age groups | ||||||
| 18–34 | 37 (2.82) | 3 (0.12) | 2.2 (1.6–3.1) | 0.1 (0.0–0.3) | 24.64 (7.59–80.0) | <0.0001 |
| 35–49 | 308 (7.54) | 82 (1.01) | 6.1 (5.5–6.8) | 0.1 (0.0–0.3) | 8.09 (6.34–10.33) | <0.0001 |
| 50–59 | 1518 (16.29) | 483 (2.59) | 14.4 (13.7–15.2) | 2.1 (1.9–2.2) | 7.32 (6.60–8.11) | <0.0001 |
| 60–69 | 5633 (27.31) | 2611 (27.31) | 29.1 (28.4–29.9) | 5.8 (5.5–6.0) | 5.37 (5.12–5.62) | <0.0001 |
| 70–79 | 11 468 (35.67) | 7285 (11.36) | 51.8 (50.9–52.8) | 13.3 (13.0–13.6) | 4.04 (3.93–4.16) | <0.0001 |
| >80 | 10 987 (32.36) | 8857 (13.23) | 88.1 (86.5–89.8) | 26.3 (25.8–26.9) | 3.32 (3.22–3.41) | <0.0001 |
Abbreviations: CI, confidence interval; HR, hazard ratio; n, numbers.
Note: Values are expressed as number (percentage) if not otherwise stated.
Discussion
We investigated the long‐term risk of HF in patients with a first‐time diagnosis of AF without other cardiovascular risk factors and comorbidities prospectively over a 30 year time period. This nationwide case–control study of 227 811 patients with AF and 452 712 controls in Sweden showed that individuals with a first‐time diagnosis of AF and without traditional cardiovascular risk factors at inclusion had a more than three‐fold increased risk for HF onset over a mean observation time of 9.1 years. Furthermore, this HF risk within the first year in AF patients younger than 50 years was more than 100 times greater than that in non‐AF controls. The HR compared with controls was lower with increasing age and longer follow‐up time. Inclusion of other non‐cardiovascular coexisting conditions such as cancer and COPD in multivariable models showed similar results. There were distinct sex differences in incidence rate of HF throughout all age categories. Furthermore, when compared with controls, women with AF had a higher relative risk for HF onset. Notably, women aged 18–34 years with AF had 24 times higher risk for HF onset, whereas men in the same age category had only nine‐fold increased risk in relation to their same age‐group controls.
As presented in the current European Society of Cardiology guidelines for management of AF, left ventricular dysfunction and HF are known AF‐related outcomes with frequency between 20% and 30%, which is congruent with our findings that 27% of patients developed HF during the study period. 1 Previous studies evaluating risk factors for HF have generally been conducted among patients with numerous coexisting conditions. A study by LaMori et al. showed that up to 36% of patients had at least one risk factor according to the CHA2DS2‐VASc score. 15 In the present study, nearly three‐fourths of patients in the registry (73.5%) had cardiovascular risk factors and comorbidities. Those AF patients were excluded in order to study the impact attributable solely to AF.
The complex relationship between AF and HF is still not well understood; different mechanisms have been proposed and data on possible causality are still lacking. 16 HF may be a primary underlying cause of AF, and these two conditions often coexist and affect each other. Frustaci et al. investigated 14 young patients with ‘lone’ AF refractory to initial medical therapy. 17 After diagnostic workup including myocardial biopsy, the investigators confirmed evidence of myocarditis, diffuse cardiomyopathic process, and non‐specific diffuse fibrotic changes that were seen even in other patients with supraventricular tachycardia and were not specific for AF alone. However, patients with previously diagnosed myocarditis were excluded from our analysis, as were patients with pre‐existing cardiomyopathies. Coexisting arrythmias, including AF, are common in patients with CHD and valve disease, 18 , 19 patients who also were excluded from our analysis. We also excluded patients with renal failure because renal failure and HF also often coexist because of their shared pathophysiological mechanisms. 20 In summary, the results of risk assessment in this study present an estimation of the attributable risk of ‘lone’ AF for HF onset.
Most previous studies that investigated the risk of HF onset indicate that women have higher risk, but several studies have shown neutral results between the sexes. However, those studies included different types of patients with AF and did not consider the effect of comorbidities or age on the risk for HF onset. Additionally, these patients were older, and the observation time in most of these reports was less than a third of that in this study. We found that the incidence rate per 1000 person‐years for new‐onset HF, however, was higher in men with AF than in women per age categories and increased with increasing age. In contrast, the HR compared with controls was higher among women, which could be explained by a lower rate of HF among the controls in women. Because this was a population with an otherwise low cardiovascular risk profile, the results of our study highlight the contribution of AF as a major factor in the development of HF. 10
This study also showed that the incidence rate per 1000 person‐years for HF onset was highest in the eldest patients with AF. However, the excess risk for HF onset was highest among younger patients with AF. The HR when compared with controls without AF was also highest within the first year after AF diagnosis and then decreased over time. A recently published study from Sweden 21 showed that 24.5% of patients with new‐onset HF aged <55 years had AF at baseline, whereas only 0.7% of the age‐paired controls had AF, from which can be derived an approximate risk ratio of 35, which is congruent with the results of this study in the age category 50–59 years (Supporting Information, Table S3 ).
Women with HF have more pronounced symptoms and signs of volume overload, 22 , 23 , 24 , 25 which often is the cause of hospitalization but also highlights the importance of female sex as an important factor in the contribution of AF to new‐onset HF. Nonetheless, the excess risk for HF onset was highest in women 18–34 years of age, and it was 24 times higher (vs. nine times higher in men) when compared with the general population.
Among non‐cardiovascular risk factors in this population, COPD/asthma and cancer were the most common in this population. COPD occurs in approximately 20% of patients with HF and has an important impact on both the symptoms and the outcomes of patients living with HF. In addition, optimal treatment of COPD may improve cardiac function. 26 Furthermore, HF may occur in patients with cancer as a result of the cancer itself, as a result of the cancer therapy, or even as a result of patient's cardiovascular background. 27 Accordingly, in our Model 3, we adjusted for these conditions but without significant effect on the excess risk besides in women (Table 2 ). Regardless, only a few women were in the youngest age category, which is important to note when interpreting these results.
Strengths and limitations
One major strength of this study is its very large study population and nearly 100% long‐term follow‐up. 13 We included all patients hospitalized and treated for AF at specialized outpatient clinics in Sweden and compared them with controls from the general population without AF. Besides, the proportion of patients at risk for HF onset after AF diagnosis in this study was congruent with previously reported data. 1 Consequently, due to the size and integrity of the studied population, this study provides generalizable results. By excluding patients with established cardiovascular disease and risk factors and correlating for several non‐cardiovascular comorbidities at baseline, we minimized the risk of confounding factors. Thereby, we were able to estimate the attributable risk of AF for new‐onset HF in relation to controls from the general population without AF. However, the results of this study should be interpreted carefully and in regard to inclusion and exclusion criteria. Furthermore, validation of AF diagnosis in the NPR was 97% 13 and HF diagnosis of patients treated in cardiology wards was 91% and internal medicine wards was 86%. 28
This study also had several limitations. First, this was a registry‐based study and the analyses performed were defined by the design of the registries; therefore, we were not able to consider potentially important clinical data such as patient symptoms, laboratory data, physical activity, smoking, hyperlipidaemia, or staging of different comorbidities. Other risk factors and comorbidities are largely diagnosed and treated in primary care rather than hospitals, and therefore, they might not be accounted for in this study. Second, we did not adjust for time‐updated comorbidities. Regardless, the highest risk compared with the general population was during the first year and decreased over the observation time, more pronouncedly in patients than in controls, indirectly suggesting a minor contribution of those factors in development of HF in this otherwise population with low cardiovascular risk profile. Third, we do not have data from outpatient settings up to 2001 and may have missed some patients with milder forms of HF diagnosed in primary care settings or outpatient clinics from 1987 to 2001, suggesting that we may have underestimated the magnitude of the total risk for HF onset in this population. However, because most patients with AF who develop HF at some point are in contact with a cardiologist in hospital settings, we believe this possibility should not have significantly influenced our results.
Conclusions
This nationwide case–control study of AF patients without traditional cardiovascular risk factors showed a three‐fold higher long‐term risk for HF onset compared with controls from the general population. Younger women (18–34 years old) had 24 times the risk for developing HF, whereas men in the same age range had a nine‐fold increased risk in relation to controls. Within the first year of diagnosis, AF patients younger than 50 years of age had a more than 100 times increased risk of HF compared with age‐paired non‐AF controls.
Clinical implication
The long‐term risk for HF onset is substantial even in a population of patients with newly diagnosed AF and otherwise low cardiovascular risk profile, and risk‐score tools to predict and prevent HF onset in these patients are needed. The excess risk related to female sex should be considered, particularly among younger patients who otherwise could be considered healthy without traditional cardiovascular risk factors or coexisting conditions. The high risk of developing HF in young women of reproductive age is also of great relevance and important to preventing complications related to pregnancy.
Conflict of interest
The authors declare no conflicts of interest.
Funding
This study was financed by grants received from the Swedish Research Council (Vetenskapsrådet) (Grant Number 2019‐00193) and from the Swedish state under the agreement between the Swedish government and the County Councils (Health and Medical Care Committee of the Regional Executive Board, Region Västra Götaland) (ALFGBG‐721351 and ALFGBG‐965023).
Supporting information
Table S1. International Classification of Diseases (ICD)‐9 codes and ICD‐10 codes used to identify comorbidities in the National Patient Register.
Table S2. ICD codes with descriptions used to detect medical records with cardiomyopathy diagnoses.
Table S3. Risk for heart failure in patients diagnosed with atrial fibrillation without other cardiovascular comorbidities or risk factors within 1 year and 5 years when compared to controls from the general population in Sweden.
Figure S1. Study population: Patients identified in the National Patient Register with a first‐time diagnosis of atrial fibrillation and without major cardiovascular conditions prior to or at the time of the index registration.
Abbreviations: AF: atrial fibrillation; CHD: congenital heart disease; HF: heart failure; VHD: valve heart disease. CV risk profile: Patients with, e.g., peripheral artery disease, myocardial infarction, aortic plaque, hypertension, diabetes mellitus, prior stroke, transient ischemic attack, thromboembolism, or vascular disease (e.g., peripheral artery disease, myocardial infarction).
Figure S2. Patients identified in the National Patient Register registered with a first‐time diagnosis of atrial fibrillation and matched controls from the Total Population Register in Sweden from the period 1987 to 2017 by age category.
Figure S3. Time to heart failure diagnosis after index registration for atrial fibrillation within 1 year in patients and controls by age category. Data from the National Patient Register and Total Population Register in Sweden.
Figure S4. Time to heart failure diagnosis after index registration for atrial fibrillation within 5 years in patients and controls by sex. Data from the National Patient Register and Total Population Register in Sweden.
Acknowledgements
We thank John Daniel from Edanz (www.edanz.com/ac) for editing a draft of this manuscript.
Basic, C. , Hansson, P.‐O. , Sandström, T. Z. , Johansson, B. , Fu, M. , and Mandalenakis, Z. (2023) Heart failure outcomes in low‐risk patients with atrial fibrillation: a case–control study of 680 523 Swedish individuals. ESC Heart Failure, 10: 2281–2289. 10.1002/ehf2.14375.
References
- 1. Hindricks G, Potpara T, Dagres N, Arbelo E, Bax JJ, Blomström‐Lundqvist C, Boriani G, Castella M, Dan GA, Dilaveris PE, Fauchier L, Filippatos G, Kalman JM, la Meir M, Lane DA, Lebeau JP, Lettino M, Lip GYH, Pinto FJ, Thomas GN, Valgimigli M, van Gelder I, van Putte B, Watkins CL, ESC Scientific Document Group . 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio‐Thoracic Surgery (EACTS): the Task Force for the diagnosis and management of atrial fibrillation of the European Society of Cardiology (ESC) developed with the special contribution of the European Heart Rhythm Association (EHRA) of the ESC. Eur Heart J. 2021; 42: 373–498. [DOI] [PubMed] [Google Scholar]
- 2. Benjamin EJ, Muntner P, Alonso A, Bittencourt MS, Callaway CW, Carson AP, Chamberlain AM, Chang AR, Cheng S, das S, Delling FN, Djousse L, Elkind MSV, Ferguson JF, Fornage M, Jordan LC, Khan SS, Kissela BM, Knutson KL, Kwan TW, Lackland DT, Lewis TT, Lichtman JH, Longenecker CT, Loop MS, Lutsey PL, Martin SS, Matsushita K, Moran AE, Mussolino ME, O'Flaherty M, Pandey A, Perak AM, Rosamond WD, Roth GA, Sampson UKA, Satou GM, Schroeder EB, Shah SH, Spartano NL, Stokes A, Tirschwell DL, Tsao CW, Turakhia MP, VanWagner L, Wilkins JT, Wong SS, Virani SS, American Heart Association Council on Epidemiology and Prevention Statistics Committee and Stroke Statistics Subcommittee . Heart disease and stroke statistics—2019 update: a report from the American Heart Association. Circulation. 2019; 139: e56–e528. [DOI] [PubMed] [Google Scholar]
- 3. Staerk L, Sherer JA, Ko D, Benjamin EJ, Helm RH. Atrial fibrillation: epidemiology, pathophysiology, and clinical outcomes. Circ Res. 2017; 120: 1501–1517. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Ziff OJ, Carter PR, McGowan J, Uppal H, Chandran S, Russell S, Bainey KR, Potluri R. The interplay between atrial fibrillation and heart failure on long‐term mortality and length of stay: insights from the, United Kingdom ACALM registry. Int J Cardiol. 2018; 252: 117–121. [DOI] [PubMed] [Google Scholar]
- 5. Lam CSP, Yancy C. Universal definition and classification of heart failure: is it universal? Does it define heart failure? J Card Fail. 2021; 27: 509–511. [DOI] [PubMed] [Google Scholar]
- 6. Xanthopoulos A, Giamouzis G, Skoularigis J, Triposkiadis F. Heart failure with reduced, mildly reduced, or preserved left ventricular ejection fraction: has reasoning been lost? World J Cardiol. 2022; 14: 438–445. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Sugumar H, Nanayakkara S, Prabhu S, Voskoboinik A, Kaye DM, Ling LH, Kistler PM. Pathophysiology of atrial fibrillation and heart failure: dangerous interactions. Cardiol Clin. 2019; 37: 131–138. [DOI] [PubMed] [Google Scholar]
- 8. Meluzin J, Starek Z, Kulik T, Jez J, Lehar F, Wolf J, Dusek L, Leinveber P, Novak M. Prevalence and predictors of early heart failure with preserved ejection fraction in patients with paroxysmal atrial fibrillation. J Card Fail. 2017; 23: 558–562. [DOI] [PubMed] [Google Scholar]
- 9. Lip GY, Nieuwlaat R, Pisters R, Lane DA, Crijns HJ. Refining clinical risk stratification for predicting stroke and thromboembolism in atrial fibrillation using a novel risk factor‐based approach: the euro heart survey on atrial fibrillation. Chest. 2010; 137: 263–272. [DOI] [PubMed] [Google Scholar]
- 10. McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M, Burri H, Butler J, Čelutkienė J, Chioncel O, Cleland JGF, Coats AJS, Crespo‐Leiro MG, Farmakis D, Gilard M, Heymans S, Hoes AW, Jaarsma T, Jankowska EA, Lainscak M, Lam CSP, Lyon AR, McMurray JJV, Mebazaa A, Mindham R, Muneretto C, Francesco Piepoli M, Price S, Rosano GMC, Ruschitzka F, Kathrine Skibelund A, ESC Scientific Document Group , de Boer RA, Christian Schulze P, Abdelhamid M, Aboyans V, Adamopoulos S, Anker SD, Arbelo E, Asteggiano R, Bauersachs J, Bayes‐Genis A, Borger MA, Budts W, Cikes M, Damman K, Delgado V, Dendale P, Dilaveris P, Drexel H, Ezekowitz J, Falk V, Fauchier L, Filippatos G, Fraser A, Frey N, Gale CP, Gustafsson F, Harris J, Iung B, Janssens S, Jessup M, Konradi A, Kotecha D, Lambrinou E, Lancellotti P, Landmesser U, Leclercq C, Lewis BS, Leyva F, Linhart A, Løchen ML, Lund LH, Mancini D, Masip J, Milicic D, Mueller C, Nef H, Nielsen JC, Neubeck L, Noutsias M, Petersen SE, Sonia Petronio A, Ponikowski P, Prescott E, Rakisheva A, Richter DJ, Schlyakhto E, Seferovic P, Senni M, Sitges M, Sousa‐Uva M, Tocchetti CG, Touyz RM, Tschoepe C, Waltenberger J, Adamo M, Baumbach A, Böhm M, Burri H, Čelutkienė J, Chioncel O, Cleland JGF, Coats AJS, Crespo‐Leiro MG, Farmakis D, Gardner RS, Gilard M, Heymans S, Hoes AW, Jaarsma T, Jankowska EA, Lainscak M, Lam CSP, Lyon AR, McMurray JJV, Mebazaa A, Mindham R, Muneretto C, Piepoli MF, Price S, Rosano GMC, Ruschitzka F, Skibelund AK. 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J. 2021; 42: 3599–3726. [DOI] [PubMed] [Google Scholar]
- 11. Ludvigsson JF, Otterblad‐Olausson P, Pettersson BU, Ekbom A. The Swedish personal identity number: possibilities and pitfalls in healthcare and medical research. Eur J Epidemiol. 2009; 24: 659–667. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Ludvigsson JF, Almqvist C, Bonamy AK, Ljung R, Michaëlsson K, Neovius M, Stephansson O, Ye W. Registers of the Swedish total population and their use in medical research. Eur J Epidemiol. 2016; 31: 125–136. [DOI] [PubMed] [Google Scholar]
- 13. Ludvigsson JF, Andersson E, Ekbom A, Feychting M, Kim JL, Reuterwall C, Heurgren M, Olausson PO. External review and validation of the Swedish national inpatient register. BMC Public Health. 2011; 11: 450. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Brooke HL, Talbäck M, Hörnblad J, Johansson LA, Ludvigsson JF, Druid H, Feychting M, Ljung R. The Swedish cause of death register. Eur J Epidemiol. 2017; 32: 765–773. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. LaMori JC, Mody SH, Gross HJ, daCosta DiBonaventura M, Patel AA, Schein JR, Nelson WW. Burden of comorbidities among patients with atrial fibrillation. Ther Adv Cardiovasc Dis. 2013; 7: 53–62. [DOI] [PubMed] [Google Scholar]
- 16. Wijesurendra RS, Casadei B. Mechanisms of atrial fibrillation. Heart. 2019; 105: 1860–1867. [DOI] [PubMed] [Google Scholar]
- 17. Frustaci A, Chimenti C, Bellocci F, Morgante E, Russo MA, Maseri A. Histological substrate of atrial biopsies in patients with lone atrial fibrillation. Circulation. 1997; 96: 1180–1184. [DOI] [PubMed] [Google Scholar]
- 18. Hernández‐Madrid A, Paul T, Abrams D, Aziz PF, Blom NA, Chen J, Chessa M, Combes N, Dagres N, Diller G, Ernst S, Giamberti A, Hebe J, Janousek J, Kriebel T, Moltedo J, Moreno J, Peinado R, Pison L, Rosenthal E, Skinner JR, Zeppenfeld K, ESC Scientific Document Group . Arrhythmias in congenital heart disease: a position paper of the European Heart Rhythm Association (EHRA), Association for European Paediatric and Congenital Cardiology (AEPC), and the European Society of Cardiology (ESC) Working Group on Grown‐up Congenital heart disease, endorsed by HRS, PACES, APHRS, and SOLAECE. Europace. 2018; 20: 1719–1753. [DOI] [PubMed] [Google Scholar]
- 19. Kline J, Costantini O. Arrhythmias in congenital heart disease. Med Clin North Am. 2019; 103: 945–956. [DOI] [PubMed] [Google Scholar]
- 20. Schefold JC, Filippatos G, Hasenfuss G, Anker SD, von Haehling S. Heart failure and kidney dysfunction: epidemiology, mechanisms and management. Nat Rev Nephrol. 2016; 12: 610–623. [DOI] [PubMed] [Google Scholar]
- 21. Basic C, Rosengren A, Alehagen U, Dahlström U, Edner M, Fu M, Novak M, Zverkova Sandström T, Schaufelberger M. Young patients with heart failure: clinical characteristics and outcomes. Data from the Swedish Heart Failure, National Patient, Population and Cause of Death Registers. Eur J Heart Fail. 2020; 22: 1125–1132. [DOI] [PubMed] [Google Scholar]
- 22. Martínez‐Sellés M, Doughty RN, Poppe K, Whalley GA, Earle N, Tribouilloy C, McMurray J, Swedberg K, Køber L, Berry C, Squire I, Meta‐Analysis Global Group In Chronic Heart Failure (MAGGIC) . Gender and survival in patients with heart failure: interactions with diabetes and aetiology. Results from the MAGGIC individual patient meta‐analysis. Eur J Heart Fail. 2012; 14: 473–479. [DOI] [PubMed] [Google Scholar]
- 23. Lam CSP, Arnott C, Beale AL, Chandramouli C, Hilfiker‐Kleiner D, Kaye DM, Ky B, Santema BT, Sliwa K, Voors AA. Sex differences in heart failure. Eur Heart J. 2019; 40: 3859–3868c. [DOI] [PubMed] [Google Scholar]
- 24. Lam CS, Carson PE, Anand IS, Rector TS, Kuskowski M, Komajda M, McKelvie R, McMurray J, Zile MR, Massie BM, Kitzman DW. Sex differences in clinical characteristics and outcomes in elderly patients with heart failure and preserved ejection fraction: the Irbesartan in Heart Failure with Preserved Ejection Fraction (I‐PRESERVE) trial. Circ Heart Fail. 2012; 5: 571–578. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Galvao M, Kalman J, DeMarco T, Fonarow GC, Galvin C, Ghali JK, Moskowitz RM. Gender differences in in‐hospital management and outcomes in patients with decompensated heart failure: analysis from the Acute Decompensated Heart Failure National Registry (ADHERE). J Card Fail. 2006; 12: 100–107. [DOI] [PubMed] [Google Scholar]
- 26. Hohlfeld JM, Vogel‐Claussen J, Biller H, Berliner D, Berschneider K, Tillmann HC, Hiltl S, Bauersachs J, Welte T. Effect of lung deflation with indacaterol plus glycopyrronium on ventricular filling in patients with hyperinflation and COPD (CLAIM): a double‐blind, randomised, crossover, placebo‐controlled, single‐centre trial. Lancet Respir Med. 2018; 6: 368–378. [DOI] [PubMed] [Google Scholar]
- 27. Zamorano JL, Lancellotti P, Rodriguez Muñoz D, Aboyans V, Asteggiano R, Galderisi M, Habib G, Lenihan DJ, Lip GYH, Lyon AR, Lopez Fernandez T, Mohty D, Piepoli MF, Tamargo J, Torbicki A, Suter TM, ESC Scientific Document Group . 2016 ESC Position Paper on cancer treatments and cardiovascular toxicity developed under the auspices of the ESC Committee for Practice Guidelines: the Task Force for cancer treatments and cardiovascular toxicity of the European Society of Cardiology (ESC). Eur Heart J. 2016; 37: 2768–2801. [DOI] [PubMed] [Google Scholar]
- 28. Ingelsson E, Arnlöv J, Sundström J, Lind L. The validity of a diagnosis of heart failure in a hospital discharge register. Eur J Heart Fail. 2005; 7: 787–791. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1. International Classification of Diseases (ICD)‐9 codes and ICD‐10 codes used to identify comorbidities in the National Patient Register.
Table S2. ICD codes with descriptions used to detect medical records with cardiomyopathy diagnoses.
Table S3. Risk for heart failure in patients diagnosed with atrial fibrillation without other cardiovascular comorbidities or risk factors within 1 year and 5 years when compared to controls from the general population in Sweden.
Figure S1. Study population: Patients identified in the National Patient Register with a first‐time diagnosis of atrial fibrillation and without major cardiovascular conditions prior to or at the time of the index registration.
Abbreviations: AF: atrial fibrillation; CHD: congenital heart disease; HF: heart failure; VHD: valve heart disease. CV risk profile: Patients with, e.g., peripheral artery disease, myocardial infarction, aortic plaque, hypertension, diabetes mellitus, prior stroke, transient ischemic attack, thromboembolism, or vascular disease (e.g., peripheral artery disease, myocardial infarction).
Figure S2. Patients identified in the National Patient Register registered with a first‐time diagnosis of atrial fibrillation and matched controls from the Total Population Register in Sweden from the period 1987 to 2017 by age category.
Figure S3. Time to heart failure diagnosis after index registration for atrial fibrillation within 1 year in patients and controls by age category. Data from the National Patient Register and Total Population Register in Sweden.
Figure S4. Time to heart failure diagnosis after index registration for atrial fibrillation within 5 years in patients and controls by sex. Data from the National Patient Register and Total Population Register in Sweden.