Abstract
Objectives. To study mortality rates among men and women with atrial fibrillation (AF) and concomitant chronic heart failure (CHF) prescribed different classes of cardiovascular drugs in primary health care. Design. A cohort of men (n = 1159) and women (n = 1155) aged 45 years or above and diagnosed with both AF and CHF from patient records from 75 primary care centers in Sweden were included in the study. Regression models with mortality as the outcome were used, with adjustment for a propensity score comprising age, cardiovascular co-morbidities, education, marital status, and pharmacotherapy. We analysed using Cox regression with hazard ratio (HR), and Laplace regression with years until 10% of the patients had died, with 95% confidence intervals (95% CI). Independent variables were prescribed cardiovascular drugs. Results. Individuals prescribed anticoagulants versus no treatment gained 1.95 years (95% CI 0.47–3.43), anticoagulants versus antiplatelets 1.26 years (95% CI 0.42–2.10), calcium channel blockers 1.17 years (95% CI 0.21–2.14), and statins 1.49 years (95% CI 0.39–2.59). Among patients 80 years or above no significant effect by anticoagulants was seen, HR 0.73 (95% CI 0.43–1.23). Conclusions. Our findings suggest that life may be prolonged in patients with AF and concomitant CHF in primary care prescribed anticoagulants, calcium channel blockers, and statins.
Introduction
Atrial fibrillation (AF) is the most common arrhythmia in the population with an estimated prevalence of 2% in Sweden (Citation1), and almost 3% in individuals above 20 years (Citation1,Citation2). The concomitant presence of chronic heart failure (CHF) in AF patients has been shown to increase mortality (Citation3).
Different aspects of AF vary in men and women. AF seems to be more common among men, accounting for 56% of all AF patients in Stockholm County (Citation1), and men also develop AF on average five years earlier than women (Citation4). Women with AF, on the other hand, exert both a higher relative risk of stroke and of mortality than men with AF (Citation5).
Anticoagulant therapy is preferred to antiplatelet therapy due to the superiority in stroke prevention (Citation6,Citation7), as the risk of stroke is markedly increased in AF patients. However, when looking at all-cause mortality, no difference between antiplatelets and anticoagulants was found in a Cochrane review based on results from randomized trials (Citation7). In patients with AF, concomitant diastolic CHF increases the risk of death and stroke (Citation8), which is why treatment with non-vitamin K oral anticoagulants is preferred in this patient group.
The primary aim of the present study was to compare the effects of antiplatelets and anticoagulants on mortality, also by sex and age groups in concomitant AF and CHF. A further aim was to understand the effects on mortality of cardiovascular drugs that are commonly prescribed to these patients.
Methods
Design
The study used individual-level patient data from 75 primary health care centers (PHCCs), 48 of which were located in Stockholm County. Individuals attending any of the participating PHCCs between 2001 and 2008 were included in the study. We used Extractor software (http://www.slso.sll.se/SLPOtemplates/SLPOPage1____10400.aspx; accessed September 19, 2010) to extract individual electronic patient records (EPRs). National identification numbers were replaced with new unique serial numbers to ensure anonymity. The files were linked to a database constructed using the Total Population Register, the Inpatient Register and the Swedish Cause of Death Register, which contains individual-level data on age, gender, education, and hospital admissions for all residents registered in Sweden. Thus, a new research database containing clinical data and information on socioeconomic status on the individuals (n = 1,098,420) registered at the 75 PHCCs was created. Data from the Cause of Death Register were used for the follow-up.
Ethical approvals were obtained from regional boards at Karolinska Institutet and the University of Lund.
Study population
All persons diagnosed with both AF and CHF who attended the 75 PHCCs between January 1st 2001 and December 31st 2007 were included in this study. They were identified by the presence of the ICD-10 (10th version of the WHO's International Classification of Diseases) codes for AF (I48) and CHF (I50 and I110) in the medical records. In total, 2,314 individuals, 1,159 men and 1,155 women, aged 45 years or above were identified and included.
Outcome variable
Time to death in the period from registration of AF diagnoses and December 31st 2007.
Pharmacotherapies
Drugs prescribed during the assessment period were recorded by Anatomic Therapeutic Chemical Classification. With regard to anti-thrombotic drugs (B01A), patients were divided into three groups: no treatment, anti-platelet treatment (B01AC), including acetylsalicylic acid (ASA; B01AC6, B01AC30), and anticoagulant treatment (B01AA) (with or without anti-platelet treatment). Diuretic drugs (C03) were recorded as thiazides or related agents, and were also registered when in combination with other drugs (C03A, C03B, C03E, C09B, and C09DA), loop diuretics (C03C) or potassium-saving diuretics (C03D), including aldosterone antagonists (C03DA). Furthermore, the following cardiovascular agents were recorded: heart-active drugs (C01), beta blockers (C07), calcium channel blockers (C08), and RAS-blocking agents (C09). Lipid-lowering drugs (C10A), including statins (C10AA), were also recorded.
Demographic and socioeconomic variables
Gender: Men and women.
Individuals were divided into the following age groups 45–54, 55–64, 65–74, 75–84 and > 85 years. Individuals younger than 45 years were excluded.
Educational level was categorized as 9 years or below (partial or complete compulsory schooling), 10–12 years (partial or complete secondary schooling), and above 12 years (college and/or university studies).
Marital status was classified as married, unmarried, divorced, or widowed.
Co-morbidities
We identified the following cardiovascular co- morbidities from the EPRs among individuals in the study population: hypertension (I10–15), coronary heart disease (CHD; I20–25), non-rheumatic valvular diseases (I34–38), cardiomyopathy (I42), cerebrovascular diseases (CVDs; I60–69), including intracranial bleeding (I60–62), diabetes mellitus (E10–14), and hypothyroidism (E03).
Statistical analyses
Differences in means and distributions between men and women were compared using Student's t-test, chi-square analysis, and Fisher's exact test. Age- adjustment for the prescribed drugs was performed using logistic regression.
Follow-up analyses were performed, firstly using Cox regression with hazard ratios (HRs) and 95% confidence interval (95% CI), using time to death as the outcome. Secondly, Laplace regression was used to calculate the difference in years until death for the first10% of the participants prescribed pharmacotherapies versus those without these particular drugs (Citation9). Because different distributions and mathematical calculations are used to obtain results in Cox and Laplace regression, respectively, putting emphasis on findings significant with both methods may reduce the risk of chance findings. The regression models were also tested for possible interactions between sex and each pharmacotherapy, but no such interaction was found.
Based on a preliminary analysis stratified at below 80 and 80 years or above of age, all prescribed cardiovascular drug classes were tested separately using 80 years as a cut-off. The cut-off for age 80 years has commonly been referred to as a relevant cut-off age for stroke prophylaxis (Citation10), as well as a marker of increased prevalence and incidence of AF (Citation11). Furthermore, the death rate was too low in women aged less than 75 years in the present study (n = 18) to provide reliable results.
Antithrombotic drugs (antiplatelets or anticoagulants), digitalis, diuretics (thiazides, loop diuretics, and aldosterone antagonists), beta blockers, RAS-blocking agents, calcium channel blockers and statins were used as main outcome factors in separate analyses. A unique propensity score was constructed for each drug class to balance all other cardiovascular drug classes, as well as co-morbidities (hypertension, CHD, CVDs, and diabetes), age, sex (in sex-adjusted models), and socioeconomic factors (educational level and marital status) (Citation12). We excluded specific anti-arrhythmic agents, potassium-saving agents other than aldosterone antagonists and lipid-lowering agents other than statins due to their low prescription rates. We excluded valvular diseases and cardiomyopathy from the analyses due to the low numbers of cases and uneven distributions.
A p-value for two-sided tests of less than 0.01 was considered statistically significant due to the multiple comparisons between men and women. A two-sided p-value of less than 0.05 was considered statistically significant for variables in the Cox regression and Laplace regression analyses. All analyses were performed in STATA 11.2, with an amendment for Laplace regression provided by Professor Bottai (Citation9).
Results
Characteristics of the study population (n = 2,314) are shown separately for men (n = 1,159) and women (n = 1,155) in . A total of 214 men (18.5%) and 241 women (20.9%) died during follow-up. Overall, men were significantly younger and had more education than did women, while significantly more women were widowed than were men. Hypertension was more common among women, while cardiomyopathy was more common among men. The mean follow-up time was 3.8 years (standard deviation [SD] 2.1), and HRs were calculated based on 8,793 person-years at risk.
Table I. Data for patients aged 45 years or above with diagnoses of AF and concomitant chronic heart failure (n = 2,314) in primary care attending the 75 PHCCs between January 1st 2001 and December 31st 2007.
shows rates of prescription of pharmacotherapies in men and women, with p-values for crude and age-adjustment shown. When considering age-adjusted p-values, greater proportions of men were prescribed RAS-blocking agents and ACE inhibitors, and greater proportions of women were prescribed digitalis, diuretics in general and also thiazides, potassium-saving diuretics and aldactone antagonists, and beta blockers in general and also beta1 -selective agents in general.
Table II. Prescription data for patients aged 45 years or above with diagnoses of AF and concomitant chronic heart failure (n = 2,314) in primary care attending the 75 PHCCs between January 1st 2001 and December 31st 2007.
Cox regression models for subjects stratified by age into two groups, below 80 years and 80 years or above, are shown in , and Laplace regression models in . Results were statistically significant with decreased risk in Cox and increased survival in Laplace regression models for anticoagulants versus no treatment and versus antiplatelets, both for all subjects and those below 80 years of age, for calcium channel blockers for all subjects and for statins for all subjects and for those below 80 years of age. For calcium channel blockers, a decreased risk in Cox regression model was also found among those 80 years or above of age, and increased survival in Laplace regression models for RAS-blocking agents for all subjects and for statins also for those 80 years or above of age.
Table III. Cox regression models for patients (n = 2,314) aged 45–104 years with AF and concomitant chronic heart failure attending the 75 PHCCs between January 1st 2001 and December 31st 2007, stratified by age.
Table IV. Laplace regression with years to first 10% mortality models for patients (n = 2,314) aged 45–104 years with AF and concomitant chronic heart failure attending the 75 PHCCs between January 1st 2001 and December 31st 2007, stratified by age.
We also tested heart-active and vessel-active calcium channel blockers separately. For heart-active agents Cox regression model showed HR 0.44 (95% CI 0.21–0.95) and Laplace regression model + 2.24 years (95% CI 0.90–3.57), and vessel-active agents, Cox regression model showed HR 0.76 (95% CI 0.53–1.10) and Laplace regression model + 0.90 years (95% CI 0.07; 1.73).
Discussion
The main finding of this study was a decreased mortality associated with anticoagulants both in relation to no treatment and to antiplatelets. Decreased mortality was also found in individuals prescribed calcium channel blockers and statins. The effects of anticoagulants were seen only among those younger than 80 years of age. In contrast, positive effects of calcium channel blockers were seen only among the elderly.
One might have expected an overall decreased mortality associated with many of the prescribed pharmacotherapies, especially RAS-blockers and beta blockers. However, the results in this observational study could be influenced by different factors, and competing risk factors could contribute to mortality especially among elderly individuals. AF seems to predict mortality in CHF patients only among those below 75 years of age (Citation13), and the decreased mortality owing to a more effective treatment of heart failure also seems to be less pronounced among elderly (Citation14). AF as a comorbid condition could also be a marker of frailty, with a worse metabolic profile and higher mortality (Citation15).
One finding was the lack of significant effect by anticoagulants on all-cause mortality among individuals aged 80 years or above, which actually is in line with the earlier findings. In the BAFTA study, which was performed among patients aged 75 years or above, a superiority of warfarin over ASA in the reduction of stroke (yearly RR 0.48) and major vascular events (RR 0.73) was shown (Citation16). However, all-cause mortality was not affected when comparing anticoagulants with ASA (RR 0.95, 95% CI 0.72–1.26). Furthermore, a Cochrane review based on randomized trials found no difference on all-cause mortality in AF patients between anticoagulants and antiplatelets (OR 0.99, 95% CI 0.83 to 1.18) (Citation7). However, we have earlier shown that there were no consistent association between CHADS2 scores and anticoagulant prescriptions in the present total primary care AF cohort, which of course may have affected our results (Citation17).
The effect by calcium channel blockers is also of interest. We have no good explanation to the positive finding of these agents. In contrast, no certain significant effect was found by beta blockers and RAS-blocking agents, both recommended in treatment of CHF, according our criteria, that is, a significant effect in both Cox and Laplace regression analysis (Citation18). Anyway, a significant effect was seen by RAS-blocking agents in the Laplace regression model, and beta blockers were borderline significant (p-value for the Cox regression model was 0.10, and for the Laplace regression model 0.053). Results of calcium channel blockers in relation to heart failure in earlier studies have been conflicting; one review reported a reduced rate of CHF in individuals prescribed antihypertensive agents (Citation19), but yet another review reported an increased risk of incident CHF (Citation20). Calcium channel blockers may, however, be beneficial in patients with diastolic CHF (Citation21). This type of CHF is regarded to be more common among elderly, especially women, and also to be more common in primary care. The decreased mortality associated with calcium channel blockers also seemed to be more associated with heart-active agents. Our findings may of course be confounding by indication, that is, individuals with a more severe CHF, or systolic CHF, may be prescribed RAS-blockers and/or beta blockers, why individuals with a less severe CHF, or diastolic CHF, may be prescribed calcium channel blockers. The prescription rate of calcium channel blockers was considerably lower than that of RAS-blockers and of beta blockers.
The reduced mortality associated with prescription of statins in line with results from some earlier clinical studies of CHF patients, concluding that statins were independently and significantly associated with lower mortality (Citation22), also among older subjects (Citation23). The effect of statins on AF in individuals with CHF is however uncertain, with controversial results in different studies (Citation24). Besides, it cannot be excluded that the positive finding for survival could be due to confounding by indication, that is, severely ill patients not being prescribed statins to the same extent.
There are several limitations of this study which must be kept in mind when interpreting the results. The study sample is a special group, that is, patients with concomitant diagnoses of AF and CHF registered in primary health care. In another study was found, that 36% of all registered AF patients in Stockholm County were not registered with a diagnosis in primary health care (Citation1). We do not know the coverage of patients with concomitant diagnoses of AF and CHF, where some patients may not being diagnosed, and other might be cared for in other forms of health care. Besides, we do not know the validity of registered diagnoses in primary health care, and there might be both over- and under-estimation of diagnosis of especially CHF. Results cannot be generalized to AF or CHF patients in general, or to patients in other settings. Another limitation is the study design, that is, a retrospective observational study. The findings may have been subject to confounding by indication and to survival bias (Citation25). All these mentioned factors could have affected the results and yielded discrepant findings. Yet another limitation of this study was the small sample size, which should be kept in mind when interpreting results in sub-samples of the present study. Besides, the use of propensity scores seems to produce treatment effect estimates of more extreme magnitudes compared to RCTs (Citation26). Furthermore, drugs prescribed by other caregivers were not included in the patient records, which may have weakened the associations between prescription of certain medications and mortality in this study. We did not have access to doses of the prescribed drugs, which is especially of importance for diuretic drugs. Furthermore, length of drug treatment was not recorded. Severity of CHF and CHD was not classified in the patient records, even if it is probable that these patients from primary care had less severe CHF than patients from hospital care, and besides, probably more often diastolic CHF. As severity of CHF is an important factor for mortality this is also a major limitation of the study. Besides, data on ejection fraction and criteria for diagnosis of CHF were not available. Moreover, AF could not be classified as paroxysmal, persistent or permanent and heart rhythm could not be classified as sinus rhythm or fibrillation rhythm. Additionally, we had not accessed renal function.
A major strength of this study was that we were able to link clinical data from individual EPRs to data from national demographic and socioeconomic registers with less than 1% of information missing. While many previous follow-up studies of AF have used hospital data, the current study used data from primary care, which may better reflect the risks associated with AF in society. Moreover, randomized controlled trials often exclude individuals with co-morbidities, such as AF patients with concomitant diabetes and CHF. In the current study, we had the possibility to include these patients in the analyses, which means that the findings are more representative of the variety of patients encountered in clinical practice today. An additional strength of this study was the use of a propensity score with covariates relevant to mortality, as it enabled adjustment with a low risk of over-fitting regression models (Citation12).
In conclusion, our findings suggest that life may be prolonged in patients in patients with AF and CHF in primary care prescribed anticoagulants, calcium channel blockers, and statins. However, given the retrospective design of this study and the possible confounding by indication, our findings must be interpreted with caution. We found a lack of positive effect of anticoagulants on survival among subjects aged 80 years or above, which could imply that the effect on total mortality of anticoagulants in elderly patients with AF and CHF could be overestimated. More studies focusing on AF patients from the primary care setting also including data on severity of CHF and on ejection fraction are needed to confirm or reject our findings.
Acknowledgments
This work was supported by ALF funding awarded to Jan Sundquist and Kristina Sundquist and by grants from the Swedish Research Council (awarded to Kristina Sundquist), the Swedish Freemasons Foundation (Jan Sundquist), and the Swedish Council for Working Life and Social Research (Jan Sundquist).
Declaration of interest: The authors report no declarations of interest. The authors alone are responsible for the content and writing of the paper.
References
- Forslund T, Wettermark B, Wandell P, von Euler M, Hasselstrom J, Hjemdahl P. Risk scoring and thromboprophylactic treatment of patients with atrial fibrillation with and without access to primary healthcare data: experience from the Stockholm health care system. Int J Cardiol. 2013;170:208–14.
- Friberg L, Bergfeldt L. Atrial fibrillation prevalence revisited. J Intern Med. 2013;274:461–8.
- Seiler J, Stevenson WG. Atrial fibrillation in congestive heart failure. Cardiol Rev. 2010;18:38–50.
- Humphries KH, Kerr CR, Connolly SJ, Klein G, Boone JA, Green M, et al. New-onset atrial fibrillation: sex differences in presentation, treatment, and outcome. Circulation. 2001; 103:2365–70.
- Michelena HI, Powell BD, Brady PA, Friedman PA, Ezekowitz MD. Gender in atrial fibrillation: Ten years later. Gend Med. 2010;7:206–17.
- Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med. 2007;146:857–67.
- Aguilar MI, Hart R, Pearce LA. Oral anticoagulants versus antiplatelet therapy for preventing stroke in patients with non-valvular atrial fibrillation and no history of stroke or transient ischemic attacks. Cochrane Database Syst Rev. 2007:CD006186.
- Jang SJ, Kim MS, Park HJ, Han S, Kang DH, Song JK, et al. Impact of heart failure with normal ejection fraction on the occurrence of ischaemic stroke in patients with atrial fibrillation. Heart. 2013;99:17–21.
- Bottai M, Zhang J. Laplace regression with censored data. Biom J. 2010;52:487–503.
- Friberg L, Hammar N, Ringh M, Pettersson H, Rosenqvist M. Stroke prophylaxis in atrial fibrillation: who gets it and who does not? Report from the Stockholm Cohort-study on Atrial Fibrillation (SCAF-study). Eur Heart J. 2006;27:1954–64.
- Fuster V, Ryden LE, Cannom DS, Crijns HJ, Curtis AB, Ellenbogen KA, et al. ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation: full text: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines and the European Society of Cardiology Committee for Practice Guidelines (Writing Committee to Revise the 2001 guidelines for the management of patients with atrial fibrillation) developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society. Europace. 2006;8:651–745.
- Senn S, Graf E, Caputo A. Stratification for the propensity score compared with linear regression techniques to assess the effect of treatment or exposure. Stat Med. 2007;26: 5529–44.
- Baldasseroni S, Orso F, Fabbri G, De Bernardi A, Cirrincione V, Gonzini L, et al. Age-dependent prognostic significance of atrial fibrillation in outpatients with chronic heart failure: data from the Italian Network on Congestive Heart Failure Registry. Cardiology. 2010;116:79–88.
- Al Suwaidi J, Asaad N, Al-Qahtani A, El-Menyar A, Al-Mulla AW, Singh R, Albinali HA. Effect of age on outcome on patients hospitalized with heart failure: from a 20-year registry in a middle-eastern country (1991–2010). Congest Heart Fail. 2012;18:320–6.
- Fumagalli S, Tarantini F, Guarducci L, Pozzi C, Pepe G, Boncinelli L, et al. Atrial fibrillation is a possible marker of frailty in hospitalized patients: results of the GIFA Study. Aging Clin Exp Res. 2010;22:129–33.
- Mant J, Hobbs FD, Fletcher K, Roalfe A, Fitzmaurice D, Lip GY, et al. Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomised controlled trial. Lancet. 2007;370:493–503.
- Carlsson AC, Wandell P, Sundquist K, Johansson SE, Sundquist J. Differences and time trends in drug treatment of atrial fibrillation in men and women and doctors’ adherence to warfarin therapy recommendations: a Swedish study of prescribed drugs in primary care in 2002 and 2007. Eur J Clin Pharmacol. 2013;69:245–53.
- McMurray JJ, Adamopoulos S, Anker SD, Auricchio A, Bohm M, Dickstein K, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The Task Force for the Diagnosis and Treatment of Acute and Chronic Heart Failure 2012 of the European Society of Cardiology. Developed in collaboration with the Heart Failure Association (HFA) of the ESC. Eur Heart J. 2012;33:1787–847.
- Costanzo P, Perrone-Filardi P, Petretta M, Marciano C, Vassallo E, Gargiulo P, et al. Calcium channel blockers and cardiovascular outcomes: a meta-analysis of 175,634 patients. J Hypertens. 2009;27:1136–51.
- Shibata MC, Leon H, Chatterley T, Dorgan M, Vandermeer B. Do calcium channel blockers increase the diagnosis of heart failure in patients with hypertension? Am J Cardiol. 2010;106:228–35.
- Aziz F, Tk LA, Enweluzo C, Dutta S, Zaeem M. Diastolic heart failure: a concise review. J Clin Med Res. 2013;5: 327–34.
- Gastelurrutia P, Lupon J, de Antonio M, Urrutia A, Diez C, Coll R, et al. Statins in heart failure: the paradox between large randomized clinical trials and real life. Mayo Clin Proc. 2012;87:555–60.
- Komajda M, Hanon O, Hochadel M, Lopez-Sendon JL, Follath F, Ponikowski P, et al. Contemporary management of octogenarians hospitalized for heart failure in Europe: Euro Heart Failure Survey II. Eur Heart J. 2009;30:478–86.
- Goette A. Upstream therapy for atrial fibrillation in heart failure. Heart Fail Clin. 2013;9:417–25.
- Delgado-Rodriguez M, Llorca J. Bias. J Epidemiol Community Health. 2004;58:635–41.
- Dahabreh IJ, Sheldrick RC, Paulus JK, Chung M, Varvarigou V, Jafri H, et al. Do observational studies using propensity score methods agree with randomized trials? A systematic comparison of studies on acute coronary syndromes. Eur Heart J. 2012;33:1893–901.