Heart & Lung 44 (2015) 189e198

Contents lists available at ScienceDirect

Heart & Lung journal homepage: www.heartandlung.org

Gender differences in stroke, mortality, and hospitalization among patients with atrial fibrillation: A systematic review Deborah W. Chapa, PhD, ACNP-BC a, *, Bimbola Akintade, PhD, ACNP-BC, MBA, MHA b, Sue A. Thomas, RN, PhD b, Erika Friedmann, PhD b a b

George Washington University, School of Nursing, 2030 M Street NW, Suite 300, Washington, DC 20036, USA University of Maryland, Baltimore, School of Nursing, 655 West Lombard Street, Baltimore, MD 21201, USA

a r t i c l e i n f o

a b s t r a c t

Article history: Received 22 May 2014 Received in revised form 15 January 2015 Accepted 16 January 2015 Available online 19 February 2015

Objective: To conduct a systematic review of gender differences in stroke, mortality, and hospitalization for patients with atrial fibrillation and/or flutter (AF/Afl). Methods: Full texts, published, peer-reviewed, English language articles from 1999 through July 2014 were examined. Articles with populations of patients with AF/Afl were included if they conducted longitudinal analysis of any of three outcomes: stroke, mortality, and hospitalization, and reported or compared at least one of the outcomes according to gender. Results: Seventeen articles were included: sixteen on stroke, nine on mortality, and one on hospitalization. In nine articles women had more strokes (RRs 0.89e1.9). Findings about gender differences in mortality (RRs 0.69e2.8) and hospitalizations were equivocal. Conclusions: Few articles examine differences in outcomes between men and women with AF/Afl. Given the prevalence of AF/Afl and health care costs it is vital to determine gender differences to evaluate appropriate therapies to decrease stroke, mortality, and hospitalizations. Ó 2015 Elsevier Inc. All rights reserved.

Keywords: Atrial fibrillation Stroke Systematic review Survival Female

Introduction Atrial fibrillation (AF) and atrial flutter (Afl) are major global health problems that are increasing in prevalence and affect more than 3 million Americans.1,2 Atrial fibrillation is an atrial arrhythmia that results in rapid randomized contractions of small areas of the atrial myocardium, thereby causing an irregular, and often rapid, ventricular rate. In Afl the atria beat regularly and faster than the ventricles. Atrial flutter frequently precedes the onset of AF. These arrhythmias are associated with an increased risk of stroke, dementia, heart failure (HF), and mortality.3e6 Estimated annual costs of managing AF/Afl is $6.65 billion, which includes $4.88 billion in hospitalization expenses and $1.53 billion in outpatient management costs.7 Atrial fibrillation/Afl are primarily diseases of the elderly with the prevalence doubling with each decade of life after the age of 60 years and occurring in 11% of the population over 80 years old.2,8 The high prevalence in the older population is anticipated to increase the burden of AF on the health-care system with an estimation of a 2.5 fold increase in patients with AF over the next 50 years.9 Prevalence rates of AF are increasing not only

* Corresponding author. E-mail address: [email protected] (D.W. Chapa). 0147-9563/$ e see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.hrtlng.2015.01.008

because of an increase in the older population but also increasing incidence of AF. The reason for the increased incidence/prevalence is unclear and is postulated due to increased risk factors of diabetes and obesity, increased diagnosis of AF, and improvements in management of hypertension and stroke.10 The incidence of AF in men is slightly higher in all age groups. The lifetime risk for developing AF for men and women over 55 is 23.8% for men and 22.2% for women.11 Though a great deal of attention is focused on gender differences in other cardiac diseases including coronary artery disease (CAD), much less attention is focused on gender differences in AF, the most common sustained cardiac arrhythmia.12 Although many individuals with CAD develop AF/Afl they frequently exist independently of each other. In CAD, different presentation and management strategies are proposed for women and men. Women present more often than men with angina, have higher resting heart rates, higher systolic heart rates, and are less likely to have undergone cardiac investigations.13 Women are more likely to receive calcium channel blockers but fewer angiotensin converting enzymes (ACE) inhibitors, and fewer lipid lowering drugs than men. However, women remain at higher risk for subsequent adverse events.13 Women at high-risk for acute coronary syndrome (ACS) experience less coronary angiography, angioplasty and coronary artery bypass surgery compared to men. They do not have higher incidence of cardiovascular death,

190

D.W. Chapa et al. / Heart & Lung 44 (2015) 189e198

recurrent myocardial (MI) or stroke but do have increased numbers of refractory ischemia and rehospitalizations.14 Women who have percutaneous coronary intervention for non-ST elevation ACS have increased risk of major adverse events compared to men.15 The extensive research on gender differences in CAD has revealed significant clinical findings. The gender differences is AF/Afl have not been carefully examined. In a multicenter study of 150 patients with AF/Afl, female gender emerged as an indicator of poor health related quality of life (HRQOL) for patients with AF/Afl.16 In another study evaluating outcomes related to gender, limited-use data from the National Institutes of Health/National Heart, Lung, and Blood Institute Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) clinical trial provided a sample of 693 patients with AF, 262 women and 431 men.17 Clinical status predicted mortality in women. However, for men, HRQOL and clinical status predicted mortality. Both clinical status and HRQOL predicted hospitalizations for men and women.17 The goal of this paper was to evaluate differences according to gender in predictors of stroke, mortality, and hospitalization in patients with AF/Afl. If differences in predictors according to gender are present in these outcomes, further evaluating these differences may lead to reduced costs, improved function or quality of life of patients with AF/Afl. If there are differences between men and women they will have important implications for disease management. Methods

after the AFFIRM study which set standards for treatment.4 Search selection strategies were conducted in a stepwise fashion: two reviewers independently examined all titles for inclusion criteria. Consensus was reached and abstracts reviewed. Finally, selected full text articles were randomly assigned to three reviewers. Each article was examined by two reviewers. Bibliographies of full-text articles were searched to locate additional articles and eight were found. Data extraction Two reviewers who were part of the research team with both clinical and methodological expertise extracted data from each study to complete detailed evidence tables. Data were rechecked against the original articles for accuracy. If discrepancies were discovered, these were discussed by the team, resolved and corrected. Quality assessment The Qualitative Assessment Tool for Quantitative Studies (QATQS) was used to rate study quality.20 This tool evaluates study bias in eight areas including: sample selection, study design, confounding, blinding, data collection methods, withdrawals and dropouts, intervention integrity and analysis. The first six areas are scored and result in an overall methodological rating of study quality as strong, moderate, or weak. The QATQS is well validated for use in public health nursing and randomized and nonrandomized trials.21,22 Two reviewers independently rated the quality with an agreement rate of 100%.

Design Article evaluation A systematic review was conducted guided by processes recommended by the Evidence Based Practice Centers funded by the Agency for Healthcare Research and Quality. Processes were developed to identify and select relevant articles, review and rate the individual articles, and then synthesize results and grade the evidence. No meta-analysis was planned as considerable heterogeneity across articles was anticipated with regard to patient samples, definitions of outcomes, length of follow-up, and settings.

The results of all analyses included in each article that were directly relevant to the specified outcomes were summarized along with study details and presented in tables for each outcome. An additional table (Table 1) was created that used the raw numbers of male and female participants and the raw number of participants of each gender who experienced each outcome to create an unadjusted risk ratio. This provides a standard metric to facilitate comparison of all articles within and between outcomes.

Literature search Results A literature search was conducted as recommended by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement18,19 Study eligibility criteria were established a priori. Inclusion criteria were full text, published, peer-reviewed, English language and human articles originating from any country. Articles about populations with AF/Afl were included if they conducted longitudinal analysis of three outcomes: stroke, mortality, and hospitalization reported or compared at least one of the outcomes according to gender, and included only participants over 18 years old. Case studies, qualitative studies, cross-sectional studies, dissertations or reviews were excluded. Studies of interventions or outcomes for caregivers or providers of patients with AF/Afl also were excluded. Search strategies were adapted from Cochrane16 and the National Institute for Health and Clinical Excellence17 protocols to systematically search MEDLINE (OVID), CINAHL (EBSCOHost), and Cochrane Libraries of Systematic Reviews and Clinical Trials (OVID) databases from 1999 through July, 2014. The searches were designed for high sensitivity to locate any study of patients with AF/ Afl that examined gender as a predictor of stroke, mortality, or hospitalization or provided data about the rates of these outcomes for the two genders separately. Limiting the search to the last 15 years was chosen to capture a timely body of research that occurred

The literature searches of the databases located 489 records and examination of references within the articles evaluated located eight additional articles (Fig. 1). Ninety four articles were duplicates. Title evaluation eliminated 235 articles. Abstract evaluation eliminated 120 articles. Full text review eliminated 31 articles. Two articles23,24 included the same sample. Only the article that focused directly on gender differences was retained.24 The final list included 17 articles.12,24e39 Ten of them included multiple outcomes.12,24e27,33,35e38 Sixteen articles included the outcome of stroke,12,24,26e39 nine articles included the outcome of mortality,24e27,33,35e38 and one article included the outcome of hospitalization.12 Specific numbers of women and men could not be ascertained from one article which included 229,530 individuals with AF/Afl and provided age and gender standardized rates but did not provide raw data about rates according to gender.12 The sixteen remaining articles included 122,795 women and 128,010 men with AF/Afl. Stroke Sixteen articles either included data that provided differences between men and women in rates of stroke or provided the data in

D.W. Chapa et al. / Heart & Lung 44 (2015) 189e198

191

Table 1 Summary of numbers of women and men and number of outcome events and average age and the risk ratio for women to men for each outcome (mortality, stroke, hospitalization) in each article. Author citation Stroke Avgil Tsadok28 Cha29 Dagres27 Fang (ATRIA)31 Friberg33 Friberg34 Gomberg (SPORTIF)24 Humphries12 Inoue (J-Rhythm)36 Melgaard37

Poli32 Poli35 Rienstra (RACE)26 Salam38 Shen30 Wang39 Mortality Asbach25 Dagres27 Friberg33 Gomberg (SPORTIF)24 Inoue (J-Rhythm)36 Melgaard37

Poli35 Rienstra (RACE)26 Salam38 Hospitalization Humphries12

Men N

Men age mean,
or (SD)

Men N outcome

Women N

Women age mean
(SD)

Women N outcome

Risk ratio (F:M)

95% CI

39,398 458 3084 8850 12,830 50,135 5072 NA 5241 47,453 5585 (30e50 yo) 20,536 (50e65 yo) 21,358 (>65 yo) 423 1361 330 2432 10,730 462

77.2 (72.2e2.4) NR 64
13 58.6% > 69 67
8.4 74.7
13.5 69.8
9 NR 69
10 NR

1734 NA 37 NA 360 6625 112 NA 92 NA

80.2 (74.8e5.4) NR 70
12 73.9% > 69 69
6.8 80.9
10.6 73.4
8 NR 73
9 NR

2559 NA 49 NA 240 8095 72 NA 34 NA

1.32

1.24

1.90

1.82

1.19

2.77

7.8 19.0 1.45

5.8 17.0 1.08

1.43 20.1 1.949

1.12 0.1.23 (30e50 yo) 0.89 (50e65 yo) 0.98 (>65 yo)

0.75 0.58 0.80 0.91

1.67 1.09 0.98 1.05

NR NR 67
9 54.5
15.7

NR NR 71
8 59
15

1.01 0.80 0.56

3.42 1.8 1.97

NR

20 67 13 6 NA 47

1.86 1.20 1.020

NR

19 45 22 9 NA 64

44,115 237 2249 6828 16,204 50,667 2257 NA 2165 26,849 2348 (30e50 yo) 9313 (50e65 yo) 14,659 > 65 yo) 239 1654 192 1417 8131 406

1.73

1.16

2.59

886 3084 12,830 5072 5241 47,453 5585 (30e50 yo) 20,536 (50e65yo) 21,358 (>65 yo) 1361 330 2431

NR 64
13 67
8.4 69.8
9 69
10 NR

NA 169 1395 301 152 NA

NR 70
12 69
6.8 73.4
8 73
9 NR

NA 115 815 93 53 NA

0.9294 2.8 0.6943 1.474 0.85 (30e50 yo) 0.74 (50e65 yo) 0.84 (>65 yo)

0.74 2.0 0.55 1.05 0.68 0.08 0.80

1.17 4.0 0.87 2.08 1.09 0.79 0.87

NR 67
9 54.5
15.7

125 23 97

741 2249 16,204 2257 2165 26,849 2348 (30e50 yo) 9313 (50e65 yo) 14,659 > 65 yo) 1654 192 1417

NR 71
8 59
15

157 13 63

1.0 0.96

0.80 0.50

1.80 1.85

NA

NR

NA

NA

NR

NA

ATRIA ¼ AnTicoagulation and Risk Factors in Atrial Fibrillation, NA ¼ Not available; data were reported as age adjusted differences only, EPICA ¼ Prevalence of heart failure in Southwestern Europe, J-Rhythm ¼ Registry of Japanese patients with atrial fibrillation, NR ¼ Not reported as means RACE ¼ Rate control versus Electrical cardioversion for persistent atrial fibrillation, SD ¼ standard deviation, SPORTIF ¼ Stroke Prevention using ORal Thrombin Inhibitor in atrial Fibrillation, yo ¼ years old All articles were of moderate quality as determined by the QATOS rating.20

a way that these rates could be computed (Tables 1 and 2).12,24,26e39 All articles were of moderate quality determined by the QATOS rating.20 One study provided only age/gender standardized rates.12 The remaining fifteen articles included a total of 136,355 women and 140,344 men.24,26e39 Among women strokes ranged from 0.438 to 8.4%,32 and among men strokes ranged from 0.438 to 6.7%26 (Table 1). Stroke risk ratios for women ranged from 1.026 to 1.9.32 In nine articles12,24,27,28,31e34,39 women had significantly more strokes than men. In all articles that provided separate data for men and women, strokes occurred more frequently in women (Table 1). In five24,27,28,31,33 of the nine articles with women having higher stroke rates than men, the women were older than men and had greater numbers of stroke risk factors at baseline (Tables 1 and 2). In the other four articles12,32,34,39 comparisons were not presented and there was no evidence of exclusion of participants due to risk factors. The sixteen articles about stroke in patients with AF/Afl classified stroke outcomes as unspecified stroke in five articles,12,24,27,28,39 thromboembolic stroke in two articles,26,29 ischemic stroke in two articles,32,34 hemorrhagic stroke in two articles,30,35 combinations of ischemic, thromboembolic and hemorrhagic in four articles,24,33,34,36,37 and one classified as in hospital stroke.38 In all of the articles reporting stroke without classifying type,12,24,27,28,39 and by classifying as ischemic stroke31,32 women

had significantly higher incidence of stroke than men. In the articles reporting stroke as thromboembolic events,26,29 as hemorrhagic events30,35 and as in hospital stroke38 there were no differences between men and women. In the four articles classifying stroke as combinations of ischemic, thromboembolic or hemorrhagic, in three of the articles women had significantly higher incidence of stroke.24,31,33 One of these articles31 found the increased rate for thromboembolic stroke. In one article women in the 50e65 age group were significantly less likely to have strokes than men,37 and in the other article men in the hemorrhagic group were more likely to experience stroke.36 One clinical trial examined gender differences in stroke rates among subgroups of patients randomized to different anticoagulants.24 There was no difference in stroke rates between men and women who had been randomly assigned to Vitamin K antagonist, but women had more strokes than men in those assigned to direct thrombin inhibitor.24 Notably, there were no differences in mean age or in average number of stroke risk factors according to gender in this study.24 Mortality Nine articles24e27,33,35e38 either included data that provided differences between men and women in mortality rates or

192

D.W. Chapa et al. / Heart & Lung 44 (2015) 189e198

Fig. 1. Flow diagram: Records reviewed to examine gender differences in stroke, mortality, or hospitalization of patients with atrial fibrillation.

provided the data in a way that mortality rates could be computed. Table 3 summarizes each article according to information about the study population, follow up time comparisons made between genders in incidence of mortality and information about variables that were controlled in these comparisons. Table 1 provides raw numbers of participants and outcome events as well as average age and unadjusted mortality risk ratios according to gender for each article. All articles were of moderate quality determined by the QATOS rating.20 Eight articles included a total of 52,987 women and 77,802 men.24,26,27,33,35e38 One article that provided mortality data for men and women did not provide numbers of participants according to gender, only age adjusted mortality rates.25 Among women mortality ranged from 0.533 to 7.0%26 and among men mortality ranged from 1.133 to 7.0%.26 Mortality risk ratios for women ranged from 0.6924 to 2.833 (Table 1). A total of seven articles reported the contribution of gender to all-cause mortality (Table 3). In three articles,24,25,36 men had significantly greater all-cause mortality than women; and in one36 men had higher mortality in the 66e75 year old age group but not in younger age groups (30e50, 50e65 years). In three articles there was no significant difference in all-cause mortality27,35,38; in two of these articles27,38 all-cause mortality was adjusted for age. A total of three articles reported gender differences in cardiovascular morality (Table 3). In all articles there was no significant difference in cardiovascular mortality.26,33,36

Hospitalization Only one article12 included the outcome of hospitalization. This large cohort study included 229,530 individuals with AF/Afl and did not provide specific numbers of women and men (Table 4). A hospitalization risk ratio was not provided. The rate of hospitalization for AF/Afl was higher for men than for women. Discussion Overall the evidence from nine of the 16 articles of moderate quality (determined by the QATOS rating)20 supports women with AF/Afl having higher risk of stroke than men.12,24,27,28,31e34,39 Women with AF may have lower mortality than men after controlling for age and other risk factors; however evidence is equivocal at this time. There is insufficient information to compare rates of hospitalization of individuals with AF/Afl according to gender. The results of this review provide evidence suggesting that among patients with AF/Afl, women are at a greater risk for stroke. In five24,27,28,31,33 articles women had significantly higher stroke rates than men and the women were older than the men ranging from 2 to 6 years (Table 1) and had a greater number of stroke risk factors such as diabetes, hypertension, chronic renal disease, cardiovascular disease, heart failure at baseline. In the other four articles12,32,34,39 comparisons were not presented and there was no

Table 2 Summary of the articles that compared stroke rates of women and men with atrial fibrillation and or atrial flutter. Outcome

Population

Avgil Tsadok28 Canada cohort

Stroke (Cerebral thrombosis, embolism, artery occlusion)

Cha29 Korea cohort

Thromboembolism

Recently diagnosed AF Patients  65 years old and men rate of stroke (2.2% vs. 1.2% in men, p ¼ 0.011)

Women had a significantly higher rate of stroke (OR ¼ 1.83, 95% CI ¼ 1.10e3.03, p ¼ 0.019)

Women > men HR ¼ 2.6 (2.2e9.2)

Thromboembolism off warfarin: women > men (3.5% versus 1.8%; adjusted rate ratio [RR], 1.6; 95% CI, 1.3e1.9). Above 75 yo: women > men RR 1.8 ci: 1.4e2.3; men RR 1.6 ci: 1.0e2.3 Major hemorrhage: no difference women HR 0.8 (0.6e1.1) Women significantly greater HR ¼ 2.6 (1.3e2.4)

Women > men HR 1.47 (1.40e1.54)

Women (overall): HR 1.18 (1.12e1.24) Women: Age < 65 yo HR ¼ 1.10 (0.86e1.41), 65e74 yo HR ¼ 1.11 (0.97e1.27), 75 yo HR ¼ 1.23 (1.17e1.30)

Age, prior stroke, hypertension, congestive heart failure, coronary artery disease, diabetes mellitus, and estrogen use

Age

35 cofactors Age, hx of AF, thyroid, frequent falls, dementia, pulmonary embolism, ischemic stroke, other thromboembolism, anemia, HTN, HF, intracranial bleeding, other major bleeding, pacemaker, diabetes, MI, another vascular disease, cancer within 3 yrs, emphysema or COPD, renal disease, platelect defects or coagulopathy, liver disease, alcohol index, CHADS2, CHA2DS2-VASc, aspirin, clopidogrel only, aspirin and clopidogrel, beta blockers, class-1 Antiarrhythmic, class-3 antiarrhythmic, verapamil or dilatizem, calcium blocker dihydropyridine, digoxin, diuretics, ACE inhibitors/receptor blockers, statin

D.W. Chapa et al. / Heart & Lung 44 (2015) 189e198

Author location study design

(continued on next page)

193

194

Table 2 (continued ) Outcome

Population

Gomberg24 Europe, Australia, United States, Asia, Canada RCT

Ischemic/ Hemorrhagic

Patients with persistent or paroxysmal non-valvular AF and one additional risk factor for stroke

Humphries12 Canada cohort

Stroke

229,530 Patients 20e106 years old discharged with 1st AF/Afl from Canadian health System records

Inoue36 Japan cohort

Thromboembolism Major hemorrhagic event

Consecutive patients with AF at outpatient clinics in J-Rhythm registry

Stroke (ischemic, Melgaard37 Denmark cohort embolic, hemorrhagic)

Patients with AF discharged from hospitals listed in the Danish Patient registry from January 1, 2000 to December 31, 2011

Poli32 Italy cohort

Ischemic stroke with VKA

Patients referred with AF for the control of OAT to one university hospital anticoagulation clinic

Poli35 Italy cohort

Hemorrhagic stroke with VKA

Rienstra26 Netherlands RCT

Thromboembolism

AF patients who started VKA treatment after the age of 80 years from 27centers affiliated to the Italian Federation of anticoagulation Clinics and participated in the EPICA study Patients in RACE conducted in study with recurrent persistent AF.

Sample N 7329

7406

Female N%

Follow-up time

2257 (31%) 31 months

2165 (29%) 2 years

773,999 26,417 (36%) 5 years

780

275 (35%) 2.28 years Range 1.9 to 3.1

3015 1654

522

192 (37)

2.5
2.1 years

2e3 years

Unadjusted results for gender

Adjusted results for gender

Covariates in adjusted results

Stroke or SEE: Women > men 2.08%/year, 95% confidence interval (CI) 1.60e2.56%/year vs. 1.44%/ year, 95% CI 1.18e1.71%/ year in men p ¼ 0.016

Direct thrombin inhibitor group: women > men (2.17%/year, 95% CI 1.48e2.87%/year vs. 1.38%/year, 95% CI 1.00e1.75%/year; p ¼ 0.03) VKA group: no difference women (1.98%/year) to men (1.51%/year).

Anticoagulant drug: direct thrombin inhibitor or VKA

Stroke significantly greater in women Women 3.2% men 2.4% in 1 year c2 ¼ 148.16, p < 0.0001. Thromboembolism: No difference OR ¼ 1.24 (0.83e1.86) p ¼ 0.30 Major hemorrhagic event men > women OR ¼ 1.59 (1.05 e2.40) p ¼ 0.03

Age, diabetes at index hospitalization, stroke/TIA at index hospitalization, and stroke/TIA in previous year CHADS2 score, use of VKA, antiplatelet drugs CHADS2 score, use of VKA, antiplatelet drugs

Thromboembolism: No difference Men 1.8%, women 1.6% OR ¼ 1.12 (0.75e1.67 p ¼ 0.58 Major hemorrhagic event no significant difference men 2.1%, women 1.4% OR ¼ 1.46 (0.98e2.19) p ¼ 0.06 Not significant women: Age 30e50 years HR 1.23 (0.90 e1.88); women significantly lower age 50 e65 years HR ¼ 0.89 (0.80 e0.98); Not significant age 65e75 years HR ¼ 0.98 (0.91e1.05) 19 men (Rate 1.2e95%CI 0.7 e2.4 per 100 pt/yrs) and 21 women (rate 2.43e95% CI 1.4e3.6 per 100 pt/yrs) females > Males RR ¼ 2.0 (95% CI 1.3e3.1) p < 0.01 Major bleeding no significant Stroke/TIA: No significant difference men OR ¼ 1.5 (0.9e2.3) difference women OR 1.2 p ¼ 0.1 (0.8e1.8) p ¼ 0.3 Major bleeding Men > women OR 1.6 (1.1 e2.3) p ¼ 0.006 No difference women 6.8%, men 6.7% Absolute difference with 90% CI 0.1 (3.8e4.0)

Use of VKA

Major bleeding: Previous stroke/ TIA, previous bleeding, vascular disease, active cancer, serum creatinine > 1.5 mg/dL, history of falls

D.W. Chapa et al. / Heart & Lung 44 (2015) 189e198

Author location study design

ACE ¼ angiotensin converting enzyme, AF ¼ Atrial fibrillation, Afl ¼ Atrial flutter, CHA2DS2VASC ¼ Age, Sex, Congestive Heart Failure, Hypertension, Stroke, TIA or Thromboembolism, Vascular Disease stroke risk scheme for AF, CHADS2 ¼ Congestive heart failure, Hypertension, Age, Diabetes mellitus, Prior Stroke or TIA stroke risk scheme for AF, CI-confidence interval stroke risk scheme for AF, COPD ¼ chronic obstructive pulmonary disease, CV ¼ cardiovascular, DM ¼ Diabetes, EKG ¼ electrocardiogram, EPICA ¼ Prevalence of heart failure in Southwestern Europe, HTN ¼ hypertension, J-Rhythm ¼ Registry of Japanese patients with atrial fibrillation, RACE ¼ Rate Control versus Electrical Cardioversion for Persistent Atrial Fibrillation Study, RCT ¼ randomized control trial, HR ¼ Hazard ratio, SBP ¼ systolic blood pressure, LV ¼ left ventricular, OR ¼ odds ratio, OAT ¼ Occluded artery treatment, RR ¼ rate ratio, TIA ¼ transient ischemic attack, SEE ¼ Systemic embolic event, VKA ¼ Vitamin K antagonist.

Women > HR ¼ 1.34 (1.06e1.71) 406 (47%) Mean 4 years Members of the Framingham study cohort diagnosed with AF/Afl Stroke

868

8137 (43%) 3.35 years Intracranial hemorrhage

Shen30 United States Cohort Wang39 USA cohort

18,867

Registry of 20 years of patients hospitalized with AF in one hospital First time hospitalization for Afl no prior stroke (California) In hospital stroke Salam38 Qatar cohort

3849

1417 (37%) Mean 4.2 days

No significant difference strokes women 0.4% men 0.4%, p ¼ 0.80

Women HR: 0.86 CI: 0.63e1.17 p ¼ 0.33

Age, ethnic group, hypertension, DM, HF, and percent time treated with vitamin k antagonist Age, SBP, DM, prior stroke or TIA

D.W. Chapa et al. / Heart & Lung 44 (2015) 189e198

195

evidence of exclusions of participants due to risk factors. Articles26,29,35e38 that reported no difference or higher rates of stroke in men than women specifically selected for lower risk patients. The stroke risk scheme for AF (CHADS2) score represents a clinical prediction for stroke in patients with non-rheumatic AF.40 Patients in one study29 had CHADS2 scores of 0 or 1 on a scale of 0e6, indicating low risk for stroke. The other study30 included only patients hospitalized for the first time with AF or Afl and excluded patients with comorbidities that would increase stroke risk including history of stroke or TIA, rheumatic heart disease, and mitral valve replacement. The differences in stroke rates between men and women could be attributed partially to differences in baseline characteristics. However, in the adjusted analyses that controlled for stroke risk factors and age the disparity in risk of stroke among women persists.12,27,28,31,33 Anticoagulation therapy for women with AF is critical. In Europe, female gender is currently recognized as an independent risk factor for stroke as evidenced by current European Society of Cardiology guidelines that recommend a more complete risk evaluation scale (CHA2DS2-VASc).41 Anti-coagulation therapy is recommended for patients with CHA2DS2-VASc scores of 2 or higher. Women receive a score of 1 in this scale as female gender is considered an independent risk factor for stroke. Other risk factors given one point on this scale are congestive heart failure, hypertension, diabetes, vascular disease, age 65e74. The risk factors given two points on the CHA2DS2-VASc scale are age greater than 75 and prior stroke.41 Therefore, it is key for providers to evaluate and prescribe anticoagulation therapy for women. An additional factor that may increase women’s risk of stroke is disparities in use of ablation therapy. Fewer women are referred for catheter ablation for AF than men and more women than men fail ablation (31.5% vs. 22.5%). Women also were referred for ablation later in the course of their AF (6.51 vs. 4.85 years) and they were older when referred.42 These disparities could make a large contribution to women’s stroke rates. The nine articles24e27,33,35e38 examining predictors of mortality in AF included largely male populations, despite the larger proportions of women among the older age cohorts in which AF is most likely to occur.24e27 Women were older than men in all articles that reported mean age for women and men (Table 1). One study examined separate age cohorts and found differences in mortality only in the oldest (65 years and older) age group.37 Women also had more risk factors (higher heart failure and hypertension but less diabetes) than men in six24,26,27,33,36,38 of the nine articles about mortality. Men may have experienced higher mortality because in most age groups women outlive men.43 There is no evidence to support differences in cardiovascular mortality between the genders. Additional multivariate data are needed to evaluate differences in mortality according to gender in patients with AF while taking into account differences in age and other risk factors for these outcomes. The nine articles are insufficient to evaluate appropriate predictors of all-cause mortality or cardiovascular morality. Additional research is therefore required to fill this gap in knowledge. Longitudinal studies including the roles of hormones, genetics, and valvular heart disease may elucidate causes of differences in mortality between men and women. Only one large clinical trial12 examined rates of hospitalization for AF/Afl. This study also included more men than women. The rate of one year hospitalization for AF/Afl was higher for men than for women. In patients with AF/Afl, male gender, age, coronary artery disease, stroke, and heart failure predicted low health related quality of life and living alone predicted increased hospitalizations for men with AF/Afl.12 More studies are required to determine the impact and reasons for increased hospitalizations in men so that interventions for prevention can be facilitated.

196

Table 3 Summary of the articles that compared mortality rates of women and men with atrial fibrillation. Outcome

Population

Sample Female N N%

Asbach25 Germany cohort

Mortality

Patients with AF diagnosed prior to pacemaker insertions followed at one department of cardiology

1627

Dagres27 Europe cohort

Mortality

5333

Friberg33 Denmark cohort

CV Mortality

Gomberg24 Europe Australia United States, Asia Canada RCT

All-cause mortality

Inoue36 Japan cohort

All-cause Mortality CV mortality

Patients with AF with EKG or Holter within 12 months either hospitalized or not hospitalized Patients from 182 centers from 35 European Society of Cardiology countries. Individuals with AF in a random sample of Copenhagen residents obtained in 1976e1978, 1981-9e1983, 1991e1994 Patients with persistent or paroxysmal non-valvular AF and one additional risk factor for stroke: Hypertension, age > or equal to 75 years, previous stroke, TIA or systemic embolic event, LV dysfunction LV ejection fraction or þ65 years with CAD or age >þ65 and diabetes. Consecutive patients with AF at outpatient clinics in J-Rhythm registry

Melgaard37 Denmark cohort

Mortality

Patients with AF age 30e75 years discharged from hospitals listed in the Danish Patient registry from January 1, 2000 to December 31, 2011

Poli35 Italy cohort

All-cause Mortality with VKA

Rienstra26 NetherlandsRCT

CV Mortality

Elderly patients with AF who started VKA treatment after the age of 80 years from 27centers affiliated to the Italian Federation of anticoagulation Clinics that participated in the EPICA study Patients with recurrent persistent AF in RACE study

Salam38 Qatar cohort

In hospital Mortality Patients hospitalized with AF in one hospital’s 20 year registry

276

7329

7406

73,999

3015

522

3849

Follow-up time Unadjusted results for gender

741 (45%) Up to 35 years

2257 (31%) 31 months

Age and stroke risk

All-cause: men significantly higher OR ¼ 1.78 (1.25e2.50) p < 0.002 CV: no significant difference men OR ¼ 0.96 (0.56e1.56) p ¼ 0.893

CHADS2 score, use of VKA or direct thrombin inhibitor drugs CHADS2 score, use of VKA or direct thrombin inhibitor drugs

No significant difference male OR ¼ 1.12 (0.77e1.65), p ¼ 0.55

Age, DM, HTN, chronic renal impairment, cardiogenic shock

Men > women; men 301 deaths over 7831 patient years (3.84%/ year, 95% CI 3.41e4.28%/year) women: vs. 93 deaths over 3507 patient-years (2.71%/year, 95% CI 2.16e3.25%/year; p ¼ 0.002)

All-cause: men > women: men ¼ 2.9%, women ¼ 2.0%, OR ¼ 1.45 (1.05e2.08) p ¼ 0.026 CV: no difference; men:0.0%, women 1% OR ¼ 0.90 (0.55e1.55) p ¼ 0.764 26,417 (36%) 5 years Women no difference age 30e50 years, HR ¼ 0.86 (0.68e1.09); 50e65 years, HR ¼ 0.74 (0.68e0.79); Women < men age 65e75 years HR ¼ 0.84 (0.80e0.87) 1654 (55%) 2.5
2.1 years No difference RR 1.00 (0.8e1.3) p ¼ 0.70

1417 (37%) Mean 4.2 days

No significant differences in mortality (5% in women vs. 5% in men, p ¼ 0.567)

No significant difference women HR ¼ 1.1 (0.6e1.8)

2165 (29%) 2 years

192 (37%) 2e3 years

Covariates in adjusted results

Median survival time men 70.0 months significantly less than for women 91.9 months, p < 0.01 (KaplaneMeier)

2249 (42%) 1 year

110 (36%) 5 years

Adjusted results for gender

No difference women 41%e(21.4) men 19.1% (63) Absolute difference with 90% CI 2.3 (4.5e9.0) No significant difference p ¼ 0.49

AF ¼ Atrial fibrillation, CHADS2 ¼ Congestive heart failure, Hypertension, Age, Diabetes mellitus, Prior Stroke or TIA stroke risk scheme for AF, CI ¼ confidence interval, CV ¼ cardiovascular, DM ¼ diabetes mellitus, LV ¼ left ventricular, EKG ¼ electrocardiogram, EPICA ¼ Prevalence of Heart Failure in Southwestern Europe Study, HTN ¼ hypertension, J-Rhythm ¼ Registry of Japanese patients with atrial fibrillation, RACE ¼ Rate Control versus Electrical Cardioversion for Persistent Atrial Fibrillation Study, RCT ¼ randomized control trial, HR ¼ Hazard ratio, LV ¼ left ventricular, OR ¼ odds ratio, TIA ¼ transient ischemic attack, VKA ¼ Vitamin K antagonist.

D.W. Chapa et al. / Heart & Lung 44 (2015) 189e198

Author location study design

D.W. Chapa et al. / Heart & Lung 44 (2015) 189e198

197

Table 4 Summary of the articles that compared hospitalization rates of women and men with atrial fibrillation and or flutter. Author location Outcome study design Humphries12 Canada cohort

Population

Sample N

229,530 First hospitalization Patients 20 to 106 yo discharged for AF/Afl with 1st AF/Afl from Canadian health System records

Female Follow-up N% time

Unadjusted results for gender

Adjusted results for gender

Covariates in adjusted results

At all ages men > women

Age adjusted

AF ¼ atrial fibrillation; Afl ¼ atrial flutter; yo ¼ years old.

Limitations There were several limitations to this review. First, there were not enough articles to conduct a meta-analysis; a synthesis using a qualitative assessment tool for quantitative articles was used in this review. We did not assess publication bias.44,45 For example, results may be biased by review of English language articles only. Finally, most articles included many more men than women; and only two28,35 included more women than men. Two articles24,29 included fewer than 35% women. Although the total numbers of men (128,010) and women (122,795) were almost equal. The articles with large differences in numbers of men and women may not adequately assess gender differences. Conclusion There is a dearth of articles examining differences in outcomes between men and women with AF/Afl. There are differences in rates of stroke between men and women with AF/Afl. Stroke is higher in women, after adjusting for age and other risk factors. This information on gender disparities in strokes may be important for treatment of patients with AF/Afl. Disparities in catheter ablation and use of anticoagulation therapy in women may contribute to women’s increase risk of stroke. This emphasizes the need for stroke prevention education and promotion of healthy life styles. Eating a healthy diet, moderate alcohol consumption, never smoking, keeping physical active and maintaining a healthy body mass decreased women’s risk of stroke by 54%.46 Further study is needed to confirm increased hospitalizations for men with AF/Afl and a possible need for interventions to decrease the rate of hospitalizations for men with AF/Afl. Atrial fibrillation is growing in prevalence and health care costs for AF are rising.8,47,48 If all studies of outcome reported incidence rates according to gender, the effect of gender on outcomes in AF/Afl would be clear. Given the prevalence of AF/Afl and associated health care costs it is vital to examine the differences of predictors based on gender to evaluate appropriate therapies that would decrease incidence of stroke, mortality and hospitalization in the AF/Afl population. References 1. Heeringa J, van der Kuip DA, Hofman A, et al. Prevalence, incidence and lifetime risk of atrial fibrillation: the Rotterdam study. Eur Heart J. 2006;27(8):949e953. 2. Shea JB, Sears SF. A patient’s guide to living with atrial fibrillation. Circulation. 2008;117(20):e340ee343. 3. Wyse DG, Waldo AL, DiMarco JP, et al. A comparison of rate control and rhythm control in patients with atrial fibrillation. N Engl J Med. 2002;347(23):1825. 4. Curtis AB, Gersh BJ, Corley SD, et al. Clinical factors that influence response to treatment strategies in atrial fibrillation: the Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) study. Am Heart J. 2005;149(4):645e649. 5. Shea JB, Maisel WH. Cardioversion. Circulation. 2002;106(22):e176ee178. 6. Waktare JE. Atrial fibrillation. Circulation. 2002;106(1):14e16. 7. Coyne KS, Paramore C, Grandy S, Mercader M, Reynolds M, Zimetbaum P. Assessing the direct costs of treating nonvalvular atrial fibrillation in the United States. Value Health. 2006;9(5):348e356. 8. Wolf PA, Mitchell JB, Baker CS, Kannel WB, D’Agostino RB. Impact of atrial fibrillation on mortality, stroke, and medical costs. Arch Intern Med. 1998;158(3):229.

9. Go AS, Hylek EM, Phillips KA, et al. Prevalence of diagnosed atrial fibrillation in adults. J Am Med Assoc. 2001;285(18):2370e2375. 10. Colissa S, Crow A, Petkun W, et al. Estimates of current and future incidence and prevalence of atrial fibrillation in the U.S. adult population. Am J Cardiol. 2013;112(8):1142e1145. 11. Lloyd-Jones Donald M, Wang TJ, Leip EP, et al. Lifetime risk for development of atrial fibrillation the Framingham heart study. Circulation. 2004;110(9): 1042e1046. 12. Humphries KH, Jackevicius C, Gong Y, et al. Population rates of hospitalization for atrial fibrillation/flutter in Canada. Can J Cardiol. 2004;20(9):869e876. 13. Ferrari R, Abergel H, Ford I, et al. Gender-and age-related differences in clinical presentation and management of outpatients with stable coronary artery disease. Int J Cardiol. 2013;167(6):2938e2943. 14. Anand S, Xie C, Mehta S, et al. Differences in the management and prognosis of women and men who suffer from acute coronary syndromes. J Am Coll Cardiol. 2005;46(10):1845e1851. 15. Glaser R, Selzer F, Jacobs AK, Laskey WK. Effect of gender on prognosis following percutaneous coronary intervention for stable angina pectoris and acute coronary syndromes. Am J Cardiol. 2006;98(11):1446e1450. 16. Akintade BF, Chapa D, Friedmann E, Thomas SA. The influence of depression and anxiety symptoms on health-related quality of life in patients with atrial fibrillation and atrial flutter. J Cardiovasc Nurs; 2015:66e73. 17. Chapa DW, Akintade B, Schron E, Friedmann E, Thomas SA. Is health-related quality of life a predictor of hospitalization or mortality among women or men with atrial fibrillation. J Cardiovasc Nurs. 2014;29(6):555e564. 18. Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: explanation and elaboration. Ann Intern Med. 2009;151(4): W-65. 19. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Ann Intern Med. 2009;151(4):264e269. 20. Effective Public Health Practice Project. Quality assessment tool for quantitative studies. Webpage. Available at: URL, http://www.hamilton.ca/ HealthandSocialServices/Research/EPHPP/; 2009 June 20. Accessed 22.06.13. 21. Armijo-Olivo S, Stiles CR, Hagen NA, Biondo PD, Cummings GG. Assessment of study quality for systematic reviews: a comparison of the Cochrane Collaboration Risk of Bias Tool and the Effective Public Health Practice Project Quality Assessment Tool: methodological research. J Eval Clin Pract. 2012;18(1):12e18. 22. Thomas BH, Ciliska D, Dobbins M, Micucci S. A process for systematically reviewing the literature: providing the research evidence for public health nursing interventions. Worldviews Evid Based Nurs. 2004;1(3):176e184. 23. Ford GA, Choy AM, Deedwania P, et al. Direct thrombin inhibition and stroke prevention in elderly patients with atrial fibrillation: experience from the SPORTIF III and V Trials. Stroke. 2007;38(11):2965e2971. 24. Gomberg-Maitland M, Wenger NK, Feyzi J, et al. Anticoagulation in women with non-valvular atrial fibrillation in the stroke prevention using an oral thrombin inhibitor (SPORTIF) trials. Eur Heart J. 2006;27(16):1947e1953. 25. Asbach S, Olschewski M, Faber TS, Zehender M, Bode C, Brunner M. Mortality in patients with atrial fibrillation has significantly decreased during the last three decades: 35 years of follow-up in 1627 pacemaker patients. Europace. 2008;10(4):391e394. 26. Rienstra M, Van Veldhuisen DJ, Hagens VE, et al. Gender-related differences in rhythm control treatment in persistent atrial fibrillation: data of the Rate Control versus Electrical Cardioversion (RACE) study. J Am Coll Cardiol. 2005;46(7):1298e1306. 27. Dagres N, Nieuwlaat R, Vardas PE, et al. Gender-related differences in presentation, treatment, and outcome of patients with atrial fibrillation in Europe: a report from the Euro Heart Survey on Atrial Fibrillation. J Am Coll Cardiol. 2007;49(5):572e577. 28. Avgil Tsadok M, Jackevicius CA, Rahme E, Humphries KH, Behlouli H, Pilote L. Sex differences in stroke risk among older patients with recently diagnosed atrial fibrillation. J Am Med Assoc. 2012;307(18):1952e1958. 29. Cha MJ, Oh GC, Hahn S, Choi EK, Oh S. Thromboembolic risk evaluation in patients with atrial fibrillation using a modified CHADS2 scoring system. J Cardiovasc Electrophysiol. 2012;23(2):155e162. 30. Shen AY, Yao JF, Brar SS, Jorgensen MB, Chen W. Racial/ethnic differences in the risk of intracranial hemorrhage among patients with atrial fibrillation. J Am Coll Cardiol. 2007;50(4):309e315. 31. Fang MC, Singer DE, Chang Y, et al. Gender differences in the risk of ischemic stroke and peripheral embolism in atrial fibrillation: the AnTicoagulation and

198

32.

33.

34.

35.

36.

37.

38.

D.W. Chapa et al. / Heart & Lung 44 (2015) 189e198 Risk factors in Atrial fibrillation (ATRIA) study. Circulation. 2005;112(12): 1687e1691. Poli D, Antonucci E, Grifoni E, Abbate R, Gensini GF, Prisco D. Gender differences in stroke risk of atrial fibrillation patients on oral anticoagulant treatment. Thromb Haemost. 2009;101(5):938e942. Friberg J, Scharling H, Gadsboll N, et al. Comparison of the impact of atrial fibrillation on the risk of stroke and cardiovascular death in women versus men (The Copenhagen City Heart Study). Am J Cardiol. 2004;94(7):889e894. Friberg L, Benson L, Rosenqvist M, Lip G. Assessment of female sex as a risk factor in atrial fibrillation in Sweden: nationwide retrospective cohort study. BMJ. 2012;344. Poli D, Antonucci E, Testa S, et al. Gender differences of bleeding and stroke risk in very old atrial fibrillation patients on VKA treatment: results of the EPICA study on the behalf of FCSA (Italian Federation of Anticoagulation Clinics). Thromb Res. 2013 Jan;131(1):12e16. Inoue H, Atarashi H, Okumura K, et al. Impact of gender on the prognosis of patients with nonvalvular atrial fibrillation. Am J Cardiol. 2014;113(6): 957e962. Melgaard L, Rasmussen LH, Skjoth F, Lip GY, Larsen TB. Age dependence of risk factors for stroke and death in young patients with atrial fibrillation: a nationwide study. Stroke. 2014;45(5):1331e1337. Salam A, AlBinali HA, Al-Mulla AW, et al. Women hospitalized with atrial fibrillation: gender differences, trends and outcome from a 20-year registry in a Middle Eastern country (1991e2010). Int J Cardiol. 2013 Sep30;168(2): 975e980.

39. Wang T, Massaro JM, Levy D, et al. A risk score for predicting stroke or death in individuals with new-onset atrial fibrillation in the community: the Framingham Heart Study. J Am Med Assoc. 2003;290(8):1049e1056. 40. Gage BF, Waterman AD, Shannon W, et al. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. J Am Med Assoc. 2001;285(22):2864e2870. 41. European Heart Rhythm Association, European Association for Cardio-Thoracic Surgery, Camm AJ, Kirchhof P, Lip GYH, et al. Guidelines for the management of atrial fibrillation: the Task Force for the Management of Atrial Fibrillation of the European Society of Cardiology (ESC). Eur Heart J. 2010;31(19):2369e2429. 42. Patel D, Mohanty P, Di Biase L, et al. Outcomes and complications of catheter ablation for atrial fibrillation in females. Heart Rhythm. 2010;7(2):167e172. 43. Arias E. United States live tables 2009. Natl Vital Stat Rep. 2014;62(7):1e63. 44. Guyatt GH, Oxman AD, Montori V, et al. GRADE guidelines: rating the quality of evidence of publication bias. J Clin Epidemiol. 2011;64(12):1277e1282. 45. Pham B, Klassen TP, Lawson ML, Moher D. Language of publication restrictions in systematic reviews gave different results depending on whether the intervention was conventional or complementary. J Clin Epidemiol. 2005;58(8):769e776. 46. Larsson S. Healthy lifestyle cut stroke risk in half for women. Neurology. 2014;83(19):1699e1704. 47. Le Heuzey JY, Paziaud O, Piot O, et al. Cost of care distribution in atrial fibrillation patients: the COCAF study. Am Heart J. 2004;147(1):121e126. 48. Stewart S, Murphy N, Walker A, McGuire A, McMurray JJV. Cost of an emerging epidemic: an economic analysis of atrial fibrillation in the UK. Heart. 2004;90(3):286e292.