Hypertension in Middle Eastern Arab and South Asian Patients with Atrial Fibrillation: From a 20-Year Hospital Registry in Qatar (1990-2010).

Background: The vast majority of literature on atrial fibrillation (AF) is based on studies from the developed world that mainly includes Caucasian patients. Data on AF in other ethnicities is very limited. The aim of this hospital-based study is to evaluate the effect of concomitant hypertension (HTN) on the characteristics and outcomes of Middle Eastern Arab and South Asian patients with AF in the state of Qatar. Materials and Methods and
Results: During the 20-year period, 3850 AF patients were hospitalized; 1483 (38.5%) had HTN, and 2367 (61.5%) without HTN. Patients with HTN were 11 years older, compared to non-HTN patients, and had a significantly higher prevalence of diabetes mellitus, chronic kidney disease, and dyslipidemia, compared to non-HTN patients. Furthermore, underlying coronary artery disease and heart failure were significantly more common but not valvular and rheumatic heart diseases which were more common in those without HTN. The rates of in-hospital mortality and stroke were significantly higher in the presence of HTN (5.3% versus 3.5%, and 0.7% versus 0.2% respectively, P = 0.001), compared to non-HTN patients. Conclusions: HTN is significantly associated with more comorbidities and worse clinical outcomes when it coexists with AF in hospitalized Middle Eastern Arab and South Asian patients. Copyright:

INTRODUCTION

Atrial fibrillation (AF) is the most common arrhythmia encountered in clinical practice.[1] Epidemiologic studies have shown that the presence of AF increases risk of mortality by 1.5-1.9 times in the age-matched general population.[2] The presence of AF in patients with acute coronary syndrome (ACS),[3] heart failure,[4] stroke,[5] renal failure,[6] and heart transplant[7] is associated with poor prognosis and high mortality risk.

Hypertension (HTN) is one of the most common conditions in the general practice.[8] It is an independent predictor for the development of AF.[9] Given that uncontrolled HTN increases the risk of cardiovascular diseases, coronary artery disease (CAD), stroke, and chronic kidney disease (CKD),[810] it may be hypothesized that the coexistence of HTN and AF may be associated with higher morbidity and mortality.

Most of the clinical studies[11] on patients with AF have been conducted in the developed countries with a predominant Caucasian population. Furthermore, to the best of our knowledge, studies[12] that reported the clinical characteristics and outcomes of hospitalized AF patients with HTN are scarce. The present hospital-based study aims to examine the clinical characteristics and outcomes of AF patients with HTN and compare them to those without HTN at a tertiary general hospital in the State of Qatar between 1990 and 2010.

MATERIALS AND METHODS

Study setting

Qatar is a small country in the Middle East located in the Arabian Gulf with a population of 1.7 million as per the 2010 census, and 2.07 million by the end of November 2013.[13] The population is comprised of Qatari and other Middle Eastern Arabs (<40%), as well as other ethnic groups. The vast majority of non-Arabs are South Asians mainly from India, Pakistan, Nepal, and Bangladesh.[13] This study enrolled patients from the Department of Cardiology and Cardiovascular Surgery at Hamad General Hospital (HGH), which was the only tertiary care hospital in the State of Qatar up until February 2010 when the tertiary cardiac hospital of the country became operational (the Heart Hospital). The clinical data of patients admitted to the coronary care unit (CCU) and cardiology ward at HGH was entered into the departmental database since January 1990. For the purpose of this study, the de-identified data between January 1990 and December 2010 were collected and analyzed. Patients with AF were selected for the purpose of this study and were divided into two groups according to the presence or absence of HTN and compared.

Definitions

The diagnosis of AF was based on the presence of AF on electrocardiography which was confirmed by two independent physicians during the index hospitalization. Diabetes mellitus was diagnosed based on patients’ past medical history, use of antidiabetic agents, or an incidental diagnosis of diabetes mellitus during the index hospitalization. Diagnosis of Congestive Heart Failure (CHF) was according to the Framingham criteria,[1415] and that of acute myocardial infarction according to the World Health Organization criteria.[16] The diagnosis of HTN was based on patients’ past medical history when HTN was newly diagnosed during the index hospitalization, according to The Seventh Report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 7) criteria.[17] Dyslipidemia was diagnosed based on the recommendations of the third report of the National Cholesterol Education Program (NCEP) for detection evaluation and treatment of high blood cholesterol in adults, also called the NCEP Adult Treatment Panel III (ATP III).[18] CKD was defined as serum creatinine level 1.5 times the upper limits of normal range.[19]

Statistical analysis

Patient's characteristics were expressed as mean and standard deviation for continuous variables and as frequencies and percentages for categorical variables. The characteristics of AF patients with HTN were compared with that of AF patients without HTN using Chi-square test and Student's t-test for the categorical and continuous variables, respectively. Multivariate analysis was performed by using a logistic regression model to evaluate the predictors of mortality. Adjusted odds ratio, 95% confidence interval, and P values were reported. All P values were derived from a two-sided test and considered significant when <0.05. Data analyses were carried out using the Statistical Package for Social Sciences version 19.0 (SPSS Inc., USA).

RESULTS

During the 20-year period, 3850 patients with AF were hospitalized, among which 1483 (38.3%) had HTN on presentation while 2367 (61.5%) did not. The characteristics of patients in both groups are presented in Table 1. The mean age and body mass index (BMI) of AF patients with HTN were significantly higher than that of those without HTN (63 ± 12 vs. 52 ± 16, P < 0.001 and 32 ± 7 vs. 29 ± 10; P < 0.001; respectively). Similarly, the proportion of females was significantly higher in those with HTN (47.3% vs. 30.2%, P < 0.001).

Table 1

Patients’ demographics, clinical characteristics and outcomes.

Variable	AF + HTN n (%)	AF alone n (%)	P
	1483 (38.3)	2367 (61.5)
Patient characteristics at admission
Age in year (mean±SD)	63±12	52±16	0.001
BMI (kg/m2) (mean±SD)	32±7	29±10	0.001
Female gender	702 (47.3)	715 (30.2)	0.001
Ethnicity (%)
Middle Eastern Arabs	1191 (80.3)	1666 (70.4)
South Asians	158 (10.7)	390 (16.5)
Others	134 (9)	311 (13.1)	0.001
Cardiovascular risk factors
Current smoker	119 (8)	358 (15.1)	0.001
Dyslipidemia	214 (14.4)	151 (6.4)	0.001
Diabetes mellitus	754 (50.8)	403 (17)	0.001
CKD	127 (8.6)	55 (2.3)	0.001
Dyslipidemia	214 (14.4)	151 (6.4)	0.001
Prior MI	218 (14.7)	199 (8.4)	0.001
Cardiac Diagnosis
ACS	167 (11.3)	122 (5.2)	0.001
Heart failure	344 (23.2)	351 (14.8)	0.001
VHD	33 (2.2)	133 (5.6)	0.001
RHD	11 (0.7)	86 (3.6)	0.001
Presenting symptom
Palpitation	591 (39.9)	1133 (47.9)	0.001
SOB	483 (32.6)	586 (24.8)	0.001
Chest pain	250 (16.9)	198 (8.4)	0.001
Dizziness	89 (6)	159 (6.7)	0.38
LVEF (mean±SD)	44±14	45±13.5	0.25
Hospital days (mean±SD)
CCU stay (days)	2.8±2	2±2	0.001
Total hospital stay (days)	5±4	4±4	0.001
In hospital outcome (%)
Death	78 (5.3)	82 (3.5)	0.007
Stroke	11 (0.7)	5 (0.2)	0.01

ACS, acute coronary syndrome; AF, atrial fibrillation; BMI, body mass index; CABG, coronary artery bypass graft; CCU, coronary care unit; CKD, chronic kidney disease; HTN, hypertension; LVEF, left ventricular ejection fraction; MI, myocardial infarction; RHD, Rheumatic heart diseases; SD, standard deviation; SOB, shortness of breath; VHD, valvular heart diseases.

Hypertensive patients with AF were more likely to have history of dyslipidemia (14.4% vs. 6.4%, P < 0.001), diabetes mellitus (50.8% vs. 17%, P < 0.001), CKD (8.6% vs. 2.3%, P < 0.001), and prior myocardial infarction (14.7% vs. 8.4%, P < 0.001) as compared to nonhypertensive patients. Interestingly, the history of smoking was significantly less common among those with HTN (8% vs. 15.1%, P < 0.001).

Upon hospital presentation, shortness of breath and chest pain were more common (32.6% vs. 24.8%, P < 0.001 and 16.9% vs. 8.4%, P < 0.001; respectively) in AF patients with HTN, whereas palpitation was more common in those without HTN (47.9% vs. 39.9%, P < 0.001).

Cardiac diagnosis of ACS (11.3% vs. 5.2%, P < 0.001) and heart failure (23.2% vs. 14.85, P < 0.001) was significantly higher in AF patients with HTN. While the diagnosis of valvular (VHD) (5.6% vs. 2.2%, P < 0.001) and rheumatic heart disease (RHD) (3.6% vs. 0.7%, P < 0.001) was higher in non-hypertensive patients.

Medication use prior to, during hospitalization and at hospital discharge is presented in Table 2. Aspirin use was significantly higher among nonhypertensive AF patients prior to hospital presentation (18.5% vs. 4.3, P < 0.001). However, aspirin use in hypertensive AF patients during hospital stay and on discharge surpassed that of the non-hypertensive ones (59.7% vs. 39.5%, P < 0.001 and 62.4% vs. 43.8%, P < 0.001; respectively).

Table 2

Medications used before presentation, during admission and at discharge.

Medications in %	Before Admission			During Admission			At Discharge
	AF + HTN	AF no + HTN	P	AF + HTN	AF no + HTN	P	AF+HTN	AF no + HTN	P
Aspirin	612 (4.3)	439 (18.5)	0.001	885 (59.7)	934 (39.5)	0.001	9.26 (62.4)	10.37 (43.8)	0.001
Warfarin	17 (1.1)	4 (2)	0.04	279 (18.8)	428 (18.1)	0.57	525 (35.4)	802 (33.9)	0.33
Clopidogrel	75 (5.1)	28 (1.2)	0.001	127 (8.6)	64 (2.7)	0.001	94 (6.3)	62 (2.6)	0.001
B blocker	242 (16.3)	129 (5.4)	0.001	394 (26.6)	501 (21.2)	0.001	276 (18.6)	382 (16.1)	0.05
CCB	132 (8.9)	33 (1.4)	0.001	250 (16.9)	136 (5.7)	0.001	318 (21.4)	151 (6.4)	0.001
Digoxin	321 (21.6)	466 (19.7)	0.14	589 (39.7)	1230 (52)	0.001	537 (36.2)	992 (41.9)	0.001
ACE inhibitors/ARBs	263 (17.7)	124 (5.2)	0.001	587 (39.6)	449 (19)	0.001	693 (46.7)	486 (20.5)	0.001
GPIIb/IIIa inhibitors	--	--	--	9 (0.6)	7 (0.3)	0.14	--	--	--
Unfractionated heparin	--	--	--	352 (23.7)	701 (29.6)	0.001	--	--	--
LMWH (enoxaparin)	9 (0.6)	2 (01)	0.003	199 (13.4)	245 (10.4)	0.004	6 (0.4)	10 (0.4)	0.93
Antiarrhythmics	366 (24.7)	502 (21.2)	0.01	834 (56.2)	1583 (66.9)	0.001	700 (47.2)	1164 (49.2)	0.23
Statins	--	--	--	--	--	--	382 (25.8)	231 (9.8)	0.001

AF=Atrial Fibrillation; GP=glycoprotein; HTN=Hypertension; LMWH=low molecular weight heparin; CCB=calcium channel blockers; ACE=angiotension converting enzyme inhibitor, ARB=angiotensin receptor blocker.

A small number of patients were on warfarin prior to hospitalization, and its use was significantly higher among non-hypertensive AF patients (2% vs. 1.1%, P = 0.04). However, warfarin use increased during hospitalization and upon hospital discharge without significant difference between the two groups.

The use of unfractionated heparin was significantly higher among nonhypertensive patients during hospitalization (29.6% vs. 23.7%, P < 0.001), while the use of low molecular weight heparin (LMWH) was significantly higher among the hypertensive group (13.4% vs. 10.4%, P = 0.004). Small number of patients was on LMWH prior to hospitalization with higher reported usage among the non-hypertensive AF patients (1% vs. 0.6%, P = 0.003). Few patients were prescribed LMWH upon hospital discharge with no significant difference in usage between the groups (0.4% vs. 0.4%, P = 0.93).

Antiarrhythmic drugs use was significantly higher more common among AF patients with HTN prior to hospitalization (24.7% vs. 21.2%, P = 0.01). Although prescribing antiarrhythmic drugs was similar between the study groups upon hospital discharge, significantly more patients without HTN were prescribed these drugs during hospitalization (66.9% vs. 56.2%, P < 0.001).

The use of other cardiac medications such as clopidogrel, beta-blockers, calcium channel blockers, and angiotensin-converting enzyme inhibitors (ACEI)/angiotensin receptor blockers (ARB) was significantly higher among AF patients with HTN prior to and during hospitalization, as well as upon hospital discharge.

Hypertensive AF patients stayed longer in CCU and in hospital (2.8 ± 2 vs. 2 ± 2, P < 0.001 and 5 ± 4 vs. 4 ± 4, P < 0.001; respectively). They fared worse in terms of death (5.3% vs. 3.5%, P < 0.007) and stroke (0.7% vs. 0.2%, P < 0.01) as compared to nonhypertensive group. After adjustment for multiple variables, advanced age was the only independent predictor of mortality among AF patients with HTN [Table 3].

Table 3

Multivariate analysis of predictors of in-hospital mortality in patients with AF and hypertension.

Variable	Adjusted OR	95% CI	P
Age	1.05	1.02-1.07	0.001
Female gender	0.88	0.54-1.45	0.63
Current Smoking	1.89	0.87-4.10	0.11
Diabetes mellitus	1.43	0.89-2.30	0.14
MI	1.36	0.77-2.39	0.29
CKD	1.69	0.87-3.26	0.12

DISCUSSION

The present study examined hospitalized AF patients over two decades and found that 38.3% of them had concomitant HTN. This group of patients as compared to those without HTN were older, more females, and had higher BMI and prevalence of other comorbidities such as diabetes, CKD, dyslipidemia, prior myocardial infarction. Moreover, ACS and heart failure were more common among them but not VHD or RHD. Consequently, they had longer CCU and hospital length of stay with a higher risk of both stroke and mortality.

Most of the observational studies on patients with AF have reported the characteristics of ambulatory patients,[1920] while only a few has reported that of hospitalized patients.[212223] The characteristics of patients as reported in these studies are summarized in Table 4.[20212425262728293031] The mean age of patients with AF in studies conducted in different regions have ranged from 64 years to 82 years,[20212223242526272829303132] whereas the patients in both groups of this study were younger, i.e., 52 and 63 years old, with comparable mean BMI (29 and 32 kg/m2) to that reported in the published data (range of 25-32 kg/m2) as shown in Table 4. Smoking among patients with AF ranged from 5.5% in European subjects[20] to 16% in the Japanese counterparts,[26] while the percentages in the present study varied between the two groups, i.e., 8% and 15.1%.

Table 4

Characteristics of patients with atrial fibrillation as reported in medical literature.

Author/year	Patients studied	Number of patients	Mean age (SD)	Gender (M: F)	Co existing conditions
					HTN (%)	DM (%)	CKD (%)	CAD (%)	CHF (%)	CVA (%)	Dyslipid (%)	VHD (%)	BMI (Kg/m2)	Smoking (%)
Hurley et al/2007 [18]	To asses HTN in pt >75yrs with AF in primary care setting, a secondary analysis of BAFTA study	3059	82 (5)	51:49	57.5	12.9	NA	35.1 h/o MI, 25.9 H/O angina	25.8	12.3	NA	16.8 (RHD 6.4)	NA	5.5
Lee et al/2007 [28]	Patients hospitalized with AF in Taiwan	162,340	73.8	55.3: 44.7	14.8	16.3	1-1.6	21.4	23.8	15.2	Na	14.8	NA	NA
Miyasaka et al/2007 [19]	Patients with AF in Olmsted county, Minnesota	4618	73.1 (14.4)	51:49	80	18	17	38	9.7	9.5	37	NA	25.8 to 28	NA
Nabauer et al/2009 [20]	Patients with AF enrolled in AFNET registry, Germany	9582	68.4 (11)	60:40	69	21.6	11.6	28	36	6.4	46.2	36.3 (RHD 3.7)	NA	NA
Uchiyama et al/2010 [21]	Analysis of J-TRACE registry	2242	70.5	68.8: 31.2	57.7	17.2	Na	3.4	21.8	14.3	26.9	ex	32.8% with BMI >25	16.1
Meinertz et al/2011[22]	Patients with AF enrolled in ATRIUM registry, Germany	3667	72 (9)	58:42	84	35	20	35	43	10.2	Na	NA	46% over wt, 32% obese	5.5
Steg et al/2011 [23]	Patients with AF enrolled in Realise AF study	10523	66.6 (12.1)	56:44	72.2	21.3	3.9	32.3	45.8	14.1	46.3	26.7	NA	10.2
Amerena et al/2012[24]	Ambulatory AF patients in RECORD AF- AP study	2629	64 (13)	60:40	58	18	6	19	25	9	37	23	25.5±4.9	13
De Vos et al/2012 [25]	Ambulatory AF patients in RECORD AF study	2137	65.1 (12)	53:47	70	15	6	19	20	8	46	17	28.2±5	14
Olesen et al/2012 [26]	Danish patients discharged from hospital with a diagnosis of AF	132372	66 – 76	53:47	37	9	3.4	NA	17.7	14.1	NA	ex	NA	NA

AF=Atrial Fibrillation, BMI=Body Mass Index, CAD=Coronary Artery Disease, CHF=Congestive Heart Failure, CVA=Cerebrovascular Accident, DM=Diabetes Mellitus, ex=exclusion criteria, HTN=Hypertension, VHD=Valvular Heart Disease, NA=Data not available

The prevalence of comorbidities such as diabetes mellitus among patients with AF varied from 9% to 35% in the respective studies. The overall prevalence of diabetes mellitus in this study was 30% with significant difference between the study groups, i.e., 17% and 50.8%.

The prevalence of heart failure among AF patients has ranged from 10% to 46% in previous reports [Table 4], whereas the prevalence of CHF among hospitalized AF patients in Taiwan was 23.8% between 1997 and 2002.[21] In the United States, CHF was more prevalent (21%) in hospitalized patients older than 65 years compared to younger ones (13%) between 1985 and 1999.[33] In Europe, the prevalence of heart failure among German ambulatory and hospitalized patients with permanent AF was 37%.[25] The prevalence of chronic heart failure in Scotland was similar between hospitalized males and females (13%) in 1986. However, by 1996, the prevalence in male patients reached a lower rate of 7%, whereas in females the rate did not change.[22] The overall prevalence of heart failure in the present study was 18% which varied between the study groups, i.e., 14.8% and 23.2%.

The prevalence of CAD among ambulatory and hospitalized patients with AF has varied globally from 3.4% to 38% as presented in Table 4. The history of prior myocardial infarction in the current study was 10.8% among all patients and varied between the different groups, i.e., 8.4% and 14.7%. The reported acute coronary events rates in Scottish patients with AF were 1.7% and 1.1% in 1986 and 1996, respectively.[22] The overall rate of acute events in Qatar as reported in the current study was 7.5% with different reported rates between the study groups, i. e., 5.2% and 11.5%.

The reported rate ranges of VHD and RHD were 14.8%–36% and 3.7%–6.4%, respectively [Table 4]. The overall rates in the current study were 4.3% and 2.5%, respectively, with significantly lower rates of both diseases in the hypertensive group.

Medication use was mainly reported in the studies of ambulatory patients with AF.[202425262728293031] The rates of antithrombotic agents’ use were 19%-51% for aspirin, 28%–83% for warfarin, and none reported for clopidogrel. The rates of the use of other medications were 7.2%-–75% for beta-blockers, 7.3%–34% for calcium channel blockers, 9.3%–35% for digoxin, 12%–83% for ACEI/ARB, and 11%-–61% for antiarrhythmic drugs. Comparisons between medication use on hospital discharge in this study and that in the available published data are shown in Figure 1.

Figure 1

Medication use among patients with atrial fibrillation as reported in clinical studies from different geographic regions and its comparison with the two patient groups in the study

Patients in Taiwan who were hospitalized with a primary diagnosis of AF stayed in hospital for 8 days, while those admitted with secondary diagnosis of AF stayed for 12 days.[21] In Scotland, females admitted to hospital with primary or secondary diagnosis of AF stayed longer as compared to their male counterparts (5 vs. 3 days or 11 vs. 8 days, respectively).[22] Data from the current study did not include information about AF diagnosis, but the mean length of hospital stay was 4 or 5 days in the different groups.

In-hospital mortality rates among all patients hospitalized for AF were 0.5% to 1.0% in the United States between 1996 and 2001.[23] In Taiwan, the rate of in-hospital mortality was lower in patients hospitalized for AF as a primary diagnosis compared to that for those hospitalized with AF as secondary diagnosis (7% vs. 10.7%) between 1997 and 2002.[21] In 1996, a Scottish study reported gender disparity in the in-hospital mortality risk, with higher rates in females compared to that in males in both primary and secondary diagnosis of AF (3.4% vs. 1.8% and 11.8% vs. 7.8%, respectively).[22] The overall mortality rate in the current study was 4.2% with variation between the study groups, i.e., 3.5% and 5.3%.

Limitations

The observational retrospective nature of the data analysis subjects the study to bias. It was difficult to report the different types of AF and whether AF was a primary or secondary diagnosis. Moreover, creatinine clearance was not used to define CKD.

CONCLUSIONS AND FUTURE DIRECTION

The present study demonstrates that hospitalized AF patients who have coexistent HTN were at higher risk of morbidity and mortality. Evaluation and implementation of optimal management strategy may be instrumental in improving clinical outcomes. Ideally, the characteristics of these patients should be investigated in properly designed prospective studies.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.