The association of hypertension with asymptomatic cardiovascular organ damage in rheumatoid arthritis

Abstract

Background: The association of hypertension with asymptomatic cardiovascular organ damage in patients with rheumatoid arthritis (RA) has been little studied by echocardiography.

Methods: Echocardiography was done in 134 RA patients and 102 healthy controls. Left ventricular (LV) geometry was considered abnormal if LV mass index or relative wall thickness was increased. LV diastolic dysfunction was considered present if septal early diastolic tissue velocity <8 cm/s. Systemic arterial compliance (SAC) was assessed from stroke volume index/pulse pressure ratio.

Results: The hypertensive RA patients (n = 72) had higher inflammatory activity, older age and more diabetes than the normotensive RA patients (n = 62) (all p < 0.05). Rates of abnormal LV geometry, LV diastolic dysfunction and lower SAC were higher among the hypertensive RA patients (p < 0.05), but similar between normotensive RA patients and controls. Hypertension was associated with a 3-fold higher prevalence both for abnormal LV geometry (odds ratio 2.89 [95% confidence interval 1.09–7.63], p = 0.03) and for diastolic LV dysfunction (odds ratio 2.92 [95% confidence interval 1.14–7.46], p = 0.03) as well as lower SAC (β = 0.31, p = 0.001) independent of age, gender, diabetes and inflammatory activity measured by erythrocyte sedimentation rate.

Conclusion: The presence of asymptomatic cardiovascular organ damage in RA patients is closely associated with hypertension independent of inflammatory activity.

Keywords:

• Hypertension
• left ventricular geometry
• rheumatoid arthritis
• systemic arterial compliance

Introduction

Patients with rheumatoid arthritis (RA) have a high risk of cardiovascular (CV) disease, attributed to a combination of systemic inflammation, side-effects of RA medication and high burden of CV risk factors.[1,2] The prevalence of hypertension is increased with up to 30% in RA patients compared to the general population, making hypertension one of the major modifiable CV risk factors in RA patients.[3] It is well documented that clinical CV events are preceded by asymptomatic CV organ damage in hypertensive patients without autoimmune disease.[4–6] In RA patients, untreated or uncontrolled hypertension, was associated with the presence of subclinical CV organ damage as assessed by the 12-lead electrocardiogram and urinary albumin:creatinine ratio in RA patients.[7] However, the association of hypertension with the presence of asymptomatic CV organ damage has not been much explored in RA patients using echocardiography. This was the aim of this study.

Materials and methods

Study population

RA patients were diagnosed by the 1987 American College of Rheumatology criteria [8] and recruited from the 15-year follow-up of the European Research on Incapacitating Diseases and Social Support (EURIDISS) cohort and the 10-year follow-up of the Oslo RA register cohort.[9,10] Details about these cohorts have been published previously.[9–11] All surviving participants were invited to participate in the study and 152 (57%) accepted. For the present analysis, only RA patients without established CV disease were eligible, and 92 from the EURIDISS cohort and 42 from the Oslo RA register cohort were included.

Statistics Norway selected 329 community control subjects without inflammatory joint disease matching the RA patients for sex, age and residential area, and of those 132 (40%) showed up for echocardiography. The clinical examinations revealed that a majority of the community controls had established CV disease or hypertension, leaving only 47 (14%) of the original control subjects to be included. In addition, 55 healthy control subjects were recruited from a control database available at the Echocardiography Core Laboratory at Haukeland University Hospital, Bergen, Norway.[12,13] The study protocol was approved by the Norwegian Regional Committee for Medical and Health Research Ethics, and all participants signed an informed consent according to the Declaration of Helsinki.

Inflammatory markers

Erythrocyte sedimentation rate (ESR) was measured from fasting blood samples by the Westergren method and C-reactive protein (CRP) by the COBAS 6000 (Roche diagnostics, Basel, Switzerland). Anti-cyclic citrullinated peptide and rheumatoid factor antibodies were analyzed from frozen serum or plasma by enzyme-linked immunosorbent assays (Inova Diagnostics, San Diego, CA). RA disease activity was assessed by the Disease Activity Score in 28 joints.[14] The participant’s self-reported medical history, smoking status and current use of medication were collected using a standardized questionnaire and quality assured by the consultant cardiologist (A.G.S.).

Blood pressure

Brachial blood pressure was measured following the European Society of Hypertension guidelines with an OMRON M7 apparatus (Kyoto, Japan).[15] Hypertension was defined as use of antihypertensive medication, known hypertension or systolic blood pressure ≥140 mmHg and/or diastolic blood pressure ≥90 mmHg at the clinic visit.[15]

Echocardiography

Echocardiography was performed following a standardized protocol, and analysed using offline digital workstations with Image Arena software version 4.1 (TomTec Imaging Systems GmbH, Unterschleissheim, Germany) at the Echocardiography Core Laboratory. All the studies were proof-read by the same reader (E. G.). Quantitative echocardiography was performed following current guidelines.[16] Left ventricular (LV) mass was indexed for height^2.7, and LV hypertrophy was considered present when exceeding the prognostically validated cut-off values of LV mass index ≥46.7 g/m^2.7 in women and ≥49.2 g/m^2.7 in men.[5] A relative wall thickness ≥0.43 was defined as concentric geometry.[4] LV geometry was classified into 4 groups based on the presence of normal or elevated LV mass index or relative wall thickness.[16] LV diastolic function was assessed from LV filling and early septal mitral annulus velocity (e′).[17] LV diastolic dysfunction was considered present e′ < 8 cm/s.[17] Mitral deceleration index was assessed as mitral valve deceleration time/peak early velocity.[18] Systemic arterial compliance (SAC) was assessed as stroke volume index/pulse pressure ratio using a previous prognostically validated method and indexed for height in the allometric power of 2.04.[19,20]

Statistics

The statistical analyses were done using IBM SPSS statistics version 22.0 (IBM, Armonk, NY). For continuous variables, normally distributed data are expressed as mean and standard deviation and as median and interquartile range for non-normally distributed data. Categorical variables are presented as percentages. Non-normally distributed variables were log transformed for comparison in univariable and multivariable analyses. Comparisons between groups were done using the one-way Analysis of Variance with Scheffe’s post hoc test, general linear model with Sidak’s post hoc test, two-sample Student’s t-test or Chi-square test as appropriate. Covariates of SAC, LV diastolic dysfunction and abnormal LV geometry were identified in multivariable linear and logistic regression analyses, respectively. A two-tailed p value of ≤0.05 was considered statistically significant in all analyses.

Results

Clinical characteristics

The average RA disease activity was low in the study population (Table 1). The hypertensive RA patients (n = 72) were older, had higher blood pressure and body mass index, and included more subjects with diabetes compared to the normotensive RA patients (n = 62) and the healthy control subjects (n = 102; all p < 0.05) (Table 1). The normotensive RA patients had comparable blood pressure, body mass index and prevalence of diabetes as the healthy control subjects, but were older and more often female (Table 1).

Table 1. Clinical characteristics of hypertensive RA patients, normotensive RA patients and control subjects.

Variables	Hypertensive RA patients (n = 72)	Normotensive RA patients (n = 62)	Controls (n = 102)
Age (years)	66 ± 10	57 ± 11*	52 ± 10*†
Women (%)	69	86	65†
Body mass index (kg/m^2)	26.8 ± 5.3	24.2 ± 3.7*	24.3 ± 3.5*
Heart rate (beats/min)	70 ± 13	66 ± 9	63 ± 11*
Current smoking (%)	23	20	19
Diabetes (%)	15	2*	0*
Systolic blood pressure (mmHg)	147 ± 18	119 ± 12*	118 ± 11*
Diastolic blood pressure (mmHg)	85 ± 9	74 ± 6*	74 ± 8*
Mean blood pressure (mmHg)	106 ± 10	89 ± 8*	88 ± 8*
Pulse pressure (mmHg)	62 ± 16	45 ± 8*	45 ± 8*
Hypertensive patients on treatment (%)	58	–	–
Angiotensin-converting enzyme inhibitors (%)	7	–	–
Angiotensin receptor blockers (%)	38	–	–
Calcium antagonists (%)	19	–	–
β-blockers (%)	24	–	–
α-blockers (%)	5	–	–
Diuretics (%)	19	–	–
RA disease duration (years)	17 ± 2	17 ± 2	–
Rheumatoid factor positive (%)	59	50	–
Anti-cyclic citrullinated peptide antibodies positive (%)	55	50	–
ESR (mm/h), median (IQR)	14 (8,26)	11 (7,16)	–
C-reactive protein (mg/l), median (IQR)	4 (2,11)	2 (1,6)*	–
Disease Activity Score in 28 joints	2.7 ± 0.9	2.4 ± 0.9*	–
Disease-modifying antirheumatic drugs (%)	69	57	–
Use of Prednisolone (%)	35	26	–

Data are expressed as mean ± standard deviation unless otherwise stated. ESR: erythrocyte sedimentation rate; IQR: interquartile range; RA: rheumatoid arthritis.

* p < 0.05 versus hypertensive RA patients,

† p < 0.05 versus normotensive RA patients.

Abnormal left ventricular geometry

The hypertensive RA patients had more abnormal LV geometry than the normotensive RA patients and healthy control subjects (both p < 0.05) while LV geometry was comparable between normotensive RA patients and healthy controls (Table 2 and Figure 1). In multivariable analyses, the main covariables of abnormal LV geometry were hypertension and higher age (p < 0.05) both among RA patients and in the total study population (Table 3).

Figure 1. LV geometry in hypertensive and normotensive groups of RA patients and in controls. LV: left ventricular; RA: rheumatoid arthritis.

Table 2. Echocardiographic characteristics of hypertensive RA patients, normotensive RA patients and control subjects.

Variables	Hypertensive RA patients (n = 72)	Normotensive RA patients (n = 62)	Controls (n = 102)
Interventricular septum thickness end-diastole (cm)	1.02 ± 0.27	0.88 ± 0.20*	0.91 ± 0.17*
Posterior wall end-diastolic thickness (cm)	0.89 ± 0.16	0.77 ± 0.15*	0.79 ± 0.16*
LV diameter at end-diastole (cm)	4.6 ± 0.6	4.7 ± 0.5	4.7 ± 0.5
LV mass/height^2.7 (g/m^2.7)	37.7 ± 12.5	30.8 ± 9.1*	31.8 ± 8.3*
LV relative wall thickness (ratio)	0.39 ± 0.10	0.33 ± 0.07*	0.34 ± 0.08*
Ejection fraction (%)	67 ± 6	67 ± 4	67 ± 4
Stroke volume index (ml/m^2.04)	22.4 ± 5.9	22.6 ± 4.8	22.6 ± 4.8
SAC ([ml/m^2.04]/mmHg)	0.39 ± 0.14	0.52 ± 0.13*	0.52 ± 0.12*
E/A ratio	1.0 ± 0.5	1.2 ± 0.3	1.4 ± 0.5*†
Septal e′ velocity (cm/s)	7 ± 2	9 ± 2*	9 ± 2*
Septal e′ <8 cm/s (%)	66	26*	27*
E/e′ ratio	11.4 ± 5.4	8.4 ± 2.2*	7.6 ± 2.5*
E/e′ ≥15 (%)	13	0*	3*
Mitral deceleration index (ms/[cm/s])	2.61 ± 0.96	2.78 ± 0.95	2.95 ± 0.98

Data are expressed as mean ± standard deviation or as percentages. e′: peak early diastolic tissue velocity at septal border of mitral annulus; E/A: early/atrial transmitral peak velocities; LV: left ventricular; RA: rheumatoid arthritis; SAC: systemic arterial compliance.

* p < 0.05 versus hypertensive RA patients,

† p < 0.05 versus normotensive RA patients.

Table 3. Covariates of abnormal LV geometry in RA patients and controls.

	Multivariable analysis RA population			Multivariable analysis Total study population
Independent variable	OR	95% CI	p	OR	95% CI	p
Hypertension	2.89	1.09–7.63	0.03	3.22	1.25–8.30	0.02
Age (years)	1.06	1.02–1.11	0.007	1.05	1.01–1.08	0.005
Female gender	0.75	0.30–1.91	0.55	0.78	0.38–1.60	0.50
lnESR (mm/hr)	1.15	0.67–1.98	0.61	–	–	–
Having RA	–	–	–	0.70	0.26-1.86	0.48
Diabetes	0.75	0.19-2.91	0.68	0.78	0.21-2.93	0.71

CI: confidence interval; lnESR: natural logarithm of erythrocyte sedimentation rate; LV: left ventricular; OR: odds ratio; RA: rheumatoid arthritis.

Left ventricular diastolic dysfunction

Ejection fraction was normal both for RA patients and controls, except in two hypertensive RA patients where it was mildly reduced. Hypertensive RA patients had lower septal e′, higher mitral early peak filling velocity/ e′ (E/ e′) ratio and higher prevalence of LV diastolic dysfunction than the normotensive RA patients and controls (all p < 0.05) (Table 2). In multivariable logistic analyses, hypertension and higher age were independently associated with the presence of LV diastolic dysfunction both among RA patients and in the total study population (p < 0.05, Table 4).

Table 4. Covariates of LV diastolic dysfunction (septal e′ <8 cm/s) among RA patients and controls.

	Multivariable analysis RA population			Multivariable analysis Total study population
Independent variable	OR	95% CI	p	OR	95% CI	p
Hypertension	2.92	1.14–7.46	0.03	2.55	1.00–6.48	0.05
Age (years)	1.12	1.06–1.17	<0.001	1.13	1.08–1.18	<0.001
Diabetes	3.45	0.59–20.02	0.17	3.44	0.59–20.15	0.17
Female gender	1.95	0.67–5.67	0.22	1.13	0.50–2.52	0.77
lnESR (mm/hr)	0.96	0.55–1.68	0.89	–	–	–
Having RA	–	–	–	0.50	0.19–1.30	0.16

CI: confidence interval; e′: peak early diastolic tissue velocity at septal border of mitral annulus; lnESR: natural logarithm of erythrocyte sedimentation rate; OR: odds ratio; RA: rheumatoid arthritis.

Systemic arterial compliance

SAC was significantly lower among the hypertensive RA patients than in the other groups (Table 2). The combination of hypertension and higher inflammatory activity (taken as the upper tertile of ESR) had the lowest levels of SAC and the highest LV relative wall thickness in RA patients (Figure 2). Lower SAC was associated with higher LV relative wall thickness (p < 0.001), but not with higher LV mass both in RA patients and in controls (Figure 3). In multivariable analysis, hypertension and higher inflammatory activity assessed by ESR were the main covariables of lower SAC in RA patients (all p < 0.05, Table 5). Replacing ESR with the Disease Activity Score in 28 joints in another multivariable model did not alter the association between hypertension and lower SAC (data not shown).

Figure 2. The combination of hypertension and higher ESR was particularly associated with higher LV relative wall thickness (A) and lower SAC (B). ESR: erythrocyte sedimentation rate; LV: left ventricular; RA: rheumatoid arthritis; SAC: systemic arterial compliance.

Figure 3. Lower SAC was associated with higher LV relative wall thickness (A): but not with higher LV mass (B) both among RA patients and controls. LV: left ventricular; RA: rheumatoid arthritis; SAC: systemic arterial compliance.

Table 5. Covariates of lower SAC in the total study population and among RA patients.

	Multivariable analysis RA patients		Multivariable analysis Total study population
	Multiple R^2 = 0.30, p < 0.001		Multiple R^2 = 0.25, p < 0.001
Independent variable	Beta	p	Beta	p
Hypertension	0.31	0.001	0.32	<0.001
lnESR (mm/h)	0.20	0.01	–	–
Age (years)	0.20	0.02	0.21	0.002
Female gender	−0.09	0.28	−0.16	0.01
Heart rate (beats/min)	0.12	0.15	0.12	0.05
Having RA	–	–	−0.04	0.60
Diabetes	0.09	0.23	0.06	0.35

lnESR: natural logarithm of erythrocyte sedimentation rate; RA: rheumatoid arthritis; SAC: systemic arterial compliance.

Discussion

This study demonstrates that in RA patients with low disease activity, the presence of asymptomatic CV organ damage like abnormal LV geometry, LV diastolic dysfunction and reduced SAC are all closely associated with the presence of concomitant hypertension. Interestingly, the combination of higher inflammatory activity and hypertension was particularly associated with asymptomatic CV organ damage among RA patients. These findings emphasize the importance of hypertension for the development of CV organ damage and subsequent CV disease in RA patients.

The suggested interaction between hypertension and inflammation seen in this study is supported by previous publications.[21,22] Higher levels of inflammatory cytokines have been found in patients with resistant hypertension compared to mildly hypertensive or normotensive patients.[21] Also, higher CRP levels have been found to predict incident hypertension independent of baseline blood pressure in the general population.[22] Further, a pathogenetic role of the immune system for the development of subclinical CV disease has been suggested both in the normal population and in RA patients.[23] In line with this, we recently demonstrated that higher disease activity was associated with higher relative wall thickness in RA patients.[24]

Whether having RA influences LV geometry independent of confounders has been inconsistently reported, probably reflecting differences in clinical characteristics like age, obesity, hypertension and diabetes between the study populations.[25,26] In this study population, excluding RA patients with known CV disease, having RA was not independently associated with abnormal LV geometry after adjusting for hypertension and other covariables.

LV diastolic dysfunction is regarded as an early sign of cardiac organ damage in hypertension.[15] In this study, hypertension and higher age were the main covariables of LV diastolic dysfunction also among RA patients. In contrast, no independent association between having RA and reduced LV diastolic function was found, contrasting previous reports in RA patients using more traditional measures for assessing diastolic function.[27,28]

The reported association of lower SAC with LV concentric remodeling in RA patients adds to previous reports in hypertensive patients.[29,30] In RA patients, this has been little studied, but two previous studies reported that arterial stiffness was not associated with higher LV mass, in line with our findings.[25,31] Lower SAC will contribute to the development of concentric geometry through increased load on the LV.[32] This is clinically important since LV concentric geometry has been associated with increased risk of CV death, stroke and myocardial infarction in hypertensive patients.[4] Both hypertension and higher ESR were independently associated with lower SAC in this study, indicating that reduced SAC is multifactorial in RA patients. The role of inflammation for the development of arterial stiffness in RA patients has previously been documented in an positron emission tomography study that showed that ¹⁸F-fluorodeoxyglucose uptake in the aortic wall was reduced by anti-tumour necrosis factor-α therapy in parallel with arterial stiffness.[33]

There are some limitations to the present results. In particular, the cross-sectional study design is unsuited to identify a causal association between hypertension and asymptomatic CV organ damage. Since the patients were recruited from a follow-up cohort, selection bias cannot be excluded. The strengths of this study are that we used an echocardiographic core laboratory as recommended by guidelines [34] and that both patient and controls were free of established cardiac disease.

In conclusion, this study demonstrates that in RA patients with low disease activity, the presence of asymptomatic CV organ damage like abnormal LV geometry, LV diastolic dysfunction and reduced SAC is closely associated with hypertension independent of disease related inflammation, pointing to the importance of hypertension management in RA.

Funding information

The study was funded by grants from the Norwegian Extra Foundation for Health and Rehabilitation (Oslo, Norway) and the South-Eastern and Western Norway Regional Health Authorities (Oslo and Bergen, Norway).

Acknowledgements

The authors are grateful to Sella Provan and Asle Hirth for their contributions with data collection.

Disclosure statement

The authors have no conflict of interest related to this work.