Abstract
Objective The prevalence of hypertension, a major cardiovascular risk factor, is increased in patients with rheumatoid arthritis (RA) and may be driven by immune activation. The purpose of this study was to determine if ambulatory 24-hour blood pressure (BP) is elevated in RA vs control participants and whether it is associated with immune activation.
Methods We conducted a cross-sectional study of 46 patients with RA and 23 control participants. Participants wore an ambulatory BP monitor that obtained diurnal BP every 15-30 minutes and nocturnal BP every 30 minutes. Inflammatory mediators in plasma were measured using an inflammation proteomics panel. Differences in BP measurements were assessed by Mann-Whitney U test, and association with inflammatory mediators was assessed by Spearman correlation.
Results Patients with RA and control participants had similar office BP, but median ambulatory systolic BP (SBP) measurements (24-hour [RA 121 mmHg vs control 116 mmHg; P = 0.01], diurnal [RA 128 mmHg vs control 120 mmHg; P = 0.003], and nocturnal [RA 112 mmHg vs control 103 mmHg; P = 0.002]) were higher in patients with RA. Patients with RA also had higher nocturnal diastolic BP (DBP; RA 63 mmHg vs control 57 mmHg; P = 0.02), but other DBP measurements were similar. Nocturnal BP dipping was less in patients with RA (12%) compared to control participants (16%; P = 0.02). In patients with RA, higher 24-hour and nocturnal SBPs and less nocturnal dipping were strongly correlated with a wide range of inflammatory mediators.
Conclusion Despite similar office measurements, 24-hour and nocturnal SBP measurements were higher in patients with RA than in control participants and were strongly associated with inflammation.
The prevalence of hypertension (HTN) is approximately 80% higher in patients with rheumatoid arthritis (RA) compared to control subjects.1 This is important because HTN is a major contributor to the 48% increased incidence of cardiovascular disease (CVD) present in patients with RA compared to the general population.2 For example, in a metaanalysis of 7872 patients with RA, the presence of HTN was associated with 84% higher risk of myocardial infarction compared to patients with RA without HTN.3 Although awareness of the importance of controlling traditional risk factors such as HTN in RA has increased, office blood pressure (BP) measurements provide limited information.
Ambulatory 24-hour BP monitoring is helpful to determine whether antihypertensive medications are controlling BP sufficiently and to identify white-coat HTN and masked HTN.4 Ambulatory 24-hour BP monitoring also measures nocturnal BP, which is a risk factor for cardiovascular (CV) mortality that is independent of the average 24-hour BP and traditional risk factors.5
We previously found that ambulatory 24-hour BP was higher in patients with systemic lupus erythematosus (SLE) compared to control participants, and that the nocturnal systolic BP (SBP) was the most significantly elevated measure and was strongly associated with markers of immune activation.6 Moreover, several studies suggest that elevated nocturnal BP is associated with autoimmune disease.6-8 Immune activation may be a driving factor for excess HTN in patients with autoimmune diseases such as SLE and RA. Indeed, mice lacking B and T cells are resistant to developing HTN, demonstrating that the immune system plays a major role in HTN.9
The purpose of this study was to determine if ambulatory 24-hour BP measurements are (1) elevated in patients with RA vs control subjects and (2) associated with immune activation in RA.
METHODS
Study subjects. The study population included 46 patients with RA and 23 control participants enrolled in the Chronic Heart Outcomes in Rheumatic Diseases study at Vanderbilt University Medical Center. Patients with RA were recruited from the Vanderbilt University Medical Center Rheumatology Clinic and control participants were recruited from a word-of-mouth referral and a volunteer database maintained by Vanderbilt University Medical Center. Subjects were frequency matched for age within 5 years and sex. Patients with RA met the 2010 American College of Rheumatology (ACR)/European Alliance of Associations for Rheumatology (EULAR) classification criteria for RA.10 Control participants could not have a diagnosis of RA or other inflammatory autoimmune disease but could have other conditions such as HTN and other CVDs. All study participants had to be ≥ 18 years of age and provide written informed consent. Individuals were excluded if they could not operate the 24-hour BP device, had atrial fibrillation, were receiving an anticoagulant, or had a condition such as lymphedema that could be potentially worsened by the 24-hour BP measurement. All patients provided informed consent to participate in the study. This study was approved by the Vanderbilt University Medical Center Institutional Review Board (no. 110365).
Clinical assessments. Clinical information was obtained from the participants and their medical records. The office BP at the time of the study visit was measured twice after participants were seated and at rest for at least 5 minutes and the average of the 2 measurements was used. HTN was defined as a prior clinical diagnosis of HTN or an office BP of ≥ 140/90 mmHg at the time of the study. Venous blood was drawn for measurement of cytokines and high-sensitivity C-reactive protein (hsCRP), which was measured by the Vanderbilt University Medical Center laboratory.
Twenty-four–hour ambulatory BP measurement. Ambulatory BP was measured every 15-30 minutes during the day and every 30 minutes at night (10:00 PM to 6:00 AM) using the card(x)plore BP monitor (Meditech). We defined diurnal and nocturnal readings based on each participant’s reported sleep period. The percent nocturnal dip was calculated as 100 × (1 − [average nocturnal/average diurnal SBP]). Participants who did not achieve ≥ 10% nocturnal dip were classified as “nondippers.”11,12 Masked HTN was defined as not having diagnosis of HTN or office BP ≥ 140/90 mmHg, but having average 24-hour BP ≥ 130/80 mmHg, diurnal BP ≥ 135/85 mmHg, or nocturnal BP ≥ 120/70 mmHg.13 White-coat HTN was defined as having no prior diagnosis of HTN or use of antihypertensive agent, and having an office BP ≥ 140/90 mmHg but having an average 24-hour BP < 130/80 mmHg, diurnal BP < 135/85 mmHg, or nocturnal BP < 120/70 mmHg.4
Cytokine measurement. In patients with RA, proinflammatory cytokines and inflammatory mediators were measured in plasma by the Olink Target 96 Inflammation panel, which is a multiplex immunoassay (Olink Bioscience). Only those cytokines and inflammatory mediators that were above the limit of detection for > 75% of study participants were included for analysis in the study.
Statistical methods. In a prior study, the average of the 24-hour SBP was approximately 124 (SD 14) mmHg.5 Based on these data, 46 patients with RA and 23 control participants would provide 80% power to detect an average difference of approximately 10 mmHg between patients with RA and control subjects. We studied more patients with RA than controls to better define the relationships between cytokines and BP in RA. Sample size and power were calculated using the PS Power and Sample Size Program version 3.1.6.14 Data analysis was conducted using SPSS version 27 (IBM Corp.). Heatmaps demonstrating Spearman correlation and cluster analysis between Olink analytes and BP were created with R version 4.3.1 (R Foundation for Statistical Computing). Data are expressed as median (IQR) and n (%) unless otherwise specified.
RESULTS
Subjects. Patients with RA and control participants were of similar age (median 52 [IQR 40-58] and 50 [IQR 35-58] years, respectively) and racial and ethnic backgrounds (Table 1). More patients with RA had HTN (43%) compared to control participants (13%). Similarly, 43% of patients with RA and 13% of control subjects used an antihypertensive drug. Patients with RA had significantly higher hsCRP than controls (2.65 [IQR 0.98-9.75] mg/L vs 1.10 [IQR 0.60-2.30] mg/L; P = 0.02). RA disease activity tended to be low to moderate based on the Disease Activity Score in 28 joints based on C-reactive protein (DAS28-CRP; 2.94 [IQR 2.19-3.60] units). Commonly used RA drugs were methotrexate (MTX; 61%), tumor necrosis factor inhibitor (TNFi) agents (54%), and leflunomide (LEF; 15%; Table 1).
Demographics of patients with RA and control subjects.
Office and 24-hour ambulatory BP measurements in patients with RA compared to control participants. Office SBP was similar in patients with RA and control participants (122 [IQR 112-139] mmHg and 121 [IQR 116-126] mmHg, respectively; P = 0.61), as was office diastolic BP (DBP; Table 2).
Office and 24-hour BP measurements in patients with RA and control subjects.
In contrast, 24-hour average SBP was significantly higher in patients with RA than in control participants (121 [IQR 114-133] mmHg and 116 [IQR 110-120] mmHg, respectively; P = 0.01; Table 2). Both diurnal SBP (RA 128 [IQR 119-136] mmHg vs control 120 [IQR 114-126] mmHg; P = 0.003), and nocturnal SBP (RA 112 [IQR 102-126] mmHg vs control 103 [95-109] mmHg; P = 0.002) were higher in patients with RA than in control participants (Table 2 and Figure 1). Nocturnal DBP was significantly higher in patients with RA (63 [IQR 58-69] mmHg) than in control participants (57 [IQR 52-64] mmHg; P = 0.02), but the 24-hour average DBPs and diurnal DBPs were not (Table 2).
Average nocturnal SBP was significantly elevated in RA (N = 46) vs control subjects (N = 23). Box plots represent median and 25th and 75th percentiles, and whiskers represent the range excluding outliers > 1.5× above or below the IQR. Significance was tested using Mann-Whitney U test. P = 0.002. RA: rheumatoid arthritis; SBP: systolic blood pressure.
The nocturnal dip percentage, which examines the beneficial drop in BP that should occur overnight,15,16 was significantly smaller in patients with RA (12% [IQR 6-17]) than in control participants (16% [IQR 10-21]; P = 0.02). A total of 43% of patients with RA and 22% of control participants were classified as nondippers (P = 0.08; Table 2). Masked HTN was present in 31% of patients with RA and 20% of control participants without known HTN (P = 0.51; Table 2). Very few (2 RA and 0 control) subjects had elevated office BP without a prior diagnosis of HTN or use of an antihypertensive agent. Among these 2 subjects with RA, ambulatory BP monitoring confirmed a diagnosis of HTN in 1 and white-coat HTN in the other.
The differences in the ambulatory BP measurements remained significant independent of age in adjusted analyses (Table 2).
Twenty-four–hour ambulatory BP measurements based on RA medication use. Among patients with RA, use of MTX was associated with a significantly greater percent nocturnal dip (13% [IQR 8-18]) compared to those who did not use MTX (9% [IQR 3-14]; P = 0.02), and this remained significant independent of age (P = 0.02). There were no other significant differences in the ambulatory BP measurements based on MTX use (Supplementary Table S1, available with the online version of this article). There were no significant differences in ambulatory BP measurements among patients with RA who did and did not use LEF, TNFi agents, glucocorticoids (GCs), or nonsteroidal antiinflammatory drugs (all P > 0.05).
Relationship between 24-hour ambulatory BP measurements and inflammation. There was strong positive correlation between numerous markers of inflammation measured by the Olink Inflammation 96 panel and 24-hour ambulatory BP measurements in patients with RA (Figure 2; Supplementary Table S2, available with the online version of this article). In particular, the average 24-hour SBP and nocturnal SBPs were strongly positively associated with many analytes; similarly, less (ie, worse) nocturnal dipping was also associated with higher concentrations of many analytes in patients with RA (Supplementary Table S2). The broad association of nocturnal SBP with numerous inflammatory mediators precluded examining specific pathway contributions. Two of the analytes, CUB-domain containing protein 1 (CDCP1) and osteoprotegerin (OPG), were not only significantly associated with 24-hour SBP and nocturnal SBP in this RA study, but also in a previous, similar study in SLE.6
Relationship between BP measurements and inflammation among patients with RA, measured using the Olink Inflammation 96 panel using serum. Olink analytes, which were detectible in > 75% of patient samples, were analyzed. Heatmap demonstrates Spearman ρ and cluster analysis. Red indicates positive correlation. Blue indicates negative correlation. BP: blood pressure; DBP: diastolic BP; RA: rheumatoid arthritis; SBP: systolic BP.
There was no significant association between measures of RA disease activity, such as tender and swollen joint counts, DAS28-CRP scores, or hsCRP concentrations, and office or 24-hour ambulatory BP measurements (all P > 0.05; Supplementary Table S3, available with the online version of this article).
DISCUSSION
This study found that although office BP levels were similar among patients with RA and control participants, patients with RA had significantly higher average 24-hour, diurnal, and nocturnal SBP; higher average nocturnal DBP; and impaired nocturnal dipping compared to control participants. The study also found that plasma levels of many immune mediators were strongly associated with higher 24-hour SBP and nocturnal SBP as well as a smaller nocturnal dip in patients with RA.
Because of increased awareness of higher CV risk among patients with RA, rheumatologists and primary care physicians have increased their efforts to address treatable CV risk factors, such as HTN. This may be why, despite the higher prevalence of HTN in patients with RA, office BP measurements were similar in patients and control participants. However, ambulatory monitoring revealed significant differences in BP measurements, particularly nocturnal SBP, between patients with RA and control subjects. Ambulatory BP monitoring identifies a risk factor5 that may contribute to the excess CV risk observed in patients with RA, but which is seldom measured in clinical practice.
In a population-based cohort of > 11,000 adults with a median of 13.8 years follow-up, every 20 mmHg increase in office SBP was associated with a 19% increase in CV mortality after adjustment for standard CV risk factors including sex, age, BMI, smoking, alcohol use, serum cholesterol, antihypertensive use, history of CVD, and diabetes mellitus.5 In contrast, every 20 mmHg increase in 24-hour average SBP was associated with a 48% increase in CV mortality after adjustment for standard CV risk factors.5 Moreover, even after adjusting for 24-hour average SBP and the other standard CV risk factors used in the prior analyses, every 20 mmHg increase in nocturnal SBP was associated with 26% increase in CV mortality.5 Thus, higher 24-hour ambulatory SBP and particularly nocturnal SBP in patients with RA may contribute substantially to the excess CV mortality in RA and deserve further investigation.
The pattern of altered ambulatory BP findings in the current study are similar to those observed in a previous study we performed in patients with SLE. In that study, as was the case in the current study, despite similar office BP measurements, the 24-hour and nocturnal SBPs were higher and the percent nocturnal dip was smaller in patients with SLE than in control participants.6 Further, in both studies of patients with autoimmune disease, inflammatory mediators were most strongly linked to nocturnal SBP.
A similar ambulatory BP pattern was also observed by others in a study of 91 patients with RA (average DAS28 score = 3.5 units) and 50 control participants without HTN or any known health problems. The patients with RA and control participants also had similar office SBP, but patients with RA had higher nocturnal SBP and a smaller nocturnal dip. In that study, the impaired nocturnal dipping was associated with higher age, erythrocyte sedimentation rate, DAS28 score, and current use of GCs.7 Moreover, an uncontrolled study of 71 patients with RA (average DAS28-CRP = 4.8 units) reported a nocturnal BP dip (average 5.6%) that was inversely associated with DAS28-CRP and was smaller than the nocturnal dip in BP in the current study (median 12%).8
The observation that nocturnal BP is strongly associated with a range of inflammatory mediators in RA and SLE is intriguing because it may reveal underlying immune mechanisms and suggest novel treatment options. Nocturnal HTN has been linked to risk factors such as poor sleep quality, high-salt diet, reduced physical activity, and diseases such as diabetes mellitus, obstructive sleep apnea, and chronic kidney disease.17 Addressing these underlying risk factors may be helpful to control the nocturnal BP and dipping.17 Also, 1 study reported that taking BP medications in the evening improved nocturnal BP control and nocturnal dipping and reduced CV events by 45% compared to daytime administration18; however, results of that study have been questioned.19 Nevertheless, our findings highlight the potential associations between autoimmune disease and BP, particularly nocturnal BP.
Further questions include whether more aggressive RA treatment would lower BP and which immunomodulatory therapies might preferentially affect BP. Some disease-modifying antirheumatic drugs (DMARDs) used to treat RA may affect BP positively or negatively. For example, LEF commonly increases BP despite controlling inflammation.20 In contrast, MTX and hydroxychloroquine decreased office BP in an observational prospective study of veterans with RA, whereas other DMARDs, such as TNFi agents and LEF, demonstrated little improvement or higher office BP.21 In the current study, MTX users had a larger nocturnal dip in BP than nonusers and a prior observational study of 56 MTX and 30 non-MTX users observed that MTX users had lower 24-hour and diurnal SBPs, but nocturnal dipping was not reported.22
Other DMARDs may have beneficial effects on ambulatory BP measurements, although studies are conflicting. In a study of 25 hospitalized patients with RA who started a DMARD (most commonly a TNFi agent) for routine clinical care, nocturnal dipping was increased and nocturnal SBP was decreased without significant change in diurnal BP after 4 weeks of treatment.8 Another study of 16 hospitalized patients with RA who started the TNFi agent infliximab as part of routine care reported a drop of approximately 7 mmHg in average 24-hour SBP after 2 weeks of treatment, which was mainly attributable to changes in diurnal BP rather than nocturnal BP,23 and it was not correlated with changes in DAS28-CRP or its components, such as CRP.23 Another uncontrolled study of 15 patients with rheumatic disease (4 RA, 5 psoriatic arthritis, 2 ankylosing spondylitis, and 2 inflammatory bowel disease–associated arthritis) who were starting a TNFi agent as part of routine care showed no effect on the 24-hour BP measurements 3 months after starting the study.24 The current cross-sectional study found no difference based on TNFi agent use. The studies demonstrating an effect of TNFi agents were among hospitalized patients without a comparator DMARD, making it challenging to know if the change was attributable to the TNFi agent or general changes in a hospitalized patient. It is possible that targeting other inflammatory mediators might result in more clear results.
In the current study, there was no significant correlation of any BP measurements with DAS28-CRP or hsCRP, which was possibly due to lower disease activity in the current study population or confounding effects of medications. However, many inflammatory mediators were associated with BP, particularly nocturnal SBP in RA and impaired nocturnal dipping.
Among the inflammatory mediators examined, this study and a prior SLE study6 observed that CDCP1 and OPG were significantly associated with 24-hour SBP and nocturnal SBP both in patients with RA and SLE. Also known as CD318, CDCP1 is a potential target for treatment of autoimmune diseases and is present on dendritic cells,25 epithelial cells, a variety of tumor cells, and synovial tissues, and its soluble form is found in synovial fluid of patients with RA.26 CDCP1 is a ligand for CD6, which regulates T cells.26 An anti-CD6 agent is in development for RA,27 psoriasis,28 and coronavirus disease 2019,29-31 with some promising results; however, BP changes have not been reported in these studies and to our knowledge there have been no clinical trials targeting CDCP1. There is less information on CDCP1 and its associated proteins, such as CD6, with BP.
The other protein identified, OPG, is known for its role in inhibiting bone resorption. It is regulated in part by inflammatory cytokines such as TNF and decreases with use of TNFi drugs.32 We previously found that OPG was elevated in patients with RA and associated with SBP and coronary atherosclerosis.33 Other studies reported that OPG was significantly increased in patients with HTN who were nondippers compared to those whose nocturnal BP dipped at least 10%, and it was associated with hypertensive damage, such as left ventricular systolic dysfunction.34 The relationship between OPG and BP has been attributed to OPG decreasing nitric oxide release, leading to elevated BP.35 Anti-OPG antibody therapy reversed pulmonary HTN in mice, interestingly without adverse bone effects,36 but its effect on systemic BP is unknown to our knowledge.
Strengths of our study include the use of control subjects who represented people of similar age and sex who had similar comorbidities compared to patients with RA. For example, participants with HTN and other CVDs were permitted among both RA and control groups. Another strength is that we examined a broad range of inflammatory mediators in RA. The major limitations of our study were that patients with RA had relatively well-controlled disease, as is the standard of care these days, and that the sample size—although appropriately powered to observe a significant difference in 24-hour average SBP—had limited power to address the effects of particular medications or to consider subgroups of patients. Another limitation is that we did not examine home BP monitoring, which may be more convenient to perform compared to ambulatory BP monitoring. Last, the study was not designed to conclusively determine if elevated nocturnal BP observed in patients with RA may have been due to pain and interrupted sleep from RA, though the BP measurements were not associated with typical RA disease activity measures such as DAS28-CRP score.
In conclusion, patients with RA had higher nocturnal and 24-hour SBP than control subjects even though office BP did not differ in the 2 groups, and higher nocturnal BP and a small nocturnal dip in BP were associated with a range of inflammatory mediators.
Footnotes
This study was funded by the following grants: National Institutes of Health (NIH) National Heart, Lung, and Blood Institute no. R01 HL140145 (to CMS); NIH National Institute of Arthritis and Musculoskeletal and Skin Diseases no. R21 AR080372 (to MJO); and NIH National Center for Advancing Translational Sciences no. 5UL1TR002243-03.
The authors declare no conflicts of interest relevant to this article.
- Accepted for publication May 6, 2024.
- Copyright © 2024 by the Journal of Rheumatology








