Sage Journals: Discover world-class research

Abstract

Background:

Hypertension (HTN) is the most important modifiable risk factor for the development of cardiovascular events (CVEs). Patients with axSpA are also associated with an increased risk of future CVE.

Objectives:

To ascertain whether baseline early-stage HTN is a predictor of future CVE in addition to inflammation in patients with axial spondyloarthritis (axSpA).

Design:

A retrospective cohort study.

Methods:

Patients with axSpA were recruited from 2001 to 2017. Patients with at least 2 years of follow-up and without prior CVE were divided into three groups according to the calculated mean blood pressure (BP) over the first 2-year follow-up period (adjusted mean BP) (⩾140/90, 130–139/80–89, and <130/80 mm Hg). They were followed from baseline until the end of 2020 or the occurrence of a first CVE. Multivariate Cox regression analyses adjusting for baseline and time-varying variables were used to assess the relationship between mean BP and CVE.

Results:

Out of the 437 patients fulfilling the inclusion criteria, 49 (11.2%) and 132 (30.2%) had an adjusted mean BP ⩾ 140/90 and 130–139/80–89 mm Hg, respectively, and 256 (58.6%) were pre-HTN. After a median follow-up of 12 (7–18) years, 56 (12.8%) CVEs were documented. The incidence rates were 21.4, 14.2, and 5.9 per 1000 patient-years for the three groups, respectively. Baseline adjusted mean BP of 130–139/80–89 mm Hg was independently associated with the occurrence of CVE after adjusting for the baseline covariates as well as time-varying high inflammatory burden.

Conclusion:

Baseline-defined early-stage HTN carries excessive risk of developing CVE which may be due to untreated inflammatory burden. Early antihypertensive therapy should target this BP level to minimize their future risk of CVE.

Graphical abstract

Keywords

2017 ACC/AHA classification axial spondyloarthritis cardiovascular risk hypertension

Introduction

It is well established that patients with axial spondyloarthritis (axSpA) have an increased cardiovascular (CV) risk compared with the general population.^1,2 Various conventional CV risk prediction algorithms underestimated CV risk in rheumatoid arthritis (RA) and psoriatic arthritis (PsA), most likely because the inflammatory component was not considered.^3–6 The European Alliance of Associations for Rheumatology (EULAR) recommended an intrinsic multiplication factor of 1.5 for patients with RA.⁷ However, the modified CV risk scores still underestimated CV risk in axSpA patients.^8,9 Therefore, there is an increasing recognition that inflammation plays an important role in driving the development of CV events (CVE) in addition to traditional CV risk factors in inflammatory arthritis.^9,10

As one of the most important modifiable risk factors for the development of CVE,^11,12 hypertension (HTN) can be detected in up to 40% of patients with axSpA.^13,14 Higher systolic blood pressure (SBP) is consistently associated with an increased risk of cardiovascular disease (CVD).¹⁵ A meta-analysis including 61 prospective studies reported that each 20 mm Hg increase in SBP (or each 10 mm Hg increase in diastolic blood pressure (DBP)) was associated with a twofold increased risk of death from CVD.¹⁶ Therefore, the American College of Cardiology/American Heart Association (ACC/AHA) recommended a lower threshold of SBP and DBP for arterial HTN diagnosis in adults in 2017.¹⁷ Accordingly, blood pressure (BP) of 130–139/80–89 mm Hg is classified as early-stage HTN and BP ⩾140/90 mm Hg is classified as stage 2 HTN. With this new definition, the prevalence of HTN increased by 15.5%–49.6%.^18,19 However, there were controversies regarding its clinical application.^20–23

Of note, a more recent meta-analysis of randomized clinical trials (RCTs) revealed that lowering BP in patients with baseline SBP less than 140 mmHg was not beneficial in reducing mortality and CVD.²⁴ As our previous study showed that baseline inflammatory burden (baseline disease duration and delay of diagnosis) is associated with incident HTN in patients with axSpA,²⁵ it is possible that the baseline state of early-stage HTN (130–139/80–89 mm Hg) may carry excessive risk of future CVE. A recent study has confirmed that patients with systemic lupus erythematosus (SLE) with a baseline-defined mean BP of 130–139/80–89 mm Hg under this new definition had a significantly higher incidence of atherosclerotic vascular events compared with the normotensives. However, the association between baseline-defined early-stage HTN with future CVE in addition to time-dependent inflammation remains unknown.

Given the relationship between untreated inflammatory burden and future incident HTN in our previous study, we hypothesize that baseline-defined early-stage HTN is associated with excess CVE in patients with axSpA. The present study was designed to address whether a BP of 130–139/80–89 mm Hg at baseline is a predictor of future CVE after controlling baseline traditional CV risk factors and time-dependent inflammation.

Methods

Study design and ethics

This is a retrospective cohort study to address the relationship between early-stage HTN and CVE in axSpA. Data were retrieved from case notes and the city-wide electronic health record (EHR) database. Briefly, patients who were diagnosed with axial spondylarthritis and fulfilled the 2009 ASAS classification criteria for axSpA,²⁶ and aged between 18 and 60 years were recruited from the rheumatology clinic of a regional hospital in Hong Kong from January 2001 to December 2017. The reason we established the upper age limit cutoff is based on a meta-analysis that reported a higher age-specific hazard ratio (HR) for CV death within this age group compared to older individuals in the general population.¹⁶ All patients should have at least 2 years of follow-up after recruitment, and at least one clinic visit afterward. Patients with an established CVE or with missing BP measurements during the first 2 years were excluded. The baseline year was defined as the beginning of the 3rd year time point after recruitment. Patients who lost to the follow-up were also excluded. For patients diagnosed before 2001, the baseline year was 2003; for those who were diagnosed after 2001, the baseline year was 2 years after diagnosis. All individuals were followed from the baseline year until the occurrence of the first CVE or the end of 2020.

The study was approved by The Joint Chinese University of Hong Kong—New Territories East Cluster Clinical Research Ethics Committee (No. 2020. 519). The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.²⁷ The study was conducted in compliance with the Declaration of Helsinki and the International Council for Harmonisation guideline for Good Clinical Practice. Written informed consents were waived.

BP categories

Recorded BP (SBP/DBP) (measured in the sitting position with an aneroid sphygmomanometer after at least 5 min rest) at each clinic visit and the antihypertensive medication history were retrieved from the city-wide EHR. To adjust for the fluctuation of BP over time, an “adjusted mean BP” was defined at baseline and calculated from BP measurements over a period of 2 years prior to the baseline visit for all enrolled axSpA patients in this study. The calculation has been validated and elaborated previously.^28,29 The formula was:

\frac{\sum_{i = 2}^{n} (\frac{x_{i} + x_{i - 1}}{2}) t_{i}}{\sum_{i = 2}^{n} t_{i}}

In short, x_i and x_i_− 1 represented the SBP or DBP level at visit i and i − 1, and t_i represented the time interval between visit i − 1 and i. By incorporating the t_i in its calculation, the adjusted mean BP considered the length of time that SBP and DBP were presumed to have remained at a particular level.

Patients were then categorized into three different groups according to their baseline adjusted mean BP: Group 1: adjusted mean BP <130/80 mm Hg (normotensives); Group 2: adjusted mean BP 130–139/80–89 mm Hg (early-stage HTN); and Group 3: adjusted mean BP ⩾140/90 mm Hg (HTN).

BP measurement at each visit

BP was measured by a nurse or trained healthcare assistant in the non-dominant arm using a digital automatic BP monitor (Colin 8800 NIBP monitor [Colin Medical Instruments Corp., USA] with a cuff of the appropriate size, which was subsequently replaced by the A&D Medical TM-2657-P [A&D Company, Limited, Japan]) following standard recommended procedures. Measurement was initiated after a 5-min rest. If the BP reading was suboptimal, an extra measurement would be subsequently implemented. If both BP readings are suboptimal, clinicians may measure extra BP readings for the patients and select the most appropriate BP reading based on clinical judgment.

Clinical data

Each patient was followed at 6–12 months intervals and clinical data were regularly updated. Inflammatory markers including erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) were measured. Disease activity and function were assessed using the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI)³⁰ and Bath Ankylosing Spondylitis Functional Index (BASFI),³¹ respectively. At baseline, demographics (age, sex, smoking, and drinking history), clinical variables (disease duration, disease activity, inflammatory markers), traditional CV risk factors, and medication history were retrieved from case notes and the EHR.

Time-varying inflammatory burden (BASDAI, ESR, and CRP) and drug history (disease-modifying anti-rheumatic drugs (DMARDs) including the use of biologic DMARDs (bDMARDs), conventional synthetic DMARDs (csDMARDs), nonsteroidal anti-inflammatory drugs (NSAIDs)) were also retrieved from the EHR at a yearly interval starting from baseline visit till the end of the study.

Endpoint

The study endpoint was defined as the first occurrence of a CVE, including angina, myocardial infarction (MI), ischemic heart disease (IHD), ischemic and hemorrhagic stroke, acute and chronic heart failure (HF), transient ischemic attack (TIA), percutaneous transluminal coronary angioplasty or carotid endarterectomy, and CVE-related deaths. These data were retrieved from the EHR.

Statistical analysis

Variables were presented as counts (percentages), mean ± SD or for normal distributed data, or median with interquartile range for non-normally distributed data. The baseline demographic and clinical characteristics of the three groups of patients with different adjusted mean BP were compared using one-way ANOVA analysis for continuous variables and χ² tests for categorical variables. When a significant difference (p < 0.05) was observed, the Scheffe test was used for post hoc comparison at a statistically significant adjusted level of p < 0.0083. CVE-free survival for the three groups was assessed using the Kaplan–Meier survival analysis. The survival distribution was compared using the log-rank test. Univariate and multivariable Cox proportional hazards regressions were used to investigate the association between the development of a first CVE and the potential risk factors (baseline characteristics listed in Table 1 with p < 0.05) after adjusting for the baseline CV risk scores (which were used to account for traditional CV risk factors). Baseline statistical analysis was performed using SPSS V.26.0 (IBM, USA) for Windows. Inflammatory markers, BASDAI, and drug use were also analyzed as time-dependent predictors (being updated at each visit) using time-varying Cox proportional hazard regression analyses (R version 4.2, available at https://www.r-project.org/, in the package named “survival” and “survminer” with the counting process method). p < 0.05 was considered significant.

Table 1.

Baseline characteristics of different hypertension categories.

Variables	Group 1, <130/80 mm Hg (N = 256)	Group 2, 130–139/80–89 mm Hg (N = 132)	Group 3, ⩾140/90 mm Hg (N = 49)	p-Value
Cardiovascular risk factors
Gender (male), n (%)	182 (71.1)	118 (89.4)**	40 (81.6)	<0.001*
Age (years)	32 (11)	37 (11)**	42 (10),*	<0.001*
BMI (kg/m²)	26 (5)	27 (5)	29 (5)	0.207
Smoker (current), n (%)	68 (26.6)	35 (26.5)	9 (18.4)	0.467
Drinker, n (%)	47 (18.4)	40 (30.3)**	6 (12.2)	0.006*
Adjusted mean systolic BP (mm Hg)	115 (109–122)	132 (127–135)**	143 (140–145),*	<0.001*
Adjusted mean diastolic BP (mm Hg)	70 (65–75)	81 (80–84)**	90 (82–91),*	<0.001*
Total cholesterol (mmol/L)	4.6 (0.9)	4.9 (0.9)**	4.8 (0.9)	0.008*
Triglyceride (mmol/L)	0.9 (0.7–1.3)	1 (0.8–1.6)	1 (0.8–1.7)**	0.002*
HDL (mmol/L)	2.1 (10.4)	1.4 (0.4)	1.4 (0.4)	0.676
LDL (mmol/L)	2.7 (2.2–3.1)	2.9 (1.3)**	2.9 (2.2–3.3)	0.014*
Fasting glucose (mmol/L)	5.1 (0.9)	5.5 (4.9–5.6)**	5.6 (1.6)**	0.001*
TC/HDL ratio	3.4 (1.1)	3.8 (1.3)**	3.6 (1.2)	0.007*
FRS^a	1.5 (0.6–3.3)	3.8 (1.6–7.1)**	5.5 (3.2–9.6)**	<0.001*
QRISK3^a	2.2 (4.4)	2.4 (4)	2.6 (2.9)	0.722
SCORE^a	1 (1–1)	1 (1–2)**	1 (1–3)**	<0.001*
Disease-related parameters
Disease duration (years)	0 (0–1)	0 (0–2)**	0 (0–2)	0.021*
Radiological axSpA, n (%)	200 (78.1)	108 (81.8)	45 (91.8)**	0.078
Family history of SpA, n (%)	45 (17.6)	24 (17.4)	12 (20.4)	0.884
HLA-B27+ve, n (%)	169 (66)	79 (59.8)	24 (49)	0.424
BASDAI (0–10),	4.8 (2.3)	4.7 (2.3)	5.5 (2.1)	0.103
BASDAI ⩾4, n (%)	162 (63.3)	84 (63.6)	37 (75.5)	0.247
BASDAI ⩾5, n (%)	120 (46.9)	66 (50)	32 (65.3)	0.062
BASDAI ⩾6, n (%)	87 (34)	46 (34.8)	23 (46.9)	0.217
BASFI (0–10)	4 (2.6)	4.3 (2.5)	4.6 (2.5)	0.233
Peripheral arthritis, n (%)	62 (24.2)	32 (24.2)	12 (24.5)	0.999
Enthesitis, n (%)	36 (14.1)	19 (14.4)	5 (10.2)	0.746
Dactylitis, n (%)	6 (2.3)	6 (4.5)	2 (4.1)	0.474
Uveitis, n (%)	41 (16)	29 (22)	9 (18.4)	0.352
CRP (mg/L)	20.3 (2.9–26.1)	13 (5.2–26.5)	11.7 (6.2–30.6)	0.455
ESR (mm/first hour) ESR ⩾20, n (%)	25.5 (12–46) 154 (60.2)	26.5 (14–45.5) 87 (65.9)	35 (16–71) 34 (69.4)	0.084 0.330
Medications
NSAIDs, n (%)	208 (81.3)	113 (85.6)	39 (79.6)	0.489
COX-2, n (%)	36 (14.1)	16 (12.1)	5 (10.2)	0.712
csDMARDs, n (%)	57 (22.3)	28 (21.2)	7 (14.3)	0.455
bDMARDs, n (%)	40 (15.6)	30 (22.7)	4 (8.2)***	0.047*
Antihypertensives, n (%)	6 (2.3)	18 (13.6)**	11 (22.4)**	<0.001*
Statin, n (%)	2 (0.8)	7 (5.3)**	1 (2)	0.019*
Steroids, n (%)	5 (2)	3 (2.3)	3 (6.1)	0.228

Data are expressed as n (%), mean (SD), or median (IQR).

Baseline data of 437, 437, and 372 patients were available for calculating the FRS, SCORE, and QRISK3, respectively.

Statistically significant at p < 0.05 between all groups.

Statistically significant at p < 0.0125 with group 1.

***

Statistically significant at p < 0.0125 with group 2.

Results

Baseline characteristics in different HTN category groups

Of the 437 patients included, 256 (58.6%) were normotensive (group 1), 132 patients (30.2%) had baseline-defined early-stage HTN (group 2), and 49 patients (11.2%) had baseline-defined HTN (group 3) at baseline. The baseline characteristics of different BP groups are shown in Table 1. Compared to the normotensive patients (group 1), patients with HTN (groups 2 and 3) were older and had higher sugar and lipid levels. Compared to group 1, longer disease duration and a higher prevalence of radiographic sacroiliitis were observed in groups 2 and 3, respectively. Fewer patients in group 3 (8.2%) used bDMARDs compared with group 2 (22.7%) (p = 0.047).

Association of baseline adjusted mean BP and future CVE

After a median follow-up of 12 (7–18) years (a total of 5494 patient-years), 56 (12.8%) CVEs were documented, including 7 (1.6%) CVE-related deaths. The incidence rate of new-onset CVEs is 10.2 cases per 1000 person-years. Recorded CVEs were IHD (n = 18), CHF (n = 12), ischemic stroke (n = 11), acute MI (n = 7) (including MI-related revascularization procedures (n = 3)), unstable angina (n = 3), TIA (n = 3), and hemorrhagic stroke (n = 2). The overall incidence of CVEs was 13 (26.5%), 24 (18.1%), and 19 (7.4%) and the incidence rates were 21.4, 14.2, and 5.9 per 1000 patient-years in group 3, group 2, and group 1, respectively. The Kaplan–Meier curves for the CVE-free survival (censored at 18 years) of the three groups are shown in Figure 1, demonstrating significant differences between the three groups (p < 0.001) using the log-rank test. Post hoc analysis showed that significant differences were present between group 1 and group 2 (p = 0.005) and between group 1 and group 3 (p < 0.001). There was no difference between groups 2 and 3 (p = 0.188).

Figure 1.

Kaplan–Meier analysis for CVE-free cumulative survival among the three patient groups (green line: ⩾140/90 mm Hg, red line: 130–139/80–89 mm Hg, blue line < 130/80 mm Hg). The differences were statistically significant among all groups (p = 0.003* between group 1 and group 2, log-rank χ² (1 df) = 9.075; p < 0.001* between group 1 and group 3, log-rank χ² (1 df) = 17.326; p < 0.001* among three groups, log-rank χ² (1 df) = 18.048). There is no difference between groups 2 and 3 (p = 0.143, log-rank χ² = 2.151).

The univariate Cox regression model revealed that both HTN groups (groups 2 and 3) had a significantly increased risk of developing a first CVE (Table 2). In the multivariable Cox regression models, patients in group 2 (HR: 2.09–2.35) and group 3 (HR: 3.17–3.80) remained at higher risk of developing CVE after adjusting for traditional CV risk scores, inflammation, and treatment used at baseline (Table 2).

Table 2.

Univariate and multivariate Cox regression analyses for baseline predictors of CVE.

<130/80	HR (95% CI)	p-Value	Model 1		Model 2		Model 3
	HR (95% CI)	p-Value	HR (95% CI)	p-Value	HR (95% CI)	p-Value	HR (95% CI)	p-Value
	Ref		Ref	Ref	Ref	Ref	Ref	Ref
130–139/80–89	2.56 (1.34, 4.89)	0.004*	2.09 (1.06, 4.14)	0.034*	2.35 (1.15, 4.79)	0.019*	2.21 (1.12, 4.33)	0.022*
⩾140/90	4.29 (2.03, 9.09)	<0.001*	3.34 (1.53, 7.30)	0.002*	3.80 (1.69, 8.52)	0.001*	3.17 (1.43, 7.01)	0.004*
Baseline disease duration	1.02 (0.98, 1.06)	0.438	–	–	–	–	–	–
bDMARDs	0.50 (0.20, 1.25)	0.137	0.54 (0.21, 1.38)	0.195	0.60 (0.23, 1.55)	0.290	0.52 (0.20, 1.33)	0.173
Antihypertensives	2.83 (1.42, 5.67)	0.003*	0.93 (0.39, 2.21)	0.86	1.17 (0.52, 2.64)	0.699	1.35 (0.62, 2.91)	0.45
Statin	3.82 (1.37, 10.61)	0.01*	1.53 (0.40, 5.86)	0.537	3.22 (1.08, 9.55)	0.035*	2.67 (0.89, 7.99)	0.08
FRS	1.07 (1.05, 1.09)	<0.001*	1.06 (1.02, 1.09)	0.001*
QRISK3	1.05 (0.99, 1.12)	0.083			1.06 (0.99, 1.13)	0.122
SCORE	1.23 (1.11, 1.37)	<0.001*					1.16 (1.02, 1.31)	0.022*

Statistically significant at p < 0.05.

bDMARDS, biological disease-modifying antirheumatic drugs; CI, confidence interval; CVE, cardiovascular event; FRS, Framingham Risk Score; HR, hazard ratio; SCORE, systematic coronary risk evaluation.

Predictors for CVE in time-dependent analysis

In the time-dependent univariable Cox proportional hazard regression model (Table 3), groups 2 and 3 had a significantly increased CVE risk with HRs of 2.59 and 4.21, respectively. Higher inflammatory burden as reflected by ESR ⩾20 (p = 0.039, HR: 2.17, 95% confidence interval (CI): 1.04, 4.35) was associated with an increased risk of developing CVE.

Table 3.

Univariate analysis with time-dependent Cox proportional hazard regression for BP and CVE.

Time-varying variable^a	Total person-time interval	Time-dependent HR (95% CI)	p-Value^b
<130/80 mm Hg	5311	Ref	Ref
130–139/80–89 mm Hg		2.59 (1.36, 4.93)	0.004*
⩾140/90 mm Hg		4.21 (1.99, 8.90)	<0.001*
ESR mm/h	2981	1.01 (0.99, 1.02)	0.582
ESR ⩾20 mm/h		2.17 (1.04, 4.35)	0.039*
CRP mg/L	3136	0.99 (0.96, 1.02)	0.518
CRP >3 mg/L		1.24 (0.58, 2.68)	0.578
BASDAI	2038	1.17 (0.99, 1.39)	0.067
BASDAI ⩾ 4		2.21 (0.99, 4.95)	0.053
BASDAI ⩾ 6		1.38 (0.58, 3.29)	0.463
NSAIDs	5311	1.29 (0.74, 2.27)	0.366
Cox2		1.48 (0.66, 3.34)	0.341
NonCox2		1.21 (0.68, 2.17)	0.523
bDMARDs	5311	0.48 (0.12, 1.97)	0.307
csDMARDs	5311	1.43 (0.64, 3.20)	0.385
Age	N/A	1.10 (1.06, 1.13)	<0.001*
Sex	N/A	1.44 (0.64, 3.20)	0.376
Smoking	N/A	1.11 (0.61, 2.04)	0.729

Variables are time-varying unless otherwise specified.

Variables with a p-value of <0.05 are included in the multivariable analysis.

Statistically significant at p < 0.05.

BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; bDMARDS, Biological Disease-Modifying Antirheumatic Drugs; BP, blood pressure; CRP, C-reactive protein; csDMARDs, conventional synthetic disease-modifying antirheumatic drugs; CVE, cardiovascular event; ESR, erythrocyte sedimentation rate; HR, hazard ratio; NSAIDs, nonsteroidal anti-inflammatory drugs.

Using time-dependent multivariable Cox regression analysis, group 2 (HR: 1.94, 95% CI: 1.00, 3.75, p = 0.048 in model 1 and HR: 2.81, 95% CI: 1.13, 7.01, p = 0.026 in model 2) and group 3 (HR: 2.19, 95% CI: 1.01, 4.75, p = 0.046 in model 1 and HR: 4.34, 95% CI: 1.62, 11.61, p = 0.004 in model 2) remained at higher risk of developing CVE after adjusting for age, sex, and high inflammatory burden (Figure 2(a) and (b)). Moreover, high ESR level (ESR ⩾20) was also the independent predictor of future CVE (HR: 2.52, 95% CI: 1.04, 6.12, p = 0.041) (Figure 2(a) and (b)).

Figure 2.

Forest plot illustrating the multivariate time-varying Cox regression for the prediction for CVE with (Figure 2b) and without (Figure 2a) adjusting for inflammatory markers.

Subgroup analysis

In subgroup analysis, we stratify patients into different age group to ascertain the impact of early-stage HTN on CVE in different age groups. Group 2 (HR: 2.13, 95% CI: 1.10, 4.14, p = 0.026 in model 1 and HR: 3.07, 95% CI: 1.22, 7.70, p = 0.017 in model 2) and group 3 (HR: 2.29, 95% CI: 1.05, 4.97, p = 0.037 in model 1 and HR: 4.14, 95% CI: 1.54, 11.16, p = 0.005 in model 2) was independently associated with future CVE in the time-dependent multivariable analysis after adjustment for high inflammatory burden, age groups, and sex (Figure 3(a) and (b)). Patients who were aged 40–60 years (HR: 5.38–10.39) at presentation and high ESR level (HR: 2.44, 95% CI: 1.16, 5.10, p = 0.018) were also independent predictors for future CVE (Figure 3(a) and (b)).

Figure 3.

Forest plot illustrating the multivariate time-varying Cox regression for the prediction for CVE stratified by age groups with (Figure 3b) and without (Figure 3a) adjusting for inflammatory markers.

Discussion

Our study found that patients with baseline-defined early-stage HTN (group 2, adjusted mean BP 130–139/80–89 mm Hg) had a persistently increased risk of future CVE after adjusting for multiple baseline traditional risk factors and medications and time-dependent high inflammatory burden. As our previous findings have confirmed that baseline inflammatory burden (baseline disease duration and delay of diagnosis) is associated with incident HTN in patients with axSpA,²⁵ suggesting a potentially significant impact of untreated inflammatory burden at baseline, not time-varying inflammatory burden, on the increased BP. The increased BP may proceed to the state of early-stage HTN and lead to future CVE, which may be underestimated in clinical practice.

With the new definition of HTN, it is notable that axSpA patients with baseline-defined early-stage HTN and HTN had a higher prevalence of traditional and axSpA-related CV risk factors. When inflammation that changes over time are taken into consideration in the time-dependent Cox-regression analysis, baseline-defined early-stage HTN remains as significance predictor for future CVE. This finding suggests that in patients with chronic inflammatory arthritis such as axSpA, while inflammation is conformed to be associated with a relative increase in CV risk, there is an excessive risk of future CVE dependent on the state of early-stage HTN which may due to the effect of untreated inflammatory burden. Using the old definition of HTN (BP ⩾140/90), it has also been reported that inflammatory burden (longer disease duration and higher disease activity) in axSpA was associated with a higher prevalence of HTN.^32,33 Therefore, our findings together with previous evidence may provide more insights into how inflammatory burden in axSpA patients interacts with increased BP. Future studies are needed to address the association between untreated disease duration, delay of diagnosis, and the state of early-stage HTN in axSpA.

A meta-analysis suggested that higher SBP or DBP was associated with higher CVE-related death rates at younger age group (40–69 years) compared to older age group in the general population.¹⁶ As the majority of our axSpA patients in the clinic belong to this age group, subgroup analysis in our study found that baseline-defined early-stage HTN also has a higher impact in these middle-aged patients and conferred approximately twofold to threefold risk of future CVE. However, it would be better to confirm these findings in larger scale, prospective cohort studies before people can consider applying a lower threshold, such as the 130–139/80–89 mm, in routine management of our axSpA patients.

A study from the American National Survey reported an age-standardized prevalence of 45.4% and an increase of over 20% in the total number of individuals with HTN with the 2017 definition.³⁴ Similarly, the prevalence of HTN in our study was 12.2% according to the previous definition and increased dramatically to 43% using the new threshold. A latest meta-analysis including prospective cohorts also revealed optimal control of early-stage HTN could reduce the risk of CVE for more than 10%.³⁵ HTN is the most common comorbidity (pooled prevalence 23%), in patients with axSpA using the old definition.¹⁴ This new definition inevitably increases the prevalence of HTN in patients with axSpA. Whether treatment of the state of early-stage HTN could potentially reduce the excessive CV risk in the future would definitely need to be addressed in future studies.

The strength of our study includes the long-term follow-up (a median of 12 years) to investigate the effect of early-stage HTN on CVE in axSpA with time-varying adjustment of high inflammatory burden and traditional risk factors. Moreover, adjusted mean BP over the first 2 years prior to baseline was utilized instead of the traditional approach of averaging BP of ⩾2 readings in two separate occasions,³⁶ as the cumulative HTN exposure over time was more reliable than BP at single time point.²⁸ However, there are several limitations to the study. First, missing data are inevitable due to the retrospective design. Body mass index was not included in the analyses as body height was not routinely recorded in the EHR. Second, the relatively small sample size may limit the power of time-dependent analysis of multiple covariates and the interaction between inflammation and traditional CV risk factors. Third, causality cannot be definitively delineated in a longitudinal study, our results should be interpreted with caution, and further larger prospective study is needed to clarify whether early-stage HTN should be the new target of BP management in patients with axSpA. Also, given the limited number of females 98/437 (22.4%), there may be uncertainty regarding the effect of gender. In addition, patients were diagnosed exclusively as AS prior to 2009, as the 2009 ASAS classification criteria were implemented that year. Subsequently, 84/437 (19.2%) patients met the ASAS classification criteria for non-radiographic axial spondyloarthritis (nr-axSpA). Whether the results from this study are applicable to patients with nr-axSpA will need to be confirmed by future studies. Lastly, whether ASDAS may be a better predictor of CVE in these patients with or without early-stage HTN deserves further study.

Conclusion

In patients with axSpA, the state of early-stage HTN carries excessive risk of future CVE, which may be due to untreated inflammatory burden. Antihypertensive therapy targeting BP < 130/80 should be initialed in these patients to minimize their future risk of CVE.

Footnotes

Acknowledgements

We would like to show our gratitude to all medical staffs and research assistants. Furthermore, we would like to thank all the supports from family members.

Declarations

ORCID iDs

Lin-Hong Shi

Steven Ho Man Lam

Lai-Shan Tam

Supplemental material

Supplemental material for this article is available online.

References

Haroon

Michael Paterson

, et al. Patients with ankylosing spondylitis have increased cardiovascular and cerebrovascular mortality: a population-based study. Ann Intern Med 2015; 163(6): 409–416.

Szabo

Levy

Rao

, et al. Increased risk of cardiovascular and cerebrovascular diseases in individuals with ankylosing spondylitis: a population-based study. Arthritis Rheum 2011; 63(11): 3294–3304.

Arts

EEA

Popa

Den Broeder

, et al. Performance of four current risk algorithms in predicting cardiovascular events in patients with early rheumatoid arthritis. Ann Rheum Dis 2015; 74(4): 668–674.

Arts

EEA

Popa

Den Broeder

, et al. Prediction of cardiovascular risk in rheumatoid arthritis: performance of original and adapted SCORE algorithms. Ann Rheum Dis 2016; 75(4): 674–680.

Ozen

Sunbul

Atagunduz

, et al. The 2013 ACC/AHA 10-year atherosclerotic cardiovascular disease risk index is better than SCORE and QRisk II in rheumatoid arthritis: is it enough? Rheumatology (Oxford) 2016; 55(3): 513–522.

Ernste

Sánchez-Menéndez

Wilton

, et al. Cardiovascular risk profile at the onset of psoriatic arthritis: a population-based cohort study. Arthritis Care Res 2015; 67(7): 1015–1021.

Peters

Symmons

DPM

McCarey

, et al. EULAR evidence-based recommendations for cardiovascular risk management in patients with rheumatoid arthritis and other forms of inflammatory arthritis. Ann Rheum Dis 2010; 69(2): 325–331.

Navarini

Margiotta

DPE

Caso

, et al. Performances of five risk algorithms in predicting cardiovascular events in patients with psoriatic arthritis: an Italian bicentric study. PLoS One 2018; 13(10): e0205506.

Shi

Lam

, et al. High inflammatory burden predicts cardiovascular events in patients with axial spondyloarthritis: a long-term follow-up study. Ther Adv Musculoskelet Dis 2022; 14: 1759720X221122401.

10.

Lam

SHM

Cheng

, et al. DAPSA, carotid plaque and cardiovascular events in psoriatic arthritis: a longitudinal study. Ann Rheum Dis 2020; 79(10): 1320–1326.

11.

Yusuf

Hawken

Ounpuu

, et al. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case–control study. Lancet 2004; 364(9438): 937–952.

12.

Lim

Vos

Flaxman

, et al. A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012; 380(9859): 2224–2260.

13.

Brophy

Cooksey

Atkinson

, et al. No increased rate of acute myocardial infarction or stroke among patients with ankylosing spondylitis-a retrospective cohort study using routine data. Semin Arthritis Rheum 2012; 42(2): 140–145.

14.

Zhao

Robertson

Reich

, et al. Prevalence and impact of comorbidities in axial spondyloarthritis: systematic review and meta-analysis. Rheumatology (Oxford) 2020; 59(Suppl. 4): iv47–iv57.

15.

Rutan

Kuller

Neaton

, et al. Mortality associated with diastolic hypertension and isolated systolic hypertension among men screened for the multiple risk factor intervention trial. Circulation 1988; 77(3): 504–514.

16.

Lewington

Clarke

Qizilbash

, et al.; Prospective Studies Collaboration. Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet 2002; 360(9349): 1903–1913.

17.

Whelton

Carey

Aronow

, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: a report of the American College of Cardiology/American Heart Association Task Force on clinical practice guidelines. Circulation 2018; 138(17): e484–e594.

18.

Millett

ERC

Peters

SAE

Woodward

. Sex differences in risk factors for myocardial infarction: cohort study of UK Biobank participants. BMJ 2018; 363: k4247.

19.

Critselis

Chrysohoou

Kollia

, et al.; ATTICA Study Group. Stage 1 hypertension, but not elevated blood pressure, predicts 10-year fatal and non-fatal CVD events in healthy adults: the ATTICA Study. J Hum Hypertens 2019; 33(4): 308–318.

20.

Wang

Liu

. Global impact of 2017 American College of Cardiology/American Heart Association Hypertension Guidelines: a perspective from China. Circulation 2018; 137(6): 546–548.

21.

Mancia

Corrao

. Global impact of the 2017 American College of Cardiology/American Heart Association Hypertension Guidelines: a perspective from Italy. Circulation 2018; 137(9): 889–890.

22.

Nadar

Stowasser

. New guidelines with few takers: will the new American guidelines ever be accepted? J Hum Hypertens 2018; 32(6): 387–389.

23.

Kario

Wang

. Could 130/80 mm Hg be adopted as the diagnostic threshold and management goal of hypertension in consideration of the characteristics of Asian populations? Hypertension 2018; 71(6): 979–984.

24.

Brunstrom

Carlberg

. Association of blood pressure lowering with mortality and cardiovascular disease across blood pressure levels: a systematic review and meta-analysis. JAMA Intern Med 2018; 178(1): 28–36.

25.

Shi

Lam

, et al. Inflammation is associated with incident hypertension in patients with axial spondyloarthritis: a longitudinal cohort study. Clin Exp Hypertens 2023; 45(1): 2205056.

26.

Rudwaleit

van der Heijde

Landewé

, et al. The development of Assessment of SpondyloArthritis international Society classification criteria for axial spondyloarthritis (part II): validation and final selection. Ann Rheum Dis 2009; 68(6): 777–783.

27.

von Elm

Altman

Egger

, et al.; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol 2008; 61(4): 344–349.

28.

Nikpour

Urowitz

Ibanez

, et al. Importance of cumulative exposure to elevated cholesterol and blood pressure in development of atherosclerotic coronary artery disease in systemic lupus erythematosus: a prospective proof-of-concept cohort study. Arthritis Res Ther 2011; 13(5): R156.

29.

Tselios

Gladman

, et al. Impact of the new American College of Cardiology/American Heart Association definition of hypertension on atherosclerotic vascular events in systemic lupus erythematosus. Ann Rheum Dis 2020; 79(5): 612–617.

30.

Garrett

Jenkinson

Kennedy

, et al. A new approach to defining disease status in ankylosing spondylitis: the Bath Ankylosing Spondylitis Disease Activity Index. J Rheumatol 1994; 21(12): 2286–2291.

31.

Calin

Garrett

Whitelock

, et al. A new approach to defining functional ability in ankylosing spondylitis: the development of the Bath Ankylosing Spondylitis Functional Index.J Rheumatol 1994; 21(12): 2281–2285.

32.

Sommerfleck

Schneeberger

Citera

. Comorbidities in Argentine patients with axial spondyloarthritis: is nephrolithiasis associated with this disease? Eur J Rheumatol 2018; 5(3): 169–172.

33.

Derakhshan

Goodson

Packham

, et al. Increased risk of hypertension associated with spondyloarthritis disease duration: results from the ASAS-COMOSPA study. J Rheumatol 2019; 46(7): 701–709.

34.

Dorans

Mills

Liu

, et al. Trends in prevalence and control of hypertension according to the 2017 American College of Cardiology/American Heart Association (ACC/AHA) guideline. J Am Heart Assoc 2018; 7(11): e008888.

35.

Han

Chen

Liu

, et al. Stage 1 hypertension by the 2017 American College of Cardiology/American Heart Association hypertension guidelines and risk of cardiovascular disease events: systematic review, meta-analysis, and estimation of population etiologic fraction of prospective cohort studies. J Hypertens 2020; 38(4): 573–578.

36.

James

Oparil

Carter

, et al. 2014 evidence-based guideline for the management of high blood pressure in adults: report from the panel members appointed to the Eighth Joint National Committee (JNC 8). JAMA 2014; 311(5): 507–520.

Early-stage hypertension defined by the 2017 ACC/AHA blood pressure guideline carries excessive cardiovascular risk in axial spondyloarthritis patients