Cardiovascular Disease in Patients with Ankylosing Spondylitis from the Rheumatology Outpatient Clinic of the UFMS-affiliated Hospital

Background: Cardiovascular diseases are a major cause of morbidity and mortality today. Despite its wide distribution, it presents particularly prevalent in certain groups of individuals, particularly when exposed to a higher degree of inflammation, giving increased cardiovascular risk. Rheumatic diseases expose their holders to this increased cardiovascular risk condition; however only recently have been associated with spondyloarthritis, particularly ankylosing spondylitis (AS). For being a classically autoimmune disease related to HLA histocompatibility system, AS may present phenotypic variations in different ethnic groups with possible diverse cardiovascular consequences.


Introduction
Cardiovascular diseases (CVD) are currently a major cause of morbidity and mortality.Recent data from the World Health Organization have shown that 30% of the total of deaths worldwide in past decades result from CVD, approximately 17 million individuals. 1,2he traditional risk factors for CVD, such as diabetes, Silva Junior et al.

Cardiovascular disease and ankylosing spondylitis
Int J Cardiovasc Sci.2019;32(1)10-18 Original Article metabolic syndrome, dyslipidemia, hypertension, smoking and family history, have been well established as predictors of cardiovascular events, and measures for their control and monitoring are recommended to reduce and fight CVD. 3 However, in certain clinical conditions, such as rheumatic diseases, the traditional risk factors do not seem to contribute to the total incidence of CVD. 4 Despite the well described higher incidence of CVD in rheumatic diseases, especially rheumatoid arthritis, systemic sclerosis, systemic vasculitis, antiphospholipid syndrome and systemic lupus erythematosus, it was only recently that the association of CVD and ankylosing spondylitis (AS) gained attention. 5,6e greater mortality of individuals with AS compared to that of the general population seems to be related to CVD. 7,8 Early atherosclerosis has been reported in association with chronic inflammation and with AS. 9 Some studies have shown that CVD and their traditional risk factors are more prevalent in patients with AS, 10,11 which is still controversial when interpreting inflammation as the major responsible for the higher cardiovascular impairment.
[14] The characteristic racial miscegenation of the Brazilian population increases the interest for such studies, because it can lead to differences in the prevalence of CVD in that population, considering the association between AS and the HLA system.This study was aimed at estimating the prevalence of CVD in individuals with AS from the Spondyloarthritis Outpatient Clinic of the Rheumatology Service of the Mato Grosso do Sul Federal University hospital (HU-UFMS), in addition to correlating the traditional risk factors and the inflammatory process of CVD.

Patients and methods
This is a quantitative descriptive study carried out from March to October 2015 with individuals with AS selected by convenience and consecutively from the patients regularly cared for at the Spondyloarthritis Outpatient Clinic of the Rheumatology Service of the HU-UFMS, which has records of 170 patients diagnosed with spondyloarthritis (psoriatic arthritis, AS, reactive arthritis, and enteropathic arthritis).Of those 170 patients, 55 were diagnosed with AS.A control group (CG) was formed by convenience sampling with employees of the UFMS without rheumatic disease and was similar to the AS group regarding age and sex.The inclusion and exclusion criteria allowed the selection of 42 individuals for the AS group and 50 for the CG.
Before data collection, all individuals provided written informed consent, duly registered in the Ethics Committee in Research of the UFMS (CAAE: 34043614.8.0000.0021).
Individuals with the following characteristics were excluded: diabetes, indigenous background, hypothyroidism or neoplasia, illiteracy, and pregnant women.All participants were screened for CVD and cardiovascular risk factors.
All individuals underwent cardiology clinical examination, laboratory tests, electrocardiography, echocardiography and carotid doppler examination on the same day, and within 20 days from the initial interview, which collected data from medical history and AS.The laboratory tests, performed after a 10-12-hour fasting, comprised the following measurements: total cholesterol and fractions, glycemia, glycated hemoglobin, uric acid, microalbuminuria, erythrocyte sedimentation rate (ESR), high-sensitivity C-reactive protein (hs-CRP) and interleukin 6 (IL-6).
The criteria to assess cardiovascular risk according to the V Brazilian Guideline on Dyslipidemia and Atherosclerosis Prevention comprised the presence of dyslipidemia, metabolic syndrome, arterial hypertension, current smoking, family history of early CVD, cardiac hypertrophy, and elevated levels of glycated hemoglobin, uric acid and hs-CRP. 15rdiovascular disease was classified as clinically manifest or subclinical.Individuals with manifest CVD were those with history of myocardial infarction, stroke, coronary artery bypass graft surgery (CABG), peripheral revascularization surgery or procedure (PRV), or peripheral obstructive arterial disease (POAD), echocardiographic evidence of segmental, systolic, diastolic (over grade I) or hypertrophic cardiomyopathy, arrhythmias or bundle-branch and atrioventricular blocks on the electrocardiogram, and presence of carotid plaque obstruction > 50% of the lumen on Doppler.Subclinical CVD was identified by an ankle-brachial index (ABI) < 0.9 or > 1.3, a carotid intima-media thickness (CIMT) >1 mm, but without significant plaque obstruction (> 50%) or microalbuminuria > 30 mg/g.Disease activity was assessed by using the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) and Ankylosing Spondylitis Disease Activity Score (ASDAS).The ASDAS-CRP and ASDAS-ESR indices16 were calculated after blood collection, using the information obtained when assessing the Bath Ankylosing Spondylitis Metrology Index (BASMI) and the Bath Ankylosing Spondylitis Functional Index (BASFI).In addition, the disease activity indices were correlated with the serum levels of IL-6, hs-CRP and uric acid, and compared to the occurrence of clinically manifest or subclinical CVD.
The AS group individuals were also assessed regarding the occurrence of CVD up to the age of 40 years and after 40 years.

Statistical analysis 17
The results referring to the quantitative variables are presented as mean ± standard deviation, while the results of the categorical variables are presented as relative frequency followed by absolute frequency.The quantitative variables were compared between individuals with and without spondyloarthritis by use of parametric Student t test for independent samples (nonpaired) because most of the samples were normally distributed (Shapiro-Wilk test; p > 0.05).Student t test was used to compare quantitative variables between age groups.The association between the qualitative variables and the presence or absence of spondyloarthritis was assessed by use of chi-square test, which was also used to assess the association between the qualitative variables and the patients' age groups.The linear correlation between some quantitative variables was assessed by use of Pearson's linear correlation test.The results of the other variables were presented as descriptive statistics or tables and graphs.The SPSS statistical program, version 22.0, was used for statistical analysis, considering a 5% significance level.

General characteristics
There was no significant difference between individuals with or without AS regarding age, sex, skin color, educational level, weight, height, blood pressure and body mass index (Table 1).

Risk factors for cardiovascular disease
Table 2 shows the results regarding the cardiovascular risk factors in the two groups.There was no significant difference between the groups regarding the variables family history, smoking habit, alcoholism, abdominal circumference, systemic arterial hypertension, total cholesterol, HDL-cholesterol, triglycerides, glycated hemoglobin, hs-CRP, microalbuminuria, metabolic syndrome, and IL-6.The LDL-cholesterol levels in the AS group, however, were higher than those in the CG (p = 0.012).The uric acid levels in the AS group were lower than those in the CG (p = 0.019).
All individuals were stratified into low, intermediate and high cardiovascular risk.Despite the higher frequency of high cardiovascular risk in the AS group, the difference was not significant; thus, there was no To assess the cardiovascular risk of well-controlled patients, and thus with lower inflammation level, the global cardiovascular risk was calculated separately in patients with ASDAS-CRP < 2 and low disease activity level, but no significant statistical difference was found between the CG and the AS group patients with that characteristic.

Manifest cardiovascular disease
Most individuals assessed in this study had no clinically manifest CVD, with prevalence between 2% and 26%, and valvular dysfunctions were the most frequent change, although all of them were mild.The percentage of individuals with right bundle-branch block was higher in the AS group than in the CG.The following variables showed no association with the presence or absence of AS: stroke, acute myocardial infarction, CABG, angioplasty, PRV/POAD, presence of carotid plaques and other findings on echocardiogram and electrocardiogram.None of the individuals in the CG and AS group showed significant carotid obstruction (plaque obstruction ≥ 50% of the lumen) (Table 3).

Subclinical cardiovascular disease
None of the individuals achieved the cutoff values for the occurrence of microalbuminuria (> 30 mg/g) or altered ABI (< 0.9 and > 1.3); in addition, the mean values of ABI and microalbuminuria were similar in both groups.Carotid plaques without significant obstruction (< 50% of the lumen) were slightly more frequent in the AS group [CG: 12.2% (n = 6); AS: 21.4% (n = 9)], but with no significant statistical difference (p = 0.239).However, as shown in Figure 1, CIMT was higher in the AS group (p = 0.018), and the cutoff value to determine altered CIMT (> 1 mm) was more often observed in the AS group [CG: 2.3 (n = 1); AS: 24.2 (n = 8)] (p = 0.003).This difference was observed when individuals with carotid plaques were excluded.Considering only the individuals with carotid plaques, no significant difference regarding CIMT was observed between the groups (CG: 0.78 ± 0.13; AS: 1.45 ± 2.12) (p = 0.464).
The hs-CRP and IL-6 levels were analyzed in both groups as one of the inflammation markers.The IL-6 showed no statistically significant difference (CG: 4.62 ± 3.41, AS: 4.84 ± 5.20; p = 0.806).Although more elevated in the AS group, the hs-CRP levels showed no significant difference as compared to those in the CG (CG: 2.35 ± 2.21, AS: 16.10 ± 56.85; p = 0.130).
There was moderate and positive significant linear correlation between the hs-CRP values obtained and those of ASDAS-CRP (Pearson, p < 0.001; r = 0.657), in addition to a positive significant, although weak, linear correlation between the hs-CRP values obtained and those of ASDAS-ESR (p = 0.007; r = 0.418).On the other hand, there was no significant linear correlation between the other inflammation markers and the disease activity indices assessed in this study (p-value between 0.177 and 0.875).
There was no linear correlation between the values obtained in ASDAS-CRP and the CIMT measures (p = 0.932, r = -0.014).In addition, there was no linear correlation between the values of hs-CRP and CIMT (p = 0.625, r = -0.079)or between time since diagnosis and CIMT (p = 0.152, r = -0.225).The lack of statistical significance in the linear correlation between CIMT and ASDAS-CRP, hs-CRP and time since diagnosis persisted even when individuals with and without plaque were separated.
In addition, the AS group patients were assessed according to age group (up to 40 years and over 40 years).There was no significant association between age group and manifest CVD.Regarding subclinical CVD, however, the percentage of patients older than 40 years with carotid plaques was significantly higher than that of those under 40 years.Microalbuminuria, ABI and CIMT showed no statistically significant difference according to age group (Table 4).
There was no linear correlation between time since diagnosis and CIMT in the age group under 40 years (p = 0.688, r = -0.130).However, there was a significant positive correlation between time since diagnosis and CIMT in individuals older than 40 years (p = 0.049, Original Article r = -0.362).Among individuals with AS older than 40 years and no carotid plaque, the correlation between time since diagnosis and CIMT was also positive (Figure 2).

Discussion
The higher frequency of cardiovascular impairment in rheumatic diseases, whose major background is chronic inflammation, explains the higher cardiovascular morbidity and mortality of those patients. 18However, the role of inflammation in the occurrence of CVD remains controversial in AS.
Although the prevalence of cardiovascular impairment in AS is not high, 12 there is consensus that the mortality rate of those patients is higher than that of the general population, mainly due to higher cardiovascular mortality. 7,8In addition, the more accurate definition of the weight of each factor involved, such as chronic inflammation, traditional cardiovascular risk factors and genetics, still lacks.The cyclic characteristic of AS throughout life, the time of exposure to disease, the effect of treatment, and the variety of exposure to cardiovascular risk factors can interfere with the CVD occurrence in that population, hindering the better understanding of the correlation between AS and heart diseases.
Unlike rheumatoid arthritis, whose cardiovascular risk score is higher, 9 publications on AS are controversial.Some authors have explained the higher prevalence of subclinical atherosclerotic disease, such as a higher CIMT, in patients with AS because of the higher exposure to cardiovascular risk factors, 7 while others, have reported that the cardiovascular impairment was independent of the presence of the traditional risk factors for CVD. 19 our study, both groups were similarly exposed to the traditional risk factors for CVD, except for mean In addition, it is worth noting the considerably higher mean value of hs-CRP of the AS group individuals, indicating their higher cardiovascular risk.Furthermore, a good part of those with well-controlled rheumatic disease and low inflammation (ASDAS < 2) were at high cardiovascular risk and out of the recommended LDLcholesterol target (< 70 mg/dL).
We do not know any Brazilian study using this cardiovascular risk stratification approach for those patients, but it is worth noting that some researchers have sought cardiovascular risk markers specific for AS, 23 according to the recommendations established for rheumatoid arthritis and systemic lupus erythematosus. 9e lack of correlation between markers of inflammation, activity and disease impairment in the AS group can be partially explained by the low mean age of the patients or the use of immune biologics in 80% of them.By being a cross-sectional study might have contributed to that, although other authors have found no correlation between inflammation and early atherosclerosis when cardiovascular risk factors were excluded. 13,21

Conclusions
There is no difference in the prevalence of manifest CVD in patients with AS as compared to that of the CG.However, subclinical CVD is more prevalent in AS patients when assessed by use of CIMT, has no relation to the AS activity, but to the time since AS diagnosis and to more advanced ages.Most patients with AS are not only at higher cardiovascular risk, even when clinically controlled, but their CVD prevention is inadequate as well.

Time since diagnosis (years)
CIMT

Figure 1 -
Figure 1 -Graph showing the carotid intima-media thickness (CIMT) of individuals with and without ankylosing spondylitis.Each column represents the mean, and the bar represents the standard error of the mean.* Significant difference as compared to individuals without spondyloarthritis (Student t test, p = 0.018).

Figure 2 -
Figure 2 -Dispersion graph showing the moderate positive linear correlation between time since diagnosis and carotid intima-media thickness (CIMT), considering only patients with ankylosing spondylitis, without plaque, aged ≥ 40 years.Each symbol represents the value for both variables for a single individual.The dashed line represents the line of linear regression between those variables.The p value and r were obtained with Pearson's linear correlation test.

Table 1 -Sociodemographic and physical data of the individuals with and without spondyloarthritis assessed in this study Variables Spondyloarthritis p value No Yes
SBP: systolic blood pressure; DBP: diastolic blood pressure.The results are expressed as mean ± standard error of the mean or relative frequency (absolute frequency).P value in the Student t test (quantitative variables) or the chi-square test (qualitative variables).Silva Junior et al.Cardiovascular disease and ankylosing spondylitisInt J Cardiovasc Sci.2019;32(1)10-18Original Article

Table 2 -Cardiovascular risk factors of the individuals with and without spondyloarthritis assessed in this study
± standard error of the mean or relative frequency (absolute frequency).P value in the Student t test (quantitative variables) or the chi-square test (qualitative variables).

Table 3 -Results regarding manifest cardiovascular disease of the individuals with and without spondyloarthritis assessed in this study
Time since diagnosis ranged from 2 to 34 years (mean of 10.76 ± 8.74 years).The time since symptom onset ranged from 3 to 47 years (mean of 19.36 ± 11.02 years).
AMI: acute myocardial infarction; CABG: coronary artery bypass graft surgery; PRV: peripheral revascularization surgery or procedure; POAD: peripheral obstructive arterial disease; ECHO: echocardiography; ECG: electrocardiography.The results are expressed as relative frequency (absolute frequency).P value in the chi-square test.