Target Organ Damage and Cardiovascular Risk in a Hypertensive Roma Sample Population in Romania

Abstract

Background:

The largest European Roma community resides in Romania, but there is still little published data on cardiovascular (CV) risk factors and disease in this group. This study addresses the prevalence of arterial hypertension, associated CV disease risk, and target organ damage (TOD) in a Roma community from Bucharest, Romania.

Methods:

This is a cross-sectional community-based participatory research to assess for CV risk factors, TOD and CV disease, including 806 Roma ethnics (18-83 years) integrated in the local community, 36.16% males. Evaluation included physical examination with blood pressure, pulse wave velocity and anklebrachial measurements, laboratory tests, ECG, echocardiography and fundoscopy.

Results:

Prevalence of hypertension was 33.62%, awareness 76.38%, higher in females (p>0.01), and control rate 44.39%. Compared to age-matched normotensives, hypertensives had more left ventricle hypertrophy and more frequently increased pulse pressure. Differences in TOD were attenuated between newly and previously diagnosed, controlled and uncontrolled, hypertensives. Cardiovascular disease was almost absent in normotensives. Ten-year risk for fatal CV disease followed an increasing trend from normotension to long standing hypertension.

Conclusion:

This is the first dedicated study to thoroughly assess TOD and risk for fatal CV disease in a Romanian Roma population. Hypertension was less prevalent than in the general population, with similar awareness, possibly as a consequence of integration in the surrounding community. Fatal CV disease risk followed the trend of increasing prevalence of risk factors, and hypertension played an important role in its modulation.

Keywords:hypertension, Roma population, Romani, gypsy, Romania, target organ damage, cardiovascular risk.

INTRODUCTION

In all its inhomogeneity, Roma ethnicity associates a high risk cardiovascular (CV) profile as a consequence of clustering of modifiable risk factors in all countries that have researched this topic (1). Data on mortality is still scarce, but so far, we know their life-expectancy is at least ten years shorter than that of their surrounding communities (2), and it appears that they are experiencing modernization and epidemiological transition, with CV diseases (CVD) emerging as top causes of morbidity and mortality (3). With more efforts being made to integrate the Roma in local communities around the world (4), they transition to a westernized lifestyle. These characteristics take on importance on a public health level, as this embrace will lead to a significant rise in CVD prevalence. The largest community of Roma inhabitants in the European Union is found in Romania (5), but here epidemiologic data on CVD in this group is still limited. In our recent study, we have shown a similar high prevalence of CV risk factors in a Roma population which was highly integrated in the surrounding community, but nonetheless retained a high CV risk profile independent of socio-economic status (6).

In the current paper, we focus on prevalence of arterial hypertension, target organ damage (TOD), and further evaluate associated risks in Roma ethnics from this large sample.

METHODS

We carried out a cross-sectional populationbased study on Roma subjects from the capital city, in two phases – one in the autumn of 2012 and the second in the autumn of 2013 – as part of a larger socially-oriented project, aiming to facilitate medical services to Roma ethnics. The community-based participatory research included 806 participants, and the procedure, diagnostic criteria and population characteristics are described in detail in a previously published paper (6).

The present study provides a comprehensive snap-shot of the CV profile of its sample population, which included 2.57%, and, respectively, 1.63% of Roma residents in the surveyed area. In each subject, the evaluation algorithm included anthropometric measures and surveyor-assisted completion of a questionnaire, on age, level of education, marital status, and behaviors regarding smoking and physical exercise. Further, each underwent brief medical history, clinical evaluation, blood pressure (BP) and ankle-brachial index (ABI) measurements, laboratory work-up focused on lipid profile, ECG, and echocardiography. In the second phase of the project, additional dip-stick microalbuminuria, fundoscopy, and pulse wave velocity (PWV) measurements were also performed.

Ten-year risk of fatal CVD was calculated in those >40 years according to the 2016 European Guidelines on CVD prevention (7), integrating SCORE risk (chart for high risk regions of Europe) and comorbidities.

Diagnostic criteria/cut-off values were based on the 2013 ESH/ESC Guidelines for the management of arterial hypertension and ADA standards of medical care in diabetes 2016. Hypertension control was defined as blood pressure <140/90 mmHg in previously diagnosed hypertensives having taken antihypertensive medication during the previous two weeks.

Trained medical personnel informed the subjects regarding the medical evaluation, procedures and testing. Written fully informed consent was obtained from all participants. The study was approved by the local Ethics Committee.

Statistical analysis

Statistical analysis was performed using IBM SPSS Statistics v23 on a significance level .0.05. Categorical variables are presented as counts and percentages, and continuous variables are presented as mean values ( standard deviation), or median (interquartile range). Between-group comparisons were made using a non-parametric test (Mann-Whitney U, Kruskal-Wallis), when comparing groups of unequal variances or with non-normal distribution, or chi-square X2 for categorical variables. Continuous variables were compared using independent samples t test. Propensity score matching was performed to control for covariates.Comparison of population proportions was done using a Z score.

RESULTS

The entire population included 806 subjects aged 18-83 years (average age 44.54±14.57, males 36.8%), from an urban community, integrated in the local non-Roma community (6). Here, prevalence of hypertension was 33.62%, with no significant gender differences (for age adjusted prevalence see Table S1). Hypertension in this Roma population was less prevalent when compared with reports of SEPHAR III (33.62% versus 45.1%, p<0.01) (8). However, the earlier SEPHAR II study was conducted roughly during the same period which makes it a better comparator. In this case, prevalence of hypertension was similar in our Roma population with the general Romanian population, but when compared with the non-Roma community in Bucharest, it was less than that reported in SEPHAR II (33.62% versus 41%, p 0.044), and similar in SEPHAR I, having no data for Bucharest yet reported from SEPHAR III (9).

Awareness recorded in those previously diagnosed with hypertension was 76.38%, higher among females than males (83.2% versus 64.3%, p <0.01), and similar to that reported for the general population (8, 9).

The subgroup of hypertensives comprised 271 adults aged 55.88±10.34 years, with 36.16% males.

The prevalence of CV risk factors among them is shown in Table 2, with dyslipidemia and an abnormal lipid profile (93.35%), abdominal obesity (72.69%), and smoking (58.67%, including ex-smokers) as top three major determinants of risk.

In terms of TOD, a reduced ABI <0.9 was the most prevalent of markers, present in over half of all hypertensives (55.35%), here with no gender differences. It was followed by left ventricle hypertrophy (LVH), found in almost one of two hypertensives (46.49%) from both projects. Pulse wave velocity measurements showed similarly high prevalence of increased arterial stiffness (56.34%), followed by hypertensive retinopathy at fundoscopy (42.96%).

Personal history of CVD revealed that in 18.31% of cases, coronary heart disease had been previously documented, and another 13.38% had already suffered myocardial infarction, more frequent in males (p <0.001). Previously diagnosed heart failure was reported in 16.2% of cases. All those with documented atrial fibrillation (7.75%) were known hypertensives.

When assessing ten-year risk of fatal CVD in those >40 years it appears that 61.96% of hypertensives are at either high or very high risk, with a considerable difference between genders, favoring women in this scenario (p <0.001).

The hypertensive versus normotensive Roma

When comparing hypertensive versus normotensive Roma from the entire population, hypertensive status is associated with a higher CV burden in terms of risk factors (dyslipidemia, obesity, abdominal obesity, diabetes mellitus, smoking), TOD (LVH, moderate to severe renal impairment, hypertensive retinopathy, increased arterial stiffness), established CVD, and high total CV risk.

Factors associated with HTN on logistic regression were only age, physical activity, and obesity. The propensity score matching to control for these led to the selection of 342 subjects from the original population (172 hypertensives versus 170 normotensives). For these subgroups, analysis showed significant differences in a composite variable including all TOD markers tested (76.7% versus 61.8% in hypertensives versus normotensives, p 0.003). In matched subgroups, LVH and increased pulse pressure were significantly associated with hypertensive status (40.1% versus 25.9%, p 0.005, respectively, 9.9% versus 3.5%, p >0.001), and there was no difference in CV risk factors. Hypertension was nevertheless associated with documented CVD and a higher total CVD risk (Table S2).

Hypertension diagnosis and control

In the case of newly versus previously diagnosed hypertensives, the first had lower CV burden, with less obesity (39.1% vs 59.4%, p 0.004), less abdominal obesity (59.4% vs 76.8%, p 0.006), but more hyperuricemia (21.9% vs 10.6%, p 0.021), and no history of previous stroke (0% vs 12.4%, p 0.047). Eventually, there was no difference in fatal CVD risk between these subgroups.

Hypertension control rate was 44.39% in previously diagnosed subjects. Controlled and uncontrolled HTN shared similar CV risk factors, disease and total CVD risk. In those with controlled hypertension however, among TOD parameters, LVH was significantly less prevalent (36.3% vs 51.7%, p 0.018).

Going from normotension to newly diagnosed hypertension to known hypertension, there was an increasing trend in CV risk in the entire population (p>0.001) (Figure 1).

DISCUSSION

The prevalence of hypertension in Romania was evaluated in the SEPHAR study series placing it at 45.1% in the general population (8). In our population, it was less prevalent than in the general Romanian population, a difference notable through all age groups, including 65 years (8, 9).

A lower prevalence of hypertension has been reported in other traveler populations as well. Iraqi migrants had lower office BP levels compared to the natives in Sweden (10), and in Pima Indians hypertension was less frequent than in native Caucasians (11). These features are seen after migration to the receiving country, usually accompanied by major changes in lifestyle and diet. Research in Pima Indians has shown they have baseline lower sympathetic nervous system activity (SNSA) (12), a dissociation between hyperinsulinemia and SNSA leading to a blunted SNS response to weight gain (13), and a more rapid cortisol mediated inhibition of SNS as a possible better CV adaptation to stress by limiting sympathoexcitation (14). All these act as natural breaks in the development of hypertension, thus providing Pima Indians with a genetic protective trait against it (11). The high prevalence of obesity with lower prevalence of hypertension in Roma, an ethnic population descended from a warm climate (15), may suggest similar physiological pathways with such populations, but have not been explored so far. In addition, genetic risk scores for hypertension proved to be lower in Roma versus non-Roma subjects in a large representative random sample in Hungary (16). Consequently, literature is still lacking strong evidence to fully explain this paradox of high CV risk profile with low hypertension in the Roma.

Results point towards a “Roma paradox” of similarly high CV profile combined with lower prevalence of hypertension. This may be a consequence of: 1) lower density of hypertensives in older age groups as a result of death of the more diseased at younger ages in a population having ten years shorter life expectancy; 2) bias in our sample selection of healthier, more aware subjects involved in maintaining personal health, leaving out the more unaware diseased; 3) a possible particular genetic profile limiting the development of hypertension. Sample bias can be objected by the comparably high CV risk profile seen in the general Romanian population, our Roma population having similar prevalence of abnormal lipid profiles, obesity, and diabetes.

The similarities in TOD between matched normotensive versus hypertensive subgroups is noteworthy, as TOD in normotensives is not risk-free (17). These facts emphasize that the high risk CV profile takes its toll even before sustained hypertension ensues and accelerates vascular ageing and microvascular damage. The differences in TOD came from LVH and increased pulse pressure, the first a predictor of development of hypertension in normotensives (18, 19), the latter a measure of vascular ageing and marker of higher CV risk (20), proving a significant difference in vascular age with the occurrence of hypertension.

There was close to no CVD among normotensives, underlining the role of hypertension as a promoter of CVD in the Roma population as well. Finding atrial fibrillation strictly in previously diagnosed hypertensives, a group probably exposed to high BP for a longer duration, comes in line with the view that it is a manifestation of TOD, considering LVH and left atrium dilation as pathological pathways to arrhythmia (21, 22).

When comparing newly diagnosed with known hypertension, the difference in CV risk was considerably attenuated, the first having less stroke and obesity, but similar TOD, CVD and total CVD risk. The presence of similar TOD in newly diagnosed versus known hypertensives may be hypothesized as a delayed diagnosis after a previous period of uncontrolled BP values.

With the exception of obesity, females had lower CV burden – less hyperglycemia, hyperuricemia, and a lower nicotinic burden, better hypertension awareness and control, less history of myocardial infarction. Eventually, this was also evident in the significantly lower risk of fatal CVD. In a population with few elders, age pyramids of Roma show lower density of males with increasing age (23). Gender differences in total CV risk in Roma have not been evaluated in the literature, however these results generate a hypothesis of a lower male survival due to the significant contrast in risk of fatal CV events.

LIMITATIONS

The cross-sectional design of the study leads to only a snap-shot of the health status of this Roma population. Unfortunately, all health studies on this ethnicity are similar, and a prospective design is difficult, due to their inconsistence in accessing medical care and their migratory behavior. This leads to an important bias in the analysis of CV risk across different moments of hypertensive status, which requires a prospective design to accurately assess outcomes over time.

Another limitation was the possibility of sample bias involving healthier and more aware individuals, leaving out those with worse health status. Should this be the case, considering our results portray a population with less HTN but a majority of high/very high risk individuals, the remaining population would be suffering from even more severe CV disease.

CONCLUSION

Roma ethnicity has been proven to associate a high CV burden, and our study confirms this same profile in a hypertensive integrated community from Romania. Similar to literature from neighboring countries, our study found less HTN than in the surrounding community. Dyslipidemia, obesity, and smoking were the top three determinants of CV risk in hypertensives. Gender differences in CV risk factors (hyperglycemia, hyperuricemia, nicotinic burden) favoring females were also reflected in lower risk of fatal CVD. Hypertensives had a high burden of TOD, while normotensives had close to no CV disease. There was a progressive trend towards higher CV risk from normotension to sustained HTN and a lack of risk factors control despite awareness. With longer exposure to hypertension the differences in organ damage and CV disease became increasingly evident.

Authors’ contributions: Emma Weiss and Elisabeta Badila contributed equally to the current study.

Conflict of interests: none declared.

Financial statement: Funding for the SUPORT projects was insured by grants from The Romanian Government National Agency for Roma in 2012 and 2013.

TABLE 1.

Hypertensive population characteristics and cardiovascular risk factors

TABLE 2.

Target organ damage, cardiovascular disease and risk among hypertensives

FIGURE 1.

Cardiovascular risk, organ damage, and CVD for different hypertensive status

TABLE S1.

Age adjusted prevalence of hypertension

TABLE S2.

Cardiovascular risk factors, cardiovascular disease and total cardiovascular risk in matched hypertensive versus non-hypertensive subjects