Trajectories of Blood Pressure in Patients with Established Coronary Artery Disease over 20 years

Objective To evaluate changes in blood pressure (BP) values in patients with established coronary artery disease (CAD) over 20 years (1997–2017). Materials and Methods Consecutive patients aged <71 years and hospitalized for acute coronary syndrome or myocardial revascularization procedures were recruited and interviewed 6–18 months after their discharge from the hospital. BP was measured in 1997–1998, 1999–2000, 2006–2007, 2011–2013, and 2016–2017. The same five hospitals took part in the surveys at each time point. Results We examined 412 patients in 1997–1998, 427 in 1999–2000, 422 in 2006–2007, 462 in 2011–2013, and 272 in 2016–2017. The proportion of patients with BP at the recommended goal was 49.2% in 1997–98, 44.5% in 1999–2000, 44.7% in 2006–07, 51.1% in 2011–13, and 58.8% in 2016–17 (p < 0.001). Mean systolic and diastolic BP decreased significantly independent of age, sex, and education (systolic BP: 137.9 ± 21.4 mmHg in 1997–98, 139.5 ± 21.6 mmHg in 1999–2000, 136.1 ± 20.3 mmHg in 2006–07, 134.8 ± 22.0 mmHg in 2011–13, and 134.2 ± 18.6 mmHg in 2016–17, p < 0.001; diastolic BP: 83.4 ± 11.0 mmHg in 1997–98, 84.8 ± 12.0 mmHg in 1999–2000, 85.2 ± 11.0 mmHg in 2006–07, 80.9 ± 12.5 mmHg in 2011–13, and 81.1 ± 10.4 mmHg in 2016–17; p < 0.001). Conclusion The analysis of five multicenter surveys provides evidence of a decrease in BP in patients with established CAD over two decades. This trend is independent of age, sex, and the education level of the patients.


Introduction
Cardiovascular disease is the leading cause of death in developed countries [1]. Despite advances in the pharmacological and invasive treatment of coronary artery disease (CAD) in recent years, mortality following myocardial infarction remains high [2]. e main causes include inadequate control of risk factors, insufficient lifestyle changes, and suboptimal pharmacotherapy [3,4]. Recently published results from Europe and North America showed that about half of all CAD patients have their blood pressure (BP) over 140/90 mmHg [3,[5][6][7]. e current guidelines strongly recommend a BP reduction in patients with CAD and hypertension [8,9]. e current optimal BP values for patients with established cardiovascular disease have been a subject of debate for decades [10][11][12][13]. e present analysis aimed to assess trends in BP values in patients with established CAD over 20 years (1997-2017).

Methods
We analyzed the data of the participants of the five surveys assessing cardiovascular prevention following hospitalization due to CAD that were conducted in [1997][1998]  e same five hospitals serving the city and surrounding districts (1.2 m inhabitants) participated in each survey. e methods used in these surveys had been published previously and were similar on each occasion [5,[14][15][16][17]. In brief, the study sample in each survey consisted of consecutive patients hospitalized for myocardial infarction, unstable angina, percutaneous coronary intervention, or coronary artery bypass grafting. Since the age limit in the first (1997)(1998) and second (1999)(2000) surveys were <71 years we excluded from the present analysis all older participants participating in the other three surveys. e patients were examined at 6-18 months after the index hospitalization. Data on patients' medical history and the medications used were obtained using a standard questionnaire. Education of the participants was assessed based on the number of years of schooling as well as a categorical variable (at least secondary education vs. lower than secondary education). Height and weight were measured in a standing position without shoes and heavy outerwear using standard scales with a vertical ruler. e scales were calibrated at the start of each survey. Body mass index was calculated according to the following formula: body mass index � weight (kg)/(height (m)) 2 .
Blood pressure was measured twice, on the right arm in a sitting position after at least five minutes of rest. e average of the two readings was used in the analysis. High blood pressure was defined using two approaches. First, we analyzed the proportions of patients achieving recommended BP (according to the European Society of Hypertension/European Society of Cardiology guidelines) at the time of each survey (BP goal of <140/90 mmHg (<130/80 mmHg in diabetics) in 1997-1998, 1999-2000, 2006-2007 and 2011-2013; <140/90 mmHg (<140/85 mmHg in diabetics) in 2016-2017) [8,[18][19][20][21]. e second definition of high pressure included all cases of systolic BP at least 140 mm of Hg or diastolic BP at least 90 mm of Hg. Participants were considered hypertensive if they had high blood pressure (≥140/90 mm of Hg) during the follow-up interview or if they were treated with antihypertensive drugs at that time and had been diagnosed with hypertension during index hospitalization [22]. e patients were considered to have undiagnosed hypertension if they had hypertension (as defined above) that has not been diagnosed during the index hospitalization. In addition, selfreported use of BP-lowering drugs was analyzed. e survey protocols were approved by the institutional Bioethics Committee. Every patient signed an informed consent form.

Statistical Analysis.
Categorical variables were reported as percentages and continuous variables as means with standard deviations (SD). e Pearson χ 2 test was applied to all categorical variables. Normally distributed continuous variables were compared by the Student's t-test or analysis of variance. Variables without normal distributions were evaluated by the Mann-Whitney U test or the Kruskal-Wallis analysis of variance. Temporal trends were evaluated with linear regression for continuous variables and logistic regression for categorical variables with subsequent surveys coded as an independent variable. Multivariable logistic regression analysis was used to calculate the odds ratios of having BP at goal in surveys II, III, IV, and V compared to the first survey. As a sensitivity analysis, we used two definitions of high blood pressure (see above). In addition, we performed the analysis limited to patients with hypertension only. Generalized linear models as implemented in the Statistica 13 software (TIBCO Software, Palo Alto, USA) were used to provide the least square means after adjusting for the covariates including sex, age, education, and employment, index event, and body mass index. A twotailed P value of less than 0.05 was regarded as indicating statistical significance.  Table 1.

Results
e participants of the fifth survey were older, better educated, and more often in employment compared to the participants of the first survey. No significant differences were observed in sex distribution between the surveys. On the other hand, the mean number of BP-lowering drugs increased significantly (  Figure 1 presents the systolic and diastolic BP categories by the survey. During the observation period the mean systolic BP, diastolic BP, and mean arterial pressure values decreased significantly (Table 2). However, this trend was presently starting from the third survey onwards when systolic BP was taken into the account and from fourth survey onwards when diastolic BP was considered.    p < 0.001). ese trends were independent of age, sex, and education of the participants (Table 3). e trends were significant in patients without diabetes (systolic BP p < 0.001, diastolic BP p < 0.001) and in patients with diabetes when systolic BP (p < 0.001), but not diastolic BP  Table 2). e increase was independent of co-factors and was visible starting from the fourth survey (Table 4). An even greater increase was shown when we limited the analysis to patients with hypertension only (from 27.7% in 1997-1998 to 53.4% in 2016-2017; Figure 2).

Discussion
ere is strong scientific evidence that the long-term survival of coronary patients may be improved by providing optimal cardiovascular prevention [8,9]. Indeed, according to the recently published results of the EUROASPIRE V survey, there is a considerable potential for further improvement in cardiovascular risk in patients with CAD [3]. Our results showed a decrease in BP in patients with established coronary artery disease in the 21st century. Such trend paralleled the mean number of BP-lowering agents prescribed and was evident beginning from 2006 when regarding systolic BP and from 2011 when diastolic BP was considered. ese findings were confirmed by multivariate analyses. Despite a favorable trend in the proportion of patients with high BP, BP values remained high in 41% of the most recent survey participants. is suggests a great potential for further reduction in cardiovascular risk in patients with hypertension through better management of the condition. Although we could not compare data on the lifestyle of all survey participants, the increase in the mean body mass index may suggest that patients with established CAD have inappropriate lifestyle habits, including diet and physical activity [23]. e present results point to improvement in cardiovascular prevention in patients with coronary artery disease.
We found an increasing prevalence of hypertension in patients with CAD. is may reflect the trends present in the general population of Poland [24]. e proportion is higher than in the general population (in the case of the Polish general population 33-43% of subjects are hypertensive [24,25]) since hypertension is a risk factor for CAD and since the average age of our population was relatively high. e estimated prevalence of hypertension may be associated with an error related to different criteria of hypertension diagnosis that were applied by different hospitals participating in the study. e present results may also have been influenced by variation in BP levels, which might have led to patients being misclassified as hypertensive/normotensive. Elevated BP and/or the intake of BP-lowering drugs are used for hypertension diagnosis in many studies assessing the prevalence of hypertension in the general population [24,26]. ese studies did not consider the errors associated with the intake of BP-lowering drugs for reasons other than hypertension (angina, heart failure, history of myocardial infarction, peripheral artery disease) because the prevalence of these conditions is much lower than the prevalence of hypertension. In the case of the present study population, this definition could not be applied since 92.3% of the subjects were taking at least one drug with BP-lowering potential. On the other hand, a diagnosis of hypertension on discharge from the hospital did not qualify a patient as being hypertensive or normotensive during the follow-up examination since 30.3% of patients without a diagnosis of hypertension during hospitalization had a BP level of at least 140/90 mmHg 6-18 months after discharge, thereby classifying them as hypertensive. Indeed, a physician's diagnosis of hypertension should be treated with caution when used to identify CAD patients with hypertension [22]. e definition of hypertension used in this analysis does not include patients with a history of elevated BP (or even with a hypertension diagnosis) and who had normal (<140/ 90 mm of Hg) BP during the control visit and who were not taking antihypertensive medication. Similarly, such patients were not considered hypertensive in other large epidemiological studies [24,26,27]. Subjects with elevated BP during hospitalization but without a diagnosis of hypertension at discharge were considered to be hypertensive if their BP was ≥140/90 mmHg during the follow-up visit. If, however, the BP of these subjects was below 140/90 mm of Hg it can be reasonably assumed that no diagnosis of hypertension at discharge was correct. is is because elevated BP values in patients with acute coronary episodes or during the period close to percutaneous or surgical coronary intervention do not always reflect actual hypertension in everyday conditions and should not be used as a basis for diagnosing hypertension. e definition of hypertension used in the present study does not apply to the subjects whose discharge letters did not include a hypertension diagnosis and who during the control visit were on antihypertensive medication and had normal BP levels. It can thus be assumed that the percentage of patients with CAD and undiagnosed hypertension may be higher than the present findings suggest.
In general, there are three reasons for poor BP control: the patient, the physician, and the healthcare system. e patient may not adhere to the medication regiment or may be resistant to it. As the mean number of antihypertensive drugs was lower in those achieving the BP goal (1.9 ± 1.0 vs. 2.1 ± 1.0), this may indicate lower compliance with medication in those not achieving the BP goal or that their hypertension was somewhat resistant to treatment. Low awareness of high BP as an important cardiovascular risk factor may also be one of the reasons. Low medication persistence is considered a major factor limiting the long-term effectiveness of antihypertensive treatment [28]. A quite frequent cause of poor BP control is clinical inertia [29]. When physicians are asked what they should do to combat inadequate BP control, they usually respond with the correct answer (increase the dosage, add an antihypertensive drug, or switch to another antihypertensive drug). However, when this issue is observed in a real-world setting, no action is taken in many cases [30]. e healthcare system may also be partially responsible. is may be due to a low drug reimbursement rate, different rules for antihypertensive drug reimbursement, time constraints, the poor working conditions of physicians and nurses, etc.
Numerous studies have assessed the quality of medical care in connection with prevention and treatment of CAD, including the control of hypertension [3,6,7,31]. An analysis of these data suggests that a high proportion of CAD patients have BP levels above the recommended goal in most European countries. Although most patients receive BPlowering drugs, a large proportion of them have elevated BP values. To the best of our knowledge, the present study is the first to present data allowing to estimate the operational efficacy of the same hospitals in terms of secondary   International Journal of Hypertension prevention of CAD events over a 20 year observation period. is provides a unique opportunity to track changes in the BP control of patients with CAD living in a specific area over a long period and who were treated in the same hospitals.

Limitations of the Study.
Our analysis has some limitations. Firstly, we were unable to assess the influence of changes in BP on the risk of cardiovascular complications. Secondly, participants were limited to those who had experienced an acute CAD event or had undergone a revascularization procedure. erefore, our results should not be directly applied to other CAD patients. irdly, the survey participants lived in a defined geographical area. Although the applicability of our results to other regions is uncertain, the trends described above reflect changes over time in the general population in Poland as well as in CAD patients from other European countries [24,31]. However, an important advantage of our analysis is that our results are not based solely on abstracted medical record data but took into account face-to-face interviews and examinations using the same protocol and standardized methods and instruments. erefore, to best of our knowledge, the presented results provide the most up-to-date and reliable information on BP trajectories in patients with CAD over the last 20 years.

Conclusion
e analysis of five multicenter surveys provides evidence of a decrease in BP in patients with established CAD over a period of two decades. e trend is independent of age, sex, and the education level of patients.

Data Availability
Data are available upon request.

Conflicts of Interest
e authors report no conflicts of interest.