The impact of diabetes mellitus and hypertension on clinical outcomes in a population of Iranian patients who underwent percutaneous coronary intervention: A retrospective cohort study

Abstract

There are heterogeneous data regarding the impact of diabetes mellitus (DM) and hypertension (HTN) on clinical outcomes after percutaneous coronary intervention (PCI). This study explored the effect of history of DM (hDM) and HTN (hHTN), separately and in combination with each other, on major adverse cardiac events (MACE) in short‐, mid‐, and long‐term intervals after PCI. Between 2000 and 2017, 1799 patients who had PCI were registered. They were categorized in four different groups: hDM, hHTN, hDM + hHTN, and no hDMQuery no hHTN. Incidence of myocardial infarction, revascularization, and coronary death totally considered as MACE was sought in short‐ (<24 hours), mid‐ (24 hours up to 6 months), and long‐term (more than 6 months) intervals after PCI. Among the subjects, 176 had hDM, 648 had hHTN, 370 had hDM + hHTN, and 605 were in no hDM no hHTN group. The median follow‐up time was 66.5 months. Time‐to‐event (time to the first MACE) was not significantly different between four groups. hHTN group was older and hDM group was younger at the time of enrollment PCI. Female gender was dominant only in hDM + hHTN group. Of the total, 130 patients (7.22%) experienced MACE. There was no MACE in short term, 23.07% of the MACEs were in mid‐term, and the remaining happened in long term. However, according to the rate ratio, incidence rate of MACE in mid‐tem was significantly higher than the long term. Also, MACE occurrence was significantly higher in hDM + hHTN and hHTN groups than the no hDM no hHTN group. Our study showed that the history of HTN significantly increases post‐PCI MACE rather than the history of DM. Having history of both DM and HTN synergistically raised MACE incidence. Incidence of MACE per month was higher in mid‐term than the long‐term interval.

1. INTRODUCTION

Non‐communicable diseases attributed to 73.4% of the total death in 2017 and the largest portion allocated to cardiovascular disorders. Of that, ischemic heart disease has assigned a considerable share in mortality. In spite of progression in the treatment of coronary heart disease (CHD) during the past three decades, its related death has increased by 22.3% from 2007 to 2017.1 Different therapies from anti‐anginal medications and lipid‐lowering drugs to invasive procedures such as percutaneous coronary intervention (PCI) and coronary artery bypass graft (CABG) have dramatically increased CHD survival. The number of PCI and CABG procedures was approximately equal in 1990 with a sharp rise in PCI since then.2

Among the several risk factors predisposing individuals to CHD, hypertension (HTN) is the most common one and is considered as a strong predictor of cardiovascular events.3, 4 High blood pressure increases the risk of all clinical presentations of CHD including angina pectoris, myocardial infarction (MI), and sudden death, nearly two‐ to threefold.4 In a meta‐analysis study, the rate of long‐term death was reported 8.55% in hypertensive compared to 5.81% in normotensive patients.4 Diabetes mellitus (DM) is known as another noticeable risk factor for CHD.5 The risk of being affected by CHD is intensified two‐ to fourfold at the presence of DM.6 It has been shown that HTN and DM are associated with increased risk of mortality after MI.3, 7 Notwithstanding, results of the studies were heterogeneous regarding the impact of HTN and DM, alone or in combination, on PCI outcomes.8, 9

It was reported that about 60% of diabetic patients suffered hypertension 10 and this rate is varied based on different factors including BMI and ethnicity.11 More than 80% of cardiovascular mortality around the world happens in low‐ and middle‐income countries.12 Considering high prevalence of these two disorders among Iranian adult population,4, 7 the aim of the present study is to determine the effect of DM and/or HTN on the rate of major adverse cardiac events (MACE) in CHD patients after PCI. Such understanding is necessary in terms of reducing the accompanied financial and social burdens.

2. METHODS

This study is in accordance with the Helsinki declaration and has been approved by the Research Ethics Committee of Shiraz University of Medical Sciences. This retrospective cohort was carried out on 1799 subjects who have undergone PCI in hospitals affiliated with Shiraz University of Medical Sciences between 2000 and 2017. Drug‐eluting stent (DES) was used in all patients. The criteria of doing angioplasty were based on ≥75% narrowing in LAD, diagonal, LCX, OM, RCA, PDA, and PLV branch. Doses of 80‐100 mg/kg heparin were given at the time of procedure. The loading dose of 600 mg clopidogrel and 325 mg aspirin (ASA) were given before the procedure followed by 150 mg clopidogrel and 325 mg ASA for 3 weeks. Clopidogrel (75 mg/day) and ASA (80 mg for non‐diabetic or 160 mg in case of DM) were administered at least for one year in all subjects and in diabetic patients at the minimum of 2 years. Also, atorvastatin 40‐80 mg daily was given to all patients for 3 weeks and continued lifelong for 20‐40 mg daily.

Study subjects were classified in four different groups including the following

• hDM group: Patients who had history of DM or receiving antihyperglycemic medications before enrollment PCI. DM was defined as fasting plasma glucose of ≥126 mg/dL (7.0 mmol/L) or 2‐h plasma glucose of ≥200 mg/dL (11.1 mmol/L) during oral glucose tolerance test or A1C of ≥6.5% (48 mmol/mol) or in a patient with classic symptoms of hyperglycemia or hyperglycemic crisis, a random plasma glucose of ≥200 mg/dL (11.1 mmol/L).13

• hHTN group: Patients who had history of HTN or receiving antihypertensive medications before enrollment PCI. HTN was defined as systolic blood pressure of ≥130 mm Hg or diastolic blood pressure of ≥80 mm Hg or both.14

• hDM + hHTN group: Patients with hDM and hHTN.

• no hDM no hHTN group: Patients without hDM and hHTN.

All the patients were tightly controlled for their hypertension and/or hyperglycemia. Subjects entered in the study after enrollment PCI. They were asked to attend in the clinic every 2‐3 months. If they were not adhering to the schedule, a nurse made phone calls to themselves (or their relatives if they were not accessible) and arranged a visit. Exercise tolerance test, technetium scans, echocardiography, and, if necessary, coronary angiography were done at follow‐up visits. The intended clinical outcomes to be followed up were MI, repeated coronary revascularization (CABG and PCI), and coronary death, totally considered as MACE. In all tables, age was regarded as the age of participants at the time of enrollment PCI. Time‐to‐event defines as the time of enrollment PCI to the time of MACE incidence and were classified in three categories: short‐, mid‐, and long term which assigned to less than 24 hours, 24 hours up to 6 months, and more than 6 months after PCI, respectively.

2.1. Statistical analyses

Data were presented as mean ± sd for continues and number(%) for categorical variables. Chi‐square test was used for comparing categorical variables. t test and analysis of variance (ANOVA) with post hoc test (Tukey) were recruited for comparing continues variables. Cox proportional‐hazards regression was used to find association between survival time and predictor variables. In order to compare MACE incidence rate between time intervals, rate ratio was measured. P value of <.05 assumes to be statistically significant. Analyses were performed using the statistical Package for Social Sciences version 16.0 (SPSS Inc) and survival package in R version 3.4.3 software.

3. RESULTS

Out of all, 61.2% was male. The age range was 31 to 94 years with a mean of 60.14 ± 11.21. The maximum follow‐up time was 201.47 months with a median of 66.5 months and a total of 121 623 person‐months. Minimum and maximum time‐to‐event was 15 days and 150.6 months, respectively, with a median of 49.91 months. Among all the patients, 7.2% have experienced MACE, of which 6.2% had MI, 69.2% had revascularization, and 24.6% passed away due to coronary death. The MACE rate was 1.06 per 1000 month.

Table 1 shows baseline characteristics of the participants. The majority of participants were in the hHTN group. Age of the participants at the time of enrollment was significantly different among four groups. hDM group was younger and hHTN group was older than the others. Multiple comparison approach revealed that the age difference was between the following groups:

• no hDM no hHTN and hHTN groups,

• no hDM no hHTN and hDM + hHTN groups,

• hDM and hHTN groups,

• and finally, hDM and hDM + hHTN groups.

Table 1.

Baseline characteristics of the participants

Variables	Patients (n = 1799)				P value
	No hDM no hHTN (n = 605)	hDM (n = 176)	hHTN (n = 648	hDM + hHTN (n = 370)
Age at enrollment PCI (y)	57.3 ± 11.14	56.94 ± 10.85	62.85 ± 11.84	61.34 ± 9.80	<.001
Gender (Male)	482 (79.66)	118 (67.04)	345 (53.24)	156 (42.24)	<.001
MACE	27 (4.5)	11 (6.3)	47 (7.3)	45 (12.2)	<.001
Time‐to‐event (mean ± SD in mo)	52.4 ± 39.80	49.18 ± 57.50	52.38 ± 43.20	50.54 ± 47.4	.994

Data were presented as mean ± sd for age and time‐to‐event; Number (%) for other variables.

Bold values imply statistical significance.

The larger part of individuals in no hDM no hHTN, hDM, and hHTN groups were male. However, females were more in hDM + hHTN group. Of the hDM, hHTN, and hDM + h HTN groups, 6.3%, 7.3%, and 12.2% experienced MACE, respectively, while these events happened only in 4.5% of the no hDM no hHTN group. Also, differences in time‐to‐event were not statistically significant among groups.

Table 2 shows univariate and multivariate Cox proportional hazard model analysis of MACE incidence with respect to age, gender, and different groups. Unlike gender, age affected MACE occurrence. MACE rate in hDM + hHTN group was significantly higher compared to no hDM no hHTN (P value <.001, HR = 3.774). While hDM group did not show any statistical difference with no hDM no hHTN group in terms of MACE incidence, it was significantly higher in hHTN group.

Table 2.

Covariates effect on MACE in univariate and multivariable approach in total data

	Patients (n = 1799)		Unadjusted P value	Adjusted p value (multivariate analysis)
	MACE (n = 130)	no MACE (n = 1669)		HR	L	U	P value
Age at enrollment PCI (y)	57.82 ± 11.22	60.32 ± 11.23	.015	0.981	0.965	0.997	.019
Gender (Male)	81 (62.3)	1020 (61.11)	.788	1.077	0.745	1.557	.692
no hDM no hHTN	27 (20.76)	578 (34.63)	<.001	1	‐	‐	‐
hDM	11 (8.46)	165 (9.88)		1.542	0.763	3.114	.228
hHTN	47 (36.15)	601 (36.00)		2.031	1.246	3.311	.005
hDM + hHTN	45 (34.61)	325 (19.47)		3.774	2.299	6.195	<.001

Data were presented as mean ± sd for age and number (%) for other variables.

Male is the reference for gender and no hDM no hHTN group is the reference with regard to hDM and hHTN status.

Abbreviations: CI, confidence interval; HR, hazard ratio; L, lower bound; U, upper bound.

Bold values imply statistical significance.

We did not observe any events in short term at all. Table 3 demonstrates multivariate approach of MACE rate in mid‐term. The MACE rate was 3.31 per 1000 person‐month with a total of 34 patients being affected. This rate was significantly different in hDM + hHTN group compared to no hDM no hHTN group (P value <.001, HR = 3.053). MACE rate was 0.79 per 1000 person‐month (96 patients) in long term (Table 4). In this interval, MACE incidence was remarkably higher in hDM + hHTN and hHTN group compared to no hDM no hHTN group. Survival probability curve (Figure 1) is in line with the aforementioned results. MACE incidence rate was compared between mid‐term and long term (Table 5). According to this table, incidence rate of MACE was significantly different between the two time intervals among all the four groups. In total, MACE incidence rate of mid‐term was 4.17 times higher than the long term (95%CI: 2.73‐6.22).

Table 3.

MACE incidence rate analysis based on age, gender, hDM, and hHTN by multivariable approach in mid‐term

	Patients (n = 1799)		HR (95% CI)			Adjusted P value
	MACE (n = 34)	no MACE (n = 1765)	HR	L	U
Age at enrollment PCI (y)	58.38 ± 10.91	60.18 ± 11.25	0.987	0.956	1.018	.370
Gender (Male)	21(61.76)	1080(61.18)	1.052	0.513	2.161	.885
no hDM no hHTN	4(11.76)	601(34.05)	1	‐	‐	‐
hDM	3(8.82)	173(9.80)	2.783	0.621	12.466	.181
hHTN	9(26.47)	639(36.20)	2.684	0.811	8.885	.106
hDM + hHTN	17(50.0)	353(20.0)	9.375	3.053	28.785	<.001

Data were presented as mean ± sd for age and number (%) for other variables.

Male is the reference for gender and no hDM no hHTN group is the reference with regard to hDM and hHTN status.

Abbreviations: CI, confidence interval; HR, hazard ratio; L, lower bound; U, upper bound.

Bold values imply statistical significance.

Table 4.

MACE incidence rate analysis based on age, gender, hDM, and hHTN by multivariable approach in long term

	Patients (n = 1765)		HR (95% CI)			Adjusted P value
	MACE (n = 96)	no MACE (n = 1669)	HR	L	U
Age at enrollment PCI (y)	57.63 ± 11.37	60.23 ± 11.23	0.984	0.965	1.003	.099
Gender (Male)	60 (62.5)	1020 (61.11)	1.084	0.709	1.662	.712
no hDM no hHTN	23 (23.95)	578 (34.63)	1	‐	‐	‐
hDM	8 (8.33)	165 (9.88)	1.323	0.590	2.966	.496
hHTN	37 (38.94)	601 (35.37)	1.914	1.118	3.279	.018
hDM + hHTN	28 (29.16)	325 (19.47)	2.795	1.577	4.956	<.001

Data were presented as mean ± sd for age and number (%) for other variables.

Male is the reference for gender and no hDM no hHTN group is the reference with regard to hDM and hHTN status.

Abbreviations: CI, confidence interval; HR, hazard ratio; L, lower bound; U, upper bound.

Bold values imply statistical significance.

Figure 1.

Survival probability curve

Table 5.

Comparison of MACE incidence rate between mid‐ and long‐term intervals

Groups	No.	Mid‐term			Long term			Rate ratio (95% CI)
		No. of person‐month at risk	No. of MACE	Incidence rate (per 1000 person‐mo)	No. of person‐mo at risk	No. of MACE	Incidence rate (per 1000 person‐mo)
no hDM no hHTN	624	3573	12	3.30	43 561	33	0.75	4.43 (2.08‐8.80)
hDM	176	996	3	3.01	11 672	9	0.76	3.92 (0.68‐15.72)
hHTN	642	3676	10	2.72	44 038	31	0.70	3.84 (1.69‐8.13)
hDM + hHTN	357	2106	9	4.46	21 631	23	1.05	4.22 (1.71‐9.47)
Total	1799	10 261	34	3.31	120 902	96	0.79	4.17 (2.73‐6.22)

Rate ratio is incidence rate of MACE in mid‐term over long term.

Abbreviation: CI, confidence interval.

Bold values imply statistical significance.

4. DISCUSSION

The present study is among the unique one in terms of long follow‐up phase (17 years with a median month of 66.5) collecting data from 1799 patients. The main finding of this study is that coexistence of HTN history and DM history in CHD patients increased the risk of MACE after PCI. MACE rate was significantly higher in hDM + hHTN compared to no hDM no hHTN group both in mid‐ and long‐term intervals. No MACE was detected in short term while 74% of events happened in long term. However, rate ratio showed that MACE incidence was more likely to happen in mid‐term than long term. We found that the occurrence of MACE after PCI was significantly higher in hHTN than no hDM no hHTN group. The role of hHTN is more striking than hDM in development of MACE in our study. There is no statistical difference in post‐PCI outcomes between hDM and no hDM no hHTN group.

There are divergent reports about the magnitude of the role of hypertension on post‐PCI outcomes from weak influence to a strong predictor of mortality.9, 15, 16 In some researches, hypertension is said to be in association with poor prognosis after PCI 8, 17, 18 while neutral effects of hypertension were observed in some others.9, 19, 20 Meanwhile, hypertension is accused for some malfunctions in cardiovascular system such as malignant ventricular arrhythmia, development of atherosclerosis, and sudden cardiac death after MI. Even transient moderate HTN after infarction is assumed to be the underlying cause of myocardial dysfunction.21 Possibly, these are among the known and unknown factors that determine the hypertension‐related events after PCI.

With respect to DM, it enhances the risk of cardiovascular events after MI.22 It was reported that CHD mortality becomes higher two‐ to fivefold due to DM compared to age‐adjusted non‐diabetic individuals.23 Case‐fatality rate after acute coronary syndrome was noticeably higher in diabetic than non‐diabetic patients.24, 25, 26 Adverse outcomes like mortality and repeated revascularization after PCI were shown to be related to the history of DM.27 A meta‐analysis declared that in‐hospital, short‐, and long‐term mortality was occurred in 3.02%, 4.12%, and 9.24% of diabetic patients, respectively, compared to 1.59%, 2.46%, and 5.35% of non‐diabetic counterparts. Multi‐vessel coronary disease and chronic total occlusion in diabetic patients possibly lead to severe stent thrombosis, stroke, silent myocardial infarction, and other major adverse events after PCI.28

Simultaneous existence of DM and HTN increases the risk of cardiovascular diseases up to fourfold compared to non‐diabetic normotensive subjects.11 In a study, the risk of mortality in acute coronary syndrome patients with history of either HTN or DM was higher than the same patients without such a history. They believed that DM was the main culprit for mortality, and an additive rise for death rate was emerged when hypertension was accompanied.15 Metabolic disorders importantly high plasma glucose and high blood pressure augment each other and coexistence of both accelerate the progression of atherosclerosis.29, 30

Diabetes mellitus and HTN are known as the complication of each other probably due to the intertwined pathophysiological pathways.31 They increase the activity of sympathetic nervous system. This causes vasoconstriction via the action of norepinephrine on vascular smooth muscle cells on one hand and hyperactivity of renin‐angiotensin‐aldosterone system (RAAS) at the other hand. Both conditions exacerbate the hypertension state. Also, growing body of evidence implies that diabetes could rise after RAAS overactivation.32, 33 DM promotes atherosclerosis through several adverse functions including platelet hyperactivity, reduced fibrinolytic capacity, increased concentrations of hemostatic proteins, and endothelial dysfunction.34 Irreversible glycosylation of collagen raised by hyperglycemia results in excessive cross‐linkage and stiffness of the vessels.32 Inflammatory state raised by high blood glucose leads to the production of superoxide and eventually endothelial dysfunction.35, 36 Also, insulin stimulates migration and proliferation of vascular smooth muscle cells.32 All the aforementioned issues have roles in worsening the coronary involvement by confining the blood supply and facilitate MACE occurrence.

It is noteworthy that hyperinsulinemia is the possible reason for left ventricular hypertrophy and decreased ventricular function.32 Left ventricular hypertrophy was said to be a more significant predictor of death rather than left ventricular systolic function and even severe CAD.37 Moreover, of the consequences of diabetic nephropathy are albuminuria which is reported to be remarkably associated with CAD, myocardial infarction, and death.32

Nevertheless, there are limited and discrepant evidences on the separate and combined impact of DM and HTN on post‐PCI outcomes. Some researchers agreed with this notion that DM as well as HTN increased the rate of mortality in patients with acute MI. Reports showed that cardiovascular events were worse in comorbid individuals than those with only each of two.8 HTN not per se, but in combination with DM, was related to increased mortality after PCI.9 Diabetic hypertensive patients suffered multi‐vessel disease more than those with non‐diabetic normotensive ones after acute MI.38 Low ventricular ejection fraction is one of the resultant effect of concurrent existence of DM and HTN.8 Also, severe hypertrophy in left ventricle and consequent impairment in systolic or diastolic function were developed more simple in hypertensive diabetic patients in comparison to subjects with each one at a time. It may be postulated that these are the underlying reasons in charge of relevant complications due to HTN and DM in CAD patients after PCI.

Hypertension and DM have challenged global health and undermine the efforts raised by improvements in healthcare interventions.33 In fact, the consequences of DM and HTN have reached to a concerning level that needs international attention in order to reduce or even stop their catastrophic effects. To obtain such a goal, comprehensive investigations with considering population differences provide suitable substrate to make evidence‐based decisions.

Some limitations of the current study are as follows: The onset and severity of DM and HTN did not consider in our study. Shortcomings in identification of patients with respect to DM and HTN might be occurred. Inability to check adherence of participants to antihypertensive or antidiabetic therapies and heterogeneity of medications should be kept in mind. Possibly, there were some confounder variables that did not come into account. Defining the exact time of MACE incidence became impossible in some cases due to lack of cooperation.

5. CONCLUSION

The effect of HTN history was more pronounced than DM history on MACE occurrence after PCI. Post‐PCI MACE incidence was higher in patients having both histories of DM and HTN than other subjects. Incidence of MACE was more probable in six months after PCI compared to longer time intervals.

CONFLICT OF INTEREST

The Authors declares that there is no conflict of interest.

AUTHORS CONTRIBUTION

MJZ, SSM and SK designed the study. SSM and SK acquired data. MS, EB and IRJ had roles in data interpretation. IRJ wrote the initial draft. All the authors reviewed the manuscript critically.

Zibaeenezhad MJ, Mohammadi SS, Sayadi M, Khorshidi S, Bahramali E, Razeghian‐Jahromi I. The impact of diabetes mellitus and hypertension on clinical outcomes in a population of Iranian patients who underwent percutaneous coronary intervention: A retrospective cohort study. J Clin Hypertens. 2019;21:1647–1653. 10.1111/jch.13705