|Home | About | Journals | Submit | Contact Us | Français|
High blood pressure (BP) levels expose patients treated with percutaneous coronary interventions (PCI) to very high risk of 10-year cardiovascular morbidity and mortality.
To investigate the role of BP levels at the time of PCI on the risk of in-stent restenosis (ISR).
We retrospectively included 796 patients previously treated with PCI, who underwent repeated angiography for recurrent angina or reversible myocardial ischemia. Patients were stratified into either case (n = 354) and control (n = 442) groups in the presence or absence of ISR (defined as in-stent diameter stenosis ≥50%). BP levels were measured at the time of first and second procedures. Normal BP levels were defined for <140/90mm Hg.
Patients with normal BP showed significantly higher ISR-free survival (Log-rank: 5.937; P = 0.015). Both systolic (HR (95% CI): 0.731 (0.590–0.906)) and systolic/diastolic BP (HR (95% CI): 0.757 (0.611–0.939)) were significantly and independently associated with lower risk of ISR at Cox-regression analysis, adjusted for potential confounding factors, including stent type and concomitant medications. Patients with ISR showed lower rates of normal systolic/diastolic BP values (166 (47%) vs. 254 (57%); P = 0.003) compared to controls. They also received higher stent number (1.40±0.74 vs. 1.24±0.51; P < 0.001) with higher stent length (24.3±15.6 vs. 21.7±13.9mm; P = 0.012), and lower rate of drug-eluting stents (DESs) (210 (48%) vs. 139 (40%); P = 0.025) compared to controls.
Normal BP at the time of PCI is associated with nearly 24% risk reduction of ISR as evaluated in a new angiography in patients with coronary artery disease.
High blood pressure (BP) is associated with significant increased risk of coronary artery disease and major cardiovascular events.1 This increased risk is often enhanced by the concomitant presence of additional cardiovascular risk factors, organ damage, and other comorbidities.2,3 Despite effective and well-tolerated BP lowering therapies, hypertensive patients remain at higher risk of acute coronary syndromes, recurrent angina, and revascularizations as compared with normotensive patients.4 A suboptimal control of BP and off-target therapeutic regimens might potentially contribute to a worse clinical outcome in these treated hypertensive patients. In turn, excessive BP reductions have been advocated as a potential contributor to increased risk of coronary events in treated hypertensive patients, particularly in those with underlying coronary artery disease.5–8 As a consequence, both European guidelines9,10 and United States recommendations11 have recently revised the BP targets to be achieved in high risk hypertensive patients with known coronary artery disease, recommending a BP control of below 140mm Hg for the systolic and below 90mm Hg for the diastolic BP levels. In the same documents, more aggressive BP lowering strategies in patients with coronary artery disease are discouraged.9–11
Percutaneous coronary interventions (PCI) have increasingly been performed in patients with coronary artery disease and have become the standard optimal treatment in patients with myocardial infarction.12 This widespread use of PCI has been boosted by the more effective drug-eluting stents (DESs) that have significantly contributed to reduce the incidence of in-stent restenosis (ISR) and repeat revascularization as compared to bare-metal stents (BMSs).13 Yet, albeit reduced, ISR is far from disappeared even with DES, as a consequence of the application of this technology in more challenging clinical and lesion settings.14 Beyond the known procedural and technical factors, BP levels and its therapeutic control may play an important role in modulating the risk of ISR and recurrences after PCI. High BP may, in fact, promote the restenosis process through different mechanisms, including shear stress, endothelial dysfunction, accelerated atherosclerosis, accumulation of growth factors, and the higher prevalence of concomitant cardiovascular and metabolic risk factors.15–19 While it is clear the prognostic role of BP on increased coronary risk,20 as well as the role of admission BP on the incidence in recurrent myocardial ischemia after myocardial infarction,21 the hypothesis that BP levels may interfere with the healing process after PCI appears reasonable.
Therefore, the aim of our study was to retrospectively investigate potential impact of BP levels measured at the time of the first PCI on the risk of developing ISR over time. In addition, we evaluated the potential interaction between baseline BP levels with the type of stent implanted (i.e., BMS vs. DES) and the risk of ISR.
This is a retrospective study including consecutive patients previously treated with PCI (index procedure) and stent implantation, who underwent repeated coronary angiography between 2004 and 2011 for recurrent angina or reversible myocardial ischemia during exercise or stress test at the Division of Cardiology of Sant’Andrea Hospital of Rome, Italy.
According to the study design, enrollment of the study population was performed at the time of second coronary angiography. After this procedure, included patients were stratified into either case or control groups on the basis of the presence or absence of ISR, respectively. Then, anthropometric and clinical characteristics, including BP levels, at the time of the first PCI were retrospectively collected by our clinical databases and centrally analyzed.
To be included in this study, a written informed consent to personal clinical data management and analysis was signed. Patients were excluded in case of: (a) additional revascularizations (either percutaneous or surgical) occurred before the index procedure; (b) secondary or malignant hypertension; and (c) neurological or mental disorders, endocrine, renal, immunological, or solid cancers, and any other acute or chronic disease which may had at least, in part, an impact on BP levels.
The study conformed to the Declaration of Helsinki and its subsequent modifications. The confidentiality of the data of each patient included in the present study was carefully and strictly protected. Informed consent and a clear comprehension of the procedure was obtained from all patients and the study was approved by the local Ethical Committee.
Clinic BP levels were noninvasively measured by 2 physicians (A.F. and G.T.) who are hypertension specialists of the European Society of Hypertension and blinded for the study groups. Alternatively, BP measurements were performed by 2 trained nurses, who are active in our Hypertension Center. BP levels were predominantly measured in the supine position at least 3 times, using the mean of the last 2 measurements. Clinic BP measurements were performed early in the morning (between 6:00 am and 10:00 am), so that the vast majority of the patients included in the study had already taken their medications. All BP measurements were performed 3–6 hours before the first PCI and before the second coronary angiography with automated, validated devices, following the recommendations of international guidelines.10 This standardized approach should have limited potential influence of patients’ reactivity or measurements’ imprecisions. In addition, the staff involved in BP measurements had been long trained to avoid common bias (e.g., rounding numbers).
Interventional cardiologists were blinded for both baseline and final BP levels considered for the present analysis. It should be also noted, however, that BP assessments were routinely performed before, during and after coronary procedures following the standard protocol adopted in our CathLab. BP levels did not affect the therapeutic choice to perform or not to perform a PCI. Based upon the main international recommendations, other factors (anatomical, clinical, procedural) influenced the therapeutic decisions during the procedure.
Normal systolic/diastolic BP thresholds were defined as <140/90mm Hg.10
Diagnosis of hypercholesterolemia was made in the presence of total cholesterol levels above 190mg/dl or low-density lipoprotein cholesterol levels above 130mg/dl, while hypertriglyceridemia was considered for triglyceride levels above 150mg/dl, or stable lipid-lowering drug treatment in both conditions.22,23 In addition, low levels of high-density lipoprotein cholesterol levels were defined when they were below 40mg/dl in men and below 50mg/dl in women.23,24 Finally, diabetes was defined in the presence of fasting plasma glucose levels above 126mg/dl or 2-hour post-load glucose of 200mg/dl or more during an oral glucose tolerance test.25–27 As for BP measurements, all these parameters were derived from the clinical database of each patient included in the study protocol, if available, and measured at the time of the first and second coronary angiographies. In particular, at the time of the first coronary angiography, all patients received written detailed recommendations for proper life-style changes and correction of modifiable risk factors, as well as pharmacological therapies, aimed at achieving the recommended therapeutic targets of major cardiovascular risk factors (i.e., lipid and glucose profile), according to the available guidelines’ recommendations.22–27
PCI procedure and decision of the stent type to be implanted were based on the clinical choice of the interventional cardiologist. In case of multivessel PCI, the same type of stent was used. All patients were prescribed acetylsalicylic acid 75–100mg once a day lifelong. A 300–600mg clopidogrel loading dose was administered to all patients, followed by 75mg once a day for at least 1 month in case of BMS, and for at least 1 year in case of DES implantation.28,29 Therapeutic agents for secondary prevention, such as antihypertensive, lipid-lowering, and glucose-lowering agents, were also prescribed according to physicians’ clinical experience, patients’ characteristics, and mostly current guidelines.25,30,31
Coronary angiography at the follow-up with possible repeated PCI was performed only if clinically indicated, due to recurrence of angina and/or signs of myocardial ischemia. ISR was defined based on the angiographic evidence of ≥50% stenosis within the previously stented segment. Target lesion revascularization was defined as a repeated PCI performed within the stent or 5mm proximal to and distal to the previously stented segment.
Baseline characteristics are reported as counts and percentages for dichotomous variables or as mean ± SD for continuous variables. Verification of normal distribution of data was accomplished using histograms and Kolmogorov–Smirnov test. Baseline comparability of the 2 study groups was examined with the use of a T-test for quantitative variables and a chi-square test for qualitative variables. Time-dependent occurrence of events was investigated with Kaplan–Meier curves and Cox proportional-hazard models in the overall population sample, stratified according to the presence or absence of clinic systolic/diastolic BP control or according to type of stent (either DES or BMS). Differences in survival were compared using the log-rank test. To evaluate the significance of predictors of ISR, hazard ratio (HR), and 95% confidence interval (CI) were derived from 2 Cox proportional-hazard regression models (model 1 and model 2), both adjusted for age, gender, and major risk factors. BP levels entered as systolic BP in model 1 and systolic/diastolic BP in model 2. The proportional-hazards assumption was checked by plotting Schoenfeld residuals summarized by a loess smoothing line against survival time for continuous variables, and by the log-minus-log test of proportionality for categorical variables. Both models were assessed for multicollinearity. All tests were 2-sided, and P value of <0.05 was considered statistically significant. All calculations were generated using SPSS, version 20.0 (SPSS, Chicago, IL).
For the present analysis, we identified 796 patients previously treated with PCI between 1998 and 2011, who underwent repeated coronary angiography for recurrent angina or evidence of reversible myocardial ischemia. Among them, 354 (44.5%) presented with ISR, while the remaining 442 (55.5%) not showing ISR were used as controls. None of the patients included in the study underwent surgical revascularization. Clinical characteristics of the overall patient population are reported in Table 1.
There were no significant differences between the 2 groups, with the only exception of dyslipidaemia that was less frequent in the restenosis than in the control group. Despite this, both total cholesterol and high-density lipoprotein cholesterol levels were slightly higher in the restenosis group. No significant differences in the drug therapy between the 2 groups were observed with regard to major antihypertensive drug classes, including angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, calcium-antagonists, and diuretics, with the exception of beta-blockers that were significantly more commonly used in the control compared to restenosis group. Use of acetylsalicylic acid was slightly larger in the control group, whereas use of clopidogrel was significantly more frequent in the restenosis group. Statins were uniformly used in both groups. In addition, there were no relevant differences between the 2 groups with regard to glucose and lipid profile, renal function, and distribution of cardiovascular risk factors.
BP levels and profiles are reported in Table 2. Mean systolic BP levels were in the high-normal range in both groups, though significantly higher in the ISR group as compared to control group. Mean diastolic BP levels were within the normal BP range, with no significant differences between the 2 groups. Proportions of patients having normal systolic and systolic/diastolic BP at the time of the index procedure were significantly lower in the restenosis group, whereas no significant differences were observed with regard to proportions of patients having normal diastolic BP levels between the 2 groups. No significant differences were observed with regard to baseline and follow-up heart rate between the 2 groups.
Angiographic characteristics are reported in Table 3. Compared with the control group, coronary stenosis treated at the index procedure were more frequently located in the left circumflex artery and in small vessels, and less frequently located at coronary bifurcation in the ISR group. In this latter group on average were implanted: (a) more stents per patient with more over-lapping and over-gapping; (b) longer stents with slightly smaller diameter; and (c) less frequent DES compared to control group.
Repeated angiography was performed at a median follow-up of 8 (4–18) months. In the ISR group, we observed higher in-segment diameter stenosis, higher rate of total occlusions, and target lesion revascularization as compared with control group. Rate of nontarget vessel revascularization was similar between the 2 groups.
At Kaplan–Meier analyses in the overall population patients with normal BP levels showed significantly higher ISR-free survival, either for the systolic or for the systolic/diastolic BP levels (Figure 1). Even after stratifying for the type of the implanted stent, we observed that patients with normal systolic BP levels showed higher ISR-free survival, both in those who received DES and in those who received BMS (Figure 2).
Multivariable Cox-regression analysis adjusted for several confounder factors, including age and gender, confirmed that normal systolic (0.731 (0.590–0.906); P = 0.004) or systolic/diastolic (0.757 (0.611–0.939); P = 0.011) BP levels were associated with lower risk of ISR (Table 4). Other factors associated with ISR were total cholesterol (P = 0.023), acetylsalicylic acid (P = 0.022), clopidogrel (P < 0.001), beta-blockers (P < 0.001), and lesion location in small vessels (P = 0.001).
In patients treated with BMS, adjusted multivariable Cox-regression analysis confirmed the significant association between normal systolic (0.705 (0.534–0.930), P = 0.013) and systolic/diastolic BP levels (0.713 (0.540–0.941), P = 0.017), with lower risk of ISR. In patients treated with DES, the latter association achieved borderline significance with normal systolic (0.707 (0.501–0.996), P = 0.048) BP levels, while it was not significant for normal systolic/diastolic BP (0.736 (0.522–1.036), P = 0.079).
We retrospectively analyzed the potential association between baseline systolic/diastolic BP levels and risk of developing coronary ISR in patients previously treated with PCI and undergoing repeated coronary angiogram due to recurrent symptoms and/or reversible signs of myocardial ischemia. Our findings demonstrate a significant and independent protective role of both systolic and systolic/diastolic BP in terms of reduced incidence of ISR in this high-risk patient population. This protective effect of BP was more pronounced in patients treated with BMS and less pronounced in patients treated with DES. Although limited by the retrospective nature of the study design, to our knowledge these findings demonstrate for the first time a correlation between clinic BP levels and ISR. These results may be, in our opinion, quite relevant for a more complete understating and management of the process of ISR after PCI, and deserve some comments.
First of all, BP targets in high-risk hypertensive patients with coronary artery disease have recently been questioned and revisited in the latest European guidelines10 and United States recommendations.11 The change in recommendations followed the demonstration of a J-curve phenomenon for cardiovascular and renal outcomes in patients with underlying coronary artery disease, who received intensive antihypertensive therapy aiming at a very tight BP control reaching levels below 130/80mm Hg or lower.5–8 The latest sets of guidelines suggest now to achieve in these patients BP targets below 140mm Hg for the systolic and below 90mm Hg for the diastolic BP levels.10,11
Despite the emphasis given in the guidelines to patients with coronary artery disease, in most of the clinical trials and available registries investigating the incidence and clinical characteristics of patients presenting with ISR after PCI, data on BP control are often not available and its relative impact on ISR is generally disregarded.
Our analysis confirmed, for the first time, the beneficial effects of BP levels maintained below the thresholds of 140/90mm Hg in these high-risk patients with coronary artery disease treated with PCI. The majority of the patients included in our study were hypertensive (about 79%) and under treatment with optimal pharmacologic therapies, including renin-angiotensin system inhibitors (about 86%), beta-blockers (55%), or calcium-channel blockers (38%). In this latter regard, although a slightly higher use of beta-blockers in the ISR group was recorded when compared to the control group, there was no significant difference in heart rate between the 2 groups. In spite of the lack of difference for this parameter, we cannot exclude that beta-blockers, through their multiple properties, including BP reduction, may have contributed in part to the observed results. In fact, they are recommended as a first choice treatment in hypertensive patients with coronary artery disease.10,11
We also observed that normal systolic and systolic/diastolic BP levels at the index procedure remained significantly and independently associated with lower ISR rate during the follow-up, even after adjustment for concomitant medications (i.e., beta-blockers, acetylsalicylic acid, and clopidogrel) and other important confounders (i.e., total cholesterol, left circumflex artery stenosis, and small vessels location). This association was particularly pronounced when patients were implanted at the index procedure with BMS, while it was less pronounced and limited only to systolic BP control in patients implanted with DES. These observations further emphasize the need and relevance of a complete secondary prevention strategy, which must include the achievement of effective BP control in patients at high risk for recurrent coronary events. In this regard, our findings confirm and further extend to the risk of coronary ISR, previous data indicating systolic baseline BP levels as powerful predictor of mortality and major cardiovascular events.32–35 Increased systolic BP levels have, indeed, demonstrated to be independently related to high risk of developing not only major cardiovascular complications, mostly including coronary artery disease, myocardial infarction and angina pectoris, but also stroke, congestive heart failure, and renal failure.32–35 Despite diastolic BP has a prominent role in the regulation of coronary flow, systolic BP represents the major factor implicated for the risk of coronary complications. Thus, it may not be surprising that diastolic BP resulted to not be related to ISR, since coronary flow occurs during diastole and thus lower diastolic BP could be even associated with higher risk of myocardial infarction. This is indeed been shown in a number of previous studies.36–38
Our study is the first to demonstrate that abnormal systolic BP levels are related to increased risk of developing coronary ISR in patients previously treated with coronary revascularization. On the other side, the strong association between BP and ISR observed especially in patients implanted with BMS underscores the importance of having baseline systolic and diastolic BP levels below the normal thresholds as useful parameters to identify those patients at particular higher risk of coronary ISR and recurrent revascularization. Of interest, in patients implanted with DES this association is weak and only limited to systolic BP, allowing to speculate that either an optimal BP or DES implantation should be advocated in patients at higher risk of coronary ISR.
This is a retrospective study including all patients undergoing repeat coronary angiography after previous PCI. We recognize the possible selection bias deriving from the lack of systematic control angiography that might have contributed to the inability in detecting clinically silent coronary ISR. There was no independent Core Lab for event assignment and evaluation. Although of potential clinical relevance, information on BP levels before (6 months) the index procedure were not considered among the aims of the current analysis. Data on glucose and lipid profile, renal function, and distribution of cardiovascular risk factors did not show relevant differences between the 2 groups either at the time of the second angiography or compared with baseline observation. Given the importance of ambulatory BP monitoring to integrate the individual BP profile, international guidelines have only recently recommended this investigation for the clinical management of hypertension in specific clinical conditions but not in all patients with hypertension or in those with coronary artery disease. Thus, systematic evaluation of ambulatory BP monitoring was not originally considered in our study, which closely reflects current clinical practice in these patients. Finally, different dosages, as well as different molecules, within the drug classes of antihypertensive, antiplatelet, and lipid-lowering agents may have different impacts on the main outcomes of the analysis. In spite of these limitations, data sustain the work hypothesis and strongly support the need for prospective, specifically designed investigations.
To our knowledge, our study firstly demonstrated that normal BP levels measured at the time of the first coronary procedure were associated with reduced risk of ISR in coronary artery disease patients previously treated with PCI and presenting with recurrent symptoms or signs of myocardial ischemia. In our population sample, total cholesterol, use of antiplatelet agents and beta-blockers, and specific locations of the previously implanted coronary stents were associated with a highly significant reduction in adjusted analyses for the risk of ISR. Of note, having BP above the normal values (i.e., 140/90mm Hg) was, in fact, associated with a 24% risk reduction of coronary ISR and repeat revascularization, especially in patients implanted with BMS. On the basis of our current observations, systematic evaluation of BP levels and control of BP within the normal range should be recommended, when a PCI procedure is undertaken.
The authors declared no conflict of interest.