Prognostic Impact of Baseline Low Blood Pressure in Hypertensive Patients With Stable Coronary Artery Disease of Daily Clinical Practice

Authors


Martín Ruiz Ortiz, PhD, MD, C/San Adolfo 18, 3°A, 14005 Córdoba, Spain
E-mail:maruor@gmail.com

Abstract

J Clin Hypertens (Greenwich). 2012;00:00–00 ©2012 Wiley Periodicals, Inc.

The authors’ aim was to investigate the prognostic value of first-visit systolic and diastolic blood pressure (SBP/DBP) in hypertensive patients with stable coronary artery disease (sCAD) in conditions of contemporary daily clinical practice. From February 1, 2000, to January 31, 2004, 690 consecutive hypertensive patients with sCAD (mean age 68±10 years, 65% male) were prospectively followed in the outpatient cardiology clinic for major events (acute coronary syndrome, revascularization, stroke, heart failure, or death) and associations with baseline SBP/DBP were investigated. At first visit, median SBP/SDP were 130/75 mm Hg (interquartile range, 25–75; 120–140/70–80 mm Hg). After 25 months of follow-up (median), 19 patients died (2.8%); 10 from cardiovascular causes (1.5%), 87 patients experienced a coronary event (13%), and 130 patients (19%) a major event. After adjusting for baseline variables, DBP <75 mm Hg or SBP <130 mm Hg resulted in independent predictors of major events (hazard ratio [HR], 1.52; 95% confidence interval [CI], 1.07–2.16, P=.02; HR, 1.68; 95% CI, 1.18–2.40, P=.004, respectively), coronary events (HR, 1.78; 95% CI, 1.15–2.75, P=.009; HR, 1.84; 95% CI, 1.20–2.83, P=.005, respectively), and cardiovascular mortality (HR, 7.02; 95% CI, 1.26–39.04, P=.03; HR, 9.26; 95% CI, 1.33–64.32, P=.02, respectively). In this study, a low first-visit SBP or DBP was associated with an adverse prognosis in hypertensive patients with sCAD of contemporary daily clinical practice.

In hypertensive patients with stable coronary artery disease (sCAD), recent post hoc analysis of clinical trials,1–5 old observational data,6–9 and meta-analysis10 suggest that reduction of systolic or diastolic blood pressure (SBP or DBP) to very low values may be accompanied by an increase in the incidence of cardiovascular events (the so called J-curve phenomenon). However, little is known about the prognostic impact of baseline blood pressure (BP) in patients with sCAD of contemporary daily clinical practice. Our objective was to test the prognostic value of first-visit SBP and DBP in a prospective observational cohort of patients with sCAD in this setting.

Methods

From February 1, 2000, to January 31, 2004, all consecutive patients with sCAD who were monitored at two cardiology outpatient clinics offered by a university hospital and who agreed to participate were included in this prospective study. This study was designed to determine the prognosis of sCAD in the present setting and to identify any associated predictive factors.11 We selected all hypertensive patients at baseline (SBP/DBP >140/90 mm Hg or taking antihypertensive medication) for this analysis.

Patients were deemed to have sCAD when they met ≥1 of the following criteria: a history of acute coronary syndrome (ACS), ie, unstable angina or acute myocardial infarction; surgical or percutaneous coronary revascularization at least 3 months before inclusion; chest pain with exercise test, myocardial isotopic perfusion scan or stress echocardiogram compatible with ischemia; or coronary angiography showing >70% stenosis of the lumen diameter of an epicardial vessel without serious valve disease.

Patient clinical information, including SBP and DBP measurement, was prospectively collected at the first appointment. The patients were managed according to their doctors’ judgment, based on the current scientific societies recommendations12–14 and followed-up annually via outpatient appointments. Major events were recorded, including all-cause death, hospitalization for ACS, the need for revascularization (unforeseen in the initial assessment), stroke, or admission for heart failure. Near the end of the study, review of medical records, telephone interviews, and contacts with family doctors were used in an effort to minimize missing data and patients lost to follow-up. Causes of death for patients who died in the hospital were obtained from the analysis of medical records. For other cases of death, information was obtained from family doctors and family members. No systematic consultation of death certificates was undertaken. The study adhered to the Helsinki recommendations for medical studies.

Baseline and follow-up data were introduced into a database created using SPSS v.12.0 software (SPSS Inc, Chicago, IL). Parametric quantitative data are presented as means±standard deviation; nonparametric data are presented as medians (interquartile range 25–75). Qualitative variables are expressed as percentages. We compared patients with a first-visit SBP equal to or above the median of the population with the rest of the series, and patients with first-visit DBP equal to or above the median of the population with the rest of the series. The sample was also divided into quartiles according to the first-visit SBP and DBP. Finally, a separate analysis was performed for patients with SBP >140 mm Hg and/or DBP >90 mm Hg at baseline vs the rest of the series. Comparisons of quantitative parametric variables between subgroups were made using the Student t test or analysis of variance; nonparametric quantitative variables were compared using the Mann-Whitney test or the Wilcoxon test as required. Chi-square or Fisher exact test were used to analyze differences between qualitative variables. Four main end points were defined: major events (mortality, ACS, need for coronary revascularization, stroke, or admission for heart failure), coronary events (ACS or the need for coronary revascularization), cardiovascular mortality, and all-cause death. Two additional secondary end points were also considered: hospital admission for heart failure and stroke. Multivariate Cox proportional hazard ratio models were constructed, with the main and secondary end points as the dependent variables and first-visit SBP and DBP as the independent variables. The models initially involved all those variables that, in comparisons between the SBP or DBP subgroups, differed with a significance of P<.15; and additionally, some selected variables of known prognostic importance. These variables included age, sex, prior revascularization, prior ACS, tobacco use, functional class, basal rhythm, and pulse pressure and prescription of statins, β-blockers, angiotensin-converting enzyme (ACE) inhibitors/receptor blockers, platelet aggregation inhibitors, diuretics, nitrates, and calcium channel blockers. Ejection fraction was not included because it was not available in 185 patients (27%) at initial visit. However, due to its well-known prognostic significance, additional models are presented, restricted to patients with ejection fraction available at first visit. Those variables that showed no significant association with the dependent variable in backward stepwise regression were eliminated. The final models therefore involved only the SBP or DBP and the variables that were demonstrated to be independent predictors of events. The results are presented as hazard ratios (HRs) with 95% confidence intervals (CIs). All statistical analyses were performed under the guidance of the Research Support Unit at our center. All calculations were made using SPSS v. 12.0 software. Significance was set at P<.05.

Results

The study involved 690 patients. The median age was 70 (63–75) years; 65% of the patients were men; and 6% of patients were 50 years and younger, 28% were 51–65 years, 59% were 66 to 80 years, and 7% were older than 80. At first visit, median SBP and DBP were 130 mm Hg and 75 mm Hg (p25–75, 120–140/70–80 mm Hg, respectively). SBP/DBP was <140/90 mm Hg in 61% of patients, and ≤140/90 mm Hg in 77% of the series.

Patients with DBP <75 mm Hg (n=313) were older, received ACE inhibitors/receptor blockers less frequently, and there was a trend to more frequent prescription of calcium channel blockers than patients with DBP ≥75 mm Hg (n=377). Other basal features were similar in both groups (Table I) although patients in the higher quartile of DBP were more often active smokers.

Table I.   Basal Features of Patients According to the Diastolic Blood Pressure at First Visit
DBP DichotomousDBP Expressed in Quartiles
 <75 mm Hg≥75 mm HgP Value<70 mm Hg70–74 mm Hg75–80 mm Hg>80 mm HgP Value
n=313n=377n=119n=194n=280n=97
  1. Abbreviations: ACE, angiotensin-converting enzyme; ACS, acute coronary syndrome; ARB, angiotensin receptor blockers; bpm, beats per minute; CAD, coronary artery disease; DPB, diastolic blood pressure; LVEF, left ventricle ejection fraction; RHR, resting heart rate; SBP, systolic blood pressure.

Demographics
 Age, y69±967±10.00469±1069±967±1066±9.02
 Men65%66%.9365%66%63%72%.46
Risk factors
 Diabetes33%38%.1934%32%38%37%.59
 Hypercholesterolemia80%82%.6876%83%81%85%.40
 Tobacco use (former) 3% (35%) 6% (31%).11 5% (40%) 2% (32%) 5% (31%) 9% (32%)<.05
 Family history of CAD 7% 9%.4912% 5%10% 5%.08
Medical history
 Previous ACS80%82%.1785%81%82%81%.84
 Prior revascularization41%37%.3945%38%40%31%.23
 Surgical11%12%.5812%10%13%11%.91
 Percutaneous27%23%.1932%24%24%19%.15
 Both 3% 2%.89 1% 4% 3% 1%.34
 Angina + ischemia12%12%.85 9%14%11%12%.65
Cardiac parameters
 Functional class I80%81%.8080%80%80%81%.99
 RHR, bpm68±1169±12.5768±1268±1168±1271±12.07
 SBP, mm Hg126±15139±16<.001121±15129±13134±12152±16<.001
 DBP, mm Hg67±582±6<.00161±470±179±290±6<.001
 Pulse pressure, mm Hg60±1457±13.0260±1559±1355±1263±14<.001
 Basal rhythm  .75    .39
 Sinus rhythm93%94% 91%95%93%96% 
 Atrial fibrillation 7% 6%  9% 5% 7% 4% 
 LVEF0.57±0.140.57±0.13.620.55±0.150.58±0.140.57±0.130.59±0.13.33
 LVEF ≤0.30 8% 5%.2110% 6% 6% 2%.20
Medical treatment prescribed at first visit
 Platelet aggregation inhibitors90%93%.2091%92%90%91%.82
 Nitrates74%68%.1175%73%69%64%.29
 Statins62%66%.3161%63%66%66%.77
 β-Blockers66%66%.9463%67%65%68%.84
 ACE inhibitor/ARB54%64%.0159%52%62%69%.02
 Calcium channel blockers55%48%.0955%55%45%58%.06
 Diuretics35%37%.5642%30%36%40%.15
 Digoxin 5% 4%.37 8% 4% 4% 5%.17

Patients with SBP <130 mm Hg (n=244) presented previous ACS more frequently, were prescribed β-blockers more often, and ACE inhibitors/receptor blockers and calcium channel blockers less frequently. Patients with SBP ≥130 mm Hg (n=446) were older, presented with diabetes mellitus and stable angina with objective evidence of ischemia as reason for inclusion and higher ejection fraction and resting heart rate at first visit more frequently. Other basal features were similar in both groups, although patients in the higher quartile of SBP were more frequently active smokers and presented a trend towards higher rates of prior surgical revascularization and lower rates of prior percutaneous coronary interventions (Table II).

Table II.   Basal Features of Patients According to SBP at First Visit
SBP DichotomousSBP Expressed in Quartiles
 <130 mm Hg≥130 mm HgP Value≤120 mm Hg121–130 mm Hg131–140 mm Hg>140 mm HgP Value
n=244n=446n=204n=169n=159n=158
  1. Abbreviations: ACE, angiotensin-converting enzyme; ACS, acute coronary syndrome; ARB, angiotensin receptor blockers; bpm, beats per minute; CAD, coronary artery disease; DPB, diastolic blood pressure; LVEF, left ventricular ejection fraction; RHR, resting heart rate; SBP, systolic blood pressure.

Demographics
 Age, y67±1169±9.0167±1168±1069±868±9.04
 Men67%64%.4567%68%64%63%.78
Risk factors
 Diabetes28%40%.00128%31%40%45%.004
 Hypercholesterolemia84%80%.2376%81%85% .40
 Tobacco use (former) 5% (36%) 5% (31%).37 5% (40%) 2% (32%) 5% (31%) 9% (32%) 
 Family history of CAD10% 7%.1412% 5% 9% 7%.08
Medical history
 Previous ACS88%79%.00389%82%80%77%.02
 Prior revascularization41%38%.4941%37%47%30%.02
 Surgical 9%13%.07 8%10%14%16%.09
 Percutaneous28%23%.1229%25%30%13%.001
 Both 4% 2%.13 3% 2% 3% 1%.69
 Angina + ischemia 7%14%.007 7%14%12%17%.04
Cardiac parameters
 Functional class I78%82%.2477%85%81%79%.22
 RHR, bpm66±1170±12.00166±1068±1270±1270±11.004
 SBP, mm Hg116±8142±12<.001115±7129±2138±2156±10<.001
 DBP, mm Hg70±877±9<.00170±874±775±782±10<.001
 Pulse pressure, mm Hg46±965±11<.00145±855±763±773±12<.001
 Basal rhythm  .86    .96
 Sinus rhythm94%94% 94%94%93%93% 
 Atrial fibrillation 6% 6%  6% 6% 7% 7% 
 LVEF0.55±0.150.59±0.13.0060.55±0.150.56±0.140.59±0.120.59±0.13.06
 LVEF ≤0.3011% 4%.00110% 6% 4% 3%.05
Medical treatment prescribed at first visit
 Platelet aggregation inhibitors91%91%.8491%92%90%92%.82
 Nitrates75%68%.0377%72%67%65%.08
 Statins66%64%.5165%68%62%63%.75
 β-Blockers71%63%<.0571%65%59%67%.12
 ACE inhibitor/ARB55%62%<.0553%53%59%77%<.001
 Calcium channel blockers44%55%.00443%53%56%55%.05
 Diuretics36%36%.9134%28%39%44%.03
 Digoxin 5% 5%.90 4% 4% 4% 6%.68

Patients with SBP >140 mm Hg and/or DBP >90 mm Hg (n=159) presented more often with diabetes (45% vs 33%, P=.004) and with a higher rate of diuretic (43% vs 34%, P=.03) and ACE inhibitors/angiotensin receptor blocker use (77% vs 54%, P<.0005) than the rest of the series (n=531), but there was a lower frequency of previous ACS (76% vs 84%, P=.02) and coronary revascularization (30% vs 41%, P=.01). Resting heart rate was also higher in this subgroup (70±11 beats per minute vs 68±12 beats per minute, P=.03), as was, obviously, SBP (156±10 mm Hg vs 126±11 mm Hg, P<.0005), DBP (82±10 mm Hg vs 73±8, P<.0005), and pulse pressure (73±11 mm Hg vs 54±11 mm Hg, P<.0005). Other basal features were similar in both groups.

After a median follow-up of 25 (12–40) months, vital and major event status was known in 686 patients (99.4%), and these data came from a visit to the outpatient clinic in 552 patients (80%). SBP and DBP significantly increased in patients with first-visit values below the median and significantly decreased in patients with first-visit values above the median (116±8/66±5 mm Hg vs 122±13/71±8 mm Hg and 142±12/81±6 mm Hg vs 134±16/75±9 mm Hg, P<.001). Nineteen patients died (2.8%); 10 from cardiovascular causes (1.5%), 29 patients were hospitalized for heart failure, 17 patients experienced a stroke, 87 patients a coronary event (12.7%), and 130 patients (19%) had a major event (Table III). At median follow-up, patients with DBP <75 mm Hg presented a significantly lower probability of survival free of major events (75% vs 84%, P=.02), coronary events (83% vs 90%, P=.009), and cardiovascular death (97% vs 99%, P=.03) but not of admission for heart failure (93% vs 97%, P=.11), stroke (97% vs 97%, P=.77), or overall survival (96% vs 98%, P=.13) than those with DBP ≥75 mm Hg. Patients with SBP <130 mm Hg presented a lower probability of survival free of major events (74% vs 84%, P=.009) and coronary events (81% vs 89%, P=.005), but not of cardiovascular death (97% vs 99%, P=.30), admission for heart failure (94% vs 96%, P=.59), stroke (98% vs 97%, P=.43) or overall survival (96% vs 98%, P=.22) than those with higher SBP (Figure). At median follow-up, we did not find significant differences in survival free of any event for patients with SBP >140 mm Hg and/or DBP >90 mm Hg at entry vs the rest of the series (96% vs 98%, P=.13 for stroke; 98% vs 95%, P=.29 for heart failure admission; 90% vs 85%, P=.16 for coronary events; 85% vs 79%, P=.12 for all major events; 100% vs 98%, P=.29 for cardiovascular death; and 99% vs 97%, P=.16 for all-cause death).

Table III.   Events at Follow-Up
 DBP <75 mm HgDBP ≥75 mm HgP ValueSBP <130 mm HgSBP ≥130 mm HgP Value
n=311n=375n=241n=445
No. (%)No. (%)No. (%)No. (%)
  1. Abbreviations: ACS, acute coronary syndrome; CABG, coronary artery bypass grafting; CV, cardiovascular; DBP, diastolic blood pressure; PCI, percutaneous coronary intervention; SBP, systolic blood pressure.

Death12 (3.9)7 (1.9).139 (3.7)10 (2.2).22
CV death8 (2.6)2 (0.5).035 (2.1)5 (1.1).30
Coronary events51 (16.4)36 (9.6).00941 (17.0)46 (10.3).005
ACS45 (14.5)33 (8.8).0236 (14.9)42 (9.4).01
PCI20 (6.4)7 (1.9).00215 (6.2)12 (2.7).01
CABG4 (1.3)3 (0.8).522 (0.8)5 (1.1).77
Stroke7 (2.2)10 (2.7).754 (1.7)13 (2.9).42
Heart failure17 (5.5)12 (3.2).1111 (4.6)18 (4.0).59
All major events71 (22.8)59 (15.8).0257 (23.7)73 (16.4).009
Figure FIGURE.

 Probability of survival free of major events (A and C) and coronary events (B and D) according to first visit diastolic (A and B) and systolic (C and D) blood pressure. DBP indicates diastolic blood pressure; SBP, systolic blood pressure.

After adjusting for baseline clinical and treatment variables, lower values of DBP and SBP were independently associated with higher rates of major events, coronary events, admission for heart failure, and cardiovascular mortality, and results were consistent in analysis performed as continuous or dichotomous variables, expressed as quartiles, or including ejection fraction in the model (Table IV, Table V, and Table VI). Stroke rate was only significantly associated with a higher SBP, expressed as a continuous variable. Association of BP with all-cause death was only statistically significant for SBP expressed as quartiles (adjusted or not by ejection fraction) and for dichotomous SBP and continuous DBP and SBP in those models adjusted by ejection fraction. β-Blockers, ACE inhibitors/angiotensin receptor blockers, and calcium channel blockers, which were prescribed differently among SBP subgroups, were not independently associated with any of the outcomes after multivariate analysis. Diuretics prescription was independently associated with higher rates of admission for heart failure (HR, 9.78; 95% CI, 2.75–34.86; P<.0005) in final multivariate models and also after adjusting for ejection fraction (HR, 8.04; 95% CI, 2.31–27.90; P=.001).

Table IV.   Associations of Blood Pressure at First Visit With Coronary Events and Major Events in Final Multivariable Models
 Event Rate, n/Na (%)Models Not Adjusted by LVEFModels Adjusted by LVEF
Hazard Ratiob (95% CI)P ValuebHazard Ratiob (95% CI)P Valueb
  1. Abbreviations: CI, confidence interval; DBP, diastolic blood pressure; LVEF, left ventricular ejection fraction; SBP, systolic blood pressure. an/N: Number of events/number of patients in each subgroup of first visit blood pressure with follow-up data (for blood pressure as a continuous variable, the whole series with follow-up data). bCox proportional hazard method.

Coronary events
 DBP (continuous)87/686 (12.7)0.96 (0.94–0.99).0020.96 (0.93–0.98).002
DBP (dichotomous)
 <75 mm Hg51/311 (16.4)1.78 (1.15–2.75).0092.08 (1.28–3.39).003
 ≥75 mm Hg36/375 (9.6)1 (reference) 1 (reference) 
DBP (quartiles)
 <70 mm Hg23/119 (19.3)1 (reference) 1 (reference) 
 70–74 mm Hg28/192 (14.6)0.67 (0.38–1.17).160.74 (0.41–1.36).33
 75–80 mm Hg28/278 (10.1)0.46 (0.27–0.81).0070.44 (0.23–0.81).009
 >80 mm Hg8/97 (8.2)0.38 (0.17–0.86).020.30 (0.11–0.81).02
 SBP (continuous)87/686 (12.7)0.96 (0.94–0.99).0020.96 (0.93–0.98).001
SBP (dichotomous)
 <130 mm Hg41/241 (17.0)1.84 (1.20–2.83).0051.80 (1.12–2.90).02
 ≥130 mm Hg46/445 (10.3)1 (reference) 1 (reference) 
SBP (quartiles)
 ≤120 mm Hg37/203 (18.2)1 (reference) 1 (reference) 
 121–130 mm Hg23/167 (13.8)0.52 (0.29–0.92).030.56 (0.31–1.04).07
 131–140 mm Hg11/158 (7.0)0.21 (0.09–0.47)<.0010.15 (0.06–0.38)<.001
 >140 mm Hg16/158 (10.1)0.18 (0.07–0.46)<.0010.12 (0.04–0.36)<.001
Major events
 DBP (continuous)130/686 (19.0)0.97 (0.95–0.99).0020.97 (0.95–0.99).002
DBP (dichotomous)
 <75 mm Hg71/311 (22.8)1.52 (1.07–2.16).021.66 (1.12–2.47)0.01
 ≥75 mm Hg59/375 (15.8)1 (reference) 1 (reference) 
DBP (quartiles)
 <70 mm Hg30/119 (25.2)1 (reference) 1 (reference) 
 70–74 mm Hg41/192 (21.4)0.77 (0.47–1.24).280.80 (0.47–1.35).39
 75–80 mm Hg47/278 (16.9)0.60 (0.38–0.96).030.58 (0.35–0.98).04
 >80 mm Hg12/97 (12.4)0.45 (0.23–0.88).020.39 (0.18–0.85).02
 SBP (continuous)130/686 (19.0)0.97 (0.95–0.99).0020.97 (0.95–0.99).002
SBP (dichotomous)
 <130 mm Hg57/241 (23.7)1.68 (1.18–2.39).0041.83 (1.23–2.72).003
 ≥130 mm Hg73/445 (16.4)1 (reference) 1 (reference) 
SBP (quartiles)
 ≤120 mm Hg51/203 (25.1)1 (reference) 1 (reference) 
 121–130 mm Hg29/167 (17.4)0.48 (0.29–0.80).0040.52 (0.30–0.89).02
 131–140 mm Hg25/158 (15.8)0.35 (0.19–0.65).0010.29 (0.14–0.57)<.001
 >140 mm Hg25/158 (15.8)0.23 (0.10–0.49)<.0010.21 (0.09–0.50)<.001
Table V.   Associations of Blood Pressure at First Visit With Cardiovascular Mortality and All-Cause Death in Final Multivariable Models
 Event Rate, n/Na (%)Models Not Adjusted by LVEFModels Adjusted by LVEF
Hazard Ratiob (95% CI)P ValuebHazard Ratiob (95% CI)P Valueb
  1. Abbreviations: CI, confidence interval; DBP, diastolic blood pressure; LVEF, left ventricular ejection fraction; SBP, systolic blood pressure. an/N: Number of events/number of patients in each subgroup of first visit blood pressure with follow-up data (for blood pressure as a continuous variable, the whole series with follow-up data). bCox proportional hazard method.

Cardiovascular mortality
 DBP (continuous)10/686 (1.5)0.90 (0.83–0.97).0060.92 (0.86–0.98).01
DBP (dichotomous)
 <75 mm Hg8/311 (2.6)5.96 (1.09–32.66).045.73 (1.05–31.22).04
 ≥75 mm Hg2/375 (0.5)1 (reference) 1 (reference) 
DBP (quartiles)
 <70 mm Hg6/119 (5.0)1 (reference) 1 (reference) 
 70–74 mm Hg2/192 (1.0)0.17 (0.03–0.90).040.20 (0.04–1.04).06
 75–80 mm Hg1/278 (0.4)0.05 (0.01–0.50).010.06 (0.01–0.62).02
 >80 mm Hg1/97 (1.0)0.11 (0.01–1.22).070.13 (0.01–1.19).07
 SBP (continuous)10/686 (1.5)0.90 (0.83–0.97).0060.92 (0.86–0.98).01
SBP (dichotomous)
 <130 mm Hg5/241 (2.1)9.26 (1.33–64.32).027.78 (1.09–55.38).049
 ≥130 mm Hg5/445 (1.1)1 (reference) 1 (reference) 
SBP (quartiles)
 ≤120 mm Hg5/203 (2.5)1 (reference) 1 (reference) 
 121–130 mm Hg1/167 (0.6)0.11 (0.01–1.08).060.11 (0.01–1.10).067
 131–140 mm Hg3/158 (1.9)0.10 (0.01–0.79).030.16 (0.02–1.33).09
 >140 mm Hg1/158 (0.6)0.01 (0.00–0.18).0030.01 (0.00–0.31).008
All cause death
 DBP (continuous)19/686 (2.8)0.95 (0.91–1.00).060.93 (0.88–0.99).02
DBP (dichotomous)
 <75 mm Hg12/311 (3.9)2.12 (0.83–5.39).122.65 (0.90–7.76).08
 ≥75 mm Hg7/375 (1.9)1 (reference) 1 (reference) 
DBP (quartiles)
 <70 mm Hg6/119 (5.0)1 (reference) 1 (reference) 
 70–74 mm Hg6/192 (3.1)0.64 (0.20–2.04).450.43 (0.12–1.52).19
75–80 mm Hg6/278 (2.2)0.44 (0.14–1.38).160.28 (0.08–1.01).05
 >80 mm Hg1/97 (1.0)0.18 (0.02–1.54).120.17 (0.02–1.46).11
 SBP (continuous)19/686 (17.66)0.99 (0.96–1.01).310.94 (0.88–0.99).02
SBP (dichotomous)
 <130 mm Hg9/241 (3.7)1.81 (0.73–4.46).204.59 (1.10–19.13).04
 ≥130 mm Hg10/445 (2.2)1 (reference) 1 (reference) 
SBP (quartiles)
 ≤120 mm Hg9/203 (4.4)1 (reference) 1 (reference) 
 121–130 mm Hg2/167 (1.2)0.12 (0.02–0.61).010.07 (0.01–0.62).02
 131–140 mm Hg6/158 (3.8)0.30 (0.08–1.16).080.31 (0.07–1.39).13
 >140 mm Hg2/158 (1.3)0.03 (0.00–0.29).0020.02 (0.00–0.26).003
Table VI.   Associations of Blood Pressure at First Visit With Admission for Heart Failure and Stroke in Final Multivariable Models
 Event Rate, n/Na (%)Models Not Adjusted by LVEFModels Adjusted by LVEF
Hazard Ratiob (95% CI)P ValuebHazard Ratiob (95% CI)P Valueb
  1. Abbreviations: CI, confidence interval; DBP, diastolic blood pressure; LVEF, left ventricular ejection fraction; SBP, systolic blood pressure. an/N: Number of events/number of patients in each subgroup of first visit blood pressure with follow-up data (for blood pressure as a continuous variable, the whole series with follow-up data). bCox proportional hazard method.

Admission for heart failure
 DBP (continuous)29/686 (4.2)0.93 (0.89–0.97).0010.93 (0.89–0.98).003
DBP (dichotomous)
 <75 mm Hg17/311 (5.5)3.02 (1.26–7.25).012.92 (1.12–7.61).03
 ≥75 mm Hg12/375 (3.2)1 (reference) 1 (reference) 
DBP (quartiles)
 <70 mm Hg9/119 (7.6)1 (reference) 1 (reference) 
 70–74 mm Hg8/192 (4.2)0.45 (0.15–1.34).150.51 (0.17–1.51).22
 75–80 mm Hg11/278 (4.0)0.26 (0.09–0.71).0090.28 (0.09–0.82).02
 >80 mm Hg1/97 (1.0)0.13 (0.02–1.07).060.15 (0.02–1.25).08
 SBP (continuous)29/686 (4.2)0.93 (0.89–0.97).0010.93 (0.88–0.97).002
SBP (dichotomous)
 <130 mm Hg11/241 (4.6)2.20 (0.90–5.36).082.36 (0.89–6.26).09
 ≥130 mm Hg18/445 (4.0)1 (reference) 1 (reference) 
SBP (quartiles)
 ≤120 mm Hg9/203 (4.4)1 (reference) 1 (reference) 
 121–130 mm Hg5/167 (3.0)0.35 (0.10–1.30).120.54 (0.14–2.08).37
 131–140 mm Hg10/158 (6.3)0.30 (0.07–1.24).100.31 (0.07–1.28).10
 >140 mm Hg5/158 (3.2)0.07 (0.01–0.51).0080.11 (0.02–0.75).02
Stroke
 DBP (continuous)17/686 (2.5)1.03 (0.98–1.09).271.04 (0.99–1.08).23
DBP (dichotomous)
 <75 mm Hg7/311 (2.3)0.66 (0.23–1.86).430.46 (0.13–1.58).22
 ≥75 mm Hg10/375 (2.7)1 (reference) 1 (reference) 
DBP (quartiles)
 <70 mm Hg2/119 (1.7)1 (reference) 1 (reference) 
 70–74 mm Hg5/192 (2.6)1.52 (0.29–8.01).621.62 (0.08–1.01).66
 75–80 mm Hg5/278 (1.8)0.98 (0.18–5.43).981.73 (0.02–1.46).61
 >80 mm Hg5/97 (5.2)3.84 (0.74–20.07).117.93 (0.84–75.23).07
 SBP (continuous)17/686 (2.5)1.03 (1.00–1.06).0481.03 (1.00–1.07).043
SBP (dichotomous)
 <130 mm Hg4/241 (1.7)0.99 (0.21–4.61).990.93 (0.17–5.13).93
 ≥130 mm Hg13/445 (2.9)1 (reference) 1 (reference) 
SBP (quartiles)
 ≤120 mm Hg4/203 (2.0)1 (reference) 1 (reference) 
 121–130 mm Hg4/167 (2.4)0.94 (0.20–4.51).940.94 (0.14–6.24).95
 131–140 mm Hg2/158 (1.3)0.40 (0.05–3.04).380.49 (0.06–4.42).53
 >140 mm Hg7/158 (4.4)1.03 (0.12–8.70).981.24 (0.11–13.91).86

Discussion

The main finding of the present study is that first-visit low DBP and SBP were significantly and independently associated with an adverse prognosis in this contemporary series of hypertensive patients with sCAD of daily clinical practice. Other studies have shown a J-shaped relationship of mean follow-up on-treatment BP with events, but some of them are post hoc analyses of randomized clinical trials,1–5 where patients are selected and may not be fully representative of daily clinical practice populations. Other observational studies have been performed in the absence of contemporary and aggressive management for achieving SBP and DBP goals and included patients with and without cardiovascular disease.6–9 The present study adds valuable information about the prognostic value of first visit BP in the specific setting of hypertensive patients with sCAD of contemporary daily clinical practice.

Several points are worth noting. First of all, we did not find a true J-shaped relationship with BP in our study, but we did find a significant increase of events in patients with first-visit low DBP and SBP, without higher rates in those with higher BP. Two factors could explain this observation: first, our population included relatively few patients with first-visit SBP/DBP >140/90 mm Hg (only 23% of the series); and second, we found a significant decrease in BP values at last visit in patients with SBP/DBP above the median. Only stroke rate was significantly associated with higher SBP at entry, assessed as a continuous variable, either adjusted or not by ejection fraction, as other investigators have shown.2,4,15 In addition, although low DBP and SBP were independently associated with major events, the main impact appears to be related to coronary events and cardiovascular mortality, as previously reported by others,2,3,15 and also with admissions for heart failure.4 Third, our results were consistent after adjusting by left ventricular systolic function, as others have found,16 and this makes it less likely that the adverse prognostic impact of first visit low blood pressure would be related to impaired cardiac function, as suggested by other studies.17 Fourth, we noticed an increase in events not just for low DBP but also for SBP, and even after controlling for pulse pressure. So, a high pulse pressure as an epiphenomenon of increased arterial stiffness does not appear to be responsible for the association of low BP with a worse prognosis in our series, as argued by other investigators.18 Fifth, 61% of our patients presented first-visit SBP/DBP <140/90 mm Hg, more than double of that found in previous general population surveys,19 perhaps due to more aggressive management of BP in a contemporary outpatient cardiology clinic. The possibility of effectively reducing BP in many patients makes it crucial to be sure of “how far should we go” in this strategy. Sixth, we found some significant differences in some medications use among different SBP/DBP subgroups, but this fact does not appear to be the cause of the observed association of BP with outcomes: β-blockers, calcium channels blockers, and ACE inhibitors/angiotensin receptor blockers were not significantly associated with outcomes in multivariate models, and diuretics, which were more frequently prescribed in higher quartiles of SBP, were associated with admissions for heart failure, although this event was predicted by lower SBP. A possible explanation could be that diuretic use was probably more common in patients with baseline compensated heart failure than in the rest of the series. And last, although our findings could support the hypothesis that diastolic hypotension could lead to coronary hypoperfusion in patients with compromised coronary flow reserve such as those with sCAD,3 our work can not be taken as evidence of a causative role of low BP in events, but, as an observational one, as an hypothesis-generating study, that must be tested with appropriately designed randomized clinical trials.

Several practice guidelines have recommended goals of SBP/DBP of <130/80 mm Hg for high-risk patients, as those with diabetes or established cardiovascular disease.20–23 These opinions and recommendations have recently been reappraised by a task force document of the European Society of Hypertension on the basis of a critical analysis of available evidence from trials.24 It has been concluded that there is no solid evidence in favor of the recommendation that high-risk patients should require a SBP target <130 mm Hg and should receive antihypertensive agents even when their BP is in the high-normal range. The results of recent trials also have questioned the “lower the better” hypothesis for BP. In the Action to Control Cardiovascular Risk in Type 2 Diabetes (ACCORD) trial of patients with diabetes,25 intensive BP lowering (to 120 mm Hg SBP) was not associated with a reduction in the risk of cardiovascular outcomes when compared with the standard therapy group (to 140 mm Hg SBP). Tight control of SBP among patients with diabetes and sCAD was not associated with improved cardiovascular outcomes compared with usual control in an observational subgroup analysis of 6400 of the 22.576 participants in the International Verapamil SR-Trandolapril Study (INVEST).26 Even in some study findings that are traditionally cited as evidence against the J-Curve, such as the Hypertension Optimal Treatment (HOT) study,27 the analysis performed by Cruickshank28 in the high-risk patient group with coronary ischemia found a 22% increase in the risk of myocardial infarction when the DBP was <80 mm Hg compared with <85 mm Hg. Our results are concordant with these findings.

Limitations

Our work has some limitations. Our relative small sample size could have precluded the observation of a higher rate of events in patients with higher BP. On the other hand, we have no data on follow-up of treatment BP, which has been the variable studied in some previous studies.1–5 However, the main point of our study is the prognostic impact of a simple and readily available factor, first-visit BP, even when it appears that the effect of treatment in follow-up was appropriate (lowering SBP and DBP in patients with values above the median and the opposite in patients with low values). Finally, the adjudication for out-of-hospital death is based on communication with family and physicians, and the accuracy and robustness of cause of death in these case scenarios could be limited. So, our findings regarding cardiovascular death should be taken with caution.

Conclusions

In this cohort of hypertensive patients with sCAD in a contemporary daily clinical practice setting, an SBP/DBP <130/75 mm Hg at first visit was significantly and independently associated with an adverse prognosis. Although further studies should confirm our findings, this simple and readily available variable could be taken into account when stratifying risk in this population.

Disclosures:  This work received a grant of the Andalusian Society of Cardiology, Seville, Spain.

Ancillary