Abstract
Despite broad treatment recommendations, there are no published reports comparing efficacy of different antihypertensive agents in patients with isolated systolic hypertension or isolated diastolic hypertension. This study was a secondary analysis of the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial. We compared the use of chlorthalidone, amlodipine, or lisinopril on the primary outcome of combined coronary heart disease, stroke, or all-cause mortality in patients with isolated systolic hypertension or isolated diastolic hypertension.
Introduction
Current United States guidelines recommend pharmacologic treatment with either a thiazide diuretic, an angiotensin converting enzyme (ACE) inhibitor/angiotensin receptor blocker (ARB), or a dihydropyridine calcium channel blocker as first-line options for patients with a systolic blood pressure (SBP) ≥130mmHg or a diastolic blood pressure (DBP) ≥80mmHg.1 Although these broad guidelines apply to patients with isolated systolic hypertension (ISH), and patients with isolated diastolic hypertension (IDH), there is no consensus on the optimal choice of antihypertensive medication in these distinct populations. There may be differences in efficacy based on differences in disease pathophysiology and phenotypes. ISH is associated with increased stroke volume and aortic stiffness,2 occurs more often in older individuals with advanced cardiovascular disease (CVD), and those with wide pulse pressures. In contrast, IDH is associated with greater peripheral vascular resistance,3 presents more often in younger patients with fewer cardiovascular disease risk factors, and in patients with a narrow pulse pressure.4 In the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT), chlorthalidone notably reduced SBP to a greater extent than amlodipine or lisinopril, and amlodipine reduced DBP greater than chlorthalidone or lisinopril,5 which may also suggest varying benefit based on hypertension type. Therefore, we set out to determine whether certain antihypertensive agents had differential effects on clinical outcomes in participants with ISH or IDH.
Methods
Our study was a secondary analysis of the ALLHAT trial, a randomized controlled trial evaluating chlorthalidone, amlodipine, or lisinopril in the treatment of high-risk participants with hypertension, the methods of which have been described previously.5 We defined ISH as a SBP ≥140mmHg and a DBP <90mmHg, and we defined IDH as a SBP <140mmHg and a DBP ≥90mmHg.6 The primary outcome was the composite outcome of all-cause mortality, combined coronary heart disease (CHD), and stroke. Separate secondary outcomes were all-cause mortality, combined CHD, stroke, heart failure (HF), and achievement of normotension (SBP<140mmHg & DBP <90mmHg) at 6 years. We excluded patients who had a primary outcome prior to the onset of the study (n=380). All covariates except for age were operationalized as categorical variables in order to account for missing data. Similar to the original ALLHAT trial, data were graphed using Kaplan-Meier Curves and truncated at 7 years, however analyses including log-rank tests and cox regression models used complete follow up times.5 We devised cox regression models adjusting for age, sex, race, ethnicity, BMI, diabetes, smoking status, educational attainment, aspirin use, alcohol use, total cholesterol, serum potassium, left ventricular hypertrophy on ECG, history of CHD, history of myocardial infarction or stroke, and history of CABG, coronary angioplasty, or coronary revascularization. Risk-adjusted logistic regression was used to estimate the outcome of normotension. Given prior studies which showed potentially greater adverse outcomes with an ACE inhibitor or an ARB in Black compared with White participants,7–9 we conducted subgroup analyses by race on the primary outcome in our cohorts of ISH and IDH. In order to account for multiple comparisons, we used a 2-sided p-value of <0.0178 to denote significance throughout the study, similar to the original ALLHAT trial.5
Results
After exclusion criteria, there were 12,845 participants with ISH and 1,259 participants with IDH during the baseline visit. Participants with ISH had an average age (SD) of 69 (8) years, 51% were male, 62% were White, and 33% were Black; 5857 (46%) participants were treated with chlorthalidone, 3496 (27%) participants were treated with amlodipine, and 3492 (27%) participants were treated with lisinopril. Of the participants with IDH during the baseline visit, the average age (SD) was 63 (7) years, 61% were male, 55% were White, and 40% were Black; 583 (46%) participants used chlorthalidone, 334 (27%) participants used amlodipine, and 342 (28%) participants used lisinopril. When comparing baseline characteristics by type of medication used, there were no significant differences among almost all characteristics in participants with ISH or IDH, including age, sex, race, and ethnicity (Table 1). Participants with IDH using amlodipine were more likely to have total cholesterol <200 mg/dl and less likely to have total cholesterol ≥200 mg/dl than participants using chlorthalidone or lisinopril (p=0.004). In terms of total events, there was significantly greater risk of HF with amlodipine compared to chlorthalidone among patients with ISH (9.5% vs 7.1%, p<0.001).
Table 1.
Baseline characteristics and outcomes by choice of antihypertensive agent, stratified by isolated systolic hypertension and isolated diastolic hypertension. CABG, coronary artery bypass graft; CHD, coronary heart disease; ECG, electrocardiogram; LVH, left ventricular hypertrophy; MI, myocardial infarction
| Isolated Systolic Hypertension | Isolated Diastolic Hypertension | |||||||
|---|---|---|---|---|---|---|---|---|
|
| ||||||||
| Chlorthalidone (n=5857) | Amlodipine (n=3496) | Lisinopril (n=3492) | P-Value | Chlorthalidone (n=583) | Amlodipine (n=334) | Lisinopril (n=342) | P-Value | |
|
| ||||||||
| Baseline Characteristics | ||||||||
|
| ||||||||
| Mean age, years (SD) | 69 (8) | 69 (8) | 69 (8) | 0.40 | 63 (7) | 63 (7) | 63 (7) | 0.19 |
|
| ||||||||
| Female Sex (%) | 2928 (50) | 1694 (49) | 1714 (49) | 0.34 | 234 (40) | 131 (39) | 124 (36) | 0.50 |
|
| ||||||||
| Race (%) | 0.36 | 0.27 | ||||||
| White | 3650 (62) | 2198 (63) | 2154 (62) | 314 (54) | 203 (61) | 173 (51) | ||
| Black | 1926 (33) | 1142 (33) | 1160 (33) | 241 (41) | 117 (35) | 147 (43) | ||
| American | 5 (0) | 10 (0) | 5 (0) | 2 (0) | 0 (0) | 1 (0) | ||
| Indian/Alaskan | 61 (1) | 30 (10) | 37 (1) | 10 (2) | 5 (2) | 10 (3) | ||
| Asian/Pacific Islander | 215 (4) | 116 (3) | 136 (4) | 16 (3) | 9 (3) | 11 (3) | ||
| Other | ||||||||
|
| ||||||||
| Hispanic (%) | 0.40 | 0.10 | ||||||
| Yes | 853 (15) | 518 (15) | 559 (16) | 129 (22) | 75 (23) | 56 (16) | ||
| No | 4973 (85) | 2958 (85) | 2912 (83) | 451 (77) | 259 (78) | 283 (83) | ||
| Unknown | 31 (1) | 20 (1) | 21 (1) | 3 (1) | 0 (0) | 3 (1) | ||
|
| ||||||||
| Obesity Class (%) | 0.49 | 0.84 | ||||||
| Not Obese | 3502 (60) | 2066 (59) | 2093 (60) | 321 (55) | 202 (61) | 205 (60) | ||
| Class I | 1433 (25) | 850 (24) | 843 (24) | 157 (27) | 80 (24) | 83 (24) | ||
| Class II | 542 (9) | 346 (10) | 326 (9) | 65 (11) | 30 (9) | 33 (10) | ||
| Class III | 307 (5) | 173 (5) | 169 (5) | 33 (6) | 17 (5) | 18 (5) | ||
| Unknown | 73 (1) | 61 (2) | 61 (2) | 7 (1) | 5 (2) | 3 (1) | ||
|
| ||||||||
| Type 2 Diabetes (%) | 2359 (40) | 1401 (40) | 1410 (40) | 0.97 | 178 (31) | 89 (27) | 77 (23) | 0.03 |
|
| ||||||||
| Current Smoker (%) | 1165 (20) | 689 (20) | 653 (19) | 0.35 | 144 (25) | 80 (24) | 92 (27) | 0.65 |
|
| ||||||||
| Years of Education (%) | 0.83 | 0.57 | ||||||
| ≤6 years | 702 (12) | 432 (12) | 447 (13) | 63 (11) | 52 (16) | 42 (12) | ||
| >6 to 12 years | 3284 (56) | 1959 (56) | 1922 (55) | 298 (51) | 165 (49) | 176 (52) | ||
| >12 years | 1474 (25) | 871 (25) | 869 (25) | 182 (31) | 98 (29) | 103 (30) | ||
| Unknown | 397 (7) | 234 (7) | 254 (7) | 40 (7) | 19 (6) | 21 (6) | ||
|
| ||||||||
| Aspirin Use (%) | 2269 (39) | 1370 (39) | 1326 (38) | 0.57 | 185 (32) | 103 (31) | 101 (30) | 0.78 |
|
| ||||||||
| Total Cholesterol (%) | 0.29 | 0.004 | ||||||
| <200 mg/dl | 1966 (34) | 1196 (34) | 1220 (35) | 197 (34) | 145 (43) | 101 (30) | ||
| ≥200 mg/dl | 3575 (61) | 2132 (61) | 2074 (59) | 362 (62) | 176 (53) | 227 (66) | ||
| Unknown | 316 (5) | 168 (5) | 198 (6) | 24 (4) | 13 (4) | 14 (4) | ||
|
| ||||||||
| Potassium Level (%) | 0.46 | 0.75 | ||||||
| <3.5 mmol/l | 167 (3) | 99 (3) | 82 (2) | 17 (3) | 13 (4) | 8 (2) | ||
| ≥3.5 to <5 mmol/l | 4801 (82) | 2885 (83) | 2872 (82) | 498 (85) | 281 (84) | 285 (83) | ||
| ≥5 mmol/l | 571 (10) | 347 (10) | 341 (10) | 44 (8) | 27 (8) | 35 (10) | ||
| Unknown | 318 (5) | 165 (5) | 197 (6) | 24 (4) | 13 (4) | 14 (4) | ||
|
| ||||||||
| LVH on ECG (%) | 926 (16) | 598 (17) | 585 (17) | 0.22 | 96 (17) | 56 (17) | 44 (13) | 0.27 |
|
| ||||||||
| History of CHD (%) | 0.05 | 0.84 | ||||||
| Yes | 1650 (28) | 905 (26) | 907 (26) | 121 (21) | 79 (23) | 71 (21) | ||
| No | 4169 (71) | 2570 (74) | 2556 (73) | 456 (78) | 251 (75) | 268 (78) | ||
| Unknown | 38 (1) | 21 (1) | 29 (1) | 6 (1) | 4 (1) | 3 (1) | ||
|
| ||||||||
| History of MI/stroke (%) | 1460 (25) | 835 (24) | 789 (23) | 0.04 | 110 (19) | 72 (22) | 67 (20) | 0.61 |
|
| ||||||||
| History of CABG, coronary angioplasty, or revascularization (%) | 905 (16) | 481 (14) | 542 (16) | 0.05 | 58 (10) | 40 (12) | 41 (12) | 0.52 |
|
| ||||||||
| Outcomes | ||||||||
|
| ||||||||
| Primary Outcome (%) | 1810 (31) | 1060 (30) | 1092 (31) | 0.68 | 129 (22) | 79 (24) | 85 (25) | 0.63 |
| 6-Year Rate per 100 Persons | 42.9 | 41.8 | 43.7 | 30.2 | 32.2 | 33.3 | ||
|
| ||||||||
| All-Cause Mortality (%) | 1001 (17) | 553 (16) | 564 (16) | 0.23 | 64 (11) | 40 (12) | 43 (13) | 0.75 |
| 6-Year Rate per 100 Persons | 20.8 | 19.1 | 19.6 | 13.4 | 14.6 | 15.1 | ||
|
| ||||||||
| Combined CHD (%) | 1053 (18) | 631 (18) | 662 (19) | 0.46 | 73 (13) | 48 (14) | 52 (15) | 0.48 |
| 6-Year Rate per 100 Persons | 24.5 | 24.4 | 26.0 | 16.8 | 19.3 | 20.2 | ||
|
| ||||||||
| Stroke (%) | 316 (5) | 177 (5) | 191 (5) | 0.71 | 23 (4) | 15 (4) | 11 (3) | 0.69 |
| 6-Year Rate per 100 Persons | 6.7 | 6.4 | 7.0 | 5.0 | 5.8 | 4.0 | ||
|
| ||||||||
| Heart Failure (%) | 416 (7) | 331 (9) | 276 (8) | <0.001 | 15 (3) | 18 (5) | 15 (4) | 0.08 |
| 6-Year Rate per 100 Persons | 9.1 | 12.2 | 10.2 | 3.3 | 7.0 | 5.6 | ||
Mean follow-up time was 4.7 years. In risk-adjusted cox regression analyses, there were no significant differences between choice of antihypertensive agent and hazards of the primary outcome in participants with ISH (adjusted hazards ratio [aHR]=0.98, 95% CI 0.91–1.06 for amlodipine vs chlorthalidone; and aHR=1.02, 95% CI 0.95–1.10 for lisinopril vs chlorthalidone). There were also no differences in hazards of the primary outcome across antihypertensive agents in participants with IDH (aHR=1.01, 95% CI 0.76–1.34 for amlodipine vs chlorthalidone; and aHR=1.10, 95% CI 0.83–1.46 for lisinopril vs chlorthalidone) (Figure 1). There were no significant differences in separate risk-adjusted outcomes of all-cause mortality, combined CHD, and stroke by antihypertensive agent among participants with ISH or IDH. Among participants with ISH, the use of amlodipine resulted in significantly greater hazards of HF compared with chlorthalidone (aHR=1.39, 95% CI 1.20–1.61, p<0.001), but there was no significant difference in hazards of HF among participants using lisinopril compared with chlorthalidone (aHR=1.13, 95% CI 0.97–1.32). Among participants with IDH, use of amlodipine or lisinopril did not result in significantly greater hazards of HF compared with chlorthalidone (aHR=2.17, 95% CI 1.07–4.41, p=0.03 for amlodipine and aHR=2.25, 95% CI 1.06–4.77, p=0.03 for lisinopril). Achievement of normotension at six years was also not significantly different by medication in both participants with ISH (OR=1.01, 95% CI 0.89–1.15 for amlodipine vs chlorthalidone; and OR=0.97, 95% CI 0.85–1.10 for lisinopril vs chlorthalidone) or IDH (OR=0.92, 95% CI 0.59–1.42 for amlodipine vs chlorthalidone; and OR=0.77, 95% CI 0.49–1.21 for lisinopril vs chlorthalidone). In subgroup analyses by race, White and Black participants with ISH had significantly greater hazards of HF with amlodipine compared with chlorthalidone (HR=1.29, 95% CI 1.07–1.54, p=0.01 for White participants, and HR=1.66, 95% CI 1.30–2.13, p<0.001 for Black participants). There were no significant differences on risk of HF between amlodipine and chlorthalidone stratified by race among participants with IDH (HR=3.31, 95% CI 1.19–9.20, p=0.02 for White participants and HR=1.85, 95% CI 0.56–6.06, p=0.31 for Black participants).
Figure 1.
Clinical events over time in participants with Isolated Systolic Hypertension (A) and Isolated Diastolic Hypertension (B) across different anti-hypertensive treatments. Figure 1 demonstrates unadjusted Kaplan-Meier survival curves for the primary composite outcome of all-cause mortality, combined coronary heart disease, or stroke, by choice of antihypertensive agent in participants with A. Isolated Systolic Hypertension and B. Isolated Diastolic Hypertension
Discussion
In this secondary analysis of a large trial database of participants with hypertension, we found that the use of chlorthalidone, amlodipine, or lisinopril resulted in nearly equivalent outcomes for the treatment of participants with ISH or IDH. Our expanded primary outcome was not significantly different between the three groups despite a large sample size. The overall results are comparable to the results of the total population with hypertension from the initial ALLHAT study, which also showed no significant differences between the three medications on most outcomes of interest. We did not observe differences in outcomes between antihypertensive agents when comparing White versus Black participants with ISH or IDH.
There was significantly greater HF risk with amlodipine compared with chlorthalidone in participants with ISH but not in participants with IDH. Despite the statistical significance, a 2% greater absolute risk represents a modest clinical difference and should be interpreted within that context. In the original ALLHAT trial, the risk of HF was also greater with amlodipine than chlorthalidone.5 Given these results, should chlorthalidone be the preferred antihypertensive agent? The ensuing Systolic Blood Pressure Intervention Trial (SPRINT) used chlorthalidone as the first line agent overall, and amlodipine as the first line calcium channel blocker.10 A post-hoc analysis of SPRINT showed greater monotherapy failure with chlorthalidone than amlodipine, though this was attributed to individual patient-level decision-making by the investigative teams.11 Another recent real-world retrospective cohort study of >700,000 participants showed that chlorthalidone resulted in greater hypokalemia and acute renal failure compared with hydrochlorothiazide, without significant cardiovascular benefit.12 Thus, the modest heart failure benefit seen here with chlorthalidone must be balanced against the potential for an increased risk of adverse renal and electrolyte events compared with other commonly used agents.
The clinically similar results observed across outcomes in ALLHAT between antihypertensive agents promotes that blood pressure lowering, irrespective of agent, is the key intervention in cardiovascular risk reduction in ISH or IDH. Any differences in efficacy and safety between antihypertensive agents is likely modest; however, care should be taken to monitor adverse effects of antihypertensive drug therapy.
Limitations of this study are that approximately 40% of participants were using other antihypertensive agents in addition to the primary drug by year 5, which may have confounded the results. Furthermore, the lower sample size among participants with IDH may have caused analyses to be underpowered to detect significant differences. Since this was not a pre-specified analysis, these findings should be considered exploratory. Nevertheless, these results lend credence to the flexible treatment recommendations for ISH and IDH in current national hypertension guidelines.
Disclosures:
Wang SY: None
Hanna JM: None
Gongal P: None
Onuma O: None
Nanna MG: Dr. Nanna reports funding from the American College of Cardiology Foundation supported by the George F. and Ann Harris Bellows Foundation and from the National Institute on Aging/National Institutes of Health from R03AG074067 (GEMSSTAR award).
References
- 1.Whelton PK, Carey RM, Aronow WS, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. May 15 2018;71(19):e127–e248. doi: 10.1016/j.jacc.2017.11.006 [DOI] [PubMed] [Google Scholar]
- 2.McEniery CM, Yasmin, Wallace S, et al. Increased stroke volume and aortic stiffness contribute to isolated systolic hypertension in young adults. Hypertension. Jul 2005;46(1):221–6. doi: 10.1161/01.HYP.0000165310.84801.e0 [DOI] [PubMed] [Google Scholar]
- 3.Huang M, Long L, Tan L, et al. Isolated Diastolic Hypertension and Risk of Cardiovascular Events: A Systematic Review and Meta-Analysis of Cohort Studies With 489,814 Participants. Front Cardiovasc Med. 2021;8:810105. doi: 10.3389/fcvm.2021.810105 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Orias M, Tabares AH, Peixoto AJ. Hypothesis: it is time to reconsider phenotypes in hypertension. J Clin Hypertens (Greenwich). May 2010;12(5):350–6. doi: 10.1111/j.1751-7176.2010.00266.x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). Jama. Dec 18 2002;288(23):2981–97. doi: 10.1001/jama.288.23.2981 [DOI] [PubMed] [Google Scholar]
- 6.Chobanian AV, Bakris GL, Black HR, et al. Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension. Dec 2003;42(6):1206–52. doi: 10.1161/01.HYP.0000107251.49515.c2 [DOI] [PubMed] [Google Scholar]
- 7.Julius S, Alderman MH, Beevers G, et al. Cardiovascular risk reduction in hypertensive black patients with left ventricular hypertrophy: the LIFE study. J Am Coll Cardiol. Mar 17 2004;43(6):1047–55. doi: 10.1016/j.jacc.2003.11.029 [DOI] [PubMed] [Google Scholar]
- 8.Materson BJ, Reda DJ, Cushman WC, et al. Single-drug therapy for hypertension in men. A comparison of six antihypertensive agents with placebo. The Department of Veterans Affairs Cooperative Study Group on Antihypertensive Agents. N Engl J Med. Apr 1 1993;328(13):914–21. doi: 10.1056/nejm199304013281303 [DOI] [PubMed] [Google Scholar]
- 9.Yamal JM, Oparil S, Davis BR, et al. Stroke outcomes among participants randomized to chlorthalidone, amlodipine or lisinopril in ALLHAT. J Am Soc Hypertens. Nov 2014;8(11):808–19. doi: 10.1016/j.jash.2014.08.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.A Randomized Trial of Intensive versus Standard Blood-Pressure Control. New England Journal of Medicine. 2015;373(22):2103–2116. doi: 10.1056/NEJMoa1511939 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Vakil D, Zinonos S, Kostis JB, et al. Monotherapy treatment with chlorthalidone or amlodipine in the systolic blood pressure intervention trial (SPRINT). J Clin Hypertens (Greenwich). Jul 2021;23(7):1335–1343. doi: 10.1111/jch.14296 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hripcsak G, Suchard MA, Shea S, et al. Comparison of Cardiovascular and Safety Outcomes of Chlorthalidone vs Hydrochlorothiazide to Treat Hypertension. JAMA Internal Medicine. 2020;180(4):542–551. doi: 10.1001/jamainternmed.2019.7454 [DOI] [PMC free article] [PubMed] [Google Scholar]