Abstract
Background and Aims
Although many angiotensin receptor blockers (ARBs) are widely used, comparative data regarding their impact on clinical outcomes are limited. We aimed to compare the clinical effectiveness of seven ARBs on long‐term cardiovascular outcomes in Korean patients with hypertension.
Methods
Using the Korean National Health Insurance Service database, the data of 780,785 patients with hypertension without cardiovascular disease (CVD) who initiated ARB treatment (candesartan, fimasartan, irbesartan, losartan, olmesartan, telmisartan, or valsartan) in 2014 and underwent this treatment for more than 6 months, were analyzed. Cox‐regression analysis was performed using Losartan as a comparator, as it was the most widely used drug, by adjusting age, sex, diabetes, dyslipidemia, smoking, alcohol drinking, exercise, body mass index, systolic blood pressure, albuminuria, estimated glomerular filtration rate, and concomitant medications. The occurrence of mortality and the rate of major adverse cardiovascular events (MACEs) of the six ARBs was compared with that of losartan.
Results
The median follow‐up duration was 5.94 (interquartile range, 5.87–5.97) years. In the crude analysis of all‐cause mortality and MACEs, fimasartan exhibited the lowest event rates. In the Cox‐regression analysis with adjustment, there was no significant difference in all‐cause mortality among ARBs. The risk of MACEs with ARBs was similar to that with losartan, although the risks with irbesartan (hazard ratio [HR], 1.079; 95% confidence interval [CI], 1.033–1.127; p = 0.007) and candesartan (HR: 1.066; 95% CI, 1.028–1.106; p = 0.015) were slightly higher.
Conclusion
In a Korean population of patients with hypertension without CVD, six different ARBs showed similar efficacy to losartan in terms of long‐term mortality and MACEs. Further well‐designed prospective studies are required to confirm our findings.
Keywords: angiotensin receptor blocker, cardiovascular event, hypertension, mortality, prognosis
1. INTRODUCTION
Angiotensin receptor blockers (ARBs) have a protective effect on the cardiovascular system and effectively lower blood pressure (BP). Additionally, ARBs are well‐tolerated and are recommended as a first‐line choice of antihypertensive medication. 1 , 2 , 3 In particular, ARBs have less common adverse effects, such as dry cough, which has a high incidence in Asian populations treated with angiotensin‐converting enzyme inhibitors (ACEIs). 4 ARBs are currently the most prescribed antihypertensive drugs in many countries, including South Korea. 2 , 5
Currently, nine ARBs are available in the global market. While most ARBs share a common molecular structure which translates into the class effect, each ARB also has a different chemical structure associated with additional benefits. 6 , 7 , 8 , 9 For example, losartan, candesartan, and valsartan exhibit strong cardiovascular protective effects in patients with heart failure with reduced ejection fraction. 10 , 11 , 12 Valsartan is more beneficial in terms of long‐term cardiovascular prognosis in patients with myocardial infarction. 13 Owing to their renoprotective effects, losartan and irbesartan have been suggested for the treatment of diabetic nephropathy. 14 , 15
In clinical practice, the prevalence of simple hypertension without complications is much higher than that of hypertension with complications. 16 Therefore, information on the use of ARBs for uncomplicated hypertension is very important. However, there is no evidence concerning which ARBs are the most suitable for patients with hypertension who do not have compelling indications, such as heart failure and myocardial infarction. As ARBs are widely prescribed antihypertensive drugs, a guide for choosing them for patients without cardiovascular disease (CVD) would be valuable to clinicians.
Recently developed ARBs, such as olmesartan and fimasartan, have strong BP‐lowering effects and are widely used in clinical practice. However, evidence of their effectiveness in improving cardiovascular prognosis beyond antihypertensive effect is scarce. Prognostic information associated with the use of these new ARBs will greatly help clinicians treat hypertension.
We conducted a retrospective analysis to compare the effects of different ARB types, including new‐generation ARBs, on long‐term mortality and cardiovascular events in patients with hypertension.
2. METHODS
2.1. Study patients
The Korean government has operated the National Health Insurance Service (NHIS) to provide medical insurance services to all Korean residents since 1989. 17 The NHIS has demographic, socioeconomic, and disability registration data to make decisions regarding eligibility and premium charging. Additionally, the NHIS has detailed data on healthcare utilization (procedures, drugs, and other treatments) submitted by medical providers for reimbursement. Using these data, the NHIS established the National Health Information Database (NHID) in 2012 to support public health policies and research. 18 We used the NHID provided by the NHIS (NHIS‐NHID). This study was conducted after obtaining approval from the Institutional Review Board (IRB) of NHIS (research management number: NHIS‐2020‐1‐110) and Seoul National University Boramae Medical Center (IRB No. GFIRB 2019‐304). As this study was conducted using the database provided by the NHIS, the IRB of Seoul National University Boramae Medical Center waived the need to obtain informed consent. This study was performed in accordance with the relevant guidelines and regulations. Among the individuals included in the NHIS‐NHID from January 1, 2014 to December 31, 2014, a total of 1,336,150 patients with the following inclusion criteria were identified: (1) diagnosis of hypertension (ICD diagnostic code: I10), (2) no history of CVD (ischemic heart disease, stroke, and heart failure), (3) health check‐up in 2014 and presence of major clinical and laboratory data, (4) initiation of ARB administration as an antihypertensive medication in 2014, and (5) ARB administration for >6 months. Administration of other classes of antihypertensive drugs was allowed. Among the nine ARBs marketed worldwide, azilsartan, which was not yet introduced into Korea in 2014, and eprosartan, which has a small number of prescriptions (n = 7300), were not included. Further exclusion criteria were as follows: (1) the occurrence of mortality before the date of prescription (n = 356) and (2) missing demographic characteristics and examination results (n = 547,709). Finally, we analyzed the data of 780,785 patients.
2.2. ARBs
The ARBs investigated were candesartan, fimasartan, irbesartan, losartan, olmesartan, telmisartan, and valsartan, according to the Anatomical Therapeutic Chemical codes maintained by World Health Organization (Supporting Information: Table S1).
2.3. Collection of clinical variables
Demographic, clinical, and laboratory data were collected from the health check‐up database in 2014. Body mass index (BMI) was calculated by dividing the weight (kg) by the square of height (m2). Systolic BP (SBP) and diastolic BP were measured using an oscillometric device in the right upper arm. The measurement was performed three times at the right upper arm, and the average of lower two values were taken, the values reflecting “during drug use.” Cardiovascular risk factors, including diabetes mellitus, dyslipidemia, smoking, alcohol consumption, and household income levels, were obtained using diagnostic codes and questionnaires. After overnight fasting for approximately 12 h, the blood glucose, total cholesterol, high‐density lipoprotein cholesterol, low‐density lipoprotein cholesterol, triglyceride, and creatine levels were measured. Urinalysis was performed and the presence and degree of proteinuria were assessed. The Chronic Kidney Disease Epidemiology Collaboration equation was used to calculate the estimated glomerular filtration rate (eGFR). Considering impacts on cardiovascular outcomes, information on medication of other antihypertensive agents (calcium channel blockers, beta‐blockers, ACEIs, and diuretics), antithrombotic agents, and statins was also obtained in cases where the medication was prescribed for more than 6 months at the time of initial ARB prescription.
2.4. Clinical outcomes
All‐cause mortality and major adverse cardiovascular events (MACEs) were the main outcome variables in this study. All‐cause mortality was determined from the date of death. MACEs were defined as cardiac death, nonfatal myocardial infarction, stroke, hospitalization for heart failure, and coronary revascularization. The Korean Standard Classification of Diseases (KCD‐7‐based ICD‐10) was used to define each MACE (Supporting Information: Table S2). The first date when the above ICD‐10 codes were present in the claims data was defined as the event date.
2.5. Statistical analysis
The χ 2 test for categorical variables and the Kruskal–Wallis test for continuous variables were performed to evaluate the differences in the distribution of demographic characteristics, incidence of MACEs, and distribution of all‐cause mortality depending on the type of ARB. Hazard ratios (HR) and 95% confidence intervals (CI) of all‐cause mortality and MACEs were calculated by using a Cox proportional‐hazard model adjusted for age, sex, diabetes (E10–E14), dyslipidemia (E78), smoking, alcohol drinking, exercise, household income, BMI, SBP, eGFR, and concomitant medications, including calcium antagonists, beta‐blockers, ACEIs, diuretics, antithrombotic agents, and statins. The proportional assumption of the Cox analysis was conducted for Cox proportional‐hazard modeling. In univariate analysis, a log‐rank test was conducted. Subsequently, multiple analyses were conducted using the Cox proportional hazard model. All statistical analyses were performed using SAS 9.4 software (SAS Institute). All two‐sided p < 0.05 were considered statistically significant.
3. RESULTS
Table 1 shows the distribution of demographic and clinical characteristics according to ARB type of study patients. A total of 58,892 (7.5%), 21,041 (2.7%), 37,170 (4.8%), 298,461 (38.2%), 100,348 (12.9%), 125,730 (16.1%), and 139,143 (17.8%) patients were taking candesartan, fimasartan, irbesartan, losartan, olmesartan, telmisartan, and valsartan, respectively. The study included 442,694 (56.7%) men and 338,091 (43.3%) women (average age, 59.62 years). The average age was the highest in the losartan group (60.75 years) and the lowest in the fimasartan group (57.66 years). The proportion of women was the highest in the losartan group (46.2%) and the lowest in the telmisartan group (39.4%). The average SBP of all study patients was 129.11 mmHg. The average SBP was the highest in the valsartan group (130.17 mmHg) and the lowest in the fimasartan group (127.37 mmHg). The prevalence of diabetes mellitus in all patients was 24.4%. The irbesartan group had the highest prevalence of diabetes mellitus (31.0%) while the fimasartan group had the lowest (21.1%). The prevalence of dyslipidemia in all patients was 38.2%, with the candesartan group exhibiting the highest prevalence (45.4%) and the losartan group exhibiting the lowest (35.4%). The proportion of smokers and drinkers was the highest in the telmisartan group and the lowest in the irbesartan group. The proportion of patients taking calcium antagonists, beta‐blockers, ACEIs, diuretics, antithrombotic agents, and statins was 22.5%, 8.8%, 0.5%, 6.1%, 26.2%, and 30.6%, respectively.
Table 1.
Baseline clinical characteristics of study patients
| Characteristic | Total (n = 780,785) | Candesartan (n = 58,892) | Fimasartan (n = 21,041) | Irbesartan (n = 37,170) | Losartan (n = 298,461) | Olmesartan (n = 100,348) | Telmisartan (n = 125,730) | Valsartan (n = 139,143) | p value | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Age (years), median (Q1, Q3) | 59.0 (52.0, 67.0) | 60.0 (53.0, 68.0) | 57.0 (50.0, 65.0) | 60.0 (54.0, 68.0) | 60.0 (54.0, 68.0) | 58.0 (52.0, 66.0) | 58.0 (52.0, 66.0) | 58.0 (52.0, 66.0) | <0.001a | ||||||||
| Sex | <0.001* | ||||||||||||||||
| Male, n (%) | 442,694 | 56.7 | 32,448 | 55.1 | 11,819 | 56.2 | 20,220 | 54.4 | 160,732 | 53.9 | 58,953 | 58.8 | 76,198 | 60.6 | 82,324 | 59.2 | |
| Female, n (%) | 338,091 | 43.3 | 26,444 | 44.9 | 9,222 | 43.8 | 16,950 | 45.6 | 137,729 | 46.2 | 41,395 | 41.3 | 49,532 | 39.4 | 56,819 | 40.8 | |
| BMI (kg/m²), median (Q1, Q3) | 25.2 (23.2, 27.3) | 25.1 (23.2, 27.1) | 25.1 (23.2, 27.2) | 25.1 (23.2, 27.1) | 25.0 (23.1, 27.1) | 25.4 (23.5, 27.5) | 25.3 (23.4, 27.4) | 25.2 (23.3, 27.4) | <0.001a | ||||||||
| SBP (mmHg), median (Q1, Q3) | 130.0 (120.0, 138.0) | 129.0 (120.0, 137.0) | 128.0 (119.0, 136.0) | 130.0 (120.0, 138.0) | 130.0 (120.0, 139.0) | 128.0 (120.0, 136.0) | 129.0 (120.0, 137.0) | 130.0 (120.0, 139.0) | <0.001a | ||||||||
| DBP (mmHg), median (Q1, Q3) | 80.0 (72.0, 85.0) | 80.0 (70.0, 84.0) | 80.0 (71.0, 85.0) | 80.0 (71.0, 85.0) | 80.0 (72.0, 85.0) | 80.0 (70.0, 84.0) | 80.0 (71.0, 85.0) | 80.0 (73.0, 86.0) | <0.001a | ||||||||
| Diabetes (yes), n (%) | 190,528 | 24.4 | 16,370 | 27.8 | 4,435 | 21.1 | 11,532 | 31.0 | 71,122 | 23.8 | 23,475 | 23.4 | 29,727 | 23.6 | 33,867 | 24.3 | <0.001* |
| Dyslipidemia (Yes), n (%) | 398,485 | 38.2 | 26,728 | 45.4 | 8,608 | 40.9 | 16,616 | 44.7 | 105,665 | 35.4 | 36,678 | 36.6 | 47,678 | 37.9 | 56,512 | 40.6 | <0.001* |
| Smoke (yes), n (%) | 143,226 | 18.3 | 9,701 | 16.5 | 3,946 | 18.8 | 5,828 | 15.7 | 51,707 | 17.3 | 19,598 | 19.5 | 24,918 | 19.8 | 27,528 | 19.8 | <0.001* |
| Drink (yes), n (%) | 233,103 | 29.9 | 15,814 | 26.9 | 6,370 | 30.3 | 9,860 | 26.5 | 83,784 | 28.1 | 32,402 | 32.3 | 40,831 | 32.5 | 44,042 | 31.7 | <0.001* |
| Physical activity (yes), n (%) | 287,360 | 36.8 | 21,311 | 36.2 | 7,852 | 37.3 | 13,193 | 35.5 | 111,244 | 37.3 | 36,712 | 36.6 | 45,923 | 36.5 | 51,125 | 36.7 | <0.001* |
| Income quartiles | <0.001* | ||||||||||||||||
| Lowest, n (%) | 158,555 | 20.3 | 11,338 | 19.3 | 4,335 | 20.6 | 7,182 | 19.3 | 63,322 | 21.2 | 20,103 | 20.0 | 24,845 | 19.8 | 27,430 | 19.7 | |
| Second, n (%) | 141,405 | 18.1 | 9,910 | 16.8 | 3,883 | 18.5 | 6,184 | 16.6 | 55,782 | 18.7 | 18,126 | 18.1 | 22,701 | 18.1 | 24,819 | 17.8 | |
| Third, n (%) | 185,401 | 23.8 | 13,419 | 22.8 | 5,100 | 24.2 | 8,501 | 22.9 | 71,642 | 24.0 | 23,985 | 23.9 | 29,478 | 23.5 | 33,276 | 23.9 | |
| Highest, n (%) | 295,424 | 37.8 | 24,225 | 41.1 | 7,723 | 36.7 | 15,303 | 41.2 | 107,715 | 36.1 | 38,134 | 38.0 | 48,706 | 38.7 | 53,618 | 38.5 | |
| FBS (mg/dl), median (Q1, Q3) | 101.0 (92.0, 116.0) | 102.0 (92.0, 116.0) | 100.0 (91.0, 113.0) | 102.0 (93.0, 119.0) | 101.0 (92.0, 116.0) | 102.0 (92.0, 116.0) | 102.0 (93.0, 116.0) | 101.0 (92.0, 116.0) | <0.001a | ||||||||
| Total cholesterol (mg/dl), median (Q1, Q3) | 186.0 (163.0, 212.0) | 182.0 (158.0, 208.0) | 186.0 (163.0, 211.0) | 182.0 (158.0, 207.0) | 187.0 (164.0, 212.0) | 188.0 (164.0, 214.0) | 188.0 (164.0, 213.0) | 185.0 (161.0, 211.0) | <0.001a | ||||||||
| HDL (mg/dl), median (Q1, Q3) | 50.0 (43.0, 60.0) | 50.0 (43.0, 60.0) | 51.0 (43.0, 60.0) | 50.0 (42.0, 59.0) | 51.0 (43.0, 60.0) | 50.0 (43.0, 59.0) | 50.0 (42.0, 59.0) | 50.0 (43.0, 60.0) | <0.001a | ||||||||
| LDL (mg/dl), median (Q1, Q3) | 106.0 (84.0, 129.0) | 102.0 (80.0, 126.0) | 105.0 (83.0, 128.0) | 102.0 (81.0, 125.0) | 106.0 (85.0, 130.0) | 107.0 (84.0, 131.0) | 107.0 (84.0, 131.0) | 104.0 (82.0, 128.0) | <0.001a | ||||||||
| TG (mg/dl), median (Q1, Q3) | 124.0 (88.0, 179.0) | 122.0 (87.0, 174.0) | 123.0 (87.0, 178.0) | 121.0 (86.0, 173.0) | 124.0 (88.0, 177.0) | 128.0 (90.0, 184.0) | 125.0 (89.0, 180.0) | 125.0 (88.0, 180.0) | <0.001a | ||||||||
| GFR (ml/min), median (Q1, Q3) | 81.0 (70.0, 94.0) | 81.0 (69.0, 94.0) | 82.0 (71.0, 96.0) | 80.0 (68.0, 93.0) | 81.0 (69.0, 93.0) | 81.0 (70.0, 94.0) | 82.0 (71.0, 96.0) | 82.0 (71.0, 95.0) | <0.001a | ||||||||
| Proteinuria | <0.001* | ||||||||||||||||
| Negative (−), n (%) | 725,690 | 92.9 | 54,556 | 92.6 | 19,946 | 94.8 | 33,643 | 90.5 | 278,075 | 93.2 | 93,991 | 93.7 | 117,165 | 93.2 | 128,314 | 92.2 | |
| Trace (±), n (%) | 23,719 | 3.0 | 1880 | 3.2 | 520 | 2.5 | 1312 | 3.5 | 8901 | 3.0 | 2859 | 2.9 | 3707 | 3.0 | 4540 | 3.3 | |
| Positive (+1), n (%) | 18,836 | 2.4 | 1,448 | 2.5 | 366 | 1.7 | 1182 | 3.2 | 7136 | 2.4 | 2148 | 2.1 | 2848 | 2.3 | 3708 | 2.7 | |
| Positive (+2), n (%) | 9162 | 1.2 | 744 | 1.3 | 162 | 0.8 | 705 | 1.9 | 3251 | 1.1 | 967 | 1.0 | 1455 | 1.2 | 1878 | 1.4 | |
| Positive (+3), n (%) | 2783 | 0.36 | 219 | 0.37 | 39 | 0.19 | 263 | 0.71 | 908 | 0.30 | 322 | 0.32 | 462 | 0.37 | 570 | 0.41 | |
| Positive (+4), n (%) | 595 | 0.08 | 45 | 0.08 | 8 | 0.04 | 65 | 0.17 | 190 | 0.06 | 61 | 0.06 | 93 | 0.07 | 133 | 0.10 | |
| Calcium channel blockers (yes), n (%) | 175,253 | 22.5 | 16,224 | 27.6 | 4219 | 20.1 | 11,023 | 29.7 | 80,508 | 27.0 | 18,771 | 18.7 | 18,162 | 14.5 | 26,346 | 18.9 | <0.001* |
| Beta‐blockers (yes), n (%) | 68,705 | 8.8 | 6245 | 10.6 | 1288 | 6.1 | 3904 | 10.5 | 24,380 | 8.2 | 8044 | 8.0 | 10,233 | 8.1 | 14,611 | 10.5 | <0.001* |
| ACEIs (yes), n (%) | 4060 | 0.52 | 341 | 0.58 | 142 | 0.67 | 237 | 0.64 | 1280 | 0.43 | 474 | 0.47 | 592 | 0.47 | 994 | 0.71 | <0.001* |
| Diuretics (yes), n (%) | 47,779 | 6.1 | 3302 | 5.6 | 1485 | 7.1 | 2046 | 5.5 | 17,074 | 5.7 | 5616 | 5.6 | 8070 | 6.4 | 10,186 | 7.3 | <0.001* |
| Antithrombotic agents (yes), n (%) | 204,531 | 26.2 | 16,825 | 28.6 | 4,270 | 20.3 | 11,593 | 31.2 | 78,708 | 26.4 | 26,611 | 26.5 | 30,908 | 24.6 | 35,616 | 25.6 | <0.001* |
| Statins (yes), n (%) | 239,239 | 30.6 | 22,323 | 37.9 | 6725 | 32.0 | 13,422 | 36.1 | 84,854 | 28.4 | 28,848 | 28.8 | 37,632 | 29.9 | 45,435 | 32.7 | <0.001* |
Abbreviations: ACEI, angiotensin‐converting enzyme inhibitor; BMI, body mass index; DBP, diastolic blood pressure; FBS, fasting blood sugar; GFR, glomerular filtration rate; HDL, high‐density lipoprotein; LDL, low‐density lipoprotein; SBP, systolic blood pressure; TG, triglyceride.
p value was obtained using χ 2 test.
p value was obtained using the Kruskal–Wallis test.
The incidence of all‐cause mortality and MACEs is presented in Table 2. During the median follow‐up period of 5.94 (interquartile range, 5.87–5.97) years, the all‐cause mortality and MACE rates of the study patients were 2.9% and 5.4%, respectively. The all‐cause mortality rate was the highest in the losartan group (3.2%) and the lowest in the fimasartan group (2.3%). The incidence of MACEs was the highest in the irbesartan group (6.17%) and the lowest in the fimasartan group (4.6%). A similar trend was observed with the individual MACEs; the incidence rate was the highest in the irbesartan group (6.2%) and the lowest in the fimasartan group.
Table 2.
Incidence of all‐cause mortality and MACE
| Total (n = 780,785) | Candesartan (n = 58,892) | Fimasartan (n = 21,041) | Irbesartan (n = 37,170) | Losartan (n = 298,461) | Olmesartan (n = 100,348) | Telmisartan (n = 125,730) | Valsartan (n = 139,143) | p value* | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| n | % | n | % | n | % | n | % | n | % | n | % | n | % | n | % | ||
| All‐cause mortality | 22,291 | 2.9 | 1701 | 2.9 | 473 | 2.3 | 1080 | 2.9 | 9566 | 3.2 | 2619 | 2.6 | 3114 | 2.5 | 3738 | 2.7 | <0.001 |
| MACE | 42,216 | 5.4 | 3445 | 5.9 | 961 | 4.6 | 2293 | 6.2 | 16,932 | 5.7 | 5119 | 5.1 | 6159 | 4.9 | 7307 | 5.3 | <0.001 |
| Cardiac death | 4121 | 0.53 | 325 | 0.55 | 76 | 0.36 | 236 | 0.63 | 1750 | 0.59 | 447 | 0.45 | 562 | 0.45 | 725 | 0.52 | <0.001 |
| Nonfatal MI | 7501 | 0.96 | 608 | 1.0 | 178 | 0.85 | 379 | 1.0 | 2958 | 0.99 | 928 | 0.92 | 1154 | 0.92 | 1296 | 0.93 | 0.025 |
| Stroke | 27,900 | 3.6 | 2230 | 3.8 | 631 | 3.0 | 1531 | 4.1 | 11,265 | 3.8 | 3327 | 3.3 | 4076 | 3.2 | 4840 | 3.5 | <0.001 |
| HF | 8335 | 1.1 | 739 | 1.3 | 200 | 1.0 | 479 | 1.3 | 3319 | 1.1 | 1060 | 1.1 | 1141 | 0.91 | 1397 | 1.0 | <0.001 |
| Coronary revascularization | 150 | 0.02 | 11 | 0.02 | 3 | 0.01 | 6 | 0.02 | 74 | 0.02 | 20 | 0.02 | 16 | 0.01 | 20 | 0.01 | 0.189 |
Abbreviations: HF, heart failure; MACE, major adverse cardiovascular event; MI, myocardial infarction.
p value was obtained using the χ 2 test.
The risks for all‐cause mortality and MACEs according to different ARBs, compared with those of losartan, are presented in Table 3. In the crude model, the HRs were lower with values of 0.901, 0.701, 0.905, 0.813, 0.771, and 0.838 in the candesartan, fimasartan, irbesartan, olmesartan, telmisartan, and valsartan groups, respectively. For MACEs, the candesartan and irbesartan groups showed a higher risk with HR of 1.039 and 1.098, while fimasartan, olmesartan, telmisartan, and valsartan showed a lower risk than that of the losartan group in the crude model with values of 0.810, 0.901, 0.864, and 0.932, respectively. In the adjusted model, all‐cause mortality risks were similar among all ARBs. Although not substantial, the HRs of MACEs were 1.066 (95% CI, 1.028–1.106; p = 0.015) and 1.079 (95% CI, 1.033–1.127; p = 0.007) in the candesartan and irbesartan groups, respectively, compared with HR of the losartan group. There was no significant difference in the risk of MACEs among other ARBs. The results of the analysis of individual MACE are presented in Supporting Information: Table S3.
Table 3.
Risks for all‐cause mortality and MACE according to different ARBs compared to those associated with losartan
| Crude model | Adjusted model | |||||||
|---|---|---|---|---|---|---|---|---|
| HR | 95% CI | p value | HR | 95% CI | p value | |||
| Low | High | Low | High | |||||
| All‐cause mortality (ref. = losartan) | ||||||||
| Candesartan | 0.901 | 0.856 | 0.948 | 0.003 | 0.994 | 0.943 | 1.046 | 0.613 |
| Fimasartan | 0.701 | 0.639 | 0.769 | <0.001 | 0.975 | 0.889 | 1.070 | 0.858 |
| Irbesartan | 0.905 | 0.850 | 0.964 | 0.009 | 0.947 | 0.889 | 1.008 | 0.180 |
| Olmesartan | 0.813 | 0.778 | 0.849 | 0.065 | 1.004 | 0.961 | 1.049 | 0.256 |
| Telmisartan | 0.771 | 0.740 | 0.802 | <0.001 | 0.965 | 0.927 | 1.005 | 0.331 |
| Valsartan | 0.838 | 0.807 | 0.870 | 0.781 | 0.993 | 0.956 | 1.031 | 0.538 |
| MACE (ref. = losartan) | ||||||||
| Candesartan | 1.039 | 1.002 | 1.078 | <0.001 | 1.066 | 1.028 | 1.106 | 0.015 |
| Fimasartan | 0.810 | 0.759 | 0.864 | <0.001 | 1.001 | 0.938 | 1.069 | 0.400 |
| Irbesartan | 1.098 | 1.051 | 1.146 | <0.001 | 1.079 | 1.033 | 1.127 | 0.007 |
| Olmesartan | 0.901 | 0.873 | 0.930 | 0.001 | 1.017 | 0.986 | 1.050 | 0.567 |
| Telmisartan | 0.864 | 0.839 | 0.890 | <0.001 | 0.995 | 0.966 | 1.024 | 0.019 |
| Valsartan | 0.932 | 0.907 | 0.958 | 0.262 | 1.022 | 0.994 | 1.050 | 0.771 |
Note: The following variables were controlled in the adjusted model: age, sex, body mass index, systolic blood pressure, diabetes mellitus, dyslipidemia, cigarette smoking, alcohol drinking, physical activity, household income, glomerular filtration rate, and the use of calcium channel blockers, beta‐blockers, angiotensin‐converting enzyme inhibitors, diuretics, antithrombotics, and statins.
Abbreviations: ARB, angiotensin receptor blocker; CI, confidence interval; HR, hazard ratio; MACE, major adverse cardiovascular event.
Subgroup analyses were conducted according to age, sex, and diabetes status (Supporting Information: Table S4). The age subgroups included patients aged ≥ 70 years and those aged < 70 years. Several ARBs had different HRs for all‐cause mortality and MACEs in the crude model, but there was no substantial difference in the adjusted model, which was similar to the results of the overall study population. We conducted stratified analysis with only ARB, ARB with one class of antihypertensive drug, and ARB with two or more classes of antihypertensive drugs (Supporting Information: Table S5). In the analysis including only the ARB group for MACEs, the HR of the candesartan group was 1.256 (95% CI, 1.035–1.524; p = 0.109) and that of the irbesartan group was 1.238 (95% CI, 0.978–1.569; p = 0.241). There was no substantial difference in all‐cause mortality and MACEs in other groups.
4. DISCUSSION
In this analysis of the real‐world data of a large Korean population of patients with hypertension without CVD, seven different ARBs had similar mortality rates during a median follow‐up period of 5.94 years. Our subgroup analysis of geriatric patients (aged >70 years) and patients with diabetes also demonstrated no substantial differences among ARBs. Although some ARBs showed better effects than those of losartan in terms of MACE occurrence, the degree of difference was only marginal and, thus, less likely to be clinically important. We also showed that newer ARBs, such as fimasartan and olmesartan, had a cardiovascular prognosis similar to that of other ARBs. To our knowledge, this is the first study to compare the efficacy of seven different ARBs on long‐term clinical outcomes in patients with hypertension. In particular, the results of our study are noteworthy as they also provide data on the long‐term prognostic effects of fimasartan and olmesartan, which have scarcely been reported.
Most of the data from existing studies comparing ARBs relate to the short‐term efficacy of the BP‐lowering effect or pleiotropic effect of ARBs. 19 , 20 , 21 , 22 , 23 , 24 Few studies have compared the long‐term clinical outcomes of different types of ARBs. A study of 6876 geriatric patients with heart failure compared the effects of five ARBs (candesartan, irbesartan, losartan, telmisartan, and valsartan) and found that losartan was associated with a lower survival rate than that of other ARBs. 25 Antoniou et al. 26 investigated 54,186 patients with diabetes and showed that those taking telmisartan and valsartan were associated with a lower risk of development of acute myocardial infarction, heart failure, and stroke than those taking irbesartan, losartan, and candesartan. Our study differs from these two studies in that we included patients with hypertension and also showed the effectiveness of the new ARBs, fimasartan, and olmesartan. Therefore, our findings are applicable to a broader population. One observational study from Canada showed that patients with hypertension treated with irbesartan had the lowest rate of developing cardiovascular events compared with those receiving other ARBs. 27 However, the primary goal of this study was to compare the effects of ARBs with other antihypertensive drug classes, and only four ARBs (candesartan, irbesartan, losartan, and valsartan) were analyzed in this sthsrudy. Moreover, the study population taking ARBs was relatively small (n = 3490). A study in Taiwan that analyzed a large number of patients (n = 690,463) from claims data compared the effects of six ARBs (candesartan, irbesartan, losartan, olmesartan, telmisartan, and valsartan) and found that olmesartan did not increase long‐term cardiovascular risk compared with that by losartan. 28 This study deserves attention in that it elucidates the effect of olmesartan, unlike previous studies. However, the study population was more heterogeneous as patients with underlying CVD were not excluded, contrary to the case in our study.
ARBs introduced in the early phase have been used in many large‐scale randomized clinical trials (RCTs) with long‐term follow‐up. Most of these studies were designed to compare other classes of antihypertensive drugs, which led to compelling indications for various medical conditions. 11 , 29 , 30 , 31 Validating the effectiveness of certain drugs with a well‐designed study is important for clinicians prescribing the drug. However, it is difficult to conduct large‐scale RCTs on newly developed ARBs because many ARBs supported by strong evidence are already available. In this situation, retrospective analyses using real‐world data may be helpful. In particular, claims data enable analysis of the data of many patients. Newer ARBs, including olmesartan, fimasartan, and azilsartan, are known to have higher potency compared with previous ARBs. 32 , 33 Many head‐to‐head studies have revealed more potent antihypertensive effects, particularly when compared with losartan or valsartan. 22 , 34 , 35 However, these newer ARBs do not have compelling evidence of long‐term cardiovascular outcomes, and the clinical use of these drugs is limited. Our study demonstrates that newer ARBs have similar long‐term clinical outcomes to those of earlier ARBs, based on the data of many patients in the real world. We believe that these data will help clinicians use new ARBs.
Most ARBs share a common molecular structure, which contributes to the class effect of ARBs. 36 However, ARBs have considerable differences in their molecular structures, which can show differences in their clinical benefits. For example, the number of hydrogen bonds that affect binding affinity is different among ARBs, and this can translate into various antihypertensive effects. 37 Several studies comparing the antihypertensive effect of ARBs have shown differences in the magnitude of BP reduction. 34 , 35 This difference in BP control may affect patients' prognoses. In patients with acute myocardial infarction, comparative analysis revealed that insurmountable ARBs, which have a longer and more stable duration of action, were more effective on long‐term clinical outcomes compared with surmountable ARBs. 38 A study with a Swedish heart failure registry showed that patients taking insurmountable ARB (candesartan) had better all‐cause mortality rates compared with surmountable ARB (losartan). 38 Unlike these studies, our results revealed no significant differences in cardiovascular outcomes of six ARBs compared with those of losartan, suggesting that the difference of ARBs is not clinically relevant in patients without CVD. Further studies are warranted to assess whether the class effect applies to certain diseases.
This study had several limitations. First, it included patients receiving other classes of antihypertensive medications, in addition to ARBs. Although the concomitant use of other antihypertensive medications was adjusted for the multivariable analysis, we also conducted stratified analysis. Some ARBs showed worse outcomes than those of others in the analysis of the ARB group. We could not find clinically meaningful explanations for the observed difference, which was also not consistent with the findings of previous studies. Second, although we included patients receiving ARBs for more than 6 months, the on‐treatment duration might have varied among ARBs. Third, the characteristics of the seven ARBs differ, reflecting current practice in that certain ARBs are used for compelling indications and there is preference for a stronger agent in younger patients. These may partly explain the difference in the crude and adjusted models. Fourth, several variables, such as BP and BMI, have a dynamic nature. Although the measurement of these variables and the initiation of ARBs were performed in the same year, the timing does not exactly match and the dynamic nature could not be reflected. Especially, BP could not reflect the efficacy of each ARB, thus, limiting interpretation of our results. Finally, as this is a real‐world study with a large sample size, the interpretation of the study findings should not depend solely on the p value. Thus, we considered the sample size, 95% CI, and clinical significance of the observed effect when interpreting the results.
5. CONCLUSIONS
In this study, which targeted a large number of Korean adults without CVD using national health data, we found that six ARBs (candesartan, fimasartan, irbesartan, losartan, olmesartan, telmisartan, and valsartan) had comparable effects in terms of long‐term cardiovascular risk compared with that of losartan. These findings also suggest that fimasartan and olmesartan, newly developed strong ARBs, may also have similar cardiovascular effects compared to those of earlier traditional ARBs. Better‐designed prospective studies are needed to confirm our findings.
AUTHOR CONTRIBUTIONS
Wonjae Lee: Data curation; formal analysis; investigation; methodology; project administration; validation; writing – original draft; writing – review and editing. Jeehoon Kang: Investigation; methodology; supervision; validation; writing – review and editing. Jun‐Bean Park: Investigation; methodology; supervision; validation; writing – review and editing. Won‐Woo Seo: Investigation; methodology; supervision; validation; writing – review and editing. Seung‐Yeon Lee: Validation; writing – original draft. Woo‐Hyun Lim: Investigation; methodology; supervision; validation; writing – review and editing. Ki‐Hyun Jeon: Investigation; methodology; supervision; validation; writing – review and editing. In‐Chang Hwang: Investigation; methodology; supervision; validation; writing – review and editing. Hack‐Lyoung Kim: Conceptualization; data curation; formal analysis; investigation; methodology; project administration; supervision; validation; writing – review and editing.
CONFLICT OF INTEREST
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
ETHICS STATEMENT
This study was approved by the IRB of NHIS (research management number: NHIS‐2020‐1‐110) and Seoul National University Boramae Medical Center (IRB No. GFIRB 2019‐304).
TRANSPARENCY STATEMENT
The lead author Hack‐Lyoung Kim affirms that this manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned (and, if relevant, registered) have been explained.
Supporting information
Supporting information.
ACKNOWLEDGMENTS
The authors would like to thank Editage (www.editage.co.kr) for English language editing and Ji Lee of The Way Health Care, Seoul, South Korea for the statistical analyses.
Lee W, Kang J, Park J‐B, et al. Long‐term mortality and cardiovascular events of seven angiotensin receptor blockers in hypertensive patients: analysis of a national real‐world database: a retrospective cohort study. Health Sci Rep. 2023;6:e1056. 10.1002/hsr2.1056
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from National Health Information Database provided by the National Health Insurance Service. Restrictions apply to the availability of these data, which were used under a policy of the NHIS. The data and materials other than the raw data underlying the study are available from the corresponding author, Dr. Hack‐Lyoung Kim, upon reasonable request.
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: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation. 2018;138:426. [DOI] [PubMed] [Google Scholar]
- 2. Williams B, Mancia G, Spiering W, et al. 2018 ESC/ESH Guidelines for the management of arterial hypertension. Eur Heart J. 2018;39:3021‐3104. [DOI] [PubMed] [Google Scholar]
- 3. Lee HY, Shin J, Kim GH, et al. 2018 Korean Society of Hypertension Guidelines for the management of hypertension: part II‐diagnosis and treatment of hypertension. Clin Hypertens. 2019;25:20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Dicpinigaitis PV. Angiotensin‐converting enzyme inhibitor‐induced cough. Chest. 2006;129:169S‐173S. [DOI] [PubMed] [Google Scholar]
- 5. Kim HC, Cho SMJ, Lee H, et al. Korea hypertension fact sheet 2020: analysis of nationwide population‐based data. Clin Hypertens. 2021;27:8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Israili Z. Clinical pharmacokinetics of angiotensin II (AT1) receptor blockers in hypertension. J Hum Hypertens. 2000;14:S73‐S86. [DOI] [PubMed] [Google Scholar]
- 7. Brunner‐La Rocca HP, Vaddadi G, Esler MD. Recent insight into therapy of congestive heart failure: focus on ACE inhibition and angiotensin‐II antagonism. JACC. 1999;33:1163‐1173. [DOI] [PubMed] [Google Scholar]
- 8. Plum J, Bünten B, Németh R, Grabensee B. Effects of the angiotensin II antagonist valsartan on blood pressure, proteinuria, and renal hemodynamics in patients with chronic renal failure and hypertension. JASN. 1998;9:2223‐2234. [DOI] [PubMed] [Google Scholar]
- 9. Klingbeil AU, Schneider M, Martus P, Messerli FH, Schmieder RE. A meta‐analysis of the effects of treatment on left ventricular mass in essential hypertension. Am J Med. 2003;115:41‐46. [DOI] [PubMed] [Google Scholar]
- 10. Pitt B, Segal R, Martinez FA, et al. Randomised trial of losartan versus captopril in patients over 65 with heart failure (evaluation of losartan in the elderly study, ELITE). Lancet. 1997;349:747‐752. [DOI] [PubMed] [Google Scholar]
- 11. Cohn JN, Tognoni G. A randomized trial of the angiotensin‐receptor blocker valsartan in chronic heart failure. N Engl J Med. 2001;345:1667‐1675. [DOI] [PubMed] [Google Scholar]
- 12. Granger CB, McMurray JJ, Yusuf S, et al. Effects of candesartan in patients with chronic heart failure and reduced left‐ventricular systolic function intolerant to angiotensin‐converting‐enzyme inhibitors: the CHARM‐Alternative trial. Lancet. 2003;362:772‐776. [DOI] [PubMed] [Google Scholar]
- 13. Pfeffer MA, McMurray JJV, Velazquez EJ, et al. Valsartan, captopril, or both in myocardial infarction complicated by heart failure, left ventricular dysfunction, or both. N Engl J Med. 2003;349:1893‐1906. [DOI] [PubMed] [Google Scholar]
- 14. Brenner BM, Cooper ME, de Zeeuw D, et al. Effects of losartan on renal and cardiovascular outcomes in patients with type 2 diabetes and nephropathy. N Engl J Med. 2001;345:861‐869. [DOI] [PubMed] [Google Scholar]
- 15. Parving HH, Lehnert H, Bröchner‐Mortensen J, Gomis R, Andersen S, Arner P. The effect of irbesartan on the development of diabetic nephropathy in patients with type 2 diabetes. N Engl J Med. 2001;345:870‐878. [DOI] [PubMed] [Google Scholar]
- 16. Oliveras A, de la Sierra A. Resistant hypertension: patient characteristics, risk factors, co‐morbidities and outcomes. J Hum Hypertens. 2014;28:213‐217. [DOI] [PubMed] [Google Scholar]
- 17. Kwon S. Thirty years of national health insurance in South Korea: lessons for achieving universal health care coverage. Health Policy Plan. 2008;24:63‐71. [DOI] [PubMed] [Google Scholar]
- 18. Lee Y, Han K, Ko SH, Ko KS, Lee KU. Data analytic process of a nationwide population‐based study using National Health Information Database established by National Health Insurance Service. Diabetes Metab J. 2016;40:79‐82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Nixon RM, Müller E, Lowy A, Falvey H. Valsartan vs. other angiotensin II receptor blockers in the treatment of hypertension: a meta‐analytical approach. Int J Clin Pract. 2009;63:766‐775. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Kassler‐Taub K. Comparative efficacy of two angiotensin II receptor antagonists, irbesartan and losartan, in mild‐to‐moderate hypertension. Am J Hypertens. 1998;11:445‐453. [DOI] [PubMed] [Google Scholar]
- 21. Smith DHG, Cramer MJM, Neutel JM, Hettiarachchi R, Koval S. Comparison of telmisartan versus losartan: meta‐analysis of titration‐to‐response studies. Blood Press Monit. 2003;8:111‐117. [DOI] [PubMed] [Google Scholar]
- 22. White WB, Weber MA, Sica D, et al. Effects of the angiotensin receptor blocker azilsartan medoxomil versus olmesartan and valsartan on ambulatory and clinic blood pressure in patients with stages 1 and 2 hypertension. Hypertension. 2011;57:413‐420. [DOI] [PubMed] [Google Scholar]
- 23. Chrysant SG, Chrysant GS. The pleiotropic effects of angiotensin receptor blockers. J Clin Hypertens. 2006;8:261‐268. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Toth PP. Pleiotropic effects of angiotensin receptor blockers: addressing comorbidities by optimizing hypertension therapy. J Clin Hypertens. 2011;13:42‐51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Hudson M, Humphries K, Tu JV, Behlouli H, Sheppard R, Pilote L. Angiotensin II receptor blockers for the treatment of heart failure: a class effect? Pharmacotherapy. 2007;27:526‐534. [DOI] [PubMed] [Google Scholar]
- 26. Antoniou T, Camacho X, Yao Z, Gomes T, Juurlink DN, Mamdani MM. Comparative effectiveness of angiotensin‐receptor blockers for preventing macrovascular disease in patients with diabetes: a population‐based cohort study. Can Med Assoc J. 2013;185:1035‐1041. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Petrella R, Michailidis P. Retrospective analysis of real‐world efficacy of angiotensin receptor blockers versus other classes of antihypertensive agents in blood pressure management. Clin Ther. 2011;33:1190‐1203. [DOI] [PubMed] [Google Scholar]
- 28. Lin JW, Chang CH, Caffrey JL, Wu LC, Lai MS. Examining the association of olmesartan and other angiotensin receptor blockers with overall and cause‐specific mortality. Hypertension. 2014;63:968‐976. [DOI] [PubMed] [Google Scholar]
- 29. Dahlöf B, Devereux RB, Kjeldsen SE, et al. Cardiovascular morbidity and mortality in the losartan intervention for endpoint reduction in hypertension study (LIFE): a randomised trial against atenolol. Lancet. 2002;359:995‐1003. [DOI] [PubMed] [Google Scholar]
- 30. Pfeffer MA, Swedberg K, Granger CB, et al. Effects of candesartan on mortality and morbidity in patients with chronic heart failure: the CHARM‐Overall programme. Lancet. 2003;362:759‐766. [DOI] [PubMed] [Google Scholar]
- 31. Yusuf S, Teo KK, Pogue J, et al. Telmisartan, ramipril, or both in patients at high risk for vascular events. N Engl J Med. 2008;358:1547‐1559. [DOI] [PubMed] [Google Scholar]
- 32. Ojima M, Igata H, Tanaka M, et al. In vitro antagonistic properties of a new angiotensin type 1 receptor blocker, azilsartan, in receptor binding and function studies. J Pharmacol Exp Ther. 2011;336:801‐808. [DOI] [PubMed] [Google Scholar]
- 33. Paik SH, Chi YH, Lee JH, Han HS, Lee KT. Pharmacological profiles of a highly potent and long‐acting angiotensin II receptor antagonist, fimasartan, in rats and dogs after oral administration. Biol Pharm Bull. 2017;40:992‐1001. [DOI] [PubMed] [Google Scholar]
- 34. Oparil S, Williams D, Chrysant SG, Marbury TC, Neutel J. Comparative efficacy of olmesartan, losartan, valsartan, and irbesartan in the control of essential hypertension. J Clin Hypertens. 2001;3:283‐318. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Chung WB, Ihm SH, Jang SW, et al. Effect of fimasartan versus valsartan and olmesartan on office and ambulatory blood pressure in Korean patients with mild‐to‐moderate essential hypertension: a randomized, double‐blind, active control, three‐parallel group, forced titration, multicenter, phase IV study (Fimasartan Achieving Systolic Blood Pressure Target (FAST) study). DDDT. 2020;14:347‐360. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Miura S, Karnik SS, Saku K. Review: angiotensin II type 1 receptor blockers: class effects versus molecular effects. J Renin Angiotensin Aldosterone Syst. 2011;12:1‐7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Zhang H, Unal H, Gati C, et al. Structure of the angiotensin receptor revealed by serial femtosecond crystallography. Cell. 2015;161:833‐844. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Jeong HC, Jeong MH, Ahn Y, et al. Comparative assessment of angiotensin II type 1 receptor blockers in the treatment of acute myocardial infarction: surmountable vs. insurmountable antagonist. Int J Cardiol. 2014;170:291‐297. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supporting information.
Data Availability Statement
The data that support the findings of this study are available from National Health Information Database provided by the National Health Insurance Service. Restrictions apply to the availability of these data, which were used under a policy of the NHIS. The data and materials other than the raw data underlying the study are available from the corresponding author, Dr. Hack‐Lyoung Kim, upon reasonable request.