Antihypertensive effectiveness of combination therapy with losartan/hydrochlorothiazide for ‘real world’ management of isolated systolic hypertension

Abstract

Objectives:

The guidelines for hypertension require the presence of compelling indications for pharmacological management of hypertension associated with various diseases. Data mainly obtained through randomized controlled trials have provided evidence supporting effectiveness of the combination of losartan (Lo) and hydrochlorothiazide (HCTZ) for management of hypertensive patients. However, there have been few reports discussing the effectiveness of Lo/HTCZ (losartan 50 mg/hydrochlorothizide 12.5 mg) in the ‘real world’ in the management of isolated systolic hypertension (ISH). This study was designed to investigate the ‘real world’ effectiveness of Lo/HTCZ-based treatment of ISH associated with various diseases.

Methods:

This was a retrospective, uncontrolled analysis of data derived from a large, cross-sectional web-based clinical database collected by physicians.

Results:

Of 24,825 eligible patients, 20,726 were followed during a 6-month period. Among these, subjects for analysis included those with systolic blood pressure (SBP) >140 mmHg and diastolic BP (DBP) <90 mmHg; patients with diabetes mellitus and chronic kidney disease were excluded. A total of 15,846 patients were analysed. Among the various complications, hypercholesterolemia was the most frequent concomitant cardiovascular (CV) risk factor (48.1%), followed by obesity (16.3%). Associated clinical conditions were cerebrovascular diseases (9.6%), ischemic heart disease (7.9%) and left ventricular hypertrophy (4.6%). Total numbers of patients exhibiting any type of complications were 62% (≤64 years old), 69% (65–74 years old) and 67% (≥75 years old) (stratification of age). Mean SBP/DBP measurements (mm Hg) were 156/78 at the start, 140/72 at 1 month and 134/72 at 6 months. Blood pressure (BP) reductions associated with various diseases were similar among patients. Laboratory data including serum levels of total cholesterol, uric acid, hemoglobin A1c and serum potassium did not change during the study. Adverse effects such as orthostatic hypotension and considerable reductions in BP (>30 mmHg SBP) were rare.

Conclusions:

Lo/HTCZ is safe and effective in reducing and improving BP control in a ‘real world’ setting. Treatment with Lo/HTCZ enabled a substantial proportion of hypertensive patients with associated diseases to achieve the recommended goal of <140 mm Hg.

Keywords

cerebrovascular disease dyslipidemia ischemic heart disease left ventricular hypertrophy obesity

Introduction

In hypertension, the goal of treatment is prevention of cardiovascular diseases (CVD) [Ruilope, 2012]. Two major risk factors for CVD are hypertension and dyslipidemia, followed by other risk factors such as age, diabetes mellitus (DM), metabolic syndrome, chronic kidney disease (CKD) and smoking. Several guidelines recommend usage of antihypertensive drugs for treatment of special diseases [Kotchen, 2014]. In the real world, physicians are faced with treatment of various categories of patients who may have several risk factors in combination. Moreover, among a growing population of elderly patients, the predominant form of hypertension treated is isolated systolic hypertension (ISH) [Chaudhry et al. 2012; Turgut et al. 2013]. In addition, a large-scale clinical study has frequently focused on the effects of one drug on reduction of blood pressure (BP) and laboratory data, including adverse effects [Ogihara et al. 2011; Weycker et al. 2007].

It is generally considered that the combination of losartan (Lo) and hydrochlorothiazide (HCTZ) elicits BP reduction synergistically, through complementary mechanisms: thiazides act as an ‘enhancer’ of the antihypertensive effect of renin–angiotensin system (RAS) inhibition via combination therapy and detrimental effects such as electrolyte imbalance and impaired insulin sensitivity are mutually cancelled by the combination [MacKay et al. 1996].

Based on these characteristics, the present study demonstrates the utility of Lo/HCTZ in ‘real world’ management of elderly subjects with ISH [Riva and Lip, 2013]. Also in the present study, the effects of individual agents on other risk factors such as total cholesterol, uric acid, serum creatinine and serum potassium are examined, since there are few reports discussing the effects of antihypertensive drug combinations on these parameters [Khanna et al. 2008; Wile, 2012].

Methods

Methodology of the study

This was a retrospective, uncontrolled analysis of data derived from a large, cross-sectional, web-based clinical database assembled by physicians [Suzuki et al. 2012]. Physicians did not receive any specific indications on digit preferences or how to submit clinical data and BP levels in the database. Care-Net^® has developed a web-based drug monitoring system aimed at local collection, central recording and real-time analysis of drug prescriptions for a new drug. Physicians have their own identification number and password, which are required for any login to the system and web-based drug prescription. Clinical data, including anthropometric parameters, BP levels and heart rate are also required.

In light of these considerations, and cognizant of the potential limitations of findings derived from a web-based, noncontrolled drug monitoring system, we retrospectively analysed and summarized the main data derived from the registry on Lo/HCTZ [Lo 50 mg/HCTC 12.5 mg) prescriptions written by 3401 general practitioners collected over 6 months at some point between December 2006 and December 2009. The operation center identified the physician. If the physician was not identified, the registration was cancelled. After online registration, the identification (ID) numbers and passwords given to all physicians were required whenever they accessed the website. Care-Net^® provided a comprehensive electronic patient record system in which the participants were able to register their patients. In a single interface, patient information including ID number, age, gender, underlying diseases, past history, BP levels and the results of laboratory tests for urine protein, serum creatinine, total cholesterol, fasting blood glucose, hemoglobin A1c (HbA1C), uric acid and serum potassium were registered.

Patient selection

Patients eligible for this study were aged over 40 years and exhibited BP higher than 140/90 mm Hg or 130/80 mm Hg [patients with DM or CKD in spite of treatment with antihypertensive drugs including angiotensin receptor blockers (ARBs]. After registration, the study committee evaluated subjects registered by the physicians and then communicated a final decision to the physicians. After the start of treatment with Lo/HCTZ, physicians were encouraged to enter BP information monthly over the next 6 months. At the end of the study, the physicians were expected to report on the same laboratory data as those registered at the start of the study. The physicians reported all registered data as figures and compared trends in BP changes. Furthermore, the physicians were obliged to report any adverse effects that occurred during the study. In the analysis, patients with DM, CKD or diastolic BP (DBP) >90 mmHg were excluded because we had discussed the influence of these parameters as previously reported [Suzuki et al. 2012]. Also, in the present analysis, patients with DBP >90 mmHg represented less than 5% of the total.

Population enrolment and follow up

At the first visit, a total of 15,846 patients received a prescription of Lo/HCTZ. On termination of the analysis 13,520 (85%) outpatients had a follow-up visit at 6 months. According to this distribution, clinical data on BP levels, BP control and responder rates were analysed only in those patients who had valid data for these parameters at the time of predefined follow up (i.e. 15,846 outpatients for paired comparisons between baseline observation and 1 month follow up, and 13,520 outpatients for paired comparisons between either 1 month and 6 month follow-up visits or baseline visit and 6-month follow up).

Obesity was defined as a body mass index (BMI) >28 kg/m². The presence of hypercholesterolemia was defined on the basis of the following diagnostic criteria: serum total cholesterol values >240 mg/dl and/or history of using lipid-lowering drugs. Patients were considered to have DM if they had been diagnosed previously by a physician, or if they were receiving insulin or oral antidiabetic medication.

CKD was defined as an estimated GFR <60 ml/min/1.73 m² or the presence of proteinuria.

Left ventricular hypertrophy was defined by an echocardiographically assessed left ventricular mass indexed to the body surface area, greater than 125 g/m² in males and 110 g/m² in females.

Ischemic heart disease (prior myocardial infarction) was generally defined according to the presence during the referred acute phase of two of the following: (1) symptoms (e.g. chest pain) lasting longer than 15 min; (2) transient increase in serum concentrations of enzymes or markers indicating cardiac damage; and (3) electrocardiographic changes typical of myocardial ischemia (new persistent ST-segment elevation or pathological Q waves in two contiguous leads).

Nonfatal stroke was defined as a neurological deficit with sudden onset and persistence of symptoms for more than 24 h, or leading to death with no apparent causes other than vascular.

Definition of markers of organ damage was based on recommendations derived from currently available clinical guidelines [Ogihara et al. 2009].

Adverse effects

Adverse effects such as marked reduction of BP as assessed by physicians were reported as possible.

Statistical analysis

Data are expressed as mean ± standard deviation. Changes over time in BP were compared between the mean values before and after treatment using Dunnett’s test. Any patients having only pretreatment data were excluded from the analysis using Dunnett’s test. Continuous variables were compared using analysis of variance among groups. All statistical tests were two-sided with an α level of 0.05 and were performed using SAS Release 8.2 (SAS Institute Inc., Cary, NC, USA).

Results

General characteristics

Figure 1 shows the patient flow chart. Of 15,846 eligible patients, 13,520 were followed for 6 months. Table 1 shows general patient characteristics. This population was almost equally distributed between males (53.2%) and females (46.8%). Patients were stratified as: ≤64 years old 6591 (41.6%); 65–74 years old 4468 (28.2%); and ≥75 years old 4787 (30.2%).

Figure 1.

Flow diagram of patients in the study. ‘Lost’: no notification from the physicians; ‘Discontinue’: the physicians discontinue administration of Lo/HCTZ; ‘Changes’: the physicians change from Lo/HCTZ to other antihypertensive drugs.

Table 1.

Characteristics of patients.

	Total	15,846	100%
Gender	Male	8430	53.2%
	Female	7416	46.8%
Age	≤64	6591	41.6%
	65−74	4468	28.2%
	≥75	4787	30.2%
	hypercholesterolemia	10,360	48.1%
	Obesity	2535	16.3%
	IHD	1251	7.9%
	LVH	728	4.6%
	Stroke	1505	9.6%

IHD, ischemic heart disease; LVH, left ventricular hypertrophy.

Hypercholesterolemia was the most frequent concomitant cardiovascular risk factor (48.1%) followed by obesity (16.3%). Associated clinical conditions including stroke (9.6%), ischemic heart disease (7.9%) and left ventricular hypertrophy (4.6%) were relatively less frequent.

All patients eligible for this study received one of the ARBs available in Japan. Physicians discontinued ARBs and started Lo/HCTZ at the start of the study. Half of the patients received ARBs as monotherapy and the others received combination therapy.

Changes in BP

Average BP reductions during 6 months are shown in Figures 2 –5. In Figure 2, changes in systolic BP (SBP) for all patients with (n = 10,376; 65.4%) or without (n = 5,479; 34.6%) complications are shown. SBP decreased at 1 month and gradually decreased further toward the end of the study (Figure 2).

Figure 2.

Prevalence of complications and comorbidities in participating patients with stratification of age.

Figure 3.

Levels of systolic blood pressure (SBP) in response to treatment with Lo/HCTZ with and without complications and/or comorbidities.

Figure 4.

(a) SBP levels in response to treatment with Lo/HCTZ in patients with hypercholesterolemia during the study. (b) SBP levels in response to treatment with Lo/HCTZ in patients with obesity during the study. (c) SBP levels in response to treatment with Lo/HCTZ in patients with stroke during the study. (d) SBP levels in response to treatment with Lo/HCTZ in patients with ischemic heart disease during the study. (e) SBP levels in response to treatment with Lo/HCTZ in patients with left ventricular hypertrophy during the study.

Figure 5.

Changes in prevalence (%) of target blood pressure <150 mmHg and <140 mmHg during the study in patients with age stratification.

At the start of the study, the levels of SBP were highest in the eldest patient group (≥75 years old) followed by the group aged 65–74 years and lowest in the youngest patient group (≤64 years old). However, at the end of the study, there were no significant differences among the three groups (Figure 3).

Changes in SBP over time in patients with various complications are shown in Figure 4. Similar trends in SBP changes are observed in all types of complication. In Figure 5, changes are shown in prevalence of the target BP during the study with stratification of patient age. If SBP levels were <150 mmHg, >90% of patients were able to achieve the target. In case of SBP <140 mmHg, >60% of patients were included.

Changes in laboratory findings

The laboratory findings related with various complications did not change during the study period (Table 2).

Table 2.

Changes in laboratory findings during the study period.

	0	1 month	2 months	3 months	6 months
Total cholesterol (mg/dl)	197.9 ± 35.2	194.2 ± 33.3	192.8 ± 33.5	192.1 ± 33.2	190.0 ± 32.0
Blood sugar (mg/dl)	119.0 ± 39.6	129.2 ± 47.2	125.6 ± 44.3	120.5 ± 39.2	117.2 ± 35.8
Uric acid (mg/dl)	5.6 ± 1.4	5.9 ± 1.4	5.9 ± 1.5	5.8 ± 1.4	5.8 ± 1.4
Serum creatinine (mg/dl)	0.84 ± 0.44	0.91 ± 0.54	0.92 ± 0.62	0.91 ± 0.5	0.89 ± 0.46
Serum potassium (mEq/l)	4.2 ± 0.9	4.3 ± 0.8	4.2 ± 0.5	4.1 ± 0.3	4.4 ±0.5

Adverse effects

Incidence of orthostatic hypotension was reported as 85 cases (1.7%) ≥75 years old, 67 cases (1.4%) 65–74 years old and 77 cases (1.1%) ≤64 years old (age stratification).

Discussion

This study using a website registration system revealed ‘real world’ management of hypertension during a 180-day follow-up period. First, apart from patients with DM and/or CKD, the most commonly complication was dyslipidemia followed by obesity. Second, compared with the prevalence of dyslipidemia, the prevalence of obesity significantly decreased with age. On the contrary, prevalence of hypertension-related target organ damage consistent with cerebrovascular and ischemic heart diseases significantly increased with age. These complications and comorbidities apparently reflect the ‘real world’ situation of patients. In the patient ‘real world’, Lo/HCTZ reduced SBP to a similar level in all age groups in spite of various complications and comorbidities.

Previously reported landmark hypertension trials have focused on patients with special complications and/or comorbidities or age profiles [Denker and Cohen, 2013]. In the present study, the effects of Lo/HCTZ on SBP of patients with a relatively wide age range were found to be similar at any age or complication.

Several studies have reported the achievement rate of target BP in elderly hypertensive patients. Saito and colleagues [2011] demonstrated that the BP target of <140/90 mm Hg was achieved in a total of 8897 patients in the Japanese Benidipine Research on Antihypertensive Effects in the Elderly (J-BRAVE) study [Saito et al. 2011] which comprised 57.5% of patients aged 65–74 years and 56.6% aged ≥75 years. In the Japanese Trial to Assess Optimal Systolic Blood Pressure in Elderly Hypertensive Patients (JATOS) [JATOS Study Group 2008; Kawano et al. 2011], strict treatment to a target BP <140 mm Hg showed no difference in the composite endpoint of CVD or renal failure, although there was a nonsignificant trend towards a higher all-cause mortality rate. Similar findings was reported in a study of over 3000 elderly hypertensive patients where strict control (target SBP <140 mm Hg) was not superior to moderate SBP control (target SBP 141–149 mm Hg) in reducing incidence of a composite cardiovascular endpoint [Okin et al. 2012].

In the Japan Hypertension Evaluation with Angiotensin II Antagonist Losartan Therapy (J-HEALTH) study [Shimamoto et al. 2008], BP was controlled at <140/90 mm Hg in 46% of patients. Kushiro and colleagues showed that administration of olmesartan in elderly hypertensive patients resulted in control of BP at <140/90 mm Hg in 49.8% [Kushiro et al. 2009]. In achievement of target BP levels, the present data are similar or superior to the previously reported data.

In the present study, when subgroups of hypertensive patients with several complications and comorbidities were analysed, no significant difference was found in BP reduction. This may be due to the broad spectrum of Lo/HTCZ actions. When effects of Lo/HTCZ on laboratory data including levels of uric acid, creatinine, blood glucose, total cholesterol and potassium were followed up, there were no significant alterations during the study period. These data illustrate the usefulness of Lo/HTCZ in ‘real world’ management of ISH.

Two reports demonstrated that serum uric acid increased in patients with a normal baseline uric acid but decreased in patients with hyperuricemia after they switched to Lo/HCTZ [Kita et al. 2010; Hosoya et al. 2012]. This study confirmed these data in relatively large numbers of patients.

In addition, total cholesterol levels were not changed during the study period, a finding already reported by several investigators [Hirose et al. 2011; Racine et al. 2010; Ibuki et al. 2014].

Furthermore, past history of CVD also did not affect the BP lowering action of Lo/HTCZ. In a post hoc analysis of the Systolic Hypertension in the Elderly Program (SHEP) that investigated various levels of in-trial BP and risk of stroke, participants with SBP <150 mmHg experienced a 38% reduction in stroke incidence compared with those with SBP >150 mmHg, while those with those with SBP <140 mmHg had a 22% risk reduction that was not statistically significant [Perry et al. 2000]. These findings not only suggest that BP reduction in the elderly is effective, but also that there may be a moderate BP threshold below which there is no additional benefit. In analysis of the present data, BP <150mmHg was achieved in 90% of patients with Lo/HTCZ in patients with stroke and ischemic heart disease, indicating that Lo/HTCZ was very effective in those patients. A recent meta-analysis revealed that all major classes of BP lowering agents may diminish recurrent stroke risk and that the degree of BP reduction may be more important than the class of agent used [Lu et al. 2009]. In the present study, Lo/HTCZ reduced SBP effectively, confirming that Lo/HTCZ is a candidate drug combination of choice for prevention of recurrent stroke.

Lo/HTCZ is well known to reduce LVH in hypertensive patients, although in the present study, regression of LVH was not evaluated. Regression of LVH has been reported in parallel with reduction of BP in the case of Lo/HTCZ, which is promising for patients with LVH, even for those who are elderly.

Finally, the reported incidence of adverse effects including orthostatic hypotension and excessive reduction of BP was unexpectedly lower compared with previous data. There are several possible reasons: (1) participating physicians did not hesitate to change and/or withdraw Lo/HTCZ, because there were no strict regulations in this study; (2) Lo/HTCZ does not have a sufficiently strong antihypertensive action to reduce BP dramatically; and (3) Lo/HTCZ is an adequate drug combination for management of ISH, as our previous data clearly showed its antihypertensive action in reducing pulse pressure [Suzuki et al. 2012].

Limitations

The lack of a case control (placebo) design in the analysis may have partly affected our findings, particularly in relation to safety and tolerability. The large sample size and distribution of registered physicians may have meant that the views expressed by respondents may not be fully representative of behaviours of the entire physician community. The requirement for electronic support may have also meant that only those physicians with easy access to the web database were included, rather than all physicians. This was a fully observational study, and during the course of the follow-up period, a large number of patients did not return to the prescribing physicians. For this reason, the number of patients with available data at follow-up visits is limited. In addition, because all patients had been treated with antihypertensive regimen including one of ARBs, the depressor effects observed in this study cannot completely be attributable to the ability of Lo/HCTZ. Finally, because many factors influence a physician’s prescribing, our current finding on the use of Lo/HCTZ cannot be extrapolated to the general hypertensive population.

Conclusion

In all considered subgroups of hypertensive patients including those with dyslipidemia, obesity and hypertension-related target organ damage consistent with ischemic heart disease, cerebrovascular diseases and left ventricular hypertrophy, the changes from ARBs alone to Lo/HTCZ were effective in BP reduction.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

Conflict of interest

The authors declare no conflicts of interest in preparing this article.

References

Chaudhry

Chavez

Gasowski

Grodzicki

Messerli

(2012) Hypertension in the elderly: some practical considerations. Cleveland Clinic J Med 79: 694–704.

Denker

Cohen

(2013) What is an appropriate blood pressure goal for the elderly: review of recent studies and practical recommendations. Clin Interv Aging 8: 1505–1517.

Hirose

Kawabe

Saito

(2011) Effects of losartan/hydrochlorothiazide treatment, after change from ARB at usual dosage, on blood pressure and various metabolic parameters including high-molecular weight adiponectin in Japanese male hypertensive subjects. Clin Exp Hypertens 33: 41–46.

Hosoya

Kuriyama

Yoshizawa

Kobayashi

Otsuka

Ohno

(2012) Effects of combined antihypertensive therapy with losartan/hydrochlorothiazide on uric acid metabolism. Intern Med 51: 2509–2514.

Ibuki

Seino

Otsuka

Mizuno

(2014) The fixed-dose combination of losartan/hydrochlorothiazide elicits potent blood pressure lowering during nighttime in obese hypertensive patients. J Clin Med Res 6: 8–16.

JATOS Study Group (2008) Principal results of the Japanese trial to assess optimal systolic blood pressure in elderly hypertensive patients (JATOS). Hypertens Res 31: 2115–2127.

Kawano

Ogihara

Saruta

Goto

Ishii

(2011) Association of blood pressure control and metabolic syndrome with cardiovascular risk in elderly Japanese: JATOS study. Am J Hypertens 24: 1250–1256.

Khanna

Lefkowitz

White

(2008) Evaluation of recent fixed-dose combination therapies in the management of hypertension. Curr Opin Nephrol Hypertens 17: 477–483.

Kita

Yokota

Ichiki

Ayabe

Etoh

Kitamura

(2010) One-year effectiveness and safety of open-label losartan/hydrochlorothiazide combination therapy in Japanese patients with hypertension uncontrolled with ARBs or ACE inhibitors. Hypertens Res 33: 320–325.

10.

Kotchen

(2014) Developing hypertension guidelines: an evolving process. Am J Hypertens 27: 765–772.

11.

Kushiro

Saito

Hirata

Ishikawa

Yamashita

. (2009) Blood pressure-lowering effects of angiotensin receptor antagonist monotherapy and in combination with other anti-hypertensive drugs in primary care settings in Japan. Clin Exp Hypertens 31: 127–141.

12.

Cheng

Zhu

Shen

(2009) A systematic review of angiotensin receptor blockers in preventing stroke. Stroke 40: 3876–3878.

13.

MacKay

Arcuri

Goldberg

Snapinn

Sweet

(1996) Losartan and low-dose hydrochlorothiazide in patients with essential hypertension. A double-blind, placebo-controlled trial of concomitant administration compared with individual components. Arch Intern Med 156: 278–285.

14.

Ogihara

Kikuchi

Matsuoka

Fujita

Higaki

Horiuchi

. (2009) The Japanese Society of Hypertension guidelines for the management of hypertension (JSH 2009). Hypertens Res 32: 3–107.

15.

Ogihara

Ueshima

Nakao

Fukiyama

Oba

. (2011) Long-term effects of candesartan and amlodipine on cardiovascular morbidity and mortality in Japanese high-risk hypertensive patients: the Candesartan Antihypertensive Survival Evaluation in Japan Extension Study (CASE-J Ex). Hypertens Res 34: 1295–1301.

16.

Okin

Hille

Kjeldsen

Dahlof

Devereux

(2012) Impact of lower achieved blood pressure on outcomes in hypertensive patients. J Hypertens 30: 802–810; discussion 10.

17.

Perry

HM.

Jr. Davis

BR.

Price

TR.

Applegate

WB.

Fields

WS.

Guralnik

JM.

. (2000) Effect of treating isolated systolic hypertension on the risk of developing various types and subtypes of stroke: the Systolic Hypertension in the Elderly Program (SHEP). JAMA 284: 465–471.

18.

Racine

Hamet

Sampalis

Longo

Bastien

(2010) A 52-week prospective, cohort study of the effects of losartan with or without hydrochlorothiazide (HCTZ) in hypertensive patients with metabolic syndrome. J Hum Hypertens 24: 739–748.

19.

Riva

Lip

(2013) Which is the optimal antihypertensive combination in different diseases, a renin- angiotensin-aldosterone system inhibitor with a diuretic or with a calcium channel blocker? Curr Pharm Des 19: 3753–3765.

20.

Ruilope

(2012) Current challenges in the clinical management of hypertension. Nat Rev Cardiol 9: 267–275.

21.

Saito

Suzuki

Kageyama

Saruta

(2011) Treatment of hypertension in patients 85 years of age or older: a J-BRAVE substudy. Clin Exp Hypertens 33: 275–280.

22.

Shimamoto

Fujita

Ito

Naritomi

Ogihara

T., Shimada, K.

. (2008) Impact of blood pressure control on cardiovascular events in 26,512 Japanese hypertensive patients: the Japan Hypertension Evaluation with Angiotensin II Antagonist Losartan Therapy (J-HEALTH) study, a prospective nationwide observational study. Hypertens Res 31: 469–478.

23.

Suzuki

Shimada

Fujiwara

(2012) Combination therapy with losartan/hydrochlorothiazide for blood pressure reduction and goal attainment in a real-world clinical setting in Japan. Ther Adv Cardiovasc Dis 6: 237–244.

24.

Turgut

Yesil

Balogun

Abdel-Rahman

(2013) Hypertension in the elderly: unique challenges and management. Clin Geriatr Med 29: 593–609.

25.

Weycker

Edelsberg

Vincze

Kjeldsen

Jamerson

Khan

ZM.

. (2007) Risk of diabetes in a real-world setting among patients initiating antihypertensive therapy with valsartan or amlodipine. J Hum Hypertens 21: 374–380.

26.

Wile

(2012) Diuretics: a review. Ann Clin Biochem 49: 419–431.