Glucometabolic effects of antihypertensive drugs:
systematic review and meta-‐analysis
By: Medical student Jenny Gustafsson Supervisor: Associate professor Johan Sundström Uppsala University Medical School Uppsala, Sweden 2011-10-25
2
ABSTRACT
Title: Glucometabolic effects of antihypertensive drugs: systematic review and meta-analysis Author: Jenny Gustafsson
Objective: To quantify the effects of antihypertensive treatment with angiotensin converting- enzyme (ACE)-inhibitors, angiotensin receptor-blockers (ARB)s, alpha 2-agonists, alpha- blockers, beta-blockers, calcium antagonists and thiazides on insulin sensitivity.
Background: Previous studies have indicated that antihypertensive drugs can affect insulin sensitivity. It has been suggested that beta-blockers and thiazides worsen insulin sensitivity, that calcium antagonists, ARBs and ACE-inhibitors on average are neutral, and that alpha- blockers may improve insulin sensitivity.
Methods: We searched the databases PubMed, Scopus and reference lists of published articles. The data were extracted and summarized in a meta-analysis.
Result: We identified 103 relevant clinical trials with 207 study arms, and investigated treatment effects on systolic and diastolic blood pressures and the metabolic parameters insulin clamp glucose disposal rate, homeostasis model of insulin resistance (HOMA-IR), fasting plasma glucose, oral glucose tolerance test (OGTT) 2-hour glucose, and intravenous glucose tolerance test (IVGTT) k-value. Clamp glucose disposal rate increased with ACE- inhibitor treatment (12.7%; 95% confidence interval 6.2 to 19.2%) and ARB treatment (4.4%;
1.8 to 7.0%); was largely neutral with beta-blocker (6.1%; -7.7 to 19.9%) and calcium antagonist treatment (6.5%; -26.5 to 39.5%); and diminished with thiazides (-10.9%; 0.5 to - 22.4%). Similarly, HOMA-IR was reduced with ACE-inhibitor (-14.9%; -3.6 to -26.2%), ARB (-7.4%; 0.5 to -15.4%) calcium antagonist (-16.0%; -4.2 to -27.8%), and alpha blocker treatment (-8.2%; -1.0 to -15.5%). HOMA-IR was largely unaffected with beta-blocker treatment (0.2%; -9.2 to 9.5%). Thiazides increased HOMA-IR (13.4%; 4.1 to 22.6%). Blood pressure was reduced similarly in all antihypertensive treatment groups.
Conclusion: Treatment with ACE-inhibitors, ARBs, calcium antagonists, and alpha-blockers improved insulin sensitivity on average, treatment with beta-blockers had a neutral effect on insulin sensitivity, and treatment with thiazides worsened insulin sensitivity.
3
GLUCOMETABOLIC EFFECTS OF ANTIHYPERTENSIVE DRUGS: SYSTEMATIC REVIEW AND META-‐ANALYSIS
ABSTRACT 2
BACKGROUND 4
The interplay between hypertension and diabetes 4
Antihypertensive drug classes may have different effects on diabetes risk 5
Angiotensin converting-‐enzyme-‐inhibitors 5
Angiotensin receptor-‐blockers 5
Beta-‐blockers 5
Calcium antagonist 6
Thiazide diuretics 7
Alpha-‐blockers 7
Hypotheses and aims 8
METHODS 9
Eligibility criteria 9
Information sources 9
Literature search 9
Measures of insulin sensitivity 11
Other variables 12
Statistical analyses 12
RESULTS 13
Angiotensin converting enzyme inhibitors 13
Angiotensin receptor-‐blockers 13
Beta-‐blockers 14
Calcium antagonists 14
Thiazides 14
Other antihypertensive drugs 15
Associations of change in blood pressures with change in insulin sensitivity measures 16
DISCUSSION 17
Summary of findings 17
Divergent guidelines for patients with diabetes and hypertension 17
Comparisons with previous studies of diabetes incidence 18
Strength and Limitations 20
Conclusion 21
REFERENCES 22
Table 1. Baseline characteristics of included studies 30
FIGURES 37
4
BACKGROUND
In Sweden have 27% among the adults hypertension and at least half of all patients with diabetes type 2 have hypertension 1. Hypertension is equally common between both sexes and is a risk factor for cardiovascular disease. It is important not only to take in consideration of the grades of hypertension but also cardiovascular risk as other risk factors and disease 2, as metabolic syndrome which including obesity, insulin resistance, dyslipedemia and
hypertension3. It has been showed that insulin sensitivity can be affected by antihypertensive drugs. In which degree insulin sensitivity is affected is shown to depend on which
antihypertensive drug is used. This is an important factor to consider when it becomes
necessary to treat patients for hypertension. Previous studies have also demonstrated that with good blood pressure control can the risk of early death and cardiovascular disease reduce4.
The interplay between hypertension and diabetes
Cardiovascular disease as stroke and coronary heart disease is one of the most common causes to early death in many parts of the western world 3. Known risk factors to development cardiovascular disease is hereditary, smoking tobacco, hypertension, dyslipedemia,
overweight and insulin resistance.
Hypertension is related to the risk of diabetes mellitus type 2 and increase the risk to complications as stroke, coronary heart disease, retinopathy, and nephropathy3. It is also found that hypertensive patients have impaired glucose tolerance5.
Diabetes mellitus is a common disease 6. Because the different effect that the antihypertensive drugs have on glucose metabolism it is important to know which antihypertensive drugs that further increase the risk to effect insulin sensitivity in a negative way. For example have studies by Lithell et al, shown that it is a possibility that beta-blockers and diuretics increase the risk to development diabetes mellitus 7.
Previous studies have shown that different kind of antihypertensive drugs effect insulin sensitivity in different degrees5,8,9, 10. It is suggested that beta-blocker and diuretics have a negative effect on insulin sensitivity, calcium antagonist and ACE-inhibitor are more neutral and alpha blocker believe improve insulin sensitivity. The use of antihypertensive drugs is
5
believed to increase the risk for metabolic syndrome by either promoting an increase in weight or changes in lipid or glucose metabolism 9.
Antihypertensive drug classes may have different effects on diabetes risk Angiotensin converting-‐enzyme-‐inhibitors
ACE-inhibitor is believed to have a more neutral effect on insulin sensitivity in hypertensive patients and have a decreased prevalence in new-onset diabetes9.
Angiotensin receptor-‐blockers
ARB has not been found to have negative effect in lipid or glucose metabolism59. Therefore is ARB a good therapy for patients with risk factors as metabolic syndrome. The Losartan intervention for Endpoints Reduction in hypertension study, (the LIFE study) 4 is a dubbel- blind prospective parallel group study that compares the effect of losartan (ARB) with beta- blocker (atenolol), if needed were thiazide added for blood pressure control. The study evaluated the antihypertensive drugs and their effect on cardiovascular morbidity and mortality of patients with ventricle hypertrofi. The mean follow up was 4,8 years. It was shown that there was 25% lower new-onset diabetes in the losartan group compared with atenolol group. The losartan group was better then atenolol group in reducing cardiovascular death, mainly because of decreased risk of stroke. Atenolol didn’t differ from losartan when incident of myocardial infarction was compared. The blood pressure was reduced in both groups4. In the CHARM 10 (Candesartan in heart failure assessment of reduction in mortality and morbidity) study was it shown that the incident in candesartan treatment group of new- onset diabetes was 19% lower compared with those that were without treatment. In the VALUE 10 (Valsartan Antihypertensive long term use evolutions) study was it found that the valsartan treatment compared with amlodipine group had a 23% lower incident in new-onset diabetes.
Beta-‐blockers
Beta-blocker has been shown to reduce cardiovascular disease and is one of the first choice treatments against hypertension. But the use of beta-blockers seems also to enhance the risk of Diabetes mellitus type 279. In a few studies it has been reported that beta-blockers is believed to promoting weight gain that can effect glucose metabolism59. Overweight can lead to hypertension, insulin resistance and lipid disorder. Hyperglycaemia is believed to be due because of decreased in insulin from pancreatic B-cells. Other explanations are changes in insulin clearance and secretion and decreased peripheral blood flow.
6
Previous studies by Lithell et al, have shown that different kind of beta-blockers effects insulin sensitivity in different degree, for example propranolol, metoprolol and atenolol reduce insulin sensitivity 15-30% and celiprilol showed to increase insulin sensibility of 35%7,8. GEMINI 10 (Glycemic effects in diabetes mellitus: Carvedilol-Metoprolol comparison I hypertensives) is a study that compared metoprolol with carvedilol in hypertensive, diabetic patients that already have treatment with ACE-inhibitor or a ARB. It was found a reduction in HOMA-IR with 9% with carvedilol treatment.
ASCOT 11 (prevention of cardiovascular events with an antihypertensive regime of amlidopine adding perindopril as required versus atenolol adding bendroflumethiazide as required , in the Anglo-Scandinavian Cardiac Outcomes Trial Blood Pressure Lowering arm:
(ASCOT-BPLA)is mulitcenter randomised controlled trial) is a study with a median follow up of 5,5 years. The amlidopine regime (calcium antagonist) prevented more cardiovascular events; especially in combination with lipid lowering treatment compared with atenolol regime (beta-blocker).The incident of diabetes was less in the amlidopine group, 30%, compared with atenolol group. In both groups did mean blood pressure fell after initiation of the antihypertensive drugs.
Calcium antagonists
It is shown that calcium antagonist have no negative effects on glucose or insulin parameters in patients with diabetes or hypertension 9, and it is found that calcium antagonist have positive effects on insulin secretion 5. The INVEST study 12 (A Calcium antagonist vs a Non- Calcium antagonist Hypertension Treatment Strategy for Patient With Coronary Artery Disease) compared calcium antagonist (verapamil), if needed with a combination of ACE- inhibitor, with no calcium antagonist strategy (atenolol), if needed in combination with
thiazides, in a randomized, controlled study, mean follow-up was 2,7 years. The study showed a lower incident in new-onset diabetes (15%) and reduced insulin resistance in the calcium antagonist group compared to beta-blocker group. In the INSIGHT study 10 (Intevention as a goal in hypertension treatment) was a randomized study that compared the nifedipine
treatment with hydrochlorothiazide plus amiloride treatment. It was found that after a follow up of 4 years that the nifedipine treatment had a lower incident in new-onset diabetes
compared with hydrochlorothiazide plus amiloride treatment. However in the STOP- Hypertension 2 trial 10 (Swedish trial in old patient with hypertension) was it found that
7
treatment with beta-blocker or diuretic compared with ACE-inhibitors or calcium antagonist that their were no difference in new-onset diabetes between the groups after a follow up time after 4,5 years.
Thiazide diuretics
Thiazide diuretics are one of the most common medications for patients with hypertension 9. Thiazide diuretics are believed to worsen insulin sensitivity and it is shown that thiazides also can affect the risk to development diabetes mellitus, the mechanism is not clear9. But it is believed to causes by reduced insulin secretion and peripheral insulin sensitivity 5.
ALLHAT 1,13 (Antihypertensive and Lipid Lowering Treatment to Prevent Heart Attack Trial) is a randomized, controlled clinical trial. ALLHAT is a study with hypertensive patients and at least one more risk factor that was randomization to treatment with chlorthalidone,
amlodipine (calcium antagonist) or lisinopril (ACE-inhibitor). The mean follow up was 4,9 years. In ALLHAT913 is it shown that the incident of new-onset diabetes was highest for subjects randomized to diuretic-based strategy compared with the calcium antagonist and ACE-inhibitor treatment. ALLHAT13 also showed that thiazide diuretic and calcium antagonist had reduced incident in stroke and cardiovascular disease compared with ACE- inhibitor. Chlorthalidone was better then lisinopril on reducing blood pressure. Therefore are thiazide diuretics still an important medication in treatment of hypertension.
The ALPINE 9,14study, (Antihypertensive treatment and Lipid Profile in a North of Sweden Efficacy evaluation) was a long-term randomized parallel group study compared thiazides alone or in combination with atenolol (beta-blocker) with ARB (candersatan) alone or in combination with calcium antagonist (felodipine). ALPINE found out that thiazide group had higher fasting insulin and plasma glucose values compared with ARB group. When the study evaluated HOMA-IR it was indicating increased insulin resistance in the thiazide group.
ALPINE also found that there was a higher risk to development metabolic syndrome with thiazide/ beta-blocker treatment compared with an ARB/calcium antagonist treatment.
In the HAPPHY (Heart attack primary prevention in hypertension) study was it shown that the incident of new-onset diabetes between thiazide diuretic and beta-blocker was the same 10.
Alpha-‐blockers
Alpha-blockers have a neutral effect on insulin sensitivity in hypertensive patients 5. In a study by Pollare et al, is it shown that insulin sensitivity was improved during an i.v glucose
8
tolerance test15. In the review by Sarafidis et al, was it reported that doxazosin treatment in patients with hypertension increased clamp insulin sensitivity with 21%10. It was also reported that in another study was it shown that treatment with moxonidine increased insulin
sensitivity in a clamp with 10%.
It is important to have a careful attention to antihypertensive drugs choice in all patients but especially in overweight patients or if they have other risk factors 3. Especially important is it when the number of patients with metabolic syndrome is still increasing. A future perspective is that the children that have overweight increases which can lead to adult obesity and risk to development metabolic syndrome, which in turn will lead to a higher risk to development cardiovascular diseases as adults. Therefore it is important to take a good consideration which antihypertensive medication that is the most appropriate treatment for these patients.
Hypotheses and aims
Several smaller studies of effects of antihypertensive drugs on insulin sensitivity have been conducted, but these have never been summarized in a meta-analysis before. Because clinical hypertension guidelines advice about drug choices partly based on such studies, summarizing the evidence is of importance.
We aimed to investigate effects of different antihypertensive drug classes on measures of insulin sensitivity. Secondary aims included investigating effects of monotherapy and combination therapy with the selected drug classes, and meta-regression of effects on blood pressure with effects on insulin sensitivity.
9
METHODS
Eligibility criteria
The inclusion criteria were the following:
Randomized/non randomized clinical trials of antihypertensive drugs.
Participants at least 18 years old.
Only hypertensive subjects studied.
Treatment duration of at least four weeks.
At least seven participant per study arm.
Evaluation at baseline and follow-up of glucose metabolism with either insulin clamp, homeostasis model assessment of insulin resistance (HOMA-IR), oral glucose tolerance test (OGTT), intravenous glucose tolerance test (IVGTT), or fasting plasma insulin and glucose (enabling the calculation of HOMA-IR).
English language.
No limitations regarding the publication date or if the trial was blinded or not.
Controlled parallel-goup trials, cross-over trials, and uncontrolled studies were included. Both monotherapy and combination therapy trials were included. Participants were allowed
background antihypertensive drug treatment. Both blinded and open label studies were
included. Exclusion criteria were insufficient or absent reporting of inclusion criteria or effect measures.
Information sources
Studies were identified by searching electronic databases and scanning reference lists of reviews and original articles. The following search was applied to PubMed and Scopus. The last search was run in last May 2011.
Literature search
The following search terms were used in PubMed and Scopus, with the limits human and English language:
- Glucose metabolism AND "Antihypertensive Agents"[majr]
- ("Insulin Resistance"[majr] AND "Antihypertensive Agents"[majr]) OR ("Glucose/metabolism"[Majr] AND "antihypertensive agents"[majr])
10 Identification
Screening
Eligibility
Included
In total, 103 articles were included in the meta-analysis. No variables were added after the project started. The data extracted from the articles sometimes needed some processing before they were suitable for presentation. For example was it sometime necessary to convert units of glucose (from mg/dl to mmol/l) or insulin (from pmol/l to mU/l) (e.g. http://www.soc-
dr.org/rds/authors/unit_tables_conversions_and_genetic_dictionaries/e5184/index_en.html).
Some studies have used blood glucose, which we have converted to plasma glucose using the formula plasma glucose = blood glucose * 1.12.
No of records identified through PubMed database searching: 655.
No of additional records identified in the Scopus database: 712.
No of records after duplicates removed: (1367-19) 1348
No of Records screened:1348 No of records excluded: 1207
No of full text-articels assessed for eligibility: 141
No of full-text articles excluded n:38.
Reasons: Normotensive: 5; Lacking baseline or final values of glucometabolic parameters or blood
pressures: 17; Ambiguous data reported: 6; Reviews: 5; Studies
shorter than 4 w: 2; No antihypertensive drug: 1;
Same study arm in two different articles:2
No of studies included in qualitative synthesis: 103
No of studies included in quantitative synthesis (meta-analyses): 103
11 Measures of insulin sensitivity
The homeostasis model assessment of insulin resistance (HOMA-IR) is an established
measurement to evaluate insulin sensitivity. HOMA-IR was calculated as HOMA-IR=fasting glucose (mmol/l)*fasting insulin (mU/l) divided by 22.5. We calculated 109 HOMA-IR values from reported glucose and insulin values, and used 64 reported HOMA-IR values.
The euglycemic,hyperinsulinaemic clamp is the gold standard method for measuring insulin sensitivity and insulin secretion. The most common technique is the one reported by
DeFronzo et al. 1979 1617. After an overnight fast, blood is collected for determination of fasting glucose and insulin. The study participants get a booster of insulin in 6 min. After that, an i.v. insulin infusion is given at a rate of 56 mU/m2 body surface for 110 min (the duration of the clamp and the concentration of insulin can vary depending on technique used). During the insulin infusion, a 20% glucose solution was given so euglycemia could be maintained, i.e. plasma glucose is held constant and the glucose infusion rate is an index of glucose metabolism. Blood samples were taken regularly at 5 min intervals. Clamp values that were used in this study were whole body glucose disposal rate (M=mg/kg body weight/min) and insulin sensitivity index was calculated by dividing glucose disposal rate by the mean insulin concentration (M/I) or by the insulin concentration and the mean glucose concentration (M/G/I). The values used were calculated during the steady state phase of euglycemia when
“the glucose uptake by all the tissue in the body is equal with glucose infusion rate and are therefore a measurements of tissue sensitivity to exogenus insulin” 16. This steady state phase is during the last 60 or 30 min of the clamp. Approximately 67 M-values were included in this meta-analyses, 45 M/I- values and 9- M/G/I-values.
Oral glucose tolerance tests (OGTT) were performed in a standardized fashion, after 12 hours of fasting19. Blood samples as glucose and insulin was taken at baseline and after 120 min after the glucose load of 75 g glucose in 250 ml water. In this study, 37 OGTT-2h-glucose values were included.
An intravenous glucose tolerance test (IVGTT) is done by injecting 300 mg/kg body weight of 50% glucose solution during 1.5 min18. The disappearance rate of glucose is expressed as the K-value calculated from the formula K=(log2 x 100)/t1/2 where t1/2 is the time in min required for the glucose concentration to be halved. In this meta-analysis, 27 K-values were used.
12 Other variables
Data for final body mass index (Tabel 1) (body weight in kilograms divided by height in meters2) were not always reported in the articles. Sometimes articles stated that there were no significant different from baseline to follow up without mentioning a specific BMI value.
Lifestyle advice was in some cases given to the study population in the trials (Table 1).They had diet restriction or physical activities during the time of the study.
The articles that account for the placebo run in period (Tabel 1) are shown. It was not
however a claim on our part. Placebo run in period that is shown is the minimal weeks that are reported in the trials. We have also accounted for how many precipitancies that have diabetes mellitus (Tabel 1), this is done because it can affect the result of the glucose variables.
A minority of the articles did only shown the hole study group number and not the separately number for the group with the different drugs, if so was done we choose to write the hole study group number for the respective drug.
Some of the trials who were evaluated one type of antihypertensive drug added one more drug in the end of follow-up period because of difficulties to adjust the blood pressure. They did not evaluate both drugs separately so we have shown that data as a combination therapy.
Statistical analyses
Baseline characteristics of included studies are reported as percent or means (standard deviations). We used inverse variance-weighted random effects meta-analysis models for all analyses. Pre-specified subgroup analyses of study arms with monotherapy and combination therapy were investigated. We investigated associations of changes in systolic and diastolic blood pressures with changes in clamp glucose disposal rate, HOMA-IR, and FPG using fixed effects meta-regression. Estimates and 95% confidence intervals are given for all meta-
analyses, and β-coefficients and p-values for the meta-regressions. Two-tailed 95%
significance levels were used throughout. We used the statistical package Comprehensive Meta-Analysis V2.
13
RESULTS
A total of 103 studies with 207 study arms (Table 1) met the inclusion criteria of this meta- analysis.
Angiotensin converting enzyme inhibitors
ACE-inhibitors in monotherapy reduced systolic and diastolic blood pressures (Figures 1 and 2) (-11.6%; 95% confidence interval -9.5 to -13.7%; and -12.4%; -9.8 to -15.1%;
respectively). ACE-inhibitors in combination with other antihypertensive drugs lowered systolic and diastolic blood pressures to a similar extent (Figures 1 and 2) (-14.5%; -11.5 to 17.5%; and -12.7%; -10.0 to -15.3%). Clamp glucose disposal rate (Figure 3) was increased with ACE-inhibitor monotherapy (12.7%; 6.2 to 19.2%) and with ACE-inhibitors as part of combination therapy (18.3%; -30.5 to 67.1%), although the latter was estimated with lower precision. HOMA-IR (Figure 4) decreased (-14.9%; -3.6 to -26.2%) with ACE-inihibitor monotherapy; but was on average unaffected by the studied combination therapies including ACE-inhibitors (Figure 4) (5.2%; -25.9 to 36.2%). FPG (Figure 5) was slightly lowered (- 2.9%; -1.4 to -4.4%) by ACE-inhibitor monotherapy; but was unaffected by combination therapy including ACE-inhibitors (0.8%; -2.0 to 3.5%). ACE-inhibitor monotherapy reduced OGTT-2h-glucose (Figure 6) (-8.1%; -2.4 to -13.8%) and if anything increased intravenous glucose tolerance test (IVGTT) k-value (Figure 7) (9.1%; -15.8 to 33.9%).
Angiotensin receptor-‐blockers
ARB monotherapy reduced systolic and diastolic blood pressures (Figures 8 and 9) (-11.0%; - 9.4 to -12.6%; and -10.4%; -9.1 to -11.7%; respectively), as did combination therapies
including ARBs (Figure 8 and 9) (-12.8%; -10.5 to -15.2%; and -13.0%; -10.9 to -15.0%;
respectively). Clamp glucose disposal rate (Figure 10) was increased by ARB monotherapy (4.4%; 1.8 to 7.0%), but was on average unaffected by ARBs in combination with other drugs (-1.8%; 0.6 to -4.3%). HOMA-IR (Figure 11) tended to be reduced by ARB monotherapy (- 7.4%; 0.5 to -15.4%) and combination therapy (-3.4%; 8.4 to -15.2%), but confidence
intervals were wide. FPG (Figure 12) was unaffected by ARB monotherapy and combination therapy (-0.8%; 0.6 to -2.3%; and -0.1 %; 2.4 to -2.8%; respectively), as was OGTT by ARB monotherapy (Figure 6) (4.2%; -8.3 to 16.7%).
14 Beta-‐blockers
Beta-blockers reduced systolic and diastolic blood pressures to a similar degree in monotherapy (Figures 13 and 14) (-8.7%; -5.8 to -11.6%; and -8.8; -5.4 to -12.1%;
respectively) and slightly more in combination with other drugs (Figures 13 and 14) (-12.0%;
-10.3 to -13.8%; and -12.0%; -10.8 to -13.2%; respectively). Beta-blockers had a neutral effect on clamp glucose disposal rate (Figure 15) in monotherapy (6.1%; -7.7 to 19.9%) as well as in combination with other antihypertensive drugs (0.9%; -32.8 to 34.6%). Similarly, the effect of beta-blockers on HOMA-IR (Figure 16) was neutral with monotherapy (0.2%; - 9.2 to 9.5) and as part of combination therapy (2.5%; -20.1 to 25.0). Effects on FPG (Figure 17) were also neutral with monotherapy (0.3%; -1.2 to 1.8%) and combination therapy (- 0.7%; 4.6 to -6.1%). Beta-blockers in monotherapy had a largely neutral effect on OGTT-2h- glucose (Figure 6) (4.1%; -11.8 to 20.1%), but reduced IVGTT k-value (Figure 7) (-13.8%; - 2.9 to -24.7).
Calcium antagonists
Calcium antagonists reduced systolic and diastolic blood pressures (Figures 18 and 19) in monotherapy (-11.8%; -9.2 to -14.5%; and -12.7%; -9.4 to -16.0%; respectively) and as part of combination therapy (-11.5; -9.9 to -13.0%; and -11.2%; -8.9 to -13.6%; respectively).
Calcium antagonists tended to increase clamp glucose disposal rate (Figure 20) (6.5%; -26.5 to 39.5% in monotherapy; and 25.2%; -19.5 to 70.0% in combination with other drugs), but confidence intervals were wide. A more consistent improvement in HOMA-IR was observed (Figure 21) for calcium antagonists in monotherapy (-16.0%; -4.2 to -27.8%) and in
combination with other antihypertensive drugs (-20.7%; -9.7 to -31.6%). FPG (Figure 22) was also lowered by monotherapy (-4.4%; -0.6 to -8.3%); but was largely unaffected by
combination therapy (-0.9%; 3.6 to -5.4). OGTT-2h-glucose (Figure 6) was not altered with calcium antagonists in monotherapy (-1.7%; 4.9 to -8.3%).
Thiazides
Thiazides reduced systolic and diastolic blood pressure (Figures 23 and 24) in monotherapy (- 8.6%; -6.7 to -10.5%; and -8.1%; -4.5 to -11.7%; respectively) and as part of combination therapy (-12.8%; -11.5 to -14.0%; and -12.4; -11.0 to -13.8%; respectively). Clamp glucose disposal rate (Figure 25) appeared to be reduced by thiazides in monotherapy (-10.9%; 0.5 to - 22.4%), and if anything less so with combination therapy (-1.8%; 0.6 to -4.2%). Thiazides increased HOMA-IR (Figure 26) in monotherapy (13.4%; 4.1 to 22.6%) and in combination with other antihypertensive drugs (8.6%; 5.1 to12.1%). Thiazides also increased FPG (Figure
15
27) in monotherapy (6.9%; 3.4 to 10.4%) and as part of combination therapy (2.3%; 0.4 to 4.2%). Thiazides tended to reduce IVGTT k-value (Figure 7) (-8.3%; 11.2 to -27.9%), but confidence intervals were wide.
Other antihypertensive drugs
Alpha 2-agonists in monotherapy reduced systolic and diastolic blood pressures (Figures 28 and 29) (-12.1%; -6.9 to -12.2%; and -12.4%; -5.6 to -19.1%; respectively). Available studies indicated no certain effect on FPG (Figure 31) (7.5%; -3.1 to 18.1%) or OGTT-2h-glucose (Figure 6) (2.7%; -20.4 to 25.8%), although point estimates were slightly increased.
Alpha-blockers reduced systolic and diastolic blood pressures (Figures 28 and 29) (-9.5%; - 6.9 to -12.2%; and -9.4%; -6.5 to -12.4%; respectively). Effects on clamp glucose disposal rate (Figure 30) were largely neutral (-0.3%; 12.5 to -13.0), but alpha-blockers lowered HOMA-IR (Figure 32) (-8.2%; -1.0 to -15.5%), FPG (Figure 31) (-2.7%; -0.2 to -5.1%), and IVGTT k-value (Figure 7) (-36.5%; -26.4 to -46.6%).
The loop diuretic furosemide reduced systolic and diastolic blood pressures (Figures 28 and 29) (-8.8%; -0.8 to -16.8%; and -9.2%; -5.9 to -12.6%; respectively). Furosemide had a neutral effect on clamp glucose disposal rate (Figure 30) (-0.7%; 24.9 to -26.0%), but a tendency to increase HOMA-IR and FPG (Figures 32 and 31) (38.9%; -18.1 to 95.8%; and 4.9%; -7.5 to 17.3%; respectively).
The direct renin inhibitor aliskiren reduced systolic and diastolic blood pressures (Figures 28 and 29) (-10.6%; -8.9 to -12.3%; and -11.0%; -9.2 to -12.9%; respectively). Aliskiren
increased clamp glucose disposal rate (Figure 30) (28.6%; 17.4 to 39.7%), and if anything tended to lower HOMA-IR (Figure 32) and FPG (Figure 31) (-5.8%; 6.6 to -18.2%; and - 2.2%; 1.8 to -6.1%; respectively).
The I1-agonist rilmenidine reduced systolic and diastolic blood pressures (Figures 28 and 29) (-9.6%; -7.0 to -12.1%; and -11.9; -7.8 to -15.9%; respectively), had a largely neutral effect on HOMA-IR (Figure 32) (7.2%; -46.6 to 60.9%) and FPG (Figure 31) (1.2%; -4.9 to 7.4%), but lowered OGTT-2h-glucose (Figure 6) (-8.9%; -2.5 to -15.2%).
16
Associations of change in blood pressures with change in insulin sensitivity measures
In meta-regression analyses, change in systolic blood pressure was not related to change in clamp glucose disposal rate (Figure 33; percent change in glucose disposal by percent change in systolic blood pressure β=0.19; p=0.34). In contrast, improvement in systolic blood
pressure was associated with improvement in HOMA-IR (Figure 34; β=3.74; p<0.0001), and FPG (Figure 35; β=0.11; p=0.006).
The degree of improvement in diastolic blood pressure was consistently related to
improvement in glucometabolic variables, expressed as clamp glucose disposal rate (Figure 36; percent change in glucose disposal by percent change in diastolic blood pressure β= -0.60;
p = 0.002), HOMA-IR (Figure 37; β=2.72; p<0.0001), and FPG (Figure 38; β=0.17;
p<0.0001).
17
DISCUSSION
Summary of findings
In this meta-analysis of 103 articles with 207 study arms, was blood pressure reduced similarly in all antihypertensive treatment groups. Treatment with ACE-inhibitors, ARBs, calcium antagonists, and alpha-blockers improved insulin sensitivity on average, treatment with beta-blockers had a neutral effect on insulin sensitivity, and treatment with thiazides worsened insulin sensitivity. Effects on HOMA-IR, clamp glucose disposal rate and FPG were coherent with most drugs. OGTT-2h-glucose was lowered with ACE-inhibitor treatment and tendency to be neutral with ARB, beta-blocker and calcium antagonist. IVGTT (k-value) had a tendency to be reduced with alpha-blocker, beta-blocker and thiazides and increased with ACE-inhibitor.
Meta regression analyses indicated that the degree of improvement on blood pressure correlates whit improvement in metabolic variables as clamp glucose disposal rate, HOMA- IR and FPG. Thou the exception were correlation to systolic blood pressure and clamp glucose disposal rate.
Divergent guidelines for patients with diabetes and hypertension
Patients with diabetes mellitus classify as hypertensive with repeated blood pressure over 130/80 mmHg 20. These patients should be treated to a systolic blood pressure below 130 mmHg and diastolic blood pressure below 80 mmHg. Patients with blood pressure 140/90 mmHg or more shall have pharmacology therapy in combination with lifestyle changes. The first line treatment that is recommended from America Diabetes Assiciation (ADA) is that patients with diabetes and hypertension shall been given treatment with an ACE-inhibitor or ARB20. If the patients need multiple therapies is the recommendation to add thiazides, calcium antagonist or beta-blocker. However it is shown in European guidelines that
development of new-onset diabetes is greater in beta-blocker and diuretic treatment compared with ACE-inhibitor, ARB and calcium antagonist2. Therefore are beta-blocker and thiazide treatment in combination not recommended to patients with metabolic syndrome or patients that have a predisposal risk to development diabetes2.
European guidelines and American guidelines are overall agree on treatment strategies for patients with high blood pressure and diabetes, i.e. the first line therapy that is recommended
18
is ACE-inhibitor and ARB for patients with diabetes and if necessary in combination with other antihypertensive drugs2,20.
Comparisons with previous studies of diabetes incidence
Approximately 15-20% of hypertensive patients shows glucose tolerance that may due to antihypertensive drugs as diuretics and beta-blockers 21. Long-term studies have showed that patient that has development diabetes because of antihypertensive drugs have a higher incident in cardiovascular complications2.
The CHARM study 21 (In the Candesartan in heart failure-assessment of reduction in mortality and morbidity) showed a lower incident of new-onset diabetes with candesartan treatment compared to placebo, thou it was related to new-onset diabetes to the added
medication of diuretic in the study arms both to candesartan and placebo arm. In four studies
21 (J.Bosh for the heart outcomes prevention evaluation study investigators, 2003, personal communication) there it was evaluated an ACE-inhibitor or ARB compared to placebo. It was shown that patients that got the ACE-inhibitor or ARB treatment had lower new-onset
diabetes compared to placebo. Nevertheless, several of the patients had also treatment with diuretic and beta-blocker.
In this meta-analysis was it found that ARB had a positive effect on insulin sensitivity which is confirmed by previous studies. The negative effect on insulin sensitivity seems to be caused of thiazide treatment. In previous studies was it believed that ACE-inhibitor was more neutral but in our results was it shown that ACE-inhibitor had a slightly positive effect on insulin sensitivity.
The VALUE study 21 was a long term study that showed that incident of new-onset diabetes was 20% less with valsartan treatment compared with amlidopine treatment. Previous studies as INSIGHT study 10 found that the nifedipine treatment had a lower incident in new-onset diabetes compared with hydrochlorothiazide plus amiloride treatment. These results are consistent with our results that indicate that calcium antagonists have a slightly positive effect on insulin sensitivity compared with other antihypertensive drugs. According to previous studies was calcium antagonist believed to have a more neutral effect on insulin sensitivity.
19
ALPINE 22 identified that thiazide group, if needed with combination of beta-blocker, had higher fasting insulin and plasma glucose values compared with ARB group, if needed in combination with calcium antagonist. When the study evaluated HOMA-IR it was indicated an increased insulin resistance in the thiazide group. In the study ARIC10 (Atherosclerosis Risk in Communities) was it shown that treatment with beta-blocker had a higher risk of 28%
to development diabetes compared with those who had no treatment at all. Thiazide diuretic, calcium antagonist and ACE-inhibitor were shown not to have a higher risk to development diabetes than those without treatment. The review 21 (New-onset diabetes and
antihypertensive drugs) suggests that beta-blocker and diuretic treatment can lead to diabetes in patients that have a higher risk to develop the disease.
SCOPE10 is a study that compares diuretic and beta-blocker with candersatan. A Subgroup to the SCOPE study (Study in congnition and prognosis in the elderly) was showed that patients that didn’t got add on therapy with hydrochlorthiazide got new-onset diabetes less often and the incident was similar to candersatan and placebo treatment In a study by Taylor et al was it investigated different risk to development diabetes in three hypertensive groups, it was shown that the incident of new-onset diabetes was higher in the population that got the treatment of thiazide diuretic compared with those without treatment.
Previous studies show a higher incidence of new-onset diabetes in patients treated with
thiazide alone or in combination with beta-blocker. Those observations are consistent with the result of this meta-analysis where we have shown that insulin sensitivity adversely affected by thiazides. Nevertheless we have not found that beta-blocker had a negative impact on insulin sensitivity in monotherapy.
We found that beta-blockers had a neutral effect on insulin sensitivity. Among previous studies, ASCOT 11 showed a lower incidence of diabetes in the amlidopine group compared with atenolol group, however the beta-blocker treatment was if necessary combined with a diuretic. It is important to note the majority of those randomized to the beta-blocker regime were also treated with a thiazide. Similar strategies were used in the ALLHAT, LIFE and the INVEST studies, in which higher incidence of new-onset diabetes were observed in those treated with beta-blockers, if necessary in combination with thiazides. The observations of the present study of a neutral effect on insulin sensitivity of beta-blockers in monotherapy may
20
motivate further post-hoc studies of subgroups of patients with beta-blockers in monotherapy in those previous studies.
Antihypertensive1,10 agent as ARB, ACE-inhibitor, calcium antagonist and alpha blocker don’t have any negative effects on metabolic parameters like thiazied diuretic and beta- blocker had. It has been shown that the incident of new-onset diabetes is lower with treatment with ARB or ACE-inhibitor compared with thiazide diuretic, beta-blocker and calcium antagonist1.
Strength and Limitations
Strengths of this study include a large sample of studies, providing robust estimates. Strengths of the method include transparent, pre-specified inclusion criteria, combining considerations of relevance and considerations of bias.
This study has limitations. The quality of the included studies varies. A meta-analysis will never be better than the original data. The major weakness of this meta-analysis refers to the defects in the originally studies and we cannot exclude the presence of unmeasured or residual confounding, possible biases due to limited use of randomization and blinding, differential drop out from study arms.
The focus was often of the results of monotherapies, but in several studies it was necessary to use combination therapy to achieve blood pressure goals. The results could therefore be confounded by differential use of other antihypertensive drugs in the study arms.
The generalizability of the study results to other populations than those typically included in hypertension studies is unknown. Mainly, the study population that was elected had mild to moderate hypertension and it is a possibility that they don’t represent patients with higher blood pressure levels or manifest cardiovascular disease. The number of diabetics patients and which origin the study population have in the originally studies can have an influence on the result. We know that different origin have different risk to development diabetes and that can become a bias concerning the choice of study population. The search of the original articles have only been done in English language databases and therefore a language bias may have been introduced.
21
Another critic is that a meta-analysis can mix different kind of studies in the same analysis.
Which studies that will be included is always a judgment. But it is very important to have inclusion criteria which is based on quality of the originally studies. Nevertheless, it is possible that we have missed studies and will not be included in this meta-analysis but it will not affect the result when the missing was in a random way. Another limitation is that we combined the results of the M value, M/I value and M/I/G value of the clamp in the same group, which are potentially different from each other but we believe that the differences are relative small. Yet another limitation is that we have only investigated the antihypertensive drugs as a group and not investigated the different kind of medications within the groups, it could result in a slight different result.
Conclusion
In this meta-analysis of hypertensive persons was blood pressure in average reduced similarly in all antihypertensive treatment groups. Treatment with ACE-inhibitors, ARBs, calcium antagonists, and alpha-blockers improved insulin sensitivity, treatment with beta-blockers had a neutral effect on insulin sensitivity, and treatment with thiazides worsened insulin
sensitivity. Effects on HOMA-IR, clamp glucose disposal rate and FPG were coherent with most drugs. These results may motivate further post-hoc studies of subgroups of patients, especially with beta-blockers and thiazides in monotherapy
Interpreting our results from a clinical viewpoint, thiazide treatment in monotherapy and as part of combination therapy should likely be avoided in patients with high risk of developing type-2 diabetes mellitus. In contrast to current European treatment guidelines2, beta-blockers did not appear to worsen cardiometabolic risk in those patients in the present study.
22
REFERENCES
1. Kahan T, Dahlgren H, de Faire U, et al. Måttligt förhöjt blodtryck En systematisk litteraturöversikt. In: SBU, Care TscoTAiH, eds. Mölnlycke: Elanders Infologistics Väst AB 2007:1-534.
2. Mancia G, De Backer G, Dominiczak A, et al. 2007 Guidelines for the management of arterial hypertension: The Task Force for the Management of Arterial
Hypertension of the European Society of Hypertension (ESH) and of the European Society of Cardiology (ESC). In: Eur Heart J. England; 2007:1462-536.
3. Cannon CP. Cardiovascular Disease and Modifiable Cardiometabolic Risk Factors. Clinical Cornerstone 2008;9:24-41.
4. Dahlof 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. In: Lancet. England; 2002:995-1003.
5. Stump CS, Hamilton MT, Sowers JR. Effect of antihypertensive agents on the development of type 2 diabetes mellitus. Mayo Clin Proc 2006;81:796-806.
6. Zappe DH, Sowers JR, Hsueh WA, et al. Metabolic and antihypertensive effects of combined angiotensin receptor blocker and diuretic therapy in prediabetic hypertensive patients with the cardiometabolic syndrome. In: J Clin Hypertens (Greenwich). United States;
2008:894-903.
7. Lithell H, Pollare T, Vessby B. Metabolic effects of pindolol and propranolol in a double-blind cross-over study in hypertensive patients. Blood Press 1992;1:92-101.
8. Lithell HO, Andersson PE. Antihypertensive treatment in insulin resistant patients. Hypertens Res 1996;19 Suppl 1:S75-9.
9. Wofford MR, King DS. Relationship between antihypertensive drugs and metabolic syndrome. Metabolic Syndrome and Related Disorders 2004;2:308-14.
10. Sarafidis PA, McFarlane SI, Bakris GL. Antihypertensive agents, insulin sensitivity, and new-onset diabetes. Curr Diab Rep 2007;7:191-9.
11. Dahlof B, Sever PS, Poulter NR, et al. Prevention of cardiovascular events with an antihypertensive regimen of amlodipine adding perindopril as required versus atenolol adding bendroflumethiazide as required, in the Anglo-Scandinavian Cardiac Outcomes Trial- Blood Pressure Lowering Arm (ASCOT-BPLA): a multicentre randomised controlled trial.
In: Lancet. England; 2005:895-906.
12. Pepine CJ, Handberg EM, Cooper-DeHoff RM, et al. A calcium antagonist vs a non-calcium antagonist hypertension treatment strategy for patients with coronary artery disease. The International Verapamil-Trandolapril Study (INVEST): a randomized controlled trial. In: JAMA. United States; 2003:2805-16.
13. 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). In: JAMA. United States; 2002:2981-97.
14. Lindholm LH, Persson M, Alaupovic P, Carlberg B, Svensson A, Samuelsson O. Metabolic outcome during 1 year in newly detected hypertensives: Results of the
Antihypertensive Treatment and Lipid Profile in a North of Sweden Efficacy Evaluation (ALPINE study). Journal of Hypertension 2003;21:1563-74.
15. Pollare T, Lithell H, Selinus I, Berne C. Application of prazosin is associated with an increase of insulin sensitivity in obese patients with hypertension. Diabetologia 1988;31:415-20.
23
16. DeFronzo RA, Tobin JD, Andres R. Glucose clamp technique: a method for quantifying insulin secretion and resistance. Am J Physiol 1979;237:E214-23.
17. Fogari R, Zoppi A, Corradi L, Lazzari P, Mugellini A, Lusardi P. Comparative effects of lisinopril and losartan on insulin sensitivity in the treatment of non diabetic
hypertensive patients. Br J Clin Pharmacol 1998;46:467-71.
18. Andersson PE, Johansson J, Berne C, Lithell H. Effects of selective alfa 1 and beta 1-adrenoreceptor blockade on lipoprotein and carbohydrate metabolism in hypertensive subjects, with special emphasis on insulin sensitivity. J Hum Hypertens 1994;8:219-26.
19. De Luca N, Izzo R, Fontana D, et al. Haemodynamic and metabolic effects of rilmenidine in hypertensive patients with metabolic syndrome X. A double-blind parallel study versus amlodipine. J Hypertens 2000;18:1515-22.
20. Executive summary: Standards of medical care in diabetes--2010. In: Diabetes Care. United States; 2010:S4-10.
21. Mancia G, Grassi G, Zanchetti A. New-onset diabetes and antihypertensive drugs. In: J Hypertens. England; 2006:3-10.
22. Lindholm LH, Persson M, Alaupovic P, Carlberg B, Svensson A, Samuelsson O. Metabolic outcome during 1 year in newly detected hypertensives: results of the
Antihypertensive Treatment and Lipid Profile in a North of Sweden Efficacy Evaluation (ALPINE study). J Hypertens 2003;21:1563-74.
23. Lender D, Arauz-Pacheco C, Breen L, Mora-Mora P, Ramirez LC, Raskin P. A double blind comparison of the effects of amlodipine and enalapril on insulin sensitivity in hypertensive patients. In: Am J Hypertens. United States; 1999:298-303.
24. Lithell H, Andersson PE. Metabolic effects of carvedilol in hypertensive patients. Eur J Clin Pharmacol 1997;52:13-7.
25. Fogari R, Zoppi A, Preti P, Fogari E, Malamani G, Mugellini A. Differential effects of ACE-inhibition and angiotensin II antagonism on fibrinolysis and insulin sensitivity in hypertensive postmenopausal women. In: Am J Hypertens. United States; 2001:921-6.
26. Fogari R, Derosa G, Zoppi A, et al. Effect of delapril/manidipine vs olmesartan/
hydrochlorothiazide combination on insulin sensitivity and fibrinogen in obese hypertensive patients. In: Intern Med. Japan; 2008:361-6.
27. Fogari R, Derosa G, Zoppi A, et al. Comparison of the effects of valsartan and felodipine on plasma leptin and insulin sensitivity in hypertensive obese patients. Hypertens Res 2005;28:209-14.
28. Haenni A, Lithell H. Moxonidine improves insulin sensitivity in insulin-resistant hypertensives. J Hypertens Suppl 1999;17:S29-35.
29. Malminiemi K, Lahtela J, Malminiemi O, Ala-Kaila K, Huupponen R. Insulin sensitivity in a long-term crossover trial with celiprolol and other antihypertensive agents. J Cardiovasc Pharmacol 1998;31:140-5.
30. Eriksson JW, Jansson PA, Carlberg B, et al. Hydrochlorothiazide, but not Candesartan, aggravates insulin resistance and causes visceral and hepatic fat accumulation:
the mechanisms for the diabetes preventing effect of Candesartan (MEDICA) Study. In:
Hypertension. United States; 2008:1030-7.
31. Ding YA, Shieh SM. Antihypertensive and metabolic effects of tiapamil. J Clin Pharmacol 1990;30:262-6.
32. Lind L, Berne C, Pollare T, Lithell H. Metabolic effects of isradipine as
monotherapy or in combination with pindolol during long-term antihypertensive treatment. J Intern Med 1994;236:37-42.
33. Watanabe M, Inukai K, Sumita T, et al. Effects of telmisartan on insulin resistance in Japanese type 2 diabetic patients. In: Intern Med. Japan; 2010:1843-7.
24
34. Olsen MH, Fossum E, Hoieggen A, et al. Long-term treatment with losartan versus atenolol improves insulin sensitivity in hypertension: ICARUS, a LIFE substudy. In: J Hypertens. England; 2005:891-8.
35. Yavuz D, Koc M, Toprak A, et al. Effects of ACE inhibition and AT1-receptor antagonism on endothelial function and insulin sensitivity in essential hypertensive patients.
In: J Renin Angiotensin Aldosterone Syst. England; 2003:197-203.
36. Grassi G, Seravalle G, Dell'Oro R, et al. Comparative effects of candesartan and hydrochlorothiazide on blood pressure, insulin sensitivity, and sympathetic drive in obese hypertensive individuals: results of the CROSS study. J Hypertens 2003;21:1761-9.
37. Uchida K, Azukizawa S, Kigoshi T, et al. Effects of perindopril on glucose and lipid metabolism in patients with mild-to-moderate essential hypertension and glucose intolerance. Clin Ther 1994;16:466-70.
38. Maritz FJ, Weich HF, Schoeman HS. A comparison of the metabolic effects of captopril and atenolol on glucose, insulin and lipoproteins in patients with mild-to-moderate essential hypertension. S Afr Med J 1995;85:1342-5.
39. Demirtunc R, Duman D, Basar M. Effects of doxazosin and amlodipine on mean platelet volume and serum serotonin level in patients with metabolic syndrome: a randomised, controlled study. In: Clin Drug Investig. New Zealand; 2007:435-41.
40. Top C, Cingozbay BY, Terekeci H, Kucukardali Y, Onde ME, Danaci M. The effects of valsartan on insulin sensitivity in patients with primary hypertension. J Int Med Res 2002;30:15-20.
41. Poirier L, Cleroux J, Nadeau A, Lacourciere Y. Effects of nebivolol and atenolol on insulin sensitivity and haemodynamics in hypertensive patients. J Hypertens
2001;19:1429-35.
42. Pessina AC, Ciccariello L, Perrone F, et al. Clinical efficacy and tolerability of alpha-blocker doxazosin as add-on therapy in patients with hypertension and impaired glucose metabolism. In: Nutr Metab Cardiovasc Dis. Germany; 2006:137-47.
43. Fogari R, Zoppi A, Mugellini A, Lazzari P, Derosa G. Different effects of aliskiren and losartan on fibrinolysis and insulin sensitivity in hypertensive patients with metabolic syndrome. Horm Metab Res 2010;42:892-6.
44. Marazzi G, Volterrani M, Caminiti G, et al. Effectiveness of nebivolol and hydrochlorothiazide association on blood pressure, glucose, and lipid metabolism in hypertensive patients. Adv Ther 2010;27:655-64.
45. Jordan J, Engeli S, Boschmann M, et al. Hemodynamic and metabolic responses to valsartan and atenolol in obese hypertensive patients. In: J Hypertens. England; 2005:2313- 8.
46. Navare HA, Frye RF, Cooper-Dehoff RM, et al. Atenolol exposure and risk for development of adverse metabolic effects: a pilot study. In: Pharmacotherapy. United States;
2010:872-8.
47. Irvin MR, Lynch AI, Kabagambe EK, et al. Pharmacogenetic association of hypertension candidate genes with fasting glucose in the GenHAT Study. J Hypertens 2010;28:2076-83.
48. Zhang JL, Qin YW, Zheng X, Qiu JL, Zhao XX, Zou DJ. Combination therapy with angiotensin-converting enzyme inhibitors and indapamide impairs glucose tolerance in Chinese hypertensive patients. Blood Press 2010;19:110-8.
49. Nishimura H, Sanaka T, Tanihata Y, Naito T, Higuchi C, Otsuka K. Losartan elevates the serum high-molecular weight-adiponectin isoform and concurrently improves insulin sensitivity in patients with impaired glucose metabolism. In: Hypertens Res. Japan;
2008:1611-8.
25
50. Derosa G, Cicero AF, D'Angelo A, et al. Metabolic and antihypertensive effects of moxonidine and moxonidine plus irbesartan in patients with type 2 diabetes mellitus and mild hypertension: a sequential, randomized, double-blind clinical trial. In: Clin Ther. United States; 2007:602-10.
51. Spirou A, Rizos E, Liberopoulos EN, et al. Effect of barnidipine on blood
pressure and serum metabolic parameters in patients with essential hypertension: a pilot study.
In: J Cardiovasc Pharmacol Ther. United States; 2006:256-61.
52. Bakris G, Molitch M, Hewkin A, et al. Differences in glucose tolerance between fixed-dose antihypertensive drug combinations in people with metabolic syndrome. In:
Diabetes Care. United States; 2006:2592-7.
53. Widimsky J, Sirotiakova J. Efficacy and tolerability of rilmenidine compared with isradipine in hypertensive patients with features of metabolic syndrome. Curr Med Res Opin 2006;22:1287-94.
54. Derosa G, Cicero AF, Bertone G, et al. Comparison of the effects of telmisartan and nifedipine gastrointestinal therapeutic system on blood pressure control, glucose
metabolism, and the lipid profile in patients with type 2 diabetes mellitus and mild hypertension: a 12-month, randomized, double-blind study. In: Clin Ther. United States;
2004:1228-36.
55. Jacob S, Klimm HJ, Rett K, Helsberg K, Haring HU, Godicke J. Effects of moxonidine vs. metoprolol on blood pressure and metabolic control in hypertensive subjects with type 2 diabetes. Exp Clin Endocrinol Diabetes 2004;112:315-22.
56. Derosa G, Cicero AF, Ciccarelli L, Fogari R. A randomized, double-blind, controlled, parallel-group comparison of perindopril and candesartan in hypertensive patients with type 2 diabetes mellitus. In: Clin Ther. United States; 2003:2006-21.
57. Ostman J, Asplund K, Bystedt T, et al. Comparison of effects of quinapril and metoprolol on glycaemic control, serum lipids, blood pressure, albuminuria and quality of life in non-insulin-dependent diabetes mellitus patients with hypertension. Swedish Quinapril Group. J Intern Med 1998;244:95-107.
58. Wiggam MI, Hunter SJ, Atkinson AB, et al. Captopril does not improve insulin action in essential hypertension: a double-blind placebo-controlled study. J Hypertens
1998;16:1651-7.
59. Fogari R, Zoppi A, Lazzari P, et al. ACE inhibition but not angiotensin II antagonism reduces plasma fibrinogen and insulin resistance in overweight hypertensive patients. J Cardiovasc Pharmacol 1998;32:616-20.
60. Hunter SJ, Harper R, Ennis CN, et al. Effects of combination therapy with an angiotensin converting enzyme inhibitor and thiazide diuretic on insulin action in essential hypertension. J Hypertens 1998;16:103-9.
61. Fogari R, Zoppi A, Lazzari P, et al. Comparative effects of nebivolol and atenolol on blood pressure and insulin sensitivity in hypertensive subjects with type II diabetes. J Hum Hypertens 1997;11:753-7.
62. Giugliano D, Acampora R, Marfella R, et al. Metabolic and cardiovascular effects of carvedilol and atenolol in non-insulin-dependent diabetes mellitus and hypertension.
A randomized, controlled trial. Ann Intern Med 1997;126:955-9.
63. Hakamaki T, Lehtonen A. Metabolic effects of spirapril and atenolol: results from a randomized, long-term study. Int J Clin Pharmacol Ther 1997;35:227-30.
64. Masuo K, Mikami H, Ogihara T, Tuck ML. Metabolic effects of long-term treatments with nifedipine-retard and captopril in young hypertensive patients. In: Am J Hypertens. United States; 1997:600-10.
26
65. Bonner G, Schmieder R, Chrosch R, Weidinger G. Effect of bunazosin and atenolol on glucose metabolism in obese, nondiabetic patients with primary hypertension.
Cardiovasc Drugs Ther 1997;11:21-6.
66. Morel Y, Gadient A, Keller U, Vadas L, Golay A. Insulin sensitivity in obese hypertensive dyslipidemic patients treated with enalapril or atenolol. J Cardiovasc Pharmacol 1995;26:306-11.
67. Haenni A, Berglund L, Reneland R, Anderssson PE, Lind L, Lithell H. The alterations in insulin sensitivity during angiotensin converting enzyme inhibitor treatment are related to changes in the calcium/magnesium balance. In: Am J Hypertens. United States;
1997:145-51.
68. Crepaldi G, Carraro A, Brocco E, et al. Hypertension and non-insulin-dependent diabetes. A comparison between an angiotensin-converting enzyme inhibitor and a calcium antagonist. Acta Diabetol 1995;32:203-8.
69. Vuorinen-Markkola H, Yki-Jarvinen H. Antihypertensive therapy with enalapril improves glucose storage and insulin sensitivity in hypertensive patients with non-insulin- dependent diabetes mellitus. Metabolism 1995;44:85-9.
70. Bohlen L, Bienz R, Doser M, et al. Metabolic neutrality of perindopril: focus on insulin sensitivity in overweight patients with essential hypertension. J Cardiovasc Pharmacol 1996;27:770-6.
71. Harper R, Ennis CN, Heaney AP, et al. A comparison of the effects of low- and conventional-dose thiazide diuretic on insulin action in hypertensive patients with NIDDM.
Diabetologia 1995;38:853-9.
72. Parhofer KG, Birkeland KI, DeFronzo R, Del Prato S, Bhaumik A, Ptaszynska A. Irbesartan has no short-term effect on insulin resistance in hypertensive patients with additional cardiometabolic risk factors (i-RESPOND). In: Int J Clin Pract. England;
2010:160-8.
73. Malminiemi K. Association between serum lipids, glucose tolerance, and insulin sensitivity during 12 months of celiprolol treatment. Cardiovasc Drugs Ther 1995;9:295-304.
74. Lehtonen A. The effects of trimazosin and pindolol on serum lipids, blood glucose and serum insulin levels. Acta Med Scand 1985;218:213-6.
75. Hauger-Klevene JH, Scornavacchi JC. Improvement of glucose tolerance in hypertensive diabetic patients treated with guanfacine one year. Eur J Clin Pharmacol 1985;29:391-3.
76. Reneland R, Alvarez E, Andersson PE, Haenni A, Byberg L, Lithell H.
Induction of insulin resistance by beta-blockade but not ACE-inhibition: long-term treatment with atenolol or trandolapril. J Hum Hypertens 2000;14:175-80.
77. Eriksson JW, Jansson PA, Foley K, Lithell H. Insulin sensitivity following treatment with the alpha 1-blocker bunazosin retard and the beta 1-blocker atenolol in
hypertensive non-insulin-dependent diabetes mellitus patients. J Hypertens 1996;14:1469-75.
78. Andersson PE, Lithell H. Metabolic effects of doxazosin and enalapril in hypertriglyceridemic, hypertensive men. Relationship to changes in skeletal muscle blood flow. In: Am J Hypertens. United States; 1996:323-33.
79. Pollare T, Lithell H, Morlin C, Prantare H, Hvarfner A, Ljunghall S. Metabolic effects of diltiazem and atenolol: results from a randomized, double-blind study with parallel groups. J Hypertens 1989;7:551-9.
80. Pollare T, Lithell H, Selinus I, Berne C. Sensitivity to insulin during treatment with atenolol and metoprolol: a randomised, double blind study of effects on carbohydrate and lipoprotein metabolism in hypertensive patients. BMJ 1989;298:1152-7.
81. Lithell H, Berne C, Waern AU, Wibell L. Glucose metabolism during long-term treatment with prazosin. Diabetes Res 1985;2:297-9.
27
82. Masajtis-Zagajewska A, Majer J, Nowicki M. Effect of moxonidine and amlodipine on serum YKL-40, plasma lipids and insulin sensitivity in insulin-resistant hypertensive patients-a randomized, crossover trial. In: Hypertens Res. England; 2010:348- 53.
83. Uzunlulu M, Oguz A, Yorulmaz E. The effect of carvedilol on metabolic parameters in patients with metabolic syndrome. In: Int Heart J. Japan; 2006:421-30.
84. Bojovic L, Micic D. Effects of ramipril therapy on some components of insulin resistance syndrome in patients with essential hypertension. Med Pregl 2002;55:286-92.
85. Kuperstein R, Sasson Z. Effects of antihypertensive therapy on glucose and insulin metabolism and on left ventricular mass: A randomized, double-blind, controlled study of 21 obese hypertensives. Circulation 2000;102:1802-6.
86. Paolisso G, Tagliamonte MR, Gambardella A, et al. Losartan mediated improvement in insulin action is mainly due to an increase in non-oxidative glucose metabolism and blood flow in insulin-resistant hypertensive patients. J Hum Hypertens 1997;11:307-12.
87. Suzuki S, Ohtomo M, Satoh Y, et al. Effect of manidipine and delapril on insulin sensitivity in type 2 diabetic patients with essential hypertension. In: Diabetes Res Clin Pract. Ireland; 1996:43-51.
88. Malmqvist K, Kahan T, Isaksson H, Ostergren J. Regression of left ventricular mass with captopril and metoprolol, and the effects on glucose and lipid metabolism. Blood Press 2001;10:101-10.
89. Haenni A, Lithell H. Treatment with a beta-blocker with beta 2-agonism improves glucose and lipid metabolism in essential hypertension. In: Metabolism. United States; 1994:455-61.
90. Lind L, Berne C, Pollare T, Lithell H. Metabolic effects of anti-hypertensive treatment with nifedipine or furosemide: a double-blind, cross-over study. J Hum Hypertens 1995;9:137-41.
91. Haenni A, Andersson PE, Lind L, Berne C, Lithell H. Electrolyte changes and metabolic effects of lisinopril/bendrofluazide treatment. Results from a randomized, double- blind study with parallel groups. In: Am J Hypertens. United States; 1994:615-22.
92. Pollare T, Lithell H, Berne C. A comparison of the effects of
hydrochlorothiazide and captopril on glucose and lipid metabolism in patients with hypertension. N Engl J Med 1989;321:868-73.
93. Huang SS, Wu TC, Lin SJ, Chen JW. Combination of an ACE inhibitor and indapamide improves blood pressure control, but attenuates the beneficial effects of ACE inhibition on plasma adiponectin in patients with essential hypertension. Circulation Journal 2009;73:2282-7.
94. Rizos CV, Milionis HJ, Kostapanos MS, et al. Effects of rosuvastatin combined with olmesartan, irbesartan, or telmisartan on indices of glucose metabolism in greek adults with impaired fasting glucose, hypertension, and mixed hyperlipidemia: A 24-week,
randomized, open-label, prospective study. Clinical Therapeutics 2010;32:492-505.
95. Koh KK, Quon MJ, Lee Y, et al. Additive beneficial cardiovascular and metabolic effects of combination therapy with ramipril and candesartan in hypertensive patients. In: Eur Heart J. England; 2007:1440-7.
96. McLaughlin DM, Atkinson AB, Ennis CN, et al. Comparison of effects of combined ACE inhibitor and low-dose thiazide diuretic with ACE inhibitor alone on insulin action in patients with hypertension and Type 2 diabetes: A double-blind crossover study.
Diabetic Medicine 2008;25:631-4.
28
97. Bank AJ, Kelly AS, Thelen AM, Kaiser DR, Gonzalez-Campoy JM. Effects of Carvedilol Versus Metoprolol on Endothelial Function and Oxidative Stress in Patients With Type 2 Diabetes Mellitus. American Journal of Hypertension 2007;20:777-83.
98. Yilmaz MI, Sonmez A, Caglar K, et al. Effect of antihypertensive agents on plasma adiponectin levels in hypertensive patients with metabolic syndrome. Nephrology 2007;12:147-53.
99. Aksnes TA, Reims HM, Guptha S, Moan A, Os I, Kjeldsen SE. Improved insulin sensitivity with the angiotensin II-receptor blocker losartan in patients with hypertension and other cardiovascular risk factors. Journal of Human Hypertension 2006;20:860-6.
100. Celik T, Iyisoy A, Kursaklioglu H, et al. Comparative effects of nebivolol and metoprolol on oxidative stress, insulin resistance, plasma adiponectin and soluble P-selectin levels in hypertensive patients. Journal of Hypertension 2006;24:591-6.
101. Vitale C, Mercuro G, Castiglioni C, et al. Metabolic effect of telmisartan and losartan in hypertensive patients with metabolic syndrome. Cardiovascular Diabetology 2005;4.
102. Courtney CH, McCance DR, Atkinson AB, et al. Effect of the alpha-adrenergic blocker, doxazosin, on endothelial function and insulin action. Metabolism: Clinical and Experimental 2003;52:1147-52.
103. Kosegawa I, Inaba M, Morita T, Awata T, Katayama S. Effect of the
vasodilatory β-blocker, nipradilol, and ca-antagonist, barnidipine, on insulin sensitivity in patients with essential hypertension. Clinical and Experimental Hypertension 1998;20:751-61.
104. Tatti P, Pahor M, Byington RP, et al. Outcome results of the Fosinopril Versus Amlodipine Cardiovascular Events Randomized Trial (FACET) in patients with hypertension and NIDDM. Diabetes Care 1998;21:597-603.
105. Gokcel A, Ozer T, Aliustaoglu M, Unat T, Yildirim MY. Effects of trandolapril on insulin concentrations and other metabolic variables in hypertensive patients. Current Therapeutic Research - Clinical and Experimental 1997;58:894-902.
106. Shionoiri H, Ashino K, Yamanaka K, Shindo K, Hiroto S, Arita T. Effect of doxazosin therapy on glucose tolerance and lipid metabolism in hypertensive patients with impaired glucose tolerance. Clinical Therapeutics 1997;19:527-36.
107. Laakso M, Karjalainen L, Lempiäinen-Kuosa P. Effects of losartan on insulin sensitivity in hypertensive subjects. Hypertension 1996;28:392-6.
108. Jacob S, Rett K, Wicklmayr M, Agrawal B, Augustin HJ, Dietze GJ.
Differential effect of chronic treatment with two beta-blocking agents on insulin sensitivity:
The carvedilol-metoprolol study. Journal of Hypertension 1996;14:489-94.
109. Minami J, Abe C, Akashiba A, et al. Long-term efficacy of combination therapy with losartan and low-dose hydrochlorothiazide in patients with uncontrolled hypertension.
International Heart Journal 2007;48:177-86.
110. Shionoiri H, Takasaki I, Naruse M, et al. Effect of cilazapril therapy on glucose and lipid metabolism in patients with hypertension. Clinical Therapeutics 1995;17:1126-35.
111. Sung PK, Samet P, Yeh BK. Effects of clonidin and chlorthalidone on blood pressure and glucose tolerance in hypertensive patients. Curr Ther Res Clin Exp 1971;13:280- 5.
112. Malminemi K, Lahtela JT, Huupponen R. Effects of celiprolol on insulin sensitivity and glucose tolerance in dyslipidemic hypertension. International Journal of Clinical Pharmacology and Therapeutics 1995;33:156-63.
113. Van Mieghem W. Evaluation of glucose tolerance during treatment with celiprolol in patients with mild arterial hypertension without diabetes mellitus. Acta Clinica Belgica 1997;52:360-6.
29
114. Martínez Martín FJ. Manidipine in hypertensive patients with metabolic syndrome: The MARIMBA study. Expert Review of Cardiovascular Therapy 2009;7:863-9.
115. Ebinç H, Ozkurt ZN, Ebinç FA, Ucardag D, Caglayan O, Yilmaz M. Effects of sympatholytic therapy with moxonidine on serum adiponectin levels in hypertensive women.
Journal of International Medical Research 2008;36:80-7.
116. Topal E, Cikim AS, Cikim K, Temel I, Ozdemir R. The effect of moxonidine on endothelial dysfunction in metabolic syndrome. American Journal of Cardiovascular Drugs 2006;6:343-8.
117. Falkner B, Francos G, Kushner H. Metoprolol succinate, a selective beta- adrenergic blocker, has no effect on insulin sensitivity. Journal of clinical hypertension (Greenwich, Conn) 2006;8:336-43.
