Journal of Diabetes & Metabolism

Hypertension with concomitant type 2 diabetes mellitus (DM) represents a major modifiable risk factor for cardiovascular disorders and it is associated with a widespread metabolic disarray including dyslipidemia [1-3]. Recent studies have demonstrated that dyslipidemia antedates and predicts the risk for cardiovascular disease (CVD) [2- 4] and havealso indicated that high concentrations of low density lipoprotein cholesterol (LDL-C) and triglyceride (TG) are strong risk factors for the development of CVDs [5,6].

The combination of these factors which often co-exist with other established cardiovascular risk factors such as obesity, alcohol consumption and smoking, greatly enhances the risk of hypertensive type 2 diabetic patients developing Coronary Heart Disease (CHD); and failure to satisfactorily control these risk factors may be one of the possible explanations for the disappointing results of antihypertensive treatment on the incidence of CHD [4-9]. Consequently, recent guidelines [9-11] have emphasized comprehensive risk factor reduction together with effective BP control so as to decrease the incidence of cardiovascular events in these patients. A corollary to this approach is that antihypertensive drugs should not aggravate metabolic factors that increase the risk of CHD and thereby offset the benefits of antihypertensive therapy [4-8].

Antihypertensive agents are reported to exhibit a wide range of effect on lipid metabolism: while some adversely affect lipid status, others have a neutral or favorable effect on it [12]. Thus previous studies have reported that diuretics such as hydrochlorothiazide (HCZ) negatively affect serum lipid levels [12-16]. In contrast to the above reports, studies by Nandeesha et al. [7], Koh et al. [12], Wong et al. [17] as well as Leonetti [18] have demonstrated that the longactingdihydropyridine calcium channel blockers (CCBs) particularly Amlodipine (AML), exhibit beneficial effects on lipid profiles because they possess antioxidant property that inhibits the oxidation of LDL-C and its influx into the arterial wall to cause atherosclerotic lesions. Even though the association between antihypertensive drugs such as AML and HCZ and lipid parameters is well established, very little is known about the effects of these two antihypertensive medications on serum lipids in hypertensive type 2 diabetic Nigerians. Therefore, the present study was designed to compare the effects of these drugs on newly diagnosed hypertensive Nigerians with type 2 DM so as to determine the one preferable for the treatment of these high risk hypertensive patients.

Population

We enrolled into the study 40 type 2 diabetic Nigerians of both gender with newly diagnosed essential hypertension (stages 1 and 2) aged 43-68 years who were attending Central Hospital and Osigbemhe Hospital, both in Auchi in Edo State of Nigeria between March 2008 and March 2009. The sample size was estimated based on the number of Nigerians that are believed to have hypertension with concomitant type 2 DM [19]; and to detect a difference of 1 unit in mean change in the measured variables, between both treatment arms with a power equal to 90% using a one sample t-test at a one-sided significance level of 0.05, this requires 20 patients per group.

Eligible participants had qualifying hypertension of Blood Pressure (BP)>160/90 and ≤ 180/120 mmHg measured on at least 2 occasions in lying/supine, sitting and standing positions using standardized methods [20]. Excluded were patients with identifiable cause of the hypertension except type 2 DM, clinical evidence of cerebrovascular, cardiac, renal, hepatic, gastrointestinal or endocrinologic disease except type 2 DM, hypersensitivity to AML and HCZ or related drugs, history of smoking, alcohol intake, substance abuse or mental illness. Also excluded were patients needing any concomitant medication (apart from oral antidiabetic drugs) e.g. digitalis, non-steroidal anti-inflammatory drugs, psychotropic drugs, monoamine oxidase inhibitors or oral contraceptives, that may interact with the trial drugs and pregnant or lactating females. Controls comprised the parallel age and sex-matched hypertensives on HCZ. The research protocol was reviewed and approved by the Ethics Committees of Irrua Specialist Teaching Hospital Irrua, Nigeria (Ambrose Alli University College of Medicine Teaching Hospital) and Central Hospital Auchi, Nigeria. After suitable explanation of the study protocol in layman language, all literate patients gave informed written consent and the illiterates thumb-printed the consent form before the beginning of the study.

Study design

Subjects were examined by a standardized, pre-tested questionnaire seeking information on demographic data, the history of hypertension, DM, current drugs if any, educational and social status, dietary habits, smoking and alcohol intake, etc. The 40 patients were randomized to AML and HCZ groups each comprising 20 patients (10 males (m) + 10 females (f)) using computer program-generated random numbers. Diabetes was treated in 32 patients with glibenclamide 5 mg once daily and metformin 500 mg once or twice daily as well as in 8 patients with gliclazide 80 mg once or twice daily.

Measurements of heights (m), weights (wt) (kg) and BP (mmHg)

A stadiometer scale (Seca model, UK) was used for measuring height, with no shoes on; and a beam balance (Hackman, UK) was used to measure weight while in light clothing. Body mass index (BMI) was computed as weight divided by height squared. Systolic BP (SBP) and diastolic BP (DBP) were measured with a standard mercury sphygmomanometer (Riester Diplomat Presameter, Germany) using standardized methods [20] at the sitting, standing and supine positions; always between 8 am and 10 am. All constricting clothing on the upper arm was removed before any measurement and subjects were discouraged from talking or moving during measurements. The first phase of the Korotkoff sound was regarded as the SBP while the fifth phase was regarded as the DBP. During measurement, readings were taken two consecutive times with an interval of at least one minute and the average recorded. During the study, subjects were not told the results of BP measurement.

Pharmacotherapy intervention

Patients in AML group were treated initially with AML 5 mg and the dose was doubled after 6 weeks if BP was not controlled while in HCZ group patients were treated with HCZ 25 mg, both medications being administered once daily. The outpatient treatment lasted 48 weeks. The patients were monitored closely and outcome measures evaluated at baseline, before treatment and at the end of weeks 1, 3, 6, 12, 24, 36 and 48. Unequivocal patient identification was possible via a patient identification list consisting of the patient number, first name and surname.

The study medications AML and HCZ are licensed for longterm treatment of hypertension so dangerous side effects due to the medicaments were not to be expected. AML 5 mg and 10 mg tablets (Amlovar^R) were donated by Neimeth International Pharmaceuticals Ikeja, Nigeria: NAFDAC Reg No A4-0333; Manufacturing Date 07- 2007 and Expiry Date 07-2010. HCZ 25 mg tablets (Esidrex^R) were donated by Novartis Pharma SAS Nigerian Representative, NAFDAC Reg No OL-3705, Manufacturing Date 08-2007 and Expiry Date 08- 2010.

Course of study and methods for recording efficacy and safety

All patients were advised to maintain their usual diet (weightmaintaining no-salt-added diet) and regular physical activity but to avoid undue stress throughout the duration of the study. They were instructed to take their drugs every morning. Each patient was observed for about 2 hours after taking medication drug for the first time. Adherence in respect of intake of medication was encouraged by interviewing patients through phone calls, sporadic visits and pill counts outside the view of patients. To preclude white-coat effect, observer bias and to accurately assess the efficacy of the drugs, patients were followed up repeatedly at weeks 1, 3,6,12, 24, 36, and 48. At each visit, volunteered or spontaneous report of adverse events were assessed for severity and association with treatment; and the attending physicians/investigators also recorded any adverse events they observed themselves or elicited from the patient through careful interrogation like “How do you feel?” No patient withdrew from the study because of adverse events. Response to therapy was defined as a decrease in the mean trough sitting SBP and DBP of 10 mmHg or a drop to <90 mmHg with reduction of >5 mmHg. BP was regarded as controlled if the DBP was <80 mmHg and SBP <130 mmHg. The effects of treatment on the various variables (except height) were assessed by comparing the values at each visit with the pretreatment baseline values.

Blood sample collection and analysis

At every visit, after an overnight fast of at least 8 h, 10 ml of blood was obtained from each patient by peripheral venipuncture into a plain sterile bottle. Serum was obtained by allowing the blood to clot and then centrifuged at 5000 rpm for 5 minutes at room temperature. Serum lipid profiles were measured at baseline (week 0) and at the end of weeks 1, 3, 6, 12, 24, 36 and 48. Total cholesterol (TC) and triglyceride (TG) were determined enzymatically by the cholesterol oxidase method and the glycerol oxidase method [21], respectively; and high density lipoprotein cholesterol (HDL-C) by the phosphotungstate magnesium chloride method [21] and low density lipoprotein cholesterol (LDL-C) was calculated using the formula of Friedewald et al. [22] (Reagent kits manufactured by Randox Laboratories Ltd Ardmore, Crumlin, UK).

All data are presented as mean ± SEM or mean ± SD (for age, height and weight) using the Proc ANOVA of SAS (2004). Where significant differences were noticed, mean separation was carried out using Duncan Multiple Range Test. Correlation between two sets of variables was determined using Spearman’s rank correlation. P<0.05 was regarded as significant.

The 4 randomized treatment subgroups were comparable with regard to the main demographic and clinical characteristics (Table 1). The effects of treatment drugs on SBP and DBP in the trial subjects are presented in table 2. The duration of treatment effect on the variables was significant (P<0.0001) and AML significantly reduced SBP and DBP more than HCZ (P<0.01). Overall, the mean M vs. F SBP/DBP decrease from baseline was 27.0/17.5 vs. 29.5/20.0 mmHg for AML group and 23.5/17.5 vs. 22.0/16.5 mmHg for HCZ group, respectively.

Table 1: Demographic characteristics and baseline blood pressures of hypertensive diabetic subjects (N=20 (10M+10F) per group).

Table 2: Effects of monotherapy with AML and HCZ for 48 weeks on BP (mmHg) in type 2 hypertensive diabetic subjects.

Table 3 shows that treatment and time-dependent effects were not significant on TC although mean values were decreased in AML group and increased in HCZ group. However, gender effect was significant (P<0.0001), for levels in the M were higher than those in F. TC was positively correlated with SBP (r=0.6816, P=0.0001). The treatment and time-dependent effects of drugs on TG were similar to those on TC (Table 4). However, gender effect was insignificant. Treatment effect on HDL-C was significant (P<0.0001) and while there was a significant time-dependent increase (P<0.0001) in HDL-C concentration in AML subgroups, there was a decrease in HCZ subgroups (Table 5). By week 48, mean M vs. F percentage change in HDL-C concentration was 17.5 vs. 17.4 in AML group and -7.3 vs. -8.1 for HCZ group, respectively. Gender effect was significant (P<0.0001) for values were higher in M. HDL-C was negatively correlated with SBP (r=-0.1280, P=0.0220) and DBP (r=-0.2371, P=0.0001). Treatment effect on LDL-C was significant (P<0.0001) and although there was a time-dependent decrease in LDL-C concentration in AML subgroups and increase in HCZ subgroups, these changes in concentration were not significant (Table 6). At week 48, mean M vs. F % change in LDL-C level from baseline was -3.2 vs. -2.2 in AML group and 2.5 vs. 1.9 for HCZ group, respectively. Gender effect was significant (P<0.001) because levels in the M were higher.

Table 3: Effects of monotherapy with AML and HCZ on total cholesterol (mg/dl) in type 2 diabetic hypertensive subjects for 48-weeks.

Table 4: Effects of monotherapy with AML and HCZ on triglycerides (mg/dl) in type 2 diabetic hypertensive subjects for 48 weeks.

Table 5: Effects of monotherapy with AML and HCZ on high density lipoprotein cholesterol (mg/dl) in diabetic hypertensive subjects for 48 weeks.

Table 6: Effects of monotherapy with AML and HCZ on low density lipoprotein cholesterol (mg/dl) in diabetic hypertensive subjects for 48 weeks.

The patients who participated in this study were recruited from 6 Local Government Areas in Edo North Senatorial Zone of Edo State in the Niger-Delta region of Nigeria. These patients may be representative of many Nigerian communities burdened by high prevalence and incidence of hypertension [19]. In the present study, the mean ages of the patients reveal that they are relatively young, a phenomenon most probably due to a high premature mortality [23]. The patients also had high mean BMIs that may be explained by lack of exercise or by their high energy, low-protein, cassava, yam or maize-based diets. The high BMI values of these patients suggest that many of them may be victims of the metabolic syndrome [24]. The limitations of this study include its restriction to a small suburban district of Nigeria and the relatively small sample size. Therefore, caution should be exercised in making deductions from the results or extrapolating findings to hypertensive diabetic black patients in general.

Data from the present study confirm that, beyond their hypotensive effects, AML and HCZ exert different metabolic effects on lipid profiles in a black Nigerian cohort. These effects may differently influence the risk profile of these hypertensive patients with DM. In this study, AML significantly reduced LDL-C and increased HDL-C levels. The effects of CCBs on lipid profiles remain controversial. Several studies report that CCBs are metabolically neutral and cause little or no lipoprotein alterations [25,26]. In contrast to the above reports, the current data have demonstrated significant beneficial changes in lipid profiles, following AML treatment, confirming similar previous reports [7,12,17,18]. It is known that multiple risk factors including hypertension and hyperlipidemia, likely through oxidative stress mechanisms, cause a synergistic destructive effect on endothelium leading to decrease nitric oxide (NO) release, increase permeability of vessel wall to LDL-C to cause atherosclerosis with its clinical manifestations such as CHD [3,5]. Certain agents that possess anti-oxidant and anti-inflammatory properties, including AML and 3-hydroxyl-3-methylglutaryl coenzyme A reductase (HMG CoA reductase) inhibitor (statin) such as atorvastatin, are known to directly stimulate NO release from human endothelial cell in a highly synergistic fashion [27]. Thus, in addition to reducing BP and improving the lipid panel, the synergistic effects of this drug combination on human endothelial function has now become an additional mechanism to halt the progression of atherosclerosis, reduce patients’ overall cardiovascular risk and improve the management of patients at risk of CVD [28-30].

AML-based treatment significantly increased HDL-C levels. According to Florentin et al. [31], plasma HDL-C levels are inversely related to the risk of atherosclerosis and CVD. The anti-atherogenic function of HDL-C is largely attributed to cholesterol transport from peripheral tissues to the liver. However, several other properties, mainly anti-oxidant, anti-inflammatory, anti-proliferative, anti-thrombotic and vasodilatory are likely to contribute to the atheroprotective action of HDL-C [31]. Also, apart from the cardiovascular benefit of increased HDL-C levels, the results of a meta-analysis reported recently indicate that there is a significant inverse relationship between baseline HDL-C levels and the rate of incident cancer and that for every 10 mg/dl increase in HDL-C levels, there is a 28% reduction in the risk of cancer [32]. Adjustment for baseline LDL-C level, age, BMI, DM, gender and smoking status did not abolish this inverse association. Biological plausibility for this observation is that anti-oxidants like HDL-C are known to reduce cancer risk and in addition, through the process of immune surveillance whereby WBCs search out and mop up abnormal cells, increased levels of HDL-C with its positive effects on the immune system, may improve immune surveillance [30] that may lower cancer risk. According to Robinson [33], reverse causality does not explain away the relationship between low HDL-C and cancer as there is a dose response observed between the lower rates of cancer with increases in HDL-C, providing support again for the association. Thus until further information is available, the present evidence best supports low HDL-C as a marker for an overall increased risk of chronic disease [33].

The negative hostile effects of HCZ on lipid profiles have been well documented consistently in many studies [13,14,16,23]. However, the current use of low dose regimens of HCZ has reduced this metabolic adversity to the barest minimum, as evidenced in the current and other studies [14,16]. As observed in this study, there appears to be gender differences in lipoprotein responses to HCZ therapy, for the M values were consistently higher than the F. This may represent a new finding among this ethnic Nigerian population and clinicians need to be aware of this increased risk. This finding is in contrast with that of Boehringer et al. [34] who reported that among post-menopausal women diuretics may induce a greater short-term increase in TC and LDL-C than they do among men; no significant changes were observed among premenopausal women, suggesting that there may be a protective effect of estrogens [35]. Estrogen therapy also has been demonstrated to increase the number of hepatic LDL-C binding sites and to stimulate hepatic uptake of chylomicron remnants [36]. A plausible explanation for the observation in the present study could be that the M patients had relatively higher ages and BMIs.

The positive correlation between TC and SBP and negative correlation between HDL-C and SBP, TG and DBP and between LDL-C and DBP, may indicate the crucial roles of these variables as risk factors for the development of hypertension and CHD. Hypertension is known to be a progressive cardiovascular syndrome arising from complex and interrelated etiologies, and it seldom occurs in isolation from other cardiovascular risk factors. In fact, data from the Framingham Heart Study show that more than 80% of hypertensive patients (>140/90 mm Hg) have at least one metabolically-linked cardiovascular risk factor. These risk factors include obesity, glucose intolerance, hyperinsulinemia, elevated LDL-C and TG levels, low HDL-C levels and left ventricular hypertrophy [37]. In conclusion, in this black Nigerian cohort, apart from reducing BP more significantly than HCZ, AML appears to be superior to HCZ with regard to clinical outcomes on lipid profiles in hypertensive patients with type 2 DM and therefore seems a better alternative for treatment of these patients. However, further research is needed to confirm our findings.

The authors appreciate gratefully the Managements of the hospitals and patients used for this study as well as Neimeth International Pharmaceuticals, Nigeria for donation of amlodipine (AmlovarR) tablets and the Nigerian Representative of Novartis Pharma SAS France for donation of hydrochlorothiazide (EsidrexR) tablets for the study.