FormalPara Key Summary Points

Uncontrolled hypertension is the largest contributor to the burden of disease and a growing public health problem in sub-Saharan Africa (SSA). While current guidelines recommend using two or more antihypertensive agents in single pill combinations (SPCs) to control hypertension, data comparing these combinations are lacking in SSA.

This observational study assessed the decrease in systolic blood pressure (BP) and tolerance from baseline up to 16 weeks after initiation of two-drug SPCs in Douala. The three groups of SPCs were renin–angiotensin system inhibitors (RAASi) + diuretics (DIU), calcium channel blockers (CCB) + RAASi, and DIU + CCB.

The three two-drug SPCs were highly effective in reducing and controlling BP with low and similar rates of adverse effects. These findings suggest that SPCs represent a simple and low-cost intervention that could significantly reduce the burden of uncontrolled hypertension in SSA. Rather than effectiveness or short-term safety, the clinician’s selection must be guided by other criteria including availability, price, and patient’s clinical profile.

Introduction

Hypertension is a major preventable cause of cardiovascular disease including stroke, heart failure, and kidney disease globally [1, 2] In sub-Saharan Africa (SSA), hypertension affects about one-third of adults [3]. Despite advances in the understanding of pathophysiology and in the therapeutic strategies to manage hypertension, SSA patients continue to have increased morbidity and mortality from hypertension-mediated organ damage (HMOD), due to poor treatment and control rates. Also, compared with other ethnic groups, black patients have a higher prevalence and earlier onset of hypertension, coupled with poorer prognosis [4]. Therefore, early treatment of hypertension and the attainment of blood pressure (BP) goals within the shortest time possible are important to prevent HMOD. BP targets are usually difficult to achieve, and most patients require a combination of two or more drugs. Indeed, several studies showed that adequate BP control in most patients can only be guaranteed by a combination of at least two classes of BP-lowering medications [5, 6], preferentially in single pill combinations (SPCs) to improve adherence [7]. In line with this, the International Society of Hypertension (ISH) and the European Society of Cardiology/European Society of Hypertension (ESC/ESH) guidelines recommended the use of an SPC therapy with either two or three antihypertensive drug classes to initiate treatment in people with hypertension [8, 9]

Five major drug classes have proven evidence in their ability to reduce BP and cardiovascular morbidity and mortality in western countries [10], including beta-blockers, calcium channel blockers (CCB), renin–angiotensin system inhibitors (RAASi) (angiotensin-converting enzyme inhibitors and angiotensin receptors blockers), and diuretics (DIU) (thiazides and thiazide-like diuretics such as indapamide), but it should be noted that the last three classes are currently the mainstay of treatment [8, 9]. The Pan African Society of Cardiology (PASCAR) recommended that these same drug classes preferentially in SPCs be used to reduce the burden of hypertension in Africa [11]. Furthermore, PASCAR recently endorsed the ISH guidelines. However, data is lacking to support these recommendations in the context of SSA. Some studies, mainly of observational design and small sample size, suggested that CCB and diuretics are more efficacious as monotherapies in terms of BP-lowering efficacy than RAASi [12,13,14,15]. These studies did not focus on initiating therapy with combinations. The CREOLE trial is to our knowledge the only contemporary large randomized clinical trial on the treatment of hypertension in SSA which reported efficacy of two-drug combinations of antihypertensive agents (DIU + CCB, CCB + RAASi, and RAASi + DIU); findings showed a better efficacy of combinations with amlodipine and a low rate of adverse events for the three combinations [16]. However, these were free pill combinations, while current practice showed a prescribing craze of SPCs in SSA [17, 18]. In this study, we aim to provide real-world data on the effectiveness and short-term safety of different two-drug SPCs of BP medications used in daily cardiology practice in a low-middle-income setting.

Methods

Study Design and Setting

We conducted an in-depth retrospective analysis of the pharmacological strategies of all patients included in the hypertension registry and treated with two-drug SPCs at two reference centers in Douala (Douala General Hospital and Cardiovascular Center of Douala). The registry was initiated in January 2010, and the current analysis focused on patients included until January 2020. The institutional review board of the Faculty of Health Sciences, University of Buea, approved the study and we further obtained administrative authorization from the participating institutions and conformed to the principles outlined in the Declaration of Helsinki.

Eligibility Criteria

All patients aged 30 years or older with a clinical diagnosis of hypertension by their cardiologist were included in the registry. Participants who reported to be on two-drug SPCs and who had at least three office visits in the subsequent 4 months following initiation of drug therapy were considered for analysis. Patients were excluded if they had insufficient diagnostic evidence of hypertension, or if they had secondary hypertension. Also, patients with hypertensive disorders in pregnancy or those with clinical manifestations of HMOD (stroke, heart failure, or coronary artery disease and renal failure), and those who had a change of their active molecule in the SPCs during the 16-week follow-up period were excluded.

Study Procedure and Data Collection

During the study period, diagnosis of hypertension was based on office BP per the ESC/ESH international guidelines [8] with a threshold of systolic blood pressure (SBP) ≥ 140 mmHg and/or diastolic blood pressure (DBP) ≥ 90 mmHg. Patients diagnosed with hypertension in the two reference settings were systematically screened for other cardiovascular risk factors and clinical cardiovascular, neurological, or renal manifestations. They were invited to undergo minimum laboratory testing including fasting blood sugar, serum sodium and potassium, serum creatinine, total cholesterol, and a 12-lead standard electrocardiogram (ECG). The requirement of an ambulatory BP measurement (ABPM), repetition of laboratory measurements, and any other laboratory tests were completely left to the judgment of the attending physician. General recommendations for treatment of hypertension in our setting during the study period were to initiate a two-drug SPCs when baseline BP was at least 20/10 mmHg (systolic/diastolic) above the target of less than 140/90 mmHg, thus being at least 160/100 mmHg. When the BP target was not achieved after 4 weeks, the attending cardiologist was free to decide on dose escalation or change of medications. We collected sociodemographic data, associated cardiovascular risk factors, HMOD, weight (in kilograms), height (in meters), BP (mmHg), drug history, and antihypertensive drugs adverse events. Outcomes were changes in BP after 16 weeks of follow-up, attainment of BP target value (less than 140/90 mmHg), and drug tolerance (self-reported drug adverse event by the patient, and/or perceived drug tolerance by the cardiologist). Office BP was measured before initiating any drug therapy and at 4, 8, 12, and 16 weeks from the index date of two-drug SPC initiation. BP was measured during each visit thrice in a seated position using a validated electronic sphygmomanometer (OMRON HEM) at 2-min intervals. The recorded office BP was the average of the second and third measurements. Heart rate was also measured simultaneously. Body mass index (BMI) was calculated by dividing the weight by the height squared. Antihypertensive drug adverse events were measured clinically on the basis of self-report from the patient and on evaluation by the attending cardiologist at any visit. Regardless of the commercial name and dose of medications, participants were categorized into three subgroups based on the active substance present in the reported SPCs used for treatment. These included CCB + DIU, CCB + RAASi, and RAASi + DIU. The diuretic classes considered were either thiazide or thiazide-like diuretics.

Statistical Analysis

Data were analyzed using R version 4.2.1. Missing data for BP in the subsequent follow-up visits were replaced with BP values from the most recent visit for the corresponding participant. Baseline characteristics were compared between the different classes of two-drug SPCs, continuous variables were presented as the mean and standard deviation for symmetrically distributed data and median and interquartile range for skewed distributions. Categorical variables were presented as frequencies and percentages. Mixed repeated measures analysis of variance (ANOVA) was used to analyze BP at different time intervals. Greenhouse–Geisser and Huynh–Feldt corrections were applied to correct for violation of the assumption of sphericity. A mixed linear repeated model was used to evaluate the change of SBP from baseline to week 16, while controlling for age, gender, and baseline SBP. The chi-squared test was used to compare the proportion of patients with controlled BP. In addition a logistic regression model was fitted to evaluate BP control at week 8 and week 16 while controlling for baseline SBP, age, and gender to evaluate association between the different SPCs and SBP and the proportion of adverse drug reactions across the three groups.

Results

Of a total population of 549 patients with hypertension on combination therapy, 377 (68.7%) participants were on two-drug fixed-dose combinations at baseline, including 123 on CCB + DIU, 96 on RAASi + CCB, and 158 on RAASi + DIU (Fig. 1). Their mean age was 54.6 (± 11.3) years, similar across the different subgroups of SPCs. There were 187 (49.6%) male participants. Also, half of the participants reported a history of hypertension, and the median duration was 2.0 (IQR 1.0–7.0) years. The baseline SBP was 168.3 (± 12.5) mmHg and was slightly higher in participants on RAASi + CCB combinations compared to the other two subgroups. In addition, the baseline DBP was 99.2 (± 10.0) mmHg, with no significant differences between the different subgroups of SPCs. Table 1 summarizes the baseline characteristics of study participants by subgroup of SPCs used for treatment of hypertension.

Fig. 1
figure 1

Flowchart of participant recruitment. CCB calcium channel blockers, DIU diuretics, RAASi renin–angiotensin system inhibitor, SPCs single pill combinations

Full size image
Table 1 Baseline characteristics compared between three subgroups of patients treated with two-drug single pill combination
Full size table

Effectiveness of Different Dual Therapy Combinations on Blood Pressure

The distribution of active molecules in the different SPCs is shown in Fig. 2. Of the 283/377 (75.1%) patients taking an SPC containing a diuretic, 212/377 (56.2%) were taking indapamide and 71/377 (18.8%) were on hydrochlorothiazide. An SPC containing perindopril was prescribed in 59.8% of the 254 (67.4%) on RAASi, while amlodipine was prescribed in 95.9% of the 219 (58.1%) patients receiving an SPC with CCB.

Fig. 2
figure 2

Proportion of use of medications in the different drug classes

Full size image

A decrease of 34.3 (± 14.2) mmHg units of SBP was observed at week 16, with no significant difference between the three groups of SPCs (p = 0.118). There was a slightly higher decrease in SBP in participants on SPCs containing a CCB compared to participants on SPCs comprising RAASi + DIU . However, this was not statistically significant across the different time periods of follow-up (Table 4).

A mixed linear model adjusted for age, gender, and baseline SBP revealed that the different follow-up period had a significant main effect on the change in SBP from baseline to 16 weeks, F3,1122 = 36.03, p < 0.001, but there was no significant main effect on the type of SPC and SBP change. Similarly, no significant interaction effect was observed between the follow-up period and the type of SPC. From week 4 to week 16, there was a linear trend in SBP change, with a mean change at 4 weeks and 16 weeks of 28.7 (16.6) mmHg and 34.3 (14.2) mmHg, respectively. A post hoc analysis of the various follow-up periods revealed a significant difference in SBP change between weeks 4 and 16, p < 0.01.

Further analysis of the BP changes from baseline across the different time points revealed no significant interaction effect between follow-up time and type of SPC on SBP (Fig. 3).

Fig. 3
figure 3

Systolic blood pressure change from baseline through week 16. CCB calcium channel blockers, DIU diuretics, RAASi renin–angiotensin system inhibitor, PWC pairwise comparisons. CCB calcium channel blockers, RAASi renin–angiotensin system inhibitor

Full size image

There was no significant interaction effect between follow-up time and the group of SPCs on SBP F5.5,1028.3 = 1.55, p = 0.16. Likewise, the main effect of the type of SPC on BP control was not statistically significant F2,374 = 3.1, p = 0.05 (Fig. 4). However, there was a significant main effect of the period of follow-up on SBP F2.75,1028.3 = 943.3, p < 0.001. Pairwise post hoc comparisons for follow-up time and SBP were statistically significant across the different follow-up periods.

Fig. 4
figure 4

Systolic blood pressure during 16 weeks of follow-up. CCB calcium channel blockers, DIU diuretics, RAASi renin–angiotensin system inhibitor

Full size image

Furthermore, no significant interaction effect was observed for the SPC type and follow-up time on DBP change F5.7,1073.9 = 0.7, p = 0.6 Similarly, there was no significant main effect of the class of SPC on BP, F2,374 = 1.0, p = 0.3 (Fig. 5). There was a significant main effect of follow-up time on the DBP, F2.9,1073.1 = 594.4, p < 0.001.

Fig. 5
figure 5

Diastolic blood pressure during 16 weeks of follow-up. CCB calcium channel blockers, DIU diuretics, RAASi renin–angiotensin system inhibitor

Full size image

Rate of BP Control

Two-thirds of patients had their BP controlled at 16 weeks of treatment, with a slightly higher percentage in the participants on combination with CCB (Table 2). Logistic regression models were fitted to estimate the effect of SPCs on blood pressure control at week 8, while controlling for baseline SBP, gender, and age. With the RAASi + DIU group used as the reference group, the odds ratio of BP control at week 8 was 1.40 (95% CI 0.85, 2.33) and 0.79 (95% CI 0.46, 1.36) for the CCB + DIU and RAASi + CCB groups, respectively. Similarly, the odds ratio for BP control at week 16 were 1.27 (95% CI 0.75, 2.17) and 0.83 (95% CI 0.48, 1.45).

Table 2 Control rate of the office blood pressure across different follow-up periods for the three subgroups of patients treated with two-drug single pill combination
Full size table

Adverse Events

Rates of adverse events (Table 3) were not statistically significant among the three listed groups. The overall incidence of adverse events was 3.4%, and fatigue was the most frequently reported. Unproductive cough was reported by three participants.

Table 3 Side effects of the three subgroups of patients treated with two-drug single pill combination
Full size table

Discussion

This observational study of patients who had been receiving one of the three most recommended SPCs of antihypertensives (CCB + DIU, RAASi + CCB, and RAASi + DIU) in our hypertension registry in the city of Douala revealed three major findings. First, we found a similar reduction in the BP variables and control rates across the 16 weeks of follow-up, although there was a significant advantage of BP control rates for SPCs with CCB + DIU at 8 weeks. Second, two-thirds of patients had their BP controlled at 16 weeks of treatment. Third, the overall incidence of adverse events was low (3.4%) with similar rates among the three groups. The above suggests that in black patients residing in SSA, the three most recommended SPCs are equally effective in reducing and controlling BP with a slight advantage for combinations with CCB in reducing BP, as described by the CREOLE trial [16], and all three are equally safe (Table 4).

Table 4 Mean systolic blood pressure decrease from baseline across different follow-up periods for the three subgroups of patients treated with two-drug single pill combination
Full size table

Approximately 150 million Africans are hypertensive, and most are either not aware, not treated, or poorly controlled, and since this number may grow, the PASCAR proposed strategies for intensive BP control [19]. In populations residing in SSA, hypertension develops at younger ages, is often more severe in terms of BP levels, and is associated with low control rates, resulting in a higher incidence of organ damage including heart disease, kidney disease, and stroke [20, 21]. Experts advocated that hypertension control in the world and especially in Africa is feasible if SPCs were widely prescribed globally, as it could increase the number of patients with controlled hypertension by 80 million and could prevent two million stroke and heart attack events and more than 600,000 cardiovascular-related deaths over 5 years [8, 9, 19, 22]. With 62.3% of our patients having their BP controlled at 16 weeks of treatment, our findings strongly support the theory that an early prescription of effective SPCs would help better attain the BP targets.

The effectiveness of SPCs antihypertensive drugs has been proven elsewhere [7, 16], in addition to improved patient adherence (which is unfortunately low in hypertension) and more rapid and greater BP lowering compared to use of single drug medications. Whether these benefits apply to all ethnic groups at the same magnitude and for all SPCs or have similar drug adverse effects needed more investigations in the black population residing in SSA. Although it is traditionally known that black patients with hypertension usually show a reduced antihypertensive response to RAASi monotherapy, whereas they usually respond more effectively to thiazide or thiazide-like diuretics and CCBs [23], there was no special emphasis on a combination based on CCB + DIU until the CREOLE trial [16] which found that combinations with amlodipine + hydrochlorothiazide or amlodipine + perindopril are more efficacious than perindopril + hydrochlorothiazide in reducing BP in black patients as part of free two-drug combinations, with similar rates of adverse events among the three treatment groups [16]. On the basis of this trial, the ISH guidelines recommended the elective prescription of either an SPC of CCB + RAASi or CCB + DIU to treat hypertension in blacks and added a preference for ARBs due to frequent risk of angioedema with ACE inhibitors among black patients [9]. With similar rates of BP control and adverse events across the three groups over 16 weeks, our findings support the CREOLE trial in the sense that a combination of RAASi + DIU is also effective in lowering and controlling BP with no excess adverse events. Implications from these findings are that in our clinical practice settings, the three recommended combinations must be encouraged for use. If a selection was to be made, the clinician’s decision must be guided by criteria other than effectiveness or short-term safety including availability, price, and patient’s profile. The latter point is important because HMOD is common in Africa and SSA patients with hypertension have an average of 3.7 cardiovascular risk factors [24, 25].

Strengths and Limitations

This study has several limitations including non-randomization, its short treatment period, the relatively small-sized treatment subgroups, and the lack of an office BP monitoring to evaluate drug effectiveness. While comparisons between groups were limited, it should be noted that for a real-world study, our patients’ baseline characteristics were grossly similar. The SPCs used in this study were quite heterogenous (type and dosage) and we did not assess medication adherence which could have affected BP control. In addition, by excluding individuals with clinical manifestations of HMOD and those who had their treatment changed during follow-up, we likely selected a population with higher chances of having a better response to antihypertensives. Although these shortcomings may limit our ability to extrapolate our findings of the observed effectiveness and tolerance of the three SPCs evaluated, these findings are consistent with previous literature demonstrating that SPCs are effective in improving hypertension control and are well tolerated. In addition, further long-term studies in western countries demonstrated that using SPCs to treat hypertension is associated with a significant reduction in cardiovascular events [7]. Large-scale studies are needed in our populations to ascertain the long-term tolerance and effectiveness of SPCs.

Conclusion

The findings of our study suggest that the three two-drug fixed-dose combinations CCB + DIU, CCB + RAASi, and RAASi + DIU are highly effective in reducing and controlling BP in patients without clinical evidence of HMOD, with low and similar rates of short-term adverse effects. Although long-term data is still needed to investigate whether this translates into a similar reduction in major cardiovascular and renal events, we believe that SPCs rather than free dual pill therapy represents a simple and potentially low-cost intervention that could significantly reduce the burden of hypertension in our setting. Specific interventions to address the problem of hypertension in Cameroon must therefore include the incorporation of these SPCs in the essential medicines list of the country, and they must be made widely available and affordable. However, we recognize that in individuals with mildly elevated BP, monotherapy would still be appropriate.