Artigo em XML
Referências do artigo
Tradução automática
Enviar este artigo por email
Citado por SciELO 
Similares em
SciELO 
Permalink
KEY POINTS
• A central problem of arterial hypertension is to improve blood pressure control rates to reduce the disease burden.
• Combinations of antihypertensives in a single-pill (SPCs), also called “fixed-dose combinations”, have been shown to improve adherence and persistence to medication compared to the administration of the same antihypertensives but in several pills, also called “free-dose combinations”.
• However, there are no randomized controlled studies showing that SPCs reduce cardiovascular events and mortality compared with free-dose combinations in patients without differences in adherence/ persistence.
Elevated blood pressure (BP) continues to be the leading cause of death worldwide1, causing around 10.4 million deaths each year, and it is estimated that this situation may worsen after the COVID-19 pandemic2. On the other hand, control rates are low worldwide, particularly in low- and middle-income countries3, a situa tion that is also likely to worsen as a result of COVID-194.
For these reasons, measures that help to im prove arterial hypertension control are likely to have a favorable impact on the burden of this disease. In this sense, current recommendations and guidelines for the treatment of arterial hy pertension5-7 propose to combine antihyperten sive drugs to improve BP control, highlighting the potential advantages of antihypertensive drugs combinations in a single pill (SPCs), also known as “fixed-dose combinations”.
Although SPCs improve adherence and per sistence8, they have some disadvantages; it could hinder the dosage adjustments, leading to potential overdosing or underdosing9, and may reduce the ability to identify the cause of an ad verse event related to one of the components. Furthermore, SPCs could produce therapeutics mistakes such as duplication in the prescription of some drugs10. Finally, although this kind of treatment could reduce per hospitalization cost, the direct cost in pharmacy may be higher than the individual use of antihypertensive drugs in two or more pills, also named “free-dose combi nations” (Free-DCs)11. Indeed, SPCs has problems for social security coverage in many countries as Argentina where a recently published multicen tric study did not find a better level of adherence in patients who used fixed‐dose antihyperten sive combinations12.
Although it can be speculated that SPCs, through better adherence compared with Free- DCs, should reduce even more cardiovascular (CV) events and mortality, there is little information regarding this issue. For this reason, the aim of this review was to provide a descriptive analysis the differences in CV outcomes be tween SPCs antihypertensive drugs treatments vs. Free-DCs treatments.
Methods
We performed a review of the full text articles pub lished in PubMed in the period February 1, 2011, to Abril 30, 2023, which compared CV outcomes in hypertensive patients treated with antihypertensive drugs in SPCs vs. Free-DCs. The search terms were: fixed-dose combina tions; single-pill combinations; cardiovascular outcomes, fixed-dose combinations; cardiovascular outcomes, single-pill combinations; clinical outcomes, fixed-dose combinations; clinical outcomes, single-pill combina tions; cardiovascular outcomes, fixed-dose combinations, antihypertensive; cardiovascular outcomes, single-pill combinations, antihypertensive; clinical outcomes, fixed-dose combinations, antihypertensive; clinical outcomes, single-pill combinations, antihypertensive.
The inclusion criteria for the studies to be reviewed were the following: a) being studies carried out in hy pertensive patients, b) comparing treatments with anti hypertensive drugs in SPCs vs. combinations of similar numbers and classes of antihypertensive drugs in ≥ two pills (Free-DCs) and, c) analyze cardiovascular events and/ or mortality. Studies carried out in hypertensive patients but selected for some conditions (chronic kidney disease (CKD), ischemic heart disease, diabetes mellitus, resistant hypertension, etc.), and those studies that, in addition to antihypertensive, also contain other drugs (statins, aspi rin, polypill, etc.), were excluded.
With the search terms previously mentioned 354 stud ies were identified, 341 articles excluded for the reasons detailed in Figure 1. After removing duplicates, 8 stud ies remained for revision [15-22]; 2 studies13,14, that were found in the references of other articles, were also added to the analysis (Fig. 1). Results of each study on charac teristics, adherence, BP levels, achieved target BP and CV outcomes were summarized in a narrative analysis.
Analyzed studies
The characteristics of the 10 studies are shown in Ta ble 1. None of them had a randomized controlled design. Most studies compared combinations of two antihyper tensive drugs14-20, two compared combinations of two or more antihypertensives13,22 and, another study, combina tions of three antihypertensives21. Most of the combina tions were of angiotensin receptor blockers (ARBs) or an giotensin-converting enzyme inhibitors (ACEIs) + calcium channels blockers (CCBs)14,17-19,22 or combinations of ARBs or ACEIs + diuretics [15,16,20].
Tabla 1 Characteristics of the studies which evaluated cardiovascular outcomes and mortality with single-pill antihypertensive drugs combinations versus free-dose combinations
Belsey JD13 investigated CV event-free survival in an observational study on a retrospective cohort in United Kingdom subjects. Combination of acute myocardial in farction (AMI), stroke, mortality, heart failure (HF), acute coronary syndrome, transient ischemic attack, bypass, and percutaneous coronary angioplasty were the events considered over a minimum follow-up period of 5 years after initiation of therapy. Authors compared patients treated with SPCs of two or more any antihypertensive drugs of different classes (n = 9929) with another group (n = 18 665), with two or more antihypertensive drugs of different classes in Free-DCs. The groups remained un balanced after matching. There were 2173 events in SPCs group (21.9%) and 6285 in Free-DCs group (33.7%); Higher CV event-free survival in the SPCs group was observed, 27% risk reduction in the crude data (HR 0.73, 95% CI 0.70- 0.76) and 26% in the adjusted model (HR 0.74, 95% CI 0.70- 0.77). It is important to note that both before and after treatment the BPs were higher in the Free-DCs group: be fore treatment Free-DCs 159.7 ± 17.4 mmHg / 91.5 ± 9.1 mmHg vs SPCs 154.9 ± 24.3 mmHg / 90.5 ± 12.8 mmHg (p > 0.001); after treatment Free-DCs 148.2 ± 13.0 mmHg / 82.7 ± 7.1 mmHg vs SPCs 147.8 ± 12.5 mmHg / 82.4 ± 7.5 mmHg (p = 0.011 for systolic BP, p = 0.001 for diastolic BP). Furthermore, target achievement (BP ≤ 140/90 mmHg) was also worse in Free-DCs group (23.8% vs. 24.9%, p = 0.04). Adherence was not reported in this study.
Ferrario CM et al14 in a retrospective observational cohort study compared three cohorts: 1-SPCs of amlo dipine + olmesartan (n = 4864), 2- SPCs of amlodipine + benazepril (n = 12 051), and 3- Free-DCs of amlodipine + ARBs (n = 7748). Means ages were different, 53.8 ± 11.2, 56.0 ± 11.91 and 60.7 ± 12.52 years old for cohort 1, 2 and 3, respectively (p < 0.001 between cohorts). Mean days follow-up were also different, 543.0 ± 113.7, 625.4 ± 133.1 and, 585.3 ± 132.4, p < 0.001. They found 35% higher risk (HR 1.35, 95% CI 1.15-1.59) in the primary combination outcome of HF, stroke, AMI, diagnosis of ischemic heart disease [IHD] and surgery related to AMI and IHD with Free-DCs compared to SPCs of amlodipine + olmesartan. Significant differences were observed in HF (HR 1.32, 95% CI 1.09-1.60), stroke (HR 1.64, 95% CI 1.49-2.36) and AMI/ IHD-related surgery (HR 1.43, 95% CI 1.01-2.03). It should be noted that there was a higher percentage (44.12%) of days covered [PDC] with medication > 80% (good-adher ence indicator) in SPCs amlodipine + olmesartan group compared with those in the SPCs amlodipine + benaz epril (36.46%, p < 0.001) and Free-DCs (19.53%, p < 0.001). The study does not report BP values or percentage of pa tients who achieved treatment goals.
Ho CT et al15 conducted an observational real-world study from the National Health Insurance Research Da tabase of Taiwan. Patients newly diagnosed with hyper tension, and prescribed with SPCs (n = 13 176) versus Free-DCs (n = 4392) of ACEIs or ARBs + thiazide diuretic were compared. Propensity score matching to eliminate differences in baseline characteristics of the two groups was performed. Primary endpoint was the combination of all-cause mortality, AMI, stroke, and coronary revascu larization. Secondary endpoints included hospitalization for HF, a new diagnosis of CKD and the start of dialysis. All patients were followed up for at least one year or till the occurrence of clinical endpoints, whichever came first. Mean duration of follow‐up were 887.89 (456.09) days and 830.22 (462.50) days (p < 0.001) for SPCs and Free-DCs groups, respectively. SPCs reduced 15% the primary end point (HR 0.85, 95% CI 0.74-0.97), and also the hospitaliza tion for HF (HR 0.76, 95% CI 0.6-0.95) and initiation of di alysis (HR 0.69, 95% CI 0.53-0.89) compared with Free-DCs regimens. Percentage of patients with good-adherence (PDC > 80%) was higher in SPCs group (35.4%) compared with Free-DCs group (28.2%) (p < 0.001). It is important to note that the benefit in CV and renal outcomes observed with SPCs, compared to Free-DCs, disappears when both groups are compared but with a PDC > 80%. Unfortunate ly, BP values and the percentage of patients that reached BP goals are not reported.
Sicras Mainar A et al16 performed a multicenter obser vational study on hypertensive patients > 30 years old, from six primary care centers and two hospitals, in Cata luña. The authors investigated the relationships between CV events incidence and compliance, persistence, and BP control level. Patients were followed for two years and treated with SPCs of ACEIs or ARBs + diuretics (n = 1112) vs. Free-DCs from the same classes of antihypertensive drugs (n = 493).
Those with Free-DCs treatments, compared with SPCs treatments, were older (70.7 ± 12.0 vs. 68.7 ± 12.1 years old, p <0.001) and had more IHD (13.2% vs. 9.1%, p = 0.044), organ failure cardiac, hepatic or renal (12.8% vs. 8.6%, p = 0.042), and diabetes mellitus (30.6% vs. 28.3%, p = 0.013). Cumulative incidence of ischemic or hemorrhagic stroke and transient ischemic attack in Free-DCs was 4.6% vs. 2.4% in SPCs (p = 0.041). SPCs treatment, compared with Free-DCs, showed better therapeutic good-adherence (PDC ≥ 80%: 56.3% vs. 41.8%, p <0.001), longer treatment persistence (62.1%, 95% CI 56.3-67.9 vs. 49.7%, 95% CI 38.5-60.9) and higher BP control (48.9%, 95% CI 43.0-54.8 vs. 46.7%, 95% CI 35.6-57.8). Patients with SPCs also had a significant reduction in systolic and diastolic BP between the beginning and the end of the study (139.6 ± 16.0 vs. 136.9 ± 17.6 mmHg, p = 0.030, and 79.1 ± 10.1 vs. 77.1 ± 10.5 mmHg, p < 0.001). In contrast, those with Free-DCs only significantly reduced the diastolic BP (139.0 ± 18.3 versus 138.1 ± 16.4 mmHg, p = NS, and 78.5 ± 9.8 versus 77.6 ± 10.6 mmHg, p = 0.046).
Simons EL et al17 in an observational study of 12 433 Australians, found 4-year mortality of 8% in the group initially treated with SPCs of amlodipine + perindopril (n = 9340) and 18% in the group initially treated with the same drugs in Free-DCs (n = 3093). In the follow-up period, witching to similar classes of antihypertensive drugs (calcium channel blockers + ACEIs or ARBs) was ac cepted. It should be noted Free-DCs group had a higher risk due to older age (71.5 vs. 67.8 years old, p < 0.001), longer duration of antihypertensive treatment before the study, greater evidence of diabetes mellitus (20% vs. 16%, p < 0.001) and hyperlipidemia (51% vs. 43%, p < 0.001). In the Free-DCs group mortality risk was higher than in the SPCs group (HR 2.81, 95% CI 2.42-3.26 in the univariate model; HR 1.83, 95% CI 1.55-2.16 in the adjusted multi variate model). Although SPCs group had higher median persistence than Free-DCs (42 months, 95% CI 33 to > 43 months vs. 7 months, 95% CI 5-9, respectively), mortal ity risk was not adjusted for persistence time. Authors acknowledge that this very large difference in mortality was unexpected and is likely to be an overestimate, pos sibly due to residual confounding by other unmeasured variables. This study also does not provide BP values or report the percentage that achieved BP targets.
Tung YC et al18 in a retrospective analysis of 16 505 Asians from Taiwan, compared two hypertension treat ment strategies: a SPCs of amlodipine + valsartan (n = 3301) vs. Free-DCs of ARBs + CCBs (n = 13 204). To identify an appropriate Free-DCs combination group, propensity score matching was utilized. After a mean follow-up of 15.2 months, SPCs group had significantly lower hospitalization rates (14.57% vs. 18.43%, p < 0.001). SPCs group also had a better CV event-free survival (HR 0.83, 95% CI 0.73-0.94) at expense of a decrease in HF rates (2.12% vs. 3.26%, p < 0.001), malignant dysrhythmia (0.18% vs. 0.42%, p = 0.021), and percutaneous coronary interven tion (0.76% vs. 1.26%, p = 0.015), but not from AMI, stroke and coronary artery bypass grafting. Although SPCs group, compared with Free-DCs group, had higher PDC (80.35% vs. 72.57%, p < 0.001) and better persistence (266 vs. 225 days, p < 0.001), CV outcomes were not adjusted for adherence/persistence. This study does not provide information about BP values or the percentage of patients who achieved BP goals.
Using National Health Insurance Research Database of Taiwan, Tung YC et al19, compared patients taking SPCs of ARBs + CCBs (n = 1136) vs. Free-DCs (n = 4544). Propensity score matching was performed to balance the potential difference in the two study groups. Mean follow-up dura tion was 2.1 years. SPCs were associated with a higher percentage of patients with good-adherence (PDC ≥ 80%: 64.97% vs. 56.88%, p < 0.001). Similarly, medication per sistence was better in SPCs group than in Free-DCs group (293.8 vs. 275.1 days, p < 0.001.). SPCs group had a 28% re duction in the risk of the primary CV event (HR 0.72, 95% CI 0.54-0.95), that was a combination of total mortality, AMI, stroke, coronary revascularization, hospitalization for unstable angina, and sudden cardiac arrest resuscita tion. SPCs was also associated with better CV event-free survival (log-rank p = 0.021) but lost statistical signifi cance when groups were compared according to similar adherence levels. At the secondary endpoints, SPCs group was associated with 29% lower risk for HF hospitalization (HR 0.71, 95% CI 0.51-0.99). There were no differences be tween the two groups in risks of a new diagnosis of CKD or the start of dialysis. This study does not provide BP re cords.
In a Canadian observational study with a median follow-up of 1826 days, Verma AA et al20 used a high-di mensional propensity score matching to identify compa rable groups23 in a retrospective cohort of hypertensive patients. They found, in the intention-to-treat analysis, 11% reduction in the risk (HR 0.89, 95% CI 0.81-0.97) of reaching the primary event (combination of death or hos pitalization due to AMI, HF, or stroke) when were used SPCs of ARBs or ACEIs + hydrochlorothiazide (n = 6675) vs. Free-DCs of the same classes of antihypertensive drugs (n = 6675). Risk reduction was due to a 15% decrease in the risk of death (HR 0.85, 95% CI 0.77-0.94). There were no significant differences in risk of AMI, stroke, and HF. The PDC was significantly higher in the group with SPCs than in the group with Free-DCs (70%, IQR 19 ± 98 and 42%, IQR 11 ± 91, respectively, p < 0.01). It should be noted that the statistical significance of the reduction in primary outcome does not hold when only the adherent individu als of the groups are compared. This study does not pro vide BP values or report the percentage of subjects who achieved BP targets.
Wang X et al21, in a retrospective cohort of 10 836 elder ly hypertensive subjects from Texas Medicare Advantage Plus, evaluated the risk of hospitalization for CV events in patients treated with a triple combination of antihy pertensive drugs. The follow-up period was one year. The study comparing any triple SPCs therapy (n = 336) with any double SPCs plus a third agent (n = 470), and any free-dose triple combination therapy (n = 10 030). Risk of hospitalization for CV events was higher in the double SPCs plus a third agent, and free-dose triple combination therapy groups (HR 3.82, 95% CI 1.80-8.12 and HR 3.65, 95% CI 1.43-9.31, respectively). It should be noted that in patients treated with triple SPCs numerous variables associated with the risk of hospitalization were less fre quent (diabetes, HF, depression, hyperlipidemia, history of previous hospitalization, neuropathy). In this study, adherence was not significantly associated with a lower risk of hospitalization for CV disease, although the time at which it was evaluated (6 months) could be considered inappropriate. Unfortunately, the authors do not provide BP values or the percentage that reached BP goals.
The recently published START study, by Schmieder RE et al22, have provided further evidence in favor of SPC. In this retrospective cohort study, data from hypertensive patients ≥18 years from a German statutory health fund (AOK PLUS) treated with renin-angiotensin system com binations given as single pill or identical multipills cover ing the years 2012 to 2018 were compared after 1:1 pro pensity score matching. More than 160 000 patients with hypertension and who received one of the following four antihypertensive combination ramipril/amlodipine, can desartan/amlodipine, valsartan/amlodipine, or valsartan/ amlodipine/hydrochlorothiazide were identified. After propensity score matching, data from 28 999 hyperten sive patients with one of the 4 SPCs combinations were compared to 28 999 hypertensive patients with identi cal drug combinations given as Free-DCs. Differences in incidence of outcomes and time to first respective event were reported as incidence rate ratios (IRRs) and hazard ratios (HRs). In all 4 comparisons between SPCs and Free- DCs, the authors found a lower mortality rate in the SPC: valsartan/amlodipine IRR, 0.761 (95% CI, 0.683-0.848); candesartan/amlodipine IRR, 0.538 (95% CI, 0.284-0.980); candesartan/amlodipine IRR 0.526 (95% CI, 0.463-0.596); valsartan/amlodipine/hydrochlorothiazide IRR, 0.515 (95%CI 0.375-0.709). Furthermore, patients treated with any of the 4 SPCs analyzed had a lower risk for a pre defined composite outcome consisting of all-cause death and all-cause hospitalization, lower than Free-DCs (p < 0.001). Regarding the specific cardiovascular end points, a significant lower event rate was observed in 15 out of 20 comparisons performed. Comparing the 4 drug combination groups, patients on SPCs had a significantly lower incidence (P<0.05) of coronary artery disease and heart failure. However, only ramipril/amlodipine showed superiority in SPC to prevent AMI, but this combination had no advantages in the prevention of cerebrovascular disease. Cardiovascular mortality was not communicat ed. The percentage of patients that were persistent to an tihypertensive drug combinations 1 year after start of ob servation was significantly higher under SPCs. Compared with the respective SPC group, the proportion of patients who were persistent using Free-DC was 20% less in the valsartan/amlodipine group, 30% less in the candesartan/ amlodipine with Free-DCs, 24% less in the candesartan/ amlodipine group, and 49% less in the valsartan/amlo dipine/hydrochlorothiazide group with Free-DC. Since no adjustment by adherence/persistence was made, the benefit observed could be related to the improvement in these variables. Unfortunately, the authors do not provide BP values or the percentage that reached BP goals.
Discussion and conclusions
Despite the “advice” of the guidelines on arte rial hypertension about the convenience of us ing SPCs, several observations should be consid ered concerning the CV outcomes and mortality.
First, no appropriate size, controlled and randomized trials have been carried out that compare primary or secondary prevention of CV events and mortality with similar numbers and classes of antihypertensive drugs in SPCs vs. Free-DCs. The information available comes only from observational studies, databases or retrospective cohorts. Despite the matching of groups and adjustments for confounding variables, there is no doubt that biases or hid den confounding variables may persist in these types of studies.
Second, the important role played by better adherence/persistence of treatment with SPCs use, which has been recently revised8, seems clear in reducing CV events. To such a point that studies showed a lower risk of mortality and CV events with SPCs do not reach to sta tistical significance when only adherent pa tients are included in the comparisons or af ter stratifying the groups by similar adherence levels15,19,20.
Third, it would be expected that the groups with better adherence would have not only low er BP values and higher percentages of hyper tension control, but also better coverage and/ or health systems. Thus, use of SCPs could be a marker of social advantage and this association explains the reduction in all-cause mortality observed in some studies. Unfortunately, eight of the ten studies do not report BP values or the percentage of individuals who achieved BP goals with treatment. Furthermore, one of the two studies that showed the BP values and the percentage of patients who reached the target BP13, does not describe how BP was measured. The other study16 found in the group with SPCs better adherence and persistence, significant BP decrease, better BP control, and lower incidence of stroke, but did not adjust the outcomes for adherence.
Supporting the relationship between SPCs use and CV event reduction, a prospective open-label study showed that a triple SPC therapy, containing in a single pill ACEI + diuretic + CCB, was more effective than Free-DCs in left ven tricular mass index reduction and left ventricu lar hypertrophy regression24, although this is a small study.
Therefore, with the information available to date, it is not possible to confirm a relationship, although there probably exists, between SPCs use, greater adherence, greater BP decrease, bet ter BP control and reduction in CV events and mortality.
Finally, the use of SPCs does not necessarily ensure good-adherence. Other predictors of non-adherence to treatment with antihypertensive drugs should also be considered, such as low ed ucational level12, young age25, female sex24, num ber of medications25 and use of diuretics12,25. In consequence, measures to improve adherence to treatments5 and hypertension control should be added to the use of SPCs, such as self-moni toring of BP at home, telemonitoring, reminders, elimination of barriers to medication access (as direct costs) and multidisciplinary approaches including nurses, pharmacists and community health workers25.
In conclusion, randomized and controlled studies are necessary to evaluate if SPCs of anti hypertensive drugs reduce CV events and mor tality, beyond the improvement of the adherence to antihypertensive treatment. Furthermore, there is also no evidence that the sequence SPCs → greater adherence → greater BP decrease → better BP control → reduction in CV events and mortality is fulfilled. Thus, to reduce CV out comes, it seems prudent not only to advise for a wide use of SPCs, but also strongly recommend more comprehensive measures that can im prove the adherence to chronic treatments.
