Articles

Quarter-dose quadruple combination therapy for initial

treatment of hypertension: placebo-controlled, crossover,
randomised trial and systematic review
Clara K Chow, Jay Thakkar, Alex Bennett, Graham Hillis, Michael Burke, Tim Usherwood, Kha Vo, Kris Rogers, Emily Atkins, Ruth Webster,
Michael Chou, Hakim-Moulay Dehbi, Abdul Salam, Anushka Patel, Bruce Neal, David Peiris, Henry Krum*, John Chalmers, Mark Nelson,
Christopher M Reid, Mark Woodward, Sarah Hilmer, Simon Thom, Anthony Rodgers

Summary
Background Globally, most patients with hypertension are treated with monotherapy, and control rates are poor Lancet 2017; 389: 1035–42
because monotherapy only reduces blood pressure by around 9/5 mm Hg on average. There is a pressing need for Published Online
blood pressure-control strategies with improved efficacy and tolerability. We aimed to assess whether ultra-low-dose February 9, 2017
http://dx.doi.org/10.1016/
combination therapy could meet these needs.
S0140-6736(17)30260-X
See Comment page 989
Methods We did a randomised, placebo-controlled, double-blind, crossover trial of a quadpill—a single capsule
*Deceased
containing four blood pressure-lowering drugs each at quarter-dose (irbesartan 37·5 mg, amlodipine 1·25 mg,
The George Institute for Global
hydrochlorothiazide 6·25 mg, and atenolol 12·5 mg). Participants with untreated hypertension were enrolled from Health, University of Sydney,
four centres in the community of western Sydney, NSW, Australia, mainly by general practitioners. Participants were Sydney, NSW, Australia
randomly allocated by computer to either the quadpill or matching placebo for 4 weeks; this treatment was followed (Prof C K Chow PhD,
by a 2-week washout, then the other study treatment was administered for 4 weeks. Study staff and participants were J Thakkar MBBS,
A Bennett MPhil,
unaware of treatment allocations, and masking was achieved by use of identical opaque capsules. The primary Prof G Hillis PhD, K Vo BSc,
outcome was placebo-corrected 24-h systolic ambulatory blood pressure reduction after 4 weeks and analysis was by K Rogers PhD, E Atkins PhD,
intention to treat. We also did a systematic review of trials evaluating the efficacy and safety of quarter-standard-dose R Webster PhD, A Salam PhD,
blood pressure-lowering therapy against placebo. This trial is registered with the Australian New Zealand Clinical Prof A Patel PhD,
Prof B Neal PhD, D Peiris PhD,
Trials Registry, number ACTRN12614001057673. The trial ended after 1 year and this report presents the final analysis. Prof J Chalmers PhD,
Prof M Woodward PhD,
Findings Between November, 2014, and December, 2015, 55 patients were screened for our randomised trial, of whom Prof A Rodgers PhD); Westmead
21 underwent randomisation. Mean age of participants was 58 years (SD 11) and mean baseline office and 24-h Hospital, Sydney, NSW,
Australia (Prof C K Chow,
systolic and diastolic blood pressure levels were 154 (14)/90 (11) mm Hg and 140 (9)/87 (8) mm Hg, respectively. One J Thakkar); Kildare Road Medical
individual declined participation after randomisation and two patients dropped out for administrative reasons. The Centre, Sydney, NSW, Australia
placebo-corrected reduction in systolic 24-h blood pressure with the quadpill was 19 mm Hg (95% CI 14–23), and (M Burke PhD); The University
of Sydney, Sydney, NSW,
office blood pressure was reduced by 22/13 mm Hg (p<0·0001). During quadpill treatment, 18 (100%) of 18 participants
Australia (Prof T Usherwood MD,
achieved office blood pressure less than 140/90 mm Hg, compared with six (33%) of 18 during placebo treatment Prof S Hilmer PhD,
(p=0·0013). There were no serious adverse events and all patients reported that the quadpill was easy to swallow. Our Prof A Rodgers); The University
systematic review identified 36 trials (n=4721 participants) of one drug at quarter-dose and six trials (n=312) of of Western Australia, Perth,
WA, Australia (Prof G Hillis);
two drugs at quarter-dose, against placebo. The pooled placebo-corrected blood pressure-lowering effects were
Johns Hopkins Bloomberg
5/2 mm Hg and 7/5 mm Hg, respectively (both p<0·0001), and there were no side-effects from either regimen. School of Public Health,
Baltimore, MD, USA
Interpretation The findings of our small trial in the context of previous randomised evidence suggest that the benefits of (M Chou MBBS); Imperial
College, London, UK
quarter-dose therapy could be additive across classes and might confer a clinically important reduction in blood pressure.
(H-M Dehbi MRes,
Further examination of the quadpill concept is needed to investigate effectiveness against usual treatment options and Prof S Thom PhD, Prof B Neal);
longer term tolerability. Department of Epidemiology
and Preventive Medicine,
Monash University, Melbourne,
Funding National Heart Foundation, Australia; University of Sydney; and National Health and Medical Research
Vic, Australia (Prof H Krum);
Council of Australia. Menzies Institute for Medical
Research, University of
Introduction Many factors contribute to poor blood pressure control, Tasmania, Hobart, Tas,
Australia (Prof M Nelson PhD);
High blood pressure is a leading cause of preventable including low adherence rates, complex guidelines Curtin University, Perth, WA,
morbidity and mortality,1 and the benefits of blood recommending multiple up-titration steps, and treatment Australia (Prof C M Reid PhD);
pressure-lowering treatments are well established.2,3 inertia. Most treated patients only receive monotherapy,4 Charles Perkins Centre,
Despite the plethora of blood pressure-lowering medicines which has low potency even at high doses.5 Furthermore, University of Sydney, Sydney,
NSW, Australia (Prof C K Chow,
available, and the fact that, once found to have high blood the increasingly strong evidence showing benefits of more Prof B Neal); and Royal Prince
pressure, most patients receive some treatment, findings intensive blood pressure lowering6,7 highlights the need for Alfred Hospital, Sydney, NSW,
of multiple large-scale population studies show poor blood new treatment strategies that are more efficacious while Australia (Prof A Patel,
pressure control in many patients globally.4 remaining tolerable. Low-dose combination therapy holds Prof B Neal)

www.thelancet.com Vol 389 March 11, 2017 1035

 Articles

Correspondence to:
Prof Clara K Chow, The George
Institute for Global Health—
Australia, Camperdown,
NSW 2050, Australia
cchow@georgeinstitute.org.au
Research in context
Evidence before this study
Findings of a systematic review and meta-analysis of
354 randomised, double-blind, placebo-controlled trials of
blood pressure-lowering therapy showed that
half-standard-dose achieved almost 80% as much blood
pressure lowering as did standard dose, and that the blood
pressure-lowering effect of different classes of drugs was
additive. Although most benefits were maintained at half-dose,
most side-effects were avoided. Findings of another trial
showed that four drugs at quarter-dose achieved greater blood
pressure reduction than did each component at standard dose.
Added value of this study
We systematically reviewed the literature on
placebo-controlled, quarter-dose blood pressure-lowering
therapy and found that one and two drugs at quarter-dose led
to placebo-corrected blood pressure reductions of 5/2 mm Hg
and 7/5 mm Hg, respectively. These reductions were not
associated with any difference in side-effects compared with
placebo. Our trial provides the first placebo-controlled data on
four quarter doses, indicating a 22/13 mm Hg reduction in
blood pressure.
Implications of all the available evidence
Our study provides proof-of-concept for an innovative
approach of using ultra-low-dose quadruple combination
therapy to achieve substantial blood pressure reductions.
Further studies are needed to examine the generalisability of
these findings and assess the longer term effects on efficacy,
safety, and tolerability compared with usual care.

For the British National considerable promise in this regard since at low doses
Formulary see https://www.bnf. most side-effects are avoided and most benefits are
org
maintained.8 However, there is uncertainty about effects at
ultra-low doses and whether combinations can achieve
clinically relevant blood pressure reductions. We, therefore,
sought to assess efficacy and tolerability of ultra-low-dose
combination therapy by conducting a systematic review of
quarter-dose blood pressure-lowering therapies and a
randomised trial of a quadpill containing four common
blood pressure-lowering medications, each at quarter-
dose.
Methods
Systematic review
We did a systematic review of all randomised trials of
quarter-dose blood pressure therapy, identifying
potentially relevant studies from searches of Embase,
MEDLINE, and the Cochrane Central Registry of
Controlled Trials, with each source searched from
inception to June, 2016; we also searched the websites of
the US Food and Drug Administration and the European
Medicines Agency. MEDLINE search terms are provided
See Online for appendix in the appendix (pp 1, 2). We searched trial registers for
any ongoing trials, including WHO’s International
Clinical Trials Registry platform, the Australian New
Zealand Clinical Trials Registry, and the Clinical Trials
Registry—India. We also retrieved studies from
reference lists of key clinical trials, systematic reviews,
and published articles, and reviewed the reference
lists of eligible studies and systematic reviews
(appendix pp 3–6, 14). We included randomised controlled
trials of adult participants (≥18 years of age) in which
quarter-standard-dose blood pressure-lowering drugs
were compared with placebo for the following drug classes:
angiotensin-converting-enzyme inhibitors; angiotensin
receptor II blockers; β blockers; calcium-channel blockers;
and thiazide and thiazide-like diuretics. Quarter-dose was
a quarter of the standard dose, defined as the most
frequently reported usual maintenance dose recorded by
the British National Formulary, Martindale: The Complete
Drug Reference,9 and the Monthly Index of Medical
Specialties.10 Two reviewers (AB, MC) extracted data
independently using a standard extraction form. A third
reviewer (AR) resolved any differences. We analysed data
using Comprehensive Meta-Analysis, version 3 (Biostat,
Englewood, NJ, USA). We used a fixed-effect model to
estimate the effects of quarter-dose blood pressure-
lowering therapy on blood pressure lowering and on
adverse events, compared with placebo. We assessed the
effect on blood pressure using the mean change in systolic
blood pressure and diastolic blood pressure from baseline
to end of study, with stan­ dar­
d­isation to a baseline of
150/95 mm Hg.8 Adverse events included all that were
reported by trials at follow-up.
Clinical trial design and participants
The Quadpill study was a randomised, placebo-controlled,
double-blind, crossover trial (figure 1). We recruited
participants from four centres in the community, predomi­
nantly through general practices, in western Sydney, NSW,
Australia. We judged participants eligible if they met the
following inclusion criteria: adults aged 18 years and older;
office systolic blood pressure greater than 140 mm Hg or
diastolic blood pressure greater than 90 mm Hg, or both,
on two readings on separate days; baseline ambulatory
systolic blood pressure greater than 135 mm Hg or diastolic
blood pressure greater than 85 mm Hg, or both; and not
taking any blood pressure drugs. Exclusion criteria
included: definite contraindication to one or more
component agents in the quadpill; the responsible clinician
judged that a change in current therapy would place the
patient at risk; severe or accelerated hypertension;
pregnancy; inability to provide informed consent; and
medical illness with anticipated life expectancy less than
3 months.
The study protocol was approved by the human research
and ethics committee at the University of Sydney.
We obtained informed consent from all participants.

1036 www.thelancet.com Vol 389 March 11, 2017


Randomisation and masking
We randomly allocated participants in a 1:1 ratio to either:
a group receiving the quadpill for 4 weeks, followed by a
2-week placebo washout, then placebo for 4 weeks; or a
group receiving placebo, then washout, then the quadpill
for the same periods. The quadpill was a single
encapsulated pill containing four common blood pressure-
lowering drugs, each at quarter-standard-dose, as defined
above (irbesartan 37·5 mg, amlodipine 1·25 mg, hydro­
chlorothiazide 6·25 mg, and atenolol 12·5 mg).
We selected the most commonly used drug from each
class in Australia. We obtained quarter-doses by halving
half-dose tablets using a pill-splitting device, without
crushing, and we weighed the pills to ensure accuracy of
halving doses. We then encapsulated the quarter doses
using gelatine capsules (DBCaps; Capsugel, Morristown,
NJ, USA). All trial medicines were prepared and packaged
at a manufacturing facility licensed with a Certificate of
Good Manufacturing Practice by the Therapeutic Goods
Administration of Australia.
Treatment allocations were done at random via a
computer-assisted randomisation sequence and were
masked to study staff enrolling participants, care
providers, outcome assessors, and participants. The
placebo capsule appeared identical and contained
four placebo tablets of similar weight to those in the
quadpill. We gave participants a single daily capsule
quadpill or placebo throughout the trial. We instructed
patients to take the capsules at the same time each day,
preferably in the morning. In addition to the study drugs,
we provided all participants with education on healthy
lifestyle options, as recommended by local blood pressure
management guidelines that were current at the time.11
Procedures
We did 24-h ambulatory blood pressure monitoring
four times (figure 1): at baseline (off study drug);
at 4 weeks (period one treatment or placebo); at 6 weeks
(after 2-week placebo washout); and at 10 weeks (period
two treatment or placebo). Ambulatory blood pressure
monitoring machines were calibrated according to the
manufacturer’s specification. We recorded office blood
pressure three times at each study visit using an Omron
T9P blood pressure monitor (HEM-759-C1; Omron
Healthcare, Hoofddorp, Netherlands). We averaged the
second and third readings for study analysis. We took
readings while participants were seated and rested, but
not while they had been left unattended.11 Furthermore,
at week 4 and week 10, we did blood biochemistry
analyses and administered a questionnaire for clinical
side-effects and medication compliance. At study end,
we assessed drug acceptability and tolerability.
We recorded all adverse events and asked patients
specifically about clinical adverse events possibly
associated with blood pressure-lowering drugs—ie,
dizziness, blurred vision, syncope or collapse, chest pain
or angina, shortness of breath, cough, wheeze, pedal
www.thelancet.com Vol 389 March 11, 2017 1037
Articles
oedema, skin rash, or itching. The study doctor (JT) and
a clinical cardiologist, in consultation with the principal
investigator if needed, judged the severity of adverse
events and whether they were related to study treatment.
We provided study drugs and investigations at no cost to
participants and we reimbursed nominal amounts to
cover travel and parking costs.
Outcomes
The primary outcome was reduction in mean 24-h
systolic blood pressure at 4 weeks using ambulatory
blood pressure monitoring. Secondary outcomes
included: reduction in mean 24-h diastolic blood pressure
and in daytime and night-time systolic and diastolic
blood pressure at 4 weeks; reduction in office systolic and
diastolic blood pressure, as measured by a standardised
automated blood pressure cuff; the proportion of
participants with controlled blood pressure at 4 weeks
(defined as <135/85 mm Hg 24-h ambulatory blood
pressure and <140/90 mm Hg office blood pressure);
adverse events and prespecified adverse events with
laboratory-associated parameters (ie, rise in alanine
aminotransferase and aspartate aminotransferase of
more than three times the upper limit of normal, or
doubling if baseline levels were known to be elevated;
a drop in estimated glomerular filtration rate by more
than 20%, as estimated from serum creatinine; and a
change in levels of sodium, potassium, and uric acid);
and assessment of acceptability and tolerability.
Patients with untreated high blood pressure
(two office blood pressure measures on two
different days >140/90 mm Hg)
Baseline visit (week 0)
Clinical questionnaire, 24-h ambulatory blood pressure,
blood tests
Randomise
Week 0–4 Quadpill* Placebo
Visit 2 (week 4): 24-h blood pressure, blood tests, adverse events
Week 4–6 2-week washout 2-week washout
Visit 3 (week 6): 24-h blood pressure
Week 6–10 Placebo Quadpill*
Final visit (week 10): 24-h blood pressure, blood tests, adverse events,
acceptability questionnaire
Figure 1: Study design
*Quadpill comprises irbesartan 37·5 mg, amlodipine 1·25 mg, hydrochlorothiazide 6·25 mg, and atenolol 12·5 mg.
 Articles

prognostic covariates. We reported compliance to study

55 screened drug using data for pills (doses) taken and missed doses
over the study period. We used a linear mixed model to
34 not randomised estimate the effect of the treatment on change in mean
11 medically ineligible 24-h systolic blood pressure from baseline for each
10 too busy or declined participation treatment period (primary outcome), according to the
5 white-coat hypertension
9 not contactable Kenward and Roger approach.12 We included all available
data in the model; no missing data were imputed. If a
patient had missing data for one period, we used data
21 eligible for randomisation
from the available period. We did a sensitivity analysis
including only patients with data available from both
1 patient declined study drug initiation periods. We also adjusted the denominator degrees of
freedom of Kenward and Roger13 to optimise for the small
sample size.
20 patients in trial
We tested for carry-over with an unpaired t test of the
main outcome, with order as an effect. We tested period
2 participants withdrew effect by using a paired t test comparing the main
outcome in period one (baseline to week 4) with main
outcome in period two (week 6 to week 10) from the
18 participants completed 10 weeks
same patient. As well as the Kenward and Roger
approach, we also did a sensitivity analysis using a
Figure 2: Trial profile standard paired t test to compare the change in primary
outcome between different periods.
We analysed continuous secondary endpoints with
Participants (n=21)
baseline values (eg, daytime/night-time ambulatory
Age (years) 58 (11)
systolic/diastolic blood pressure) similarly to the primary
Sex endpoint. We analysed other continuous variables without
Male 10 (48%) a baseline value in each period with a paired t test. We have
Female 11 (52%) reported counts and percentages of all adverse events.
University education 9 (43%) We tested for interaction of treatment effect with age
24-h systolic blood pressure/diastolic blood 140 (9)/87 (8) (≤60 vs >60 years), sex, and body-mass index (BMI ≤30 vs
pressure (mm Hg)
>30 kg/m²). We also did subgroup analyses for each
Office systolic blood pressure/diastolic blood 154 (14)/90 (11)
pressure (mm Hg)
variable. We did trial analyses using SAS version 9.4.
Time since diagnosis of hypertension (months) 4·2 (5·4)
This trial is registered with the Australian and
New Zealand Clinical Trials Registry, number
Diabetes 2 (10%)
ACTRN12614001057673.
Hyperlipidaemia 5 (24%)
Previous myocardial infarction 0
Role of the funding source
Coronary artery revascularisation 0
The funder had no role in study design, data collection,
Cerebrovascular disease 0
data analysis, data interpretation, or writing of the report.
Previous depression 4 (19%)
KV, KR, CKC, and AR had full access to all data in the
Current smoker 5 (46%)
study. CKC and AR had final responsibility for the
Data are mean (SD) or number of patients (%). decision to submit for publication.
Table 1: Baseline characteristics of trial participants
Results
Between November, 2014, and December, 2015,
Statistical analysis 55 patients were screened for the Quadpill trial, and
We planned a sample size of 50 patients to provide 21 individuals were judged eligible and randomly
90% power at an α of 0·05, to detect a difference in allocated to a treatment group (figure 2). Baseline
systolic blood pressure of 12 mm Hg between the characteristics of the study population are shown in
quadpill and placebo, assuming an SD of the within- table 1. One patient declined participation before study
patient difference of 12 mm Hg and taking into account drug initiation and two participants withdrew at the end
the possibility of a 10% loss to follow-up. The study ended of the first treatment period for social reasons (figure 2).
at 1 year at the end of the budget and staffing time Therefore, 18 patients had complete data for the primary
allocated and the original sample size was not reached. outcome.
Analyses were by intention to treat. All tests were two- The difference in mean 24-h systolic blood pressure
sided. All statistical analyses were unadjusted for between quadpill and placebo periods was 18·7 mm Hg

1038 www.thelancet.com Vol 389 March 11, 2017

 Articles

Quadpill treatment period Placebo treatment period Difference* p value

(95% CI)
Baseline (week 0 or End of treatment Baseline (week 0 End of treatment
week 6) (week 4 or week 10) or week 6) (week 4 or week 10)
24-h systolic blood pressure 138·4 (7·3) 119·6 (7·6) 137·1 (10·4) 138·2 (10·0) 18·7 (14·3–23·0) <0·0001
24-h diastolic blood pressure 86·7 (10·6) 73·3 (8·7) 85·1 (9·4) 87·6 (11·9) 14·2 (11·5–16·9) <0·0001
Daytime ambulatory systolic blood 141·7 (7·7) 121·4 (7·9) 140·3 (11·6) 143·7 (10·5) 22·3 (17·7–26·9) <0·0001
pressure
Daytime ambulatory diastolic blood 89·9 (11·0) 75·7 (9·2) 87·9 (9·5) 91·1 (12·5) 15·3 (12·6–18·1) <0·0001
pressure
Night-time ambulatory systolic 128·8 (13·4) 114·4 (9·0) 126·2 (9·2) 125·4 (13·4) 10·4 (2·6–18·3) 0·0128
blood pressure
Night-time ambulatory diastolic 77·7 (12·9) 66·8 (8·9) 77·8 (10·0) 79·4 (13·1) 12·5 (7·9–17·1) <0·0001
blood pressure
Office systolic blood pressure 149·9 (16·7) 122·1 (8·8) 145·8 (10·2) 144·6 (12·2) 22·4 (16·5–28·3) <0·0001
Office diastolic blood pressure 87·4 (10·0) 71·8 (8·9) 86·1 (11·3) 84·8 (12·1) 13·1 (8·9–17·3) <0·0001

Data are mean (SD), unless otherwise stated. *Difference in change between quadpill and placebo period.

Table 2: Effects of quadpill and placebo on blood pressure variables

Study drug allocated when Treatment period when Severity* Action taken Outcome Relation to
adverse event occurred adverse event occurred study drug*
Gastrointestinal illness Quadpill First Mild None Resolved Not related
Headache Quadpill First Mild None Resolved Not related
Dry nose Placebo Second Mild None Resolved Not related
Vertigo Neither Between first and second Mild None Resolved Not related
Dizziness Quadpill First Mild Temporarily discontinued Resolved Related
study drug
Increased urinary Quadpill First Mild None Resolved Possibly related
frequency†
Increased urinary Placebo Second Mild None Resolved Possibly related
frequency†
Respiratory-tract infection Quadpill Second Mild None Resolved Not related

*Judged by the study doctor (JT) and a clinical cardiologist, in consultation with the principal investigator if needed. †Reported by one male patient during the intervention
period and the same patient in the placebo period; we instructed him to consult a local doctor for urological assessment.

Table 3: Adverse events

(95% CI 14·3–23·0), and in 24-h diastolic blood pressure
was 14·2 mm Hg (11·5–16·9). Similarly, the difference in
office systolic and diastolic blood pressure was 22·4 mm Hg
(16·5–28·3) and 13·1 mm Hg (8·9–17·3), respectively
(table 2). Daytime ambulatory systolic blood pressure,
daytime ambulatory diastolic blood pressure, night-time
ambulatory systolic blood pressure, and night-time
ambulatory diastolic blood pressure were all significantly
lower with the quadpill (table 2). All participants (18/18
[100%]) achieved office systolic and diastolic blood pressure
less than 140/90 mm Hg while on the quadpill, compared
with six (33%) of 18 while on placebo (risk ratio [RR] 3·01,
95% CI 1·54–5·89; p=0·0013). Ambulatory blood pressure
less than 135/85 mm Hg was achieved by 15 (83%) of
18 participants while on the quadpill compared with
seven (39%) of 18 while on placebo (RR 2·14, 95% CI
1·25–3·65; p=0·0053).
Tests for both a carry-over effect (p=0·868) and a period
effect (p=0·308) were not significant. There were no
significant interactions by age, sex, or BMI. In one
sensi­tivity analysis, using a standard comparison (paired
t test), results were virtually identical, with a difference in
mean 24-h systolic blood pressure between the quadpill
and placebo of 18·7 mm Hg (95% CI 14·3–23·0;
appendix p 12). Similarly, in a second sensitivity analysis,
in which we only included patients who did not have
missing data (n=18), results were also virtually identical,
with the difference in mean 24-h systolic blood pressure
of 18·7 mm Hg (95% CI 14·2–23·2).
Treatment compliance was high, with a mean number
of capsules missed in the last week of 0·2 (SD 0·4) for
the quadpill and 0·3 (0·6) for placebo. 18 participants
who finished the study completed the end-of-study
acceptability questionnaire, with all reporting the study
medication was either very easy (n=13) or easy (n=5) to
swallow. Moreover, all 18 participants reported it was
either very likely (n=10) or likely (n=8) they would take
the quadpill if available for use.

www.thelancet.com Vol 389 March 11, 2017 1039

 Articles
Number of Trials Participants Reduction in diastolic blood pressure
drugs at (mm Hg) (95% CI)
quarter-dose
One 36 4721
Two 6 312
Three 0 0
Four 1 18
20 10
Favours quarter-dose
Figure 3: Efficacy of combination therapy with one, two, three, and four drugs at quarter-dose on blood pressure-lowering, compared with placebo
Data for one and two drugs at quarter-dose are from the systematic review. Data for four drugs at quarter-dose are from the trial reported here. No data were
available for three drugs at quarter-dose (n/a).
There were no serious adverse events and no patients
had a prespecified adverse event. One participant
reported dizziness while on the quadpill, causing
temporary discontinuation of treatment, and one
reported increased urinary frequency in quadpill and
placebo periods, which were judged related or possibly
related to study treatment (table 3).
Mean heart rate was lower with the quadpill treatment
than with placebo, with a difference between groups of
6·5 beats per minute (95% CI 2·3–10·6). There was a
difference between the quadpill and placebo with respect
to changes in creatinine (4·4 mmol/L, 95% CI 0·9–7·8;
p=0.02), urate (0·03 mmol/L, 0·01–0·04; p=0·003), and
glucose (0·2 mmol/L, 0·02–0·4; p=0·04), but no patient
had more than a 12% increase in any variable
(appendix p 13). Levels of alanine aminotransferase,
aspartate aminotransferase, sodium, potassium, total
cholesterol, or LDL-cholesterol did not differ between the
quadpill and placebo.
In the systematic review, 36 trials (4721 participants)
were identified that reported the efficacy of one blood
pressure-lowering drug at quarter-dose compared
with placebo (appendix pp 7–11). Pooling these data,
quarter-dose blood pressure-lowering drugs reduced
systolic blood pressure by 4·7 mm Hg (95% CI 3·9–5·4)
and diastolic blood pressure by 2·4 mm Hg (1·9–2·8;
figure 3). Adverse events were reported in 14 of these
trials (n=1838), with no increase noted in adverse events
for one drug at quarter-dose compared with placebo (RR
1·0, 95% CI 0·88–1·10). Furthermore, six trials (n=312)
were identified in which two drugs at quarter-dose were
compared with placebo. The pooled reduction in systolic
and diastolic blood pressure was 6·7 mm Hg (95% CI
4·8–8·6) and 4·4 mm Hg (3·3–5·5), respectively. No
increase in side-effects was noted with two drugs at
quarter-dose compared with placebo (RR 0·93, 95% CI
0·29–2·9). No trials of three or four drugs at quarter-dose
versus placebo were identified in the systematic review.
Discussion
Our trial findings show that a capsule containing four
blood pressure-lowering drugs each at quarter-dose
reduced 24-h ambulatory blood pressure by roughly
1040 www.thelancet.com Vol 389 March 11, 2017
Reduction in systolic blood pressure
(mm Hg) (95% CI)
2·4 (1·9–2·8) 4·7 (3·9–5·4)
4·4 (3·3–5·5) 6·7 (4·8–8·6)
n/a n/a
13·1 (8·9–17·3) 22·4 (16·5–28·3)
0 25·0 12·5 0
Favours placebo Favours quarter-dose Favours placebo
19/14 mm Hg and achieved an office blood pressure less
than 140/90 mm Hg in all participants. When large
effects are seen in small trials, it is especially important
to review in the context of past evidence; our systematic
review findings and those of previous similar trials14,15 are
consistent with our finding of a large benefit with
minimum side-effects.
As far as we are aware, our findings are the first
placebo-corrected results showing the full effects of
four drugs at quarter-dose. Together with our systematic
review finding that one or two drugs at quarter-dose
produces no increase in side-effects compared with
placebo—our clinical trial findings suggest considerable
potential advantages for a single capsule containing
multiple blood pressure-lowering drugs at ultra-low
dose. A major novel feature of the proposed quadpill
approach is the theoretical advantage of initiating therapy
with a highly effective and tolerable combination versus
starting with fewer agents and titrating up (which in
practice is rarely done well). Thus, initiation of
combination therapy at very low doses might, for a
worthwhile proportion of patients, achieve the blood
pressure targets seen in SPRINT16,17 without the multiple
titration steps. However, as yet, there is no direct head-to-
head evidence for the quadpill approach being better
than that used currently, and comparison of the quadpill
with existing stepped-care approaches is required in
randomised clinical trials.
The main limitations of our trial are the small sample
size, short follow-up, and minimum power to assess side-
effects. A major barrier to recruitment was identifying
untreated individuals with elevated blood pressure within
the settings in which we work. The trial did not aim to
assess the contributions of each component or the
comparison with other strategies. One further issue is the
definition of quarter-dose. The strengths of our study
include the crossover design and use of ambulatory blood
pressure monitoring, hence maximising statistical power,
and the use of randomisation and placebo control to
minimise bias.
We recorded significant increases in creatinine, urate,
and glucose in our study, although no patient had more
than a 12% increase in any measure. There were no

longer term follow-up data and any clinical implications
are uncertain. The small increases in urate are consistent
with previous dose-response analyses for effects of
hydrochlorthiazide18 and observations with respect to
glucose and atenolol.19 Lower systemic pressure can
reduce glomerular perfusion pressure and lead to longer
term renal benefits for people with raised intraglomerular
pressure and proteinuria.20–23 However, trials have also
shown an increase in adverse renal outcomes with
intensive blood pressure lowering.7,24,25 Long-term
randomised data are required to determine the clinical
implications of the creatinine differences observed in
this study.
There has been one previous trial of four-drug
quarter-dose blood pressure lowering, involving
110 untreated individuals with blood pressure greater
than 140/90 mm Hg.14 In that trial, a 26/15 mm Hg
reduction was recorded in blood pressure, from a
baseline of 160/96 mm Hg, with therapy comprising
amlodipine 1·25 mg, atenolol 12·5 mg, bendro­
flumethiazide 0·625 mg, and captopril 50 mg. That trial
was unable to estimate a placebo-corrected reduction in
blood pressure but did report significantly greater
reductions with the four-drug quarter-dose therapy than
those seen with each mono­ therapy at standard dose.
Compared with individual agents, the combination
showed a greater systolic blood pressure reduction than
amlodipine (8 mm Hg, 95% CI 1–14), atenolol (9 mm Hg,
2–16), bendroflumethiazide (11 mm Hg, 4–18), and
captopril (7 mm Hg, 1–14). In that trial, no side-effects
were reported in the quadpill group and the only
two withdrawals were in the atenolol group. The only
other trial to date of low-dose antihypertensive therapy
with more than two agents assessed three-drug half-dose
therapy versus placebo in a crossover trial and showed a
similarly large blood pressure reduction of 18/10 mm Hg
(p<0·001).15 The placebo-corrected reduction with one-
drug and two-drug standard-dose therapy at similar
blood pressure levels is around 9/5 mm Hg and
17/9 mm Hg, respectively.26
Poor blood pressure control is a global problem.4,27
Initiating treatment with two-drug combination therapy
has been advocated28 as a more effective means to achieve
blood pressure control rapidly and with fewer clinic
visits.29 Our study draws on the same underlying
principles but extends the idea further to initiating
treatment with multiple ultra-low-dose agents in a single
capsule.30 By comparison with existing approaches to
blood pressure-lowering therapy, administration of a
single quadruple combination capsule is likely to achieve
more blood pressure-lowering than up-titrating
monotherapy, since doubling the dose for blood pressure
drugs from half-dose or from standard dose provides
only about 1–2 mm Hg further reduction in systolic or
diastolic blood pressure.8 Moreover, a quadpill approach
could address treatment inertia related to the clinician
and patient because it reduces the reliance on stepped
www.thelancet.com Vol 389 March 11, 2017 1041
Articles
titration, which is rarely completed in practice. A quadpill
also addresses the individual variation in responsiveness
to different agents through provision of a combination
with a range of modes of action. Improved adherence is
also likely as a result of both decreased pill burden31 and
use of lower doses to minimise side-effects.8
In summary, our study is the first placebo-controlled
trial to indicate that quarter-dose four-drug combination
therapy could be efficacious in lowering blood pressure.
It presents a novel approach that could achieve
substantially greater blood pressure control with a single
pill, which could have widespread clinical applicability.
Further trials are required to assess contributions of
different components, and the long-term efficacy and
safety in a broader population, both for initial treatment
and among patients with inadequate control or side-
effects while receiving monotherapy.
Contributors
CKC is the chief investigator of the clinical trial, led the writing of the
protocol and successful funding application, supervised JT, and drafted
the report. JT is a PhD student who primarily implemented the trial
protocol. AB, MB, and TU supported trial recruitment. KV ran all
statistical analysis, supervised by KR, who was the primary writer of the
statistical analysis plan. CKC, AR, and GH contributed to trial design.
AR and CKC had the idea for the trial. AB drafted the protocol and data
collection forms for the systematic review, did the search, data
abstraction, and data checking as first reviewer, led the statistical
analysis, and drafted the systematic review report. CKC contributed to
the idea for the systematic review, revision of the protocol, and review of
data analyses. MC contributed to the literature search, trial
identification, data abstraction, and data checking as second reviewer,
and reviewed data analyses. H-MD contributed to data checking as
second reviewer, and review of data analyses. EA assisted with data
checking and analysis. AR had the idea for the systematic review and
supervised research staff working on the project. RW, AS, AP, BN, DP,
HK, JT, JC, MN, CMR, GH, MW, SH, and ST contributed to review of
the protocol and data analyses. All authors contributed to critical review
of this report.
Declaration of interests
CKC is supported by a National Health and Medical Research Council
(NHMRC) Career Development Fellowship, co-funded by a National
Heart Foundation Future Leader Fellowship and the Sydney Medical
Foundation. BN reports grants for a clinical trial from Abbvie,
Dr Reddy’s Laboratories, Jannsen, Merck Schering-Plough, and Roche;
speaking fees from Abbott, Novartis, Pfizer, Roche, and Servier; travel
fees from Janssen, Roche, and Servier; fees for advisory board
membership from Janssen; is Chair of the Steering Committee for
two ongoing large-scale trials of an SGLT2 inhibitor and member of
the Steering Committee for a third; and is supported by an NHMRC
Principal Research Fellowship; all outside the submitted work. All
honoraria and travel fees are paid to BN’s institution, not as personal
fees. JC reports research grants and honoraria from Servier for the
ADVANCE trial, outside the submitted work. AP was supported by a
Senior Research Fellowship and Program Grant from NHMRC, during
the conduct of the study. DP reports grants from NHMRC, the
National Heart Foundation of Australia, and University of Sydney,
during the conduct of the study. ST reports personal fees from Amgen,
Lilly, Pfizer, and Sanofi, outside the context of the submitted work; and
acknowledges support by the UK National Institute of Health Research
(NIHR) Biomedical Research Centre at Imperial College Healthcare
NHS Trust and Imperial College London. MW reports consultant fees
from Amgen, outside the submitted work. George Health Enterprises,
the social enterprise arm of The George Institute for Global Health,
has applied for patents in this research area, on which CKC and AR
are named as inventors; George Health Enterprises has also received
investment to develop fixed-dose combinations containing aspirin,
 Articles
statins, and blood pressure-lowering drugs. AB, JT, GH, MB, TU, KV,
KR, EA, RW, MC, H-MD, AS, HK, MN, CMR, and SH declare no
competing interests.
Acknowledgments
The Quadpill study was supported by a Vanguard Grant and
Ross Hohnen prize from the National Heart Foundation of Australia
(grant number 100227), University of Sydney Bridging Grant, and
National Health and Medical Research Council of Australia (NHMRC)
programme grant. We thank the general practitioners who referred
patients to the trial (Nicholas Bennett, Yvette Castellano, and
Christopher Davis); all participants; Peter Rushton, Yvonne Stanford,
and other staff at Kildare Road Medical Centre; and project staff,
Elizabeth Knight, Helen Monaghan, Laurent Billot (chair of Data
Monitoring and Safety Committee), Craig Rogers (Pharmaceutical
Packaging Professionals), and Nick Karrasch (Trialfacts).
References
1 Lim SS, Vos T, Flaxman AD, et al. A comparative risk assessment of
burden of disease and injury attributable to 67 risk factors and risk
factor clusters in 21 regions, 1990–2010: a systematic analysis for
the Global Burden of Disease Study 2010. Lancet 2012; 380: 2224–60.
2 Turnbull F, for the Blood Pressure Lowering Treatment Trialists’
Collaboration. Effects of different blood-pressure-lowering
regimens on major cardiovascular events: results of
prospectively-designed overviews of randomised trials. Lancet 2003;
362: 1527–35.
3 Staessen JA, Wang JG, Thijs L. Cardiovascular protection and blood
pressure reduction: a meta-analysis. Lancet 2001; 358: 1305–15.
4 Chow CK, Teo KK, Rangarajan S, et al. Prevalence, awareness,
treatment, and control of hypertension in rural and urban
communities in high-, middle-, and low-income countries. JAMA
2013; 310: 959–68.
5 Wald DS, Law M, Morris JK, Bestwick JP, Wald NJ.
Combination therapy versus monotherapy in reducing blood
pressure: meta-analysis on 11,000 participants from 42 trials.
Am J Med 2009; 122: 290–300.
6 Xie X, Atkins E, Lv J, et al. Effects of intensive blood pressure
lowering on cardiovascular and renal outcomes: updated systematic
review and meta-analysis. Lancet 2016; 387: 435–43.
7 Wright JT Jr, Williamson JD, Whelton PK, et al. A randomized trial
of intensive versus standard blood-pressure control. N Engl J Med
2015; 373: 2103–16.
8 Law MR, Wald NJ, Morris JK, Jordan RE. Value of low dose
combination treatment with blood pressure lowering drugs:
analysis of 354 randomised trials. BMJ 2003; 326: 1427.
9 Brayfield A. Martindale: the complete drug reference.
http://www.pharmpress.com/product/MC_MART/
martindale-the-complete-drug-reference (accessed Jan 18, 2017).
10 MIMS. Monthly index of medical specialties. http://www.mims.
com/resources/portal/common/document/mims/mimsau.htm
(accessed Nov 17, 2016).
11 National Heart Foundation of Australia. Guide to management of
hypertension 2008: assessing and managing raised blood pressure
in adults—updated December, 2010. https://heartfoundation.org.
au/images/uploads/publications/
HypertensionGuidelines2008to2010Update.pdf (accessed
Jan 18, 2017).
12 Kenward MG, Roger JH. The use of baseline covariates in crossover
studies. Biostatistics 2010; 11: 1–17.
13 Kenward MG, Roger JH. An improved approximation to the
precision of fixed effects from restricted maximum likelihood.
Comput Stat Data Anal 2009; 53: 2583–95.
1042 www.thelancet.com Vol 389 March 11, 2017
14 Mahmud A, Feely J. Low-dose quadruple antihypertensive
combination: more efficacious than individual agents: a preliminary
report. Hypertension 2007; 49: 272–75.
15 Wald DS, Morris JK, Wald NJ. Randomized Polypill crossover trial
in people aged 50 and over. PLoS One 2012; 7: e41297.
16 Wright JT Jr, Whelton PK, Reboussin DM. A randomized trial of
intensive versus standard blood-pressure control. N Engl J Med
2016; 374: 2294.
17 Drawz PE, Pajewski NM, Bates JT, et al, for the SPRINT Study
Research Group. Effect of intensive versus standard clinic-based
hypertension management on ambulatory blood pressure: results
from the SPRINT (Systolic Blood Pressure Intervention Trial)
ambulatory blood pressure study. Hypertension 2017; 69: 42–50.
18 Peterzan MA, Hardy R, Chaturvedi N, Hughes AD. Meta-analysis of
dose-response relationships for hydrochlorothiazide,
chlorthalidone, and bendroflumethiazide on blood pressure, serum
potassium, and urate. Hypertension 2012; 59: 1104–09.
19 Chang SW, Gong Y, McDonough CW, et al. Melatonin pathway and
atenolol-related glucose dysregulation: is there a correlation?
Clin Transl Sci 2016; 9: 114–22.
20 Wright JT Jr, Bakris G, Greene T, et al. Effect of blood pressure
lowering and antihypertensive drug class on progression of
hypertensive kidney disease: results from the AASK trial. JAMA
2002; 288: 2421–31.
21 Hunsicker LG, Adler S, Caggiula A, et al. Predictors of the
progression of renal disease in the Modification of Diet in Renal
Disease Study. Kidney Int 1997; 51: 1908–19.
22 Lewis EJ, Hunsicker LG, Clarke WR, et al. Renoprotective effect of
the angiotensin-receptor antagonist irbesartan in patients with
nephropathy due to type 2 diabetes. N Engl J Med 2001; 345: 851–60.
23 Keane WF, Brenner BM, de Zeeuw D, et al. The risk of developing
end-stage renal disease in patients with type 2 diabetes and
nephropathy: the RENAAL study. Kidney Int 2003; 63: 1499–507.
24 Peralta CA, McClure LA, Scherzer R, et al. Effect of intensive versus
usual blood pressure control on kidney function among individuals
with prior lacunar stroke: a post hoc analysis of the Secondary
Prevention of Small Subcortical Strokes (SPS3) randomized trial.
Circulation 2016; 133: 584–91.
25 Cushman WC, Evans GW, Byington RP, et al. Effects of intensive
blood-pressure control in type 2 diabetes mellitus. N Engl J Med
2010; 362: 1575–85.
26 Law MR, Morris JK, Wald NJ. Use of blood pressure lowering drugs
in the prevention of cardiovascular disease: meta-analysis of
147 randomised trials in the context of expectations from
prospective epidemiological studies. BMJ 2009; 338: b1665.
27 Sever PS, Messerli FH. Hypertension management 2011: optimal
combination therapy. Eur Heart J 2011; 32: 2499–506.
28 Byrd JB, Zeng C, Tavel HM, et al. Combination therapy as initial
treatment for newly diagnosed hypertension. Am Heart J 2011;
162: 340–46.
29 Feldman RD, Zou GY, Vandervoort MK, Wong CJ, Nelson SA,
Feagan BG. A simplified approach to the treatment of uncomplicated
hypertension: a cluster randomized, controlled trial. Hypertension
2009; 53: 646–53.
30 Salam A, Webster R, Singh K, et al. TRIple pill vs Usual care
Management for Patients with mild-to-moderate Hypertension
(TRIUMPH): study protocol. Am Heart J 2014; 167: 127–32.
31 Webster R, Patel A, Selak V, et al. Effectiveness of fixed dose
combination medication (“polypills”) compared with usual care in
patients with cardiovascular disease or at high risk: a prospective,
individual patient data meta-analysis of 3140 patients in
six countries. Int J Cardiol 2016; 205: 147–56.