Content uploaded by Leo Akioyamen
Author content
All content in this area was uploaded by Leo Akioyamen on Mar 11, 2018
Content may be subject to copyright.
Comparative Efficacy of Angiotensin II
Antagonists in Essential Hypertension:
Systematic Review and Network
Meta-analysis of Randomised
Controlled Trials
Bernice Tsoi, MSc, PhD
a,b
, Leo E. Akioyamen, BScN, MD
c[12_TD$DIFF]*
,
Ashley Bonner, MSc, PhD
a[13_TD$DIFF],d
, Claudia Frankfurter, BHSc, MD
c[14_TD$DIFF]
,
Mitchell Levine, MD, MSc
a,b
, Eleanor Pullenayegum, MSc, PhD
a,e
,
Ron Goeree, MSc
a,b
, Daria O’Reilly, PhD
a,b
a
Department of Clinical Epidemiology & Biostatistics, McMaster University, Hamilton ON, [15_TD$DIFF]Canada
b
Programs for Assessment of Technology in Health (PATH) Research Institute, St. Joseph’s Healthcare, Hamilton ON, [17_TD$DIFF]Canada
c
Faculty of Medicine, University of Toronto, Toronto ON, [19_TD$DIFF]Canada
d
Statistics for Integrative Genomics and Meta-Analysis (SIGMA) Lab, McMaster University, Hamilton, ON, [15_TD$DIFF]Canada
e
Child Health Evaluative Sciences Program, The Hospital for Sick Children (SickKids), Toronto ON, [21_TD$DIFF]Canada
Received 21 January 2017; received in revised form 24 May 2017; accepted 23 June 2017; online published-ahead-of-print xxx
©2017 Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) and the Cardiac Society of Australia and New Zealand (CSANZ).
Published by Elsevier B.V. All rights reserved.
*Corresponding author at: Faculty of Medicine, University of Toronto | Mississauga Academy of Medicine, 3359 Mississauga Road – Terrence Donnelly Health
Sciences Complex, Mississauga, ON, L5L 1C[23_TD$DIFF]6, Canada, Email: sele.akioyamen@mail.utoronto.ca
Background Evidence on the long-term clinical benefits of individual members of angiotensin II receptor blockers is
limited given the lack of head-to-head studies. We conducted a network meta-analysis to determine the
comparative efficacy of different members within this drug class with respect to outcomes of (i) blood
pressure reduction (at 24 and 52 weeks) and (ii) prevention of cardiovascular disease [24_TD$DIFF](>104 weeks).
Methods A systematic literature review was conducted – Protocol registration: (PROSPERO – CRD42014007067) – to
identify relevant literature from the following databases: Cochrane Library, PubMed, MEDLINE and
Embase; searched from inception to July 2016. Randomised controlled trials (RCTs) were included if they
reported long-term effectiveness relating to blood pressure, mortality, myocardial infarction or stroke.
Eligible studies included those with placebo or specific active-treatment comparators (either another angio-
tensin II receptor blockers or hydrochlorothiazide). A Bayesian random-effects network model was used to
combine direct within-trial comparisons between treatment groups with indirect evidence from other trials.
Results Thirty-six studies were identified, representing 28 unique trials. Blood pressure reduction, based on [25_TD$DIFF]12
studies (n = 807) with fixed dosing regimen, was found to be similar amongst members of the angiotensin
receptor blocker drug class at both 24 and 52 weeks. A network meta-analysis of five studies (n = 16,716)
with a treat-to-target approach found that prevention of all-cause mortality, stroke and myocardial infarc-
tion was similar across the angiotensin-receptor blockers therapies initiated.
Heart, Lung and Circulation (2017) xx, 1–17
1443-9506/04/$36.00
http://dx.doi.org/10.1016/j.hlc.2017.06.721
ORIGINAL ARTICLE
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
Introduction
Hypertension is a prevalent medical condition that is a major
modifiable risk-factor for cardiovascular disease. Although
most clinical guidelines recommend an optimal blood pres-
sure of <140/90 mmHg in low-to-moderate risk individuals
and <150/90 mmHg in the elderly [1–3], meeting these tar-
gets can be challenging. Previous studies have shown that, in
addition to lowering blood pressure, antihypertensive ther-
apies confer protection against the development of cardio-
vascular disease [4,5]. However, these studies are only
partially relevant clinically, given that care providers often
do not simply face the issue of whether therapy should be
initiated but rather which specific therapy should be pre-
scribed. Existing systematic reviews and meta-analyses that
compare across antihypertensive classes [6–10] can partly
address this issue by providing clarity on the appropriate-
ness of certain classes in different patient populations and
under different indications. However, several agents may
exist within a single drug class and an important question
on selection still remains: can they all be considered
equivalent?
Most, if not all, national guidelines for the management
of hypertension recommend angiotensin II receptor block-
ers(ARBs)asfirst-linetherapy[2,3,11,12].ARBsinhibit
the actions of angiotensin II through selective binding
of type 1 (AT
1
) receptors in vascular smooth muscle
[13]. The results of treatment are effective reductions in
blood pressure [14] and known ancillary properties inde-
pendent of blood-pressure lowering, such as slowing the
progression of renal disease [15–18].Yet,debateremains
unresolved regarding differences in the efficacy of agents
within this drug class. Structural and chemical differences
have been identified and some clinical studies have sug-
gested that not all ARBs are equal, with newer agents
having superior and more rapid blood pressure control
when compared to losartan or valsartan at four and eight
weeks [19,20].
Clinically meaningful differences within this drug class
would have implications for optimising therapeutic decision-
making. Yet most studies are typically two-armed placebo-
controlled trials that do not examine comparative effective-
ness. Furthermore, meta-analyses attempting to address
comparative effectiveness to date have been limited, as their
analyses are constrained to direct evidence pooled from
short-term (12 weeks) trials addressing only blood pressure
outcomes [21–23]. In order to appropriately assess the ques-
tion of intra-class superiority, the ideal approach would
entail multiple treatment (network) meta-analyses in which
both direct and indirect evidence are combined to generate
an estimate on the comparative effectiveness of individual
ARBs.
The purpose of this study is to provide a comprehensive
and updated systematic review on the comparative effec-
tiveness of ARBs in reducing blood pressure and cardiovas-
cular event rates (i.e., myocardial infarction (MI), stroke,
cardiovascular- and all-cause mortality) in patients with
hypertension using network meta-analysis (NMA). We
examine long-term effectiveness by comparing blood pres-
sure outcomes at 24 and 52 weeks and cardiovascular event
rates at >104 weeks.
Methods
Literature Search Strategy
We followed a pre-specifiedstudy protocol (CRD42014007067)
and [26_TD$DIFF]the Preferred Reporting Items for Systematic Reviews
and Meta-Analyses (PRISMA) consensus statement through-
out the design, implementation, and reporting of this study
[24].
We searched OVID Medline (1966 to present; In-Process
and Other Non-Indexed Citations), Embase (1974 to July
2016), PubMed (for non-Medline records) and The Cochrane
Library. A search filter was applied to restrict search results
to randomised controlled trials, meta-analyses and system-
atic reviews. The search strategy used controlled vocabulary,
including MeSH terms and keywords, related to ‘‘hyper-
tension”; ‘‘angiotensin receptor blockers” and the generic
names of pharmacological agents (Appendix I). Addition-
ally; the references of all retrieved articles and any relevant
systematic reviews were hand-searched. We retrieved only
English-language studies. An unrestricted timeframe was
chosen to capture all relevant publications with the latest
search performed as of 9 July 2016. We imported citations
into [27_TD$DIFF]reference management software for de-duplication and
title/abstract screening.
Selection Criteria
Titles and abstracts were independently screened in dupli-
cate to assess eligibility according to pre-defined inclusion/
exclusion criteria. Potentially relevant studies then under-
went full-text screening. To be included in this analysis,
eligible trials had to fulfill the following criteria (Appendix
II): randomised-control trial design; enrolling adult patients
(18 years of age) with essential hypertension (no further
restriction was imposed for age, gender or other co-morbid-
ities); comparing ARBs to either another ARB, hydrochloro-
thiazide or a placebo regimen (with baseline concurrent
drugs unrelated to hypertension permitted).
Conclusions Current evidence is insufficient to show differences in any members within the angiotensin II receptor
blocker drug classwith respect to bloodpressuring lowering effects or a reduction in cardiovascular diseases.
Keywords Angiotensin II type 1 receptor blockers Systematic review Meta-analysis Blood pressure
Myocardial infarction Stroke
2B. Tsoi et al.
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
[28_TD$DIFF]Outcomes of interest included either blood pressure,
following at least six months of treatment, or adverse cardio-
vascular events (i.e. total and cardiovascular-related mortal-
ity, MI, stroke), following at least two years of treatment.
Given the restriction for trials with longer durations, studies
involving concomitant antihypertensive medication (e.g.,
usual antihypertensive treatment), step-care or combination
therapy were included. Any disagreements were resolved by
consensus between the independent screeners.
Study Quality Assessment
Two authors independently assessed risk of bias using the
relevant components recommended by the Cochrane
Collaboration.
[29_TD$DIFF]Data Extraction Strategy
Data from each included trial were extracted by one reviewer
using a structured form that was subsequently checked by a
second reviewer. Disagreements were resolved by discus-
sion. In this form, details on the trial design (e.g. timeframe,
inclusion/exclusion criteria), trial population (e.g. age, gen-
der proportion, baseline risk factors) and trial results (e.g.
blood pressure measurements, absolute value and relative
risk for cardiovascular and cerebrovascular events) were
collected. [30_TD$DIFF]Study authors were contacted to answer queries
and provide additional information.
If a study resulted in multiple publications, data were
extracted from both the primary and secondary papers.
However, the analysis included only the main paper (defined
as the one with the largest sample size) unless the secondary
paper reported a different follow-up period or a separate
outcome of interest.
Statistical Analysis
Given the nature of the included studies, NMA was possible.
This type of analysis, which combines both direct and indi-
rect treatment comparisons, can summarise RCTs of several
different treatment strategies and provide point estimates
(and 95% confidence interval (CI)) of their association for a
given endpoint. The NMA was conducted using a Bayesian
random-effects generalised linear model [28] with a consis-
tency assumption for the treatment effects. The generalised
linear model framework allowed us to handle continuous
and binary outcome variables. We used R software version
3.1.1 [29] with R package gemtc version 0.6 [30] to specify the
model and interface with Just Another Gibbs Sampler (JAGS)
software version 3.4.0 [31] to execute Bayesian estimation of
the model parameters through a Markov chain Monte Carlo
(MCMC) process. The default of vague prior distributions
for treatment effect and heterogeneity parameters was cho-
sen. As per convention [31], we set an adaption phase of
20,000 samples, a burn-in phase of 100,000 samples, and a
thinning interval of 10, resulting in 10,000 samples being
used for inference in the MCMC process. To ensure conver-
gence for all model parameters, four chains were run and
assessed by the Gelman-Rubin-Brooks plot and diagnostic
test [32].
Results
5184 unique publications were identified, of which 5102 were
excluded following title/abstract screening. Eighty-two docu-
ments underwent full-text screening of which 36 met the full
inclusion criteria (Figure 1), representing 28 unique trials.
Three clinical trials that were initially selected could not be
subsequently included in the narrative synthesis as specific
details regarding blood pressure were not available [19,33,34].
All remaining included studies reported on blood pressure
reduction and, amongst these, five further reported on out-
comes of cardiovascular morbidity and mortality (Table 1).
Table 1 summarises the general characteristics and base-
line demographics of the studies identified in our systematic
review. The majority of the studies were two-armed trials,
with the exception of five [31_TD$DIFF]three-arm trials. As the study
objective was to assess long-term efficacy, mean study dura-
tion ranged from 24 to 243 weeks.
Heterogeneity between studies was observed in terms of
the patient population (i.e., age, gender, co-morbidities) and
the dosing regimen (i.e. fixed dosing or treat-to-target). A
variety of patient populations were enrolled including: dia-
betes mellitus without [35–41] or with additional comorbid-
ities [15–17,42–44]; metabolic syndrome [45–47]; and left-
ventricular hypertrophy [48,49]. Certain studies focussed
solely on the elderly [50–55]; while others centred on over-
weight and/or obese patients [42,43,49,56] (Table 1). Some of
the identified studies evaluated the efficacy of ARBs in low-
ering blood pressure by either adhering to the same dosage
strength throughout the study’s duration [35,37,41–
43,46,49,53,56–61] or by employing a forced dose-doubling
regimen applied to all patients regardless of their treatment
response [27]. However, a greater majority of the trials
attempted to attain a particular blood pressure target using
treat-to-target methodology. In the latter case, all patients
received an ARB and the dosage of their respective ARB was
doubled and/or additional antihypertensives were added in
order to achieve a pre-defined blood pressure goal [15–
17,36,38–40,44,45,47,48,52,54,57,62–65]. Blood pressure goals
varied considerably across these studies: ranging from a
systolic blood pressure between <130 to <160 mmHg and
a diastolic blood pressure between <80 to <95 mmHg.
Although the treat-to-target studies are described narra-
tively in this review, they were excluded from NMA for the
outcome of blood pressure reduction specifically. As a large
portion of patients in the placebo group received active
treatment, this made it difficult to interpret the treatment
effect attributable to ARB alone. In such studies, the differ-
ence in blood pressure reduction achieved would be smaller
partly because background treatment with other antihyper-
tensive drugs was permitted.
The quantitative NMA therefore addresses two separate
questions according to the outcome: (i) in studies involving
a fixed dosing regimen, what is the comparative effective-
ness of individual ARBs in terms of controlling blood pres-
sure following at least 24 weeks of treatment?; and (ii) in
treat-to-target trials, what is the long-term comparative
Efficacy of Angiotensin II Antagonists in Essential Hypertension 3
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
effectiveness of ARB-initiated therapy alongside conven-
tional antihypertensive treatment in preventing cardiovas-
cular diseases following at least two years of treatment?
Risk of Bias Assessment
Table 2 presents the risk of bias assessment for the included
studies. Nearly all were associated with an uncertain risk of
selection bias given poor reporting on the methods of ran-
domisation and the approach to preserve allocation conceal-
ment. A little over half of the studies reported blinding
although, in the majority of the cases, the method by which
blinding was ensured was not adequately described. The
least-likely risk of bias was observed in terms of selective
outcome reporting as most studies reported the results of the
outcomes that were set out either in their methods or in
published study protocols. Other sources of bias found in
some studies included imbalance in the prognostic factors
despite randomisation.
Efficacy: Blood Pressure Change
As previously mentioned, treat-to-target trials were excluded
from the analysis of blood pressure given that such study
designs do not aim to compare absolute blood pressure
efficacy but, rather, the secondary effects of treatment, such
as collateral benefits (e.g., reduction in cardiovascular mor-
bidity and mortality), at the same or similar levels of blood
pressure control. Consequently, treat-to-target trials have
limited utility when evaluating blood pressure since a spe-
cific blood pressure target would have been pre-defined for
all treatment arms within a trial.
Figure 1 PRISMA Diagram of the Clinical Review Search Strategy.
4B. Tsoi et al.
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
Table 1 Characteristics and baseline data of included studies (secondary articles from main trial also presented).
First Author Study name Mean
follow-up,
wk
Number of
participants
[5_TD$DIFF]analysed
Intervention, dose Titration to reach BP goal Demographic Details Baseline Blood Pressure
(mmHg)
Dose
doubling?
Concomitant
medication?
Mean
age (SD)
Female
(%)
BMI
(SD)
Concomitant
diseases
SBP (SD) DBP (SD)
Brenner [14] RENAAL 177 751 Losartan, 50 mg od Yes Yes 60 (7) 38.5 30 (6) Diabetes +
nephropathy
152 (19) 82 (10)
762 Placebo 60 (7) 35.2 29 (6) 153 (20) 82 (11)
DeRosa [45,46] 52 95 Telmisartan, 40 mg od NR NR 56 48 27.6 (1.1) Diabetes + MS 135 (4) 86 (4)
5
152 Irbesartan, 150 mg od 55 52 27.7 136 (4) 84 (3)
41.2
DeRosa [38] 52 40 Telmisartan, 40 mg od NR NR 54 45 26.9 (1.2) Diabetes 143 (5) 92 (3)
5
39 Eprosartan, 600 mg od 55 49 26.4 (1.3) 144 (5) 91 (4)
3
40 Placebo 53 50 26.2 143 (4) 92 (4)
51.1
Foulquier [69] TRANSCEND 243 2547 Telmisartan, 80 mg od NR Yes NR NR NR 143.4 NR
2551 Placebo NR Yes NR NR NR 143.5 NR
Galzerano [62] 52 40 Telmisartan, 80 mg od No No 55 43.9 NR 157 (8) 96 (6)
8
25 HCTZ, 25 mg od 53 46.4 NR 154 (10) 95 (7)
7
Hasegawa [66] TALENT 52 29 Telmisartan, 40 mg od Yes Yes 59.1 (10.3) 31 25.7 (4.9) 152.1 (16.5) 90.0 (13.3)
28 Losartan, 50 mg od 56.4 (10.1) 32.1 25.9 (4.8) 150.6 (10.6) 92.1 (12.3)
Lewis [15] IDNT-2 104 579 Irbesartan, 75 mg od Yes Yes 59.3 (7.1) 35 31 (5.6) Diabetes +
proteinuria
160 (20) 87 (11)
569 Placebo 58.3 (8.3) 29 30.5 (5.9) 158 (20) 87 (11)
Lindholm [70] ALPINE 52 196 Candesartan, 16 mg od No Yes 54.5 (9.4) 52 27.8 (4.1) 154.7 (13.2) 96.8 (5.6)
196 HCTZ, 25 mg od 55.4 (9.6) 53 28.1 (4.2) 155 (13.5) 97 (5.7)
Lithell [57] SCOPE 194 2477 Candesartan, 8–16 mg od Yes Yes 76.4 64.8 27 166 (8.9) 90.3 (6.6)
2460 Placebo 76.4 64.2 26.9 166.5 (9) 90.4 (6.6)
Papademetriou
[54]
SCOPE (substudy) 167 754 Candesartan, 8 mg od Yes Yes 77.3 63.3 26.7 168.7 82.3
764 Placebo 76.9 65.3 26.3 169.3 82.5
Saxby [53] SCOPE (substudy: 191 112 Candesartan, 8 mg od Yes Yes 76 (4) 48.2 NR 165 (8) 88 (7)
Single [6_TD$DIFF]centre) 116 Placebo 76 (5) 46.5 NR 166 (8) 89 (7)
Trenkwalder
[56]
SCOPE (substudy: 191 1253 Candesartan, 8 mg od No No 76.4 67 NR NR NR
No add-on therapy) 845 Placebo NR NR NR
Trenkwalder
[58]
SCOPE (substudy: 167 NR Candesartan, 8 mg od NR NR NR NR NR NR NR
Pre-specified subgroup) NR Placebo NR NR NR NR NR
Efficacy of Angiotensin II Antagonists in Essential Hypertension 5
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
Table 1. (continued).
First Author Study name Mean
follow-up,
wk
Number of
participants
[5_TD$DIFF]analysed
Intervention, dose Titration to reach BP goal Demographic Details Baseline Blood Pressure
(mmHg)
Dose
doubling?
Concomitant
medication?
Mean
age (SD)
Female
(%)
BMI
(SD)
Concomitant
diseases
SBP (SD) DBP (SD)
Makris [61] 26 45 Eprosartan, 600 mg od No No 55 (10) 55.6 23.96 148.77 (9) 95.31
41 Losartan, 100 mg od 55 46.3 23.97 148.56 (9.1) 95.24
9
Menne [39] ROADMAP
(substudy)
167 2043 Olmesartan, 40 mg od No Yes 58 (8.7) 52.9 31.3 (4.9) Diabetes 138.4 (15) 81.7 (9.5)
1977 Placebo 58.2 (8.5) 55.4 31.1 (4.9) 137.7 (14.4) 81.5 (9.1)
Minami [50] 24 20 Telmisartan, 40 mg od NR No 63.1 (11.6) 61.9 24.4 (3) MS NR NR
20 Losartan, 50 mg od 63.1 61.9 24.4 (3) NR NR
11.6
Murakami [48] ADIPO 24 9 Telmisartan, 40 mg od No Yes 61.4 (4.58) 66.7 27.2 (1.6) MS 146.3 (3.6) 85 (3.8)
10 Valsartan, 80 mg od 50.4 (4.75) 40 30.8 (2.4) 144.9 (4.8) 89.5 (3.6)
Nedogoda [52] 24 30 Telmisartan, 80 mg od NR NR 47.4 (9.2) 50 31.1 (3.1) Majority: LVH (90%);
hyperchloesterolanaemia
(80%)
158 (3) 98 (3)
30 Losartan, 100 mg od 46.7 (8.2) 50 29.4 (3.6) Majority: LVH (97%)
hyperchloesterolanaemia
(87%)
157 (4) 98 (3)
Negro [59] 24 21 Irbesartan, 150 mg od NR NR 45 (9.3) 30.4 35.6 (2.8) 149.7 (4.7) 94.4 (4.2)
22 Telmisartan, 80 mg od 46.7 (8.2) 34.8 36.4 (2.4) 151.7 (4.9) 97.7 (4.2)
Neldam [55] 24 123 Candesartan, 8 mg od Yes NR 78.5 65 NR 178.9 (15.9) 101.8 (4.8)
3.3
62 HCTZ, 12.5 mg od 78.1 (3.4) 58 NR 179.8 (16.5) 101 (4)
Parving [16] IRMA2 104 195 Irbesartan, 150 mg od Yes Yes 58.4 33.8 29.9 (3.8) Diabetes +
microalbuminuria
153 (14) 90 (9)
8
194 Irbesartan, 300 mg od 57.3 (7.9) 29.4 30 (4.3) 153 (14) 91 (10)
201 Placebo 58.3 (8.7) 31.3 30.3 (4.4) 153 (15) 90 (9)
Rossing [47] IRMA2
(substudy:
104 13 Irbesartan, 150 mg od No Yes 62 10 29 (2) Diabetic +
microalbuminuria
156 (15) 91 (11)
9
Single centre) 15 Irbesartan, 300 mg od 56 20 29 (4) 159 (16) 90 (9)
9
15 Placebo 60 30 28 (5) 157 (15) 89 (7)
7
Picca [51] 26 15 Losartan, 50 mg od Yes NR 48 43.3 NR Concentric LVH 164 (7) 105 (4)
8
15 Valsartan, 80 mg od NR 171 (8) 101 (6)
Rayner [67] 24 27 Losartan, 50 mg od Yes Yes NR 60 NR 151 (8.4) 89.2 (8.4)
25 Candesartan, 8 mg od NR 59 NR 157 (16.7) 90.5 (7.84)
Rizos [49] 24 52 Telmisartan, 80 mg od NR NR 60 (10) 48.1 29 (4) MS 153 (14) 91 (10)
6B. Tsoi et al.
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
Table 1. (continued).
First Author Study name Mean
follow-up,
wk
Number of
participants
[5_TD$DIFF]analysed
Intervention, dose Titration to reach BP goal Demographic Details Baseline Blood Pressure
(mmHg)
Dose
doubling?
Concomitant
medication?
Mean
age (SD)
Female
(%)
BMI
(SD)
Concomitant
diseases
SBP (SD) DBP (SD)
48 Irbesartan, 300 mg od 60 (10) 54.2 29 (5) 151 (11) 90 (9)
51 Olmesartan, 20 mg od 58 (12) 52.9 28 (4) 151 (11) 93 (8)
Sawaki [40] 52 14 Losartan, 25 mg od NR NR 54 (7) 57.1 24.3 (2.8) Diabetes 134 (14) 78 (14)
15 Placebo 54 (9) 46.7 24.5 131 (15) 82 (12)
Schram [42] 52 20 Candesartan, 8 mg od Yes Yes 60 (7) 45.8 28.8 (3.7) Diabetes 151 (14) 94 (10)
Spoelstra-de-
man [41]
19 HCTZ, 12.5 mg od 63 (6) 33.3 29.5 (3.5) 157 (13) 93 (9)
Sica [65] 24 323 Azilsartan, 20 mg od Yes No 57.8 (12.1) 49.8 30.8 (5.7) 158.1 (14.4) 91.2 (11)
311 Azilsartan, 20 mg od 56.8 (10.7) 48.6 30.7 (5.3) 156.3 (12.5) 91.5 (10.5)
322 Valsartan, 80 mg od 58.1 (10.9) 46.3 31.2 (5.8) 157 (14) 90.8 (11.3)
Solomon [68] VALIDD 38 166 Valsartan, 160 mg od Yes Yes 61.1 (9.4) 53 30.1 (5.4) 143.5 (16.7) 85.4 (10.5)
175 Placebo 60.2 (9.5) 49 30.7 (6.1) 144.1 (15.6) 87 (10.1)
Tedesco [63,64] 96–104 42 Losartan, 50 mg od NR NR 54 (9) 54.4 NR 157 (9) 96 (9)
27 HCTZ, 25 mg od 56 (7) 45.4 NR 158 (10) 97 (7)
Tsutamoto [60] 52 25 Olmesartan, 20 mg od No No 68.2 (12.3) 40 NR 134 (15) 77 (9.7)
25 Candesartan, uncertain 67.7 (7.8) 36 NR 130 (21) 73 (7.5)
Uzu [43] 24 14 Telmisartan, 80 mg od NR NR 57 (7) NR 26.9 (3.7) Diabetes 138 (11) 83 (7)
14 Valsartan, 160 mg od 60 (10) NR 26.6 (3.0) 134 (12) 80 (8)
Uzu [44] 24 16 Valsartan, 160 mg od NR NR 58 (10) 75 NR Diabetes NR NR
16 Losartan, 50 mg od + HCTZ 58 (10) 31.3 NR NR NR
Abbreviations: DBP, diastolic blood pressure; HCTZ, hydrochlorothiazide; LVH, left ventricular hypertrophy; MS, metabolic syndrome; SBP, systolic blood pressure.
ADIPO, Abdominal fat Depot Intervention Program of Okayama; ALPINE, Antihypertensive Treatment and Lipid Profile in a North of Sweden Efficacy Evaluation; IDNT-2, Irbesartan Diabetic Nephropathy Trial; IRMA2,
Irbesartan in Patients With Type 2 Diabetes and Microalbuminuria; RENAAL, Reduction of Endpoints in NIDDM with the Angiotensin II Antagonist Losartan Study; ROADMAP, Randomized Olmesartan and Diabetes
Microalbuminuria Prevention; SCOPE, Study on Cognition and Prognosis in the Elderly; TALENT, Telmisartan and Losartan Cardiac Evaluation Trial; TRANSCEND, Telmisartan Randomised AssessmeNt Study in ACE
iNtolerant;VALIDD, Valsartan in Diastolic Dysfunction.
Efficacy of Angiotensin II Antagonists in Essential Hypertension 7
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
Given this, only 12 unique trials involving a fixed dosing
regimen were suitable for inclusion in the NMA on blood
pressure (n = 807 patients) (Table 3). However, even amongst
trials on the same ARB, different dosage strength may have
been studied. Short-term pharmacological studies suggest a
near flat dose-response within this class of drugs [66], but, to
remain conservative, only the highest dose equivalent for
each ARB was selected for the analysis [40,46,49,58–60].As
such, only half the number of patients (n = 460) [32_TD$DIFF]were
included in the NMA in which 407 and 53 patients received
an ARB and a hydrochlorothiazide, respectively.
As treatment efficacy in terms of blood pressure control is
dependent on the duration of therapy, two time points were
explored: the efficacy at 24 weeks [41,46,49,58] and 52 weeks
[59,60].Figure 2 shows the network of treatment comparisons
according to the time period assessed. At 24 weeks, five
members of the ARB class were studied in which only 6 of
the 15 possible pair-wise comparisons were studied directly.
At 52 weeks, only two members of the ARB class could be
studied, sharing the common comparator of hydrochlorothi-
azide. In both analyses, there was no evidence of non-conver-
gence with the MCMC process. Pairwise between-study
heterogeneity could not be assessed given that each direct
pairwise comparison was informed by only a single study.
The results on blood pressure reduction from the pairwise
comparisons between different members of the ARB class are
presented in Figures 3 and 4 following 24 and 52 weeks of
treatment, respectively. At both time points, the 95% credible
Table 2 Study Quality according to Cochrane Risk of Bias Tool (only the main study is presented) (+: low risk of bias;?:
uncertain risk of bias; : high risk of bias).
First Author Random
sequence
generation
Allocation
concealment
Blinding of
participants and
study personnel
Blinding of
outcome
assessor
Incomplete
outcome
data
Selective
reporting
Other
sources
of bias
Brenner [14] ?? ? ? +
DeRosa [45,46] +? + + + ++
DeRosa [38] ?? ? ? ? ++
Foulquier [69] +? ? ? + ++
Galzerano[62] ?? ? ? ++
Hasegawa [66] ?? ++
Lewis [15] ?? ? ? ? +
Lindholm [70] ?? ? ? +++
Lithell [57] + + + (patient) ? + + +
? (study personnel)
Makris [61] ?? ++
Menne [39] ++ + + ? ++
Minami [50] +? ++
Murakami [48] ?? ++
Nedogoda [52] ? ? + (patient) +++
(study personnel)
Negro [59] ?? +++
Neldam [55] ?? ? ? ++
Parving [16] ? ? + (patient) ? +
Rossing [47] ? (study personnel)
Picca [51] ?? ? + +++
Rayner [67] ?? +++
Rizos [49] ?? ??++
Sawaki [40] ?? ++
Schram [42] ? ? + (patient) ? + + –
? (study personnel)
Sica [65] ?? ? ? ++
Solomon [68] +? ? + + +
Tedesco [63,64] ?? ? ? ++
Tsutamoto [60] ?? ?++
Uzu [43] ++ ??+
Uzu [44] ++ ??++
8B. Tsoi et al.
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
Table 3 Summary of study results (only main study presented).
First Author Study name Mean
follow-up,
wk
Number of
analysed
Intervention,
dose
Blood pressure Clinical outcomes (no of subjects)
Definition DSBP
(mmHg)
DDBP
(mmHg)
Mortality Major CV
event
MI Stroke
All-cause CV-related
Studies that report blood pressure outcomes only: titration and additional add-on therapy not permitted
DeRosa [45,46] 52 95 Telmisartan, 40 mg od Seated, trough 6 mths: 5 6 mths: 5
12 mths: 11 12 mths: 8
152 Irbesartan, 150 mg od 6 mths: 5 6 mths: 4
12 mths: 11 12 mths: 7
DeRosa [38] 52 40 Telmisartan, 40 mg od Seated 6 mths: 4 6 mths: 5
12 mths: 8 12 mths: 8
39 Eprosartan, 600 mg od 6 mths: 4 6 mths: 2
12 mths: 7 12 mths: 4
40 Placebo 6 mths: 1 6 mths: 1
12 mths: 2 12 mths: 2
Galzerano [62] 52 40 Telmisartan, 80 mg od Ambulatory 24 13
25 HCTZ, 25 mg od 10 8
Makris [61] 26 45 Eprosartan, 600 mg od Seated 10.9 12.9
41 Losartan, 100 mg od 18 13
Nedogoda [52] 24 30 Telmisartan, 80 mg od Ambulatory 12 7
30 Losartan, 100 mg od 15 12
Negro [59] 24 21 Irbesartan, 150 mg od Ambulatory 16.6 14.2
22 Telmisartan, 80 mg od 17.1 12.9
Rizos [49] 24 52 Telmisartan, 80 mg od Seated 17 10
48 Irbesartan, 300 mg od 17 8
51 Olmesartan, 20 mg od 17 10
Sawaki [40] 52 14 Losartan, 25 mg od Seated 3.4 4.6
15 Placebo 1.7 1.1
Sica [65] 24 323 Azilsartan, 20 mg od Ambulatory 14.9 NR
311 Azilsartan, 20 mg od 15.3 NR
322 Valsartan, 80 mg od 11.6 NR
Tedesco [63,64] 96–104 42 Losartan, 50 mg od Ambulatory 10 mths: 19 10 mths: 11
22 mths: 22 22 mths: 11
27 HCTZ, 25 mg od 10 mths: 7 10 mths: 5
22 mths: 11 22 mths: 7
Tsutamoto [60] 52 25 Olmesartan, 20 mg od Not reported 6 mths: 5 6 mths: 2
12 mths: 2.6 12 mths: 2
25 Candesartan, uncertain 6 mths: 3 6 mths: 2
12 mths: 2 12 mths: 2
Uzu [43] 24 14 Telmisartan, 80 mg od Ambulatory 52
14 Valsartan, 160 mg od 32
Efficacy of Angiotensin II Antagonists in Essential Hypertension 9
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
Table 3. (continued).
First Author Study name Mean
follow-up,
wk
Number of
analysed
Intervention,
dose
Blood pressure Clinical outcomes (no of subjects)
Definition DSBP
(mmHg)
DDBP
(mmHg)
Mortality Major CV
event
MI Stroke
All-cause CV-related
Studies that report blood pressure outcomes only: titration and additional add-on therapy permitted to reach desired blood pressure goal
Hasegawa [66] TALENT 52 29 Telmisartan, 40 mg od Seated 21 13
28 Losartan, 50 mg od 18 12
Lindholm [70] ALPINE 52 196 Candesartan, 16 mg od Seated 6 mths: 20.9 6 mths: 12.8
12 mths: 21 12 mths: 13
196 HCTZ, 25 mg od 6 mths: 23.9 6 mths: 13.9
12 mths: 22.8 12 mths: 12.9
Minami [50] 24 20 Telmisartan, 40 mg od Ambulatory Not calculable Not calculable
20 Losartan, 50 mg od Not calculable Not calculable
Murakami [48] ADIPO 24 9 Telmisartan, 40 mg od Seated 9.5 7.1
10 Valsartan, 80 mg od 6.4 2.9
Neldam [55] 24 123 Candesartan, 8 mg od Seated, trough 16.3 12
62 HCTZ, 12.5 mg od 18.8 11.4
Parving [16] IRMA2 104 195 Irbesartan, 150 mg od Seated 10 7
194 Irbesartan, 300 mg od 10 8
201 Placebo 97
Rossing [47] IRMA2 (substudy: 104 13 Irbesartan, 150 mg od Seated, trough 13 8
Single centre) 15 Irbesartan, 300 mg od 13 8
15 Placebo 11 9
Picca [51] 26 15 Losartan, 50 mg od Supine 27 17
15 Valsartan, 80 mg od 33 16
Rayner [67] 24 27 Losartan, 50 mg od Seated 18.3 12
25 Candesartan, 8 mg od 24 10.1
Schram [42] 52 20 Candesartan, 8 mg od Seated 6 mths: 17 6 mths: 11
12 mths: 18 12 mths: 13
19 HCTZ, 12.5 mg od 6 mths: 21 6 mths: 11
12 mths: 20 12 mths: 10
Spoelstra-de-man [41] 52 20 Candesartan, 8 mg od Ambulatory 26 17
19 HCTZ, 12.5 mg od 23 18
Solomon [68] VALIDD 38 166 Valsartan, 160 mg od Seated 12.8 -7.1
175 Placebo 9.7 5.5
Tsutamoto [60] 52 25 Olmesartan, 20 mg od Not reported 6 mths: 5 6 mths: 2
12 mths: 2.6 12 mths: 2
25 Candesartan, uncertain 6 mths: 3 6 mths: 2
12 mths: 2 12 mths: 2
Uzu [44] 24 16 Valsartan, 160 mg od Ambulatory Not calculable Not calculable
16 Losartan, 50 mg od + HCTZ Not calculable Not calculable
10 B. Tsoi et al.
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
Table 3. (continued).
First Author Study name Mean
follow-up,
wk
Number of
analysed
Intervention,
dose
Blood pressure Clinical outcomes (no of subjects)
Definition DSBP
(mmHg)
DDBP
(mmHg)
Mortality Major CV
event
MI Stroke
All-cause CV-related
Studies reporting both blood pressure and final clinical outcomes: titration and additional add-on therapy permitted to reach desired blood pressure goal
Brenner [14] RENAAL 177 751 Losartan, 50 mg od Not specified 12 mths: 6 12 mths: 4 155 NR 268 68 NR
24 mths: 9 24 mths: 5
762 Placebo 12 mths: 3 12 mths: 2 158 NR 247 50 NR
24 mths: 9 24 mths: 5
Foulquier [69] TRANSCEND 243 2547 Telmisartan, 80 mg od Not specified 7.4 NR NR 193 NR 97 102
2551 Placebo 3.5 NR NR 186 NR 130 125
Lewis [15] IDNT-2 104 579 Irbesartan, 75 mg od Seated 20 10 87 138
569 Placebo 14 10 93 144
Lithell [57] SCOPE 194 2477 Candesartan, 8–16 mg od Seated 21.7 10.8 259 145 242 70 89
2460 Placebo 18.5 9.2 266 152 268 63 115
Papademetriou [54] SCOPE (substudy) 167 754 Candesartan, 8 mg od Seated 22.2 6 82 47 75 23 20
764 Placebo 20.2 4.8 90 48 85 25 35
Saxby [53] SCOPE (substudy: 191 112 Candesartan, 8 mg od Seated 24 14 NR NR NR 3 4
Single centre) 116 Placebo 17 12 NR NR NR 5 8
Trenkwalder [56] SCOPE (substudy: 191 1253 Candesartan, 8 mg od Seated 21.8 11 NR NR NR NR NR
No add-on therapy) 845 Placebo 17.2 8.4 NR NR NR NR NR
Trenkwalder [58] SCOPE (substudy: 167 NR Candesartan, 8 mg od Seated Based on
subgroup
Based on
subgroup
Based on
subgroup
Based on
subgroup
Based on
subgroup
Based on
subgroup
Based on
subgroup
Pre-specified subgroup) NR Placebo Based on
subgroup
Based on
subgroup
Based on
subgroup
Based on
subgroup
Based on
subgroup
Based on
subgroup
Based on
subgroup
Menne [39] ROADMAP (substudy) 167 2043 Olmesartan, 40 mg od Seated 12.1 72514NRNRNR
1977 Placebo 8.2 4.9 14 3 NR NR NR
Abbreviations: CV[8_TD$DIFF], cardiovascular; DBP, diastolic blood pressure; HCTZ[9_TD$DIFF], hydrochlorothiazide; MI[10_TD$DIFF], myocardial infarction; SBP[11_TD$DIFF], systolic blood pressure.
ADIPO, Abdominal fat Depot Intervention Program of Okayama; ALPINE, Antihypertensive Treatment and Lipid Profile in a North of Sweden Efficacy Evaluation; IDNT-2, Irbesartan Diabetic Nephropathy Trial; IRMA2,
Irbesartan in Patients With Type 2 Diabetes and Microalbuminuria; RENAAL, Reduction of Endpoints in NIDDM with the Angiotensin II Antagonist Losartan Study; ROADMAP, Randomized Olmesartan and Diabetes
Microalbuminuria Prevention; SCOPE, Study on Cognition and Prognosis in the Elderly; TALENT, Telmisartan and Losartan Cardiac Evaluation Trial; TRANSCEND, Telmisartan Randomised AssessmeNt Study in ACE
iNtolerant;VALIDD, Valsartan in Diastolic Dysfunction.
Efficacy of Angiotensin II Antagonists in Essential Hypertension 11
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
interval indicates no difference between members of the ARB
class.
An advantage to the Bayesian approach is its capability of
estimating probabilities that each strategy would be the best
treatment compared to the others. Common to NMA is a rank
ordering which indicates the order in which the various
treatments are most to least likely efficacious. However,
we felt it was inappropriate and unlikely helpful to rank
treatments because of the imprecision in the estimates gen-
erated by the NMA. This would lead to considerable uncer-
tainty in the ranks. Rank probabilities for each treatment
were similar [72] and presenting such results may potentially
be misleading.
Efficacy: Final Clinical Outcomes
All trials reporting on final cardiovascular outcomes were
included in the NMA (Table 3)[15,16,36,54,65]. These were
all multi-national, two-armed, placebo-controlled RCTs that
compared ARB-initiated therapy to usual antihypertensive
therapy; adhering to treat-to-target designs in order to opti-
mise blood pressure goals. Participants were balanced in
gender (51.5% females) with the mean study duration being
2.68 years (range: 2.6–4.7). In total, 16,716 individuals were
randomly assigned to either one of the five ARBs or to
placebo, with subsequent titration and additional add-on
therapy permitted to reach the desired blood pressure goal.
For this NMA, placebo represented the bridging group.
Figure 2 Network of the direct comparisons available reporting the efficacy of ARB on blood pressure reduction. The size of
each treatment node is proportional to the number of analysed participants (sample size). The networks for systolic and
diastolic blood pressure were the same.
Figure 3 Relative treatment effects of ARBs at 24 weeks in terms of the absolute change [weighted mean difference (95%
credible interval)] in systolic blood pressure (blue) and diastolic blood pressure (light green). To interpret: weighted-mean
difference (WMD) <0 favours the column-defining treatment (e.g. eprosartan, compared to irbesartan, is associated with
increase in blood pressure by +3.14/ + 2.06 mmHg.) To obtain WMDs for comparisons in the opposite direction, the reverse
must be taken (eg. losartan, compared to irbesartan, is associated with a blood pressure changed of +2.02/ + 3.97 mm Hg).
Significant results are bolded.
12 B. Tsoi et al.
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
Figure 5 shows the network of eligible comparisons for the
NMA. Of the 15 possible pair-wise comparisons between the
five members of the ARB class and placebo, only five have
been studied directly in individual trials. There was no evi-
dence of non-convergence. Between-study heterogeneity
could not be assessed given that each direct pairwise com-
parison was informed by a single study.
Figures 6 and 7 summarise the results of the Bayesian
NMA. Across the outcomes of interest, the 95% credible
intervals of the pairwise comparison of individual members
within the ARB class suggests that members within the ARB
class have no differences in treatment effects with respect
to the risk of stroke, MI and mortality (i.e., all-cause of
CV-related).
Discussion
In the absence of direct, head-to-head evidence comparing
the long-term clinical efficacy of different members of the
ARB class, a NMA was conducted. Our findings show that
there was no difference in the comparative efficacy between
agents with respect to blood pressure control or the incidence
of cardiovascular events following long-term treatment.
These results summarise the experience so far from long-
term clinical trials on this drug class by incorporating both
direct and indirect comparisons, including those that have
never been directly compared quantitatively in previous
trials or reviews. Traditional meta-analyses published to date
tend to selectively compare a few members within the drug
class [67–72] and consider only the direct evidence within
trials. Our study is unique in that it considers the totality of
the available evidence base in conducting this analysis.
Some trials have suggested that certain agents within the
ARB class can lower blood pressure to a different extent
[14,19,73–77] and occasionally, traditional meta-analyses
have appeared claiming superiority of one agent over
another on certain outcomes [22,67–69,71]. However, these
meta-[33_TD$DIFF]analyses are limited, given issues of selection bias
and inappropriate methods of pooling data which violate
randomisation (i.e., such as combining results from a single
arm of a trial). Our study lends support to existing meta-
analyses that have shown comparable blood pressure lower-
ing capacity amongst different agents within the ARB class
[23,70,72,78,79]. This includes, specifically, one of the largest
pooled meta-analyses to date on this topic by Conlin et al.
[23]. A total of 11,281 patients were studied by combining
43 trials on four ARBs (i.e. candesartan, irbesartan, losartan,
valsartan) and it was found that the effectiveness of the
individual ARBs, following four to six weeks of
Figure 4 Relative treatment effects of ARBs at 52 weeks
in terms of the absolute change [weighted mean differ-
ence (95% credible interval)] in systolic blood pressure
(blue) and diastolic blood pressure (light green). To
interpret: weighted mean difference (WMD) <0 favours
the column-defining treatment (e.g. losartan, compared
to telmisartan, is associated with change in blood pres-
sure of 1.99/ + 0.99 mm Hg.) To obtain WMDs for
comparisons in the opposite direction, the reverse must
be taken. Significant results are bolded.
Figure 5 Network of the direct comparisons available
reporting the efficacy of ARB-initiated therapy on pre-
vention of cardiovascular-related mortality. The size of
each node is proportional to the number of analysed
participants (sample size). The network may differ
depending on the outcome reported as not all studies
reported on all final clinical outcomes of interest.
Figure 6 Relative treatment effects [odds ratio (95%
credible interval)] of ARBs on myocardial infarction
(MI) (grey) and stroke prevention (light green). To inter-
pret: odds ratio (OR) >1 favours the column-defining
treatment. To obtain ORs for comparisons in the oppo-
site direction, reciprocals should be taken (eg. OR for MI
in irbesartan compared with candesartan is 1/
1.59 = 0.63). Significant results are bolded.
*ND, no data available to conduct a comparison for that
particular outcome.
Efficacy of Angiotensin II Antagonists in Essential Hypertension 13
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
administration, in lowering blood pressure was similar
across agents [23]. Amongst individual agents, the absolute
weighted mean diastolic and systolic blood pressure reduc-
tion ranged from 8.2 to 8.9 mmHg and from 10.4 to
11.8 mmHg, respectively (i.e., this represents a maximum
difference between individual ARBs of 0.7 mmHg and
1.3 mmHg for diastolic and systolic blood pressure respec-
tively). The NMA technique employed in our analysis goes
beyond traditional meta-analysis in that it allows dissection
of the association of individual members of this drug class on
blood pressure by considering all evidence simultaneously
together. Furthermore, our NMA studied a longer treatment
period (i.e., 24 weeks) and found that the effectiveness of
individual members of the ARB class in lowering blood
pressure remained similar to what has been reported follow-
ing a shorter duration of treatment (i.e., 8 to 12 weeks [23]).
Treatment for hypertension is considered preventative as
elevated blood pressure does not manifest, in itself, as a
symptomatic illness. Rather, hypertension is of concern
due to its role as a modifiable risk factor for cardiovascular
events. To address whether members of the ARB class had
differing effect on the risk of cardiovascular disease, a second
NMA on these outcomes was conducted. This is novel as no
previous studies have done such a comparison within the
ARB drug class. Given that all long-term trials reporting final
clinical outcomes were treat-to-target designs, it is important
to understand that the original studies, in fact, involved a
comparison between ARB-initiated antihypertensive treat-
ment versus non-ARB initiated antihypertensive treatment.
Two main conclusions can be drawn from the analysis of
the final clinical outcomes. Firstly, the data provides no clear
evidence of any differential effects between individual mem-
bers of the ARB class with respect to prevention of major
cardiovascular diseases. In addition, as the common treat-
ment group linking this network was placebo (which, in such
study designs reflect usual antihypertensive management),
the findings suggest that ARB, as a class, had similar efficacy
to non-ARB initiated anti-hypertensive therapy across the
final clinical outcomes that were studied. To explore whether
ARB-initiated treatment led to other beneficial outcomes over
conventional hypertensive therapy without an ARB, a poten-
tial analysis of interest would be to compare the number of
additional antihypertensive drugs required to achieve blood
pressure control. However, this outcome was found to be
poorly and inconsistently reported across studies and no
further analysis was feasible. The hypothesis is that, if
patients on ARB-initiated therapy required fewer drugs to
achieve the desired blood pressure target than conventional
non-ARB therapy, this may prove to be clinically meaningful
to patients since this may lower treatment burden, increase
compliance and lead to cost-savings.
Strengths and Limitations
Hypertension is a relatively asymptomatic illness. It is there-
fore important to ensure that the downstream benefits of
preventing a primary clinical event are balanced against
the potential side effects from treatment [78]. We did not
investigate safety outcomes in this study. As the safety pro-
file of ARBs, as a class, have been described as placebo-like
with low rates of severe adverse events [20], we did not
analyse tolerability or safety.
As in other meta-analyses, given the lack of data within
each trial, we did not adjust our analyses for compliance to
the assigned treatment. Despite an understanding that the
risk for hypertensive-related cardiovascular complications is
dependent on both an individual’s blood pressure and the
duration in which blood pressure is adequately controlled,
compliance was difficult to assess given the lack of reporting.
To better address the impact of compliance on the efficacy of
particular drugs, a meta-regression could have been
Figure 7 Relative treatment effects [odds ratio (95% credible interval)] of ARBs on all-cause (grey) and CV-related (light
green) mortality. To interpret: odds ratio (OR) >1 favours the column-defining treatment. To obtain ORs for comparisons in
the opposite direction, reciprocals should be taken. Significant results are bolded.
*ND, no data available to conduct a comparison for that particular outcome.
14 B. Tsoi et al.
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
performed. In such an analysis, patient-level data including
important covariates such as the degree of compliance would
be combined by regression techniques to better understand
the impact of each covariate on the outcomes of interest.
However, adding covariates to the models would increase
model complexity and is ill-advised in situations where there
are small networks of trials. Furthermore, it would have
required access to patient-level data from all trials involved.
Finally, it is worth noting that our current analyses are
challenged by scant primary data on blood pressure lower-
ing, cardiovascular event rates, and related mortality. How-
ever, no current minimums are set for the number or sample
size of trials included in NMA [81,82], and clinically infor-
mative insights have been gained from similar high-quality
investigations [83] examining the clinical benefits of antihy-
pertensive treatment. Still, our findings must be cautiously
interpreted. Although there was no evidence of non-conver-
gence in any of the network models, the precision of our
estimates may have been affected by the small number of
studies involved in each network. Additionally, the wider
credible intervals produced by our analyses suggest that our
analyses may lack power. Our outcomes of interest are also
based on a small number of events, making it possible for
calculated odds ratios to [34_TD$DIFF]have been considerably affected by
even small difference between studies in how events were
classified. There was also no evidence available on the long-
term efficacy of azilsartan and eprosartan that could be
incorporated into this NMA.
Our review also has several strengths. Among them are a
protocol-driven approach adherent with best practice in sys-
tematic review conduct and reporting; a comprehensive lit-
erature search of multiple electronic databases; attempts to
contact authors to solicit missing data; double data abstrac-
tion; quality assessment of the primary studies using vali-
dated tools; and, appropriate methods for combining effect
estimates. Our NMA offers one of the most comprehensive
compilations of data specifically on the topic of comparative
ARB effectiveness, providing a balanced analysis of the evi-
dence base.
Conclusions
Our findings demonstrate that, overall, ARBs were capable of
lowering blood pressure and that the evidence, albeit limited,
suggest no difference between members within this class in
terms of their ability to control blood pressure. No individual
ARB offered significantly greater protection from cardiovas-
cular morbidity and mortality. Additional well-designed,
long-term, head-to-head comparative trials may be required
to better address whether the efficacy between individual
members in this drug class indeed varies given the potential
lack of power in this analysis. Until such studies are con-
ducted, prudence is advised, as there is no evidence to
support any preferential claims. Even compared to non-
ARB based treatment, there is little support for the superior-
ity of ARB-initiated therapy. Rather, this study highlights the
general role of blood pressure lowering therapies to reduce
the incidence of cardiovascular disease risks, independent of
which agent is used.
Declarations
Ethics Approval and Consent to
Participate
Not applicable.
Consent for Publication
Not applicable.
Availability of Data and Material
Data used to conduct the analyses is available in supplemen-
tal files.
Competing Interests
The authors have no competing interests to declare.
Funding
BT was supported by an Award from the Father Sean
O’Sullivan Research Centre, St. Joseph’s Healthcare Hamil-
ton and the CIHR Drug Safety and Effectiveness Cross-Dis-
ciplinary Training program. LA was supported by the
Comprehensive Research Education for Medical Students
Scholar Program at the University of Toronto. The remaining
authors have no potential conflicts of interest to disclose.
Authors’ Contributions
BT conceived and designed the study. BT, LA, CF conducted
the study. BT and AB conducted the analysis under the
guidance of EP, ML, and DOR. All authors read, edited
and approved the final version of the manuscript and agree
to be accountable for the work[35_TD$DIFF].
Acknowledgements
None.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.hlc.
2017.06.721.
References
[1] Daskalopoulou SS, Rabi DM, Zarnke KB, Dasgupta K, Nerenberg K,
Cloutier L, et al. The 2015 Canadian Hypertension Education Program
recommendations for blood pressure measurement, diagnosis, assess-
ment of risk, prevention, and treatment of hypertension. Can J Cardiol
2015;31:549–68.
Efficacy of Angiotensin II Antagonists in Essential Hypertension 15
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
[2] Mancia G, Fagard R, Narkiewicz K, Redon J, Zanchetti A, Bo
¨hm M, et al.
2013 ESH/ESC guidelines for the management of arterial hypertension:
the Task Force for the Management of Arterial Hypertension of the
European Society of Hypertension (ESH) and of the European Society
of Cardiology (ESC). Eur Heart J 2013;34:2159–219.
[3] James PA, Oparil S, Carter BL, Cushman WC, Dennison-Himmelfarb C,
Handler J, et al. 2014 evidence-based guideline for the management of
high blood pressure in adults: report from the panel members appointed
to the Eighth Joint National Committee (JNC 8). JAMA 2014;311:507–20.
[4] Collins R, MacMahon S. Blood pressure, antihypertensive drug treatment
and the risks of stroke and of coronary heart disease. Br Med Bull
1994;50:272–98.
[5] Gueyffier F, Froment A, Gouton M. New meta-analysis of treatment trials
of hypertension: improving the estimate of therapeutic benefit. J Hum
Hypertens 1996;10:1–8.
[6] Fretheim A, Odgaard-Jensen J, Brørs O, Madsen S, Njølstad I, Norheim
OF, et al. Comparative effectiveness of antihypertensive medication for
primary prevention of cardiovascular disease: systematic review and
multiple treatments meta-analysis. BMC Med 2012;10:33.
[7] Law MR, Morris JK, Wald NJ. Use of blood pressure lowering drugs in
the prevention of cardiovascular disease: meta-analysis of 147 random-
ised trials in the context of expectations from prospective epidemiologi-
cal studies. BMJ 2009;338. b1665–b1665.
[8] Psaty BM, Lumley T, Furberg CD, Schellenbaum G, Pahor M, Alderman
MH, et al. Health outcomes associated with various antihypertensive
therapies used as first-line agents: a network meta-analysis. JAMA 2003;
289:2534–44.
[9] Wright JM, Musini VM. First-line drugs for hypertension. Cochrane
Database Syst Rev 2009;CD001841.
[10] Akioyamen L, Levine M, Sherifali D, O’Reilly D, Frankfurter C, Pulle-
nayegum E, et al. Cardiovascular and cerebrovascular outcomes of long-
term angiotensin receptor blockade: meta-analyses of trials in essential
hypertension. J Am Soc Hypertens JASH 2016;10:55–69. e1.
[11] Hackam DG, Quinn RR, Ravani P, Rabi DM, Dasgupta K, Daskalopoulou
SS, et al. The 2013 Canadian Hypertension Education Program recom-
mendations for blood pressure measurement, diagnosis, assessment of
risk, prevention, and treatment of hypertension. Can J Cardiol 2013;
29:528–42.
[12] National Clinical Guideline Centre (UK). Hypertension: The Clinical
Management of Primary Hypertension in Adults: Update of Clinical
Guidelines 18 and 34 [Internet]. London: Royal College of Physicians
(UK); 2011 [cited 2017 Jan 20]. Available from: http://www.ncbi.nlm.nih.
gov/books/NBK83274/.
[13] Burnier M. Angiotensin II type 1 receptor blockers. Circulation
2001;103:904–12.
[14] Lacourcie
`re Y, Asmar R. A comparison of the efficacy and duration of
action of candesartan cilexetil and losartan as assessed by clinic and
ambulatory blood pressure after a missed dose, in truly hypertensive
patients: a placebo-controlled, forced titration study. Candesartan/Los-
artan study investigators. Am J Hypertens 1999;12:1181–7.
[15] Brenner BM, Cooper ME, de Zeeuw D, Keane WF, Mitch WE, Parving
HH, et al. Effects of losartan on renal and cardiovascular outcomes in
patients with type 2 diabetes and nephropathy. N Engl J Med 2001;
345:861–9.
[16] Lewis EJ, Hunsicker LG, Clarke WR, Berl T, Pohl MA, Lewis JB, 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.
[17] Parving HH, Lehnert H, Bro
¨chner-Mortensen J, Gomis R, Andersen S,
Arner P, et al. The effect of irbesartan on the development of diabetic
nephropathy in patients with type 2 diabetes. N Engl J Med 2001;345:
870–8.
[18] Viberti G, Wheeldon NM, MicroAlbuminuria Reduction With VALsar-
tan (MARVAL) Study Investigators. Microalbuminuria reduction with
valsartan in patients with type 2 diabetes mellitus: a blood pressure-
independent effect. Circulation 2002;106:672–8.
[19] Ball KJ, Williams PA, Stumpe KO. Relative efficacy of an angiotensin II
antagonist compared with other antihypertensive agents: Olmesartan
medoxomil versus antihypertensives. J Hypertens Suppl Off J Int Soc
Hypertens 2001;19:S49–56.
[20] Smith DHG, Dubiel R, Jones M. Use of 24-hour ambulatory blood pres-
sure monitoring to assess antihypertensive efficacy: a comparison of
olmesartan medoxomil, losartan potassium, valsartan, and irbesartan.
Am J Cardiovasc Drugs Drugs Devices Interv 2005;5:41–50.
[21] Smith DHG. Comparison of angiotensin II type 1 receptor antagonists in
the treatment of essential hypertension. Drugs 2008;68:1207–25.
[22] Nixon RM, Mu
¨ller E, Lowy A, Falvey H. Valsartan vs. other angiotensin II
receptor blockers in the treatment of hypertension: a meta-analytical
approach. Int J Clin Pract 2009;63:766–75.
[23] Conlin PR, Spence JD, Williams B, Ribeiro AB, Saito I, Benedict C, et al.
Angiotensin II antagonists for hypertension: are there differences in
efficacy? Am J Hypertens 2000;13:418–26.
[24] Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, Ioannidis
JPA, et al. The PRISMA Statement for Reporting Systematic Reviews and
Meta-Analyses of Studies That Evaluate Health Care Interventions:
Explanation and Elaboration. PLoS Med 2009;6:e1000100.
[27] Sica D, White WB, Weber MA, Bakris GL, Perez A, Cao C, et al. Compari-
son of the novel angiotensin II receptor blocker azilsartan medoxomil vs
valsartan by ambulatory blood pressure monitoring. J Clin Hypertens
2011;13:467–72.
[28] Dias S, Sutton AJ, Ades AE, Welton NJ. Evidence synthesis for decision
making 2: a generalized linear modeling framework for pairwise and
network meta-analysis of randomized controlled trials. Med Decis Mak
Int J Soc Med Decis Mak 2013;33:607–17.
[29] R Development Core Team. R: A language and environment for statisti-
cal computing. Vienna, Austria: R Foundation for Statistical Computing;
2009.
[30] Valkenhoef G van, Kuiper J. gemtc: Network Meta-Analysis Using
Bayesian Methods [Internet]. 2016 [cited 2017 Jan 20]. Available from:
https://cran.r-project.org/web/packages/gemtc/index.html.
[31] JAGS: A Program for Analysis of Bayesian Graphical Models using Gibbs
Sampling [Internet]. [cited 2017 Jan 20]. Available from: https://www.
researchgate.net/publication/2567033_JAGS_A_Program_for_
Analysis_of_Bayesian_Graphical_Models_using_Gibbs_Sampling.
[32] Brooks SP, Gelman A. General Methods for Monitoring Convergence of
Iterative Simulations. J Comput Graph Stat 1998;7:434–55.
[33] Eijkelkamp WBA, Zhang Z, Remuzzi G, Parving H-H, Cooper ME, Keane
WF, et al. Albuminuria is a target for renoprotective therapy independent
from blood pressure in patients with type 2 diabetic nephropathy: post
hoc analysis from the Reduction of Endpoints in NIDDM with the
Angiotensin II Antagonist Losartan (RENAAL) trial. J Am Soc Nephrol
JASN 2007;18:1540–6.
[34] Yoneda T, Takeda Y, Usukura M, Oda N, Takata H, Yamamoto Y, et al.
Aldosterone breakthrough during angiotensin II receptor blockade in
hypertensive patients with diabetes mellitus. Am J Hypertens 2007;20:
1329–33.
[35] Derosa G, Ragonesi PD, Mugellini A, Ciccarelli L, Fogari R. Effects of
telmisartan compared with eprosartan on blood pressure control, glucose
metabolism and lipid profile in hypertensive, type 2 diabetic patients: a
randomized, double-blind, placebo-controlled 12-month study. Hyper-
tens Res Off J Jpn Soc Hypertens 2004;27:457–64.
[36] Menne J, Izzo JL, Ito S, Januszewicz A, Katayama S, Chatzykirkou C, et al.
Prevention of microalbuminuria in patients with type 2 diabetes and
hypertension. J Hypertens 2012;30:811–8. discussion 818.
[37] Sawaki H, Terasaki J, Fujita A, Nakagawa S, Kanatsuna N, Sadahiro K,
et al. A renoprotective effect of low dose losartan in patients with type 2
diabetes. Diabetes Res Clin Pract 2008;79:86–90.
[38] Spoelstra-de Man AME, van Ittersum FJ, Schram MT, Kamp O, van Dijk
RA, Ijzerman RG, et al. Aggressive antihypertensive strategies based on
hydrochlorothiazide, candesartan or lisinopril decrease left ventricular
mass and improve arterial compliance in patients with type II diabetes
mellitus and hypertension. J Hum Hypertens 2006;20:599–611.
[39] Schram MT, van Ittersum FJ, Spoelstra-de Man A, van Dijk RA, Schalk-
wijk CG, Ijzerman RG, et al. Aggressive antihypertensive therapy based
on hydrochlorothiazide, candesartan or lisinopril as initial choice in
hypertensive type II diabetic individuals: effects on albumin excretion,
endothelial function and inflammation in a double-blind, randomized
clinical trial. J Hum Hypertens 2005;19:429–37.
[40] Uzu T, Sakaguchi M, Yokomaku Y, Kume S, Kanasaki M, Isshiki K, et al.
Effects of high sodium intake and diuretics on the circadian rhythm of
blood pressure in type 2 diabetic patients treated with an angiotensin II
receptor blocker. Clin Exp Nephrol 2009;13:300–6.
[41] Uzu T, Sakaguchi M, Tsuda A, Kadota A, Yokomaku Y, Kume S, et al.
Effects of blood pressure and the renin-angiotensin system on platelet
activation in type 2 diabetes. J Diabetes Investig 2010;1:196–201.
[42] Derosa G, Fogari E, D’Angelo A, Cicero AFG, Salvadeo SA, Ragonesi PD,
et al. Metabolic effects of telmisartan and irbesartan in type 2 diabetic
patients with metabolic syndrome treated with rosiglitazone. J Clin
Pharm Ther 2007;32:261–8.
[43] Derosa G, Cicero AFG, D’Angelo A, Ragonesi PD, Ciccarelli L, Piccinni
MN, et al. Telmisartan and irbesartan therapy in type 2 diabetic patients
treated with rosiglitazone: effects on insulin-resistance, leptin and tumor
16 B. Tsoi et al.
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721
necrosis factor-alpha. Hypertens Res Off J Jpn Soc Hypertens
2006;29:849–56.
[44] Rossing K, Christensen PK, Andersen S, Hovind P, Hansen HP, Parving
H-H, et al. Comparative effects of Irbesartan on ambulatory and office
blood pressure: a substudy of ambulatory blood pressure from the
Irbesartan in Patients with Type 2 Diabetes and Microalbuminuria study.
Diabetes Care 2003;26:569–74.
[45] Murakami K, Wada J, Ogawa D, Horiguchi CS, Miyoshi T, Sasaki M, et al.
The effects of telmisartan treatment on the abdominal fat depot in
patients with metabolic syndrome and essential hypertension: Abdomi-
nal fat Depot Intervention Program of Okayama (ADIPO). Diab Vasc Dis
Res 2013;10:93–6.
[46] Rizos CV, Milionis HJ, Kostapanos MS, Florentin M, Kostara CE, Elisaf
MS, et al. Effects of rosuvastatin combined with olmesartan, irbesartan,
or telmisartan on indices of glucose metabolism in Greek adults with
impaired fasting glucose, hypertension, and mixed hyperlipidemia: a
24-week, randomized, open-label, prospective study. Clin Ther 2010;32:
492–505.
[47] Minami J, Furukata S, Ishimitsu T, Matsuoka H. Comparison of therapies
between fixed-dose telmisartan/hydrochlorothiazide and losartan/
hydrochlorothiazide in patients with mild to moderate hypertension.
Int Heart J 2009;50:85–93.
[48] Picca M, Agozzino F, Pelosi G. Effects of losartan and valsartan on left
ventricular hypertrophy and function in essential hypertension. Adv
Ther 2004;21:76–86.
[49] Nedogoda SV, Ledyaeva AA, Chumachok EV, Tsoma VV, Mazina G,
Salasyuk AS, et al. Randomized trial of perindopril, enalapril, losartan
and telmisartan in overweight or obese patients with hypertension. Clin
Drug Investig 2013;33:553–61.
[50] Saxby BK, Harrington F, Wesnes KA, McKeith IG, Ford GA. Candesartan
and cognitive decline in older patients with hypertension: a substudy of
the SCOPE trial. Neurology 2008;70:1858–66.
[51] Papademetriou V, Farsang C, Elmfeldt D, Hofman A, Lithell H, Olofsson
B, et al. Stroke prevention with the angiotensin II type 1-receptor blocker
candesartan in elderly patients with isolated systolic hypertension: the
Study on Cognition and Prognosis in the Elderly (SCOPE). J Am Coll
Cardiol 2004;44:1175–80.
[52] Neldam S, Forse
´n B, Multicentre Study Group. Antihypertensive treat-
ment in elderly patients aged 75 years or over: a 24-week study of the
tolerability of candesartan cilexetil in relation to hydrochlorothiazide.
Drugs Aging 2001;18:225–32.
[53] Trenkwalder P. The Study on COgnition and Prognosis in the Elderly
(SCOPE)–recent analyses. J Hypertens Suppl Off J Int Soc Hypertens
2006;24:S107–14.
[54] Lithell H, Hansson L, Skoog I, Elmfeldt D, Hofman A, Olofsson B, et al.
The Study on Cognition and Prognosis in the Elderly (SCOPE): principal
results of a randomized double-blind intervention trial. J Hypertens
2003;21:875–86.
[55] Trenkwalder P, Elmfeldt D, Hofman A, Lithell H, Olofsson B, Papade-
metriou V, et al. The Study on COgnition and Prognosis in the Elderly
(SCOPE) – major CV events and stroke in subgroups of patients. Blood
Press 2005;14:31–7.
[56] Negro R, Formoso G, Hassan H. The effects of irbesartan and telmisartan
on metabolic parameters and blood pressure in obese, insulin resistant,
hypertensive patients. J Endocrinol Invest 2006;29:957–61.
[57] Tsutamoto T, Nishiyama K, Yamaji M, Kawahara C, Fujii M, Yamamoto
T, et al. Comparison of the long-term effects of candesartan and olme-
sartan on plasma angiotensin II and left ventricular mass index in
patients with hypertension. Hypertens Res Off J Jpn Soc Hypertens
2010;33:118–22.
[58] Makris TK, Stavroulakis G, Papadopoulos DP, Krespi P, Hatzizacharias
A, Zilidis A, et al. Eprosartan effect on fibrinolytic/hemostatic variables
in arterial hypertension: a comparative study to losartan. Drugs Exp Clin
Res 2004;30:125–32.
[59] Galzerano D, Tammaro P, Cerciello A, Breglio R, Mallardo M, Lama D,
et al. Freehand three-dimensional echocardiographic evaluation of the
effect of telmisartan compared with hydrochlorothiazide on left ventric-
ular mass in hypertensive patients with mild-to-moderate hypertension:
a multicentre study. J Hum Hypertens 2004;18:53–9.
[60] Tedesco MA, Ratti G, Aquino D, Limongelli G, di Salvo G, Mennella S,
et al. Effects of losartan on hypertension and left ventricular mas s: a long-
term study. J Hum Hypertens 1998;12:505–10.
[61] Tedesco MA, Ratti G, Mennella S, Manzo G, Grieco M, Rainone AC, et al.
Comparison of losartan and hydrochlorothiazide on cognitive function
and quality of life in hypertensive patients. Am J Hypertens 1999;
12:1130–4.
[62] Hasegawa H, Takano H, Narumi H, Ohtsuka M, Mizuguchi T, Namiki T,
et al. Effects of telmisartan and losartan on cardiovascular protection in
Japanese hypertensive patients. Hypertens Res Off J Jpn Soc Hypertens
2011;34:1179–84.
[63] Rayner BL, Trinder YA, Baines D, Isaacs S, Opie LH. Effect of losartan
versus candesartan on uric acid, renal function, and fibrinogen in
patients with hypertension and hyperuricemia associated with diuretics.
Am J Hypertens 2006;19:208–13.
[64] Solomon SD, Janardhanan R, Verma A, Bourgoun M, Daley WL,
Purkayastha D, et al. Effect of angiotensin receptor blockade and
antihypertensive drugs on diastolic function in patients with hyper-
tension and diastolic dysfunction: a randomised trial. Lancet 2007;
369:2079–87.
[65] Foulquier S, Bo
¨hm M, Schmieder R, Sleight P, Teo K, Yusuf S, et al.
Impact of telmisartan on cardiovascular outcome in hypertensive
patients at high risk: a Telmisartan Randomised AssessmeNt Study in
ACE iNtolerant subjects with cardiovascular Disease subanalysis. J
Hypertens 2014;32:1334–41.
[66] Taddei S. RAS inhibitors’ dose-dependent efficacy: myth or reality? Curr
Med Res Opin 2015;31:1245–56.
[67] Grosso AM, Bodalia PN, Macallister RJ, Hingorani AD, Moon JC, Scott
MA. Comparative clinical- and cost-effectiveness of candesartan and
losartan in the management of hypertension and heart failure: a system-
atic review, meta- and cost-utility analysis. Int J Clin Pract 2011;65:
253–63.
[68] Meredith PA, Murray LS, McInnes GT. Comparison of the efficacy of
candesartan and losartan: a meta-analysis of trials in the treatment of
hypertension. J Hum Hypertens 2010;24:525–31.
[69] Xi G-L, Cheng J-W, Lu G-C. Meta-analysis of randomized controlled
trials comparing telmisartan with losartan in the treatment of patients
with hypertension. Am J Hypertens 2008;21:546–52.
[70] Xu F-Y, Yang B, Shi D, Li H, Zou Z, Shi X-Y. Antihypertensive effects and
safety of eprosartan: a meta-analysis of randomized controlled trials. Eur
J Clin Pharmacol 2012;68:195–205.
[71] Zheng Zhenfeng, Shi Huilan, Jia Junya, Li Dong, Lin Shan. A systematic
review and meta-analysis of candesartan and losartan in the manage-
ment of essential hypertension. J Renin-Angiotensin-Aldosterone Syst
JRAAS 2011;12:365–74.
[72] Zheng Z, Lin S, Shi H. A systematic review and meta-analysis of telmi-
sartan versus valsartan in the management of essential hypertension. J
Clin Hypertens 2010;12:414–21.
[73] Kassler-Taub K, Littlejohn T, Elliott W, Ruddy T, Adler E. Comparative
efficacy of two angiotensin II receptor antagonists, irbesartan and los-
artan in mild-to-moderate hypertension: Irbesartan/Losartan Study
Investigators. Am J Hypertens 1998;11:445–53.
[74] Lacourcie
`re Y, Krzesinski J-M, White WB, Davidai G, Schumacher H.
Sustained antihypertensive activity of telmisartan compared with val-
sartan. Blood Press Monit 2004;9:203–10.
[75] Mallion J, Siche J, Lacourcie
`re Y. ABPM comparison of the antihyperten-
sive profiles of the selective angiotensin II receptor antagonists telmi-
sartan and losartan in patients with mild-to-moderate hypertension. J
Hum Hypertens 1999;13:657–64.
[76] Mancia G, Korlipara K, van Rossum P, Villa G, Silvert B. An ambulatory
blood pressure monitoring study of the comparative antihypertensive
efficacy of two angiotensin II receptor antagonists, irbesartan and val-
sartan. Blood Press Monit 2002;7:135–42.
[77] Oparil S, Guthrie R, Lewin AJ, Marbury T, Reilly K, Triscari J, et al. An
elective-titration study of the comparative effectiveness of two angioten-
sin II-receptor blockers, irbesartan and losartan: Irbesartan/Losartan
Study Investigators. Clin Ther 1998;20:398–409.
[78] Fabia MJ, Abdilla N, Oltra R, Fernandez C, Redon J. Antihypertensive
activity of angiotensin II AT1 receptor antagonists: a systematic review of
studies with 24 h ambulatory blood pressure monitoring. J Hypertens
2007;25:1327–36.
[79] Feldman RD, Hussain Y, Kuyper LM, McAlister FA, Padwal RS, Tobe
SW. Intraclass differences among antihypertensive drugs. Annu Rev
Pharmacol Toxicol 2015;55:333–52.
[81] Salanti G, Giovane CD, Chaimani A, Caldwell DM, Higgins JPT. Evalu-
ating the Quality of Evidence from a Network Meta-Analysis. PLoS ONE
2014;9:e99682.
[82] Cipriani A, Higgins JPT, Geddes JR, Salanti G. Conceptual and Technical
Challenges in Network Meta-analysis. Ann Intern Med 2013;159:130.
[83] Palmer SC, Mavridis D, Navarese E, Craig JC, Tonelli M, Salanti G, et al.
Comparative efficacy and safety of blood pressure-lowering agents in
adults with diabetes and kidney disease: a network meta-analysis. The
Lancet 2015;385:2047–56.
Efficacy of Angiotensin II Antagonists in Essential Hypertension 17
HLC 2433 No. of Pages 17
Please cite this article in press as: Tsoi B, et al. Comparative Efficacy of Angiotensin II Antagonists in Essential Hypertension:
Systematic Review and Network Meta-analysis of Randomised Controlled Trials. Heart, Lung and Circulation (2017), http://
dx.doi.org/10.1016/j.hlc.2017.06.721