JACC: HEART FAILURE VOL. 7, NO.

4, 2019

ª 2019 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION

PUBLISHED BY ELSEVIER

Target Doses of Heart Failure

Medical Therapy and Blood Pressure
Insights From the CHAMP-HF Registry

Poghni A. Peri-Okonny, MD,a Xiaojuan Mi, PHD,b Yevgeniy Khariton, MD, MSC,a Krishna K. Patel, MD, MSC,a
Laine Thomas, PHD,b Gregg C. Fonarow, MD,c Puza P. Sharma, MBBS, MPH, PHD,d Carol I. Duffy, DO,d
Nancy M. Albert, PHD,e Javed Butler, MD, MPH,f Adrian F. Hernandez, MD,b Kevin McCague, MA,d
Fredonia B. Williams, EDD,g Adam D. DeVore, MD,b J. Herbert Patterson, PHARMD,h John A. Spertus, MD, MPHa,i

ABSTRACT

OBJECTIVES This study sought to determine the rate of use of target doses of foundational guideline-directed medical
therapy (GDMT) in a contemporary cohort of patients with heart failure with reduced ejection fraction (HFrEF) across
systolic blood pressure (SBP) categories.

BACKGROUND Patients with HFrEF are infrequently titrated to recommended doses of GDMT. The relationship
between SBP and achieving GDMT target doses is not well studied.

METHODS Patients enrolled in the CHAMP-HF (Change the Management of Patients With Heart Failure) registry
without documented intolerance to angiotensin-converting enzyme inhibitors (ACEIs), angiotensin receptor blockers
(ARBs), angiotensin receptor-neprilysin inhibitors (ARNIs), and beta blockers (BBs) were assessed at enrollment. We
estimated the proportion receiving target doses (% of target dose [95% confidence interval (CI)]) based on the most
recent American College of Cardiology/American Heart Association/Heart Failure Society of America heart failure
guidelines at baseline in all patients, and by SBP category ($110 vs. <110 mm Hg).

RESULTS Of the 3,095 patients eligible for analysis, 2,421 (78.2%) had SBP $110 mm Hg. The proportion of patients
receiving target doses were 18.7% (95% CI: 17.3% to 20.0%; BB), 10.8% (95% CI: 9.7% to 11.9%; ACEI/ARB), and 2.0%
(95% CI: 1.5% to 2.5%; ARNI). Among those with SBP <110 mm Hg (n ¼ 674), 17.5% (95% CI: 14.6% to 20.4%; BB),
6.2% (95% CI: 4.4% to 8.1%; ACEI/ARB), and 1.8% (95% CI: 0.8% to 2.8%; ARNI) were receiving target doses. Among
those with SBP $110 mm Hg (n ¼ 2,421), 19.0% (95% CI: 17.4% to 20.6%; BB), 12.1% (95% CI: 10.8% to 13.4%; ACEI/
ARB), and 2.0% (95% CI: 1.5% to 2.6%; ARNI) were receiving target doses.

CONCLUSIONS In a large, contemporary registry of outpatients with chronic HFrEF eligible for treatment with BBs and
ACEI/ARB/ARNI, <20% of patients were receiving target doses, even among those with SBP $110 mm Hg.
(J Am Coll Cardiol HF 2019;7:350–8) © 2019 by the American College of Cardiology Foundation.

H
few
eart failure with reduced ejection fraction
(HFrEF) is associated with
morbidity and mortality (1). Over the past
decades, advances in medical
significant

therapy
the beta adrenergic system with beta blockers (BBs)
and inhibition of the angiotensin pathway with
angiotensin-converting enzyme inhibitors (ACEIs)/
angiotensin receptor blockers (ARBs) have been the
have resulted in a substantial reduction in poor car- cornerstone of HFrEF treatment (5), sacubitril/valsar-
diovascular outcomes (2–4). Although inhibition of tan, which belongs to the class of angiotensin II

From the aUniversity of Missouri, Kansas City, Missouri; bDuke Clinical Research Institute, Durham, North Carolina; cUniversity of
California Los Angeles, Los Angeles, California; dNovartis Pharmaceuticals Corporation, East Hanover, New Jersey; eCleveland
Clinic, Cleveland, Ohio; fUniversity of Mississippi, University, Mississippi; gMended Hearts, Hunstville, Alabama; hUniversity of
North Carolina, Chapel Hill, North Carolina; and the iSaint Luke’s Mid America Heart Institute, Kansas City, Missouri. CHAMP-HF
is funded by the Novartis Pharmaceuticals Corporation. Drs. Peri-Okonny, Khariton, and Patel are supported by the National Heart,
Lung, and Blood Institutes of Health under award number T32HL110837. Dr. Thomas has received grants from Novartis Phar-
maceuticals Corporation. Dr. Fonarow has received grants from the National Institutes of Health; has received personal fees from
Abbott, Amgen, Janssen, Medtronic, and Novartis Pharmaceuticals Corporation; and has been on the steering committee for Get
With the Guidelines. Drs. Sharma, Duffy, and McCague are employed by and have stock ownership with Novartis Pharmaceuticals

ISSN 2213-1779/$36.00 https://doi.org/10.1016/j.jchf.2018.11.011

JACC: HEART FAILURE VOL. 7, NO. 4, 2019 Peri-Okonny et al. 351
APRIL 2019:350–8 Blood Pressure and GDMT

receptor blocker-neprilysin inhibitors (ARNIs), has Patients eligible for enrollment met the ABBREVIATIONS

recently been shown to provide additional morbidity following criteria: 1) age $18 years; 2) primary AND ACRONYMS

and mortality reduction over enalapril alone among diagnosis of HFrEF (left ventricular ejection
ACEI = angiotensin converting
patients on background BB therapy (5,6). fraction [LVEF] #40% within 12 months of enzyme inhibitor
enrollment); 3) prescribed $1 oral pharmaco-
SEE PAGE 359 ARB = angiotensin receptor
therapy for HF at the time of enrollment; and blocker

In patients with HFrEF, target medication doses 4) willingness to complete protocol re- ARNI = angiotensin receptor-
established in clinical trials are encouraged in national quirements for study visits, procedures, and neprilysin inhibitor

guidelines (5). This stems from evidence showing questionnaires. Patients were excluded if BB = beta-blocker

dose-related relationships between outcomes and they were ineligible (actively participating in GDMT = guideline-directed
ACEI (7) and BB doses (8,9). Despite this robust evi- an interventional clinical research study, medical therapy

dence, target doses of these important therapies are receiving comfort care measures only or LVEF = left ventricular
ejection fraction
often underused in clinical practice (10,11). One po- enrolled in a hospice program, had a life ex-
pectancy of <1 year, and had a history of, or MRA = mineralocorticoid
tential barrier to intensifying guideline-directed
receptor antagonist
medical therapy (GDMT) may be concerns regarding planned heart transplantation, left ventricu-
TIA = transient ischemic attack
low blood pressure, which can cause symptomatic lar assist device implantation, or dialysis).
TD = target dose
hypotension and fatigue (12). Low blood pressure, as a Additionally, we excluded those who re-
barrier to achieving targeted dosing of GDMT in ported HF medication-related side effects (new/
routine clinical care has not been described. Under- worsening cough, worsening renal dysfunction, renal
standing whether blood pressure is a barrier to failure, angioedema, dizziness/lightheadedness/hy-
achieving targeted doses of HFrEF therapy can help potension, or clinically significant increase in serum
guide how to achieve more aggressive dosing of potassium), had a known contraindication to ACEI/
evidence-based therapies. To address this gap, we ARB, ARNI, or BB, or were missing systolic blood
used a large, multicenter, contemporary outpatient pressure (SBP) and demographic information at
registry of patients with HFrEF to conduct an analysis enrollment.
to describe the intensity of heart failure (HF) medica- Additional data collected at enrollment included
tion dosing and to determine the proportion of patients patient-level demographics and clinical characteris-
who had lower blood pressures might be a barrier to the tics, medical history, laboratory results, use of HF
intensification of ACEIs/ARBs/ARNIs or BBs. medications and devices, and patient-reported health
status. Eligible sites were identified based on the
METHODS completion of a feasibility survey, which provided
investigators with the opportunity to ensure broad
STUDY DESIGN. For this cross-sectional analysis, we geographic and provider specialty representation.
used data at enrollment from the CHAMP-HF (Change Study coordinators at each site were responsible for
the Management of Patients With Heart Failure) identification and enrollment of subjects during the
registry: a prospective, observational study of out- course of a scheduled outpatient visit, with this
patients with HFrEF at 151 U.S. practice sites (13). analysis being limited to only those patients enrolled

Corporation. Dr. Albert has received personal fees from Novartis Pharmaceuticals Corporation. Dr. Butler has received grants from
the National Institutes of Health and the European Union; and has received personal fees from Abbott, Adrenomed, Amgen, Array,
AstraZeneca, Bayer, BerlinCures, Boehringer Ingelheim, Bristol-Myers Squibb, Cardiocell, CVRx, G3 Pharmaceuticals, Immolife,
Janssen, Lantheus, LinaNova, Luitpold, Medscape, Medtronic, Merck, Novartis Pharmaceuticals Corporation, NovoNordisk,
Relypsa, Roche, Sanofi, StealthPeptide, SC Pharma, V-Wave Limited, Vifor, and ZS Pharma. Dr. Hernandez has received grants
from AstraZeneca, Bristol-Myers Squibb, GlaxoSmithKline, Luitpold Pharmaceuticals, Merck, and Novartis Pharmaceuticals Cor-
poration; and has received personal fees from Bayer, Boston Scientific, and Novartis Pharmaceuticals Corporation. Dr. DeVore has
received research funding from Akros Medical, American Heart Association, NIH, Amgen, Bayer, Intra-Cellular Therapies, Luitpold
Pharmaceuticals, NHLBI, Novartis Pharmaceuticals Corporation, and Patient-Centered Outcomes Research Instititute (PCORI);
and has received personal fees from Novartis Pharmaceuticals Corporation. Dr. Patterson has received grants from Amgen, Bristol-
Myers Squibb, Merck, and Novartis Pharmaceuticals Corporation; and has received personal fees from Novartis Pharmaceuticals
Corporation, Amgen, Otsuka, Merck, Boehringer Ingelheim, and Bristol-Myers Squibb. Dr. Spertus has received grants from the
National Institutes of Health, Bayer, and Abbott Vascular; has received support from Novartis Pharmaceuticals Corporation; has
received personal fees from Bayer, AstraZeneca, United Healthcare, MyoCardia, and Janssen; has intellectual property rights for
the Kansas City Cardiomyopathy Questionnaire; and has an equity interest in Health Outcomes Sciences. The content herein is
solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. All
other authors have reported that they have no relationships relevant to the contents of this paper to disclose.

Manuscript received October 24, 2018; revised manuscript received November 20, 2018, accepted November 20, 2018.
352 Peri-Okonny et al. JACC: HEART FAILURE VOL. 7, NO. 4, 2019

Blood Pressure and GDMT APRIL 2019:350–8

DATA COLLECTION. Site coordinators interviewed

T A B L E 1 Demographic and Clinical Characteristics of the Study Population Overall and by
Systolic Blood Pressure Categories
patients to collect their sociodemographic charac-
teristics and health status, and abstracted informa-
Systolic Blood Pressure
tion from the medical record regarding medical
Overall <110 mm Hg $110 mm Hg
history and medications at enrollment. The primary
(N ¼ 3,095) (n ¼ 674) (n ¼ 2,421)
outcome for this analysis was the achievement of
Demographics
Age (yrs) target dose BB ($100% target dose [yes/no]) and
Median (Q1–Q3) 68.0 (59.0–75.0) 66.0 (57.0–74.0) 68.0 (59.0–76.0) ACEI/ARB/ARNI ($100% target dose [yes/no]) based
<40 100 (3.2) 35 (5.2) 65 (2.7) on each patient’s daily dose at enrollment compared
40–64 1,150 (37.2) 275 (40.8) 875 (36.1) with the guideline-recommended target daily doses
65–80 1,448 (46.8) 288 (42.7) 1,160 (47.9) (Online Table 1) (5,14). We limited our analysis to
80þ 397 (12.8) 76 (11.3) 321 (13.3)
these medication classes because, barring any intol-
Sex
erance, guidelines recommend all patients with
Male 2,184 (70.6) 477 (70.8) 1,707 (70.5)
Female 911 (29.4) 197 (29.2) 714 (29.5)
HFrEF be placed on maximally tolerated (target)
Race doses of these medications before considering addi-
White 2,283 (73.8) 513 (76.1) 1,770 (73.1) tional therapy (14). Medication daily dose was
Black 519 (16.8) 111 (16.5) 408 (16.9) calculated based on dose per administration and
Other/unknown 293 (9.5) 50 (7.4) 243 (10.0) frequency. SBP readings were obtained from the
Medical history
patients’ baseline clinic visit as was routine for each
Chronic renal insufficiency 552 (17.8) 127 (18.8) 425 (17.6)
participating site.
COPD 943 (30.5) 196 (29.1) 747 (30.9)
Dementia 69 (2.2) 11 (1.6) 58 (2.4)
STATISTICAL ANALYSIS. The enrollment character-
Diabetes mellitus 1,285 (41.5) 251 (37.2) 1,034 (42.7)
istics of the CHAMP-HF cohort were stratified by
Hepatic dysfunction 64 (2.1) 19 (2.8) 45 (1.9)
SBP categories (<110 vs. $110 mm Hg). We used
Peripheral artery disease 422 (13.6) 75 (11.1) 347 (14.3)
Stroke/TIA 325 (10.5) 69 (10.2) 256 (10.6) proportions for categorical variables and medians
Ischemic heart disease 1,195 (38.6) 258 (38.3) 937 (38.7) with quartiles for continuous variables. We calcu-
Myocardial infarction 1,039 (33.6) 244 (36.2) 795 (32.8) lated the proportion of patients receiving ACEIs/
Hypertrophic cardiomyopathy 214 (6.9) 22 (3.3) 192 (7.9) ARBs/ARNIs and BBs in the overall cohort and by
Restrictive cardiomyopathy 43 (1.4) 9 (1.3) 34 (1.4) SBP groups. The proportion of patients achieving
Vital signs
$100% of the target dose are presented along with
Systolic blood pressure (mm Hg) 120 (110–130) 100 (96–105) 124 (118–136)
95% confidence intervals (CIs). The proportion of
Diastolic blood pressure (mm Hg) 72 (64–80) 62 (60–69) 75 (70–80)
Heart rate (bpm) 72 (66–81) 73 (66–82) 72 (65–80)
patients on <50% and 50% to <100% were also
Clinical measures/labs calculated. The chi-square test was used to compare
NYHA functional classification the proportion of medication classes across SBP
I 339 (11.0) 64 (9.5) 275 (11.4) groups.
II 1,787 (57.7) 366 (54.3) 1,421 (58.7) Distribution of SBP was plotted for patients who
III 814 (26.3) 206 (30.6) 608 (25.1)
were not on GDMT defined as fulfilling any of the
IV 69 (2.2) 20 (3.0) 49 (2.0)
following: 1) target dose of BB but not target dose of
LVEF (%) 30 (23–36) 28 (20–33) 32 (25–37)
ACEI/ARB/ARNI; 2) target dose of ACEI/ARB/ARNI
<20 324 (10.5) 131 (19.4) 193 (8.0)
20–25 722 (23.3) 174 (25.8) 548 (22.6)
but not target dose of BB; and 3) neither target dose
26–35 1,257 (40.6) 255 (37.8) 1,002 (41.4) of BB nor target dose of ACEI/ARB/ARNI, so that the
36–40 772 (24.9) 106 (15.7) 666 (27.5) percent of patients not receiving guideline-
eGFR (ml/min/1.73 m2) 60 (50–70) 60 (47–61) 60 (51–73) recommended dosing across different SBP thresh-
<30 94 (3.0) 31 (4.6) 63 (2.6) olds could be described. A sensitivity analysis was
30–45 259 (8.4) 67 (9.9) 192 (7.9)
conducted using the following SBP cutpoints (100,
45–60 425 (13.7) 116 (17.2) 309 (12.8)
120, and 130 mm Hg) to examine the effects on the
$60 1,205 (38.9) 239 (35.5) 966 (39.9)
results. An additional sensitivity analysis excluded
Continued on the next page
those with heart rate <60 beats/min because
between December 2015 and August 2017. CHAMP-HF bradycardia may be a reason for nonintensification
was sponsored by Novartis Pharmaceuticals Corpo- of BB.
ration, and all participating sites obtained local or All analyses were performed using SAS software
central institutional review board approval before version 9.4, SAS Institute, Cary, North Carolina.
patient enrollment as well as informed consent from Analyses were performed independently by the
each participant. Duke Clinical Research Institute, and the lead
JACC: HEART FAILURE VOL. 7, NO. 4, 2019 Peri-Okonny et al. 353
APRIL 2019:350–8 Blood Pressure and GDMT

author takes responsibility for guiding data analysis

T A B L E 1 Continued
and interpretation.
Systolic Blood Pressure

Overall <110 mm Hg $110 mm Hg

RESULTS (N ¼ 3,095) (n ¼ 674) (n ¼ 2,421)

Medication at enrollment
There were 4,493 patients enrolled across 151 sites in ACEI 1,279 (41.3) 286 (42.4) 993 (41.0)
the CHAMP-HF registry between December 2015 and ARB 646 (20.9) 99 (14.7) 547 (22.6)
August 2017. After exclusion of patients ineligible for MRA 1,006 (32.5) 298 (44.2) 708 (29.2)
the registry (n ¼ 42), those with any documented Beta blocker 2,560 (82.7) 577 (85.6) 1,983 (81.9)

medication-related side effects that could limit use or ARNI 400 (12.9) 118 (17.5) 282 (11.6)
Quality of life assessment
dosing (n ¼ 1,105), specific contraindications to ACEI/
KCCQ12 68.8 (49.0–85.4) 65.4 (45.8–84.0) 69.3 (50.0–85.4)
ARB/ARNI or BB (n ¼ 84), missing SBP (n ¼ 161), and
missing demographic information (n ¼ 6), 3,095 Values are median (interquartile range) or n (%).
patients across 148 sites were included in the final ACEI ¼ angiotensin-converting enzyme inhibitor; ARB ¼ angiotensin receptor blocker; ARNI ¼ angiotensin
receptor-neprilysin inhibitor; COPD ¼ chronic obstructive pulmonary disease; eGFR ¼ estimated glomerular
study cohort (Online Figure 1). The median SBP was filtration rate; KCCQ ¼ Kansas City Cardiomyopathy Questionnaire; LVEF ¼ left ventricular ejection fraction;
MRA ¼ mineralocorticoid receptor antagonist; TIA ¼ transient ischemic attack.
120 mm Hg (interquartile range [IQR]: 110 to 130
mm Hg) and median diastolic blood pressure was 72
mm Hg (IQR: 64 to 80 mm Hg). The majority of pa-
tients were male (70.6%), older than 64 years (59.6%),
ACEI/ARB, 2.0% (95% CI: 1.5% to 2.5%) for ARNI, and
white (73.8%), and had SBP $110 mm Hg (n ¼ 2,421,
18.7% (95% CI: 17.3% to 20.0%) for BBs (Table 2).
78.2%) (Table 1). The median Kansas City Cardiomy-
Among those with SBP <110 mm Hg who were eligible
opathy Questionnaire-Short Form overall summary
for both ACEI/ARB/ARNI and BB (n ¼ 674), 6.2% (95%
score was 69 (IQR: 49 to 85.4) and most patients
CI: 4.4% to 8.1%), 1.8% (95% CI: 0.8% to 2.8%), and
(84%, n ¼ 2,601) had either New York Heart Associa-
17.5% (95% CI: 14.6% to 20.4%) were receiving target
tion functional class II or III symptoms. The median
doses of ACEI/ARB, ARNI, and BBs, respectively. A
LVEF was 30% (IQR: 23% to 36%).
greater proportion of patients were receiving <50% of
Among those with SBP <110 mm Hg, the median
target doses than were receiving >50% (Table 2,
SBP/diastolic blood pressure was 100 mm Hg (IQR: 96
Figure 2). Among those with SBP $110 mm Hg who
to 105 mm Hg)/62 mm Hg (IQR: 60 to 69 mm Hg),
were eligible for both ACE/ARB/ARNI and BB (n ¼
compared with 124 mm Hg (IQR: 118 to 136 mm Hg)/75
2,421), the proportion of patients receiving target
mm Hg (IQR: 70 to 80 mm Hg) among those with
doses were 12.1% (95% CI: 10.8% to 13.4%) for ACEI/
SBP $110 mm Hg. The median heart rate was 73
ARB, 2.0% (95% CI: 1.5% to 2.6%) for ARNI, and 19.0%
beats/min (IQR: 66 to 82 beats/min) among those
(95% CI: 17.4% to 20.6%) for BBs. Similarly, a greater
with SBP <110 mm Hg and 72 beats/min (IQR: 65 to 80
proportion of patients were receiving <50% than
beats/min) among those with SBP $110 mm Hg
were receiving >50% of target doses for each drug
(Table 1). BB use was lower (81.9% vs. 85.6%;
class despite having SBP $110 mm Hg (Table 2,
p ¼ 0.03), ARB use was higher (22.6% vs. 14.7%;
Figure 3).
p < 0.0001), ARNI use was lower (11.6% vs. 17.5%;
Finally, among patients who were already pre-
p < 0.0001), and ACEI use was similar (41.0% vs.
scribed 2 classes of medications including a BB (i.e.,
42.4%; p ¼ 0.50) in patients with SBP $110 mm Hg
an ACEI/ARB/ARNI and a BB [n ¼ 1,619]), only 8.8%
compared with those with SBP <110 mm Hg. The
(n ¼ 142) were receiving target doses of both
prevalence of comorbidities and other baseline
(Table 3, Figure 4). Among those with SBP <110
characteristics by SBP categories are presented in
mm Hg, 5.8% (n ¼ 22) were receiving target doses of
Table 1. The distribution of SBP among patients not
both an ACEI/ARB/ARNI and a BB (Table 3). For
receiving target doses had 120  17 mm Hg (Online
those with SBP $110 mm Hg, 9.7% (n ¼ 120) were
Figure 2).
receiving target doses of both an ACEI/ARB/ARNI
PROPORTION RECEIVING TARGET DOSES OF ACEIs/ and a BB (Table 3). A sensitivity analysis excluding
ARBs/ARNIs OR BBs. Among the 3,095 patients who patients with heart rate <60 beats/min produced
were eligible for ACE/ARB/ARNI and BB, 61% were qualitatively similar results (Online Tables 2a to 2c).
receiving ACEI/ARB, 12.9% were receiving ARNI, and In an additional sensitivity analysis, the proportion
82.7% were receiving a BB (Figure 1). The proportion of patients receiving target doses of ACEI/ARB
of patients receiving target doses for these medica- increased slightly with increase in SBP cutoff (11.2%,
tion classes were 10.8% (95% CI: 9.7% to 11.9%) for 13.6%, and 17.3% among patients with SBP $100
354 Peri-Okonny et al. JACC: HEART FAILURE VOL. 7, NO. 4, 2019

Blood Pressure and GDMT APRIL 2019:350–8

practice partly depends on using therapies in a

F I G U R E 1 Distribution of Percent Target Doses for Each Medication Class in the
Overall Cohort
manner similar to that tested in the clinical trials and
endorsed by guidelines. In fact, in an explicit test of
the intensity of ACEI dose, the ATLAS (Assessment of
Treatment with Lisinopril and Survival) trial found
lower rates of mortality and hospitalization with
higher versus lower doses of lisinopril (7). Similar
dose-dependent benefit has been shown with angio-
tensin blockers and BBs (8,15). We analyzed a
contemporary cohort of patients with HFrEF enrolled
in the CHAMP-HF registry because the prescription
patterns of dosing intensity according to blood pres-
sure in patients with HFrEF in contemporary U.S.
practice is not known. We found an overwhelming
majority of patients eligible for either BBs or ACEIs/
ARBs/ARNIs were not receiving target doses and
that most patients were on <50% of guideline-
recommended target doses, even after excluding
those with medication-specific intolerance or con-
traindications. Importantly, SBP did not appear to be
a barrier to intensifying treatment overall, as marked
ACEI ¼ angiotensin converting enzyme inhibitor; ARNI ¼ angiotensin receptor-neprilysin under treatment was noted in those with a
inhibitor; ARB ¼ angiotensin receptor blocker; BB ¼ beta blocker; SBP ¼ systolic blood SBP $110 mm Hg, with an overall similar pattern
pressure. noticed when we considered SBP $120 mm Hg and
SBP $130 mm Hg. Despite a majority of our cohort
(52.3%) receiving both a BB and an ACEI/ARB/ARNI,
mm Hg, $120 mm Hg, and $130 mm Hg, respec-
the proportion of patients receiving target doses of
tively) (Online Tables 3a, 4a, and 5a). The results
both types of medication (beta blockade and angio-
were otherwise similar to the primary analysis
tensin inhibition) was only slightly higher among
(Online Tables 3 to 5).
those with SBP $110 mm Hg (9.7%) compared with
DISCUSSION those having a SBP <110 mm Hg (5.8%). Finding
that <1 in 10 patients in both cohorts were optimally
Translating the benefits of medical therapies that treated with GDMT highlights an important oppor-
were previously observed in clinical trials to clinical tunity to intensify foundational HF therapies
to potentially further improve morbidity and mor-
tality (9,16).
T A B L E 2 Baseline Medication in the Overall Cohort (N ¼ 3,095) Among Patients With
SBP <110 mm Hg (n ¼ 674) and SBP $110 mm Hg (n ¼ 2,421) We used an SBP cutoff of 110 mm Hg in our
primary analyses because prior analyses using an
Not 50% to
Treated Treated <50% TD <100% TD $100% TD %TD (95% CI) SBP threshold of 110 showed that although overall
Overall cohort risk of death and hospitalizations increase below
ACEI/ARB 1,901 1,194 1,122 434 334 10.8 (9.7–11.9) this threshold in HFrEF, ARB (17), ARNI (18), and
ARNI 400 2,695 220 118 61 2.0 (1.48–2.46) BB (19) were associated with reduced mortality and
Beta blocker 2,043 1,052 942 518 578 18.7 (17.3–20.0) hospitalizations even at SBP <110 mm Hg. We also
SBP <110 mm Hg
selected this threshold to reflect a range above
ACEI/ARB 382 292 265 73 42 6.2 (4.41–8.06)
which: 1) clinicians will likely be more comfortable
ARNI 118 556 78 28 12 1.8 (0.78–2.78)
Beta blocker 483 191 248 114 118 17.5 (14.6–20.4)
initiating or intensifying HF medications that affect
SBP $100 mm Hg systemic blood pressure; and 2) further reductions
ACEI/ARB 1,519 902 857 361 292 12.1 (10.8–13.4) in SBP documented in clinical trials with uptitra-
ARNI 282 2,139 142 90 49 2.0 (1.46–2.58) tion were unlikely to result in symptomatic hypo-
Beta blocker 1,560 861 694 404 460 19.0 (17.4–20.6) tension (17,18). Yet, even with higher SBP levels in
sensitivity analyses, the proportion of patients
Values are n, unless otherwise indicated.
%TD ¼ n ($100%TD)/n (treated); CI ¼ confidence interval; TD ¼ target dose; other abbreviations as in Table 1. receiving suboptimal medication doses remained
high.
JACC: HEART FAILURE VOL. 7, NO. 4, 2019 Peri-Okonny et al. 355
APRIL 2019:350–8 Blood Pressure and GDMT

Our study adds to the body of evidence showing

F I G U R E 2 Distribution of Percent Target Doses for Each Medication Class Among
that despite the dose-response relationship for GDMT
Patients With SBP <110 mm Hg
with outcomes (20–22), suboptimal dosing of HF
medications remains highly prevalent, and provides
new insight that SBP may not be a primary barrier to
dosing intensification. Suboptimal dosing has previ-
ously been observed in a population of outpatients
with HFrEF (23) and in patients who underwent
implantable cardioverter-defibrillator implantation,
an indication predicated upon the use of GDMT at
appropriate doses (24). Patient factors (e.g., older age,
comorbid lung disease or diabetes, intolerance or
contraindications, and medication cost), and
physician factors (e.g., physician knowledge/global
self-confidence) have been suggested as potential
reasons for suboptimal dosing of HF therapy (25,26).
Our study prospectively collected some of these data
and excluded patients intolerant of ACEIs/ARBs/
ARNIs and BBs while further exploring whether blood
pressure and heart rate could have potentially pre-
cluded the institution or intensification of these HF
therapies (12,27). Abbreviations as in Figure 1.

Clinical hypotension is commonly considered at

SBP <90 mm Hg (28), and, if present, may cause cli-
nicians to defer or de-escalate therapy (especially if act on a clinically identified problem (32). Although
patients are symptomatic). Moreover, SBP ranges we posited that blood pressure might be a potential
between 90 and 110 mm Hg may still generate factor contributing to clinical inertia related to
considerable provider discomfort with respect to intensifying HF therapy, it seemed to have at most a
intensifying HF medications that reduce SBP. How- modest impact, suggesting that other factors
ever, an SBP >110 mm Hg is not considered hypo-
tension, and we posit that a greater proportion of
clinicians should be comfortable instituting or F I G U R E 3 Distribution of Percent Target Doses for Each Medication Class Among
intensifying HF medical therapy at these levels. Patients With SBP $110 mm Hg
Although low SBP has been associated with poor
prognosis in HFrEF (29–31), angiotensin inhibition
and beta blockade have shown improved outcomes,
independent of baseline blood pressure (18,21). Our
findings of lower overall use of target doses of BBs
and ACEIs/ARBs/ARNIs in those with SBP <110
mm Hg may reflect clinician hesitation to uptitrate
medical therapy due to the association of low SBP and
worse outcomes. However, blood pressure level
should not be a significant factor among those with
SBP $110 mm Hg. There is a need for qualitative
studies to further refine our understanding of how
clinicians use SBP in decision-making on intensifica-
tion of medication therapy in HF. In the interim, more
aggressive efforts to ensure appropriate medication
dosing such as the creation of electronic medical re-
cord or clinic-based performance measures may be
needed.
Clinical inertia may contribute to suboptimal
dosing of effective therapies (10). Clinical inertia is a Abbreviations as in Figure 1.

well-described concept that refers to the failure to

356 Peri-Okonny et al. JACC: HEART FAILURE VOL. 7, NO. 4, 2019

Blood Pressure and GDMT APRIL 2019:350–8

critical opportunity for developing and promoting

T A B L E 3 Summary of Patients Receiving Target BB and ACEI/ARB/ARNI
Doses Among Those on Both a BB and 1 of ACEI/ARB/ARNI (N ¼ 1,619)
strategies aimed at stimulating wide spread adop-
tion of optimal HF treatments, including treating to
Target ACEI/ARB/ARNI Target ACEI/ARB/ARNI
SBP <110 mm Hg SBP $110 mm Hg
effective medication doses. Performance measures
Target BB No Yes Total No Yes Total
are 1 way of stimulating adoption of best medical
No 263 16 279 755 116 871
practices, but current performance measures rely on
69.4 4.2 73.6 60.9 9.4 70.2 the presence or absence of use of specific medica-
Yes 78 22 100 249 120 369 tion classes and not the intensity of treatment (38).
20.6 5.8 26.4 20.1 9.7 29.8
Given the evidence that medication dose is impor-
Total 341 38 379 1,004 236 1,240
90.0 10.0 100 81.0 19.0 100 tant to achieve better outcomes in HFrEF, perfor-
mance measures should consider evolving to reflect
Values are n and %.
the achievement of appropriate doses.
BB ¼ beta-blocker; other abbreviations as in Table 1.

STUDY LIMITATIONS. Our study is a descriptive

analysis and was intended to explore a potential
contributed to the under-dosing of HF medications. rationale for not intensifying HF medications to
There is an urgent need for concurrent efforts aimed guideline-recommended doses. As a cross-sectional
at availing patients of the best medical therapy analysis, we cannot comment on changes in treat-
while simultaneously investigating factors related to ment over time (in subsequent visits), although pa-
suboptimal dosing of important HF medications in tients had been under the care of their providers
patients who can tolerate them. Despite the ad- before enrollment in CHAMP-HF. We did not include
vances in treatment, mortality rates for HFrEF other evidence-based mediations such as mineralo-
remain high (33,34), and the previously observed corticoid antagonists (MRAs), which also reduce
improvements in mortality among patients with HF blood pressure. However, MRA use is typically rec-
appears to have slowed down over recent decades ommended after optimization of ACEIs/ARBs and
(35). Given the multiple observations that subopti- BBs. Although we included a diverse cohort from 151
mal dosing of critical HF therapy is highly prevalent, practices, these findings may not be generalizable to
and the potential survival benefit associated with other U.S. patients and practices. While careful
optimal dosing of HF therapy (36,37), there is a attention was paid to adjudicating intolerances,
there may have been patients with undocumented
intolerances or preferences precluding the use of
BBs or ACEIs/ARBs/ARNIs. Further, we assessed SBP
F I G U R E 4 Distribution of Percent of Patients Receiving Target Doses of BB and ACEI
at 1 point in time. It is unknown if SBP was labile,
or ARB or ARNI Among Patients Receiving BB and ACEI or ARB or ARNI
whether physicians relied upon home SBP mea-
surements, or if patients had reported symptoms
potentially associated with symptomatic hypoten-
sion that may have led to a reduction or inability to
uptitrate BB or ACEI/ARB therapies at the time of
the clinical visit. ARNI use was low in our study and
this is likely a reflection of the recent approval of
this drug class. There was a high representation of
white male participants in our study. Whether our
results apply to other demographic groups remains
unclear. Finally, the study was not designed to
identify why optimal medication doses were not
used.

CONCLUSIONS

In a contemporary cohort of patients with HFrEF, we

found that more than 90% of patients determined to
Abbreviations as in Figure 1. be eligible for BB and ACEI/ARB/ARNI, and with
SBP $110 mm Hg were not receiving target doses of
JACC: HEART FAILURE VOL. 7, NO. 4, 2019 Peri-Okonny et al. 357
APRIL 2019:350–8 Blood Pressure and GDMT

therapies that have been previously shown to reduce PERSPECTIVES

morbidity and mortality in HFrEF. Further efforts,
which may include intensive education through the
COMPETENCY IN MEDICAL KNOWLEDGE: In a contempo-
electronic medical record, defined protocols, im-
rary cohort of patients with HFrEF, use of target doses of ACEI/
provements in transition of patients between pro-
ARB/ARNI or BB was low among patients eligible for these
viders, creation of dosing-based performance
therapies (at most 20%), even with SBP $130 mm Hg. The
measures, patient education, and other strategies are
majority of patients were not treated or were on <50% of rec-
needed to support the use of more effective,
ommended doses. No therapy was used at guideline recom-
guideline-recommended doses of HF medications in
mended doses in more than 1 in 5 patients.
patients with HFrEF.

TRANSLATIONAL OUTLOOK: Further studies into strategies

ADDRESS FOR CORRESPONDENCE: Dr. Peri-Okonny,
that improve intensification of foundational evidence-based
Saint Luke’s Mid America Heart Institute, 4401 Wor-
medical therapies in HFrEF are needed.
nall Road, Kansas City, Missouri 64111. E-mail:
periokonnyp@umkc.edu.

REFERENCES

1. Benjamin EJ, Blaha MJ, Chiuve SE, et al. Heart dysfunction: results from the HF-ACTION (Heart 18. Bohm M, Young R, Jhund PS, et al. Systolic
disease and stroke statistics-2017 update: a report Failure: A Controlled Trial Investigating Outcomes blood pressure, cardiovascular outcomes and ef-
from the American Heart Association. Circulation of Exercise Training) trial. J Am Coll Cardiol 2012; ficacy and safety of sacubitril/valsartan (LCZ696)
2017;135:e146–603. 60:208–15. in patients with chronic heart failure and reduced
ejection fraction: results from PARADIGM-HF. Eur
2. Sacks CA, Jarcho JA, Curfman GD. Paradigm 10. Bhatt AS, DeVore AD, DeWald TA, Swedberg K,
Heart J 2017;38:1132–43.
shifts in heart-failure therapy — a timeline. N Engl Mentz RJ. Achieving a maximally tolerated beta-
J Med 2014;371:989–91. blocker dose in heart failure patients: is there 19. Montero-Perez-Barquero M, Flather M,
room for improvement? J Am Coll Cardiol 2017; Roughton M, et al. Influence of systolic blood
3. Colucci WS, Packer M, Bristow MR, et al. Car-
69:2542–50. pressure on clinical outcomes in elderly heart
vedilol inhibits clinical progression in patients with
mild symptoms of heart failure. US Carvedilol failure patients treated with nebivolol: data from
11. Heywood JT, Fonarow GC, Yancy CW, et al.
Heart Failure Study Group. Circulation 1996;94: the SENIORS trial. Eur J Heart Fail 2014;16:
Comparison of medical therapy dosing in out-
2800–6. 1009–15.
patients cared for in cardiology practices with
4. Packer M, Fowler MB, Roecker EB, et al. Effect heart failure and reduced ejection fraction with 20. Wikstrand J, Hjalmarson A, Waagstein F, et al.
of carvedilol on the morbidity of patients with and without device therapy: report from IMPROVE Dose of metoprolol CR/XL and clinical outcomes in
severe chronic heart failure: results of the carve- HF. Circ Heart Fail 2010;3:596–605. patients with heart failure: analysis of the experi-
dilol prospective randomized cumulative survival 12. Parwani P, Ryan J. Heart failure patients with ence in metoprolol CR/XL randomized intervention
(COPERNICUS) study. Circulation 2002;106: low blood pressure: how should we manage trial in chronic heart failure (MERIT-HF). J Am Coll
2194–9. neurohormonal blocking drugs? Circ Heart Fail Cardiol 2002;40:491–8.

5. Yancy CW, Jessup M, Bozkurt B, et al. 2017 ACC/ 2012;5:819. 21. Metra M, Torp-Pedersen C, Swedberg K, et al.
AHA/HFSA Focused Update of the 2013 ACCF/AHA 13. DeVore AD, Thomas L, Albert NM, et al. Influence of heart rate, blood pressure, and beta-
Guideline for the Management of Heart Failure: A Change the management of patients with heart blocker dose on outcome and the differences in
Report of the American College of Cardiology/ failure: Rationale and design of the CHAMP-HF outcome between carvedilol and metoprolol tartrate
American Heart Association Task Force on Clinical registry. Am Heart J 2017;189:177–83. in patients with chronic heart failure: results from the
Practice Guidelines and the Heart Failure Society of COMET trial. Eur Heart J 2005;26:2259–68.
14. Yancy CW, Jessup M, Bozkurt B, et al. 2013
America. J Am Coll Cardiol 2017;70:776–803.
ACCF/AHA guideline for the management of heart 22. Vardeny O, Claggett B, Packer M, et al. Effi-
6. Packer M, McMurray JJ, Desai AS, et al. failure: a report of the American College of Car- cacy of sacubitril/valsartan vs. enalapril at lower
Angiotensin receptor neprilysin inhibition diology Foundation/American Heart Association than target doses in heart failure with reduced
compared with enalapril on the risk of clinical Task Force on Practice Guidelines. J Am Coll Car- ejection fraction: the PARADIGM-HF trial. Eur J
progression in surviving patients with heart fail- diol 2013;62:e147–239. Heart Fail 2016;18:1228–34.
ure. Circulation 2015;131:54–61.
15. Konstam MA, Neaton JD, Dickstein K, et al. 23. Greene SJ, Butler J, Albert NM, et al. Medical
7. Packer M, Poole-Wilson PA, Armstrong PW, Effects of high-dose versus low-dose losartan on therapy for heart failure with reduced ejection
et al. Comparative effects of low and high doses of clinical outcomes in patients with heart failure fraction: the CHAMP-HF registry. J Am Coll Cardiol
the angiotensin-converting enzyme inhibitor, lisi- (HEAAL study): a randomised, double-blind trial. 2018;72:351–66.
nopril, on morbidity and mortality in chronic heart Lancet 2009;374:1840–8.
24. Miller AL, Wang Y, Curtis J, Masoudi FA,
failure. ATLAS Study Group. Circulation 1999;100:
16. Ouwerkerk W, Voors AA, Anker SD, et al. De- Buxton AE, Wang TY. Optimal medical therapy
2312–8.
terminants and clinical outcome of uptitration of use among patients receiving implantable car-
8. Bristow MR, Gilbert EM, Abraham WT, et al. ACE-inhibitors and beta-blockers in patients with dioverter/defibrillators: insights from the National
Carvedilol produces dose-related improvements in heart failure: a prospective European study. Eur Cardiovascular Data Registry. Arch Int Med 2012;
left ventricular function and survival in subjects Heart J 2017;38:1883–90. 172:64–7.
with chronic heart failure. Circulation 1996;94:
17. Anand IS, Rector TS, Kuskowski M, Thomas S, 25. Komajda M, Follath F, Swedberg K, et al. The
2807–16.
Holwerda NJ, Cohn JN. Effect of baseline and EuroHeart Failure Survey programme—a survey on
9. Fiuzat M, Wojdyla D, Kitzman D, et al. Rela- changes in systolic blood pressure over time on the quality of care among patients with heart
tionship of beta-blocker dose with outcomes in the effectiveness of valsartan in the Valsartan failure in Europe. Part 2: treatment. Eur Heart J
ambulatory heart failure patients with systolic Heart Failure Trial. Circ Heart Fail 2008;1:34–42. 2003;24:464–74.
358 Peri-Okonny et al. JACC: HEART FAILURE VOL. 7, NO. 4, 2019

Blood Pressure and GDMT APRIL 2019:350–8

26. Fuat A, Hungin APS, Murphy JJ. Barriers to 31. Komajda M, Bohm M, Borer JS, et al. Effi- neprilysin inhibitor therapy in heart failure. JAMA
accurate diagnosis and effective management of cacy and safety of ivabradine in patients with Cardiol 2016;1:714–7.
heart failure in primary care: qualitative study. chronic systolic heart failure according to blood
37. Fonarow GC, Yancy CW, Hernandez AF,
BMJ 2003;326:196. pressure level in SHIFT. Eur J Heart Fail 2014;
Peterson ED, Spertus JA, Heidenreich PA. Poten-
16:810–6.
27. Gorelik O, Feldman L, Cohen N. Heart failure tial impact of optimal implementation of
and orthostatic hypotension. Heart Fail Rev 2016; 32. Phillips LS, Branch WT Jr., Cook CB, et al. evidence-based heart failure therapies on mortal-
21:529–38. Clinical inertia. Ann Int Med 2001;135:825–34. ity. Am Heart J 2011;161:1024–30.e3.

28. Patel PA, Heizer G, O’Connor CM, et al. Hy- 33. Curtis LH, Greiner MA, Hammill BG, et al. Early 38. Bonow RO, Ganiats TG, Beam CT, et al. ACCF/
potension during hospitalization for acute heart and long-term outcomes of heart failure in elderly AHA/AMA-PCPI 2011 performance measures for
failure is independently associated with 30-day persons, 2001–2005. Archives of internal medi- adults with heart failure: a report of the American
mortality: findings from ASCEND-HF. Circ Heart cine 2008;168:2481–8. College of Cardiology Foundation/American Heart
Fail 2014;7:918–25. Association Task Force on Performance Measures
34. Hummel SL, Pauli NP, Krumholz HM, et al. and the American Medical Association-Physician
29. Ambrosy AP, Vaduganathan M, Mentz RJ, et al.
Thirty-day outcomes in Medicare patients with Consortium for Performance Improvement. J Am
Clinical profile and prognostic value of low systolic
heart failure at heart transplant centers. Circ Heart Coll Cardiol 2012;59:1812–32.
blood pressure in patients hospitalized for heart
Fail 2010;3:244–52.
failure with reduced ejection fraction: insights from
the Efficacy of Vasopressin Antagonism in Heart 35. Tsao CW, Lyass A, Enserro D, et al. Temporal
Failure: Outcome Study with Tolvaptan (EVEREST) trends in the incidence of and mortality associated
KEY WORDS heart failure, systolic blood
trial. Am Heart J 2013;165:216–25. with heart failure with preserved and reduced
pressure, target dose
ejection fraction. J Am Coll Cardiol HF 2018;6:
30. Gheorghiade M, Vaduganathan M, Ambrosy A,
678–85.
et al. Current management and future directions
for the treatment of patients hospitalized for heart 36. Fonarow GC, Hernandez AF, Solomon SD, A PPE NDI X For supplemental tables and
failure with low blood pressure. Heart Fail Rev Yancy CW. Potential mortality reduction with figures, please see the online version of
2013;18:107–22. optimal implementation of angiotensin receptor this paper.