ORIGINAL ARTICLE

Adoption of Sacubitril/Valsartan
for the Management of Patients
With Heart Failure

See Editorial by Colaco and Kazi Lindsey R.

Sangaralingham, MPH
Downloaded from http://circheartfailure.ahajournals.org/ by guest on February 16, 2018

BACKGROUND: The US Food and Drug Administration approved the use S. Jeson Sangaralingham,
of sacubitril/valsartan in patients with heart failure with reduced ejection MS, PhD
fraction in July 2015. We aimed to assess the adoption and prescription Nilay D. Shah, PhD
Xiaoxi Yao, PhD
drug costs of sacubitril/valsartan in its first 18 months after Food and
Shannon M. Dunlay, MD,
Drug Administration approval.
MS
METHODS AND RESULTS: Using a large US insurance database, we
identified privately insured and Medicare Advantage beneficiaries who
filled a first prescription for sacubitril/valsartan between July 1, 2015,
and December 31, 2016. We compared them to patients treated with
an angiotensin-converting enzyme inhibitor or angiotensin receptor
blocker. Outcomes included adoption, prescription drug costs, and 180-
day adherence, defined as a proportion of days covered ≥80%. A total of
2244 patients initiated sacubitril/valsartan. Although the number of users
increased over time, the proportion of heart failure with reduced ejection
fraction patients taking sacubitril/valsartan remained low (<3%). Patients
prescribed sacubitril/valsartan were younger, more often male, with
less comorbidity than those taking an angiotensin-converting enzyme
inhibitor/angiotensin receptor blocker. Although a majority of prescription
costs were covered by the health plan (mean, $328.37; median, $362.44
per 30-day prescription), out-of-pocket costs were still high (mean,
$71.16; median, $40.27). By comparison, median out-of-pocket costs
were $2 to $3 for lisinopril, losartan, carvedilol, and spironolactone.
Overall, 59.1% of patients were adherent to sacubitril/valsartan. Refill
patterns suggested that nearly half of nonadherent patients discontinued
sacubitril/valsartan within 180 days of starting.
CONCLUSIONS: Adoption of sacubitril/valsartan after Food and Drug
Administration approval has been slow and may be associated with the
Correspondence to: Shannon M.
high cost. Dunlay, MD, MS, Mayo Clinic, 200
First St SW, Rochester, MN 55905.
E-mail dunlay.shannon@mayo.edu

Key Words: adoption ◼ drug

costs ◼ heart failure ◼ sacubitril
◼ valsartan
© 2018 American Heart
Association, Inc.

Circ Heart Fail. 2018;11:e004302. DOI: 10.1161/CIRCHEARTFAILURE.117.004302 February 2018 1

 Sangaralingham et al; Adoption of Sacubitril/Valsartan in Heart Failure

HF with reduced ejection fraction (HFrEF). Given these

WHAT IS NEW? results from a single, multicenter, international, double-
blind, randomized trial, the US Food and Drug Admin-
• In this claims-based cohort study, adoption of
sacubitril/valsartan was slow after Food and Drug istration (FDA) approved sacubitril/valsartan (Entresto,
Administration approval. Novartis) for use in patients with chronic HFrEF and New
• Sacubitril/valsartan was more often prescribed in York Heart Association functional classes II to IV symp-
patients with heart failure with reduced ejection toms in July 2015. Approval by the European Commis-
fraction who were younger, male, and had less sion then followed in November 2015. Less than a year
comorbidity. later, sacubitril/valsartan received a class I recommenda-
• Out-of-pocket and insurer-paid prescription drug tion for use in chronic symptomatic HFrEF in both the US
costs were much higher than other guideline- and European HF clinical practice guidelines.6,7
directed heart failure therapies. Since approval and the incorporation of sacubitril/
• In total, 40.9% of patients were nonadherent/dis-
valsartan into the HF guidelines, the adoption of sacu-
continued sacubitril/valsartan in the first 6 months
after initiation. Nonadherence and discontinuation bitril/valsartan into real-world clinical practice in the US
were more common in Black patients and patients has yet to be described. Thus, the goal of this study was
who had not previously been taking an angioten- aimed to assess the adoption, continuation, and pre-
Downloaded from http://circheartfailure.ahajournals.org/ by guest on February 16, 2018

sin-converting enzyme inhibitor or angiotensin scription cost of sacubitril/valsartan in its first year after
receptor blocker. FDA approval in individuals enrolled in US commercial
and Medicare Advantage health plans.
WHAT ARE THE CLINICAL
IMPLICATIONS? METHODS
• Slow adoption of sacubitril/valsartan may be asso-
ciated with the high cost. Data Source
• Factors contributing to the high rates of nonadher- We conducted a retrospective analysis using the OptumLabs
ence/discontinuation observed in this study require Data Warehouse, a large database including over 100 mil-
further exploration. lion individuals with private and Medicare Advantage health
plans.8,9 The database comprises of medical and pharmacy
claims for individuals in all 50 states and of all ages and eth-
nic and racial groups. Medical claims include claims for pro-

H
eart failure (HF) is complex pathophysiological fessional (eg, physician), facility (eg, hospital), and outpatient
and clinical syndrome that continues to be a prescription medication services. Pharmacy claims include
leading cause of morbidity, hospitalization, and fill date, strength, days supply, prescriber specialty, generic/
mortality.1,2 Although significant improvement in HF brand names, and paid amounts (out-of-pocket [OOP] and
survival has been observed over the last few decades,3 health plan paid amounts). The data, analytic methods, and
largely because of the identification and increased use study materials will not be made available to other research-
ers for purposes of reproducing the results or replicating the
of pharmacological therapies that improve outcomes
procedure, as access to the data requires a partnership with
such as those targeting the renin–angiotensin–aldoste- OptumLabs. Pursuant to the Health Insurance Portability and
rone system,4 the 5-year mortality and rates of hospital- Accountability Act, the use of deidentified data does not
ization remain unacceptably high.1 Because of this, the require Institutional Review Board approval.
discovery and development of novel pharmacothera-
pies remain paramount to improving outcomes in HF.
Study Populations
In 2014, a major breakthrough in HF therapeutics
First, we identified all individuals 18 years or older who filled
occurred with the publication of the results of the PAR- a prescription for sacubitril/valsartan between July 1, 2015,
ADIGM-HF trial (Prospective Comparison of Angiotensin and December 31, 2016 (sacubitril/valsartan cohort). For each
II Receptor Blocker Neprilysin Inhibitor With Angiotensin- individual, we defined the index initiation date as the date of
Converting Enzyme Inhibitor [ACEi] to Determine Impact first fill date of sacubitril/valsartan.
on Global Mortality and Morbidity in HF) which was To contextualize and understand which patients with
stopped after a median follow-up of 27 months because HFrEF were initiated on sacubitril/valsartan, we also identi-
of a significant reduction in morbidity and mortality in fied all patients with systolic HF (HFrEF cohort). We relied
participants receiving AngII receptor blocker neprilysin on billing codes for systolic HF, an approach which has been
demonstrated to be 97.7% specific for identifying individuals
inhibitor.5 Specifically, McMurray et al5 reported that
with HF and an EF <45%.10 We included those with a single
LCZ696, a first in class small molecule that delivers simul- inpatient claim or at least 2 outpatient claims on different
taneous angiotensin receptor blockade via valsartan and dates (International Classification of Diseases, 9th Edition/
neprilysin inhibition via sacubitril, significantly reduced International Classification of Diseases, 10th Edition 428.2X
all-cause mortality, cardiovascular mortality, and HF hos- or I50.2X in any position on the claim) from July 1, 2015, to
pitalization compared with enalapril in patients with December 31, 2016, a method similar to that used by Curtis

Circ Heart Fail. 2018;11:e004302. DOI: 10.1161/CIRCHEARTFAILURE.117.004302 February 2018 2

 Sangaralingham et al; Adoption of Sacubitril/Valsartan in Heart Failure

et al.11 We defined the index date as the date of first billing resume treatment with an ACEi or ARB. As such, we exam-
code for systolic HF during the study period. ined the proportion of patients that failed to refill sacubitril/
Finally, as we were interested in identifying differences in valsartan that filled an ACEi or ARB in the follow-up period.
characteristics of patients initiated on sacubitril/valsartan with Titration of sacubitril/valsartan in the 180 days after initiation
those who were eligible for initiation, we identified the sub- was determined based on the doses of initial and subsequent
set of patients with HFrEF who were treated with an ACEi fills. Available doses include 24 mg sacubitril/26 mg valsartan,
or angiotensin receptor blocker (ARB) and did not receive 49/51 mg, and 97/103 mg.
sacubitril/valsartan during the study period (HFrEF ACEi/ARB
cohort). We selected those adults with a billing code for sys- Statistical Analysis
tolic HF that filled a prescription for an ACEi or ARB. The index
Baseline characteristics of the sacubitril/valsartan and HFrEF
date was the date of first fill of an ACEi or ARB during the
ACEi/ARB cohorts are reported as frequencies with percent-
study period.
ages for categorical data and means with SDs for continuous
We required all patients to have continuous enrollment
variables. Differences in baseline characteristics in patients in
in a medical and prescription drug plan for at least 6 months the 2 cohorts were compared using χ2 tests for categorical
before their index date. variables and t tests for continuous variables. Logistic regres-
sion was used to assess predictors of use of sacubitril/valsar-
Independent Variables tan versus ACEi/ARB with results presented as odds ratios and
95% confidence intervals.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on February 16, 2018

Independent variables of interest at index included age, sex,

The proportion of beneficiaries with HFrEF (denomina-
race/ethnicity, residence region, health plan coverage, comor-
tor) who were taking sacubitril/valsartan (numerator) in each
bidities, prescription medications, and prescriber specialty.
month of the study period was calculated. Univariate and
Comorbidities were captured by International Classification
multivariate logistic regression were used to assess predictors
of Diseases, 9th Edition and International Classification of
of adherence (proportion of days covered ≥80%) to sacubitril/
Diseases, 10th Edition codes in any position on claims in
valsartan in the 180 days after initiation with data presented
the 6 months before the index date.12 Comorbidity burden
as odds ratios and 95% confidence interval. All analyses were
was assessed using the Charlson comorbidity index.12 For
conducted using SAS 9.4 (SAS Institute Inc, Cary, NC).
each patient in the sacubitril/valsartan and HFrEF ACEi/ARB
cohorts, we assessed prescriber specialty, categorizing as car-
diologist, primary care (family medicine, internal medicine), or
other. Use of other HF medications in the 90 days before index RESULTS
date was determined based on pharmacy claims. Monthly Population Characteristics
OOP and health plan (insurer) costs of sacubitril/valsartan
and other commonly used HF medications were determined The baseline characteristics of patients taking sacubitril/
based on the pharmacy claims. Total number of medications valsartan and those with HFrEF taking an ACEi/ARB in the
(other than ACEi/ARB and sacubitril/valsartan) at index date study period are shown in Table  1. A total of 102 247
was assessed by therapeutic class using pharmacy claims. The patients had HFrEF, of whom 39 598 were taking an ACEi/
total OOP cost of all medications (other than ACEi/ARB and ARB but not sacubitril/valsartan in the study period. A
sacubitril/valsartan) at index date was calculated. For those total of 2244 patients filled a new prescription for sacubi-
filling prescriptions for other than 30 days supply, a 30-day tril/valsartan in the 18 months after FDA approval, which
supply cost was calculated.
was predominantly prescribed by a cardiologist (82%). A
majority of patients were initiated on the 24/26 mg dose
Sacubitril/Valsartan Adherence and Dose  (58.6%), with fewer initially filling 49/51 mg (32%) and
Titration 97/103 mg (9.4%). Many patients (43.9%) initiated on
We obtained pharmacy fills for the 180-day period after ini- sacubitril/valsartan were not taking an ACEi or ARB in the
tiation of sacubitril/valsartan. Pharmacy claims are complete 90 days before initiation. Only a small proportion filled
with a 3-month lag in the OptumLabs Data Warehouse. their first prescription for sacubitril/valsartan within 7 days
Patients who ended coverage within 180 days of first fill were (3%) and 30 days (9%) of discharge after a hospitaliza-
excluded from analysis. Medication adherence/continuation tion, suggesting that the vast majority of patients are
was calculated by the proportion of days covered. The numer- initiated on sacubitril/valsartan in the outpatient setting.
ator was the number of days supply in the first 180 days after Compared with patients taking ACEi/ARB, those taking
initiation, and the denominator was 180. As medications sacubitril/valsartan were younger (mean age, 67.6 versus
are provided to patients in the hospital, days hospitalized 70.3 years), more often male (68.2% versus 58.5%), and
were added to the numerator. Nonadherence was defined
had fewer comorbidities. Patients filling sacubitril/valsar-
as a proportion of days covered <80%.13 We recognize that
patients may choose to stop taking a prescription medication
tan were also more likely to be using other commonly
for several reasons, such as cost and intolerance, and may prescribed HF medications at baseline, but were taking
make the decision in conjunction with their physician or on fewer total medications (Table 1). Adjusting for potential
their own. Although we cannot fully distinguish these reasons confounders, use of other HF medications was associ-
using pharmacy claims, we hypothesized that patients who ated with greater odds of receiving sacubitril/valsartan,
stopped after discussion with their physician would often whereas older age, female sex, chronic obstructive pul-

Circ Heart Fail. 2018;11:e004302. DOI: 10.1161/CIRCHEARTFAILURE.117.004302 February 2018 3

 Sangaralingham et al; Adoption of Sacubitril/Valsartan in Heart Failure

Table 1.  Baseline Characteristics of Patients With Table 1. Continued

HFrEF Taking an ACEi/ARB and Initiators of Sacubitril/
Sacubitril/
Valsartan HFrEF ACEi/ARB Valsartan P
(n=39 598) (n=2244) Value
Sacubitril/
HFrEF ACEi/ARB Valsartan P Medication use (prior 90 days)
(n=39 598) (n=2244) Value
 ACEi/ARB … 1260 (56.1%) …
Age, y 70.3 (12.1) 67.6 (12.0) <0.001
 β-blockers 26 727 (67.5%) 1942 (86.5%) <0.001
Age category, y <0.001
 Aldosterone
6891 (17.4%) 899 (40.1%) <0.001
 18–64 11 438 (28.9%) 806 (35.9%) antagonist
 65–74 12 431 (31.4%) 747 (33.3%)  Digoxin 3845 (9.7%) 371 (16.5%) <0.001
 75+ 15 729 (39.7%) 691 (30.8%)  Loop diuretics 20 629 (52.1%) 1430 (63.7%) <0.001
Female 16 424 (41.5%) 713 (31.8%) <0.001 OOP cost of sacubitril/valsartan or ACEi/ARB (per 30-day supply) <0.001
Race/ethnicity <0.001  Mean (SD) $3.07 ($8.15) $71.16 ($119.36)

 Asian 632 (1.6%) 49 (2.2%)  Median (IQR) $40.27

$2.00 ($0.65–$2.66)
($3.60–$60.00)
 Black 6855 (17.3%) 388 (17.3%)
Other medication use* <0.001
Downloaded from http://circheartfailure.ahajournals.org/ by guest on February 16, 2018

 Hispanic 2707 (6.8%) 202 (9.0%)

 Mean (SD)
 Unknown 7475 (18.9%) 331 (14.8%) therapeutic 15.6 (10.3) 13.5 (8.4)
 White 21 929 (55.4%) 1274 (56.8%) classes

Census region <0.001  Median (IQR)

therapeutic 13 (9–20) 12 (8–17)
 Missing 40 (0.1%) 1 (<0.1%) classes
 Midwest 11 534 (29.2%) 475 (21.2%) Other medication OOP* (per 30-day supply)
 Northeast 5874 (14.8%) 344 (15.3%)  Mean (SD) $112.91 ($293.5) $80.17 ($151.7) <0.001
 South 18 805 (47.5%) 1290 (57.5%)  Median (IQR) $44.96 $39.04
($12.80–$120.54) ($8.56–$98.92)
 West 3345 (8.5%) 134 (6.0%)

Charlson Comorbidity Index <0.001 ACEi/ARB indicates angiotensin-converting enzyme inhibitor/angiotensin

receptor blocker; HFrEF, heart failure with reduced ejection fraction; IQR,
 Mean (SD) 4.0 (2.4) 3.6 (2.3) interquartile range; and OOP, out-of-pocket.
 Median (IQR) 3.0 (2.0–5.0) 3.0 (2.0–5.0) *At time of initiation of sacubitril/valsartan or date of first fill of ACEi/ARB
in the study period.
Chronic conditions (6 mo baseline)

 Myocardial monary disease, cerebrovascular disease, dementia, and

8473 (21.4%) 442 (19.7%) 0.06
infarction
use of a greater total number of prescription medications
 Diabetes mellitus 18 482 (46.7%) 1009 (45.0%) 0.11
were associated with lower odds of receiving sacubitril/
 Chronic renal
10 957 (27.7%) 553 (24.6%) 0.002 valsartan. Differences by region and race/ethnicity were
disease
also observed (Figure 1)
 Chronic
pulmonary 14 769 (37.3%) 613 (27.3%) <0.001
disease
Monthly Trends in Initiation of Sacubitril/
 Cerebrovascular
disease
6987 (17.6%) 268 (11.9%) <0.001 Valsartan
 Dementia 2925 (7.4%) 42 (1.9%) <0.001 The number of patients initiating sacubitril/valsartan
 Hypertension 33 554 (84.7%) 1842 (82.1%) <0.001 steadily increased after its market debut in July 2015
(Figure 2A). However, the overall proportion of patients
 Atrial fibrillation 15 654 (39.5%) 854 (38.1%) 0.164
with HFrEF taking sacubitril/valsartan remained low
Health plan coverage <0.001
(maximum 2.3%) throughout the study period (Fig-
 Privately insured 7154 (18.1%) 541 (24.1%) ure 2B).
 Medicare
32 444 (81.9%) 1703 (75.9%)
Advantage

Prescribing provider Monthly Costs

 Cardiologist 10 933 (36.0%) 1480 (82.0%) <0.001 The average monthly costs incurred by patients and
 Primary Care 15 998 (52.7%) 174 (9.6%) insurers for sacubitril/valsartan and other common
 Other 3422 (11.3%) 151 (8.4%)
HF medications are shown in Figure 3. Both OOP and
insurer costs for sacubitril/valsartan were substantially
(Continued )
higher than costs for other guideline-directed HFrEF
medications including lisinopril, losartan, carvedilol,

Circ Heart Fail. 2018;11:e004302. DOI: 10.1161/CIRCHEARTFAILURE.117.004302 February 2018 4

 Sangaralingham et al; Adoption of Sacubitril/Valsartan in Heart Failure

nonadherence observed in Black patients, those with

chronic obstructive pulmonary disease, and patients
residing in the South. Better adherence was observed
in patients previously on an ACEi/ARB, those taking a
higher number of total medications, and those initiated
on the 97/103 mg dose. We found no association of
OOP costs of sacubitril/valsartan with adherence/con-
tinuation (odds ratio for adherence per $10 increase,
0.999; 95% confidence interval, 0.992–1.006; P=0.76).

Sacubitril/Valsartan Dose Titration

Of the 1161 patients who continued sacubitril/valsartan
for 180 days after initiation, 355 (30.6%) had a dose
titration (Figure 4). A majority of patients were started
on the 24/26 mg dose, 26% of whom titrated up in
Downloaded from http://circheartfailure.ahajournals.org/ by guest on February 16, 2018

the 180 days after initiation. A similar proportion of

those initiated on 49/51 mg were titrated up to 97/103
mg after initiation (29%), whereas a small proportion
was titrated down to 24/26 mg (5%). A small number
of patients were initiated on 97/103 mg, very few of
Figure 1. Predictors of use of sacubitril/valsartan.
whom titrated down during follow-up (3%). In total,
The adjusted odds ratios (95% CI) of receiving sacubitril/
valsartan rather than angiotensin-converting enzyme inhibi- 285 (24.5%) patients were taking 97/103 mg of sacu-
tor (ACEi)/angiotensin receptor blocker (ARB) associated bitril/valsartan by the end of the study period.
with selected baseline patient characteristics are shown. All
variables shown were included in the model. COPD indicates
chronic obstructive pulmonary disease; MI, myocardial infarc- DISCUSSION
tion; and ref, referent group.
There are several important findings from this study.
and spironolactone. A majority of prescription costs In this large population of individuals with HF enrolled
for sacubitril/valsartan were covered by the insurer in US private and Medicare Advantage health plans,
(mean, $328.37; median, $362.44; interquartile range, adoption of sacubitril/valsartan was slow in the first
$315.41–$396.06). Monthly OOP costs varied widely 18 months after FDA approval. There were system-
(mean, $71.16; median, $40.27; interquartile range, atic differences in the characteristics of individuals
$3.60–$60.00). Costs were similar in beneficiaries initiated on sacubitril/valsartan compared with oth-
enrolled in commercial and Medicare Advantage health ers with HFrEF, in that they were younger, more often
plans early after FDA approval; in more recent months, male, and had fewer comorbidities. In addition, the
OOP costs were lower in patients with commercial observed OOP costs for sacubitril/valsartan were much
plans. In contrast, median OOP costs were $2 to $3 for higher than other guideline-directed medical therapies
lisinopril, losartan, carvedilol, and spironolactone. for HFrEF and may represent an important barrier to
initiation and continuation. Finally, many patients ini-
tiated on sacubitril/valsartan discontinued use in the
Sacubitril/Valsartan Adherence/ first 180 days.
Sacubitril/valsartan is a new drug class that simul-
Continuation taneously blocks the renin–angiotensin–aldosterone
Overall, 1161 of the 1964 (59.1%) patients were system and inhibits neprilysin, the enzyme responsible
adherent to sacubitril/valsartan in the 180 days after ini- for the degradation of the natriuretic peptides and oth-
tiation. Of those who were nonadherent (n=803), 393 er peptides.14,15 Despite a class I recommendation for
(48.9%) did not refill a prescription for sacubitril/val- use of sacubitril/valsartan in the 2016 HF guidelines,7
sartan for >30 days after their supply ended, suggest- the adoption after FDA approval in this HFrEF popula-
ing that they discontinued sacubitril/valsartan. Of those tion was slow. While use slowly increased over time,
who were nonadherent/discontinued sacubitril/valsar- 18 months after FDA approval only 2.3% of patients
tan (n=803), 160 (20%) switched to an ACEi/ARB. The with HFrEF were taking sacubitril/valsartan. These find-
univariate and multivariate predictors of 180-day medi- ings can be compared with a recent report from the
cation adherence/continuation are shown in Table  2. Get With The Guidelines (GWTG) HF registry. Among
Adjusting for potential confounders, higher rates of patients hospitalized with HFrEF at GWTG-participating

Circ Heart Fail. 2018;11:e004302. DOI: 10.1161/CIRCHEARTFAILURE.117.004302 February 2018 5

 Sangaralingham et al; Adoption of Sacubitril/Valsartan in Heart Failure
Downloaded from http://circheartfailure.ahajournals.org/ by guest on February 16, 2018

Figure 2. Use of sacubitril/valsartan in the first 18 months after Food and Drug Administration (FDA) approval.
A, The number of patients filling a first-time prescription for sacubitril/valsartan by month in the first 18 months after FDA
approval. The release of the American College of Cardiology/American Heart Association Heart Failure guideline update on
May 20, 2016, is depicted by the arrow. B, The proportion of patients with heart failure with reduced ejection fraction (HFrEF)
taking sacubitril/valsartan by month after approval.

hospitals from July 2015 to June 2016, the authors the vast majority of patients are initiated on sacubitril/
examined how often sacubitril/valsartan was listed as a valsartan in the outpatient setting, as only 9% filled a
discharge medication. Similar to our report, they found new prescription within 30 days of hospital discharge.
that sacubitril/valsartan was documented in only 2% of As such, the GWTG analysis may have been detecting
patients and that older individuals and those with more prevalent users. Alternatively, patients who were pre-
comorbid conditions were less likely to be taking sacu- scribed sacubitril/valsartan at discharge may have never
bitril/valsartan.16 However, unlike the GWTG report, filled the prescription because of cost or other factors,
our study captured actual prescription fills by patients, underscoring the importance of inclusion of pharmacy
rather than documentation of sacubitril/valsartan on data in our analysis. The access to pharmacy data also
a discharge medication list. Our findings suggest that enabled us to provide novel information about adher-

Circ Heart Fail. 2018;11:e004302. DOI: 10.1161/CIRCHEARTFAILURE.117.004302 February 2018 6

 Sangaralingham et al; Adoption of Sacubitril/Valsartan in Heart Failure
Downloaded from http://circheartfailure.ahajournals.org/ by guest on February 16, 2018

Figure 3. Monthly costs of heart failure medications.

The mean out-of-pocket costs (A) and insurer-paid costs (B) for sacubitril/valsartan and other common medications used to
treat patients with heart failure with reduced ejection fraction are shown.

ence, continuation, and cost of sacubitril/valsartan, been FDA approved since 2011. These medications offer
which was not possible in GWTG. shorter treatment periods and are more effective and
It is important to contextualize the early adoption of tolerable than older regimens (interferon alfa and riba-
sacubitril/valsartan compared with other pharmacologi- virin), but are also more expensive. In one claims-based
cal therapies for patients with a chronic condition. After study, although only 10% of patients with hepatitis C
publication of the results of the RALES trial (Randomized received any treatment, by the end of 2014, essentially
Aldactone Evaluation Study) of spironolactone versus all patients treated were receiving the newer medica-
placebo in HFrEF in 1999,17 the rate of spironolactone tions.20 Thus, compared with other selected examples of
use in Medicare beneficiaries with HFrEF tripled over the novel treatments in chronic serious illness, adoption of
next 18 months.18 However, spironolactone had already sacubitril/valsartan has been slow.
been in use in clinical practice for other indications such Cost is an important potential barrier to use of sacu-
as hypertension, and as such, many physicians may have bitril/valsartan. Both the insurer-paid and OOP costs for
been more willing to adopt the medication for use in sacubitril/valsartan were several orders of magnitude
HFrEF when compared with a newly developed medica- higher than other guideline-based HFrEF medications.
tion like sacubitril/valsartan. By comparison, after FDA These high costs can contribute to multi-level barriers
approval of the first novel oral anticoagulant, dabi- to initiation and continuation. Insurers often impose
gatran, for use in nonvalvular atrial fibrillation, it was strict coverage criteria, prior authorizations, and tiered
prescribed to ≈10% of patients with nonvalvular atrial OOP coverage to deter unnecessary use. UnitedHealth-
fibrillation, constituting ≈20% of all oral anticoagulant care, the largest insurer represented in this data source,
prescriptions, over the next 18 months.19 However, has instituted both prior authorization and the highest
dabigatran offered a large convenience advantage to tier prescription coverage (tier 3) for sacubitril/valsartan,
patients over warfarin, as it did not require international such that patients are required to pay a higher cost to
normalized ratio monitoring and may have contributed fill. While intended to promote appropriate drug use,
to its strong early adoption. Finally, many novel medica- time-consuming prior authorizations can be a deterrent
tions to treat the chronic serious illness hepatitis C have to clinician prescribing.21 Finally, higher OOP costs not

Circ Heart Fail. 2018;11:e004302. DOI: 10.1161/CIRCHEARTFAILURE.117.004302 February 2018 7

 Sangaralingham et al; Adoption of Sacubitril/Valsartan in Heart Failure

Table 2.  Factors Association with Sacubitril/Valsartan if effective at improving outcomes. Insurer coverage of
Adherence sacubitril/valsartan may change after results of addi-
Univariate OR Multivariable OR tional clinical trials that are underway, and the manu-
Factors (95% CI) (95% CI) facturer (Novartis) now offers financial assistance such
Age, y as a free 30-day trial and reduced copay cards to help
 18–64 ref ref individuals afford the OOP costs.
 65–74 1.253 (1.006–1.561)* 1.096 (0.960–1.252)
Some would argue that the clear demonstration of
efficacy of sacubitril/valsartan in the PARADIGM-HF trial
 75+ 1.118 (0.896–1.395) 1.018 (0.887–1.168)
is sufficient to warrant widespread adoption in patients
Female 0.878 (0.724–1.065) 0.956 (0.863–1.059)
with chronic symptomatic HFrEF.24 However, in an era
Race/ethnicity where HF care already accounts for >18 billion dollars
 White ref ref annually in the United States,25 considering cost in addi-
 Asian 0.48 (0.375–0.614)† 1.301 (0.759–2.230) tion to efficacy is important in determining value of a
 Black 0.674 (0.49–0.928)* 0.694 (0.545–0.885)†
new treatment. In several analyses using data from PAR-
ADIGM-HF,26–29 the incremental cost-effectiveness ratio
 Hispanic 0.917 (0.478–1.759) 0.843 (0.628–1.131)
ranged from $45 017 to $50 959 per quality-adjusted life
 Unknown 0.662 (0.508–0.862)† 0.927 (0.722–1.189)
year over a lifetime, a level relatively consistent with other
Downloaded from http://circheartfailure.ahajournals.org/ by guest on February 16, 2018

Census region accepted invasive cardiovascular interventions.30 By com-

 Midwest ref ref parison, the cost-effectiveness of other HFrEF pharmaco-
 Northeast 0.848 (0.627–1.147) 0.890 (0.714–1.109) logical therapies, such as ACEi, β-blockers, and aldoste-
 South 0.785 (0.624–0.988) * 0.834 (0.707–0.984)* rone antagonists, is much more favorable (<$10 000 per
 West 1.275 (0.811–2.004) 1.337 (0.960–1.863)
quality-adjusted life year).31 However, higher costs are
expected with newly released medications, and medica-
Charlson Comorbidity
Index
0.985 (0.948–1.025) tions with higher associated costs may have value if there
is a significant gain in quality-adjusted life year. The exist-
Chronic conditions
ing cost-effectiveness analyses for sacubitril/valsartan are
 Myocardial infarction 0.966 (0.769–1.214) 0.965 (0.854–1.089)
limited to data obtained from a single clinical trial, and
 Diabetes mellitus 1.031 (0.861–1.236) 0.927 (0.837–1.028) real-world outcomes data are needed to determine true
 Chronic renal disease 0.935 (0.759–1.151) 0.954 (0.849–1.071) effectiveness. In our study, patients taking sacubitril/val-
 Chronic pulmonary
0.811 (0.664–0.991)* 0.849 (0.761–0.947)†
sartan were older (mean, 68 versus 64 years), more often
disease female (32% versus 21%), Black (17% versus 5%), and
 Cerebrovascular disease 1.009 (0.763–1.334) 1.009 (0.869–1.172) had higher prevalence of comorbidities such as diabe-
 Hypertension 1.039 (0.822–1.313) 1.084 (0.951–1.235) tes mellitus (45% versus 35%) than patients enrolled in
 Atrial fibrillation 1.034 (0.859–1.244) 0.935 (0.846–1.034) the PARADIGM-HF trial.5 The effectiveness and impact
Prior ACEi/ARB use 1.402 (1.17–1.681)† 1.146 (1.042–1.262)†
on outcomes of sacubitril/valsartan in patients who were
not well represented in PARADIGM-HF may differ.
Total number other
medications
1.047 (1.034–1.06)† 1.056 (1.042–1.070) † Even if we decide that sacubitril/valsartan is a high-
value treatment in HFrEF, patients may still struggle to
Initial sacubitril/valsartan dose
afford to continue to fill and take the medication as
 24–26 mg ref ref
prescribed. We found that over one third of patients
 49–51 mg 1.005 (0.825–1.224) 0.880 (0.751–1.031) initiated on sacubitril/valsartan were nonadherent
 97–103 mg 1.369 (0.988–1.897) 1.320 (1.055–1.652)* over the next 180 days. The refill patterns suggest
Factors associated with adherence/continuation (defined as proportion of that almost half of nonadherent patients discontinued
days covered ≥80%) in the 180 days after initiating sacubitril/valsartan are sacubitril/valsartan completely. Whether this occurred
shown. An OR >1 indicated a greater likelihood of a patient being adherent
because of cost, hemodynamic, renal, or other intol-
during the first 180 days. ACEi/ARB indicates angiotensin-converting enzyme
inhibitor/angiotensin receptor blocker; CI, confidence interval; and OR, odds erance would be of interest to investigate in future
ratio. studies. Our analyses revealed 2 factors that may have
*P<0.05.
contributed. Sacubitril/valsartan was tested in the
†P<0.01.
PARADIGM-HF trial in patients with chronic symptom-
atic HFrEF who were able to tolerate an ACEi or ARB.5
only deter patient initiation of a medication but can also However, we found that many patients initiated on
negatively impact adherence and outcomes.22,23 Most sacubitril/valsartan had not filled a prescription for an
patients with HF are taking numerous daily medications ACEi/ARB in the 3 months prior. Patients who were not
and, in the absence of unlimited financial resources, are previously on an ACEi or ARB were more likely to be
often faced with important decisions about what they nonadherent/discontinue therapy. In the PARADIGM-HF
can afford to allocate toward a single medication, even trial, almost 20% of patients had to discontinue the

Circ Heart Fail. 2018;11:e004302. DOI: 10.1161/CIRCHEARTFAILURE.117.004302 February 2018 8

 Sangaralingham et al; Adoption of Sacubitril/Valsartan in Heart Failure

Figure 4. Titration of
sacubitril/valsartan in the
first 180 days after initia-
tion.
Among those who were
adherent (proportion of days
covered ≥80%) to sacubitril/
valsartan, the proportion that
changed dose during the 180
days after initiation by starting
dose are shown.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on February 16, 2018

study during either the enalapril or sacubitril/valsar- and is specific for a reduced EF, it not very sensitive.10
tan run-in phase, primarily because of adverse events This could result in an underestimation of the number
or abnormal laboratory values.5 Our findings suggest of patients with HFrEF on an ACEi/ARB. However, many
that patients who are not taking an ACEi/ARB before patients may be taking an ACEi/ARB in the presence of a
starting sacubitril/valsartan may be less likely to tolerate reduced EF, but have no HF symptoms, and would there-
it, leading to discontinuation. It is possible that other fore not be eligible to switch to sacubitril/valsartan. As
patients were prescribed sacubitril/valsartan outside of such, the true denominator of patients eligible for sacu-
the recommended route or indication (eg, New York bitril/valsartan is likely to be smaller than the number of
Heart Association functional class I, administered with- individuals with HFrEF taking an ACEi/ARB in this study.
out 36-hour ACEi washout period), and this may have
also adversely impacted real-world tolerability. Further-
more, we found that Black patients were more often Conclusions
nonadherent. Although the reasons for nonadherence/ Despite FDA approval and a strong recommendation
discontinuation are unknown, Blacks may have a strong for use in the HF guidelines, the adoption of sacubi-
vasoactive response, possibly because of the preven- tril/valsartan in clinical practice has been slow. High
tion of natriuretic peptide breakdown via neprilysin insurer-paid and patient OOP costs represent an
inhibition as they are relatively deficient in natriuretic important potential barrier to use and continuation.
peptides.32,33 Further work is needed to understand tol- Additional real-world data are needed to determine
erability of sacubitril/valsartan in Black individuals. As the effectiveness and impact on outcomes of sacu-
previously mentioned, enrollment of Black participants bitril/valsartan in patients who were not well repre-
in PARADIGM-HF was low (5% of total participants),5 sented in PARADIGM-HF.
compared with 17% of those initiating sacubitril/valsar-
tan in our study, underscoring the importance of exami-
nation of real-world data such as these to determine ACKNOWLEDGMENTS
effectiveness of new therapies. All authors made substantial contribution to the conception or
design of the work. L.R. Sangaralingham analyzed the data.
L.R. Sangaralingham and Dr Dunlay drafted the article, and
Limitations Drs Sangaralingham, Shah, and Yao critically revised it for con-
There are important limitations to acknowledge to aid tent. All authors gave final approval of the version submitted.
in interpretation of these findings. First, the data source
includes patients enrolled in private and Medicare Advan-
tage health plans with pharmacy benefits, and findings SOURCES OF FUNDING
in other populations may differ. Second, we used billing This analysis was funded by the Mayo Robert D. and Patricia
codes to identify patients with HFrEF. Though a billing E. Kern Center for the Science of Healthcare Delivery. Dr Dun-
code for systolic HF has a high positive predictive value lay's time is funded by the National Institutes of Health (NIH;

Circ Heart Fail. 2018;11:e004302. DOI: 10.1161/CIRCHEARTFAILURE.117.004302 February 2018 9

 Sangaralingham et al; Adoption of Sacubitril/Valsartan in Heart Failure

K23 HL116643), and Dr Sangaralingham is supported by the 8. Sangaralingham LR, Shah ND, Yao X, Roger VL, Dunlay SM. Incidence
NIH (R01 HL132854). and early outcomes of heart failure in commercially insured and Medi-
care advantage patients, 2006 to 2014. Circ Cardiovasc Qual Outcomes.
2016;9:332–337. doi: 10.1161/CIRCOUTCOMES.116.002653.
9. Wallace PJ, Shah ND, Dennen T, Bleicher PA, Bleicher PD, Crown WH. Op-
DISCLOSURES tum Labs: building a novel node in the learning health care system. Health
Aff (Millwood). 2014;33:1187–1194. doi: 10.1377/hlthaff.2014.0038.
None. 10. Li Q, Glynn RJ, Dreyer NA, Liu J, Mogun H, Setoguchi S. Validity of claims-
based definitions of left ventricular systolic dysfunction in Medicare patients.
Pharmacoepidemiol Drug Saf. 2011;20:700–708. doi: 10.1002/pds.2146.
11. Curtis LH, Whellan DJ, Hammill BG, Hernandez AF, Anstrom KJ, Shea
AFFILIATIONS AM, Schulman KA. Incidence and prevalence of heart failure in el-
derly persons, 1994-2003. Arch Intern Med. 2008;168:418–424. doi:
From the Robert D. and Patricia E. Kern Center for the Science 10.1001/archinternmed.2007.80.
of Healthcare Delivery (L.R.S., N.D.S., X.Y.), Department of 12. Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, Saun-
Cardiovascular Medicine (S.J.S., S.M.D.), and Department of ders LD, Beck CA, Feasby TE, Ghali WA. Coding algorithms for defining
Health Sciences Research (N.D.S., X.Y., S.M.D.), Mayo Clinic, comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care.
2005;43:1130–1139.
Rochester, MN; and OptumLabs, Cambridge, MA (N.D.S.).
13. Hess LM, Raebel MA, Conner DA, Malone DC. Measurement of adher-
ence in pharmacy administrative databases: a proposal for standard defi-
nitions and preferred measures. Ann Pharmacother. 2006;40:1280–1288.
FOOTNOTES doi: 10.1345/aph.1H018.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on February 16, 2018

14. Hubers SA, Brown NJ. Combined angiotensin receptor antagonism and
Received June 7, 2017; accepted December 22, 2017. neprilysin inhibition. Circulation. 2016;133:1115–1124. doi: 10.1161/
Circ Heart Fail is available at http://circheartfailure. CIRCULATIONAHA.115.018622.
15. von Lueder TG, Sangaralingham SJ, Wang BH, Kompa AR, Atar D,
ahajournals.org.
Burnett JC Jr, Krum H. Renin-angiotensin blockade combined with
natriuretic peptide system augmentation: novel therapeutic con-
cepts to combat heart failure. Circ Heart Fail. 2013;6:594–605. doi:
REFERENCES 10.1161/CIRCHEARTFAILURE.112.000289.
16. Luo N, Fonarow GC, Lippmann SJ, Mi X, Heidenreich PA, Yancy CW,
1. Benjamin EJ, Blaha MJ, Chiuve SE, Cushman M, Das SR, Deo R, de Fer- Greiner MA, Hammill BG, Hardy NC, Turner SJ, Laskey WK, Curtis LH,
ranti SD, Floyd J, Fornage M, Gillespie C, Isasi CR, Jiménez MC, Jordan Hernandez AF, Mentz RJ, O’Brien EC. Early adoption of sacubitril/valsar-
LC, Judd SE, Lackland D, Lichtman JH, Lisabeth L, Liu S, Longenecker CT, tan for patients with heart failure with reduced ejection fraction: insights
Mackey RH, Matsushita K, Mozaffarian D, Mussolino ME, Nasir K, Neu- from Get With the Guidelines-Heart Failure (GWTG-HF). JACC Heart Fail.
mar RW, Palaniappan L, Pandey DK, Thiagarajan RR, Reeves MJ, Ritchey 2017;5:305–309. doi: 10.1016/j.jchf.2016.12.018.
M, Rodriguez CJ, Roth GA, Rosamond WD, Sasson C, Towfighi A, Tsao 17. Pitt B, Zannad F, Remme WJ, Cody R, Castaigne A, Perez A, Palensky J, Wittes
CW, Turner MB, Virani SS, Voeks JH, Willey JZ, Wilkins JT, Wu JH, Alger J. The effect of spironolactone on morbidity and mortality in patients with
HM, Wong SS, Muntner P; American Heart Association Statistics Commit- severe heart failure. Randomized Aldactone Evaluation Study Investigators. N
tee and Stroke Statistics Subcommittee. Heart Disease and Stroke Statis- Engl J Med. 1999;341:709–717. doi: 10.1056/NEJM199909023411001.
tics-2017 Update: a report from the American Heart Association. Circula- 18. Juurlink DN, Mamdani MM, Lee DS, Kopp A, Austin PC, Laupacis A,
tion. 2017;135:e146–e603. doi: 10.1161/CIR.0000000000000485. Redelmeier DA. Rates of hyperkalemia after publication of the Random-
2. Dunlay SM, Roger VL. Understanding the epidemic of heart failure: past, ized Aldactone Evaluation Study. N Engl J Med. 2004;351:543–551. doi:
present, and future. Curr Heart Fail Rep. 2014;11:404–415. doi: 10.1007/ 10.1056/NEJMoa040135.
s11897-014-0220-x. 19. Marzec LN, Wang J, Shah ND, Chan PS, Ting HH, Gosch KL, Hsu JC, Mad-
3. Roger VL, Weston SA, Redfield MM, Hellermann-Homan JP, Killian J, dox TM. Influence of direct oral anticoagulants on rates of oral anticoagu-
Yawn BP, Jacobsen SJ. Trends in heart failure incidence and survival in a lation for atrial fibrillation. J Am Coll Cardiol. 2017;69:2475–2484. doi:
community-based population. JAMA. 2004;292:344–350. doi: 10.1001/ 10.1016/j.jacc.2017.03.540.
jama.292.3.344. 20. Yao X, Sangaralingham LR, Ross JS, Shah ND, Talwalkar JA. Adoption of
4. Kaye DM, Krum H. Drug discovery for heart failure: a new era or the end new agents and changes in treatment patterns for hepatitis C: 2010-
of the pipeline? Nat Rev Drug Discov. 2007;6:127–139. doi: 10.1038/ 2014. Am J Manag Care. 2016;22:e224–e232.
nrd2219. 21. Fischer MA, Schneeweiss S, Avorn J, Solomon DH. Medicaid prior-autho-
5. McMurray JJ, Packer M, Desai AS, Gong J, Lefkowitz MP, Rizkala AR, rization programs and the use of cyclooxygenase-2 inhibitors. N Engl J
Rouleau JL, Shi VC, Solomon SD, Swedberg K, Zile MR; PARADIGM-HF Med. 2004;351:2187–2194. doi: 10.1056/NEJMsa042770.
Investigators and Committees. Angiotensin-neprilysin inhibition ver- 22. Eaddy MT, Cook CL, O’Day K, Burch SP, Cantrell CR. How patient cost-
sus enalapril in heart failure. N Engl J Med. 2014;371:993–1004. doi: sharing trends affect adherence and outcomes: a literature review. P T.
10.1056/NEJMoa1409077. 2012;37:45–55.
6. Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JG, Coats AJ, Falk 23. Gourzoulidis G, Kourlaba G, Stafylas P, Giamouzis G, Parissis J, Maniadakis
V, González-Juanatey JR, Harjola VP, Jankowska EA, Jessup M, Linde C, N. Association between copayment, medication adherence and outcomes
Nihoyannopoulos P, Parissis JT, Pieske B, Riley JP, Rosano GM, Ruilope in the management of patients with diabetes and heart failure. Health
LM, Ruschitzka F, Rutten FH, van der Meer P; Authors/Task Force Mem- Policy. 2017;121:363–377. doi: 10.1016/j.healthpol.2017.02.008.
bers. 2016 ESC Guidelines for the diagnosis and treatment of acute and 24. Packer M. Angiotensin neprilysin inhibition for patients with heart failure:
chronic heart failure: The Task Force for the diagnosis and treatment of what if sacubitril/valsartan were a treatment for cancer? JAMA Cardiol.
acute and chronic heart failure of the European Society of Cardiology 2016;1:971–972. doi: 10.1001/jamacardio.2016.3053.
(ESC)Developed with the special contribution of the Heart Failure Asso- 25. International R. Cardiovascular disease: a costly burden for America. pro-
ciation (HFA) of the ESC. Eur Heart J. 2016;37:2129–2200. doi: 10.1093/ jections through 2035. http://www.heart.org/idc/groups/heart-public/@
eurheartj/ehw128. wcm/@adv/documents/downloadable/ucm_491543.pdf. Accessed Janu-
7. Yancy CW, Jessup M, Bozkurt B, Butler J, Casey DE Jr, Colvin MM, Drazner ary 23, 2018.
MH, Filippatos GS, Fonarow GC, Givertz MM, Hollenberg SM, Lindenfeld 26. Gaziano TA, Fonarow GC, Claggett B, Chan WW, Deschaseaux-Voinet C,
J, Masoudi FA, McBride PE, Peterson PN, Stevenson LW, Westlake C. 2017 Turner SJ, Rouleau JL, Zile MR, McMurray JJ, Solomon SD. Cost-effective-
ACC/AHA/HFSA focused update of the 2013 ACCF/AHA guideline for the ness analysis of sacubitril/valsartan vs enalapril in patients with heart fail-
management of heart failure: a report of the American College of Cardiol- ure and reduced ejection fraction. JAMA Cardiol. 2016;1:666–672. doi:
ogy/American Heart Association Task Force on Clinical Practice Guidelines 10.1001/jamacardio.2016.1747.
and the Heart Failure Society of America. Circulation. 2017;136:e137– 27. King JB, Shah RU, Bress AP, Nelson RE, Bellows BK. Cost-effectiveness of
e161. doi: 10.1161/CIR.0000000000000509. sacubitril-valsartan combination therapy compared with enalapril for the

Circ Heart Fail. 2018;11:e004302. DOI: 10.1161/CIRCHEARTFAILURE.117.004302 February 2018 10

 Sangaralingham et al; Adoption of Sacubitril/Valsartan in Heart Failure

treatment of heart failure with reduced ejection fraction. JACC Heart Fail. 31. Banka G, Heidenreich PA, Fonarow GC. Incremental cost-effectiveness of
2016;4:392–402. doi: 10.1016/j.jchf.2016.02.007. guideline-directed medical therapies for heart failure. J Am Coll Cardiol.
28. Ollendorf DA, Sandhu AT, Pearson SD. Sacubitril-valsartan for the treat- 2013;61:1440–1446. doi: 10.1016/j.jacc.2012.12.022.
ment of heart failure: effectiveness and value. JAMA Intern Med. 32. Gupta DK, Claggett B, Wells Q, Cheng S, Li M, Maruthur N, Selvin E,
2016;176:249–250. doi: 10.1001/jamainternmed.2015.7661.
Coresh J, Konety S, Butler KR, Mosley T, Boerwinkle E, Hoogeveen R,
29. Sandhu AT, Ollendorf DA, Chapman RH, Pearson SD, Heidenreich PA.
Ballantyne CM, Solomon SD. Racial differences in circulating natriuretic
Cost-effectiveness of sacubitril-valsartan in patients with heart failure
peptide levels: the atherosclerosis risk in communities study. J Am Heart
with reduced ejection fraction. Ann Intern Med. 2016;165:681–689. doi:
10.7326/M16-0057. Assoc. 2015;4:e001831. doi: 10.1161/JAHA.115.001831.
30. Sanders GD, Hlatky MA, Owens DK. Cost-effectiveness of implantable 33. Gupta DK, de Lemos JA, Ayers CR, Berry JD, Wang TJ. Racial differenc-
cardioverter-defibrillators. N Engl J Med. 2005;353:1471–1480. doi: es in natriuretic peptide levels: the Dallas Heart Study. JACC Heart Fail.
10.1056/NEJMsa051989. 2015;3:513–519. doi: 10.1016/j.jchf.2015.02.008.
Downloaded from http://circheartfailure.ahajournals.org/ by guest on February 16, 2018

Circ Heart Fail. 2018;11:e004302. DOI: 10.1161/CIRCHEARTFAILURE.117.004302 February 2018 11

 Adoption of Sacubitril/Valsartan for the Management of Patients With Heart Failure
Lindsey R. Sangaralingham, S. Jeson Sangaralingham, Nilay D. Shah, Xiaoxi Yao and Shannon
M. Dunlay
Downloaded from http://circheartfailure.ahajournals.org/ by guest on February 16, 2018

Circ Heart Fail. 2018;11:

doi: 10.1161/CIRCHEARTFAILURE.117.004302
Circulation: Heart Failure is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX
75231
Copyright © 2018 American Heart Association, Inc. All rights reserved.
Print ISSN: 1941-3289. Online ISSN: 1941-3297

The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circheartfailure.ahajournals.org/content/11/2/e004302

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published
in Circulation: Heart Failure can be obtained via RightsLink, a service of the Copyright Clearance Center, not
the Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further information
about this process is available in the Permissions and Rights Question and Answer document.

Reprints: Information about reprints can be found online at:

http://www.lww.com/reprints

Subscriptions: Information about subscribing to Circulation: Heart Failure is online at:

http://circheartfailure.ahajournals.org//subscriptions/