Upload
ritasuzyanti
View
214
Download
0
Embed Size (px)
Citation preview
8/16/2019 ARTIKEL 1 Hypertension
1/6
Research Article
Long-term safety of nebivolol and valsartancombination therapy in patients with hypertension:
an open-label, single-arm, multicenter study
Joel M. Neutel, MDa,*, Thomas D. Giles, MDb, Henry Punzi, MDc,d, Robert J. Weiss, MDe,Huiling Li, PhDf , and Amy Finck g
aOrange County Research Center, Tustin, CA, USA;b Department of Medicine, Tulane University, New Orleans, LA, USA;
cTrinity Hypertension and Metabolic Research Institute, Carrollton, TX, USA;d UT Southwestern Medical Center, Carrollton, TX, USA;
e Maine Research Associates, Auburn, ME, USA; f Department of Biostatistics, Forest Research Institute, Jersey City, NJ, USA; and
gTrial Operations, Forest Research Institute, Jersey City, NJ, USA
Manuscript received August 28, 2014 and accepted September 16, 2014
Abstract
Long–term safety of a free–tablet combination of nebivolol and valsartan was assessed in a Phase III, open–label trial
(NCT01415505). Adults with hypertension entered a 4–week placebo run–in phase, followed by a 52–week treatment phase.
Initial dosage (Neb/Val 5/160 mg/d) was titrated up to 20/320 mg/d to achieve blood pressure (BP) goal (JNC7 criteria), with
the addition of hydrochlorothiazide (up to 25 mg/d) if needed. Safety and tolerability parameters included adverse events.
Efficacy assessments included baseline–to–endpoint change in diastolic BP and systolic BP and the percentage of patients
who achieved BP goal. All analyses were performed using descriptive statistics. Study completion rate was 60.4% (489/
810). The most frequent reason for discontinuation was insufficient therapeutic response (8.4%). Adverse events were expe-
rienced by 59.2% of patients, with the most common being headache (5.7%), nasopharyngitis (5.0%), and upper respiratorytract infection (4.6%). Three (0.4%) deaths occurred during the study; none was considered related to study medication.
Mean standard deviation changes from baseline at week 52 (observed cases) were 25.5 15.9 mm Hg (systolic BP)
and 19.0 8.7 mm Hg (diastolic BP). A total of 75.7% nebivolol/valsartan–treated and 57.8% nebivolol/valsartan/hydro-
chlorothiazide–treated completers achieved BP goal. Long–term treatment with nebivolol and valsartan in adults with hyper-
tension was safe and well–tolerated. J Am Soc Hypertens 2014;8(12):915–920. 2014 American Society of Hypertension.
All rights reserved.
Keywords: Angiotensin receptor blocker; beta-blocker; blood pressure; clinical trial.
Introduction
Over two–thirds of individuals with hypertension require
more than one drug to achieve blood pressure (BP) con-
trol.1,2 There is a dearth of data about combining b–
blockers and renin–angiotensin–aldosterone system
(RAAS) inhibitors, but it has been suggested that such com-
binations would be suboptimal, due to a partial overlap inthe mechanism of action.3
This study was funded by Forest Research Institute.
Conflict of Interest: Dr Neutel does not have any significant re-
lationships or affiliations to disclose. Dr Giles has received grants
and personal fees from Forest Research Institute, Inc. Dr Punzi has
received grants/research support from the NIH, Forest Labora-
tories, Inc., Takeda, Boehringer Ingelheim, Daiichi-Sankyo, Astra-
Zeneca, and he is also on the Speakers’ Bureau for Forest
Laboratories, Inc., Daiichi-Sankyo, and AstraZeneca. Dr Weiss
has received grant/research support from Forest Laboratories,
Inc. Wei Chen and Amy Fink are employees of Forest Research
Institute.
*Corresponding author: Joel M. Neutel, MD, 14351 Myford
Rd, Ste B, Tustin, CA 93780. Tel: (714) 550-9990; Fax: (714)
550-1226.
E-mail: [email protected]
Journal of the American Society of Hypertension 8(12) (2014) 915–920
1933-1711/$ - see front matter 2014 American Society of Hypertension. All rights reserved.
http://dx.doi.org/10.1016/j.jash.2014.09.017
mailto:[email protected]://dx.doi.org/10.1016/j.jash.2014.09.017http://dx.doi.org/10.1016/j.jash.2014.09.017http://dx.doi.org/10.1016/j.jash.2014.09.017http://dx.doi.org/10.1016/j.jash.2014.09.017http://dx.doi.org/10.1016/j.jash.2014.09.017http://crossmark.crossref.org/dialog/?doi=10.1016/j.jash.2014.09.017&domain=pdfmailto:[email protected]
8/16/2019 ARTIKEL 1 Hypertension
2/6
Nebivolol is a b1–selective antagonist with nitric oxide–
dependent vasodilatory properties4 and has been approved
for the treatment of hypertension, alone or in combination
with other antihypertensive agents. The b1:b2 receptor af-
finity for nebivolol has been estimated at 321:1 for doses
up to 10 mg/day, but the high b1 selectivity may be lost
at the highest approved, but clinically rarely used, dose of 40 mg/day.5 Compared with atenolol, nebivolol has been
shown to decrease central pulse pressure and the augmenta-
tion index,6 although the two drugs lower brachial BP, pulse
rate, and plasma renin activity to a similar extent.7 Valsar-
tan is an angiotensin II receptor blocker (ARB), and is
already a component of several fixed–dose combina-
tions.8–10 The antihypertensive efficacy of valsartan and
other ARBs is based on reduction of peripheral resistance11
and, similar to other ARBs,9 valsartan has been shown to
reduce the levels of pro–inflammatory cytokines in patients
with hypertension.12 Both nebivolol and valsartan have an
excellent safety and tolerability profile.13–15
Here we report the results of an open–label study that as-
sessed the long–term safety of nebivolol and valsartan,
administered as a free–tablet, flexible–dose combination, in
adults with stage 1 or 2 hypertension (per JNC7 criteria1).
Methods
Ethical Conduct
This study was conducted in compliance with the Interna-
tional Conference on Harmonisation of Technical Require-
ments for Registration of Pharmaceuticals for Human Use(ICH) guidelines and the US Food and Drug Administration
guidelines for good clinical practice; and in accordance with
the ethical principles that originate from the Declaration of
Helsinki and the US Food and Drug Administration Code
of Federal Regulations Title 21, section 312.120. All enrolled
patients provided voluntary, written informed consent and
Health Insurance Portability and Accountability Act (HI-
PAA) authorization prior to participating in any study proce-
dures. The institutional review boards of all participating
centers approved the study protocol, informed consent
form, and information sheet advertisements.
Study Design
This was a Phase III, multicenter, open–label, single–arm
trial (NAC-MD-02; NCT01415505) conducted in the US.
Following a 1–week screening, participants entered a 4–
week single–blind placebo run–in phase, followed by a
52–week treatment period during which they received the
free–tablet combination of nebivolol 5 mg/day and valsartan
160 mg/day. The initial dosage (Week 0) of 5/160 mg/day
nebivolol/valsartan was doubled to 10/320 mg/day at
Week 2, and it was further increased to 20/320 mg/day if pa-
tients did not achieve BP goal (SBP/DBP 140/90 mm Hg
[without diabetes] or 130/80 mm Hg [with diabetes]) after
at least 4 weeks of treatment at the 10/320 mg/d dosage. If
BP goal was not met after a minimum of 10 weeks on the
20/320 mg/d dosage, 12.5 mg/d hydrochlorothiazide
(HCTZ) was added and doubled 4 weeks thereafter to
25 mg/d if BP goal was still not achieved. Patients who
did not achieve BP goal after 14 weeks of adding HCTZ totheir regimen (starting with the 12.5 mg/d dosage) were dis-
continued from the study. After 52 weeks of treatment, all
patients underwent 1 week of down–titrating their nebivolol
dosage (20 mg to 10 mg to 5 mg to placebo, in 3–day incre-
ments, as applicable), before all study medication was
discontinued.
Participants
Men and women ages 18 years or older were eligible to
participate if they had a heart rate of 55 beats per minute
(except for patients already on b-blockers), a normal phys-ical examination at Screening, and stage 1 or 2 hyperten-
sion, with a recent DBP measurement of 90 mm Hg
and
8/16/2019 ARTIKEL 1 Hypertension
3/6
disease or active gastritis, pancreatitis, renal impairment,
major gastrointestinal tract surgery within 6 months from
Screening, or pregnancy or breastfeeding.
Safety Assessments
Safety and tolerability were assessed by recording adverseevents (AEs) and monitoring vital signs at each visit; by per-
forming physical examinations at Screening and Week 52; by
performing electrocardiograms at Screening, Weeks 0 and
52, and after the down–titration phase; by measuring clinical
laboratory parameters (blood and urine) at Screening and
Weeks 0, 6, 22, and 52. Height and weight were measured
at each visit between Weeks 0 and 52.
Efficacy Assessments
Efficacy assessments included change from baseline in
trough–seated SBP and DBP at each visit, percentage of pa-tients who achieved BP goal, percentage of DBP responders
(DBP
8/16/2019 ARTIKEL 1 Hypertension
4/6
were taking and 210/388 (54.1%) were not taking HCTZ,
experienced at least one treatment–emergent AE (TEAE).
The majority of TEAEs were considered mild or moderate
in severity. Common TEAEs (those occurring in 2% of
the Safety population) are listed in Table 2. In addition to
the 19 (2.4%) patients who reported bradycardia
(Table 2), seven (0.9%) reported sinus bradycardia, withoutco–occurrence of those two in the same individual. Two pa-
tients (0.2%) who experienced bradycardia also reported fa-
tigue, but those two AEs did not coincide in either of those
two individuals. Erectile dysfunction was reported by 3/388
(0.8%) patients not taking HCTZ and by 4/419 (1.0%) pa-
tients taking HCTZ.
During the 52–week treatment phase and the 1–week
down-titration phase, 19 (2.4%) patients experienced a
SAE, with only one SAE (bradycardia) deemed related to
study medication. A total of three (0.4%) patients died dur-
ing the study, but all deaths were deemed unrelated to the
study medication.
Clinical Laboratory Parameters
Changes from baseline in major metabolic parameters are
shown in Table 3. MeanSD changes of 1.04.1kginbody
weight and 0.3 1.4 kg/m2 in body mass index were
observed over the course of the study.
A total of 431/784 (55.0%) patients had a potentially clin-
ically significant post–baseline laboratory value, of which
the levels of high–sensitivity C–reactive protein were the
most prevalent: 77/247 (31.2%) had high–sensitivity C–reac-
tive protein levels below 0.8 mg/L and 128/247 (51.8%) hadlevels above 3.0 mg/L. High–density lipoprotein cholesterol
levels 60 mg/dL were recorded in 17.1%
(93/543) and 6.3% (34/543) participants, respectively.
Low–density lipoprotein cholesterol values 159 mg/dL were found in 11.5% (69/599) and 14.0%
(84/599) of participants. Total cholesterol levels that ex-
ceeded the upper limit of normal range by 10% were re-corded in 19.6% (121/616) of patients. A total of 20/664
(3.0%) patients had a post–baseline fasting glucose measure-
ment that represented >50% increase from baseline and was
over 1.2–fold the upper limit of normal; a total of 69/509
(13.6%) participants experienced a baseline–to–endpoint
shift of HbA1c levels from normal to high, and 30/509
(5.9%) experienced a shift from high to normal.
Efficacy
After 52 weeks of open–label treatment, the mean SD
trough–seated SBP and DBP decreased by 25.5 15.9 mm
Hg and 19.0 8.7 mm Hg, respectively. A total of 419
(51.5%) participants received HCTZ at some point in the
study (Table 4), of which 296 completed the trial.
At Week 52, 65.1% (327/502) of participants remaining
in the trial achieved the BP goal, including 75.7% (156/
206) of those who did not need HCTZ addition, 57.8%
(171/296) of those who did, 61.9% (86/139) of blacks
(vs. 66.4% [241/363] of non–blacks), and 61.1% (77/126)
of Hispanics (vs. 66.5% [250/376] of non–Hispanics).
Visit–by–visit changes in BP, the rates of BP goal achieve-
ment, DBP response, SBP response, and the percentage of
patients receiving HCTZ are shown in Figure 2.
Table 2
Common treatment–emergent adverse events (TEAE; 2% of the
total safety population), n (%)
Common TEAEs Nebivolol–Valsartan Combination
Taking
HCTZ
(n ¼ 419)
Not Taking
HCTZ
(n ¼ 388)
Total
(N ¼ 807)
Headache 27 (6.4) 19 (4.9) 46 (5.7)
Nasopharyngitis 20 (4.8) 20 (5.2) 40 (5.0)
Upper respiratory
tract infection
26 (6.2) 11 (2.8) 37 (4.6)
Dizziness 18 (4.3) 17 (4.4) 35 (4.3)
Bronchitis 16 (3.8) 8 (2.1) 24 (3.0)
Cough 14 (3.3) 8 (2.1) 22 (2.7)
Fatigue 9 (2.1) 12 (3.1) 21 (2.6)
Back pain 10 (2.4) 10 (2.6) 20 (2.5)
Urinary tract infection 13 (3.1) 7 (1.8) 20 (2.5)
Bradycardia 7 (1.7) 12 (3.1) 19 (2.4)
Sinusitis 15 (3.6) 4 (1.0) 19 (2.4)
Peripheral edema 7 (1.7) 11 (2.8) 18 (2.2)
Table 3
Changes from baseline to week 52 in glucose and lipid parameters
(safety population)
Parameter Nebivolol–Valsartan
Combination (N ¼ 782)
Baseline Change from
Baseline
Fasting glucose, mg/dL* 101.6 22.2 2.9 26.7
HDL cholesterol, mg/dL* 49.9 13.9 2.3 7.7
LDL cholesterol, mg/dL* 128.0 34.8 9.3 25.9
Total cholesterol, mg/dL* 192.6 36.7 1.2 30.9
HDL, high–density lipoprotein; LDL, low–density lipoprotein.
* Mean standard deviation.
Table 4
Distribution of dosages at endpoint, n (%), safety population
(N ¼ 807)
Nebivolol–Valsartan
Combination (mg/day)
HCTZ Addition (mg/d)
0 12.5 25
5/160 45 (5.6) – –
10/320 162 (20.1) 1 (0.1) 1 (0.1)
20/320 181 (22.4) 204 (25.3) 213 (26.4)
Total 388 (48.1) 205 (25.4) 214 (26.5)
HCTZ, hydrochlorothiazide.
918 J.M. Neutel et al. / Journal of the American Society of Hypertension 8(12) (2014) 915–920
8/16/2019 ARTIKEL 1 Hypertension
5/6
Discussion
Our data suggest that nebivolol–valsartan combination is
safe and well–tolerated for long–term use (including a
neutral metabolic effect), which is consistent with the favor-able safety and tolerability profile of the individual compo-
nents.15,16 The observed reductions in SBP and DBP are
comparable with those attained in similarly designed, 52–
week trials of valsartan combined with agents from other
classes,17,18 and consistent with the results of shorter, pla-
cebo–controlled trials in which nebivolol was added to an
ongoing treatment with an ARB or an angiotensin–convert-
ing enzyme inhibitor (ACEI).19,20 The subset of participants
in our trial who discontinued treatment due to insufficient
response had a higher proportion of blacks and individuals
with diabetes compared with the entire trial population
(blacks, 42.6% vs. 30.4%; diabetes, 42.6% vs. 17.0%), whichis in agreement with the reported difficulties of attaining BP
control in those patients.1 However, the rates of BP control
between blacks and non–blacks who completed the trial
were similar (61.9% vs. 66.4%, respectively).
The relatively low occurrence of b–blocker-typical AEs
(bradycardia, erectile dysfunction, fatigue, weight
gain)21,22 in our study is consistent with the results of shorter
head–to–head trials, in which nebivolol had lower rates of
those AEs than other b1–selective blockers (atenolol, meto-
prolol).23 In addition, the favorable safety and tolerability
profile of nebivolol15 is possibly reflected in 1.5– to 8–fold
lower rates of some common AEs (peripheral edema,
nasopharyngitis, and dizziness) reported in a similarly de-
signed, 52–week valsartan–amlodipine trial.18
Limitations of this study are inherent to the open–label
design (lack of randomization, blinding, and control groups)
but its duration and treatment scheme (titration to response,
addition of a third drug as needed) mirror clinical practice
and support a notion that the results could be generalized to
the out–of–trial population. In addition, screening was not
performed by means of ambulatory blood pressure moni-
toring, which suggests that possibly up to one–third of partic-
ipants may have had white–coat hypertension.24
In conclusion, a 52–week treatment with a free combina-
tion consisting of an ARB (valsartan) and a vasodilatory b–
blocker (nebivolol), with the addition of HCTZ as needed,
was well–tolerated. The combination resulted in significant
decreases in both systolic and diastolic BP, which were sus-
tained over the 52–week period. The majority of patients
were able to achieve recommended BP goals.
Acknowledgments
Writing assistance and editorial support for the prepara-
tion of this manuscript was provided by Prescott Medical
Communications Group, Chicago, Illinois.
References
1. Chobanian AV, Bakris GL, Black HR, Cushman WC,
Green LA, Izzo JL Jr, et al. Seventh report of the Joint
Figure 2. Blood pressure (BP) changes and percentages of responders, BP goal achievers, and recipients of additional medication
(HCTZ), by visit (intent–to–treat [ITT] population, observed cases). Diastolic blood pressure (DBP) and systolic blood pressure
(SBP) responders were those patients who achieved DBP
8/16/2019 ARTIKEL 1 Hypertension
6/6
National Committee on Prevention, Detection, Evalua-
tion, and Treatment of High Blood Pressure. Hyperten-
sion 2003;42:1206–52.
2. Gu Q, Burt VL, Dillon CF, Yoon S. Trends in antihyper-
tensive medication use and blood pressure control
among United States adults with hypertension: the
National Health and Nutrition Examination Survey,2001 to 2010. Circulation 2012;126:2105–14.
3. Gradman AH, Basile JN, Carter BL, Bakris GL,
Materson BJ, Black HR, et al. Combination therapy
in hypertension. J Am Soc Hypertens 2010;4:90–8.
4. Vanhoutte PM, Gao Y. Beta blockers, nitric oxide, and
cardiovascular disease. Curr Opin Pharmacol 2013;13:
265–73.
5. M€unzel T, Gori T. Nebivolol: the somewhat-different
beta-adrenergic receptor blocker. J Am Coll Cardiol
2009;54:1491–9.
6. Koumaras C, Tziomalos K, Stavrinou E, Katsiki N,
Athyros VG, Mikhailidis DP, et al. Effects of renin-angiotensin-aldosterone system inhibitors and
beta-blockers on markers of arterial stiffness. J Am
Soc Hypertens 2014;8:74–82.
7. Simon G, Johnson ML. Comparison of antihypertensive
and beta 1-adrenoceptor antagonist effect of nebivolol
and atenolol in essential hypertension. Clin Exp Hyper-
tens 1993;15:501–9.
8. Bains J, Smith WB. Valsartan plus hydrochlorothiazide:
a review of its use since its introduction. Expert Opin
Pharmacother 2011;12:1975–84.
9. Miura S, Saku K. Efficacy and safety of angiotensin II
type 1 receptor blocker/calcium channel blocker combi-nation therapy for hypertension: focus on a single-pill
fixed-dose combination of valsartan and amlodipine.
J Int Med Res 2012;40:1–9.
10. Neutel JM, Smith DH. Hypertension management:
rationale for triple therapy based on mechanisms of ac-
tion. Cardiovasc Ther 2013;31:251–8.
11. Israili ZH. Clinical pharmacokinetics of angiotensin II
(AT1) receptor blockers in hypertension. J Hum Hyper-
tens 2000;14(Suppl 1):S73–86.
12. Manabe S, Okura T, Watanabe S, Fukuoka T, Higaki J.
Effects of angiotensin II receptor blockade with valsar-
tan on pro-inflammatory cytokines in patients withessential hypertension. J Cardiovasc Pharmacol 2005;
46:735–9.
13. Neutel JM, Bedigian MP. Efficacy of valsartan in pa-
tients aged > or ¼65 years with systolic hypertension.
Clin Ther 2000;22:961–9.
14. Pool JL, Glazer R, Chiang YT, Gatlin M. Dose-
response efficacy of valsartan, a new angiotensin II re-
ceptor blocker. J Hum Hypertens 1999;13:275–81.
15. Weiss RJ, Saunders E, Greathouse M. Efficacy andtolerability of nebivolol in stage I-II hypertension: a
pooled analysis of data from three randomized,
placebo-controlled monotherapy trials. Clin Ther
2011;33:1150–61.
16. Black HR, Bailey J, Zappe D, Samuel R. Valsartan: more
than a decade of experience. Drugs 2009;69:2393–414.
17. Chrysant SG, Murray AV, Hoppe UC, Dattani D,
Patel S, Hsu H, et al. Long-term safety, tolerability
and efficacy of aliskiren in combination with valsartan
in patients with hypertension: a 6-month interim anal-
ysis. Curr Med Res Opin 2008;24:1039–47.
18. Smith TR, Glazer RD, Koren MJ, Wernsing M, Zhang Y.Combination therapy with amlodipine/valsartan in
essential hypertension: a 52-week, randomised, open-
label, extension study. Int J Clin Pract 2010;64:1367–74.
19. Deedwania P, Shea J, Chen W, Brener L. Effects of
add-on nebivolol on blood pressure and glucose param-
eters in hypertensive patients with prediabetes. J Clin
Hypertens (Greenwich) 2013;15:270–8.
20. Weiss RJ, Stapff M, Lin Y. Placebo effect and efficacy
of nebivolol in patients with hypertension not
controlled with lisinopril or losartan: a phase IV, ran-
domized, placebo-controlled trial. Am J Cardiovasc
Drugs 2013;13:129–40.21. Black HR, Greenberg BH, Weber MA. The foundation
role of beta blockers across the cardiovascular disease
spectrum: a year 2009 update. Am J Med 2010;123:S2.
22. Kountz DS. Are tolerability concerns a class effect of
beta-blockers in treating patients with hypertension?
Postgrad Med 2009;121:14–24.
23. Ambrosioni E, Borghi C. Tolerability of nebivolol in
head-to-head clinical trials versus other cardioselective
B-blockers in the treatment of hypertension. High
Blood Press Cardiovasc Prev 2005;12:27–35.
24. Verberk WJ, Kroon AA, Thien T, Lenders JW, van
Montfrans GA, Smit AJ, et al. Prevalence of thewhite-coat effect at multiple visits before and during
treatment. J Hypertens 2006;24:2357–63.
920 J.M. Neutel et al. / Journal of the American Society of Hypertension 8(12) (2014) 915–920
http://refhub.elsevier.com/S1933-1711(14)00806-7/sref1http://refhub.elsevier.com/S1933-1711(14)00806-7/sref1http://refhub.elsevier.com/S1933-1711(14)00806-7/sref1http://refhub.elsevier.com/S1933-1711(14)00806-7/sref2http://refhub.elsevier.com/S1933-1711(14)00806-7/sref2http://refhub.elsevier.com/S1933-1711(14)00806-7/sref2http://refhub.elsevier.com/S1933-1711(14)00806-7/sref2http://refhub.elsevier.com/S1933-1711(14)00806-7/sref2http://refhub.elsevier.com/S1933-1711(14)00806-7/sref3http://refhub.elsevier.com/S1933-1711(14)00806-7/sref3http://refhub.elsevier.com/S1933-1711(14)00806-7/sref3http://refhub.elsevier.com/S1933-1711(14)00806-7/sref4http://refhub.elsevier.com/S1933-1711(14)00806-7/sref4http://refhub.elsevier.com/S1933-1711(14)00806-7/sref4http://refhub.elsevier.com/S1933-1711(14)00806-7/sref5http://refhub.elsevier.com/S1933-1711(14)00806-7/sref5http://refhub.elsevier.com/S1933-1711(14)00806-7/sref5http://refhub.elsevier.com/S1933-1711(14)00806-7/sref5http://refhub.elsevier.com/S1933-1711(14)00806-7/sref6http://refhub.elsevier.com/S1933-1711(14)00806-7/sref6http://refhub.elsevier.com/S1933-1711(14)00806-7/sref6http://refhub.elsevier.com/S1933-1711(14)00806-7/sref6http://refhub.elsevier.com/S1933-1711(14)00806-7/sref6http://refhub.elsevier.com/S1933-1711(14)00806-7/sref7http://refhub.elsevier.com/S1933-1711(14)00806-7/sref7http://refhub.elsevier.com/S1933-1711(14)00806-7/sref7http://refhub.elsevier.com/S1933-1711(14)00806-7/sref7http://refhub.elsevier.com/S1933-1711(14)00806-7/sref8http://refhub.elsevier.com/S1933-1711(14)00806-7/sref8http://refhub.elsevier.com/S1933-1711(14)00806-7/sref8http://refhub.elsevier.com/S1933-1711(14)00806-7/sref9http://refhub.elsevier.com/S1933-1711(14)00806-7/sref9http://refhub.elsevier.com/S1933-1711(14)00806-7/sref9http://refhub.elsevier.com/S1933-1711(14)00806-7/sref9http://refhub.elsevier.com/S1933-1711(14)00806-7/sref9http://refhub.elsevier.com/S1933-1711(14)00806-7/sref10http://refhub.elsevier.com/S1933-1711(14)00806-7/sref10http://refhub.elsevier.com/S1933-1711(14)00806-7/sref10http://refhub.elsevier.com/S1933-1711(14)00806-7/sref11http://refhub.elsevier.com/S1933-1711(14)00806-7/sref11http://refhub.elsevier.com/S1933-1711(14)00806-7/sref11http://refhub.elsevier.com/S1933-1711(14)00806-7/sref12http://refhub.elsevier.com/S1933-1711(14)00806-7/sref12http://refhub.elsevier.com/S1933-1711(14)00806-7/sref12http://refhub.elsevier.com/S1933-1711(14)00806-7/sref12http://refhub.elsevier.com/S1933-1711(14)00806-7/sref12http://refhub.elsevier.com/S1933-1711(14)00806-7/sref13http://refhub.elsevier.com/S1933-1711(14)00806-7/sref13http://refhub.elsevier.com/S1933-1711(14)00806-7/sref13http://refhub.elsevier.com/S1933-1711(14)00806-7/sref13http://refhub.elsevier.com/S1933-1711(14)00806-7/sref13http://refhub.elsevier.com/S1933-1711(14)00806-7/sref14http://refhub.elsevier.com/S1933-1711(14)00806-7/sref14http://refhub.elsevier.com/S1933-1711(14)00806-7/sref14http://refhub.elsevier.com/S1933-1711(14)00806-7/sref15http://refhub.elsevier.com/S1933-1711(14)00806-7/sref15http://refhub.elsevier.com/S1933-1711(14)00806-7/sref15http://refhub.elsevier.com/S1933-1711(14)00806-7/sref15http://refhub.elsevier.com/S1933-1711(14)00806-7/sref15http://refhub.elsevier.com/S1933-1711(14)00806-7/sref16http://refhub.elsevier.com/S1933-1711(14)00806-7/sref16http://refhub.elsevier.com/S1933-1711(14)00806-7/sref17http://refhub.elsevier.com/S1933-1711(14)00806-7/sref17http://refhub.elsevier.com/S1933-1711(14)00806-7/sref17http://refhub.elsevier.com/S1933-1711(14)00806-7/sref17http://refhub.elsevier.com/S1933-1711(14)00806-7/sref17http://refhub.elsevier.com/S1933-1711(14)00806-7/sref18http://refhub.elsevier.com/S1933-1711(14)00806-7/sref18http://refhub.elsevier.com/S1933-1711(14)00806-7/sref18http://refhub.elsevier.com/S1933-1711(14)00806-7/sref18http://refhub.elsevier.com/S1933-1711(14)00806-7/sref19http://refhub.elsevier.com/S1933-1711(14)00806-7/sref19http://refhub.elsevier.com/S1933-1711(14)00806-7/sref19http://refhub.elsevier.com/S1933-1711(14)00806-7/sref19http://refhub.elsevier.com/S1933-1711(14)00806-7/sref20http://refhub.elsevier.com/S1933-1711(14)00806-7/sref20http://refhub.elsevier.com/S1933-1711(14)00806-7/sref20http://refhub.elsevier.com/S1933-1711(14)00806-7/sref20http://refhub.elsevier.com/S1933-1711(14)00806-7/sref20http://refhub.elsevier.com/S1933-1711(14)00806-7/sref21http://refhub.elsevier.com/S1933-1711(14)00806-7/sref21http://refhub.elsevier.com/S1933-1711(14)00806-7/sref21http://refhub.elsevier.com/S1933-1711(14)00806-7/sref22http://refhub.elsevier.com/S1933-1711(14)00806-7/sref22http://refhub.elsevier.com/S1933-1711(14)00806-7/sref22http://refhub.elsevier.com/S1933-1711(14)00806-7/sref23http://refhub.elsevier.com/S1933-1711(14)00806-7/sref23http://refhub.elsevier.com/S1933-1711(14)00806-7/sref23http://refhub.elsevier.com/S1933-1711(14)00806-7/sref23http://refhub.elsevier.com/S1933-1711(14)00806-7/sref24http://refhub.elsevier.com/S1933-1711(14)00806-7/sref24http://refhub.elsevier.com/S1933-1711(14)00806-7/sref24http://refhub.elsevier.com/S1933-1711(14)00806-7/sref24http://refhub.elsevier.com/S1933-1711(14)00806-7/sref24http://refhub.elsevier.com/S1933-1711(14)00806-7/sref24http://refhub.elsevier.com/S1933-1711(14)00806-7/sref24http://refhub.elsevier.com/S1933-1711(14)00806-7/sref24http://refhub.elsevier.com/S1933-1711(14)00806-7/sref23http://refhub.elsevier.com/S1933-1711(14)00806-7/sref23http://refhub.elsevier.com/S1933-1711(14)00806-7/sref23http://refhub.elsevier.com/S1933-1711(14)00806-7/sref23http://refhub.elsevier.com/S1933-1711(14)00806-7/sref22http://refhub.elsevier.com/S1933-1711(14)00806-7/sref22http://refhub.elsevier.com/S1933-1711(14)00806-7/sref22http://refhub.elsevier.com/S1933-1711(14)00806-7/sref21http://refhub.elsevier.com/S1933-1711(14)00806-7/sref21http://refhub.elsevier.com/S1933-1711(14)00806-7/sref21http://refhub.elsevier.com/S1933-1711(14)00806-7/sref20http://refhub.elsevier.com/S1933-1711(14)00806-7/sref20http://refhub.elsevier.com/S1933-1711(14)00806-7/sref20http://refhub.elsevier.com/S1933-1711(14)00806-7/sref20http://refhub.elsevier.com/S1933-1711(14)00806-7/sref20http://refhub.elsevier.com/S1933-1711(14)00806-7/sref19http://refhub.elsevier.com/S1933-1711(14)00806-7/sref19http://refhub.elsevier.com/S1933-1711(14)00806-7/sref19http://refhub.elsevier.com/S1933-1711(14)00806-7/sref19http://refhub.elsevier.com/S1933-1711(14)00806-7/sref18http://refhub.elsevier.com/S1933-1711(14)00806-7/sref18http://refhub.elsevier.com/S1933-1711(14)00806-7/sref18http://refhub.elsevier.com/S1933-1711(14)00806-7/sref18http://refhub.elsevier.com/S1933-1711(14)00806-7/sref17http://refhub.elsevier.com/S1933-1711(14)00806-7/sref17http://refhub.elsevier.com/S1933-1711(14)00806-7/sref17http://refhub.elsevier.com/S1933-1711(14)00806-7/sref17http://refhub.elsevier.com/S1933-1711(14)00806-7/sref17http://refhub.elsevier.com/S1933-1711(14)00806-7/sref16http://refhub.elsevier.com/S1933-1711(14)00806-7/sref16http://refhub.elsevier.com/S1933-1711(14)00806-7/sref15http://refhub.elsevier.com/S1933-1711(14)00806-7/sref15http://refhub.elsevier.com/S1933-1711(14)00806-7/sref15http://refhub.elsevier.com/S1933-1711(14)00806-7/sref15http://refhub.elsevier.com/S1933-1711(14)00806-7/sref15http://refhub.elsevier.com/S1933-1711(14)00806-7/sref14http://refhub.elsevier.com/S1933-1711(14)00806-7/sref14http://refhub.elsevier.com/S1933-1711(14)00806-7/sref14http://refhub.elsevier.com/S1933-1711(14)00806-7/sref13http://refhub.elsevier.com/S1933-1711(14)00806-7/sref13http://refhub.elsevier.com/S1933-1711(14)00806-7/sref13http://refhub.elsevier.com/S1933-1711(14)00806-7/sref13http://refhub.elsevier.com/S1933-1711(14)00806-7/sref13http://refhub.elsevier.com/S1933-1711(14)00806-7/sref12http://refhub.elsevier.com/S1933-1711(14)00806-7/sref12http://refhub.elsevier.com/S1933-1711(14)00806-7/sref12http://refhub.elsevier.com/S1933-1711(14)00806-7/sref12http://refhub.elsevier.com/S1933-1711(14)00806-7/sref12http://refhub.elsevier.com/S1933-1711(14)00806-7/sref11http://refhub.elsevier.com/S1933-1711(14)00806-7/sref11http://refhub.elsevier.com/S1933-1711(14)00806-7/sref11http://refhub.elsevier.com/S1933-1711(14)00806-7/sref10http://refhub.elsevier.com/S1933-1711(14)00806-7/sref10http://refhub.elsevier.com/S1933-1711(14)00806-7/sref10http://refhub.elsevier.com/S1933-1711(14)00806-7/sref9http://refhub.elsevier.com/S1933-1711(14)00806-7/sref9http://refhub.elsevier.com/S1933-1711(14)00806-7/sref9http://refhub.elsevier.com/S1933-1711(14)00806-7/sref9http://refhub.elsevier.com/S1933-1711(14)00806-7/sref9http://refhub.elsevier.com/S1933-1711(14)00806-7/sref8http://refhub.elsevier.com/S1933-1711(14)00806-7/sref8http://refhub.elsevier.com/S1933-1711(14)00806-7/sref8http://refhub.elsevier.com/S1933-1711(14)00806-7/sref7http://refhub.elsevier.com/S1933-1711(14)00806-7/sref7http://refhub.elsevier.com/S1933-1711(14)00806-7/sref7http://refhub.elsevier.com/S1933-1711(14)00806-7/sref7http://refhub.elsevier.com/S1933-1711(14)00806-7/sref6http://refhub.elsevier.com/S1933-1711(14)00806-7/sref6http://refhub.elsevier.com/S1933-1711(14)00806-7/sref6http://refhub.elsevier.com/S1933-1711(14)00806-7/sref6http://refhub.elsevier.com/S1933-1711(14)00806-7/sref6http://refhub.elsevier.com/S1933-1711(14)00806-7/sref5http://refhub.elsevier.com/S1933-1711(14)00806-7/sref5http://refhub.elsevier.com/S1933-1711(14)00806-7/sref5http://refhub.elsevier.com/S1933-1711(14)00806-7/sref5http://refhub.elsevier.com/S1933-1711(14)00806-7/sref4http://refhub.elsevier.com/S1933-1711(14)00806-7/sref4http://refhub.elsevier.com/S1933-1711(14)00806-7/sref4http://refhub.elsevier.com/S1933-1711(14)00806-7/sref3http://refhub.elsevier.com/S1933-1711(14)00806-7/sref3http://refhub.elsevier.com/S1933-1711(14)00806-7/sref3http://refhub.elsevier.com/S1933-1711(14)00806-7/sref2http://refhub.elsevier.com/S1933-1711(14)00806-7/sref2http://refhub.elsevier.com/S1933-1711(14)00806-7/sref2http://refhub.elsevier.com/S1933-1711(14)00806-7/sref2http://refhub.elsevier.com/S1933-1711(14)00806-7/sref2http://refhub.elsevier.com/S1933-1711(14)00806-7/sref1http://refhub.elsevier.com/S1933-1711(14)00806-7/sref1http://refhub.elsevier.com/S1933-1711(14)00806-7/sref1