Efficacy and safety of outpatient treatment based on the Hestia clinical decision rule with or without NT-proBNP testing in patients with acute pulmonary embolism: a randomized clinical trial

Am J Respir Crit Care Med. 2016;194:998-1006

Paul den Exter for the Vesta Study Investigators

Department of Thrombosis and Hemostasis LUMC Leiden

Conflict of Interest Disclosure Form

In accordance with the rules of the Health Care Inspectorate (IGZ)

Name:

MVHuisman

Affiliation:

LUMC Leiden  I have no potential conflict of interest to report

Type of affiliation / financial interest

Receipt of grants/research supports: Receipt of honoraria or consultation fees: Participation in a company sponsored speaker’s bureau: Stock shareholder: Other support (please specify): Scientific advisory board [

please delete the checkboxes to the left that aren’t applicable, then delete this yellow text box

]

Name of commercial company

Declarations Menno Huisman

• No pharma declaration for this study • Supported in part by a grant of the Gratama foundation •

Introduction

• Outpatient treatment has become routine practice for DVT based on high level evidence • Recent studies have indicated that

selected

patients with acute PE can be safely treated as outpatients as well • Uncertain which method to use to select

low-risk

PE patients who can safely be treated at home

Vesta: rationale

• Hestia study 1 : Outpatient treatment in 297 PE patients selected by the Hestia criteria • Mortality 1.0% • Recurrent VTE: 2.0% • Major bleeding: 0.7% • Home study 2 : Outpatient treatment in 152 PE patients selected with clinical criteria combined with NT-proBNP testing (<500ng/L): • Mortality: 0% • Recurrent VTE: 0% • Major bleeding: 0%

1. Zondag, JTH 2011; 9(8):1500-7 2. Agterof, JTH 2010; 8(6):1235-4

Vesta study objectives

1. To assess whether the Hestia clinical decision rule alone is as safe as the Hestia rule

combined

with NT-pro BNP testing to select PE patients for outpatient treatment 2. To validate the Hestia clinical decision rule

Vesta study design

• Randomized, non-inferiority open label clinical trial (Netherlands Trial Register identifier NTR2603) • Setting: 2 academic and 15 non-academic hospitals in the Netherlands between 2010 - 2014

Patients

• Aged 18 years or over • Acute, symptomatic, objectively proven (recurrent) PE • Life expectancy > 3 months • Eligible for outpatient treatment according to the Hestia clinical decision rule • Willing and able to give informed consent

Hestia clinical decision rule

Hemodynamically unstable Thrombolysis/embolectomy necessary Need for i.v. pain medication Oxygen needed to keep O2 saturation > 90% PE diagnosed during therapeutic anticoagulation Active bleeding or high risk for bleeding Documented history of HIT Creatinine clearance < 30 ml/min Uncontrolled hypertension Pregnancy Medical/social reason for treatment in the hospital? If one or more items are present, the patient can not be treated at home

Interventions

Hestia criteria met Randomization Arm A: NT proBNP testing Arm B: Direct discharge NT proBNP >500 NT proBNP <500 In hospital treatment Direct discharge Blood storage for post-hoc NT proBNP testing

Vesta – Primary and secondary endpoints

• Primary endpoint: • 30 day adverse outcome defined as: PE- or bleeding-related mortality, cardiopulmonary resuscitation or ICU admission • Secondary endpoints: • Adjudicated recurrent VTE at 3 months • Adjudicated major bleeding at 3 months • Adjudicated all-cause mortality at 3 months • Sample size: 530 patients required to demonstrate non inferiority between groups, non-inferiority margin: 3.4%, power: 80%

Flow of patients Assessed for eligibility (n=1102) )

Enrollment

Excluded: (n=552) Randomized (n=550) NT-proBNP testing Lost to follow-up (n=0)

Allocation Follow-Up n=550

Direct discharge Lost to follow-up (n=0)

Baseline charecteristics

Characteristic Age (mean ± SD) Age > 60 (n, %) Male sex (n, %) BMI (kg/m2, mean ± SD) VTE Risk factors Immobilization or recent surgery (n, %) Previous PE (n, %) Previous DVT (n, %) Active Malignancy (n, %) Estrogen use (n, %) Comorbidities COPD with therapy (n, %) Heart failure with therapy (n, %) NT-proBNP (N=275)

55 (15) 108 (39.3) 145 (52.7) 27.0 (4.6) 33 (12.2) 41 (15.1) 30 (11.0) 23 (8.5) 50 (18.4) 9 (3.3) 3 (1.1)

Direct discharge (N=275)

53 (15) 99 (36.0) 152 (55.3) 27.7 (5.2) 39 (14.4) 45 (16.6) 28 (10.3) 14 (5.2) 45 (16.6) 14 (5.2) 2 (0.7)

NT-proBNP assessment

Hestia criteria met Arm A: NT proBNP testing N=275 Arm B: Direct discharge N=275 NT proBNP >500 NT proBNP <500 NT proBNP >500 NT proBNP <500 N=34 (12%) N=241 (88%) N=23 (8%) N=252 (92%)

Hospital Home

Primary endpoint

• Direct discharge group: 3 patients: 1.1%; 95% CI: 0.2 3.2% • NT-proBNP group: 0 patients: 0%; 95% CI: 0-1.3% • Absolute difference 1.1 %: 95% CI: -0.5% - 3.2% • None of the patients who experienced the primary endpoint had elevated NT-proBNP levels at baseline

Primary and secondary endpoints

Outcome Primary endpoint* at day 30 PE related mortality ICU admission Recurrent VTE at 3 months DVT Recurrent PE Major bleeding at 3 months All-cause mortality at 3 months NT-proBNP

(n=275) 0 0 0 2 (0.73%) 1 (0.36%) 1 (0.36%) 1 (0.36%) 4 (1.5%)

Direct discharge

(n=275) 3 (1.1%) 1 (0.36%) 2 (0.73%) 3 (1.1%) 1 (0.46%) 2 (0.73%) 3 (1.1%) 3 (1.1%)

Conclusions

• Outpatient treatment of acute PE patients selected by the Hestia clinical decision rule alone is safe • Additional prognostic assessment based on NT-proBNP levels did not affect the prognosis

Conclusions

• Outpatient treatment of acute PE patients selected by the Hestia clinical decision rule alone is safe • Additional prognostic assessment based on NT-proBNP levels seemed not to affect the prognosis • Yealy editorial: Yealy DM Am J Resp Crit Care Med 2016;194: 927 - 929

Future perspective: Efficiency study

• Home PE study: Hestia vs sPESI score (NCT02811237)

Vesta Study Investigators

• LUMC, Leiden:

MV Huisman, FA Klok, PL den Exter, W Zondag

• Diakonessenhuis, Utrecht:

E Cornage, MA van de Ree

• Groene Hart Hospital, Gouda:

H Peltenburg, T Koster, F Besemer

• Medisch Spectrum Twente, Enschede:

M Eijsvogel

• Rijnland Hospital, Leiderdorp:

GJPM Jonkers, C Ootjers , K de Groot

• Medisch Centrum Haaglanden, Den Haag:

MJM de Vreede , ICM Mos

• Rode Kruis Hospital, Beverwijk:

LM Faber

• HAGA Hospital, Den Haag:

M van Gerwen, A Mairuhu

• Bronovo Hospital, Den Haag:

LT Vlasveld , T Langeveld

• Spaarne Hosptital, Hoofddorp:

C Onstijn, CF Melissant

• VUMC, Amsterdam:

HMA Hofstee, MHH Kramer

• Rijnstate Ziekenhuis, Arnhem:

MMC Hovens

• Diaconessenhuis, Leiden:

J Dolsma

• Reinier de Graaf Gasthuis, Delft:

RE Brouwer

• Van Weel Bethesda Hospital, Dirksland:

KW van Kralingen

• Ikazia Hospital, Rotterdam:

R Heller

• Amphia Hospital, Breda:

M Grootenboers

Enrollment

Assessed for eligibility Allocated to undergo NT-proBNP testing (n=279)  Consent withdrawn < 24H (n=1)  Diagnosis of PE refuted < 24H (n=3) (n=1102) ) Randomized (n=558)

Allocation

• • Excluded: (n=544) • Hemodynamically unstable (n=33) • Thrombolysis (n= 2) • High bleeding risk (n=13) • Oxygen supply (n= 182) • • Intravenous pain medication (n= 17) Already receiving anticoagulants (n=19) • Documented HIT (n=2) • Uncontrolled hypertension (n=9) • • Severe renal insufficiency (n=2) Pregnant (n=5) • Medical or social reasons (n=129) • Age < 18 years (n=2) • • Symptoms >14 days (n=63) Life expectancy < 3 months (n=10) • • Inaccessible for follow-up (n=7) Previous participation in this trial (n=4) Refusal of consent (n=15) Eligible but not randomized (n=30) Allocated to direct discharge (n=279)  Consent withdrawn < 24H (n=2)  Diagnosis of PE refuted < 24H (n = 2)

Follow-Up n=550

Lost to follow-up (n=0) Lost to follow-up (n=0)