Heart Failure Hospitalizations - ACC.21 Scientific Session ...

Reduction in Heart Failure Hospitalizations

With Ambulatory Hemodynamic Monitoring

Seen in Clinical Trials Is Maintained in the

'Real World'

Akshay S. Desai, MD, MPH*, Arvind Bhimaraj, MD, MPH, Rupinder Bharmi, MS,

Rita Jermyn, MD, Kunjan Bhatt, MD, David Shavelle, MD, Margaret M. Redfield, MD,

Robert Hull, MD, Jamie Pelzel, MD, Kevin Davis, BS, Nirav Dalal, MS, MBA,

Philip B. Adamson, MD, J. Thomas Heywood, MD

Background

• Despite guideline-directed treatment, patients with

symptomatic heart failure (HF) are at high risk for recurrent

hospitalizations and death

• Most HF hospitalizations (HFH) are congestive exacerbations

driven by progressive rise in intracardiac filling pressures

• Filling pressures typically rise weeks in advance of overt

symptoms

• Ambulatory hemodynamic monitoring may enable early

detection of impending congestion and direct intervention to

prevent hospitalization Chun S, et al. Circ Heart Fail. 2012;5:414-421

Zile MR, et al. Circulation. 2008; 118:1433-1441

CHAMPION Trial (N=550) Reduction in HFH with PA Pressure-guided therapy

CardioMEMS

PAP sensor

FDA Approved

May 2014

Chronic HF

NYHA III

Prior HFH

Abraham WT, et al. Lancet 2016; 387: 453-461.

- Treatment Group (158 HF hospitalizations)

- Control Group (254 HF hospitalizations)

Open Questions

Effectiveness vs. Efficacy

• How effective is ambulatory PAP monitoring in

reducing HF hospitalizations in ‘real world’ clinical

practice?

• Does the effectiveness vary in subgroups defined by

sex or age?

• Are comprehensive healthcare costs reduced in

practice?

Methods• Objective

– To evaluate the effectiveness of hemodynamic monitoring in

reducing health care utilization in the general use population

• Study Design

– Retrospective cohort study based on administrative claims using

100% data from the CMS Standard Analytic File

• Population

– All Fee-for-service Medicare beneficiaries undergoing PAP sensor

implantation between June 1, 2014 and June 30, 2016

– At least 6 months of continuous enrollment before and after

implantation

Statistical Methods

• Rates of HF and all-cause hospitalization during the 6 months

before and after PAP sensor implantation compared using the

Andersen-Gill method (accounting for competing risks)

• Comprehensive costs pre- and post-implant compared using a

non-parametric bootstrap model

• Analyses repeated for the smaller cohort with data available for

12 months before and after implant

• Sensitivity analyses conducted in cohorts restricted to

ambulatory implants and those completing the full analytic

period

• Data analyzed in subgroups defined by sex and age

Patient Selection1935 patients with PAP sensor implants

from June 1, 2014- June 30, 2016

1241 minimum 6-month follow-up data

11146-month cohort (primary analysis)

558 with minimum 12-month follow-up data

48012 month cohort (secondary analysis)

694 implanted after Jan 1, 2016

127 not continuously enrolled in Medicare

Part A and B or enrolled in HMO plan

556 implanted after June 30, 2015

78 not continuously enrolled in Medicare

Part A and B or enrolled into HMO plan

Patient Characteristics at Implant

6-month cohort (N=1114) 12-month cohort (N=480)

Mean age (yrs) 71.3 ± 10.8 71.4 ± 11.4

Age ≥ 75 yrs 460 (41.3%) 211 (44.0%)

Female 403 (36.2%) 180 (37.5%)

Race

White

Black

Other

902 (81.0%)

161 (14.5%)

51 (4.6%)

396 (82.5%)

69 (14.4%)

15 (3.1%)

Comorbidities

Diabetes

Hypertension

COPD

727 (65.3%)

1089 (97.8%)

861 (77.3%)

311 (64.8%)

471 (98.1%)

384 (80.0%)

Clinical Outcomes before and after PAP

sensor Implantation (6-month cohort)

1899

1020879

0

500

1000

1500

2000

All-CauseHospitalizations

HF Hospitalizations Non-HFHospitalizations

Pre-Implant Post-Implant

Clinical Outcomes before and after PAP

sensor Implantation (6-month cohort)

HR 0.55, 95% CI (0.49-0.61)

p<0.001

Cu

mu

lati

ve H

F H

osp

ital

izat

ion

s

1899

1020879

1119

381

738

0

500

1000

1500

2000

All-CauseHospitalizations

HF Hospitalizations Non-HFHospitalizations

Pre-Implant Post-Implant

Post-Implant:

139 deaths

17 VAD/transplants

NB: Event accumulation during the pre-implant interval displayed backwards

from time of implant

Clinical Outcomes before and after PAP

sensor Implantation (12-month cohort)

HR 0.66, 95% CI (0.57-0.76)

p<0.001

Cu

mu

lati

ve H

F H

osp

ital

izat

ion

s

1387

696 691

859

300

559

0

500

1000

1500

All-CauseHospitalizations

HF Hospitalizations Non-HFHospitalizations

Pre-Implant Post-Implant

Post-Implant:

106 deaths

15 VAD/transplants

NB: Event accumulation during the pre-implant interval displayed backwards

from time of implant

Healthcare Costs Before and After

PAP Sensor Implantation

$28,870

$47,690

$18,360

$34,500

$0

$10,000

$20,000

$30,000

$40,000

$50,000

$60,000

$70,000

$80,000

6-month cohort 12-month cohort

Pre-Implant Post-Implant

-$10,510

($9,703-$11,330)

-$13,190

($11,590-$14,740)

*Costs shown per patient over time interval noted

Healthcare Costs Before and After

PAP Sensor Implantation

$28,870

$47,690

$18,360

$34,500

$0

$10,000

$20,000

$30,000

$40,000

$50,000

$60,000

$70,000

$80,000

6-month cohort 12-month cohort

Pre-Implant Post-Implant

-$10,510

($9,703-$11,330)

-$13,190

($11,590-$14,740)

*Costs shown per patient over time interval noted

Subgroups of Interest

Age

Sex

Implant Type

Excluding Death

Excluding Death/VAD/Tx

Hazard Ratio for HFH (post v. pre)

Subgroups of Interest

Age

Sex

Implant Type

Excluding Death

Excluding Death/VAD/Tx

Hazard Ratio for HFH (post v. pre)

Limitations

• Nonrandomized data, no concurrent control

• Data regarding EF and NYHA Class not available

• Unable to comment on medication or hemodynamic changes

following device implantation

• Cannot exclude confounding by additional enhancements to

disease management in post-implant interval

Conclusions• In this retrospective, Medicare claims-based analysis,

cumulative HFH were 45% lower in the 6 months following PAP

sensor implantation than in the 6 months prior

• Similar HFH reductions seen in the subset of patients with

available 12 month pre- and post-implant data

• No observed variation in HFH reduction by sex or age

• Reductions in healthcare utilization translate into estimated

cost savings of $13,190 at 1 year (breakeven point of ~2 years)

• These data from an unselected population support the ‘real

world’ effectiveness of this approach to HF management

Back-up slides

Billing procedure codes used to identify PAP sensor implants

20

Billing codes used to identify HF hospitalizations

21

1899

1020879

1119

381

738

0

500

1000

1500

2000

All-CauseHospitalizations

HF Hospitalizations Non-HFHospitalizations

Pre-Implant Post-Implant

Post-Implant:

139 deaths

17 VAD/transplants

NB: Event accumulation during the pre-implant interval displayed backwards

from time of implant

Cu

mu

lati

ve A

ll-C

ause

Ho

spit

aliz

atio

ns

All-cause clinical outcomes before and after PAP sensor

Implantation (6-month cohort)

All-cause clinical outcomes before and after PAP sensor

Implantation (12-month cohort)

Cu

mu

lati

ve A

ll-C

ause

Ho

spit

aliz

atio

ns

1387

696 691

859

300

559

0

500

1000

1500

All-CauseHospitalizations

HF Hospitalizations Non-HFHospitalizations

Pre-Implant Post-Implant

Post-Implant:

106 deaths

15 VAD/transplants

NB: Event accumulation during the pre-implant interval displayed backwards

from time of implant

HF Healthcare Costs Before and After PAP Sensor Implantation

$12,410

$19,900

$4,945

$8,690

$0

$5,000

$10,000

$15,000

$20,000

$25,000

$30,000

$35,000

$40,000

6-month cohort 12-month cohort

Pre-Implant Post-Implant

-$7,433

($7,000-$7,884)

-$11,260

($10,460-$12,020)

*Costs shown per patient over time interval noted

PA Pressure Reduction in PracticeFirst 2000 Commercial PAP sensor Implants

Heywood JT, et. al. Circulation. 2017; ePub

CHAMPION Control

CHAMPION Treatment

General Use Cohort

Cha

nge

in P

AP

(A

UC

), m

mH

g-da

y

P=

0.0

47

6

P<

0.0

01