PARTNERS HF PPT - OptiVol® Fluid Status Monitoring

Report
Combined Heart Failure Device Diagnostics
Identify Patients at Higher Risk of
Subsequent Heart Failure Hospitalizations:
Results from PARTNERS HF Study
David J. Whellan, MD FACC
Associate Professor of Medicine
Director, Jefferson Coordinating Center for Clinical Research
Jefferson Medical College
on behalf of the PARTNERS HF Study Investigators
Whellan et al. JACC Vol. 55, No. 17, April 27, 2010:1803–10
Disclosure
I will not discuss off label use or investigational use
in my presentation.
I have financial relationships to disclose:
Employee of: Thomas Jefferson University
Consultant for: Medtronic
Stockholder in: NA
Research support from: Medtronic
Honoraria from: Medtronic
Background
• Despite medical interventions, there remains
a high rate of HF hospitalizations in the CRT-D
patient population.
• Identifying at risk patients is a challenge
• Recent publications have shown a single parameter
HF diagnostic identifies high risk patients
1,2
1. Small et al. J Card Fail. November 2009; 15(9):813.
2. Perego et al. Interv Card Electrophysiol 2008, 23:235-242.
Hypothesis
Routine evaluation of combined diagnostics
recorded by implantable devices can identify HF
patients at risk for subsequent heart failure
hospitalizations.
Combined Diagnostic Algorithm
• Based on 8 Diagnostic
trends which are
recorded daily
• Individual algorithms for
each trend have been
used to flag significant
observations from the
trends in Medtronic
ICDs/CRTs
OptiVol
VT/VF
therapy
AF burden
Rate during
AF
%CRT
HR
Activity
HRV
Study Design
• Prospective observational study
• Subjects with CRT ICDs
• 12-month follow-up, scheduled visits every 3
months
• Clinical and device data collected at all visits
• Limitations:
 Clinicians had access to the diagnostics
 Review and/or interventions based on trends were not
required and alerts were not utilized
Methods – Categorization of Events
• All CV and HF-related events were collected
• Events and deaths were classified and
adjudicated by an independent committee
• Primary endpoint was the number of HF
hospitalizations with pulmonary congestion
Combined Algorithm
Positive Combined Algorithm = any 2 criteria +
Parameter
Criterion
Fluid Index
AT/AF Duration
V. rate during AT/AF
Patient Activity
Night Heart Rate
HRV
CRT % Pacing
Shock(s)
≥60 ohm/days
≥6 hours & not persistent AT/AF
AT/AF ≥24 hrs & V. ≥ 90 bpm
Avg. <1 hr over 1 week
≥85 bpm for 7 consecutive days
<60 ms for 7 consecutive days
< 90% for 5 of 7 days
1 or more shocks
OR Fluid Index ≥100
Combined Algorithm
• Algorithm criteria were tested on an independent data set
from a registry (819 patients) to determine optimal # of
criteria met to trigger combined algorithm
# Criteria Evaluations
1
43%
too high
2
14%
optimal
3
3%
too few
• Combined algorithm also used prior finding that a high
(≥100) fluid index alone has higher specificity
Monthly Evaluation Model
Start*
30
Diagnostic Risk
Assessment 1
60
90
...Repeat until
End of Follow-up
HF Event
Assessment 1
Evaluation 1
Diagnostic Risk
Assessment 2
HF Event
Assessment 2
Evaluation 2
Diagnostic Risk
Assessment 3
HF Event
Assessment 3
Evaluation 3
* Day 0 = later of consent date or
60 days post-implant
• Repeated using Quarterly (90 days) and Semi-monthly (15 days) evaluations
Statistical Methods
• Cox proportional hazards model to adjust for
pre-defined clinical variables including:
 Age
 Gender
 Heart Failure Etiology
 NYHA Class*
 Diabetes
 HF Medication Regimen (Diuretics, ACE/ARB, B-Blocker)*
* Most recent prior to evaluation
• Sub-group analysis for subjects with and without a
HF event.
Results: Cohort and Event Rates
• 694 patients in this analysis cohort who had
impedance monitoring and >2 months of FU
• 60 patients (8.5%) had 78 monthly evaluation
periods with at least one HF hosp. (pulmonary)
• Low event rate: 1.4% (78/5693) of monthly
evaluations had HF hosp. (pulmonary)
Baseline Characteristics (N = 694)
Age
68 yrs
Female
33%
African American
11%
NYHA (Class III)
95%
History of AF
26%
Ischemic HF
62%
Diabetic
40%
Diuretics
83%
ACE Inhibitor or ARB
81%
Beta Blockers
89%
Combined Diagnostics Triggered
1324 monthly evaluations with combined algorithm triggered
≥ 2 Diagnostic Criteria Met
43%
% of
evaluations
when ≥2
Diagnostic
Criteria Met
(N = 960)
75%
OptiVol Fluid Index ≥100 ohm days Met
29%
67%
28%
% of triggered evaluations
62%
60%
43%
45%
30%
21%
18%
14%
15%
7%
5%
0%
AF
AF+RVR OptiVol
Low
Night
Index ≥60 Activity HR
Low
HRV
Low ICD Shock(s)
Pacing%
Evaluations with Heart Failure
Hospitalization (Pulmonary)
Kaplan-Meier HF Hospitalization Curves
6%
P < 0.0001
Hazard Ratio = 5.5 (95% CI: 3.4 – 8.8)
5%
+ Diagnostic
4%
3%
2%
1%
- Diagnostic
0
0
10
20
30
Days After Diagnostic Evaluation
Risk of a HF hosp. for pts with
+ Diagnostic was 5.5 x risk of pts w/ - Diagnostic
Multivariable Analysis
P-Value
1
Age
0.90
0.7
Gender
0.15
1
Heart Failure Etiology
0.91
1.4
NYHA Class*
0.18
1.6
Diabetes (@baseline)
0.06
1.6
Diuretics*
0.35
0.7
ACE/ARB*
0.7
0.9
Beta-Blockers*
0.90
4.8
+ Combined Diagnostic
* Before evaluation date
0
1
2
3
4
5
<0.0001
6
7
8
Hazard Ratio
Patients w/ + combined diagnostic were 4.8 times
more likely to have a HF hospitalization with
pulmonary congestion independent of other clinical variables.
Subgroup by HF Event
0.9
5.4
0.9
Subjects with
P = 0.85
5.4
Pulm HF Hosp.
-1
1
3
5
7
9
-1
1
3
5
7
9
7
9
0.9
P < 0.0001
5.4
Subjects without
Pulm. HF Hosp.
-1
1
3
5
Effect of Evaluation Frequency
Evaluation Frequency
6.9
15 Days (Semi-Monthly)
5.5
30 Days (Monthly)
3.1
90 Days (Quarterly)
-1
1
3
5
7
9
11
Hazard Ratio
• More frequent evaluations enhance risk stratification.
• Monthly evaluations provide reasonable balance of
risk stratification benefit and clinician effort.
Conclusion
• Patients with a + combined diagnostic were
5.5x more likely to have a HF hospitalization
with pulmonary congestion before the next
evaluation
• The combined algorithm is an independent
predictor of HF hospitalization in patients
without a HF hospitalization, while it provides
limited information in patients who have
experienced a HF hospitalization.
Clinical Implications
• Monthly evaluation of combined
diagnostics can identify patients at a
higher risk of a HF hospitalization within
the next month.
• Intervening by either modification of
medications (i.e. diuretic dosing) or
increase surveillance (i.e. clinic visit) may
reduce clinical events
HF Hosp. w/Pulm. Cong.
30 Day Evaluation window
Case
Study 1
AF+
Fluid +
Case Study 2
HF Hospitalization with
Pulmonary Congestion
Evaluation Window
OptiVol Index+
Activity+
HRV+

similar documents