troponin use it in all patients with acute heart failure! pro

Troponin: Use it in all patients

with acute heart failure!

PRO

W. Frank Peacock, MD, FACEPProfessor, Emergency MedicineBaylor College of Medicine

Alan Maisel, MD, FACC

• The father of BNP• Past section editor of JACC

• PI of 100’s of major biomarker studies and publications

• Founder/Director of UCSD Biomarker conference

We had to promise….

You don’t need troponin in HF?

Troponin in AHF?

• Diagnosis?– Nope

• Prognosis– Ton’s of data– Useless because its always positive?

• What to do with it?– Is troponin (+) HF patient different than one that is troponin (-)?

Tn Elevation w/o Overt Cardiac Ischemia

• Trauma• contusion, ablation, pacing, ICD firings,

cardioversion, endomyocardial biopsy, cardiac surgery, interventional closure of ASDs

• CHF• Aortic valve disease and HOCM with significant

LVH• HTN• Hypotension, often with arrhythmias• Postoperative noncardiac surgery patients who

seem to do well• Renal failure• Critically ill patients, esp with diabetes,

respiratory failure, gi bleeding, sepsis• Drug toxicity, eg adriamycin, 5 FU, herceptin,

snake venoms, carbon monoxide poisoning• Hypothyroidism• Abnormalities in coronary vasomotion, including

coronary vasospasm• Apical ballooning syndrome

• Inflammatory diseases– myocarditis, eg. Parvovirus B19, Kawasaki

disease, sarcoid, smallpox vaccination, or myocardial extension of BE

• Post PCI patients who appear to be uncomplicated

• Pulmonary embolism, severe pulmonary hypertension

• Sepsis• Burns, esp if TBSA > 30%• Infiltrative diseases including amyloidosis,

hemachromatosis, sarcoidosis and scleroderma• Acute neurological disease

– CVA, subarchnoid bleeds• Rhabdomyolysis with cardiac injury• Transplant vasculopathy• Vital Exhaustion

Cardiac Troponin and Outcome in Acute Heart Failure

W. Frank Peacock IV, MD, Teresa De Marco, MD, Gregg C. Fonarow, MD, Deborah Diercks, MD,

Janet Wynne, MS, Fred S. Apple, PhD, and Alan H.B. Wu, PhD

for the ADHERE Investigators

New England Journal of Medicine. 2008;358:2117-26.

Mortality According to Time in Hospital and Troponin Status at Presentation

Troponin-positive

Troponin-negative

Days in Hospital

Cum

ulat

ive

Mor

talit

y (%

) 25

20

15

10

5

00 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

P <0.001*

*Dashed lines show 95% CI

Peacock WF et al. N Engl J Med. 2008;358:2117-26.

N=67,928

In-hospital Mortality According to Troponin T Quartile

P <0.001

No. of Patients 1773 502 1138 1119

In-H

ospi

tal M

orta

lity

(%)

Troponin T Quartile

0

2

4

6

8

1.72.8

3.3

6.3

≤ 0.01 > 0.01-0.02 > 0.02-0.06 > 0.06


In-hospital Mortality Accordingto Troponin I Quartile

0

2

4

6

8

2.02.7

3.4

5.3

P <0.001

No. of Patients 11,090 10,367 9323 9534

In-H

ospi

tal M

orta

lity

(%)

Troponin I Quartile≤ 0.04 > 0.04-0.10 > 0.10-0.2 > 0.2


Resource*

Positive for TroponinN=4240

Negative for TroponinN=63,684 P Value

Odds Ratio (95% CI)

Intensive care unitAny time – no. (%) 1565 (37) 10,493 (16) <0.001 NAMedian days (IQR) 2.9 (1.6-5.0) 2.3 (1.2-4.1) <0.001 NAAdjusted mean days 4.1 3.7 0.007 NAHospital – daysMedian stay (IQR) 5.1 (3.2-8.3) 4.1 (2.8-6.7) <0.001 NAAdjusted mean stay 6.6 5.5 <0.001 NAProcedures – no. (%)CABG 164(4) 478 (1) <0.001 5.46 (4.54-6.57)IABP 113 (3) 192 (<1) <0.001 8.03 (6.30-10.2)Cardiac catheterization 1002 (24) 6383 (10) <0.001 3.04 (2.81-3.28)Mechanical ventilation 479 (11) 641 (1) <0.001 2.68 (2.41-2.99)

Resource Utilization According to Troponin Status

*Means were adjusted for age, blood urea nitrogen level, systolic and diastolic blood pressure, creatinine, sodium, heart rate, and dyspnea at rest. Odds ratios are for troponin-positive compared to troponin-negative and were adjusted for age, blood urea nitrogen level, systolic and diastolic blood pressure, creatinine, sodium, heart rate, and dyspnea at rest. CABG: coronary-artery bypass graft, IABP: intraaortic balloon pump, IQR: interquartile range.

Peacock WF. N Engl J Med. 2008;358:2117-26.

Conclusion• Patients reporting with acute decompensated HF

and high troponin were a high-risk cohort– More cardiac procedures– Longer hospitalization– Higher risk of in-hospital death

• Measurement of troponin adds important prognostic information to evaluation of patients presenting with acute HF– Patients with positive troponin had a twice as high risk of

death• Troponin should be considered part of early risk assessment


BACHBACH

•• BACH TrialBACH Trial–– ProspectiveProspective–– 15-center15-center–– International study of ED patients International study of ED patients

presenting with dyspneapresenting with dyspnea–– March 2007 to February 2008March 2007 to February 2008

–– N=1641N=1641

14 day mortality

• 1641 BACH SOB patients– Gold standard AHF: 568 (34.6 %)

• 52% were male• 36% had a prior history of HF

– Overall• 20 (3.5%) died by 14 days• 65 (11.4%) were dead by 90 days.

Peacock WF. Acad EM 2011;18(9):947-958

Can you predict 14 day mortality?

Variables14-Day

Survivor (n=447)

14-Day Mortality (n=19) p-value

Age (years) 70.4 ±14.1 79.1 ±8.9 0.0065Pulse oximetry (%) 94.8 ±5.7 90.9 ±8 0.0253Wheezing 61 (13.9) 8 (42.1) 0.0034Hyperlipidemia 204 (49.5) 2 (11.1) 0.0012

• 20 (3.5%) died by 14 days• 65 (11.4%) were dead by 90 days

• Non-survivors vs survivors, appear nearly clinically identical in all be 4/47 presenting characteristics


ED Treatment

VariablesAll Patients

(n=466)

14-Day Survivor (n=447)

14-Day Mortality

(n=19)p-value

Oxygen 180 (38.6) 169 (37.8) 11 (57.9) 0.1552IV vasodilators (Ntg, Ntp, nesiritide) 128 (27.5) 121 (27.1) 7 (36.8) 0.6370IV inotropes (milrinone, Dopa, NE) 6 (1.3) 6 (1.3) 0 (0) 1.0000Furosemide 303 (65) 289 (64.7) 14 (73.7) 0.4732PAC 12 (2.6) 12 (2.7) 0 (0) 1.0000ICU Admit 37 (7.9) 32 (7.2) 5 (26.3) 0.0123Hospital Admit 395 (84.8) 377 (84.3) 18 (94.7) 0.3324


Peacock WF. Acad EM 011; 18(9):947-958

AHF Mortality

Dr. Maisel, I know you were the PI of the BACH trial, but really Alan?

Lumper or Splitter?

• Lumping and splitting are opposing tendencies in any discipline which has to place individual examples into rigorously defined categories.

– A "lumper" takes a gestalt view of a definition, and assigns examples broadly, assuming that differences are not as important as signature similarities

– A "splitter" takes precise definitions, and creates new categories to classify samples that differ in key ways

To be clinically useful, a phenotypic division must have 3 characteristics

1) Can we identify a population (i.e., a phenotype), with unique characteristics?

2) Are the unique characteristics manifested in a treatment response

3) Once identified as a unique population, can we change the predicted outcomes with an intervention?

HF Phenotypic Division

Acute vs Chronic• AHF

– BP, Tn, NP, Gal 3, ST2

• CHF– EF, Tn, NP– Chronotropic incompetence

Consequence of disease• Acute Hemodynamics

(manifestations)– BP, HR– Vascular reactivity (alt.

baroreceptor control)

• Protoplasmic– Inflammation, Fibrosis

• Precipitants (etiology)– IHD, DM, HTN

Is there a Troponin (+) phenotype?

1) Can we identify a population (i.e., a phenotype), with unique

characteristics?

Serial changes in HsTnIpredict outcome in patients with ADHF

• 144 ADHF followed from hospital to 90 days post-d/c– 1˚ endpoints: all cause mortality and HF-related readmits

• TnI & BNP at 6 draws– Admit, d/c, and 4 consecutive days during hospitalization

• ROC curves at discharge determined for endpoints– TnI 23.25 ng/L (C-stat 0.648, 95% CI 0.553–0.742)– BNP 360 ng/L (C-stat 0.637, 95% CI 0.543–0.731)

Xue Y. EJHF. doi:10.1093/eurjhf/hfq210

Troponin Assay

• Verisens RUO Human cTnI Assay by Nanosphere (Northbrook, IL, USA)

• Analytical sensitivity of the assay is 0.25 ng/L• 10% CV = 8 ng/L• 99th %ile = 4.50 ng/L


Event Free by Discharge TnI

Increased mortality and readmits P<0.003

HR = 3.547


Event Free: D/C TnI Quartiles


TnI 0.7–13.7 ng/L

TnI 13.7–29.5 ng/L

TnI 87.6 ng/L ≥1000 ng/L

TnI 29.5–87.6 ng/L

P<0.019, HR 6.07P<0.007, HR 7.83P< 0.014, HR 6.550

90 Day Events by D/C TnI and BNP


Lo Lo, n=34

Hi Lo, p<0.015, HR 12.94

Lo Hi, p<0.042, HR 9.02

Hi Hi, p<0.007, HR 15.97

Cutoffs: TnI 23.25 ng/LBNP 360 ng/L

n=34, 26, 28, 56

• Endpoints had an early TnI increase• Event free had fairly stable TnI

Mean TnI Levels in Patients ± Events


Survival by Increasing (any Tn> baseline) vs. Stable or Decreasing TnI


P<0.047HR= 4.52

Xue Y, Clopton P, Peacock WF, Maisel AS. Serial changes in high-sensitive Troponin I predict outcome in patients with decompensated heart failure. EJHF 2011;13:37-42.

What is the spectrum?

• Low circulating levels of Tn, detectable now with highly sensitive assays, may indicate subclinical chronic myocardial injury and thereby identify heightened risk for pathological cardiac remodeling and the development of HF.

De Lemos JA. JACC 59(5) 2012doi:10.1016/j.jacc.2011.10.874

Italian Predictor Study Subanalysis

• Examined the relationship between hs-cTnT and– LV mass/BSA (women >95 g/m2, men > 115 g/m2 )

• N=1973 community dwelling elderly– mean age 73 (range 65–84)

• hs-cTnT (Elecsys 2010, Roche Diagnostics GmbH)– Range 3–10 000 ng L1

– 99th %ile 14 pg mL– Reports results below LOD (3 ng/L1) in 30.1% of subjects

Masson S. J Intern Med 2013; 273: 306–317.

Italian Predictor Study Subanalysis

• 24.8% of subjects had elevated LV mass.

• Median cTnT higher with elevated vs nl LV mass: – 6.6 [3.5–11.6] vs. 4.6 [3.0–8.1] ng/L (P < 0.001)

• Median hs-cTnT increased in parallel to LV mass: – Normal LV mass 4.6 [3.0–8.1]– LV hypertrophy: mild 5.8 [3.1–10.2], moderate 7.6 [3.8–13.7], severe 8.4 [3.8–17.6] (P < 0.0001)

Masson S. J Intern Med 2013; 273: 306–317.

Troponin

• Troponin at HF presentation predicts – Death at that hospitalization– Death within 2 weeks (who should go to ICU?)

• Hi Sensitive– Predicts outcomes– Serial levels identify worse prognosis

– Elevated levels associated with structural abnormalities (increased LV mass)

No Troponin in HF???

Don’t know where you got that, but you should put it

back….

Can troponin > 99th%ile identify a cohort with a

unique treatment response?

2) Are the unique characteristics manifested in a treatment response

Troponin Lab-test-ology

• 95th vs 99th %ile

• 10% CV• LOD• Sensitivityand Specificity

Hs Tn Definition

Apple FS. A new season for cardiac troponin assays: it’s time to keep a scorecard. Clin Chem 2009;55:1303–6.

HsTnT: A Biomarker for Diuretic Response in decompensated HF

• 2˚ analysis of a spironolactone RCT– N=100 ADHF patients– COBAS TnT STAT assay (Roche Diagnostics)– URL (99th %ile) 0.014 ng/mL

• Furosemide diuretic response defined as:– “Faster” if ↓ IV, or switched to po, in the 1st 3 days– “Slower” if ↑ or maintained after 3 days in hospital

Ferreira JP. Card Res Pract. 2014, ID 269604

HsTnT: A Biomarker for Diuretic Response in decompensated HF

• Overall hsTnT ↓’d from days 1 to 3 (p = 0. 039)• Patients with hsTnT increase had longer LOS (p=0.033)– “Slower diuresing” remained decompensated

• No differences in hsTnT (p = 0.955)

– “Faster diuresing” were successfully compensated• Reduction in hsTnT (p = 0.025)• HsTnT decrease was correlated with NTproBNP reduction (p = 0. 007)

Ferreira JP. Card Res Pract. 2014, ID 269604

Can Tn > 99th%ile identify a cohort with a unique

treatment response that can be mitigated?3) Once identified as a unique population, can we change the predicted outcomes with an

intervention?

RELAX-AHF

• International, multicenter, double-blind, placebo-controlled trial– Hospitalized AHF randomized within 16 h to intravenous placebo or serelaxin.

– TnI measured at baseline and days 2, 5, and 14

• hs-cTnT measured by Roche Elecsys assay– 99th %ile URL = 0.014 g/l– The lowest level with a CV of 10% was 0.013 g/l

Metra M. JACC. 2013;61:196–206

RELAX-AHF

• Overall serelaxin reduces 180-day mortality– combined studies: N=1,395– HR: 0.62; 95% CI 0.43 to 0.88 (p =0.0076).

• At baseline, hs-cTnT plasma levels >URL in 93% of patients

Metra M. JACC. 2013;61:196–206

RELAX-AHFRising TnT associated with death• Increased baseline troponin associated with increased 180 day all cause death

• Changes in hs-cTnI at day 2 were associated 180-day mortality

Metra M. JACC. 2013;61:196–206

RELAX AHF All cause death thru day 180

Subdivided by % TnT change from baseline to day 2

Metra M. JACC. 2013;61:196–206

RELAX-AHF

Placebo Serelaxin P valueRate of Tnt Increase > 20%

145/534 (27.2%)

86/522 (16.5%)

<0.0001

Metra M. JACC. 2013;61:196–206

• Serelaxin associated with decreased rate of rising troponin

Treatment has an effect

Metra M. JACC. 2013;61:196–206

Summary

• Troponin elevation in AHF identifies a subset of patients who are:– at higher risk of short/long term death– more likely to have structural cardiac abnormalities

– less responsive to diuretic therapy– May be targets for specific iv neurohormonal antagonist therapies

• wait for Relax II and TRUE-HF

• Troponin– Measure at admission

– Repeat if fail to improve or change for the worse

– Repeat if initially above the 99th %ile

Put those crazy ideas about

“No Tn in HF” back where you got them…..

troponin use it in all patients with acute heart failure! pro

Health & Medicine