123I-MIBG Imaging for Prediction of Mortality and Potentially Fatal Events in Heart Failure: The ADMIRE-HFX Study

Jagat Narula; Myron Gerson; Gregory S. Thomas; Manuel D. Cerqueira; Arnold F. Jacobson

doi:10.2967/jnumed.115.156406

Article Figures & Data

Figures

Tables

Download figure
Open in new tab
Download powerpoint
FIGURE 1.
Lower mortality risk based on preserved myocardial innervation. Prediction of risk for composite endpoint of death or death-equivalent event is improved with addition of H/M. (A) Image from 37-y-old man with nonischemic cardiomyopathy, LVEF of 28%, and BNP of 378 ng/mL. Baseline 2-variable model predicted 21% (intermediate) 2-y risk. ¹²³I-MIBG H/M was 1.69. Three-variable model (LVEF, BNP, and H/M) predicted 14% (intermediate) risk. Risk for all-cause mortality based on H/M alone (Fig. 4) was 6% (low). (B) Image from 51-y-old woman with ischemic HF, LVEF of 30%, and BNP of 166 ng/mL. Baseline 2-variable model estimated 13% (intermediate) 2-y risk for composite mortality endpoint. H/M was 1.80. Three-variable model (LVEF, BNP, and H/M) estimated 7% (low) risk. No subject in trial with H/M ≥ 1.80 experienced all-cause mortality.
Download figure
Open in new tab
Download powerpoint
FIGURE 2.
Kaplan–Meier all-cause mortality survival curves based on H/Ms of <1.25, 1.25–1.64, and ≥1.65. Mortality risk is significantly stratified by dividing population on the basis of mean H/M (1.44) ± SD (0.20). Two-year mortality was 3.1% for H/M ≥ 1.65 (n = 147), 11.8% for H/M = 1.25–1.64 (n = 660), and 19.1% for H/M < 1.25 (n = 154).
Download figure
Open in new tab
Download powerpoint
FIGURE 3.
Two-year all-cause mortality rate in relation to H/M intervals. Two-year all-cause mortality rates based on 0.1 increments of H/M show progressive decline from maximum of 29.4% for H/M < 1.10. There were no deaths among 47 subjects with H/M ≥ 1.80.
Download figure
Open in new tab
Download powerpoint
FIGURE 4.
Occurrence of fatal and potentially fatal arrhythmic events (sudden death, resuscitated arrest, ICD defibrillation) (n = 70) in relation to H/M. Peak occurrence was in 1.30–1.39 range, with progressive decline for higher H/Ms. There were no fatal or potentially fatal arrhythmic events among subjects with H/M ≥ 1.80.

Tables

Figures

View popup

TABLE 1

Baseline Clinical Characteristics of Study Subjects

Variable	No. of subjects with data	All subjects	Alive (n = 863)	Dead (n = 101)	P
Age^* (y)	964	62 ± 12	62 ± 12	67 ± 14	<0.0001
Male sex (%)	964	80	80	85	NS
Ischemic etiology (%)	964	66	65	70	NS
NYHA II (%)	964	83	83	78	NS
Ejection fraction^* (%)	964	27 ± 6	27 ± 6	25 ± 6	0.0004
Systolic blood pressure^* (mm Hg)	963	123 ± 19	123 ± 19	123 ± 20	NS
ICD at enrollment (%)	964	19	20	13	0.08
ICD by death or end of study (%)	964	43	44	32	0.02
QRS width^* (ms)	951	120 ± 29	119 ± 29	126 ± 29	0.02
ACEI (%)	964	72	72	71	NS
ARB (%)	964	23	23	22	NS
β blocker (%)	964	92	92	87	0.09
Aldosterone blocker (%)	964	39	39	36	NS
Digoxin (%)	964	24	24	27	NS
Statin (%)	964	67	67	67	NS
Sodium^* (meq/dL)	947	139 ± 3	139 ± 3	139 ± 4	NS
Creatinine^* (mg/dL)	946	1.20 ± 0.37	1.17 ± 0.34	1.41 ± 0.49	<0.0001
Glomerular filtration rate by MDRD^* (mL/min)	946	68 ± 21	69 ± 20	59 ± 24	<0.0001

↵* Data are mean ± SD.
NS = not statistically significant; ACEI = angiotensin-converting enzyme inhibitor; ARB = angiotensin receptor blocker; MDRD = modification of diet in renal disease.

View popup

TABLE 2

Cox Proportional Hazards Analysis for All-Cause Mortality

Model	No. of deaths	Variable	HR	P
Primary list of factors without H/M (n = 964)	101 (10.5%)	Age	1.04 (1.02, 1.05)	<0.001
Primary list of factors with H/M (n = 961)	101 (10.5%)	H/M	0.08 (0.03, 0.24)	<0.001
		Age	1.03 (1.01, 1.05)	<0.001
Secondary list of factors without H/M (n = 926)	97 (10.5%)	Age	1.02 (1.00, 1.04)	0.024
		Log BNP	4.80 (3.10, 7.43)	<0.001
Secondary list of factors with H/M (n = 926)	97 (10.5%)	H/M	0.23 (0.07, 0.76)	0.016
		Age	1.02 (1.00, 1.04)	0.018
		Log BNP	4.05 (2.56, 6.39)	<0.001

Data in parentheses are 95% CI.

View popup

TABLE 3

Multivariate Logistic Regression Model and ROC Curve Analysis for 1-Year All-Cause Mortality

Factor	Significant variable	Odds ratio estimate	ROC AUC	P for AUC difference
Primary list without H/M (n = 899)	Age	1.031 (1.005, 1.058)	0.583 (0.490, 0.677)
Primary list with H/M (n = 898)	Age	1.028 (1.002, 1.056)	0.684 (0.608, 0.760)	0.027
	H/M	0.034 (0.006, 0.186)
Secondary list without H/M (n = 864)	Log BNP	6.049 (3.172, 11.535)	0.740 (0.670, 0.811)
Secondary list with H/M (n = 863)	Log BNP	4.859 (2.476, 9.537)	0.754 (0.687, 0.821)	0.302
	H/M	0.130 (0.021, 0.818)

Data in parentheses are 95% CI.

View popup

TABLE 4

Multivariate Logistic Regression Model and ROC Curve Analysis for 2-Year All-Cause Mortality

Factor	Significant variables	Odds ratio estimate	ROC AUC	P for AUC difference
Primary list without H/M (n = 760)	Age	1.041 (1.021, 1.062)	0.650 (0.587, 0.713)
	Lipid-lowering drugs at baseline	0.574 (0.353, 0.934)
	NYHA classification	1.708 (1.006, 2.899)
Primary list with H/M (n = 759)	Age	1.039 (1.019, 1.060)	0.687 (0.628, 0.746)	0.058
	Lipid-lowering drugs at baseline	0.565 (0.344, 0.928)
	H/M	0.070 (0.020, 0.238)
Secondary list without H/M (n = 731)	Age	1.02 (1.00, 1.05)	0.751 (0.697, 0.805)
	Lipid-lowering drugs at baseline	0.55 (0.33, 0.92)
	Log BNP	5.45 (3.26, 9.12)
Secondary list with H/M	Age	1.03 (1.00, 1.047)	0.757 (0.704, 0.810)	0.475
	Lipid-lowering drugs at baseline	0.54 (0.32, 0.91)
	Log BNP	4.56 (2.67, 7.77)
	H/M	0.19 (0.05, 0.73)

Data in parentheses are 95% CI.

View popup

TABLE 5

Multivariate Cox Proportional Hazards Model for Composite of All-Cause Mortality, Resuscitation, and ICD Defibrillation Using Secondary List of Factors

Model	Number of events	Variable	HR	P
Without H/M (n = 926)	130 (14.0%)	LVEF	0.96 (0.93, 0.99)	0.002
		Log BNP	3.07 (2.14, 4.41)	<0.001
With H/M (n = 924)	129 (14.0%)	H/M	0.24 (0.08, 0.68)	0.008
		LVEF	0.96 (0.94, 0.99)	0.004
		Log BNP	2.56 (1.76, 3.72)	<0.001

Data in parentheses are 95% CI.

View popup

TABLE 6

Multivariate Logistic Regression Model for 2-Year Composite of All-Cause Mortality, Resuscitation, and ICD Defibrillation Using Secondary List of Factors

Model	Significant variable	Odds ratio estimate	ROC AUC
Without H/M (n = 735)	History of hypertension	1.70 (1.08, 2.66)	0.728^* (0.681, 0.776)
	Lipid-lowering drugs at baseline	0.61 (0.39, 0.97)
	LVEF	0.95 (0.92, 0.98)
	Log BNP	3.62 (2.37, 5.52)
With H/M (n = 734)	Log BNP	3.03 (1.96, 4.69)	0.733^* (0.686, 0.779)
	History of hypertension	1.62 (1.03, 2.54)
	Lipid-lowering drugs at baseline	0.613 (0.387, 0.973)
	LVEF	0.956 (0.926, 0.988)
	H/M	0.227 (0.069, 0.743)

↵* P = 0.408.
Data in parentheses are 95% CI.

View popup

TABLE 7

Net Reclassification Improvement Analysis

	Model with LVEF, log BNP, and H/M
Model with LVEF and log BNP	Low probability (<8%)	Intermediate probability (8%–24%)	High probability (>24%)
Subjects with events
Low probability (<8%)	9	4	0
Intermediate probability (8%–24%)	2	50	9
High probability (>24%)	0	3	50
Subjects without events
Low probability (<8%)	180	29	0
Intermediate probability (8%–24%)	71	347	29
High probability (>24%)	0	28	113

Net gain in reclassification (event) = 8/127 = 0.063 (P = 0.059). Net gain in reclassification (no event) = 41/797 = 0.051 (P = 0.001). Net reclassification improvement: 0.114 (P = 0.002). Table includes only subjects with results for all 3 variables.