Purpose: Single-center experiences have shown that myocardial meta-iodobenzylguanidine (mIBG) uptake has prognostic value in heart failure (HF) patients. To verify these observations using a rigorous clinical trial methodology, a retrospective review and prospective quantitative reanalysis was performed on a series of cardiac (123)I-mIBG scans acquired during a 10-year period at six centers in Europe.
Methods: (123)I-mIBG scans obtained on 290 HF patients [(262 with left ventricular ejection fraction (LVEF) < 50%)] from 1993 to 2002 were reanalyzed using a standardized methodology to determine the heart-to-mediastinum ratio (H/M) on delayed planar images. All image results were verified by three independent reviewers. Major cardiac events [MCEs; cardiac death, cardiac transplant, potentially fatal arrhythmia (including implantable cardioverter-defibrillator discharge)] during 24-month follow-up were confirmed by an adjudication committee.
Results: MCEs occurred in 67 patients (26%): mean H/M ratio was 1.51 +/- 0.30 for the MCE group and 1.97 +/- 0.54 for the non-MCE group (p < 0.001). Two-year event-free survival using an optimum H/M ratio threshold of 1.75 was 62% for H/M ratio less than 1.75, 95% for H/M ratio greater than or equal to 1.75 (p < 0.0001). Logistic regression showed H/M ratio and LVEF as the only significant predictors of MCE. Using the lower and upper H/M quartiles of 1.45 and 2.17 as high- and very low-risk thresholds, 2-year event-free survival rates were 52% and 98%, respectively. Among patients with LVEF < or = 35% and H/M > or = 1.75 (n = 73), there were nine MCEs because of progressive HF and only one because of an arrhythmia.
Conclusion: Application of a clinical trial methodology via the retrospective reanalysis of (123)I-mIBG images confirms the previously reported prognostic value of this method in HF patients, including potential identification of a quantitative threshold for low risk for cardiac mortality and potentially fatal ventricular arrhythmias.