A bi-center cardiovascular magnetic resonance prognosis study focusing on dobutamine wall motion and late gadolinium enhancement in 3,138 consecutive patients.

To the Editor: In the present study we sought to investigate the predictive value of resting and inducible wall motion abnormalities (WMA) and of late gadolinium enhancement (LGE) for hard cardiac outcomes and for revascularization procedures in 3,138 patients undergoing dobutamine cardiac magnetic resonance (DCMR) in 2 tertiary centers (University Hospital Heidelberg; and German Heart Institute, Berlin) and during a long-term follow-up. From January 2000 through June 2008, consecutive patients who were referred for clinically indicated DCMR due to suspected or known coronary artery disease were prospectively enrolled. Our patient cohort consisted of 1,369 patients enrolled in Berlin and 1,493 patients enrolled in Heidelberg who were reported in previous studies (1,2). Additional patients (n 276) were added in he present analysis, resulting in 3,138 patients in the analysis opulation. Patients were examined in clinical 1.5-T whole-body ardiac magnetic resonance scanners. Wall motion and LGE maging were performed and interpreted as described previously 1,2). Patient preparation, imaging and analysis for DCMR mounted to approximately 30 to 40 min. Cardiac death and nonfatal myocardial infarction were regisered as hard cardiac events. Other events included clinically ndicated revascularization by percutaneous coronary intervention PCI) or coronary artery bypass graft (CABG). Because the results f the magnetic resonance examination might have triggered evascularization procedures, patients with “early” revascularization rocedures within 3 months after DCMR were censored. A eparate analysis was performed in this patient subgroup, to nvestigate the effect of early revascularization procedures on ubsequent cardiac events. Data were collected in 3,138 patients during a 3.3 1.7-year range 0.5 to 9.7-year) follow-up period. Hard cardiac events ccurred in 183 (5.8%) patients, including 80 (2.5%) cardiac deaths n 44 due to intractable heart failure, n 21 due to infarction, nd n 15 who died suddenly) and 103 (3.3%) nonfatal yocardial infarctions. In 520 (16.6%) patients early revascularzation was performed within 90 days after the magnetic resonance xamination (450 by PCI, and 70 by CABG), and 257 (8.2%) atients underwent late revascularization (203 by PCI, and 54 by ABG). Noncardiac death was recorded in 43 (1.4%) patients. By multivariable analysis, inducible WMA, LGE, and resting MA were independent predictors of hard cardiac events (hazard atio [HR]: 6.5, 95% confidence interval [CI]: 4.6 to 9.3; HR: 2.2, 5% CI: 1.2 to 4.1; and HR: 1.6, 95% CI: 1.2 to 2.3, respectively, 0.001 for all). With a series of Cox proportional hazards odels, resting and inducible WMA exhibited incremental value beyond clinical parameters for the prediction of outcome (Fig. 1A). In patients where LGE was additionally performed, resting WMA, LGE, and inducible WMA exhibited incremental value (Fig. 1B). With integrated discrimination improvement analysis, inducible WMA exhibited incremental value in the first model, whereas LGE and inducible WMA showed incremental value in the second model (Figs. 1A and 1B). Patients with inducible WMA experienced a significantly higher rate of hard cardiac events (Fig. 1C), whereas patients with LGE exhibited a higher rate of subsequent hard cardiac events compared with those without LGE (Fig. 1D). In the absence of inducible ischemia by DCMR, clinical outcomes were not influenced either by PCI or CABG (Fig. 1E). Conversely, patients with inducible WMA equally profited from PCI and CABG compared with those continued on medical treatment (Fig. 1F). Within the first 3 years of follow-up excellent outcomes were recorded for patients without inducible ischemia with regard to hard cardiac events and late revascularization procedures (annual hard event rate of 0.6%, and revascularization rate of 1.6%). Between the third and fourth year, however, hard cardiac event rates increased by 2-fold (from 0.9% to 2.0%). Over the third and sixth year, annual event rates rose for both hard events and revascularization to 1.6% and 3.2%, respectively. Our DCMR findings in 3,138 patients demonstrate that: