Real-World Registry Study Confirms MADIT 2 Results

Fran Lowry

September 02, 2009

September 2, 2009 (Barcelona, Spain) — A registry study done in Germany has shed light on how implantable cardioverter-defibrillators (ICDs) are used in the real world outside of the rarefied atmosphere of well-controlled, randomized clinical trials.

Presented at the European Society of Cardiology (ESC) 2009 Congress, data from the Prevention of Sudden Cardiac Death II Registry (PreSCD Registry) showed that patients at high risk for sudden cardiac death who received an ICD one month or more after suffering an MI had a 44% lower mortality than their counterparts who did not get an ICD (p=0.053), Dr Heinz Völler (Klinikum am See, Ruedersdorf, Germany) reported.

In the sickest patients--those with a severely reduced ejection fraction of 30% or less--mortality was reduced by 47%, and this was of borderline significance (p=0.07). Patients who were implanted 11 months or more after their index MI fared the best by far. In those patients, late ICD implantation reduced mortality by 80%. On the other hand, patients who were implanted just one to three months post-MI had a twofold increase in mortality.

The registry was initiated shortly after the release of the MADIT 2 trial results in December 2002 and screened 10 612 patients one month post-MI until May 2005. Of this number, the PreSCD Registry included data on 2068 patients with 36 months of follow-up. The majority of these patients (53.7%) had an LVEF greater than 40% post-MI; 33.7% had an LVEF that ranged from 31% to 40%; and 12.6% of patients had an LVEF below 30%.

"MADIT II gave us proof of efficacy, but we had no idea of the effectiveness in clinical practice, and we felt it was important to prove the results in the real world," Völler told heartwire .

A German Paradox

Paradoxically, in a country that has the highest ICD implantation rate in Europe, very few patients got an ICD. Among 259 patients who had the lowest ejection fractions, only 57 (22%) received an ICD in the first four months after their MI. In the next three years, only 25 more patients (10%) received an ICD. In all, a total of 142 ICDs were implanted.

These data were disturbing to Dr Panos Vardas (Heraklion University Hospital, Greece), who formally discussed the paper.

Vardas, who chaired the task force that developed the 2007 ESC guidelines for cardiac pacing and cardiac resynchronization therapy, asked why there was such a low rate of implantation in guideline-eligible patients. "Germany is a country with a high rate of implantations; more than 200 units per million are implanted. So we have to ask why the results showed such low rates of implantation in a country that has the highest rate of ICD implantations in Europe."

Vardas also took issue with the fact that the registry did not provide any information about outcomes in subgroups of patients such as those with a QRS duration of more than 120 ms and also failed to provide data on cardiac mortality, presenting only total mortality rates.

Vardas told heartwire : "The registry is weak in my opinion. It did not analyze subgroups with special importance, it did not analyze cardiac mortality issues, and it does not explain the low rates of implantations. This is indeed a paradox in a country that has such a high implantation rate."

He hypothesized that this low rate could be due to the fact that the registry followed patients from rehabilitation centers and not academic, tertiary-care institutions. "The patients might have been coming from areas within Germany that have low rates of ICD implantation. It would appear that, even in Germany, there are significant regional differences in the ICD-implantation rate," he said.

It is also possible that the doctors did not follow the guidelines at this particular point in time, he added.

Völler told heartwire that he tried to differentiate between cardiac and all-cause mortality but in one-third of the deaths, the cause was unknown. "Therefore, I could not present these data."

But he did note that the registry may have been initiated too early after the MADIT trial for the results to have filtered down to community clinicians. "It is possible that patients, and especially the cardiologists and the rhythm centers, were not aware of the MADIT trial results, and that may have been why the implantation rate was so low. It may be that regional distribution differences exist in terms of ICD implantation."

To access complete presentation slides (pdfs), click here.

Völler disclosed a financial relationship with Boston Scientific, which sponsored the study.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.