ASNC: Myocardial Scarring on PET Predicts Response to Resynchronization Therapy

SAN DIEGO, Sept. 14 -- Positron emission tomography (PET) can identify patients with ischemic heart failure who are most likely to benefit from cardiac resynchronization therapy (CRT), investigators reported here.

The PET-determined count of scarred myocardial segments and total scar score predicted the likelihood of response to CRT, DaLi Feng, M.D., of the Mayo Clinic in Rochester, Minn., reported at the American Society of Nuclear Cardiology meeting here.

"By [receiver operator curve] analysis, a PET total scar score of 19 yielded the best sensitivity and specificity for predicting CRT response in ischemic heart failure patients," said Dr. Feng.

However, he noted, "Additional studies in large patient groups are needed to confirm these findings."

CRT improves symptoms, exercise capacity, left ventricular ejection fraction, and survival in patients with depressed left ventricular ejection fraction and ventricular dyssynchrony. However, as many as 30% of patients treated with CRT do not derive a benefit, and the reasons for nonresponse are not completely understood, said Dr. Feng.

PET is considered the gold standard for assessment of myocardial viability, Dr.Feng noted, but no published study has examined PET's ability to predict response to CRT.

Investigators prospectively evaluated 34 patients with ischemic heart failure who had had PET imaging prior to CRT. Various patients underwent PET studies with FDG and ¹³NH₃, FDG and Rb-82, and rest/stress perfusion with ¹³NH₃ or Rb-82.

The baseline New York Heart Association heart failure classification was 3.3. Left ventricular ejection fraction averaged 22.7%, and the mean QRS duration was 154 msec. Thirty patients completed six months of follow-up, 15 responded to CRT.

There was no difference between responders and nonresponders in baseline demographic and clinical characteristics.

Responders had significantly more viability segments (12 versus 8.8, P=0.02), fewer scar segments (4 versus 7.2, P=0.02), and lower total weighted scar score (13.6 versus 23.8, P=0.01).

In contrast to previous studies, however, the presence of lateral wall scar and total lateral wall scar score did not predict response to CRT.

"This could be because of the small sample size and insufficient power to detect small to moderate differences in the lateral wall scar between the two groups," said Dr. Feng.