ORLANDO — The best shunt to use in the Norwood operation, during the first stage of repairing hypoplastic left heart syndrome, remains a toss-up despite completion of a yearlong randomized trial with more than 500 infants.
“What we see at 12 months is a survival advantage” for a right ventricle-to-pulmonary artery shunt (RVPAS) compared with the alternative, modified Blalock Taussig shunt (MBTS), Dr. Richard G. Ohye said at the annual scientific sessions of the American Heart Association.
“The concern is that the [survival] curves begin to converge, and in the future will they remain parallel or cross?” As a result of this uncertainty about long-term survival, “concrete recommendation will have to wait for further follow-up,” said Dr. Ohye, director of pediatric cardiac surgery at the University of Michigan, Ann Arbor.
The MBTS has been the traditional option during the Norwood operation, but concerns about its safety focused on the retrograde coronary flow it allows that could potentially interfere with coronary perfusion and lead to ischemia and possibly death. The other option, the RVPAS, irequires a ventriculotomy, which could compromise right ventricular function and might also trigger arrhythmias.
Uncertainty over which shunt produced the best outcomes led to the Single Ventricle Reconstruction Trial, done through the Pediatric Heart Network at 15 U.S. sites. It randomized 555 infants scheduled for the Norwood operation to either of the two shunt strategies.
One year after randomization, the incidence of death or need for heart transplant was 26% in patients getting the RVPAS and 36% in those getting the MBTS, a statistically significant 10% absolute difference in the primary end point in favor of the RVPAS. Although this was the primary end point, it did not tell the entire story.
Follow-up continued and after an average of 32 months, 16 additional deaths or heart transplants occurred in the RVPAS group compared with 7 of these events in the MBTS infants, a trend that led to the observation that the event curves may be converging over time.
Another worrisome finding was that unintended cardiac procedures such as balloon dilatations of the shunt or neoaorta, shunt revisions, or unplanned pulmonary artery reconstructions, were significantly more common in the RVPAS infants, 54%, compared with those who received MBTS, 44%. The RVPAS also produced smaller pulmonary arteries by the Nakata index.
By most other criteria, the two procedures produced similar outcomes. Time to extubation during surgery, duration of ventilation, and total days spent in the ICU and in the hospital were identical, as was the percentage of infants who required an open sternum or extracorporeal membrane oxygenation. The incidence of nonfatal serious adverse events was similar in the two arms, as was long-term right ventricular function. Infants treated with a RVPAS had the advantage of a significantly reduced need for cardiopulmonary resuscitation, 13%, compared with 20% in the MBTS infants.
“Although 12-month, transplant-free survival is higher with RVPAS, the emergence of later mortality with RVPAS is of concern. Continued follow-up of the cohort will be important to determine intermediate and long-term outcomes,” Dr. Ohye said.
Vitals
Major Findings: At 1 year, infants who received the RVPAS had a significant1absolute reduction in death or need for transplant vs. those who received the MBTS; after 1 year that rate was higher in the RVPAS patients.
Source of Data: The Single Ventricle Reconstruction Trial, involving 555 infants.
Disclosures: The National Heart, Lung, and Blood Institute sponsored the trial. Dr. Ohye had no conflicts.