There have been many studies conducted on the efficacy of intercessionary prayer and both positive and negative results have been reported.
Methodology is often extremely problematic and many are significantly flawed. It should be noted that it is extremely difficult to run proper double-blind trails (not least because a necessary factor is God, who even if he existed could not be guaranteed to answer any prayers for intercession) and where there is no double blinding, it is impossible to determine the extent to which a placebo effect accounts for the observed effects.
A 2006 meta-study, which analysed the results of previous trials discovered that there was no benefit to patients as a consequence of intercessionary prayer.
http://www.springerlink.com/content/x7rtu32722 145572/
I know of at least one earlier meta-study that initially concluded there was some benefit, but upon further analysis and with the inclusion of more methodologically sound trials found the opposite.
It's a bit like the way that homeopaths claim there are scientific trials that show a benefit to their snake oil - they do exist, but they just happen to be the methodologically flawed ones. Strange that, innit?
In the specific case of the heart patients, it was concluded that the patients regarded the fact that they were being prayed for as indicative of the seriousness of their predicament and thus they experienced a negative reaction.