Operative Strategies to Reduce Cerebral Embolic Events During On- and Off-Pump Coronary Artery Bypass Surgery: A Stratified, Prospective Randomized Trial
In this study the authors randomized 142 patients within two separate treatment arms: (1) Patients undergoing on-pump CABG were randomized to either a single- vs. double-clamp technique for proximals; (2) patients undergiong OPCAB were randomized to either a Heartstring device or single clamp for proximals. For all 4 treatment groups, the number of embolic HITS was assessed with transcranial doppler of the middle cerebral artery.
Results:
(1) On-pump CABG: The mean number of HITS in the single-clamp group was 10.0, while that for the double-clamp group was 16.0 (p=.10). (Although the difference was not significant, the treatment groups were rather small.)
(2) OPCAB: The mean number of HITS in the Heartstring group was 30.0, whereas that for the partial clamp group was 7.0 (p<.0001). Thus, the Heartstring group had more than 4 times the embolic load by TCD.
Notably, neurocognitive outcomes were not different among the groups.


To me, the answer to this issue is simple. In patients who do not have pre-existing thrombus in their ventricle from an acute MI, an off pump operation with a LIMA-Radial T graft +/- an in situ RIMA, or LIMA-RIMA T graft, should have a stroke rate of 0%. Period.
The article does not mention whether a blower-mister was used with the Heartstring proximals. I believe that causes gaseous emboli.