The use of "hugging balloons" for the implementation of a coronary angioplasty is a well-known and much-written-about technique. It has not, on the other hand, ever been stressed that two inflated hugging balloons are not totally obstructive, leaving enough space, between the balloons and the arterial wall, for perfusion. The cross section of this free area can be as much as 12% of the total arterial cross section, when the side-by-side balloons are of a relatively similar size. Calculations show that similar total conduit cross sections are achieved when inflating either a 3.0 balloon or two 2.0 side-by-side balloons. An in vitro experiment compares how systems using 2.0 double balloons and a 3.0 autoperfusion balloon affect the distal flow of a 3 mm internal diameter conduit. A semi-rigid silastic tubing is perfused with a 5% glucose isotonic solution at various predetermined proximal pressure levels. Distal pressures and flows are measured at different stages: with no balloon, with deflated balloons, and with inflated balloons. Comparison shows that 1) the double balloon system offers little resistance to flow compared to the autoperfusion balloon; 2) this system's rate of perfusion run-off is hardly affected and is constantly better than with the autoperfusion balloon; 3) this difference is all the more marked, the lower the proximal perfusion. These results suggest that the use of a double balloon could, in some cases, provide an alternative solution to autoperfusion equipment. Further clinical studies are needed to test this hypothesis.