Enthusiasm for new therapeutic modalities frequently results in a ground swell of overuse, which recedes to leave a trough of negativism before a definitive level is established. In the case of CABG, the ground swell has become a 10-year flood tide, without an ebb, although ripples have marred the surface. It is generally agreed that suitable control populations are needed to determine the role of a therapy and that the best controls would arise from randomized prospective trials. At present, the best randomized trial is the VA study which has shown that operation does not enhance longevity for single vessel disease, but clearly does for left main coronary disease. In those individuals with two- and three-vessel disease, the VA study has not shown that surgical therapy is better than medical treatment. However, three other studies containing patients selected by the VA criteria indicate that operation provides significantly greater longevity for patients in these two categories. Acceptance of this position requires acknowlegement of the validity of the VA study and the comparison of extramural concurrent surgical series.