Prior to going on active duty as a military surgeon in 1944, Adrian Kantrowitz served a nine-month internship at the Jewish Hospital of Brooklyn, where he worked with prominent neurosurgeon Leo Davidoff. He impressed Davidoff by inventing a special surgical clamp for craniotomies, and got the surgeon's tentative approval to continue his training there after the war. By the time he was discharged in 1946, however, Davidoff had no residency openings. He advised Kantrowitz to pursue training in general surgery, which he did at Mt. Sinai Hospital and at Montefiore Hospital in New York. He was quickly drawn to the very new field of cardiac surgery.

Cardiology had developed into a distinct specialty before World War II, greatly aided by tools such as the electrocardiograph (EKG) and cardiac catheters, which allowed physicians to measure pressures within the heart chambers. By contrast, cardiac surgery was just getting started, owing largely to the difficulty of interrupting the heart's operation long enough to repair it, without risking brain damage. The limited amount of work being done on hearts was usually done by thoracic surgeons, who worked on the main blood vessels or the pericardium rather than the heart itself. But the boundaries of thoracic surgery were expanding rapidly during this period, often with dramatic results. In 1944, Alfred Blalock and Helen Taussig at Johns Hopkins had devised an arterial shunt operation that prolonged the lives of many "blue babies" born with heart defects. Dwight Harken's wartime success at getting pieces of shrapnel out of still-beating hearts inspired him to use the technique (after many animal trials) to correct mitral valve stenosis, a common outcome of rheumatic fever. (Prior to the advent of antibiotics, starting in the late 1930s, rheumatic fever often followed streptococcal infections such as strep throat and scarlet fever. This caused scarring and narrowing of the heart valve openings, particularly the mitral valve through which blood flows from the left atrium to the left ventricle. With the flow of oxygenated blood from the lungs thus seriously reduced, patients were often chronically short of breath and effectively disabled. Over a million Americans suffered from rheumatic heart disease at mid-century.) In 1948, Harken, working in Boston, Charles Bailey in Philadelphia, and Russell Brock in London independently carried out successful mitral commissurotomies to re-open the constricted valve. With the heart still beating, they made a small incision in the left atrium and, working by touch alone, opened the stenosed valves with a finger, sometimes fitted with a small blade. This was a "closed heart" procedure, carried out by opening the left chest cavity to gain access. Though risky, especially in the early years, there was soon great demand for the operation; hospitals everywhere were pressured to offer it. Meanwhile, some surgeons, e.g., John Gibbon in Philadelphia and Clarence Dennis in Brooklyn, were investigating ways to maintain circulation artificially with pump oxygenators. Others, like Wilfred Bigelow in Toronto, believed that controlled hypothermia might be used to decrease metabolic rate so that the heart could be slowed or stopped for a few minutes without danger to the patient.

Kantrowitz, observing Bailey's closed heart commissurotomy, thought that the "blind" approach seriously compromised the outcome. He resolved to find some way to open this part of the heart to the surgeon's view. A heart-lung machine, which could both oxygenate and circulate the patient's blood, would be ideal, but the early prototypes of such devices tended to leak, damage blood cells, and generate embolisms. Reasoning that only the left side of the heart needed to be bypassed to work on the mitral valve, Kantrowitz developed a mechanical left heart to do this. Assisted by his brother Arthur, then a physicist at Cornell University, Kantrowitz devised a system that delivered normally oxygenated blood from a cat's left lung into a glass chamber and then used a modified roller-type pump to route it back into the arterial tree via the femoral artery. With the right pulmonary artery temporarily tied off to keep blood from filling the left side of the heart, the mitral valve could be exposed. In 1951, Kantrowitz filmed this same procedure done on a dog, and showed the film to 350 colleagues at the New York Academy of Medicine, providing their first view of a mitral valve functioning in situ. This left-heart bypass technique was never adopted for mitral valve surgery in humans (within the next two years, surgeons would have some success doing open heart surgery with heart-lung machines or with hypothermia) but it established Kantrowitz as a promising young surgeon and researcher.

The left-heart bypass research also got Kantrowitz thinking about helping patients with diseased, rather than congenitally defective hearts. Congenital structural defects of the heart and disabling scars caused by rheumatic fever often could be surgically remedied once workable circulatory support was available; but many other patients suffered from failing hearts due to blockage of the coronary arteries (which supply blood to the heart itself), damage from previous heart attacks, or other conditions. Many investigators had explored ways to increase the blood supply to the myocardium (heart muscle), especially the crucial left ventricle, which pumps oxygenated blood from the lungs out to the body through the aorta. However, during systole, when the ventricles contract, the coronary arteries are squeezed, making them somewhat resistant to the blood flow. Kantrowitz wondered whether blood could be delivered to the coronaries during diastole, when the heart relaxed, and the coronaries were open to greater flow. To pursue this idea of "diastolic augmentation," Kantrowitz spent a year as a research fellow studying cardiac physiology with renowned physiologist Carl Wiggers at Western Reserve University School of Medicine in Cleveland. He became well versed in hemodynamics, and designed an experiment to test the efficacy of diastolic augmentation, working on dogs. The dog's blood was routed through a delay line from the femoral artery, and then the left coronary artery was filled with blood pulsed from either the aorta or with a delayed pressure pulse from the femoral artery. Kantrowitz was able to demonstrate that introducing the delayed pulse during diastole could increase the coronary blood flow by 10 to 15 percent. The potential benefits of diastolic augmentation or "counterpulsation" were clear, and he would spend much of his career developing devices that could aid failing hearts in this manner.