A cadmium telluride nuclear probe with an Elscint gamma camera was used to detect and measure left to right intracardiac shunts at the bedside in 34 patients. Fifteen also had right heart catheterisation and oximetric measurement of the shunt. For the nuclear technique 740 MBq (20 mCi) of technetium-99m pertechnetate was injected into the right antecubital vein and the pulmonary to systemic flow ratio (QP:QS) was measured by the gamma variate technique. Data were not obtained in four patients because the nuclear probe failed in three patients and one storage disc was corrupted. Data from the gamma camera were lost in another patient. When the size of the shunt measured by the nuclear probe was compared with that measured by the oximetric technique the mean difference (SD of mean difference) was 0.36 (SD 0.78) and when it was compared with the gamma camera it was 0.08 (SD 0.67). Analysis of scatter plots showed that the larger the shunt, the larger the discrepancy. Twenty four of 29 data sets showed complete agreement between the nuclear probe and gamma camera on the size of the shunt. Any differences were small. These data indicate that left to right intracardiac shunts may be measured accurately by a nuclear probe at the bedside in either the coronary care unit or outpatient department.