Detection of intracardiac left-to-right shunting in adults: A prospective analysis of the variability of the standard indocyanine green technique in patients without shunting

Intracardiac left-to-right shunting may be detected and quantitated by an oximetric analysis of blood from the right-sided cardiac chambers and prominent early recirculation of indocyanine green after it is injected into the central venous circulation and sampled from a systemic artery. Although the variability of oximetric measurements has been established in patients without shunting and a range of normal has been determined for the technique, the variability of the indocyanine green method among persons without shunting has not been clarified; as a result, a range of normal for indocyanine green has not been established. In 66 adult patients in whom hydrogen inhalation (an extremely sensitive technique for detecting even very small intracardiac left-to-right shunts) revealed no shunting, indocyanine green curves were generated by injection into the pulmonary artery as blood was sampled from a systemic artery. In these patients the percentage left-to-right shunt (that is, the percentage of pulmonary blood flow reaching the lungs through an intracardiac shunt) (determined with the equation of Carter et al) ranged from −9% to +26% (+7 ± 8%, mean ± standard deviation). There was no definable relation between the percentage left-to-right shunt and the indicator dilution measurement of cardiac output. Thus, these data establish a range of normal for the indocyanine green technique of detecting and measuring intracardiac left-to-right shunting. If this technique is to be used reliably to detect shunting, its results must demonstrate a percentage shunt in excess of +26%.