To the Editor:—
With great interest and appreciation, we read the article of Stendel et al. 1about contrast-enhanced transcranial Doppler ultrasonography (c-TCD) for detection of a patent foramen ovale (PFO) before surgery in the sitting position. Although we agree that this is a significant alternative to the gold standard set by contrast-enhanced transesophageal echocardiography, we would like to make some methodologic remarks.
Contrast-enhanced transcranial Doppler ultrasonography is an indirect approach to detect right-to-left shunting and does not allow for an exact anatomic localization of shunts. Precisely, in the mentioned diagnostic setting, a PFO is likely to cause high-intensity transient signals in c-TCD, but one should be aware that a PFO is not proven by this method. To ensure the diagnosis, contrast medium has to cross the atrial septum following the pressure gradient produced by a Valsalva maneuver during an interval of time that excludes pulmonary passage. Therefore, this interval should start with presence of contrast agent in the right atrium and should not exceed 10 cardiac cycles. Stendel et al. 1allow for 3–15 heart cycles after 5 s injection and 5 s Valsalva maneuver, following the protocol of Schwarze et al. , 2but pulmonary arteriovenous fistulas can be a reason for the detection of high-intensity transient signals during the mentioned interval.
In c-TCD, we frequently observe high-intensity transient signals that meet with the criteria to diagnose a PFO, whereas in contrast-enhanced transesophageal echocardiography, there is no evidence for it. These findings can be explained either by leakage of the capillary lung filter for microbubbles in some individuals or by pulmonary passage of contrast medium parts below standard size, despite correct preparation and handling of the D-Galactose contrast medium.
When c-TCD is positive with Valsalva maneuver, we recommend an additional contrast injection without Valsalva maneuver to exclude pulmonary passage. In the normal heart cycle, an intermittent right-to-left pressure gradient occurs in the early systole because the tricuspidal valve physiologically closes a little earlier than does the mitral valve. Therefore, with c-TCD, even at rest an atrial septal defect or a permanent PFO allowing for spontaneous right-to-left shunting can be detected. Spontaneous right-to-left shunting indicates high hemodynamic relevance of the shunt and may be a valuable finding to select surgical and anesthesiologic procedures to avoid intraoperative paradoxical air embolism.