Transcranial Doppler sonography (TCD) can detect the passage through the middle cerebral artery of microbubbles present in intravenous injected saline when an intracardiac right-to-left shunt occurs. Given the increasingly recognized importance of patent foramen ovale (PFO) as a possible anatomical factor predisposing to stroke, we tried to assess the efficiency of TCD in the evaluation of PFO in comparison to the gold standard represented by transesophageal echocardiography (TEE). Seventy-two consecutive patients scheduled for TEE also underwent TCD, simultaneously in 18 cases. A 20-ml agitated saline bolus was injected into an antecubital vein through an indwelling catheter while the patient was performing a Valsalva manoeuvre. The procedure was repeated a maximum of 10 times or until the contrast medium was tracked from the right to left atrium (TEE+) and/or the high-intensity spike typical for embolism was recorded on TCD (TCD+). Among non-simultaneous cases, the first 14 were studied with a non-standardized protocol: the resulting sensitivity and specificity of TCD as compared to TEE were both 43% (TEE+/TCD+ = 3; TEE–/TCD– = 3; TCD+/TEE– = 4; TCD–/TEE+ = 4). In all the 18 simultaneous examinations, however, there was 100% agreement between TEE and TCD (TEE+/TCD+ = 12; TEE–/TCD– = 6). Moreover, when the protocol was standardized in 40 further patients studied non-simultaneously, sensitivity and specificity were 90 and 100%, respectively (TCD+/TEE+ = 19; TCD–/ TEE– = 19; TCD–/TEE+ = 2; TCD+/TEE– = 0) Standardization was obtained by timing of Valsalva manoeuvre to injection, delay of embolic signals, caliber of venous catheters, content of air bubbles in the saline, velocity of injection. In conclusion, when properly standardized techniques are applied, TCD can be used as a screening test for the detection of PFO.