Transcranial doppler ultrasonography should it be the first choice for persistent foramen ovale screening?

We investigated 420 consecutive patients [(260 females, 160 males); mean age 34.8 ± 16.7 (13, 65) years] admitted to our department with cryptogenic stroke, transient ischaemic attacks (TIA) or other neurological symptoms between 2007 and 2014. Most patients (over 50%) had a history of TIA. The main demographic characteristics of the patients are summarized in Table 1.

All patients underwent TCD and TEE evaluation. The studies were performed on the same day by two different experienced cardiologists, unaware with respect to the result of the other study. The patients were examined in the fasting state. Written informed consent was obtained from all patients. The patients were in left lateral decubitus position during both studies. The studies were performed at resting state and during the Valsalva maneuver after contrast injection. The patients had been trained in performing the Valsalva maneuver before study. The training phase included up to 5 Valsalva maneuver attempts, with at least two resulting in septal shifting or reduced middle cerebral artery flow velocity which was followed by contrast injection and the Valsalva maneuver solicited. The study was repeated at least three times.

TCD and TEE examinations were performed according to a standardized procedure. In brief, 10 mL of air-mixed saline was injected into the right antecubital vein at three different times, while the Doppler signal was recorded during the Valsalva maneuver. During TCD the passage of contrast into the right-middle cerebral artery was recorded 25 seconds after the Valsalva maneuver. A mixture of 9 mL physiological saline and 1 mL air was agitated 10 times in 2 10-mL syringes connected to a 3-way stopcock to exchange the air-saline mix and achieve good dilution. The bolus of saline solution was prepared and injected by the same nurse using the same method in all studies. The patients started the maneuver about 5 seconds after contrast injection, pressing against the closed glottis for at least 10 seconds. They then performed a deep expiration and inspiration, followed by a deep expiration.

A baseline TCD examination was performed with a Toshiba Power Vision echo-machine using a 2-MHz probe, according to standard practice guidelines. Middle cerebral artery flow was monitored through the temporal bone window. The middle cerebral artery was identified with color Doppler in its proximal portion and insonated bilaterally (Figure 1). On TCD study, the effectiveness of the Valsalva maneuver was verified by a reduction of the middle cerebral artery flow velocity, in comparison with the basal spectrum. TCD was considered positive if at least one microembolic signal was recorded on TCD spectrum within 25 seconds from contrast injection. The shunt was defined as small (1, 10 microembolic signals), medium (>10 microembolic signals), or large (>10 microembolic signals with “curtain”).

The patients received local pharyngeal anesthesia with 10% topical lidocaine for TEE study. A TEE study was performed using a Toshiba Power Vision machine with a 5.0-MHz multiplane probe, according to a standard protocol including color flow Doppler data. The atrial septum was analyzed from the transverse mid-esophageal four-chamber view to the longitudinal biatrial-bicaval view (Figure 2). Since PFO was not clearly recognized by a bicaval view at 90° rotation, additional image planes (60°, 90° and/or 110°-130°) were used to better analyze the atrial septum. On TEE study, the effectiveness of the Valsalva maneuver was verified by a reduction in right ventricular and atrial size and by bulging of the atrial septum into the left atrium. To assess PFO, semi-quantification of the right to left shunt (RLS) was performed. On TEE the evaluation was based on counting the number of micro bubbles (MBs) moving from the right atrium to the left atrium through the PFO after the Valsalva maneuver, within the first three cardiac cycles. The studies were considered positive for PFO when at least one micro bubble was observed in the left atrium. The severity of the shunt was quantified as mild (<10 micro bubbles), moderate (10–20 micro bubbles), or severe (>20). If the quantitative results varied between the first and second study the largest number of MBs decided about the shunt size.

The SPSS 13.0 software package was used for statistical analysis. Quantitative data were expressed as the mean ± standard deviation, qualitative data as the percentage. The performance indexes used were sensitivity, specificity, predictive positive value (PPV), negative predictive value (NPV), and k concordance index, which express the agreement proportion beyond chance.

Sensitivity was defined as:

Specificity was defined as: