Effects of posture on right-to-left shunt detection by contrast transcranial doppler.

BACKGROUND AND PURPOSE There is controversy about the optimal patient position for the detection of right-to-left shunt (RLS). The study was performed to investigate which patient position best detects RLS during contrast-enhanced transcranial Doppler. METHODS We prospectively evaluated consecutive patients with ischemic stroke or TIA referred to our Noninvasive Cerebrovascular Laboratory for suspected paradoxical embolism. The standard protocol for RLS detection recommended by the International Consensus Criteria was followed. Each patient was examined at rest and after Valsalva maneuver in 4 positions: supine, right lateral decubitus, right lateral leaning, and upright sitting, in random order. RLS was graded 0 (no microbubbles [mB] detected), 1 (1-10 mB), 2 (>10 mB but no curtain), and 3 (curtain, shower of mB). Blood pressure, heart rate, and neurological symptoms were monitored. Data were analyzed using SPSS version 17. RESULTS RLS was detected in at least 1 position in 89 of 240 patients (37.1%; 95% CI, 33.1%-43.3%). The detection of at least 1 mB with normal breathing was lowest in supine position and highest in right lateral decubitus. With Valsalva maneuver, this was highest in upright sitting (20.4% versus 8.3%; P<0.0002). If mB were undetected on upright sitting position, then they may still be detected in other positions. Changes in the position of the body and the injection of agitated saline were well-tolerated. CONCLUSIONS RLS is best detected in the upright sitting position with Valsalva maneuver. If negative, then other positions may be used. Validation of our findings by other centers may be helpful.