OBJECTIVE Transesophageal echocardiography (TEE) and transcranial Doppler (TCD) are the methods of choice to study patent foramen ovale (PFO), but there are discrepancies between the 2 concerning PFO detection. No study has analyzed right-to-left shunt (RLS) quantification concordance. The 2 methods are carried out in different hemodynamic states, and the Valsalva maneuver (VM) required in each also differs. The authors compared PFO detection and concordance of RLS quantification classifications performing the 2 studies simultaneously. METHODS The authors prospectively included consecutive stroke patients undergoing TEE and applied the TCD protocol of the Consensus Conference. Echocardiographic PFO was diagnosed when at least 3 microbubbles (MBs) were detected in the left atrium within 3 heartbeats after opacification of the right atrium. RLS quantification was (1) TCD: minimum (1-10 MBs), moderate (11-25 MBs), and massive (>25 MBs) and (2) TEE: small (3-10 MBs), moderate (11-30 MBs), and large (>30 MBs). METHODS contingency tables (chi(2) and K test). Results. The authors studied 110 patients whose mean age was 56.7 +/- 12.1 years, and 60.9% were men. PFO was detected at the first VM in 30% of patients with TCD and in 31.8% with TEE. At the second VM, both methods detected the same patients (32.7%). RLS was minimum (14), moderate (5), and massive (17) in TCD and small (13), moderate (3), and large (20) in TEE. There was an almost perfect concordance in RLS quantification (K = 0.928, P = .001), with only 4 discrepancies. CONCLUSIONS Simultaneous study with TCD and TEE showed an almost perfect concordance in PFO detection and RLS quantification.