Transthoracic second harmonic two- and three-dimensional echocardiography for detection of patent foramen ovale

Transoesophageal echocardiography (TEE) with contrast administration is still considered as the reference method for the detection of patent foramen ovale (PFO) with interatrial shunt, but it is a semi-invasive exam. The aim of the present study is to evaluate a role of two- and three-dimensional transthoracic echocardiography (TTE and R3DTE) as a diagnostic alternative to transcranial Doppler ultrasound (TCD) and TEE for detection of atrial right-to-left shunt.

Seventy-five patients with history of cerebrovascular events were subjected to four diagnostic examinations: TCD, TTE, R3DTE, and TEE, with bubble contrast. Bubbles in the left atrium within three cardiac cycles were considered diagnostic for PFO and later as a pulmonary shunt. Greater than 20 bubbles in the left atrium were considered a large shunt and <20 a small shunt. Every exam was read blinded to the results of the others. From the 75 enrolled patients, 62 (82.6%) patients showed right-to-left shunt with TEE; the results were also positive in 53 patients using TCD (70.6%), in 53 using R3DTE (70.6%), and in 55 using TTE (73.3%) (P = NS). There is a statistically significant superiority for TEE in the capacity of detecting shunts compared with TCD (P < 0.024), TTE (P < 0.018), and R3DTE (P < 0.018). The TEE presents a superior ability to recognize mild/moderate interatrial shunts respect to other exams (P = 0.003), without differences for shunts of high degree. In comparison to the TEE, the sensitivity is 89% for TTE, 88% for R3DTE, and 85% for TCD; the specificity is 100% for TTE and R3DTE, and 90% for TCD; the positive predictive value is 100% for TTE and R3DTE, and 98% for TCD; and the negative predictive value is 65% for TTE, 65% for R3DTE, and 53% for TCD. Considering only for mild/moderate shunts, the diagnostic accuracy is clearly inferior (sensitivity 63% for TTE, 58% for R3DTE, and 53% for TCD).

In this cohort of patients, TEE confirms the role of 'gold standard' exam for the detection of PFO; the non-invasive methods, and the TTE in particular, present a good diagnostic accuracy, but are inferior to the TEE because of the low negative predictive value and the non-optimal detection of small shunts. If the only purpose of TEE is the detection of significative interatrial shunt, TEE can be replaced by TTE. The R3DTE presents a good diagnostic accuracy, provides a better anatomical definition of the interatrial septum, and may have a role in this setting of patients, but does not add a lot to the TTE for the diagnosis.