I am not an expert in imaging but from experiences so far:-

TEE are useful to identify anatomy of the PFO to aid in closure planning. If the shunt is not significant (>1.5) and there are no symptoms along with patient's preference to not do an invasive procedure ; then no TEE is performed.

We have seen many instances when the margins of PFO or when PFO are associated with big anatomical defects, where interventional procedures are not recommended. TEE would be recommended in those instances. 

I ketty,

according to my neurological experience, what happen with those patients could be actually summarized in two scenarios:

1. the patient is discovered to have a PFO after a TTE for any cardiological reason; then he's referred to a neurologist  

2. the patient is followed-up by a neurologist for any kind of cerebrovascular disease and PFO is suggested by shunt signs at TCD or by a TTE as part as the routine workup of cerebrovascular patients (you may consider also that they may have other reasons to perform a TTE during hospitalization, before a neurologist put a probe over his head...)

So it seems quite difficult to establish a precise sequence for everyone. Ideally, to me, it would be: TCD first, then TTE and finally TEE. TEE is useful:

- as Habib said, mainly to eventually establish surgery

- for TCD/TTE inconsistency: when TCD is positive and TTE is negative it might be that TEE comes out positive.  

One important point: TCD is useful as the first screening, from a neurological perspective, because tells you if there's a right-left shunt, accounting also for those, though rare, patients with pulmonary fistulas and no PFO.

In case of minimal shunt and absence of symptoms only TTE.In other cases TEE to establish anatomy and optimal management

In most cases one looks for a PFO if suspicion of right-to-left shunt (cryptogenic stroke, orthodexia-platypnoe syndrome). One pays a little bit more attention in young patients (below the age of 55-60 years) if PFO closure is eventually part of the treatment plan for cryptogenic stroke. Transcranial doppler or TTE with contrast and Valsalva are usefull screening tools to detect R-L shunt of any origin. If positive or if high clinical supicion a TEE is warranted to confirm the diagnosis and delineate anatomy in those patients in whom PFO closure might be indicated.

Dear Ketty,

in our cardio-Neurological Centre we use this sequence:

1. the patient who discoveres to have a PFO after a TTE for any cardiological reason, is referred to a neurologist  

2. the patient who is valueted by neurologist for any kind of cerebrovascular disease  and discovers to have cardioembolic lesions of unknown origines (cryptogenic signs) is referred to our Centre to do TTE+TCD Bubble Test with Valsalvometer (to optimize the manoevre ) (the shunt Dx-->Sn must be important, more than Grade III-V /5 of Spencer Grade in TDC or 2-3/3 in TTE)

3. Evidence of thrombophilic alterations with: Fatt II, Fatt V of Leiden and MTHFR

4. 3D-TEE for anatomical pattern of PFO or measures of a Functional ASD, only if the patient is referred to the Percutaneus Closure! Otherwise is not necessary