While EMDR has been evidenced as effective in neutralizing trauma and installing new cognitions, it accesses primary processes of experience and that's why orientation and client preparation are essential to safeguard the client from flooding experiences! As such, it's indirectly capable of being "invasive" in that way.
Although there is robust evidence for the effectiveness of EMDR as a treatment for PTSD, to dogmatically state that (1) it is the least invasive treatment for patient and therapist, and (2) should be first choice, seems (intensionally?) provocative, even insensitive and unscientific, reminiscent of the days when ECT was the first choice of psychiatric treatment. Would it not be better to argue for the prior establishment of a beneficial therapeutic relationship, as well as careful, insightful, clinical evaluation in terms of inclusion and exclusion criteria for EMDR, such as the availability of a therapeutic milieu, to minimize the likelihood of intense emotional reprocessing and seizures, before starting EMDR with persons with PTSD?
This propaganda statement is not entirely true; it can be effective - if well done; however I see more and more patients, who got some EMDR and they have reacted with dissociation; the not well trained clinician does not recognise that and thinks hge is at least not intrusive (dissociation is - per se- mainly silent!). So I would say, EMDR is efficient - with good diagnosis and a clear goal for applying the technique. And it stays a technique, however potent it is; a hammer is a very potent tool - for nails and for heads- the question is ecology!