Sir, the recent paper by Johnston and Littlewood (BDJ 2015; 218: 119–122) admirably summarises contemporary orthodontic retention regimes, but in doing so reveals that in the past 30 years there has been little progress in our understanding of why almost all cases relapse to some degree, even after prolonged retention.

While the common reappearance of lower incisor crowding is not always noticed by the patient, it is frequently accompanied by less acceptable reflected changes in the upper arch. As the authors point out it seems very likely that the reappearance of lower arch crowding is due to growth-related forward migration of the buccal segments.1 How odd then that it is now regarded as unacceptable for orthodontists to leave even small residual premolar extraction spaces at the end of fixed appliance treatment, when the evidence is that to do so will preserve labiolingual incisor alignment in the lower arch for many years until these spaces finally close?2

Less is understood about the cause of relapse of corrected rotations. As Johnston and Littlewood state this is thought to be due to the stretching of transseptal and supracrestal periodontal fibres which then try to return the tooth to its original rotated position. But why, in this area of very high cellular and collagen turnover, does the rapid replacement of these fibres not retain the tooth in its new position rather than cause its relapse? If these obdurate fibres are indeed the cause, why is it that the once popular surgical procedure of 'pericision' (circumferential supracrestal fiberotomy), be it undertaken by scalpel or YAG laser only reduces, rather than eliminates the relapse? Nevertheless, it seems this support mechanism must be implicated since, given adequate space, emerging rotated lower incisors correct spontaneously until they are half erupted and the gingival attachment becomes established.3

Francis Bacon observed that 'nature is often hidden, sometimes overcome, seldom extinguished',4 and that 'where the cause is not known the effect cannot be produced'.5 I submit that the adoption of semi-permanent retention should be regarded only as a pragmatic temporary solution to this intractable problem, for when a lingual bonded retainer fails it is often at a single tooth which the patient fails to notice until significant relapse has occurred. Surely it is incumbent on our speciality to continue to research this area to clarify the underlying causes of relapse in its various forms and devise more satisfactory solutions?