Sir, I read with interest the article entitled An audit of referral practice for patients with impacted palatal canines and the impact of referral guidelines, by my regional consultant colleague Spencer Nute and his co-author T. Hassan (BDJ 2006; 200: 493–496).

They quite rightly pointed out Prevention of an impaction is always preferable to its treatment', but that despite local educational lectures and the dissemination of guidelines in the form of algorithms, the number of patients with impacted maxillary canines who had been subsequently referred to Southend Hospital for assessment and management, both at a more appropriate age as well as having had the deciduous canine previously removed, had only slightly increased, albeit without statistical significance.

Although not directly stated in their report, the implication was that the referring GDPs had been encouraged to perform interceptive extractions of the deciduous canines for their patients before referral, on the basis that previous publications had shown that when undertaken before a mean age of just over 11, between 62%1 and 78%2 of impacted canines would spontaneously recover. They commented that 'Both studies showed that the outcome was dependent on a number of variables including the patient's age', but didn't mention that the most reliable predictor of success was in relation to the unerupted canine's position relative to the lateral incisor root which it was adjacent to.

Indeed, the prospect of success has been found to rise as high as between 73%1 to 91%2 if the crown of the canine has not overlapped the lateral incisor beyond half its root width. However, when it has, the percentages fall to between 29%1 and 64%.2

In that regard, if general dental practitioners are to be encouraged to undertake pre-referral extraction of deciduous canines in appropriate cases, it might be prudent for them to do so with sufficient knowledge as to be potentially more discerning. Otherwise, in those cases where the canine impaction is severe, and therefore less likely to respond favourably to the intervention, the indiscriminate loss of the deciduous predecessor could disadvantage the patient.

For example, in a situation where an impacted canine would be better managed through its surgical exposure, retaining the deciduous canine would not only provide a natural form of space maintenance in the interim before the successional tooth was close enough to be approximated into the line of the arch, but equally should the procedure fail, it would still remain, either to act as a substitute for the permanent tooth, assuming it was in good enough condition, or if not, to retain sufficient alveolar bone for longer, so as to facilitate all future alternative restorative options, such as the use of a single osseo-integrated implant.

Drs Hassan and Nute respond to the above three letters: We thank Drs Spary, Firth and Chate for their interest in our article.

Dr Firth feels that our advice conflicts with advice he received as a student. We would suggest that the evidence base has developed in the intervening 30 years. Recent Royal College of Surgeons evidence-based guidelines3 state that the maxillary canines should be palpable in the labial sulcus 'by the age of 10-11' years, and that 'radiographs prior to the age of 10-11 years are usually of little benefit'. We therefore stand by our assertion that patients should be referred by 12 years of age, as it should almost always be possible to diagnose the problem before then.

We agree with Dr Spary that the study by Leonardi et al. is a valuable contribution to the literature, as it was randomised, included a control group and considered the extra variable of headgear for space maintenance. When undertaking interceptive extractions, it would be advisable to consider using space maintainers, such as headgear, in the future. However, as he will be aware, it can take a considerable time from an article's submission to its publication. This was the case with our article as it was overlooked due to a clerical error at the BDJ. The article by Leonardi et al. was published after we submitted ours.

We disagree with Dr Spary that we suggested 'crowding is not a factor'. We made it clear that Ericson and Kurol4 only treated uncrowded patients. Their high success rate may have been due to adequate space, and so their findings may not be in such contrast to Leonardi et al. who tried to obtain adequate space with headgear.

Dr Spary feels that 'practitioners reading the introduction ... might decide to extract the deciduous canines without asking for ... advice'. If one takes a small section of any article out of context, one may draw incorrect conclusions. We believe Dr Spary's concerns are unjustified if our article is taken as a whole. Our title clearly states that this was an audit of referral practice and the impact of our guidelines. The gold standard clearly states the importance of timely referral. The algorithm sent to the general dentists and reproduced as Figure 2 clearly encourages the referral of patients. Neither the gold standard nor the algorithm advises dentists to perform interceptive extractions. The discussion consists of eight paragraphs covering referral patterns and their modification, and one discussing interceptive extractions. The conclusion reiterated that referral practice was poor and that our guidelines had a limited impact, not that dentists should extract without specialist advice.

We briefly discussed the literature on interceptive extractions to highlight why orthodontists want referrals at the correct age. As we were not auditing the efficacy of interceptive extractions, a detailed discussion of the procedure would not have been directly relevant. Indeed, had we done this, readers may have gained the impression that we were educating them to perform interceptive treatment without specialist advice: exactly the opposite of what we, Dr Chate and Dr Spary would wish to do.

We collected data on the absence of primary canines, as some patients are referred by primary care specialists who should be aware of best practice. These patients may have been referred later because appropriate interceptive extractions were tried unsuccessfully. It could have been unfair to the referring practitioners to assume that all 'late' referrals were due to poor management.

We carried out this project and article to encourage timely referral. This allows orthodontists and patients to have an informed discussion and decide upon the best course of action. We do not encourage general dentists to undertake treatment without specialist advice. If, like Dr Chate and Dr Spary, some readers found this aspect of our article ambiguous, then we thank them for raising the issue so that we could clarify it.