Sir, we were surprised to see the paper Partial Caries Removal and cariostatic Materials in carious Primary Molars: A Randomised Controlled Clinical Trial published (BDJ 2004, 197: 697–701) as the results were so poor. In this article the authors noted similar failure rates for all three techniques which led them to the conclusion that partial caries removal and restoration was comparable to conventional restorations. There are a number of issues which can be discussed. The overall failure rate for all three methods of restoration was very high. Out of some 86 restorations available for follow–up after 24 months 31 nearly 40% had failed.

Normally follow–up for such studies is 48 to 60 months to compare with previous similar studies1. Also, significantly more PCC/GIC and PCC/BCC failed in the first six months compared with CR. In addition some 62% of class II restorations had failed. No doubt the failure rate would have been even higher for a full 60 months study.

These failure rates are unacceptable for restorative treatment in children and would probably have been even higher had the authors included the teeth that had abscessed, or were subsequently extracted, rather than curiously withdrawing them from the study. Abscess formation is a sign of failure and these teeth should be included with the failures reported in the final analysis. Also, it is unclear how many of the children lost to follow–up might have undergone extractions of the restored primary teeth under local or general–anaesthesia by other dentists. Recent studies have shown more acceptable survival rates for restorations in primary teeth, but where restorations were performed using local analgesia, proper isolation and following the well established principles of caries removal and cavity design 2,3.

It is not surprising that the authors found that placing a material in an unprepared cavity leads to failure of the restoration in a high proportion of the cases. We should not accept such high failure rates of restorations in primary teeth when such results would have been totally unacceptable for permanent teeth. Repetitive treatment in children has implications for a child's behaviour and emphasis should be on restorative materials and techniques that seldom fail, such as stainless steel crowns4. In addition, due consideration should be given to the diagnosis of the state of the pulp especially for proximal caries. In primary teeth pulp inflammation sets in early preceding the exposure of the pulp5.

In our opinion this is yet another article in the British Dental Journal reporting on techniques for restoration of primary molars that circumvent the basic principles of restorative dentistry, which would never be considered for the restoration of permanent teeth. A high failure rate of restorations placed in primary teeth of children where the state of the pulp is not taken into account, longevity of restorative materials not considered or principles of cavity design ignored, has given rise to the myth that restorative treatment for primary teeth is futile. This is not true. Excellent success rates have been reported and following some of the principles mentioned above it is possible to achieve excellence in restorative treatment for primary teeth, which children who still get dental disease, deserve from dental health professionals.

Authors of the paper, J Foley, DJP Evans and A Blackwell respond: We refer to our recent publication (BDJ 2004; 197: 697–701) and to Professors Duggal and Curzon's letter in response to our study. Firstly, in relation to the failure rate, whilst the authors acknowledge the high failure rate of this study, this is primarily attributed to the poor performance of the PCR:BCC restorations, with 37% of restorations being lost by six months and 51% by the study censor date; the reasons for the failure rate of this restoration type are explored fully within the paper's text. Regarding the failure rate of the other restoration types, i.e. PCR:GIC and CR, 16% and 19% of restorations had failed for PCR:GIC restorations at six – and twenty–four months respectively, whilst for CR restorations, 5% and 12% of restorations failed over the same time periods; this is comparable to other studies where a plastic restoration has been used to restore a primary molar tooth6,7 and hence, to consider the results 'so poor' is, we believe, invalid. In relation to those teeth where abscess formation was noted, the authors note the ambiguity of the text and wish to clarify that these teeth were included in the statistical analysis and it was the patients who were withdrawn from the trial. Concerning those patients lost to follow–up, unfortunately, this is an inevitable difficulty with any trial which is clinically–based.

With regard to conventional restorative treatment techniques, the authors fully acknowledge that this is currently accepted best practice, in line with BSPD guidelines 8 and effective, particularly when used in specialist practice 9 and dental hospitals 10 ; the difficulty, however, appears to be that such techniques are not popular with general dental practitioners working within Primary Care, where over 90% of child dental care is provided and indeed, recent surveys throughout the United Kingdom have determined that less than 15% of cavities in primary teeth in five–year–olds are being restored. 11, 12 Furthermore, the care index (the proportion of carious primary teeth which have been treated restoratively) for five–year–olds in Scotland has fallen from 20% in 1983 to 9% in 2003. 13 Perhaps one of the conclusions which can be drawn from this relatively small prospective study is that there is both a need to train further Specialists in Paediatric Dentistry to work within the Primary Care setting, in addition to the overall expansion of the discipline within the Hospital Dental Service?