This article discusses the risks which are associated with the retention of impacted third molars and the problems which may result from their removal.  It concludes that guidelines are needed, but that currently there is no conclusive evidence to justify the prophylactic removal of asymptomatic impacted lower third molars.

By Dr Marcelo Fernandes

Health systems will never have sufficient resources to cover every demand put on the system so choices must be made on where and how resources are best allocated. There are therapeutic indications for the removal of impacted third molars, but there is no concensus about the need for removal of asymptomatic impacted third molars. The reason for the prophylactic removal of third molars is the belief that eventually most of these teeth will cause pathological conditions, so should be extracted early in life, when patients can cope better with post-operative symptoms and are likely to have fewer systemic health problems. However, recent studies have shown that many impacted third molars may remain free of disease or symptoms and therefore their removal seems to be unnecessary. Although the current literature concerning third molars is vast, the predictability that an impacted but symptom-free third molar will develop disease is difficult. Many questions concerning the management of impacted third molars still remain unanswered. The answers to those questions are of great importance, not only for clinicians but also for healthcare providers who have to make decisions regarding the allocation of resources as well as patients who will be directly affected by the decisions clinicians and healthcare providers make.`

The development of guidelines
The first formal guideline document on the management of impacted third molars was the US National Institutes of Health (NIH) Report, namely the NIH Consensus Development Conference for Removal of Third Molars, 1980.  In addition to highlighting the need for more research on the issue of impacted third molars, and particularly the prophylactic removal of asymptomatic teeth, a number of well-defined criteria for the removal of third molars emerged from the conference. The criteria were  in part, infection, nonrestorable carious lesions, cysts, tumors, and destruction of adjacent teeth and bone. The enormous number of teeth removed in the UK in the 1990s prompted researchers to evaluate the reasons for extraction; in 1998 it was estimated that around 20-30% of third molars that were removed in the UK did not follow the NIH criteria for removal and were, therefore, removed prophylactically. Around that time the Faculty of Dental Surgery of the Royal College of Surgeons of England had just published a document on the management of impacted wisdom teeth, to be used as a national guideline, and this document highlighted the fact that there was no conclusive evidence to justify the prophylactic removal of asymptomatic impacted lower third molars. This document was followed by the NICE (National Institutes of Clinical Excellence) guidelines in England and Wales (NICE Guidance on the removal of wisdom teeth 2000) and by the Scottish Intercollegiate Guidelines Network (SIGN) document [1, 2]. Both documents indicated that prophylactic extractions should be discontinued. However, the documents also emphasised that more research should be carried out in the area. In the light of these guidelines the “watchful monitoring” policy seems to be favoured by health managers in the UK, as in the short term it seems to be a less costly procedure when compared to extractions and surgical extractions. In 2005 the Cochrane Oral Health Group systematic review found no evidence to support or to refute the prophylactic removal of impacted wisdom teeth [3].
 
As a consequence of these publications the number of extractions in the UK has fallen dramatically, but this is not the case in other countries. A recent study compared the pattern of extractions of asymptomatic impacted third molars in US dental schools in 1998/1999 with that of 2004/2005. Surprisingly little change was found despite most scientific evidence discrediting prophylactic extraction. Most departments in the dental schools studied still favoured prophylactic extraction of partially erupted and unerupted third molars. Who is right? In a profession where there is a lot of pressure for all interventions to be evidence-based, where is the evidence for removing or for retaining impacted lower third molars?

Risks associated with third molar retention
A pathology-free impacted third molar that is not prophylactically removed has the potential to cause future disease if left in place. However, some studies have shown that the incidence of disease in these teeth may be overestimated by dentists. The most commonly reported pathologies associated with a non-interventionist approach are described below.

1. Pericoronitis
Many studies have looked at the association of pericoronitis and third molar impaction and this is still the main cause for extraction of these teeth. However, one of the major flaws in these studies is the fact that there is no standard definition of pericoronitis. The eruption process is also likely to cause minor gingivitis, where the symptoms may be similar to pericoronitis, and the lack of a good definition for this disease may lead researchers and clinicians to misclassify it. Still pericoronitis is undoubtedly the main problem faced by dentists when it comes to lower impacted third molars.

2. Caries
Another major indication for extraction of impacted lower third molars is caries, either in the third molar itself or in the distal surface of the second molar. However, most studies in this field were conducted in patients that were referred for removal and therefore do not represent the actual incidence of this disease in the general population. In addition, the association between caries and impacted third molars may also be overestimated; it is well known that this disease is related to
socio-economical factors.

3. Cysts and tumours
Reports in the literature on the incidence of cyst formation in third molars vary considerably, but the condition normally ranges from 0% to 11% or even higher in certain studies. This large variation makes the potential of cyst formation a weak indication for prophylactic extraction of impacted third molars. Again, cyst formation is another pathology that seems to be related to age. Neoplasic tumours associated with third molars are very rare.

4. Root resorption
Some studies have shown that when a third molar is left in situ there is a probability that it may cause resorption of the distal root of the adjacent tooth. Mesioangulation is the most likely type of impaction to cause this. Some studies also report an association between root resorption at the apex and individuals over 50 years of age. Again, these studies are retrospective and are normally caried out in secondary care settings, thus do not represent the incidence of this problem in the general population.

5. Periodontal disease
The incidence of periodontitis on the distal surface of the second molar varies from 1% to 5%. Although a higher incidence of periodontitis associated with impacted wisdom teeth has been reported among older patients, it is well known from the literature that the incidence and prevalence of this disease increase with age regardless of the presence of third molars. There is a paucity of studies relating periodontitis associated with impacted third molars with oral hygiene, which may be a confounding factor.

6. Late lower incisor crowding
One of the main controversies regarding the prophylactic removal of lower third molars is the belief that the retention of these teeth would cause late crowding of the lower incisors. However, a  randomised controlled trial found that the presence of impacted third molars had no significant clinical influence on the development of crowding. Previous studies support these findings and suggest that crowding may be caused by factors other than the presence of third molars. Finally, a review of pertinent studies related to management of third molars in an orthodontic context concluded that the influence of third molars in the alignment of the anterior dentition may be controversial and that no evidence exists which demonstrates that these teeth cause late incisor crowding.

7. Other pathologies
There are other less common pathologies that can be associated with the presence of third molars. One of the most commonly reported is pain directly related to the presence of a third molar. The reported prevalence of this condition varies greatly: between 5% and 53%. The incidence of cellulitis and osteomyelitis has
been reported to be around 5%.

There are a few other conditions which are also believed to be associated with impacted third molars. These mainly include functional disorders such asocclusal interference, cheek biting, mastication disorders, trismus and temporomandibular joint (TMJ) problems. Although these pathologies and symptoms may cause distress and pain, their relationship to third molars is not well established and there is a paucity of current literature about this subject.

Most of the studies mentioned above were performed in secondary care settings where patients had been referred for treatment, so they  may not represent the incidence of problems in the general population.

Risks associated with third molar removal
Morbidity can be inevitable after surgical procedures. Even after uncomplicated surgery, most patients subjected to third molar removal experience some morbidity. The most common post-operative adverse effects following third molar removal are listed below.

1. Post-operative symptoms
The most common symptoms after third molar removal are pain and swelling; some patients can experience pain for more than seven days. Functional limitation is also seen in many cases. Other adverse effects may include difficulty in eating, speech disorders, trismus and inability to socialise.

2. Reduction in quality of life
A number of studies are beginning to show that oral disorders can have a significant impact on the physical, social and psychological well being of people. The majority of patients take time off work after third molar surgery. Although many studies have been carried out, it seems that the impact of third molar removal on a patient’s quality of life has been underestimated by dentists when they decide on treatment options.

3. Nerve damage
Nerve damage after third molar removal can be a serious complication, especially if it subsequently becomes permanent. Different studies demonstrate great variation in the incidence of lingual nerve damage, but in general it ranges from 1% to 6%. It seems that the main risk factors for nerve damage are the distribution of the nerves and the experience of the surgeon. Many studies seem to point out that there is an increased risk of nerve damage for extractions of horizontally impacted teeth.

4. Other complications
General anesthesia is, per se, not without risk. The UK National Third Molar Project in 1998 found that most third molars in the UK were removed under general anaesthesia. This contrasted with findings from other countries such as Brazil and Sweden where most third molars are removed under local anesthesia.

In addition, surgical procedures may damage other surrounding tissues like the periodontum. Other complications that may result from third molar removal are injuries to adjacent teeth during the surgical procedure, retention of a root tip (which may require another surgical intervention), post-operative alveolar osteites and
mandible fracture.

Final considerations
In our recent study investigating the survival of impacted lower third molars (n=676) in patients attending general dental practice, we found that most teeth (83%) survived the study period (one year) symptom-free. Some of these teeth were symptomatic, but only caused minor symptoms (9%) which were not sufficient to justify extraction. Only a small percentage (8%) caused symptoms that would justify extraction according to the guidelines used (SIGN).There was a higher incidence of symptoms in younger patients when compared to patients in the older age groups [Table 1].

We also looked at the possible association between angulation and degree of impaction with the development of symptoms in the teeth studied. When angulation of impaction was considered, the distally impacted lower third molar had more symptoms when compared to the other angulations [4].

The conclusions drawn from our studies, as well as from the current literature, show that the fate of an impacted lower third molar is still, to large extent, unclear. However, such studies also indicate that in regard to the symptoms investigated, the majority of impacted lower third molars can be safely left in place until symptoms develop. These results, combined with the fact that the evidence in favour of extraction is weak, lead us to believe that prophylactic extraction should be discontinued until further longitudinal studies indicate otherwise.

The development of guidelines is paramount in order to standardise the profession and provide evidence-based dentistry. However, patients should have the opportunity to make more informed decisions about which method of treatment is preferable for them. Such decision should be based on the probability of the development of symptoms and the risks posed by intervention or non-intervention.

References
1. NICE. Guidance on the removal of wisdom teeth.York: National Institute for Clinical Excellence; 2000.
2. SIGN. Management of unerupted and impacted third molar teeth. Edinburgh: Scottish Intercollegiate Guidelines Network; 2000.
3. Mettes DTG, Nienhuijs MEL, van der Sanden WJM, Verdonschot EH, Plasschaert A. interventions for treating asymptomatic impacted wisdom teeth in adolescents and adults. Cochrane Database of Syst Rev 2005; Issue 2. Art. No.: CD003879. DOI: 10.1002/14651858. CD003879.pub2.
4. Fernandes MJ, Ogden GR, Pitts NB, Ogston SA, Ruta DA. Actuarial life-table analysis of lower impacted wisdom teeth in general dental practice. Community Dent Oral Epidemiol 2010; 38: 58–67.

The author
Dr Marcelo Fernandes,
General Dental Practitioner, Aberdeen, Scotland, UK.
e-mail: mfernandes10@hotmail.com