Introduction

This study reviews the effect of experience on the outcome of patients undergoing surgical extraction of third molar teeth. This is one of the most common procedures performed in oral and maxillofacial surgery (OMFS). The paradigm may easily be extended to other common operations such as adenoidectomy, tonsillectomy and middle ear surgery (including insertion of grommets). In their entirety they form the majority of the surgical procedures carried out each year in the United Kingdom. The most common indications for the removal of third molar teeth are pain, recurrent swelling and infection. The assessment of a symptomatic patient involves clinical examination and assessment of a two dimensional radiograph of the tooth to be removed. A decision would be formulated on whether the surgery should take place under local anaesthesia, intravenous sedation or general anaesthesia. Factors including surgical difficulty, anticipated complications, patient's preference, fear and anxiety and the surgeon's experience as well as the published NICE guidelines should be taken into account before decision making.1

Postoperative complications are a potential risk of any surgical procedure. However, with careful assessment and treatment planning, they may be minimised. This involves appropriate patient preparation, an aseptic technique, meticulous management of the hard and soft tissues, the use of controlled force when applying surgical instruments, haemostasis and adequate adherence to postoperative instructions.2,3

Common postoperative complications of third molar surgery include pain, bleeding, trismus, swelling, infection, alveolar osteitis, delayed healing and sensory disturbances in the distribution of the inferior alveolar and lingual nerves, and occasionally the mylohyoid nerve.4 Less common complications include displacement of tooth/root fragments, injuries to adjacent teeth and structures, temporomandibular joint disorders, jaw fracture, maxillary tuberosity fracture and the aspiration or swallowing of teeth.5,6

Variables that may contribute to the incidence and severity of postoperative complications include the length of operation, surgical techniques implemented, including flap design, irrigation and management of soft tissues, as well as the use of dressings, mouthwashes and prophylactic antibiotics.7,8,9,10,11 Other factors that may influence the outcome include the patient's health status, age, gender, race and anxiety.12,13 The degree of impaction of the tooth and pre-existing pathology and/or infection should also be taken into consideration as it may affect the outcome.14

Previous studies have suggested that the level of experience of the surgeon may impact upon the incidence of postoperative complications.14,15,16 Sisk et al.15 investigated the effect of the experience of the surgeon on the complication rate following surgical removal of third molar teeth by comparing specialists in an oral surgery group with residents in the same faculty. They showed that complications were numerous after removal of teeth classified as being partially or completely impacted within bone and that patients treated by less experienced surgeons had significantly higher incidences of complications.

Handelman et al.17 carried out a study to assess the postoperative complications in patients who had undergone surgical removal of third molars by OMFS residents compared with general dentistry residents. They showed that overall there was no significant difference in complication rates between the two groups.

Berge and Gilhuus-Moe14 compared postoperative complications following surgical removal of third molars in two groups of patients treated by four general dental practitioners and by a consultant oral surgeon. An increased incidence of postoperative alveolar osteitis, pain and increased duration of surgery was reported in the patients treated by the general practitioners. de Boer et al.6 found there to be higher complication rates when third molar surgery was performed by residents, with regards to alveolar osteitis, swelling and postoperative bleeding. Interestingly, patients treated by senior staff in the same study showed higher rates of postoperative infection and paraesthesia.

Most of the literature supports the hypothesis that inexperience of the surgeon is related to an increased incidence of postoperative complications (Table 1).14,15 However, some studies have failed to reveal any correlation between the experience of the surgeon and postoperative complications.17

Table 1 Disadvantages of non-apprenticeship training models and training paradigms
Full size table

In a preliminary prospective study by Jerjes et al.,1 involving 1,087 patients at University College London Hospital, the incidence of some postoperative complications in a group of patients treated by residents was statistically significantly higher than the group of patients treated by specialists. This was in terms of postoperative trismus, alveolar osteitis, infection and paraesthesia of the inferior alveolar and lingual nerves. Also, the percentages of patients that experienced a sore throat postoperatively and swelling were slightly higher but statistically insignificant. However, the incidence of postoperative bleeding was higher in the group of patients treated by specialists.

This study was designed to expand upon the preliminary data, aiming to assess the incidence of postoperative complications following the removal of third molars under local anaesthetic, in relation to the experience of the operating surgeon. The implications of the results on training and healthcare are discussed.

Materials and methods

In this prospective clinical study, 3,236 patients underwent surgical removal of impacted third molar teeth. Patients' demographics and the radiological characteristics for each case, based on plain film radiographs, were recorded. The teeth were classified to be either fully erupted, fully impacted (ie unerupted) or partially impacted (ie partially erupted).

Subjects included in the study had their third molars removed surgically at the UCLH Maxillofacial Unit and the Eastman Dental Hospital, London. The study protocol was approved by the joint UCL/UCLH committees of the ethics for human research.

All procedures were performed in similar clinics, equipped with similar surgical instruments, rotary and irrigation devices and materials (sutures and haemostatic agents). Before surgery, each patient was informed of the possible complications and provided fully informed consent. Local anaesthesia was applied (2% lidocaine with 1:80,000 adrenaline) by local tissue infiltration and/or inferior alveolar nerve block.

A standard surgical approach was implemented in all cases. An envelope mucoperiosteal flap was reflected and bone was removed buccodistally. Sectioning of the tooth was carried out when needed. The wound was carefully irrigated with sterile saline and any bony spicules removed. Following tooth extraction, the flap was then repositioned and sutured with 4/0 Vicryl. A lingual flap was not raised in any of the cases. No patient in this study required coronectomy and lingual or buccal split techniques.

During the immediate postoperative phase, all patients were given written instructions about wound care and warned of possible complications. All patients were prescribed the recommended analgesics and antibiotics and were reviewed seven days postoperatively.

The cases were distributed among specialists and residents randomly regardless of patient age, gender or even complexity of surgery. The specialists included staff grade oral surgeons, senior registrars and OMFS consultants. The residents included senior house officers and junior registrars.

Patients were excluded from the study if their records showed that they failed to attend during the postoperative review appointments. The procedures were performed by 12 OMFS surgical residents and seven surgical specialists.

Patients who presented with paraesthesia in the first week postoperatively were followed for up to two years or until they recovered sensation. Where indicated paraesthesia was assessed by standard neurological testing, with soft touch and neurological pin point sensation. Surgical exploration and micro-neurosurgical repair was applied when indicated; however, this area is beyond the scope of this study. Patients who continued to present with paraesthesia after two years were considered to have a permanent nerve dysfunction.

Statistical methods

Data were collected upon proformas. One hundred proformas were randomly selected and reviewed by an independent assessor to ensure quality of data entry. Statistical Package for Social Scientists (SPSS version 14.0) was used to analyse the data collected.

Adverse outcomes following surgery were summarised as frequencies separately, according to the grade of the surgeon undertaking the procedure. The results were cross tabulated and the Chi-squared statistic was used to test for differences in the case-mix between the surgical grades. Fisher's exact test was used for the analysis of contingency tables and therefore to measure the p-value. The Likelihood Ratio for each complication was calculated, to compare the likelihood of a patient suffering a specific complication, in relation to the grade of surgeon providing the treatment.

Results

The 3,236 patients treated had a mean age of 24.2 years. The age of the patients in this study ranged from 17 to 36 years. The distribution of patients in the various age groups was relatively equal among the residents and specialists. However, the specialists treated slightly more patients over 30 years and between 17-20 years of age (Table 2).

Table 2 Profile of treated cases and case-mix in relation to seniority of surgeon. Note: The numbers refer to the number of patients
Full size table

There were 346 more female than male patients in the study. The OMFS residents treated a higher proportion of female patients (943/1,565; 60.3%) than their senior colleagues (848/1,671; 50.7%). 2,531/3,236 (78.2%) of the teeth removed had roots that radiographically appeared at a distance of 2 mm or less from the inferior alveolar canal (IAC). The resident and specialist surgeons removed similar numbers of teeth that were close to the inferior alveolar nerve, with 1,267 and 1,262 for the junior and senior surgeons, respectively.

Both groups of surgeons treated similar numbers of fully erupted and partially impacted teeth. The specialists were noted to have removed more third molar teeth reported as fully impacted (273/368; 74.2%). The majority of the teeth extracted were partially impacted (2,572/3,236; 79.5%) and were vertically or mesioangularly impacted. The residents treated more patients with horizontally impacted teeth (163/272; 59.9%). However, the specialists removed more mesioangularly impacted teeth (822/1,518; 54.2%). The number of teeth removed that were vertically or distoangularly impacted was similar for both groups of surgeons.

A higher incidence and significant statistical difference in the group of patients treated by residents was noted with regards to some postoperative complications. These included trismus (p <0.001), sore throat (p = 0.004), delayed healing (p = 0.009), alveolar osteitis (p <0.001) and postoperative infection (p <0.001) (Table 3). Postoperative bleeding was the only complication that was reported to have a significantly higher incidence in the group of patients treated by seniors (p <0.001). No statistically significant differences were seen in either group in terms of postoperative swelling and abscess.

Table 3 Incidence of each type of postoperative complication in relation to seniority of the surgeon undertaking the procedure. Note: The numbers refer to the number of patients. 'Lip' refers to patients reporting numbness of the lip. 'Tongue' refers to patients reporting numbness of the tongue
Full size table

The group of patients treated by residents were over 20 times more likely to develop inferior alveolar and lingual nerve paraesthesia within the first month following the surgery (p <0.001). This group were also over 20 times more likely to sustain such a complication for the first two years following surgery (p <0.001).

Discussion

Third molar extraction is one of the most commonly performed procedures within the field of dento-alveolar surgery, and overall there is a fairly low rate of postoperative complications. Thus, trainees have frequent opportunities to acquire the skills to perform third molar surgery effectively, with minimal complications. The results suggest that these operations are not difficult to master but require experience to minimise adverse outcomes, falling in line with the saying 'practice makes perfect'.

In summary, patients treated by oral and maxillofacial surgery residents experienced a higher rate of postoperative complications compared with patients treated by specialists. Categories of postoperative complications where this applied included trismus, sore throat, delayed healing, infection, alveolar osteitis and nerve paraesthesia. Postoperative bleeding was the only parameter that had a significantly higher incidence in patients treated by more senior surgeons.

Previous training programmes have provided a broad range of surgical exposure to different specialities. Some knowledge of general surgical principles is often learned best through the direct observation and/or assisting of senior colleagues, building upon information gained from written learning material.

Although the current vogue of minimal access surgery is laudable the shift to a wider or alternative exposure in difficult or complicated cases should be taught as a default position. With better surgical exposure surgeons avoid excessively forceful instrumentation of the oral cavity (through a small incision) and also reduce the need for excessive opening of the mouth, with the associated complications. Junior surgeons may find it harder to initially identify difficult cases that may require alternative approaches.

These findings can also be readily translated to some of the other most common performed operations. These include tonsillectomy, adenoidectomy, grommets insertion, even skull base surgery where, in some instances, the trainee cannot be directly observed by the trainer to provide immediate feedback. This is due to obscuration of the surgical field by instruments or operator, preventing the trainee from learning the appropriate skill set from the trainer.

The findings of this study are very similar to those of the preliminary study by Jerjes et al.,1 with a greater number of parameters of postoperative complications being statistically more significant. Previous studies have suggested that a surgeon's experience may be a factor in the frequency of complications.14,15,16 This may be attributable to a surgeon's assessment and treatment planning skills, which develop over time. Surgeons should be able to identify potential complications and aim to avoid or minimise them.

The assessment of difficulty of extractions may be easier for more experienced surgeons. However, a study by Sursala et al. concluded that error in estimation of difficulty had little or no dependency on surgical experience.18 It is likely that a surgeon's inexperience will have greater bearing on other factors affecting outcome such as surgical technique, tissue handling and length of operation.19

Postoperative trismus may be related to the detrimental effect of prolonged surgery on the muscles of mastication.20 Trismus and swelling are both subjective parameters and difficult to assess quantitatively, despite being easily observed.1 They may be of varying severity and various techniques have been suggested to measure them.20,21 In this study, there was a significantly higher incidence of trismus (25 times more likely) in the group of patients treated by OMFS residents. These results are consistent with the findings of previous studies.13,14,15

There was no difference between the two groups with regards to postoperative swelling, which was also the finding of de Boer et al. in 1995.6 Baxendale et al.22 found that the administration of dexamethasone leads to a marked reduction in swelling following the removal of third molar teeth. Other studies have been carried out to assess the potential benefit of steroid in reducing postoperative complications.23,24 However, most surgeons feel that there is no need to carry out such specific measures to reduce postoperative swelling and trismus, since both of these symptoms usually subside after 3-5 days of their own accord.

The incidence of postoperative bleeding in the group of patients treated by more senior surgeons was double that of the incidence in the group of patients treated by the more junior surgeons. These findings correspond with those of the preliminary study,1 but conflict with those of several previous studies.4,6,14

On the other hand, it was found that patients treated by OMFS residents were nearly 50 times more likely to develop postoperative infection compared with patients treated by senior specialists. A possible explanation for this may be that the OMFS residents treated a higher percentage of female patients than the specialists. Female patients have been shown to have an increased tendency to develop infection following third molar removal.5,25

There was no significant difference between the two groups in terms of incidence of postoperative abscess formation. This contradicts with the results of the study by de Boer et al.6 However, also in this study, correlating with our results, postoperative sore throat and delayed healing were significantly more prevalent in patients treated by junior as opposed to senior surgeons.6

There are several factors that may influence the incidence of alveolar osteitis. Patients treated by OMFS residents were found to be over 150 times more likely to develop acute alveolar osteitis. The higher rate of postoperative alveolar osteitis in the junior surgeon group may be due to the higher number of female patients in this group as previous studies have shown a higher incidence of alveolar osteitis in female patients.2,3,25 Other associated risk factors include smoking and the use of the oral contraceptive pill (OCP). Use of the OCP may explain the reported higher rate of incidence in female patients.3 It is often difficult to accurately record the number of patients on the OCP as patients often disregard this as a medicament worth mentioning during medical history taking. Blondaeu et al.3 found that the incidence of alveolar osteitis ranges between 1-5%, irrespective of experience of the operating surgeon. Other factors that influence the likelihood of development of alveolar osteitis include patient age, medical history, smoking history, tooth position, duration of surgery and surgical technique.6,15,26

Although there is no one successful method of eliminating the occurrence of a dry socket and some of the other complications described, there are several measures that can be taken to prevent them. These include avoiding unnecessary trauma and use of excessive force during surgery, careful debridement of the socket and removal any loose fragments of bone. The provision of and adherence to postoperative instructions, especially advice regarding the avoidance of smoking following tooth removal, is essential.2,27

A number of studies have found that the administration of a course of antibiotics immediately postoperatively and rinsing with mouthwashes 12 hours after tooth removal may help to some extent in preventing and/or treating such complications.7,8,9,10,28 However, the use of prophylactic antibiotics and type of antibiotic used is controversial. Some studies have found that the routine provision of antibiotics does not provide any benefit in the prevention of postoperative complications.29 The use of a preoperative chlorhexidine rinse and resorbable gelatine or cellulose sponges have been suggested to reduce the incidence of alevolar osteitis, infection and bleeding.1,11,16,24

Numbness of the lip and tongue are symptoms that a patient may rate as having a high significance compared with some of the other complications described. The rates of nerve damage are variable. Previous studies have found that lingual nerve damage may range between 0%30-23%31 and inferior alveolar nerve damage may range between 0.4%15-8.4%.32 The incidences of temporary and permanent nerve damage in our study lie in the lower part of these ranges, irrespective of the seniority of the surgeon.

Patients treated by the OMFS residents reported a higher incidence of numbness of the tongue in the follow-up period. Permanent nerve dysfunction was considered if patients reported numbness two years following surgery. Our results show that patients treated by OMFS residents were 20 times more likely to develop this complication. Such a complication may be associated with a surgeon's experience, poor surgical technique and poor instrument handling.33

Patients treated by the OMFS residents also reported a higher incidence of numbness of the lip in the follow-up period. This is despite both groups of surgeons removing relatively equal numbers of teeth considered radiographically to be in close proximity to the inferior alveolar nerve. Patients treated by residents were found to be more over 20 times more likely to experience permanent nerve dysfunction compared with patients treated by specialists.

The incidence of permanent nerve damage of the inferior alveolar and lingual nerves was lower than reported by Baitaineh et al.34 and consistent with that reported by Sisk et al.15 Interestingly, de Boer et al.6 found that increased nerve paraesthesia occurred following third molar removal by specialists.

If a patient still reports numbness at the end of a monitoring period of two years, permanent nerve damage is considered.35 At this stage, a further radiograph is required to assess the continuity of the mandibular canal, as there may be infection of the inferior alveolar nerve. Surgical exploration, decompression or repair of the nerve may be considered. With regards to the lingual nerve, if a patient complains of permanent numbness of the tongue, surgical exploration may be indicated to assess the continuity of the nerve and possible need for microsurgical repair.36 However, when a patient reports improvement of sensation or an acceptable sensory deficit, non-interventional follow-up is advised. Surgical intervention is contraindicated in patients with central neuropathic pain, medical neuropathy, or when medically compromised or if the dysaesthesia not resolved by a nerve block.37,38

Conclusion

The 'craft' medical specialties have fallen victim to the latest 'one size fits all' changes in postgraduate training in the UK. This is evident for the several rapid reversals in policy have been observed recently. The true motive and logic of these quantum changes remains to be seen.

Dentoalveolar surgery and in particular third molar removal forms an integral element of training within the field of oral and maxillofacial surgery. Competence in third molar surgery forms a sound foundation for the skills necessary for some of the more complex surgical procedures performed by oral and maxillofacial surgeons. Since several hundreds of patients are treated per unit each year, this confers third molar removal to be an easily accessible field for the training of both junior and more senior surgeons. An ability to raise intraoral mucosal flaps, remove bone, divide teeth and suture mucosal tissue eases the progress to other surgical areas including surgical dermatology, trauma surgery, orthognathic/craniofacial surgery and surgical oncology.

Postoperative complications did occur in patients treated by both junior and more senior surgeons. However, the results of this study suggest that there is a statistically significant higher incidence of complications in some parameters following third molar removal when patients are treated by less experienced surgeons. These findings were found with in relation to trismus, sore throat, delayed healing, infection, alveolar osteitis and nerve paraesthesia.

One may question whether it is ethical to allow juniors to perform some of these most common operative interventions, in the knowledge that patients they treat are more likely to experience postoperative complications. Ethical arguments will revolve around workload and the obligation to train the next generation of 'senior surgeons'.

The General Dental Council's document on Standards for dental professionals states that the first principle of practice in dentistry is to put the patient's interests first and act to protect them and secondly, to respect a patient's choice. The conclusions of this study may have implications upon the provision of healthcare. Patients may feel that they do not wish to be treated by less experienced surgeons, due to the greater risks involved. All steps to minimise these complications must be undertaken, in order to improve patient care. In the future we have to ensure we impart not only the knowledge of how to do the procedure but also the experience in how to avoid their complications.

Further research into the influencing factors and prevention of complications is necessary. All clinicians develop their skill base with experience and even if OMFS residents are closely supervised, it is impossible to eliminate the complications outlined in this study. This study also shows that complications do occur in patients treated by more senior surgeons.

The authors would like to stress that the impact of the findings from this study upon the profession, education and research is still unknown. Higher evidence based trials are expected to reveal more parameters that can affect the complications rate in third molar surgery. Larger cross-sectional studies and comparison of clinicians at different levels within various training programmes are recommended.