Abstract
The use of prophylactic antibiotics to reduce postoperative complications in third molar surgery remains controversial. The study was a prospective, randomized, double blind, placebo-controlled clinical trial. 100 patients were randomly assigned to two groups. Each patient acted as their own control using the split-mouth technique. Two unilateral impacted third molars were removed under antibiotic cover and the other two were removed without antibiotic cover. The first group received antibiotics on the first surgical visit. On the second surgical visit (after 3 weeks), placebo capsules were given or vice versa. The second group received antibiotics with continued therapy for 2 days on the first surgical visit and on the second surgical visit (after 3 weeks) placebo capsules were given or vice versa. Pain, swelling, infection, trismus and temperature were recorded on days 3, 7 and 14 after surgery. Of 380 impactions, 6 sockets (2%) became infected. There was no statistically significant difference in the infection rate, pain, swelling, trismus, and temperature between the two groups ( p > 0.05). Results of the study showed that prophylactic antibiotics did not have a statistically significant effect on postoperative infections in third molar surgery and should not be routinely administered when third molars are removed in non-immunocompromised patients.
The debate about the use of antibiotic prophylaxis began in the early 1950s. A ltemeire et al. emphasized the necessity of determining the specific indications for prophylactic antibiotic therapy on the basis of existing knowledge.
In 2007, the American Heart Association published new guidelines for antibiotic prophylaxis in infective endocarditis , based on a growing body of evidence that for most patients, the risk of taking preventative antibiotics outweighs the benefits. Risks include adverse reactions to antibiotics and the development of drug-resistant bacteria. In March 2008, the Department of Health in the UK published the National Institute for Health and Clinical Excellence (NICE) guidance on ‘antibiotic prophylaxis against infective endocarditis’ . These guidelines recommend that antibiotic prophylaxis should not be given to adults and children with structural cardiac defects undergoing dental interventional therapy .
The use of prophylactic antibiotics to reduce postoperative complications in third molar surgery remains controversial. Some authors favour routine prophylaxis , others suggest it might be of value only in difficult cases and others report no benefit . If prophylactic antibiotics are considered, it is important to note that they only provide adequate protection if effective levels are present at the time of bacterial contamination .
The incidence of postoperative infection in oral surgery ranges from 1% to less than 6%, and most infection is minor . This low complication rate does not support the routine use of antibiotic prophylaxis if the basic principles of prophylaxis are followed because the potential for adverse reactions to antibiotic therapy outweigh the possible benefits of a decrease in the infection rate. Studies comparing infection rates after use or no use of antibiotics did not show a decrease infection rates in groups using antibiotics .
Several reports consider the use of antibiotics in third molar surgery. Most researchers used amoxicillin, metronidazole, clindamycin, cephradine, tinidazole/pivampicillin, clavulanic acid and doxycycline . They found infection rates of 1–27% . The overall incidence of infection from third molar extraction is in the range 3–5% .
There was growing concern in the Department of Maxillo-facial and Oral Surgery at University of the Western Cape about the overzealous use of antimicrobials in the removal of impacted third molars. The authors decided to evaluate the potential value of prophylactic antibiotics in third molar surgery.
Materials and methods
The study was a prospective, randomized, double blind, placebo-controlled clinical trial in which the patients acted as their own control (cross over). 100 patients with 4 impacted third molars and with no known immune compromise were included in the study. The patients were paired and assigned to two groups using radiographs and the P ell & G regory classification ( Table 1 ) for mandibular impaction and the classification described by A rcher for maxillary impaction ( Table 2 ). In this split-mouth design, divisions of the mouth, such as right (upper and lower) and left (upper and lower) quadrants constituted the experimental units, which were randomly assigned to two treatment groups. Each patient served as his or her own control, which increased statistical efficiency. In the first surgical visit, two unilateral third molars were surgically removed. The second surgical visit was scheduled 3 weeks later when the other two contralateral third molars were removed.
| (1) Relation of the tooth to the ramus of the mandible |
| Class I: Sufficient amount of space between the ramus and distal of the second molar for the accommodation of the mesiodistal diameter of the crown of the third molar |
| Class II: The space between the ramus and the distal of the second molar is less than the mesiodistal diameter of the crown of the third molar |
| Class III: All or most of the third molar is within the ramus of the mandible |
| (2) Relative depth of the third molar in bone |
| Position A: The highest portion of the tooth on a level with or above the occlusal line |
| Position B: The highest portion of the tooth below the occlusal line, but above the cervical line of the second molar |
| Position C: The highest portion of the tooth on the level with or below the cervical line of the second molar |
| (3) The position of the tooth in relation to the long axis of the second molar | |||
| (i) | Vertical | These may also occur in | |
| (ii) | Horizontal | (a) Buccal deflection | |
| (iii) | Inverted | (b) Lingual deflection | |
| (iv) | Mesioangular | (c) Torsion | |
| (v) | Distoangular | ||
| (1) Relative depth of the impacted maxillary third molars in bone |
| Class A: The lowest portion of the crown of the impacted maxillary third molar is on a line with the occlusal plane of the second molar |
| Class B: The lowest portion of the crown of the impacted maxillary third molar is between the occlusal plane of the second molar and the cervical line |
| Class C: The lowest portion of the crown of the impacted maxillary third molar is at or above the cervical line of the second molar |
| 2. The position of the long axis of the impacted maxillary third molar in relation to the long axis of the second molar | |||
| (i) | Vertical | These may also occur simultaneously in | |
| (ii) | Horizontal | (a) Buccal version | |
| (iii) | Mesioangular | (b) Lingual version | |
| (iv) | Distoangular | (c) Torsoversion | |
| (v) | Inverted | ||
| (vi) | Buccoangular | ||
| (vii) | Linguoangular | ||
| (3) Relationship of the impacted maxillary third molar to the maxillary sinus |
| Sinus approximation |
| No bone or a thin partition of bone between the impacted maxillary third molar and the maxillary sinus |
| No sinus approximation |
| 2 mm or more of bone between the impacted maxillary third molar and the maxillary sinus |
Prior to the trial, each patient was informed about the study, its aim, implications and possible complications. Signed informed consent was obtained. The patients were examined clinically and those with known immune-compromised conditions, infections or on antibiotic therapy were excluded. In each patient, two unilateral third molars were removed with antibiotic cover and the other two contralateral molars were removed without antibiotic cover. Neither patient nor surgeon was aware which teeth were removed under antibiotic cover.
The randomization process was carried out in accordance with items 8–10 of the CONSORT statement 2001 checklist for randomized controlled clinical trials (Cochrane Collaboration, Manchester, UK) . Participants were allocated to one of the two groups by asking them to pick one of the sequentially numbered, opaque sealed envelopes containing either of the two interventions. Each participant had an equal chance of being assigned to one of the two groups. During the entire double-blind study, randomization was conducted by the same hospital nurse, who had research experience.
Group I received 1 g of amoxicillin, 1 h before surgery (preoperatively) on the first surgical visit. On the second surgical visit, placebo glucose capsules (capsules of same shape and size containing glucose only) were given or vice versa. Group II received 1 g of amoxicillin, 1 h before surgery as well as 500 mg amoxicillin 8 hourly for 2 days postoperatively (the current regimen used in the department) on the first surgical visit. On the second surgical visit placebo glucose capsules were given or vice versa.
Surgery was performed under local anaesthesia. All patients were operated on by the same surgeon using a standard operating technique. An envelop mucoperiosteal flap was raised and if indicated, bone was removed on the buccal and distal aspect of the third molar with a no 8 round burr under constant sterile 0.9% saline irrigation. Tooth elevation, crown removal and or root division and elevation were carried out as required. After removal of the tooth the surgical field was meticulously rinsed with sterile 0.9% saline. The wound was closed by placing 3-0 Vicryl ® (Ethicon Inc, Piscataway, NJ, USA), interrupted sutures.
A standard regime of analgesics (ibuprofen 400 mg preoperatively, ibuprofen 400 mg 6 hourly for 2 days) plus paracetamol 500 mg and codeine phosphate 8 mg 6 hourly for 2 days) were used for both groups. A mouthwash (0.2% chlorhexidine gluconate 10 ml stat, preoperatively and postoperatively 8 hourly for 3 days) was given to all patients. It was planned to use Tilidine–HCl (50 mg) as an escape analgesic but this was not required.
In this study, pain severity was recorded on a visual analogue scale (VAS). Pain was recorded 3 times a day for 2 weeks. Patients were instructed to rate and record pain intensity on the VAS. Facial swelling was measured using visual scoring on clinical observation ( Table 3 ). Swelling was measured preoperatively, after 3 days, 7 days and 2 weeks postoperatively. Maximum mouth opening ability was measured in millimeters between the upper and lower central incisors using a Vernier-calibrated sliding caliper preoperatively, and on every postoperative visit ( Table 4 ). Clinical signs of pus collection were recorded on every visit. Temperature was recorded preoperatively and on every postoperative visit. Temperature >38 °C was considered a fever. Patients were evaluated for the presence of dry sockets at each visit. Halitosis, pain and clinical signs of clotless socket with necrotic bone were used as diagnostic criteria for a dry socket.
| Variables | Score |
|---|---|
| None (absent) | 0 |
| Mild (just visible and palpable) | 1 |
| Moderate (obvious) | 2 |
| Severe | 3 |
| Variables | Score |
|---|---|
| Non | 0 |
| 1–5 mm | 1 |
| 6–10 mm | 2 |
| 11–15 mm | 3 |
| 16–20 mm | 4 |
| >20 mm | 5 |
The methods used to evaluate pain, swelling, trismus and infection are accepted methods employed in the scientific literature . Inter-examiner variability was excluded by using only one research assistant. All assessments were done in the same clinical environment. This clinical research was approved by Research (Ethics) Committee of the University of the Western Cape.
Data management
The data was tabulated on an Excel spread sheet and analyzed using a commercially available statistical software package (SPSS ® 15.0, SPSS ® Inc.). The χ 2 test was used to compare the proportion of the nominal variables between the two treatment groups. Non-parametric analysis of variance was used to identify any significant variables in the two groups.
Results
100 patients (62 females and 33 males) were included in the study. The mean age of the study population was 26 years. Surgery was performed on 100 patients but 95 patients completed the study protocol. Five patients were excluded from the study as they did not complete the follow-up protocol. 380 impacted third molars were removed in the two groups (192 in Group I and 188 in Group II). There was no significant difference between the two groups with regard to the clinical position of the impactions or difficulity in removal of the third molars.
138 bony (Classes II and III) impacted mandibular third molars and 75 maxillary bony (Classes B and C) impacted third molars were removed in the study. 53 Class I mandibular third molar impactions and 114 Class A maxillary impacted third molars were removed.
Only 6 (2%) postoperative infections were recorded in 380 sockets ( Fig. 1 ). In Group I, four (2%) infected sockets were recorded of which three (2%) were in the placebo group and one in the stat group. No infection was recorded in this group on days 7 and 14 after surgery. One infection was recorded 3 weeks after surgery in the placebo group. In Group II, only two (1%) infected sockets were recorded. Both these infections occurred on the same patient. This patient was taking antibiotics (Group II) on the first surgical visit, and on the second surgical visit she used placebo capsules. Both of these infections were recorded on the third day after surgery. Only one patient from Group I presented with a dry socket and that was on the seventh day after surgery.
In Group I, 37 (77%) patients presented with no swelling, 5 (10%) patients had mild swelling and one patient had severe swelling ( Fig. 2 ). In the placebo subgroup, 18 (38%) patients had no swelling, while 15 (31%) patients showed mild and moderate swelling on the third day after surgery. 22 (47%) patients in Group II had no swelling. 18 patients (38%) presented with mild swelling and one patient with severe swelling. 20 (43%) patients in the placebo subgroup presented with mild swelling and 10 (21%) with moderate swelling on the third day after surgery. Two (4%) patients had severe swelling in this group. The swelling score decreased with time ( Fig. 3 ), but there was no statistically significant evidence of treatment group effect ( p > 0.05). The swelling scores in the two treatment groups were not statistically different ( Table 5 ).
