Removal of miniplates is a controversial topic in oral and maxillofacial surgery. Originally, miniplates were designed to be removed on completion of bone healing. The introduction of low profile titanium miniplates has led to the routine removal of miniplates becoming comparatively rare in many parts of the world. Few studies have investigated the reasons for non-routine removal of miniplates and the factors that affect osteosynthesis after osteotomy in large numbers of patients. The aim of the present study was to investigate complications related to osteosynthesis after bilateral sagittal split osteotomy (BSSO) in a large number ( n = 153) of patients. In addition to the rates of removal, emphasis was placed on investigating the reasons and risk factors associated with symptomatic miniplate removal. The rate of plate removal per patient was 18.6%, the corresponding rate per plate being 18.2%. Reasons for plate removal included plate-related complications in 16 patients and subjective discomfort in 13 patients. Half of the plates were removed during the first postoperative year. Smoking was the only significant predictor for plate removal. Patients undergoing orthognathic surgery should be screened with regard to smoking and encouraged and assisted to cease smoking, at least perioperatively.
Bilateral sagittal split osteotomy (BSSO) of the mandible is a standard procedure in oral and maxillofacial surgery. The first surgeon to report this procedure was Hugo L Obwegeser; he performed it on 22 April 1953 using a transbuccal approach under local anaesthesia . Since then the procedure has been refined , and osteosynthesis assisted by screws and plates has replaced intermaxillary fixation .
With the exception of the Champy system , the early plates were bulky and often made of stainless steel. Removal of the plates once they had ceased to function was advocated as part of the treatment. The Strasbourg Osteosynthesis Research Group (SORG), which was founded in 1988 as a team of independent, dedicated surgeons working for the scientific and technical advancement in oral and craniomaxillofacial surgery, made the following recommendations at their symposium in Volendam, Netherlands in 1991: ‘A plate which is intended to assist the healing of bone becomes a non-functional implant once this role is completed. It may then be regarded as a foreign body. While there is no clear evidence to date that a plate causes actual harm, our knowledge still remains incomplete. It is therefore not possible to state with certainty that an otherwise symptomless plate, left in situ , is harmless. The removal of a non-functioning plate is desirable provided that the procedure does not cause undue risk to the patient .’
Today, miniplates, in general, and titanium miniplates, in particular, are widely used for fixation of facial fractures and osteotomies. Titanium has good tissue compatibility and is also widely used for permanent dental implant therapy. Titanium is considered to be non-carcinogenic, resistant to corrosion, non-toxic and without allergic associations . Titanium miniplates may compromise radiographs, but do not need to be removed for magnetic resonance imaging (MRI) . Routine removal of titanium miniplates is rarely performed by oral and maxillofacial consultants in Finland . Occasionally plates have to be removed, for example for wound-related complications and subjective discomfort.
The aim of the present retrospective study was to investigate complications related to osteosynthesis with emphasis on the rates of symptomatic miniplate removal and the reasons and risk factors associated with removal, by studying the case records of patients who had undergone BSSO.
Materials and Methods
The case records of patients who had undergone BSSO between 1 January 1997 and 31 December 2003 at the authors’ hospital were studied. Inclusion criteria were fixation with miniplates and monocortical screws only in patients who had undergone BSSO. Patients were excluded if they had undergone fixation using screws or screws/plates made of resorbable materials or had had any other osteotomy procedure (e.g. maxillary osteotomy or any other mandibular osteotomy) apart from BSSO. All plates were inserted and removed intra-orally, without the use of a trochar. 159 patients who met the inclusion criteria were identified and selected from the hospital’s database. The case records of 6 patients were not found, leaving 153 patients (96%) for analysis.
90 patients (59%) were female and 63 (41%) were male, giving a 0.7–1 male:female ratio. At the time of osteotomy the age range was 17.0–56.4 years (average 35.1 years). The mean follow-up time was 23.2 months (range 32 days to 4.4 years). Three patients had a follow-up time of less than 100 days and seven patients a follow-up time of more than 100 days but less than 200 days.
The reasons for osteotomy were mandibular hypoplasia in 145 patients (95%), 8 of whom also had mandibular asymmetry, and mandibular hyperplasia in 7 (5%), 3 of whom also had mandibular asymmetry. One patient (0.7%) was diagnosed with skeletal class I occlusion with mandibular asymmetry.
308 titanium miniplates had been inserted in 153 patients. Two patients each had three plates inserted because of an unfavourable fracture of the mandible (i.e. bad split). The following types of plates were inserted: Martin ® 2.0 mm plate in 131 patients; Lorenz ® 2.0 mm plate in 15 patients; and Synthes ® 2.0 mm plate in 6 patients. The plates are 1.1 mm thick and the monocortical screws have a diameter of 2.0 mm. The case records of two patients did not contain data on the type of plate used.
In 148 patients (97%), an intra-oral drain was inserted until the first postoperative day to minimize postoperative hematoma. A surgical splint was used in all patients during surgery and for the first 4 postoperative weeks. Postoperative maxillomandibular fixation (MMF) was not used, but all patients had rubber elastics inserted for occlusal guidance. Peri- and postoperative antibiotics were administered to all patients. Most patients were hospitalized for 2 days. During this period, all patients received antibiotics, usually penicillin or first-generation cephalosporin administered i.v. In the event of penicillin allergy, clindamycin was used. All patients were prescribed a course of antibiotics for 1 week after leaving hospital.
The outcome variable for the present study was removal of at least one symptomatic plate and was initially categorized into three groups: not removed; removed due to plate-related complications (infection or hardware failure); and removed due to subjective discomfort (sensitivity to cold, palpability or due to discomfort about having a foreign substances/objects in the body). Subsequently, the last two categories were combined, and a dichotomous variable indicating the need for removal was created (1 = yes, 0 = no).
Explanatory variables included age in years at the time of operation, gender, smoker or non-smoker, duration of operation in minutes, type of operation (i.e. mandibular advancement, set back, or rotation), and the extent of skeletal movement (in mm). The patient’s smoking status (i.e. smoker or non-smoker) was inter alia recorded by the anaesthesiologist onto the anaesthesia forms. The duration of the operation was also recorded on this form. The type of operation and extent of skeletal movement were identified from the cephalometric prediction and the case operation files.
The χ 2 -test was used to evaluate the statistical significances of the associations between the explanatory variables and the dichotomous outcome variables. Logistic regression analysis was fitted for the outcomes, and the results were presented as odds ratios (OR), with 95% confidence intervals (CI) (SAS 9.1.3).
29 patients (19%) had at least one miniplate removed during the follow-up ( Figs 1–3 ). Seven patients were male (24%) and 22 were female (76%). Plate removal was bilateral in 28 cases despite the fact that the reason for removal was unilateral in 14 cases. 56 plates (18%) were removed due to various symptoms.
Figure 4 summarizes the reasons for plate removal. The reasons were categorized into plate-related complications for 16 patients ( Table 1 ) and subjective discomfort for 13 patients ( Table 2 ). Plate-related complications comprised infection for 12 cases and screw loosening for 4 cases. No plate fractures were observed. In the group with subjective discomfort: 5 patients wanted removal due to palpability; 4 due to cold sensitivity; and 4 because they did not want any ‘foreign objects/substances’ in the body.
|No. of patients||12||4||16|
|% of total||75.0%||25.0%||100.0%|
|Average time from insertion to removal in days||325||248||306|
|No. of patients with unilateral plate-related complication||10||3||13|
|No. of patients with bilateral plate-related complication||2||1||3|
|No. of patients with bilateral plate removal||12||4||16|
|Sensitivity to cold||Palpability||Discomfort for foreign matter||Total|
|No. of patients||4||5||4||13|
|% of total||30.8%||38.5%||30.8%||100.0%|
|Average time from insertion to removal in days||605||517||724||608|
|No. of patients with unilateral subjective discomfort||1||1|
|No. of patients with bilateral subjective discomfort||4||4||4||12|
|No. of patients with unilateral plate removal||1||1|
|No. of patients with bilateral plate removal||4||4||4||12|
29 plates (52%) were removed during the first postoperative year. All patients with infection were primarily treated as outpatients with oral antibiotics. One patient was later hospitalized because of an abscess that required incision and drainage. Of the four patients who had screw loosening, one had no symptoms and the diagnosis was made from a routine follow-up radiograph.
No statistically significant association between plate removal and age could be observed, but the mean age of the patients who had plate removal because of sensitivity to cold (40.1 years) was higher than in any other group.
There was a weak and statistically non-significant, association between plate removal and gender. Female patients underwent plate removal more often than males. In the group that had plate-related complications, plates were removed from 4 males and 12 females. In the group that had subjective discomfort, plates were removed from 3 males and 10 females. In the group that had hyperaesthesia to cold, all four patients were female.
Most patients (150, 98%) had no systemic diseases so no statistical analysis was performed on the variable of general health. One patient had rheumatoid arthritis (without immunosuppressive medication), another had juvenile-onset rheumatoid arthritis (without immunosuppressive medication), and a third patient had juvenile-onset diabetes mellitus. Of these three patients, only the individual who had rheumatoid arthritis underwent plate removal. The reason in this case was subjective discomfort (i.e. plate palpability) and plate removal was unilateral.
34 patients (22%) were smokers. The records of nine patients contained no data on their smoking status. There was an increased risk for plate removal among subjects who reported smoking at baseline compared with non-smokers (OR 2.1, 95% CI 0.9–5.1). In the adjusted analysis, smoking was the only significant predictor (OR 2.9, CI 1.1–7.9) for plate removal ( Table 3 ). 10 smokers and 18 non-smokers had their plates removed. In the group which underwent plate removal due to infection (12 cases), 3 patients were smokers.
|(95% CI)||Beta||SE||n = 132 OR (95% CI)||p|
|Female||2.1 (0.9–5.1)||0.319||0.506||1.4 (0.5–3.7)||0.529|
|Age (continuous variable)||1.0 (0.9–1.0)||−0.03||0.029||1.0 (0.9–1.0)||0.292|
|Operation time (continuous variable)||1.0 (0.99–1.01)||−0.007||0.007||1.0 (0.98–1.01)||0.336|
|Set back||4.6 (0.6–34.1)||1.015||1.147||2.8 (0.3–26.1)|
|Yes||1.7 (0.7–4.2)||0.552||0.553||1.7 (0.6–5.1)||0.318|
|Amount of movement (continuous variable)||1.0 (0.8–1.4)||0.129||0.175||1.1 (0.8–1.6)||0.458|