YBJOM-4390; No. of Pages 6
ARTICLE IN PRESS Available online at www.sciencedirect.com
British Journal of Oral and Maxillofacial Surgery xxx (2014) xxx–xxx
Recovery of mouth-opening after closed treatment of a fracture of the mandibular condyle: a longitudinal study E.T. Niezen a , I. Stuive b , W.J. Post c , R.R.M. Bos d , P.U. Dijkstra e,∗ a
University of Groningen, University Medical Centre Groningen, Groningen, The Netherlands University of Groningen, University Medical Centre Groningen, Centre for Rehabilitation, Groningen, The Netherlands c University of Groningen, Faculty of Behavioural and Social Sciences, Department of Special Needs Education and Childcare, Groningen, The Netherlands d University of Groningen, University Medical Centre Groningen, Department of Oral and Maxillofacial Surgery, Groningen, The Netherlands e University of Groningen, University Medical Centre Groningen, Department of Oral and Maxillofacial Surgery, Centre for Rehabilitation, School for Health Research, Groningen, The Netherlands b
Accepted 11 November 2014
Abstract The aim of this retrospective study was to assess recovery of mouth opening after closed treatment of fractures of the mandibular condyle, and analyse which characteristics might influence recovery. We measured mouth opening in 142 patients (mean (SD) age 30 (14) years, 96 of whom were male) during follow-up at 3, 6, 13, 26, and 52 weeks after the injury. Fractures were assessed on radiographs. Data were analysed using a multilevel analysis. Half the fractures were of the low condylar neck (n = 71). Thirty-seven patients had bilateral condylar fractures, 29 had dislocated fractures, and in 80 the fracture was displaced. One or more additional mandibular fractures were present in 68. During follow-up mean (SD) mouth opening increased to: 33.6 (9.6) at 3 weeks, 40.1 (10.0) at 6 weeks, 45.1 (9.6) at 13 weeks, 49.8 (9.5) at 26 weeks, and 52.6 (7.5) at 52 weeks. Older age, female sex, displaced fracture, bilateral fractures, additional mandibular fractures, and the interaction between follow-up time and additional mandibular fractures, were predictors of a less favourable recovery of mouth opening. Clinicians can use the results of this study to predict recovery of mouth opening after closed treatment of fractures of the mandibular condyle at first consultation. © 2014 The British Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.
Keywords: Fracture of the mandibular condyle; Recovery; Mouth opening; Closed treatment; Cohort study
Introduction There is still debate about whether open reduction and internal fixation or closed treatment is the treatment of choice for fractures of the mandibular condyle.1 Factors that influence this choice include type of fracture, associated mandibular ∗ Corresponding author at: University of Groningen, University Medical Centre Groningen, Department of Oral and Maxillofacial Surgery, Centre for Rehabilitation, PO BOX 30.001, 9700 RB Groningen, The Netherlands. Tel.: +31 50 3610297. E-mail address: [email protected] (P.U. Dijkstra).
fractures, and risk of complications such as loss of height of the ramus, malocclusion, anterior open bite, injury to the facial nerve, chronic pain, reduced mandibular function, deviation in mouth opening, restricted mouth opening, and ankylosis.1–8 Whether or not patients regain their prefracture mouth opening is not clear because it is not known. It is therefore necessary to compare mouth opening after the fracture with available normative values. Factors associated with a restricted mouth opening after fracturing the mandibular condyle include gender, age at the time of the accident, bilateral condylar fractures, number of additional mandibular fractures, exercise compliance, maxillomandibular fixation
http://dx.doi.org/10.1016/j.bjoms.2014.11.007 0266-4356/© 2014 The British Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Niezen ET, et al. Recovery of mouth-opening after closed treatment of a fracture of the mandibular condyle: a longitudinal study. Br J Oral Maxillofac Surg (2014), http://dx.doi.org/10.1016/j.bjoms.2014.11.007
YBJOM-4390; No. of Pages 6
2
ARTICLE IN PRESS E.T. Niezen et al. / British Journal of Oral and Maxillofacial Surgery xxx (2014) xxx–xxx
and closed treatment.4,5,9–13 However, others have found that closed treatment resulted in a (slightly) wider mouth opening than open reduction.14 Longitudinal studies that have analysed several of these factors together are scarce, and their influence on mouth opening, corrected for the influence of other factors (confounders), is limited.13,14 It is therefore not possible to inform patients with a condylar fracture about their expected recovery of mouth opening at the first consultation. Clinically, mouth opening is relevant because it is associated with mandibular function.10 The aim of this longitudinal study was to assess the influence of characteristics of patients and fractures on the recovery of mouth opening after closed treatment of fractures of the mandibular condyle.
Patients and methods Data from patients who were referred to the Department of Oral and Maxillofacial Surgery of the University Medical Centre Groningen between March 1998 and July 2002 with a fracture of the mandibular condyle and who had participated in a previous study were reanalysed.9 The inclusion criterion was a fracture of the mandibular condyle within the past week that was shown on radiographs (panoramic, Towne projection, transpharyngeal, or transcranial) or computed tomographic (CT) scanning. Patients with a history of psychiatric disorders, mental retardation, inability to understand Dutch, or with limited mouth opening before the fracture, were excluded. Assessment of fractures The position of the fractured segments and type of fracture were assessed on radiographs (by one senior staff member of the department). In intra-articular fractures the fracture line runs within the capsule of the temporomandibular joint. In fractures of the high condylar neck the fracture line runs through or above the mandibular notch, whereas in those of the low condylar neck the fracture line runs below the level of the mandibular notch. The fracture was recorded as dislocated if the condylar head was in front of the articular eminence, or if the proximal segment made an angle, relative to the distal segment, of 50◦ or more, laterally or medially. It was recorded as displaced when the proximal segment was displaced relative to the distal segment, and there was no overlap between the two segments (grossly displaced, minor, or not displaced). The segments were recorded as deviated if the proximal and distal segments made contact, and if there was angulation between the segments.9,10,15 Additional mandibular fractures were also recorded. Treatment Fractures were treated according to “closed treatment” principles, which are the standard procedures of the department
of Oral and Maxillofacial Surgery of the University Medical Centre Groningen, the Netherlands, as described in detail previously.9 Rigid intermaxillary fixation was not used. Fractures of other parts of the mandible were treated according to the principles of open reduction using internal fixation with plates and screws (2.0–2.3 mm Martin® system, Martin GmbH, Germany). Follow-up and assessment During follow-up (3, 6, 13, 26, and 52 weeks after injury) maximum mouth opening was measured using Vernier calipers, as the maximum interincisal distance (between teeth 11 and 41) added to the vertical overlap.16 If teeth 11 or 41 were damaged, the measurements were made between teeth 21 and 31. If the patients failed to attend follow-up a new appointment was sent. If the patient failed again, a reminder was sent. All participants gave oral and written consent. This study was approved by the medical ethics review board of the University Medical Centre Groningen. Entry of data and statistical analysis Missing data about the fractures were retrieved from medical records. If the missing data were not found in the records, the radiographs were reassessed. Missing data about vertical overlap were calculated as follows. If the vertical overlap for a patient was available from another follow-up, that vertical overlap was used. If the vertical overlap was available of two or more follow-up visits, the vertical overlap measured nearest to the “missing” one was used. If no data about vertical overlap were available, then the median vertical overlap of the total group (3 mm) was used. In case of bilateral fractures, the fracture nearest to the joint was used for analysis. We assumed that a fracture nearer to the joint would have more impact on mouth opening than a more distant fracture. To take into account the correlation between measurements within patients, we made a 2-level, multilevel analysis in MLwin (Version 2.27). In a multilevel analysis all available data for all participants are used, whereas in a repeated measurement ANOVA only complete cases without missing follow-up data are used. The highest level was “patient” and the lowest level was the repeated measurements of mouth opening during follow-up. In this way the results of a multilevel analysis can roughly be interpreted as the results of ordinary regression analysis, where the variability both between and within patients are included. With respect to the predictor variables, two groups of dummy variables were created, one group of which concerned the height of the fracture, “intra-articular fracture” and “high condylar neck fracture” were taken together, and the reference group was “low condylar neck fractures”. The other dummy variables concerned displacement – one group was “gross displacement”
Please cite this article in press as: Niezen ET, et al. Recovery of mouth-opening after closed treatment of a fracture of the mandibular condyle: a longitudinal study. Br J Oral Maxillofac Surg (2014), http://dx.doi.org/10.1016/j.bjoms.2014.11.007
YBJOM-4390; No. of Pages 6
ARTICLE IN PRESS E.T. Niezen et al. / British Journal of Oral and Maxillofacial Surgery xxx (2014) xxx–xxx
and “minor displacement”, the reference group for which was “no displacement”. (Coding: 1 = yes, 0 = no). Mouth opening was the dependent variable and possible predictors were follow-up time (weeks since the fracture), sex, age at the time of the injury, and details of the fracture (height, displacement, deviation, bilateral or unilateral, and number of additional mandibular fractures). The influence of follow-up time was first looked at on a graph in which the mean of the measured mouth opening and the 95% reference range were entered on the y axis and follow-up time on the x-axis (Fig. 1). Initially the curve was steep, which indicated a fast increase in mouth opening, and over time the curve lost its steepness, which indicated a more gradual increase. As a consequence, follow-up time was entered in the regression equation after logarithmic (e) transformation. Interactions between predictors – for instance, sex and age, were explored. Predictor variables were entered one by one in the multilevel analysis and remained in the multilevel analysis regression equation if the estimated regression coefficient (b) was significant (or the model fit of the regression equation improved significantly). Probabilities of less than 0.05 were accepted as significant.
Results Two patients who had open treatment for a fracture of the mandibular condyle were excluded. In total, 142 patients fulfilled the inclusion criterion and were entered in the study (Table 1). The number of patients for whom data were available differed for each follow-up time. For follow-up at 3 weeks, data of 73 patients were available, for 6 weeks 96 patients, for 13 weeks 97 patients, for 26 weeks 76 patients, and for 52 weeks 87 patients. A total of 429 records were available for analysis. Data about vertical overlap were missing for 19 patients (23 records). For 6 patients the median of the group (3 mm) was entered into the database. For 7 patients the vertical overbite was available for 1 follow-up, and for another 6 it was available for more than one follow-up. In the multilevel analysis follow-up time, age, sex, displacement, bilateral fractures, additional mandibular fractures, and the interaction between follow-up time and additional mandibular fractures, influenced the recovery of mouth opening (Table 2).
3
Table 1 Details of patients and fractures. Data are number (%) unless otherwise stated. Variables Mean (SD) age (years) 30 (14) Sex 96 (68) Male 46 (32) Female Site of fracture 31 (22) Head 40 (28) High condylar neck Low condylar neck 71 (50) 37 (26) Bilateral Position of fracture 33 (23) Deviated Displaced 62 (44) No Minor 31 (22) 49 (35) Gross Dislocated 29 (20) Additional mandibular fractures 74 (52) 0 58 (41) 1 7 (5) 2 3 (2) 3 Mean (SD) maximal mouth opening (mm) (weeks) 33.6 (9.6) 3 40.1 (10.0) 6 13 45.1 (9.6) 49.8 (9.5) 26 52.6 (7.5) 52 Table 2 Results of multilevel analysis with follow-up time modelled as a natural logarithmic function. Independent Ln follow-up time (weeks)* Age (years) Sex (male = 1, female = 0) Gross displacement (yes = 1, no = 0)** Minor displacement (yes = 1, no = 0)** Bilateral fractures (yes = 1, no = 0) Number of other mandibular fractures Ln follow-up time × Number of other mandibular fractures Constant
b
Se b
p value
5.547
0.534
ARTICLE IN PRESS Available online at www.sciencedirect.com
British Journal of Oral and Maxillofacial Surgery xxx (2014) xxx–xxx
Recovery of mouth-opening after closed treatment of a fracture of the mandibular condyle: a longitudinal study E.T. Niezen a , I. Stuive b , W.J. Post c , R.R.M. Bos d , P.U. Dijkstra e,∗ a
University of Groningen, University Medical Centre Groningen, Groningen, The Netherlands University of Groningen, University Medical Centre Groningen, Centre for Rehabilitation, Groningen, The Netherlands c University of Groningen, Faculty of Behavioural and Social Sciences, Department of Special Needs Education and Childcare, Groningen, The Netherlands d University of Groningen, University Medical Centre Groningen, Department of Oral and Maxillofacial Surgery, Groningen, The Netherlands e University of Groningen, University Medical Centre Groningen, Department of Oral and Maxillofacial Surgery, Centre for Rehabilitation, School for Health Research, Groningen, The Netherlands b
Accepted 11 November 2014
Abstract The aim of this retrospective study was to assess recovery of mouth opening after closed treatment of fractures of the mandibular condyle, and analyse which characteristics might influence recovery. We measured mouth opening in 142 patients (mean (SD) age 30 (14) years, 96 of whom were male) during follow-up at 3, 6, 13, 26, and 52 weeks after the injury. Fractures were assessed on radiographs. Data were analysed using a multilevel analysis. Half the fractures were of the low condylar neck (n = 71). Thirty-seven patients had bilateral condylar fractures, 29 had dislocated fractures, and in 80 the fracture was displaced. One or more additional mandibular fractures were present in 68. During follow-up mean (SD) mouth opening increased to: 33.6 (9.6) at 3 weeks, 40.1 (10.0) at 6 weeks, 45.1 (9.6) at 13 weeks, 49.8 (9.5) at 26 weeks, and 52.6 (7.5) at 52 weeks. Older age, female sex, displaced fracture, bilateral fractures, additional mandibular fractures, and the interaction between follow-up time and additional mandibular fractures, were predictors of a less favourable recovery of mouth opening. Clinicians can use the results of this study to predict recovery of mouth opening after closed treatment of fractures of the mandibular condyle at first consultation. © 2014 The British Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.
Keywords: Fracture of the mandibular condyle; Recovery; Mouth opening; Closed treatment; Cohort study
Introduction There is still debate about whether open reduction and internal fixation or closed treatment is the treatment of choice for fractures of the mandibular condyle.1 Factors that influence this choice include type of fracture, associated mandibular ∗ Corresponding author at: University of Groningen, University Medical Centre Groningen, Department of Oral and Maxillofacial Surgery, Centre for Rehabilitation, PO BOX 30.001, 9700 RB Groningen, The Netherlands. Tel.: +31 50 3610297. E-mail address: [email protected] (P.U. Dijkstra).
fractures, and risk of complications such as loss of height of the ramus, malocclusion, anterior open bite, injury to the facial nerve, chronic pain, reduced mandibular function, deviation in mouth opening, restricted mouth opening, and ankylosis.1–8 Whether or not patients regain their prefracture mouth opening is not clear because it is not known. It is therefore necessary to compare mouth opening after the fracture with available normative values. Factors associated with a restricted mouth opening after fracturing the mandibular condyle include gender, age at the time of the accident, bilateral condylar fractures, number of additional mandibular fractures, exercise compliance, maxillomandibular fixation
http://dx.doi.org/10.1016/j.bjoms.2014.11.007 0266-4356/© 2014 The British Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Niezen ET, et al. Recovery of mouth-opening after closed treatment of a fracture of the mandibular condyle: a longitudinal study. Br J Oral Maxillofac Surg (2014), http://dx.doi.org/10.1016/j.bjoms.2014.11.007
YBJOM-4390; No. of Pages 6
2
ARTICLE IN PRESS E.T. Niezen et al. / British Journal of Oral and Maxillofacial Surgery xxx (2014) xxx–xxx
and closed treatment.4,5,9–13 However, others have found that closed treatment resulted in a (slightly) wider mouth opening than open reduction.14 Longitudinal studies that have analysed several of these factors together are scarce, and their influence on mouth opening, corrected for the influence of other factors (confounders), is limited.13,14 It is therefore not possible to inform patients with a condylar fracture about their expected recovery of mouth opening at the first consultation. Clinically, mouth opening is relevant because it is associated with mandibular function.10 The aim of this longitudinal study was to assess the influence of characteristics of patients and fractures on the recovery of mouth opening after closed treatment of fractures of the mandibular condyle.
Patients and methods Data from patients who were referred to the Department of Oral and Maxillofacial Surgery of the University Medical Centre Groningen between March 1998 and July 2002 with a fracture of the mandibular condyle and who had participated in a previous study were reanalysed.9 The inclusion criterion was a fracture of the mandibular condyle within the past week that was shown on radiographs (panoramic, Towne projection, transpharyngeal, or transcranial) or computed tomographic (CT) scanning. Patients with a history of psychiatric disorders, mental retardation, inability to understand Dutch, or with limited mouth opening before the fracture, were excluded. Assessment of fractures The position of the fractured segments and type of fracture were assessed on radiographs (by one senior staff member of the department). In intra-articular fractures the fracture line runs within the capsule of the temporomandibular joint. In fractures of the high condylar neck the fracture line runs through or above the mandibular notch, whereas in those of the low condylar neck the fracture line runs below the level of the mandibular notch. The fracture was recorded as dislocated if the condylar head was in front of the articular eminence, or if the proximal segment made an angle, relative to the distal segment, of 50◦ or more, laterally or medially. It was recorded as displaced when the proximal segment was displaced relative to the distal segment, and there was no overlap between the two segments (grossly displaced, minor, or not displaced). The segments were recorded as deviated if the proximal and distal segments made contact, and if there was angulation between the segments.9,10,15 Additional mandibular fractures were also recorded. Treatment Fractures were treated according to “closed treatment” principles, which are the standard procedures of the department
of Oral and Maxillofacial Surgery of the University Medical Centre Groningen, the Netherlands, as described in detail previously.9 Rigid intermaxillary fixation was not used. Fractures of other parts of the mandible were treated according to the principles of open reduction using internal fixation with plates and screws (2.0–2.3 mm Martin® system, Martin GmbH, Germany). Follow-up and assessment During follow-up (3, 6, 13, 26, and 52 weeks after injury) maximum mouth opening was measured using Vernier calipers, as the maximum interincisal distance (between teeth 11 and 41) added to the vertical overlap.16 If teeth 11 or 41 were damaged, the measurements were made between teeth 21 and 31. If the patients failed to attend follow-up a new appointment was sent. If the patient failed again, a reminder was sent. All participants gave oral and written consent. This study was approved by the medical ethics review board of the University Medical Centre Groningen. Entry of data and statistical analysis Missing data about the fractures were retrieved from medical records. If the missing data were not found in the records, the radiographs were reassessed. Missing data about vertical overlap were calculated as follows. If the vertical overlap for a patient was available from another follow-up, that vertical overlap was used. If the vertical overlap was available of two or more follow-up visits, the vertical overlap measured nearest to the “missing” one was used. If no data about vertical overlap were available, then the median vertical overlap of the total group (3 mm) was used. In case of bilateral fractures, the fracture nearest to the joint was used for analysis. We assumed that a fracture nearer to the joint would have more impact on mouth opening than a more distant fracture. To take into account the correlation between measurements within patients, we made a 2-level, multilevel analysis in MLwin (Version 2.27). In a multilevel analysis all available data for all participants are used, whereas in a repeated measurement ANOVA only complete cases without missing follow-up data are used. The highest level was “patient” and the lowest level was the repeated measurements of mouth opening during follow-up. In this way the results of a multilevel analysis can roughly be interpreted as the results of ordinary regression analysis, where the variability both between and within patients are included. With respect to the predictor variables, two groups of dummy variables were created, one group of which concerned the height of the fracture, “intra-articular fracture” and “high condylar neck fracture” were taken together, and the reference group was “low condylar neck fractures”. The other dummy variables concerned displacement – one group was “gross displacement”
Please cite this article in press as: Niezen ET, et al. Recovery of mouth-opening after closed treatment of a fracture of the mandibular condyle: a longitudinal study. Br J Oral Maxillofac Surg (2014), http://dx.doi.org/10.1016/j.bjoms.2014.11.007
YBJOM-4390; No. of Pages 6
ARTICLE IN PRESS E.T. Niezen et al. / British Journal of Oral and Maxillofacial Surgery xxx (2014) xxx–xxx
and “minor displacement”, the reference group for which was “no displacement”. (Coding: 1 = yes, 0 = no). Mouth opening was the dependent variable and possible predictors were follow-up time (weeks since the fracture), sex, age at the time of the injury, and details of the fracture (height, displacement, deviation, bilateral or unilateral, and number of additional mandibular fractures). The influence of follow-up time was first looked at on a graph in which the mean of the measured mouth opening and the 95% reference range were entered on the y axis and follow-up time on the x-axis (Fig. 1). Initially the curve was steep, which indicated a fast increase in mouth opening, and over time the curve lost its steepness, which indicated a more gradual increase. As a consequence, follow-up time was entered in the regression equation after logarithmic (e) transformation. Interactions between predictors – for instance, sex and age, were explored. Predictor variables were entered one by one in the multilevel analysis and remained in the multilevel analysis regression equation if the estimated regression coefficient (b) was significant (or the model fit of the regression equation improved significantly). Probabilities of less than 0.05 were accepted as significant.
Results Two patients who had open treatment for a fracture of the mandibular condyle were excluded. In total, 142 patients fulfilled the inclusion criterion and were entered in the study (Table 1). The number of patients for whom data were available differed for each follow-up time. For follow-up at 3 weeks, data of 73 patients were available, for 6 weeks 96 patients, for 13 weeks 97 patients, for 26 weeks 76 patients, and for 52 weeks 87 patients. A total of 429 records were available for analysis. Data about vertical overlap were missing for 19 patients (23 records). For 6 patients the median of the group (3 mm) was entered into the database. For 7 patients the vertical overbite was available for 1 follow-up, and for another 6 it was available for more than one follow-up. In the multilevel analysis follow-up time, age, sex, displacement, bilateral fractures, additional mandibular fractures, and the interaction between follow-up time and additional mandibular fractures, influenced the recovery of mouth opening (Table 2).
3
Table 1 Details of patients and fractures. Data are number (%) unless otherwise stated. Variables Mean (SD) age (years) 30 (14) Sex 96 (68) Male 46 (32) Female Site of fracture 31 (22) Head 40 (28) High condylar neck Low condylar neck 71 (50) 37 (26) Bilateral Position of fracture 33 (23) Deviated Displaced 62 (44) No Minor 31 (22) 49 (35) Gross Dislocated 29 (20) Additional mandibular fractures 74 (52) 0 58 (41) 1 7 (5) 2 3 (2) 3 Mean (SD) maximal mouth opening (mm) (weeks) 33.6 (9.6) 3 40.1 (10.0) 6 13 45.1 (9.6) 49.8 (9.5) 26 52.6 (7.5) 52 Table 2 Results of multilevel analysis with follow-up time modelled as a natural logarithmic function. Independent Ln follow-up time (weeks)* Age (years) Sex (male = 1, female = 0) Gross displacement (yes = 1, no = 0)** Minor displacement (yes = 1, no = 0)** Bilateral fractures (yes = 1, no = 0) Number of other mandibular fractures Ln follow-up time × Number of other mandibular fractures Constant
b
Se b
p value
5.547
0.534
