The aims of this study were to evaluate volumetric changes in the nasal cavity (NC) and pharyngeal airway space (PAS) after Le Fort I maxillary distraction osteogenesis (MDO) using a three-dimensional (3D) simulation program, and to determine the effects of MDO on respiratory function during sleep with polysomnography (PSG). 3D computed tomography images were obtained and analyzed before surgery (T0) and at a mean 8.2 ± 1.2 months postsurgery (T1) (SimPlant-OMS software) for 11 male patients (mean age 25.3 ± 5.9 years) with severe skeletal class III anomalies related to maxillary retrognathia. The simulation of osteotomies and placement of distractors were performed on stereolithographic 3D models. NC and PAS were segmented separately on these models for comparison of changes between T0 and T1. PSG including the apnoea–hypopnoea index (AHI), sleep efficiency, sleep stages (weakness, stages 1–4, and rapid eye movement (REM)), and mean lowest arterial O 2 saturation were obtained at T0 and T1 to investigate changes in respiratory function during sleep. MDO was successful in all cases as planned on the models; the average forward movement at A point was 10.2 mm. Increases in NC and PAS volume after MDO were statistically significant. These increases resulted in significant improvement in sleep quality. PSG parameters changed after MDO; AHI and sleep stages weakness, 1, and 2 decreased, whereas REM, stages 3 and 4, sleep efficiency, and mean O 2 saturation increased.
The use of distraction osteogenesis (DO) for the treatment of craniomaxillofacial anomalies has become popular over the past 10 years. Corpus, ramus, premaxillary, and maxillary DO are the most successful and widely used techniques for the correction of congenital class III malocclusions when rigid fixation techniques are insufficient. With rigid placement of the distractors, distraction forces may be transmitted to insufficient bone, and due to the anatomy of the region, difficulties are often encountered, especially in maxillary distractions. Due to the experience gained with extraoral distractors, intraoral sub-periosteal distractors were developed and are now being used successfully.
There are few reports on the use of internal distraction in the facial skeleton region. The advantages of internal devices are that they are less conspicuous and more tolerable to the patient. However, the design and installation of internal maxillary distraction osteogenesis (MDO) devices are difficult. To advance the maxilla to a predetermined position, it is critical to locate the distractors parallel to each other on both sides of the maxilla. Once the device is fixed, the vector of motion cannot be changed. So, presurgical projection on a three-dimensional (3D) stereolithographic (SLA) model is crucial to identify the distraction vector.
Only a few studies have attempted to explore the effect of Le Fort I osteotomy on the pharyngeal airway space (PAS) in class III malocclusions. Harada et al. found increases in both nasopharyngeal depth and velar length after MDO. Mochida et al. showed that MDO in cleft lip and palate (CLP) patients led to an increase in the upper airway and a decrease in nasal resistance that remained stable even after 1 year. Aksu et al. found that anterior movement of the maxilla associated with MDO resulted in significant increases in posterior, superoposterior, and middle airway spaces in adult CLP patients.
The distraction procedure not only changes the volume of bone tissue, but also modifies the surrounding structures such as the nasal cavity (NC). The anatomical and aesthetic aspects of DO are crucial, but the importance of functional consequences might overcome these aspects. Efforts to improve occlusion and facial aesthetics, and consequently the patient’s quality of life, may have the opposite effect. Creating upper airway resistance as a result of surgical procedures could influence respiration. Although the pharynx is voluntarily dilated when the patient is awake, there may be trouble during sleep. The single-night sleep study with full polysomnography (PSG) is a useful examination for evaluating, detecting, and quantifying respiratory impairment and can be performed safely in a variety of clinical situations. In the light of these facts, a hypothesis has been proposed that the volume of the NC and PAS increase after Le Fort I MDO and that the dentomaxillofacial changes accompanying MDO may have an effect on sleep respiratory function.
Patients and methods
This study was approved by the institutional ethics committee and informed consent forms were signed by all patients. Eleven male adult patients admitted to the department of orthodontics were included in this study. They ranged in age from 18 to 33 years (mean age 25.3 ± 5.9 years) and none of them had any syndrome involving the skeletal structure. Only male subjects were included to eliminate the gender differences in pharyngeal airway changes. All diagnostic data, including cephalograms, photographs, and computed tomography (CT) images, were obtained and analyzed presurgery and at >6 months postsurgery (mean 8.2 ± 1.2 months). The patients had severe skeletal class III anomalies due to maxillary retrognathia (mean sella–nasion–A point angle (SNA) 72.28 ± 1.49°, A point–nasion–B point angle (ANB) −7.71 ± 1.25°), an excessive increased negative overjet (mean −11.83 ± 3.71 mm), an anterior and/or posterior cross-bite, and normal mandibular development (sella–nasion–B point angle (SNB) 80.00 ± 0.81°) ( Table 1 ). Since all patients were adult with mandibular growth and mandibular complex structure that was normal, and all had excessive negative overjet, treatment with MDO was preferred over conventional Le Fort I osteotomy and rigid fixation to avoid the risk of potential relapse.
|Before surgery (T0)||After surgery (T1)||T0–T1% increase||T0–T1|
|Mean||SD||95% CI for mean||Median||Mean||SD||95% CI for mean||Median||Mean difference||SD||Test statistics a||P -value|
|Lower bound||Upper bound||Lower bound||Upper bound|
|Overjet, mm||−11.83||3.71||−15.01||−9.83||−11||1.85||0.69||1.21||2.49||2||−13.68||1.34||Z = −2375||0.018 *|
|SNA,°||72.28||1.49||70.90||73.66||72||81.57||0.53||81.07||82.06||82||−9.29||1.11||t = −2401||0.016 *|
|SNB,°||80.00||0.81||79.24||80.75||80||79.57||0.53||79.07||80.06||80||0.43||0.20||t = 1732||0.083|
|ANB,°||−7.71||1.25||−8.87||−6.55||−8||2.00||0.81||1.24||2.75||2||−9.71||1.34||t = −2379||0.017 *|
|NC volume, mm 3||25,777.7||2322.4||23,629.8||27,925.5||25,413||29,413.4||2607.4||27,001.9||31,824.9||29,236||14.1%||−3635.7||876.6||t = −2366||0.018 *|
|PAS volume, mm 3||27,484.5||2890.9||24,810.8||30,158.2||27,318||29,494.2||3101.0||26,626.2||32,362.2||29,312||7.3%||−2009.7||796.3||t = −2366||0.018 *|
3D CT scans
All CT examinations were performed before surgery (T0) and at >6 months after distraction (T1) using a 64 detector CT scanner (Aquilion 64; Toshiba Medical Systems, Otawara, Japan), with the subjects in supine position. Scan parameters were as follows: 120 kV, 150 mA, 400 ms rotation time, slice thickness of <0.5 mm and increments of 0.4 mm, and detector collimation 64 mm × 0.5 mm (pitch 0.64). CT image data were transferred to a network computer workstation. The software SimPlant-OMS (Materialise NV, Leuven, Belgium) was used to build the SLA model and to perform the 3D analysis of the NC and PAS and was operated by the same investigator from the medical design and manufacturing centre. The CT scan images were used for simulation and virtual planning of the operation and also to build the SLA for distractor harmonization before surgery. Furthermore, the length and direction of the distraction, postoperative dental relationship, and the regions of bone that would be stabilized by the distractor were predetermined virtually.
The SLA model was built by a Spectrum Z510 colour 3D printer (Z-Corp., Burlington, MA, USA) using STL datasets. The distractor positioning and incision lines for simulation before the actual operation were carried out on the SLA models; the plates of the distractor were fitted properly and marked on the SLA model for transfer to the operating area ( Fig. 1 ).
The NC and PAS were segmented on the 3D virtual models. The pre- and post-treatment virtual models of the segmented regions were evaluated and compared. NC segments were restricted by the ostium of the paranasal sinuses, posterior airway, and nostrils ( Fig. 2 ). The borders of the PAS were formed from the following: (1) anterior: a vertical plane through the distal margin of the vomer, soft palate, base of the tongue, and anterior wall of the pharynx; (2) posterior: posterior wall of the pharynx; (3) lateral: the lateral walls of the pharynx; (4) upper: the roof of the nasopharynx; and (5) lower: a plane passing through the upper border of the larynx perpendicular to the sagittal plane ( Fig. 3 ).
With a classic and complete high Le Fort I osteotomy under general anaesthesia, down-fracture and complete mobilization of the maxillary segment was performed in all patients by the same surgeon of the department of maxillofacial surgery. Insertion of the device was started with the fixing of titanium miniplates (MODUS; Medartis, Basel, Switzerland) to the zygomatic buttresses and maxillary segment below the osteotomy line bilaterally, as simulated previously on the SLA model. Distraction cylinders (MODUS, MDO1.5; Medartis, Basel, Switzerland) were placed on the miniplates. After the activation and direction of the distractors were approved, the device was anchored to the miniplates. After a 7-day latency period, distraction was performed at a rate of 0.5 mm twice a day for approximately 20 days. Distraction was continued until the elimination of the negative overjet and including 1-mm overtreatment. The patients underwent a consolidation period of 60 days with the distractors in place. Following the consolidation phase, the distractors were removed under local anaesthesia.
All patients underwent a one-night sleep study in the sleep research centre before MDO and at >6 months after MDO. Airflow was monitored through oral and nasal thermistors and cannulae adapted for this purpose. The mean lowest arterial O 2 saturation was measured continuously by pulse oximetry using a finger probe. Body position was assessed continuously both with a closed-circuit camera and with a body position sensor.
All variables were recorded on a computerized system. Sleep parameters were recorded on a 32-channel polygraph (SomnoStar Alpha Series 4; SensorMedics Corp., Yorba Linda, CA, USA). Sleep respiratory information, including the apnoea–hypopnoea index (AHI), sleep efficiency, sleep stages (weakness, stages 1, 2, 3, and 4, and rapid eye movement (REM)), and mean lowest arterial O 2 saturation, was used for the data analysis. According to the PSG test results obtained before surgery, five of the 11 patients did not have any problem related with airway obstruction or snoring during sleep (AHI < 5). Six patients were diagnosed as simple snorers (AHI < 5); no patient was diagnosed with obstructive sleep apnoea syndrome (OSAS) at T0.
The descriptive analysis of preoperative and postoperative measurements was performed using SPSS 12.0.1 for Windows (SPSS Inc., Chicago, IL, USA). All pre- and post-treatment variables were compared with the paired-samples t -test, except for overjet and AHI, which did not pass the formal normality test (Kolmogorov–Smirnov). Non-parametric variables were compared with the Wilcoxon signed-rank test. P -values of <0.05 were considered statistically significant. Methodological errors in volumetric measurements were minimized by double-recording. All parameter measurements were repeated after 15 days by the same author; the Wilcoxon signed-rank test was then applied. The results of this analysis indicated that there were no statistically significant differences between the original and repeated measurements at the 0.05 level ( P > 0.05).
MDO was comfortable for the patients; its application was easy due to the position of the intraoral activating rod and it was removed after the consolidation period. No infection, dislocation, or device breakages occurred, and no lesions were encountered around the oral tissues during the distraction period. After MDO, skeletal alterations and occlusion at T1 were found to be compatible with the 3D simulation and SLA model, as visualized using the software.
Skeletal and volumetric measurements
The measured mean sagittal bone gain parallel to the skull base at A point was 10.2 mm (range 7.5–16.2 mm). As a result of the advancement of the maxillary complex, the mean negative overjet increased significantly from −11.83 ± 3.71 mm to 1.85 ± 0.69 mm ( P < 0.05). The mean value of the SNA angle increased significantly from 72.28 ± 1.49° to 81.57 ± 0.53° ( P < 0.05), but the mean change in SNB angle was not statistically significant ( P > 0.05). The average value of the ANB angle changed from −7.71 ± 1.25° to 2.00 ± 0.81° and this change was significant ( P < 0.05). The volume of NC (T0 25,777.7 ± 2322.4 mm 3 to T1 29,413.4 ± 2607.4 mm 3 ) and the volume of PAS (T0 27,484.5 ± 2890.9 mm 3 to T1 29,494.2 ± 3101.0 mm 3 ) increased significantly ( P < 0.05) ( Fig. 4 ; Table 1 ).