Long-Term Outcome of Orthognathic Surgery

Orthognathic surgery is treatment which often leads to great improvement in function and appearance, but is demanding to the patient, considerable resources are being spent, and risks for unwanted side effects are involved. Because of the elective nature of the treatment it is particularly important that the patient is well informed about all aspects of the provision and outcome of treatment before the decision to initiate treatment is reached. This information should be evidence-based. In this area it is, however, difficult to establish research materials that are both homogenous and of adequate size. The present chapter represents an attempt to summarize research findings from long-term follow-up of more than 1000 patients treated according to an established protocol over a period of 15 years.

stability based on initial anatomy and various surgical procedures has been proposed (Proffit et al., 1996). Relapse may be presented as the mean post-surgical change, or as the rate of subjects with a clinically significant relapse, usually defined as change of 2 mm or more. Another issue of importance is when the relapse occurs, and hence short-and longterm stability should be examined. Skeletal relapse may also to a varying extent impact on occlusal stability.

Skeletal stability
The results presented below is a compilation of findings from studies performed at the University of Oslo. The cephalometric variables have been recorded as x-and y-coordinates (Fig. 1). The x-axis was constructed 7 to the nasion-sella line. The most common osteotomies for management of Class II and III skeletal malocclusions will be addressed.

Class III: mandibular setback
The bilateral sagittal split osteotomy (BSSO) is a versatile and widely used approach for correction of mandibular prognathism. In a study of 80 (46 males, 34 females) consecutively operated patients (Mobarak et al., 2000a) this procedure appeared to be fairly stable. Three years post-surgery the mean relapse was 1.6 mm and represented 26% of the surgical setback at Pogonion (Pg) (6.3 mm). In 36% of the patients relapse was clinically significant (≥ 2 mm). Most of the relapse took place during the first 6 months after surgery (72%) (Fig. 2). The magnitude of the setback was to some extent associated with the amount of horizontal relapse (r = 0.39, P < 0.01). Gender differences in relapse were small. An alternative approach to BSSO is the vertical ramus osteotomy (VRO). Traditionally the VRO has required intermaxillary fixation for 6-8 weeks. Since 1995, stable plate fixation has routinely been used with an extraoral incision by our team (Hoegevold et al., 2001) (Fig. 3). The main morphologic indication for applying the extraoral VRO (EVRO) has been a moderate setback without rotation. Another indication is related to reduced risk for sensory disturbances. In a 1-year follow-up study post-operative changes were mainly in the form of a small anterior relapse tendency of about 10% of the surgical setback. Mean relapse was 0.6 mm (SD 1.2 mm). A relapse ≥ 2 mm was observed in 14% of the subjects (Mobarak et al., 2000b).

Class III: maxillary advancement
After the LeFort I osteotomy became common in the 1970-ies, this is an increasingly used alternative to correct Class III malocclusion. This approach allows treatment based on the diagnosis in cases where the discrepancy is caused by a retrognathic maxilla (Fig. 4). In addition to be associated with few unwanted side-effects like nerve injuries, a negative impact on the airways which may occur after mandibular setback, is avoided. In a sample of 43 individuals having had maxillary advancement of ≥ 2 mm (mean advancement 5 mm), the mean relapse was 0.9 mm (18%), and the relapse was clinically significant in 14% of the patients (Dowling et al., 2005). Almost all (89%) of the relapse took place during the first 6 months. Regression analysis identified large advancements and downward movement of the anterior maxilla as risk factors for horizontal relapse.

Class III: bimaxillary surgery
In patients with severe skeletal discrepancies single jaw surgery may not be possible or may have an unwanted effect on the patient's facial appearance. In later years, there has been a trend for increased use of a combination of mandibular setback and maxillary advancement. Severe skeletal Class III malocclusions are frequently associated with anterior open bite, which is more readily corrected with bimaxillary surgery and may also ensure optimal facial harmony ( Fig. 5). A study of stability comprised 81 patients (Jakobsone et al., 2011a) who pre-surgery had a mean negative overjet of -7 mm (range -18.9 to 3.2 mm) and a negative overbite of -1.8 (range -9.7 to 6.4 mm). Skeletal stability assessed 3 years after surgery varied depending on the direction and amount of the surgical repositioning. Maxillary advancement was stable, whereas a significant mean relapse was observed after mandibular setback (Fig. 6). About half of the 62 subjects with a setback of 2 mm or more, had a clinically significant relapse. Most of the skeletal relapse occurred during the first 6 months after surgery. Regression analysis showed that stability was increased when the setback was small and the posterior maxilla was impacted.

Class II: mandibular advancement
The BSSO is the most frequently used osteotomy for management of mandibular retrognatism. High-angle and low-angle Class II cases represent two distinct entities, with different facial patterns, treatment goals, and clinical challenges.
A study of stability analysed these two categories separately in 61 consecutive cases (Mobarak et al., 2001a). High-(n = 20) and low-angle (n = 20) patients had different patterns of surgical and post-operative skeletal changes. High-angle cases were associated with greater horizontal relapse. Relapse in the low-angle cases occurred early in the postoperative period, whereas in the high-angle cases, relapse was a more continuous process with a significant proportion occurring later in the follow-up period (Fig. 7). Increasing the anterior facial height (low-angle) was a relatively stable procedure, with on average 70% of the improvement remaining after 3 years.

Class II: bimaxillary surgery
Our studies of stability after bimaxillary surgery to correct skeletal Class II malocclusion have focused on subjects with a high-angle facial pattern as this group represents the greatest challenge. In a study of 31 patients (ML/NSL > 38) the stability of maxillary impaction and mandibular advancement varied considerably (Winter et al., 2010). Stable results (relapse < 2 mm) after maxillary impaction were observed in 77% of the subjects, whereas mandibular advancements were stable in only 42%. Despite the risk for relapse, some patients showed excellent results after 3 years (Fig. 8).

Class II: advancement genioplasty
Genioplasty allows 3-dimensional control of chin position, resulting in significant improvement of facial aesthetics whether performed separately or combined with other osteotomies. Of the actual corrections of the chin, advancement genioplasty to improve a receding chin is probably the most common. Our study of stability after advancement genioplasty included 21 subjects who had no additional osteotomies (Shaughnessy et al., 2006) (Fig. 9). Mean advancement was 8.4 mm. The results 3 years after surgery showed that this was a stable procedure with a mean relapse of only 8% of the advancement. Some of the post-surgical change could be ascribed to remodelling in the area rather than instability of the segment.

Open bite correction by isolated maxillary surgery
Open bite occurs both in Class I, II, and III malocclusions. When the sagittal discrepancy is moderate, the occlusion can usually be corrected by LeFort I impaction as the only procedure (Fig. 10). In a study of 40 consecutively operated patients (Espeland et al., 2008a) it was observed that impaction of posterior maxilla (≥ 2 mm) relapsed by one third. Inferior movement of the anterior maxilla relapsed about two thirds. Most of the relapse occurred during the first 6 months after surgery.  Fig. 11 summarizes frequencies of stable results (relapse < 2 mm) after various one-jaw osteotomies. The most stable results were observed after advancement of the maxilla, setback of the mandible (EVRO), and advancement of the chin. The least stable procedures were advancement of the mandible in patients with high-angle facial pattern and downward movement of the maxilla. In the literature a number of factors have been proposed which may explain skeletal relapse. For Class III anomalies early relapse has been associated with clockwise rotation of the proximal segment during BSSO-surgery (Proffit et al., 1996). In our study (Mobarak et al., 2000a), however, this did not seem to be responsible for marked relapse. Late relapse has often been ascribed to late mandibular growth which our findings also indicated. For Class II corrections by BSSO advancement, early relapse has been suggested to be due to suboptimal positioning of the condyle, whereas condylar resorption might be responsible for relapse occurring later. Risk factors for condylar resorption have been addressed in the literature, for example condylar morphology and bone quality (Hoppenreijs et al., 1999).

Occlusal stability
Despite some skeletal relapse after most surgical corrections, the anterior occlusion is generally stable. In the studies addressing stability referred to above, the incisor relationship was also examined. Below is a short summary of the main findings. After Class III correction by one-jaw surgery, both with setback of the mandible (Mobarak et al., 2000a) and advancement of the maxilla (Dowling et al., 2005), all patients examined had a positive overjet 3 years after the operation. Bimaxillary surgery resulted in positive overjet in 79 of 81 patients (Jakobsone et al., 2011a). Relapse of open bite was observed in 8 subjects, and in 5 of these, the negative overbite was less than 1 mm. Fig. 11. Diagrammatic representation of stability after various one-jaw surgical procedures. Bars represent frequency of patients with relapse less than 2 mm 3 years after surgery.
After Class II correction with mandibular advancement, the difference in skeletal stability between patients with low-and high-angle facial pattern was reflected in the stability of the occlusion. After 3 years 85% of the low-angle patients had overjet less than 4 mm, whereas the corresponding figure for the high-angle subjects was 60%. In the group of 41 high-angle patients treated with bimaxillary surgery, one third of those having overjet > 6 mm before surgery, still had an overjet > 6 mm after 3 years. Of 40 open-bite patients treated by isolated LeFort I impaction, 88% had positive overbite 3 years post-surgery (Espeland et al., 2008a). In all the remaining patients, the negative overbite was less than 1 mm. The skeletal relapse was counteracted by dentoalveolar compensation, which contributed to approximately 50% of the correction of the overbite. For all categories of malocclusion, the skeletal relapse generally took place during the first 6 months. A negative effect on the occlusion was avoided for most patients by dentoalveolar compensation, which underlines the importance of post-surgical monitoring and orthodontic management if signs of relapse become apparent.

Soft tissue response
Software for prediction of the soft tissue response has become an integral part of treatment planning aiming to provide a realistic estimate of the outcome of surgery. Ratios (or percentages) for soft to hard tissue changes have focused in particular on nose, lips, mentolabial fold, and chin. It is important that the data which serve to generate software prediction programs are valid. Changes in sagittal and vertical dimensions as well as lip thickness have been addressed in studies at our department.

Class III: mandibular setback
The soft tissue response for the lower lip, mentolabial fold and soft tissue chin to the underlying hard tissues was about 1:1. Some effect was also seen for the upper lip (Fig. 12).
The main effect of mandibular setback on the soft tissue profile included an increase in facial convexity, straightening and lengthening of the upper lip with a concomitant increase in nasolabial angle, and deepening of the mentolabial fold (Mobarak et al., 2001b).

Class III: bimaxillar surgery
To quantify the relative soft tissue response to the skeletal changes, a sample of 80 patients were divided in 3 subgroups according to pre-operative characteristics: 1) open bite, 2) positive overbite and the upper lip resting on upper incisors, and 3) positive overbite and the upper lip resting on lower incisors (lip block) (Jakobsone et al., 2011b). In the first two groups, there were strong correlations between the horizontal movement of upper incisors and upper lip (r = 0.77 and 0.85, respectively). The upper lip followed the maxilla with a ratio of 0.5:1. When the upper lip rested on lower incisors before surgery, the association between maxillary repositioning and upper lip changes were weak. In all groups a strong association between horizontal soft and hard tissue changes of lower lip and chin was observed. The vertical position of the tip of the nose was affected by both the vertical and horizontal repositioning of the maxilla. Several factors influencing the soft tissue response were identified. Software prediction programs should preferably take into account different pre-surgical characteristics.

Class II: mandibular advancement
The ratio for the soft to hard tissue response after mandibular advancement was about 1:1 for the chin and 0.6:1 for the lower lip. For the mentolabial fold, the ratio was on average 0.9:1, but varied somewhat according to vertical facial pattern (Mobarak et al., 2001c). Following mandibular advancement the facial profile became straighter. The mentolabial fold became more shallow, especially in low-angle subjects who also had an increase in anterior facial height.

Class II: advancement genioplasty
In our study of soft tissue response to advancement of the chin as the only surgical procedure, the results showed that the ratio for soft to hard tissue movement was 0.9:1 (Shaughnessy et al., 2006). The mentolabial depth increased as a result of treatment (Fig. 9).

Effect on upper airways
In addition to the changes in masticatory function and facial harmony, repositioning of the jaws may also impact on airway morphology. Mandibular advancement and setback osteotomies have been reported to influence the position of the hyoid bone, and consequently tongue position, pharyngeal airway morphology, and head flexion (Achilleos et al., 2000a,b). Studies have also shown that in adults, there is on average a reduction in the sagittal dimension of the minimal pharyngeal airway space with increasing age (Kollias & Krogstad, 1999). Over the last two decades the capacity of maxillo-mandibular osteotomies to induce or resolve obstructive sleep apnoea (OSA) has received increased attention. A challenge in the planning of orthognathic surgery is to balance concerns related to facial aesthetics and airway changes. A recent study of bimaxillary correction of Class III malocclusion examined whether maxillary advancement and/or impaction had the potential to compensate for the negative effect of mandibular setback on airways (Jakobsone et al., 2011c). Sagittal dimensions of the pharynx were measured on 4 different levels (Fig. 13) in 4 subgroups. The long-term change (3 years) at the various levels related to whether the maxilla was advanced and/or impacted or not, appears from Fig. 14. Advancement of the maxilla resulted in an increase of 15-20% at the nasopharyngeal level. At the hypopharyngeal level, advancement of the maxilla did to some extent compensate for the negative effect of mandibular setback. A decrease of 5-10% was observed at the oropharyngeal and retrolingual (PASmin) levels.

Sensory function
Impairment of sensory function is the most common long-term side effect after orthognathic surgery, especially following sagittal split osteotomies (Westermark et al., 1999). In our sample of 381 patients operated with bilateral sagittal split osteotomy (BSSO), 47.8% reported altered sensory function 3 years after the operation (Table 1) Among patients with impaired sensation (n=181), 40.9% indicated that the impairment was of concern to them because it affected their daily life. A mild distress was reported by 28.2%, 8.3% reported a moderate distress, and 4.4% said the situation caused severe distress. The age at the time of surgery was significantly associated with self-reported alterations in sensory function (P < 0.001). Distress due to impaired sensation also increased with age (P = 0.024) (Fig. 15).

Patients' satisfaction with treatment results
Orthognathic surgery may have an impact on quality of life (Cunningham et al., 2002). At a clinical review 3 years after surgery, patients monitored by our team fill in a questionnaire addressing their opinions of the treatment result in terms of improvement in dental and facial appearance, chewing ability, speech, impact on social life, satisfaction with overall result, and whether they would re-elect surgery with their present experience (Espeland et al., 2008b Table 3. Patients' reports on satisfaction/dissatisfaction related to surgical approach. Significant difference between approaches (P < 0.001).
Relatively more males than females were satisfied, and correction of Class III malocclusion was associated with increased frequency of satisfied patients. The patients' stated reasons for dissatisfaction were allocated to the following categories: impaired nerve function (n=16), relapse (n=18), appearance (n=17), TMJ problems (n=8), occlusal function (6), and other reasons (n=11).

Concluding remarks
Orthognatic surgery may on the basis of systematic monitoring of outcomes according to our established protocols be regarded as a treatment modality that predictably leads to the correction of severe occlusal anomalies. Skeletal corrections are also generally stable and even if some relapse may occur, the occlusion is usually not affected due to orthodontic compensation during the first post-surgical months. Most patients are also satisfied with the result. Improvement in facial soft-tissue profile and the effect on upper airways are less predictable. Class II malocclusions represents a greater challenge compared to other anomalies, especially in patients with high-angle facial pattern.

Acknowledgment
The authors are greatly indebted to Håkon Størmer for excellent support in preparation of the illustrations.