Surgical Complications of Orthognathic Surgery

Abstract

Objectives: Orthognathic surgery is popular and provides patients with improved aesthetics and functionality. The procedure is considered safe and predictable. Possible complications do exist and can be life-threatening. The aim of this study is to assess the prevalence of intra- and post-operative complications, and to define possible correlations between diagnoses. Study Design: Medical records of 103 patients who underwent orthognathic surgery in a 4.5-year period (2013–2017), at the Baruch Padeh “Tzafon” Medical Center in Poriya (PMC), Israel, were retrospectively collected. The data were analyzed for descriptive statistics and non-parametrical tests. Results: In total, there were 56 complications in 45 patients (43.7% probability of complication per patient). Major complication occurred in only one case (1%), moderate complications appeared in 8.7% of the cases, while the most prevalent minor complications were temporary nerve injury and hardware issues (17.5% and 15.5%, respectively). Statistical analysis revealed that bi-max surgery has significantly higher (OR 1.34, CI 99% 1.05–1.69, p = 0.019) prevalence of complications than one-jaw surgery, as did skeletal class II patients (OR 2.75, CI 99% 2.25–3.35, p = 0.022), as compared to skeletal class III patients. Conclusions: Serious complications seem to be rare in orthognathic surgery at PMC.

1. Introduction

Orthognathic surgery is used to correct congenital and acquired dentofacial discrepancies to improve both function and esthetics. The most commonly performed orthognathic procedures include Bilateral Sagittal Split osteotomy (BSSO) and Le Fort 1 Osteotomy (LF1). These osteotomies are considered relatively safe having low complication rates [1,2,3,4]. However, as in all surgeries, numerous complications of varying severity may occur. This may include the rare life-threatening complications, such as hemorrhage, that reported to occur in up to 2% of the cases [5,6,7].

The aim of this study is to review intraoperative and postoperative complications related to orthognathic surgery carried out over a four-year period in the Baruch Padeh “Tzafon” Medical Center (PMC) in Poriya, Israel, to determine the prevalence of intra- and post-operative complications.

2. Materials and Methods

This is a retrospective study approved by the Institutional Review Board of PMC (Approval #0026-18-POR). Inclusion criteria included patients who underwent orthognathic surgery at the Oral and Maxillofacial Surgery department at PMC between January 2013 and July 2017 (4.5 years). Exclusion criteria were syndromic patients, follow-up less than 12 months or missing data. Initially, 132 cases in total were identified and 29 cases were excluded. The final study group included 103 patients. The study was approved by the Institutional Review Board according to the Helsinki Declaration. All surgeries were performed by a team of 4 surgeons from the department, at least one of which was an attending surgeon. All cases were planned by a resident assigned to that case, then reviewed by an attending and finally approved by at departmental review. Our department utilizes manual planning, using models and casts to produce surgical splints. The photographs (lateral cephalograms) are manually measured to achieve cephalometric measurements pre- and post-operatively.

The medical records of the patients were analyzed according to established protocols [8] and neurosensory impairment was determined with von Frey tests and 2-Point Discrimination tests. Data collected included: age, sex, primary diagnosis (skeletal class I/II/III), secondary diagnosis (open bite, asymmetry), and type of surgery. Complications were defined as severe, moderate and minor using Clavien-Dindo Classification [9,10]: severe (grade IV)—major bleeding, skull base fracture; moderate (grade IIIb)—permanent [more the 9 months post-operative] nerve injury, necessity for re-operation; minor (grade I–IIIa)—bad split, condylar sagging, temporary nerve injury, temporo-mandibular joint [TMJ] [1,11,12] disorders, complications related to hardware, dental injuries, simple hemorrhage.

Statistical analysis was carried out using Microsoft Excel^® (Version 16.67) software for MS Windows, for 3 variables. The remaining analyses were descriptive. The comparison between subgroups was made using non-parametrical tests: Pearson Chi-Square test, ANOVA, t-test, Kruskal–Wallis test, Wilcoxon Signed Ranks test, Fisher’s Exact test, and Mann–Whitney U test. Adjustment factors were calculated by the Pearson method. A probability of <0.05 was considered significant.

3. Results

This is a retrospective study. The study group included 103 patients (60 male, 43 female), with a mean age of 23.7 ± 7.5 years (range 16–60 years). The most prevalent (67.9%) preoperative diagnosis was Skeletal Class III (CLIII), frequently combined with open bite (54%), followed by Skeletal Class II (CLII) (

Table 1. Diagnoses Distribution.

Skeletal Jaw Relations	Number (%) of Patients	Secondary Diagnosis	Number (% within Group) of Patients *
Class III	70 (67.9%)	None	25 (36%)
		Open Bite	38 (54%)
		Asymmetry	14 (20%)
Class II	21 (20.4%)	None	13 (62%)
		Open Bite	8 (38%)
		Asymmetry	0 (0%)
Class I	12 (11.6%)	None	9 (75%)
		Open Bite	2 (17%)
		Asymmetry	1 (8%)
Total	103 (100%)

* Possibility of results exceeding 100%, as there could be more than one secondary diagnosis.

). All patients had preoperative orthodontic treatment with standard fixed appliances. The most common surgery type, constituting over half (57%) of the surgeries, was LF1 + BSSO, followed by LF1 alone and then BSSO Genioplasty (GP) (

Table 2. Surgeries Distribution.

Surgical Procedure	Number	%
BSSO	4	3.8%
BSSO + GP	9	8.7%
LF1	14	13.6%
LF1 + GP	8	7.70%
LF1 + BSSO	55	53.4%
LF1 + BSSO + GP	5	4.8%
LF1 2p	4	3.8%
LF1 2p + Genioplasty	1	0.97%
LF1 2p + BSSO	1	0.97%
LF1 + HVRO	2	1.94%

Legend: BSSO—Bilateral Sagittal Split Osteotomy, LF1—Le-Fort 1 Osteotomy, LF1 2p—Two Piece Maxilla Le-Fort 1 Osteotomy, HVRO—High Vertical Ramus Osteotomy, GP—Genioplasty.

). The surgeries were performed by all the staff in the department—including faculty and residents—randomly assigned. During surgery, the maxilla was cut using reciprocating saw, while mandibular osteotomies were performed by Lindemann drill. All surgeries were performed under hypotensive anesthesia—using Nitrates (sodium nitroprusside, nitroglycerin) alone or combined with β-adrenoceptor antagonists (e.g., propranolol and esmolol) or calcium channel antagonists (e.g., nicardipine). In all cases, rigid internal fixation with titanium miniplates and screws were used for osteosynthesis (1.5 mm miniplates for the maxilla, 2 mm plates for the mandible).

Assessing correlations between primary diagnosis and surgery type revealed a significant (p = 0.003, Fisher’s Exact Test) connection between Skeletal Class II and Skeletal Class III diagnoses and the probability of two jaw (bimaxillary) surgery (Figure 1). The correlation is strongest for Skeletal Class III diagnosis where 70% of the patients underwent two jaw surgery (61.4% of LF1 + BSSO alone).

Maxillo-Mandibular Fixation (MMF-wires) was used to guide occlusion for the first 6 weeks post-operatively using blenderized diet.

The post-operative protocol includes hospitalization for a week, then routine check-ups twice a week for the first two weeks, then weekly for the next 6 weeks, then once a month until 6 months post-operatively and once in 3 months until 18 months post-operatively. The patients did not resume professional or physical activity up until 7–8 weeks after the surgery. Post-operative orthodontic treatment was also resumed at 8 weeks.

There were 2 incidents of a severe complication which was major bleeding after LF1 surgery, requiring 3 units of blood transfusion. Cases of moderate complications included (i) 3 incidents having permanent nerve (Inferior alveolar nerve—IAN) injury (2.9%), all of which occurred after LF1 + BSSO surgeries, after at least 12 months and (ii) 6 cases requiring a second surgery (5.7%, 2 after LF1 and LF1 + BSSO, 1 after LF1 + BSSO + GP and BSSO). In our study group the probability of moderate surgical complications was 9.6%.

The most prevalent minor complication was temporary nerve (IAN) injury (29.3%) and complications related to hardware (27.6%). There were 56 complications in total (54.4%) and as 9 patients presented more than 1 complication, only 45 patients (43.7%) presented complications—giving an odd of 0.77 per patient to suffer complications after surgery. The most prevalent complication was temporary nerve injury followed by hardware complications (loose hardware, broken hardware). The complete data including the probability of complication per patient and relative risk of complication per patient (compared to LF1 + BSSO as being the most prevalent surgery) are presented in

Table 3. Complications by type of surgery.

	BSSO	BSSO + genio	LF1	LF1 + genio	LF1 + BSSO	LF1 + BSSO + genio	LF1 2p	LF1 2p + genio	LF1 2p + BSSO	LF1 + HVRO	Total	Probability within Complications
Sev./Mod. Complications	1	0	2	0	5	1	0	0	1	0	10	17.9%
Bad Split	0	2	0	0	0	0	0	0	0	0	2	3.6%
Condylar Sag	0	1	0	0	1	0	0	0	0	0	2	3.6%
Temp. Nerve Injury	1	4	0	2	9	2	0	0	0	0	18	32.1%
Late infection	0	0	0	0	1	0	0	0	0	0	1	1.8%
Hardware Issues	0	0	1	0	13	0	1	1	0	0	16	28.6%
TMJ Disorders	0	0	0	0	0	0	0	0	0	0	0	0.0%
Dental Injuries	0	0	1	0	3	0	2	0	0	0	6	10.7%
Simple Hemorrhage	0	0	0	0	1	0	0	0	0	0	1	1.8%
Total Complications	2	7	4	2	33	3	3	1	1	0	56
Total Patients w/complications	2	5	2	2	26	3	3	1	1	0	45
Total Patients	4	9	14	8	55	5	4	1	1	2	103
Probability of complication per patient	50.0%	55.6%	14.3%	25.0%	47.3%	60.0%	75.0%	100.0%	100.0%	0.0%	43.7%

. Analysis was carried out on surgeries grouped as two-jaw and one-jaw surgery Analysis revealed that one-jaw surgery results in significantly less complications than two-jaw surgeries (p = 0.019, Fisher’s Exact Test) as is clearly demonstrated in the 1.34 OR (Odds Ratio) of a patient having complications after a two-jaw surgery (

Table 4. Complications by type of surgery stacked to bi-max and one-jaw surgery.

	Bi-Maxillary Surgery	One Jaw Surgery	Total	Rate Ratio of Complication per Type of Surgery
Sev./Mod. Complications	7	3	10	0.89
Bad Split	0	2	2	0
Condylar Sag	1	1	2	0.38
Temp. Nerve Injury	13	5	18	0.99
Late infection	1	0	1	N/A
Hardware Issues	14	2	16	2.67
TMJ Disorders	0	0	0	N/A
Dental Injuries	3	3	6	0.38
Simple Hemorrhage	1	0	1	N/A
Total Complications	40	16	56
Total Patients w/complications	33	12	45
Total Patients	72	31	103
Probability of complication per patient	45.8%	38.7%
Odds Ratio of complication per patient	1.34	1		p = 0.019
99% Confidence Interval	1.05, 1.69

). It was also found that the RR (Relative Risk) for complications in two-jaw surgery was 2.67, mainly due to higher complication rates related to hardware issues.

Analyzing for relationships between complications and initial diagnosis revealed that patients with CL III have a significantly lower (p = 0.022, Pearson Chi-Square Test) chance of post-operative complications (OR 1.69 for CL I and 2.75 for CL II) as shown in

Table 5. Complications by initial skeletal diagnosis.

	CL I	CL II	CL III	Total
Sev./Mod. Complications	2	3	5	10
Bad Split	0	2	0	2
Condylar Sag	1	1	0	2
Temp. Nerve Injury	3	4	11	18
Late infection	0	0	1	1
Hardware Issues	1	6	9	16
TMJ Disorders	0	0	0	0
Dental Injuries	1	0	5	6
Simple Hemorrhage	1	0	0	1
Total Complications	9	16	31	56
Total Patients w/complications	6	13	26	45
Total Patients	12	21	70	103
Probability of complication per patient	50.0%	61.9%	37.1%
Odds Ratio of complication per patient	1.69	2.75	1
99% Confidence Interval	1.38, 2.06	2.26, 3.35	p = 0.022

. As the surgical team involved both faculty and trainees, no difference among trainees or faculty could be distinguished.

4. Discussion

Orthognathic surgery, like any surgical procedure, is not exempt from potential complications either during surgery or postoperatively. As orthognathic procedures are elective surgery in young, healthy patients any complication is important and significant [13].

The severest complication found in this study was a major hemorrhage. Bleeding is recognized as a possibly fatal complication with a surgical prevalence of 0.2–2.2% [3,4,5,8,14,15]. Hemorrhage sources vary and include maxillary artery and vein, pterygoid artero-venous plexus, inferior alveolar or retromandibular artery or vein, palatine artery, or facial artery. Severe arterial bleeding may require blood transfusion [16,17,18]. Consistent with the literature, we had two hemorrhage episodes which occurred in two patients during the postoperative period (1.9%), one of them requiring surgical intervention to embolize with coils the internal maxillary artery the other was treated with local pressure on one of the branches of the facial artery and local application of tranexamic acid.

One of the relatively moderate complications identified in this study is permanent nerve injury. Generally, permanent damage due to the injury of the alveolar nerve occurs with a prevalence of 0.8–9.6% [8,14,19,20,21]. IAN lesions are the most common with a risk of section, traction, or compression during osteotomy, splitting, or fixation of the mandible. Another nerve that may be injured is the Infra-Orbital Nerve (ION) in maxillary osteotomies. In this study, the prevalence of permanent nerve injury was 2.9%, similar to other reports [21,22]. We did not perform any nerve repair as no obvious nerve transection was identified or noted intraoperatively. At our institution, it is not standard to look for the nerve, if it is seen then its status is documented.

Temporary injury occurred in 17.5% of cases, almost all in the IAN. This number is on the lower side of the range of previous reports [1,23].

The second type of moderate complication identified in this study was the necessity for a second surgery or re-operation. This is usually a psychologically difficult outcome for the patient. The main indications are malocclusion recurrence or nonunion of the osteotomies. In this study, 6 patients (5.8%) underwent a second surgery, similar to rates previously reported [6]. Re-operation was performed for non-union or relapse.

Accurate preoperative planning that takes in account relapse tendency and stable fixation are essential to reduce the risk of this complication. Hardware was removed in 19 cases; we did not include these in “reoperation”.

Undesired fractures (bad split) may occur due to an irregular osteotomy line or an unfavorable fracture and their incidence varies between 0.9–23% [1,13,23,24,25,26]. There are reports that wisdom teeth in line of osteotomy increase risk of bad fracture [27,28,29]. These fractures most frequently occur in the lingual or the buccal plate of the mandible or, rarely, coronoid process or mandibular condyle. Such fractures may be a risk factor for infections, bone sequestration of the fragments, delayed bone healing and nonunion. In this study, the prevalence of a bad split of the mandible is 3.6%, a relative less than other studies [18,21,29]. A careful osteotomy execution in the case of small mandibular height or thickness may prevent this outcome.

In this study we had only 1 case of late infection (1%), one of the lowest compared to previous reports [2,22,23]. This underscores the success of our departments protocol for peri-operative (2 h pre-op until up to 24 h post-op) antibiotic treatment (amoxicillin + clavulanic acid IV, clarithromycin IV for penicillin allergic patients).

Dental injuries are one of the most common complications of orthognathic surgery. Such injuries result from tooth sectioning by the drill, periodontal disease, osteolytic lesions, root absorption and angular defects. In this study, the prevalence of dental injuries is 5.8%. Using of cutting guides is becoming increasingly popular and may decrease the incidents of tooth and nerve injury [30], as well as using correct drills and blades may help to prevent dental lesions during osteotomies [5].

Complications related to surgical fixation materials or “hardware issues” may cause surgical wound dehiscence, loss of screws, loss of fixation, plate exposure, plate rupture, bone structure mobility, bone necrosis, infection related to fixation materials, and infection related to the loss of fixation [31]. In many cases, the complication requires plate and screw removal. We found hardware issues to be one of the most prevalent complications (15.5%) and indeed most of those cases (81.3%) required plate and screw removal. This is consistent with documented in the literature [2,6,21,32]. We did not include these as the complication rate because not all plate removals were due to infection as some of them were removed only due to discomfort reported by the patients.

This study demonstrates a relationship between diagnosis and surgery type, operation type and complication probability and diagnosis and complication probability. The large number of surgery types prevented statistical analysis of the complication by exact surgery type. LF1 + BSSO surgery types are more significantly associated with CL II and especially CL III surgery. The explanation for this may be that CL I defects are often managed with one-jaw surgery because there is no need for significant Antero-Posterior movement. One the other hand, LF1 surgery was not conducted at all in CL II. This makes sense since CL II almost always results from a case of mandibular deficiency with or without horizontal maxillary excess, problems which cannot be corrected with LF1 surgery alone.

As in the literature, our study demonstrates a significantly higher risk of complications after a two-jaw surgery. This may be due to either longer surgery times (leading to a higher rate of complications) and/or the greater number of structures being operated on (an intact jaw cannot cause complications) [6,21].

Consistent with other studies, it was revealed that patients with CL III have a significantly lower risk of complications compared to CL I and especially CL II patients. We hypothesize that the anterior movement of the mandible in CL II surgeries is a specific risk factor for condylar problems, as was presented in previous reports [33,34]. Another risk factor may be the size of the mandible, which is frequently retrognathic and hypoplastic in these patients, the opposite of that for CL III patients. This phenomenon leads to increase in the operation time, which in turn, may lead to increase in complications rate [35].

There are few limitations of this study. It is a retrospective study derived from the experience and surgery types that were performed in the department. As such, it has limited data about surgeries which are not often performed in this setting. The limited number of cases operated according to one particular method often precludes any sound conclusions in regard to complications by type of surgery correlations.

5. Conclusions

The data from this retrospective study show that severe complications are relatively rare in orthognathic surgery; this finding is consistent with other studies. Statistical analysis revealed interesting links between diagnoses, surgeries and complication probabilities. We found firstly that one-jaw surgery is safer that two-jaw surgery; secondly, we found that the skeletal condition with the highest risk for complications in Skeletal Class II.