Materials and methods

A systematic review was conducted, based on the PRISMA guidelines ( http://www.prisma-statement.org ). The PubMed, Cochrane Library, and Scopus databases were searched for publications in the English language, without any restriction on the type of study (all searched up to 3 June 2015). The search strategy was defined by the following terms: ‘orthognathic’ and ‘double-jaw’, ‘orthognathic’ and ‘two-jaw’, ‘orthognathic’ and ‘mandible-first’, ‘orthognathic’ and ‘bimaxillary’.

Inclusion criteria encompassed any mention of the mandible-first sequence within the abstract of any article generated by the search, without any restriction on the type of study. The exclusion criterion was the absence of any reference to the mandible-first sequence within the abstract.

The systematic search was conducted by one author (A.M.B.), and two authors (A.M.B., P.S.C.) independently performed the screening of titles and abstracts. Once an article abstract was selected according to the eligibility criteria (inclusion and exclusion), the full-text article was read, including the references. Any reference that could contribute to the purpose of the systematic review was retrieved. The two authors then presented their list of eligible studies and any difference was discussed until consensus was reached.

Results

With the application of the search criteria, the initial search identified a total of 887 abstracts from the different databases ( Fig. 1 ). After reading the full-text versions of the corresponding articles and relevant references, six articles were selected ( Table 1 ). The contents of these six articles addressed the topic of mandible-first bimaxillary orthognathic surgery. The 884 articles excluded following the review of abstracts were either duplicates or did not mention the mandible-first sequence.

Flow diagram (PRISMA format) of the screening and selection process.

The first mention of the altered sequence for bimaxillary orthognathic surgery was provided by Lindorf and Steinhäuser in 1978. They stated that bimaxillary orthognathic surgery should start with the mandible, since a stable reference (the maxilla) is needed to accurately reproduce the surgical movements predicted during model surgery to correspond to the actual surgery. The authors performed model surgery starting with the maxilla, then assembling the mandible in the desired final position for the construction of the final inter-occlusal splint; later, the maxilla was returned to its original position, and an intermediate splint fabricated.

In the 1980s, Buckley et al. highlighted the disadvantages of starting bimaxillary orthognathic surgery with the maxilla, due to the instability of an already operated maxilla that could be displaced during mandibular manipulation and fixation. With two maxillary cast models and one mandibular cast model mounted on a semi-adjustable articulator, the authors also proposed model surgery to begin with the maxilla, since surgery itself would start with the mandible having the uncut maxilla as a stable reference for the operated mandible. The need for rigid fixation instead of wire osteosynthesis for this technique was emphasized.

In the following decade, Cottrell and Wolford published their experience of commencing bimaxillary orthognathic surgery with the mandible first. The authors proposed model surgery to start with the mandible based upon the prediction of its position on the final occlusion, thus eliminating errors related to achieving centric relation for model surgery. Moreover, they suggested that even in segmental maxillary surgery, the use of a final splint would be an option; however, their preference was for direct dental inter-digitation to achieve the final planned occlusion, in order to eliminate interferences from the splint. It was suggested that these modifications would reduce the time and materials required for this planning step and also result in improved accuracy. Surgery was started with the mandible as long as rigid fixation was used, and attention was given to properly seating the condyles in their fossa during the stabilization of the proximal mandibular segments.

The lack of reliability of the mandible as a stable reference was the subject of another publication related to orthognathic surgery. In 2006, Posnick et al. reported that some patients would present difficulties in registration of centric relation, such as those with congenital incomplete formation or absence of a mandibular condyle (as in hemifacial microsomia) and those with total or partial loss after tumour resection or condylar fracture. They stated that in cases without a reliable centric relation, the model surgery sequence should start with the maxilla, and the traditional bimaxillary orthognathic surgery (maxilla-first) should be modified to the mandible-first sequence.

The controversy over the choice of the initial jaw osteotomy for bimaxillary orthognathic surgery was recently debated. Turvey stated that it should be a surgeon’s decision whether the maxilla or the mandible will be the first jaw cut in bimaxillary surgery. In addition, Perez and Ellis detailed specific instances in which it would be preferable to start the surgery with the mandible, exemplified by modification of the mandibular position used for model surgery after general anesthesia. These specific examples warranting consideration of the mandible-first sequence include the following: cases of bimaxillary orthognathic surgery in which down-grafting of the posterior maxilla is planned; when centric relation registration for model surgery is uncertain; when intermaxillary fixation is impaired by a thick intermediate splint if surgery starts with the maxilla; when fixation of the maxilla is not rigid; when concomitant temporomandibular joint surgery is planned with the orthognathic procedure. Model surgery could start with the mandible when only one set of dental models is available, or start with the maxilla when two sets of models are mounted on an articulator.

The most recent article selected was the first study to compare whether the choice of jaw operated on first would interfere with the outcomes of orthognathic surgery. In 2014, Ritto et al., using a retrospective sample, demonstrated that the maxilla could be accurately placed in the predicted position, regardless of the choice of the initial jaw operated on for bimaxillary surgery.