The aim of this study was to conduct a systematic review to evaluate the accuracy and benefits of computer-aided planning in orthognathic surgery. The search was performed in PubMed, EMBASE, Cochrane Library, LILACS, and SciELO. The articles identified were assessed independently and in a blinded manner by two authors using selection criteria and eligibility criteria. The database search yielded 375 studies. Following the application of search and eligibility criteria, a final nine studies were included in the systematic review. The level of agreement between the authors in the study selection process was substantial ( κ = 0.767) and study eligibility was considered excellent ( κ = 0.863). The accuracy of translation was <1.2 mm in the maxilla (vertical) and <1.1 mm in the mandible (sagittal), and for rotation was <1.5° in the maxilla (pitch) and <1.8° in the mandible (pitch). Two studies showed a medium potential risk of bias and six studies showed a high potential risk of bias. Computer-aided planning in orthognathic surgery was considered accurate for the studies included in this systematic review. However, the low quality of these studies means that randomized clinical trials are needed to compare computer-aided planning to conventional planning in orthognathic surgery.
Orthognathic surgery is a surgical procedure to treat facial deformities and involves osteotomy of the jaw bones to correct dentofacial disharmony and misalignment. Surgical planning to obtain the desired stability and harmony is complex and accuracy is essential.
Planning in orthognathic surgery has evolved over the past decades. This started with ‘classic’ surgical planning using cephalometric analysis of lateral radiographs, facial analysis, and plaster casts of the patient’s dental arches mounted on an articulator and surgical splints made from acrylic resin. Planning then evolved into the use of two-dimensional (2D) computer programs for cephalometric analysis of lateral radiographs and the more modern technique of computer-aided planning.
Computer-aided orthognathic surgery integrates planning and the surgical intervention using software, with three-dimensional (3D) cephalometric analysis of bone and soft tissue, performance of the surgical movements to achieve the ideal dento-skeletal harmony, and transfer of the virtual planning to the surgical setting using a surgical splint. However, although surgical planning is computer-aided, the facial analysis of patients must be performed clinically.
Some authors have reported that computer-aided planning increases the effectiveness of orthognathic surgery, with more accurate osteotomy than with the classic planning of surgery.
In order to ensure accurate results, the computer programs used for orthognathic surgery planning require experienced operators and quality computed tomography (CT) scans. It is important to note that the same anatomical points are used for cephalometric analysis in both classic and computer-aided planning. As such, the chances of error are the same for the two types of analysis, since these points are assigned by the surgeon and not the computer. However, computer-aided planning results in fewer preoperative steps for the surgeon, which lowers the number of systematic errors in the placement of osteotomies.
Randomized controlled clinical trials offer the best scientific evidence to assess the effectiveness and accuracy of computer-assisted planning for orthognathic surgery. However, if studies of this nature are not available in the literature, other types of research can be used to evaluate an intervention or generate hypotheses on it. A systematic review of intervention studies is therefore an important tool in helping to understand and quantify the accuracy of computer-aided planning in orthognathic surgery, as well as evaluating the need for new research on the subject.
The aim of this systematic review was to assess the accuracy of computer-aided planning in orthognathic surgery and to determine whether it provides greater benefits to the patient and surgical procedure than classic planning, as well as to ascertain the quality of the available literature.
Materials and methods
A systematic search was conducted of electronic and printed media (annals from conferences) on computer-aided planning for orthognathic surgery. The databases used were PubMed, EMBASE, the Cochrane Library, LILACS, and SciELO. There were no restrictions in the search strategy regarding language or year of publication. Key words and Boolean operators (‘OR’ and ‘AND’) were used to join terms (thesaurus or words) related to orthognathic surgery and computer-aided planning.
The search of PubMed was conducted using the following medical subject heading (MeSH) terms: [(‘Orthognathic Surgery’ OR ‘Orthognathic Surgery’ OR ‘Orthognathic Surgeries’ OR ‘Surgeries, Orthognathic’ OR ‘Surgery, Orthognathic’ OR ‘Maxillofacial Orthognathic Surgery’ OR ‘Maxillofacial Orthognathic Surgeries’ OR ‘Orthognathic Surgeries, Maxillofacial’ OR ‘Orthognathic Surgery, Maxillofacial’ OR ‘Surgeries, Maxillofacial Orthognathic’ OR ‘Surgery, Maxillofacial Orthognathic’ OR ‘Orthognathic Surgical Procedures’ OR ‘Orthognathic Surgical Procedure’ OR ‘Procedure, Orthognathic Surgical’ OR ‘Procedures, Orthognathic Surgical’ OR ‘Surgical Procedure, Orthognathic’ OR ‘Surgical Procedures, Orthognathic’) AND (‘Surgery, Computer-Assisted’ OR ‘Surgery, Computer-Assisted’ OR ‘Computer-Assisted Surgeries’ OR ‘Surgeries, Computer-Assisted’ OR ‘Surgery, Computer Assisted’ OR ‘Computer-Assisted Surgery’ OR ‘Computer Assisted Surgery’ OR ‘Computer-Aided Surgery’ OR ‘Computer Aided Surgery’ OR ‘Computer-Aided Surgeries’ OR ‘Surgeries, Computer-Aided’ OR ‘Surgery, Computer-Aided’ OR ‘Surgery, Image-Guided’ OR ‘Image-Guided Surgeries’ OR ‘Surgeries, Image-Guided’ OR ‘Surgery, Image Guided’ OR ‘Image-Guided Surgery’ OR ‘Image Guided Surgery’ OR ‘Computer-Aided Design’ OR ‘Computer Aided Design’ OR ‘Computer-Aided Designs’ OR ‘Design, Computer-Aided’ OR ‘Designs, Computer-Aided’ OR ‘Computer-Assisted Design’ OR ‘Computer Assisted Design’ OR ‘Computer-Assisted Designs’ OR ‘Design, Computer-Assisted’ OR ‘Designs, Computer-Assisted’ OR ‘Computer-Aided Manufacturing’ OR ‘Computer Aided Manufacturing’ OR ‘Manufacturing, Computer-Aided’ OR ‘Computer-Assisted Manufacturing’ OR ‘Computer Assisted Manufacturing’ OR ‘Manufacturing, Computer-Assisted’ OR ‘CAD-CAM’)].
The same search strategy was applied to the Cochrane Library, since this also uses MeSH terms.
For the search of EMBASE, the Emtree terms ‘orthognathic surgery’, ‘computer assisted surgery’ and ‘computer aided design’ were used to carry out a specific search: ‘orthognathic surgery’/syn AND (‘computer assisted surgery’/syn OR ‘computer aided design’/syn).
Health sciences descriptors (Descritores em Ciências da Saúde—DeCS) in the English language were used to search the LILACS and SciELO databases; the following searches were performed: LILACS (tw:(‘orthognathic surgery’)) AND (tw:(‘computer-aided design’)) and SciELO ‘orthognathic surgery’ AND ‘computer-aided design’.
A search for unpublished studies and research published in journals not indexed in major databases was conducted: (1) annals of the last three editions of the International Conference on Oral and Maxillofacial Surgery (ICOMS), promoted by the International Association of Oral and Maxillofacial Surgery, and the last three annual meetings of the American Association of Oral and Maxillofacial Surgery (AAOMS Annual Meeting). (2) Google Scholar, using MeSH terms (‘orthognathic surgery’, ‘computer aided design’, ‘computer assisted surgery’) and Boolean operators (‘AND’ and ‘OR’): ‘orthognathic surgery’ AND (‘computer aided design’ OR ‘computer assisted surgery’).
The authors of the abstracts found in the annals were contacted by e-mail for information on the results or to determine whether the study had been published.
The references of papers identified were analyzed for further studies not located in the above-mentioned searches.
The systematic search was conducted by one of the authors (O.L.H.J.), and articles were selected independently by two authors (O.L.H.J., O.E.B.) based on the title and abstract. Studies with the following characteristics were chosen for full-text reading: (1) intervention studies; (2) creation of a virtual surgical splint; (3) investigations assessing the accuracy of orthognathic surgery with a virtual surgical splint. Papers that did not meet these criteria were excluded from the analysis. When the authors disagreed on the selection of a paper, the study was read in full.
Studies for which the titles and abstracts were evaluated and that were accepted in the first selection process were submitted to an eligibility assessment.
The level of agreement between the authors was tested using Cohen’s kappa coefficient ( κ ).
Eligibility of the studies
The eligibility of the studies was checked by two authors (O.L.H.J., O.E.B.) who were blinded to the title, abstract, authorship, and origin of the papers. The following eligibility criteria were used: (1) the main theme of the paper had to centre on computer-aided planning for orthognathic surgery; (2) the trial had to be original and an intervention study; (3) the surgical procedure had to be computer-aided with virtual planning and with the design of a virtual surgical splint; (4) accuracy measures had to be presented for the surgical procedure. In the event of disagreement between the two authors, the study in question was discussed with the third, more experienced author (R.B.O.).
Investigations that did not meet the eligibility criteria were excluded from the analysis and the reason for their exclusion was reported.
When questions arose regarding the methodology or results of a paper, the author was contacted by e-mail to obtain the necessary answers.
The level of agreement between the authors was tested using Cohen’s kappa coefficient ( κ ).
Demographic and methodological data, as well as the accuracy results, were extracted from the studies that met the eligibility requirements by the two blinded authors (O.L.H.J., O.E.B). In the event of disagreement between the two authors, the study was discussed with the third author (R.B.O.). When doubts persisted, the author of the study in question was contacted by
Analysis of the methodological quality of the studies included
The quality of the papers was assessed using an adaptation of the bias analysis proposed by Clementini et al. The criteria used by these authors are related to the randomization of the sample, validation of measurements, statistical analysis, the definition of inclusion and exclusion criteria, and whether sample loss was reported in the postoperative period. In addition to these items, analysis of comparison data between interventions and blinding of the rater were included as criteria.
With respect to the risk of bias for each study analyzed, papers containing all the above-mentioned items were considered low risk, those for which one or two items were missing were deemed medium risk, and investigations that did not include three or more items were considered high risk.
The strategies used for the main search and grey literature search were applied for the last time on 13 January 2014. The manual search was conducted after papers had been included, with no need for updating. A flowchart showing the studies included at each stage of the systematic review is given in Fig. 1 .
Three hundred and fifty-seven articles were found in PubMed, 93 in EMBASE, and 28 in LILACS; no studies were found in the Cochrane Library or SciELO. Duplicate papers were removed, leaving a total of 375 possible studies. The oldest of these papers was published in 1984, with studies on computer-aided planning in orthognathic surgery increasing in subsequent years to a total of 157 publications in the last 3 years (2011–2013) ( Fig. 2 ).
With regard to the grey literature search, four studies were selected from Google Scholar, one from the ICOMS annals, and four from the annals of the AAOMS Annual Meeting. The authors of the studies found in these annals were contacted by e-mail regarding the results and to determine whether the research had been published in the form of a scientific paper. Those studies for which an author reply was not received were excluded. As a result, only the four studies found in the Google Scholar search were included in the sample submitted for eligibility assessment.
One study was found and included in the final sample for systematic review.
The titles and abstracts of the 375 selected studies were read and 34 of these were chosen for full-text reading. Whenever the two authors differed as to the selection of a paper, disagreement was resolved by choosing the broadest possible study. The level of agreement between the two authors in selecting studies to be read in full was measured at κ = 0.767.
Eligibility of the studies
As part of the eligibility assessment, 38 studies were read in full, 34 from the main search and four from the grey literature search. At the end of this analysis, eight papers were included in the sample for systematic review. The other 30 studies were excluded for the following reasons: computer-aided planning for orthognathic surgery was not the main theme of the paper ( n = 7 studies ), the paper was not an intervention study ( n = 14 studies ) or was not original ( n = 1 ), the surgical procedure did not involve a computer-assisted virtual surgical splint ( n = 3 ), and accuracy measures for the surgical procedure were not provided ( n = 5 ).
The level of agreement between the two authors for the eligibility assessment was measured at κ = 0.863.
The sample used in the systematic review consisted of nine studies, of which seven were found in the main search, one in the grey literature, and one in the manual search.
The research groups on computer-aided planning in orthognathic surgery were from different countries; however, two papers by a group from the USA and two by a research group in Spain were included in the systematic review. All of the studies were prospective and only one paper was a randomized controlled clinical trial. Six were published in 2013, and the largest sample (65 patients) was part of a multi-centre trial. The total sample from the studies included, in which CAD/CAM was used in orthognathic surgery, comprised 137 individuals; their mean age was 20–30 years and the gender (male/female) and facial deformity (class II/class III) proportions were approximately 50% ( Table 1 ).
|Author, year, and country of origin||Type of study||Sample||Age, years, mean ± SD (variation)||Gender||Type of facial deformity|
|Xia et al., 2007
|Prospective case series/pilot study||n = 5||NA||2 M, 3 F a||3 class II, 2 class III, all asymmetrical a|
|Centenero and Hernández-Alfaro, 2012
|Prospective case series||n = 16||NA||NA||9 class II, 7 class III|
|Sun et al., 2013
|Prospective case series||n = 15||NA||NA||NA|
|Li et al., 2013
|Prospective case series||n = 6||(19–30)||4 M, 2 F||NA|
|Hsu et al., 2013
|Prospective case series/multi-centre||n = 65
New York: 13
Houston: 26.7 (15–51)
Portland: 26.7 (16–46)
New York: 21.7 (16–51)
|31 M, 34 F
Houston: 23 M, 18 F
Portland: 3 M, 8 F
New York: 5 M, 8 F
|Hernández-Alfaro and Guijarro-Martínez, 2013
|Prospective case series/proof of concept study||n = 6||23.7 (19–37)||3 M, 3 F||5 class II, 1 class III|
|Shehab et al., 2013
|Prospective case series/pilot study||n = 6||23.5 (18–30)||6 F||6 class II with vertical maxillary excess|
|Zinser et al., 2013
|Non-randomized clinical trial||n = 28
Virtual splint: 8
Surgical navigation: 10
Classic splint: 10
|20.8 ± 4.9 (18–35)
Virtual splint: 21.6 ± 5.45 (19–35)
Surgical navigation: 20.5 ± 4.1 (18–32)
Classic splint: 20.6 ± 2.6 (18–26)
|15 M, 13 F
Virtual splint: 4 M, 4 F
5 M, 5 F
Classic splint: 6 M, 4 F
|5 class II, 23 class III
Virtual splint: 8 class III
Surgical navigation: 1 class II, 9 class III
Classic splint: 4 class II, 6 class III
|De Riu et al., 2014
|Randomized controlled clinical trial||n = 20
Virtual splint: 10
Classic splint: 10
|Virtual splint: (21–54)
Classic splint: (24–47)
|10 M, 10 F
Virtual splint: 3 M, 7 F
Classic splint: 7 M, 3 F
|Class II/class III: NA