Types of studies

Randomised controlled trials (RCTs) conducted in people undergoing OS. Quasi‐randomised and non‐randomised trials, observational studies, narrative reviews, commentaries and letters to editors were excluded.

Types of participants

People of any age undergoing OS in any setting.

Types of interventions

• Intervention: any type of antibiotic (penicillin and its derivatives, cephalosporins, etc.), with any regimen or mode of administration (short‐term or long term; oral, endovenous or intramuscular; preoperative or perioperative regimen).

• Comparison: placebo, or another antibiotic, or another regimen of antibiotic.

Types of outcome measures

Studies had to report any of the following outcomes to be included:

Electronic searches

In June 2014 we searched the following electronic databases to identify reports of relevant randomised clinical trials.

• The Cochrane Wounds Group Specialised Register (up to 05 June 2014);

• The Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2014, Issue 5);

• Ovid MEDLINE (1946 to May Week 4 2014);

• Ovid MEDLINE (in‐Process & Other Non‐Indexed Citations, June 04 2014);

• Ovid EMBASE (1980 to 2014 June 04);

• EBSCO CINAHL (1982 to 05 June 2014).

We used the following search strategy in Cochrane Central Register of Controlled Trials (CENTRAL):

#1 MeSH descriptor: [Orthognathic Surgery] explode all trees 8
#2 MeSH descriptor: [Orthognathic Surgical Procedures] explode all trees 61
#3 MeSH descriptor: [Osteotomy, Le Fort] explode all trees 63
#4 (orthognathic near/5 surg*):ti,ab,kw 197
#5 ((maxillary next osteotom*) or "Le Fort" or (mandibular near/5 osteotom*) or (vertical next ramus next osteotom*) or genioplast*):ti,ab,kw 210
#6 #1 or #2 or #3 or #4 or #5 343
#7 MeSH descriptor: [Antibiotic Prophylaxis] explode all trees 1226
#8 MeSH descriptor: [Anti‐Bacterial Agents] explode all trees 9253
#9 (antibiotic* or cephalosporin* or cefazolin or cefuroxime or amoxicillin* or amoxycillin* or clindamicin or clindamycin or penicillin* or levofloxacin):ti,ab,kw 20956
#10 #7 or #8 or #9 24784
#11 #6 and #10 22

We adapted this strategy to search Ovid MEDLINE, Ovid EMBASE and EBSCO CINAHL. The search strategies for Ovid MEDLINE, Ovid EMBASE and EBSCO CINAHL can be found in Appendix 1, Appendix 2 and Appendix 3 respectively. We combined the Ovid MEDLINE search with the Cochrane Highly Sensitive Search Strategy for identifying randomised trials in MEDLINE: sensitivity‐ and precision‐maximising version (2008 revision) (Lefebvre 2011). We combined the EMBASE search with the Ovid EMBASE filter developed by the UK Cochrane Centre (Lefebvre 2011). We combined the CINAHL searches with the trial filters developed by the Scottish Intercollegiate Guidelines Network (SIGN 2011). We did not restrict the search and study selection with respect to language, date of publication or study setting. If an article published in a language other than English was identified, relevant data were extracted by a translator.

We searched the following ongoing trials databases up to February 2013.

• Current Controlled Trials (http://www.controlled‐trials.com/).

• ClinicalTrials.gov (http://www.clinicaltrials.gov/).

• World Health Organization (WHO) International Clinical Trials Registry Platform (ICTRP) (http://www.who.int/ictrp/en/).

• EU Clinical Trials Register (https://www.clinicaltrialsregister.eu/)

Searching other resources

We examined reference lists of relevant articles that were identified by the electronic searches for other pertinent articles to include in the review. We also searched in Google Scholar to detect unpublished or grey literature.

To account for any delay in indexing in the electronic databases mentioned above, we also searched the last six months of the following journals, up to February 2013.

•  Journal of Oral and Maxillofacial Surgery.
•  International Journal of Oral and Maxillofacial Surgery.
•  British Journal of Oral and Maxillofacial Surgery.
•  Journal of Craniofacial Surgery.
•  Head & Neck: Journal for the Sciences & Specialties of the Head and Neck.

Finally, we handsearched the last five years of the online abstract indexes of conference proceedings of the annual meetings of the American Association of Oral and Maxillofacial Surgeons and the International Association for Dental Research up to February 2013.

Selection of studies

In a first screening, the title and the abstract of all potentially relevant articles were listed and were evaluated using a pre established selection criteria form. This process was done independently by two review authors who followed instructions especially designed for this stage, which were widely inclusive. All articles selected for full‐text screening by either review author were retrieved. The full text of all articles that potentially met the eligibility criteria were assessed independently by two review authors. Disagreements were resolved by discussion, and, if no consensus was reached, a third review author acted as arbiter. We include a study flow diagram, as recommended by the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta‐Analyses) statement (Liberati 2009), to illustrate the results of all searching activity and the process of screening and selecting studies for inclusion in the review (Figure 1). Articles that were published in a different language were evaluated by reviewers fluent in that language, with previous experience in systematic reviews.

Figure 1

---

---

Flow diagram.

Data extraction and management

Two review authors extracted the data from all selected studies independently, using a standardised form of Microsoft Excel Office for Mac 2011 created for this purpose. We extracted data regarding the characteristics of the studies relevant for this review, such as the characteristics of the population (sex, age, and selection criteria for enrolment in the trial), surgery (type and number of osteotomies, if reported), intervention and comparison (antibiotic type and regimen), and outcomes (definitions and results of the trial). Discrepancies between the data were reviewed by the two review authors, and, if needed, a third review author acted as arbiter. If data for the outcomes of interest were missing, we contacted the trial authors to obtain the information. A standard form was to be designed to request trial authors to provide specific data needed for the review, and the fields required were highlighted in each case. We planned to liaise with the Wounds Group should any translations be required.

Assessment of risk of bias in included studies

The risk of bias of the included studies was evaluated using the Cochrane risk of bias tool (Higgins 2011). All domains of this tool were used (sequence generation, allocation concealment, blinding of participants, personnel and outcome, incomplete outcome data, selective outcome reporting and other sources of bias), and the instructions published in the Cochrane Handbook for Systematic Reviews of Interventions were followed in assessing each domain and in performing the evaluation of the overall risk of bias. This evaluation was performed by two review authors independently, based on the full text of the trials. Disagreements between the review authors were discussed, and consensus had to be reached for a study to be classified as having low, high or unclear risk of bias. As allocation concealment was judged to be the most critical risk of bias domain for this review, a trial would have been classified as having overall HIGH risk of bias if either the allocation concealment domain or two or more other domains were judged to have been at high risk of bias. The same criteria were applied to classify a trial as having unclear risk of bias. Otherwise, the trial was to be classified as having low risk of bias. As it has been shown that blinding to study author and/or affiliation of the study is not associated with the overall results of evaluations (Moher 1999), the review authors were not blinded to these characteristics of the trials.

Measures of treatment effect

The measures of treatment effect that were to be used for evaluating the outcome in each trial include the following.

• SSI: Rate of infection at the surgical site was analysed as dichotomous, and risk ratios (RRs) and 95% confidence intervals (CIs) were calculated. We expected studies to report how the presence of SSI was judged and two review authors with clinical expertise determined whether this definition was consistent with the CDC definition.

• Systemic infection: We planned to analyse data as dichotomous, and RRs and 95% CIs would have been calculated; however, no data were available that would permit the review authors to do this.

• Length of hospital stay (days): We planned to analyse data as continuous, using mean difference (MD) with 95% CI. If the data were presented as median and were likely skewed, we would not have pooled the data. Data were insufficient for evaluation of this outcome.

• Adverse events: When trials reported adverse events in sufficient detail (e.g. the number of participants who experienced at least one adverse event), we analysed these data dichotomously.

Unit of analysis issues

We used the participant as the unit of analysis. No unit of analysis issues were anticipated, nor were any found.

Dealing with missing data

Each trial was analysed with regards to the presence and proportion of missing data, and it was judged that the amount of missing data was not substantial enough to change the direction of the results. Therefore, it was not necessary to impute missing data, and only complete case data were used.

Assessment of heterogeneity

The Chi² test was used to determine the presence of statistical heterogeneity, using a level of significance of 0.1. Quantification of inconsistency across studies was done using the I² statistic, and its interpretation was based on recommendations of The Cochrane Collaboration (Deeks 2011), that is, an I² between 0% and 40% might be considered as unimportant heterogeneity amongst the trials; 30% to 60% might represent moderate heterogeneity; 50% to 90% might represent substantial heterogeneity; and 75% to 100% might represent considerable heterogeneity. Clinical heterogeneity was assessed qualitatively by considering participants, setting and intervention characteristics with the help of experts. Methodological heterogeneity was evaluated using the domains of the risk of bias tool (Higgins 2011). Exploration of heterogeneity was based on subgroup analyses (detailed below).

Assessment of reporting biases

Efforts were made to detect reporting biases, if possible, in accordance with the recommendations of the Cochrane Handbook for Systematic Reviews of Interventions (Sterne 2011). Outcome reporting biases were explored by looking for published protocols of the trials included in the systematic review (where available). If a sufficient number of studies had been identified, publication bias was to have been explored using funnel plots for all outcomes and tested using the Egger test (Egger 1997).

Data synthesis

We present a narrative overview of the included trials. When appropriate, we present meta‐analyses of outcome data using RevMan 5.3. The decision to pool data in a meta‐analysis depended on the availability of outcome data and assessment of between‐trial heterogeneity. Because of anticipated and observed clinical heterogeneity between primary studies, random‐effects meta‐analysis was planned and performed. Participants were analysed in the arms to which they were randomly assigned, in accordance with the intention‐to‐treat principle.

For analytical purposes, we classified the antibiotic regimens according to time and duration of antibiotic prophylaxis as follows.

• Preoperative dose only: antibiotic was administered only once before the OS.

• Short‐term prophylaxis: antibiotic was administered before and during surgery, and during the same day of surgery.

• Long‐term prophylaxis: antibiotic was administered before, during and longer than one day after surgery.

In studies that included more than one arm of any of the categories, data from those arms were combined. The main findings are presented in a 'Summary of findings' table (Rosenbaum 2010).

Subgroup analysis and investigation of heterogeneity

We anticipated that two factors could cause heterogeneity across results of the trials, and these would be explored through subgroup analyses. The number of osteotomies was the predictor of greatest interest. Based on previous findings (Chow 2007), the a priori hypothesis for this factor is that trials in which participants had a lower number of osteotomies would show greater treatment effect in favour of antibiotic prophylaxis than trials in which participants underwent a greater number of osteotomies. Other sources of heterogeneity that would have been investigated, if possible, include mode of administration of the antibiotic and duration of surgery.

However, as statistical heterogeneity between studies was low and we detected no important clinical source of heterogeneity; we did not perform any subgroup analysis.

Sensitivity analysis

We had planned to perform a sensitivity analysis by excluding trials with high risk of bias. As the allocation concealment was judged to be the most critical risk of bias domain for this review, a trial would have been classified as having high risk of bias if either the allocation concealment domain or two or more other domains were judged to have high risk of bias. The same criteria were applied to classify a trial as having unclear risk of bias. Otherwise, the trial was to be classified as having low risk of bias. None of the trials was classified as having high risk of bias, thus no sensitivity analysis was performed.

We also planned to perform sensitivity analyses if data imputation was needed, if a meta‐analysis showed a trial with an effect that was different from the others in an important way (outlier) and if this trial had clinical features that made it different from the others. None of these scenarios were faced, thus no sensitivity analyses were performed.