Literature search

A comprehensive systematic review of the literature was performed in the bibliographic databases PubMed (National Library of Medicine, NCBI), EMBASE, and the Cochrane Central Register of Controlled Trials from inception to 15 October 2013; the review was performed in accordance with the PRISMA statement. Search terms included controlled terms from medical subject headings (MeSH) in PubMed and Emtree in EMBASE, as well as free text terms. We used free text terms only in the Cochrane register. Search terms expressing distraction osteogenesis were used in ‘AND’ combination with search terms comprising ‘mandible’ and terms for complication, failure, and morbidity ( Table 1 ). The references of the identified articles were searched for additional relevant publications.

Study selection and inclusion criteria

Two reviewers independently screened all potentially relevant titles and abstracts for pre-specified eligibility criteria. If necessary, the full text article was checked for the eligibility criteria. Differences in judgement were resolved through a consensus procedure. The full text of articles was then obtained for further review.

The articles were included if they met the following eligibility criteria: (1) clinical article, (2) mandibular distraction osteogenesis (MDO), (3) congenital deformity, and (4) a report on complications. Studies were excluded if data on complications were insufficient, no translation was available, or the publication was a non-clinical article ( Table 2 ).

Articles that were found clinically relevant to the study subject were included in the systematic review. According to their emphasis, these relevant papers were included if they described MDO in the treatment of congenital mandibular deformities. The articles were screened for the following data: type of deformity, number of patients, type of DO, distraction device, vector, and type and number of complications. The latter were classified according to the proposed classification index shown in Fig. 1 . This classification emphasizes on the severity and clinical consequences of a complication by dividing events according to spontaneous resolving or permanent complications, hospitalization or general anesthesia required for correction of the complication.

Distraction osteogenesis complication index, based on this systematic review.

The initial literature search identified a total of 973 references: 521 in PubMed, 437 in EMBASE, and 15 in the Cochrane register. After removing duplicate references ( n = 329) that were selected from more than one database, 644 papers remained. Titles and abstracts were screened for eligibility by the two reviewers; 335 articles were excluded from the review based on the abstract. The full text was obtained for 309 papers and analyzed thoroughly. Subsequent categorization produced the following clusters ( Table 2 ): (1) 124 articles concerned non-congenital deformities; (2) 57 had insufficient or no information on complications and/or methods; (3) 24 papers were non-clinical (eight scientific, 16 synopsis); (4) four papers were not relevant (three non-DO, one maxilla); (5) 11 papers had no available translation (one Russian, eight Chinese, one Japanese, one Polish); (6) five articles had an edited publication type (three discussion, two letters to the editor/authors); (7) three papers were not available in the international libraries. These seven groups were excluded from further evaluation. In the case of a paper that reported complications in a mixed population (congenital, developmental, or acquired), in which the complications could not be traced back to the exact patient subgroup, the article was excluded on the basis of insufficient data. In total 228 articles were excluded based on the eligibility criteria. Eighty-one articles on MDO for congenital deformities were included. The flowchart of the literature search and selection process through the different phases of the systematic review (PRISMA) is shown in Fig. 2 .

Flowchart of the search and selection procedure using PubMed (National Library of Medicine, NCBI), EMBASE, and the Cochrane Central Register of Controlled Trials.