Quality of English literature reviews concerning longevity of direct posterior restorations in permanent teeth – a systematic review of reviews [Protocol]

REVIEW AIM: The aim of this systematic review is to appraise the quality of existing reviews in the English dental literature in regard to general review methodology, as well as specifically to the comparison method applied, during the last 20 years concerning the compared longevity of different types of direct restorations placed in permanent posterior teeth and subsequently the validity of such reviews’ conclusions.

SYSTEMATIC LITERATURE SEARCH: Databases: MEDLINE accessed via PubMed; CENTRAL accessed via Cochrane Library; Open access sources: Biomed Central, Database of Open Access Journals (DOAJ); Regional databases: IndMed, Sabinet, Scielo; Grey-Literature sources: Scirus (Medicine), OpenSIGLE, Google Scholar. Journals, identified as not being fully indexed in any of the above electronic databases, will be hand-searched. Searching of reference lists of included articles.

SEARCH TERM DEVELOPMENT: Strings of search terms (containing MeSH and text search terms) together with Boolean operators will be developed and utilized for searching the databases.

ARTICLE SELECTION CRITERIA: (i) Reports of systematic or non-systematic reviews and meta-analyses concerning clinical trials; (ii) Publication language in English; (iii) Review topic related to the comparative longevity (failureand/or survival rate) of direct posterior restorations in vital permanent teeth (premolar and/or molar teeth); (iv) Publication period 01.01.1992 – present. Exclusion will be based on the following criteria: (i) The review constitutes in essence either a: Clinical practice guideline; audit; survey; technical report; review commentary; historical overview; (ii) Focus on adhesive bonding systems, only; (iii) Focus on endodontic treated teeth; (iv) Expert opinion with narrow focus without some form of comprehensive overview of the current trial literature; (v) Lack of some explicit comparison of longevity data between at least two different tooth restoration types.

DATA EXTRACTION: Both authors (SM and VY) will evaluate the quality of the accepted reviews independently using the 11-item AMSTAR tool; without being blinded to authors, institutions, journal names and review results. In addition, the type of longevity comparison utilized in each review will be recorded. The three types of comparisons: naïve – indirect; adjusted indirect; and direct comparison, as well as the comparison result between the compared restoration types, i.e. the longevity of Type A being larger or equal to Type B, will be recorded.

DATA ANALYSIS AND REPORTING: The unit of investigation is the review report. A ratio of the obtained AMSTAR score (achieved score/total number of assessed items) will be calculated for each accepted review. Log transformation will be utilized, to meet assumptions of linearity by computing the natural logarithm of each AMSTAR score ratio. Simple linear regression analysis will be conducted in order to quantify the relationships between: (i) year of review publication and general review quality (as expressed by the AMSTAR score ratio); (ii) general review quality and type of applied comparison. The trend lines of the regression results will be plotted in separate scatter graphs. The difference in conclusions drawn from naïve-indirect comparison and adjusted indirect/direct comparison will be quantified and discussed in light of review recommendations for clinical practice.