Root canal therapy, fixed partial dentures and implant-supported crowns, have similar short term survival rates.

DATA SOURCES

Searches using the Medline, Cochrane and Embase databases and 'citation mining' (identifying references from included studies) were carried out. In addition, experts' recommendations for data sources were followed, and the table of contents of every issue of the most recent 2 years of a given list of dental journals were reviewed, the latter representing half of the total number of original research articles in English from the past 5 years on implant-supported crown (ISC), fixed partial dentures (FPD) and root canal (RC) therapy. Publication language was limited to English and grey literature was excluded, namely proceedings of conferences not listed in Medline, Cochrane or Embase databases, meetings and lectures.

STUDY SELECTION

Comparative or noncomparative, prospective or retrospective longitudinal data were selected that related to clinical, biological, psychological and economic outcomes, as well as beneficial or harmful effects, of saving teeth by root canal treatment and/or alternative treatments, including: extracting the tooth and leaving an edentulous space or replacing the missing tooth with a fixed-partial-denture or implant-supported tooth.

DATA EXTRACTION AND SYNTHESIS

Three pairs of investigators (each team dealing with one of the three main treatment options) independently searched, selected and extracted data for analysis. Tables of evidence were created from articles that met the validity criteria. Each selected paper was given a quality score, where the maximum possible was 17. Discussion and consensus were used to resolve disagreement. Interpretation of the outcome data and classification of data according to success or survival and the type of study were verified by two statisticians. Qualitative and quantitative methods were used to integrate the evidence. The data were analysed by deciding whether and what data to combine, and measuring the statistical heterogeneity of the data using Cochrane Q and I statistics. For the purpose of comparison, clinical outcomes were grouped into three time-intervals (2-4 years, 4-6 years and >6 years). Each discipline and followup interval from individual studies were displayed in a Forest plot with Wilson Score 95% confidence intervals. Meta-analysis produced pooled points and weighted averages. Psychosocial and economic outcome data was subjected to narrative review only.

RESULTS

The 143 selected studies varied considerably in design, definition of success, assessment methods, operator type and sample size. Direct comparison of treatment types was extremely rare. Limited psychosocial data revealed the traumatic effect of loss of visible teeth. Economic data were largely absent. Success rates for ISC were higher than for RC treatment with direct or indirect restoration and FPD, respectively. Success criteria, however, differed greatly between treatment types, rendering direct comparison of success rates futile. Long-term survival rates for ISC and RC therapy were similar and superior to those for FPD.

CONCLUSIONS

The lack of studies with similar outcome criteria and comparable time intervals limited the ability to make valid comparisons between these treatments. It was possible to conclude that ISC and RC treatments have superior long-term survival compared with FPD. Limited data suggested that extraction without replacement resulted in inferior psychosocial outcomes compared with alternatives. Long-term prospective clinical trials with large sample sizes and clearly defined outcome criteria are needed.