Does Classification of Composites for Network Meta-analyses Lead to Erroneous Conclusions?

Area of Science:

• Dental Materials Science
• Clinical Dentistry
• Evidence Synthesis

Background:

• Dental composite materials are classified using various systems, including manufacturer, filler particle size, resin-monomer base, and viscosity.
• Network meta-analyses are used to rank composite material classes based on clinical trial data, aiding dentists in material selection.
• Annual failure rates (AFRs) offer an alternative method for assessing material performance without requiring classification.

Purpose of the Study:

• To evaluate the agreement of material rankings derived from different classification systems for dental composites.
• To determine if varying classification methods lead to discrepancies in material performance assessments.
• To compare rankings obtained through network meta-analysis with those derived from AFRs.

Main Methods:

• A systematic review of randomized controlled trials (2005-2015) involving composite restorations in permanent teeth was conducted.
• Network meta-analyses were performed to rank composite classes based on different criteria (manufacturer, particle size, resin-monomers, viscosity) and adhesives.
• Material combinations were also ranked using annual failure rates (AFRs).

Main Results:

• Analysis of 42 studies (6088 restorations) revealed significant agreement between rankings of material class combinations across different classification systems (R² 0.03-0.56).
• Rankings based on AFRs demonstrated low precision and poor agreement with other classification systems.
• AFRs were found to be significantly correlated with the follow-up periods of the included trials.

Conclusions:

• High agreement exists between material rankings derived from different classification systems, supporting cautious deductions about specific material performance.
• Syntheses relying on AFRs may yield unreliable results due to their dependence on follow-up duration and inherent low precision.
• The findings suggest that classification systems, when used in meta-analyses, provide consistent relative rankings for dental composites.