Effects of using different criteria for caries removal: a systematic review and network meta-analysis

Conventionally, caries excavation is performed until only hard dentine remains, while more selective and reliable criteria might be available. We aimed at systematically comparing the effects of using different excavation criteria via network meta-analysis.

Electronic databases were searched for randomised or non-randomised clinical trials (RCTs/NRCTs) evaluating excavation of cavitated lesions.

Criteria were divided into six groups: Excavation until pulpo-proximal dentine on the cavity floor was (1) either hard on probing, (2) slightly softened on probing, (3) not stainable by caries-detector-dye, or until (4) self-limiting polymer burs, (5) fluorescence-assisted devices or (6) chemo-mechanical gels indicated termination of the excavation. Evaluation of risk of complications, risk of pain/discomfort, excavation time, and number of remaining bacteria were then undertaken using Bayesian network meta-analysis.

28 studies (19 RCTs, 9 NRCTs) with 1782 patients (2555 lesions), most of them investigating primary teeth, were included. Risk of complications was highest when excavating until only non-stainable dentine remained, and lowest when not attempting to remove all softened dentine. Risk of pain significantly decreased if self-limiting chemo-mechanical excavation or fluorescence-assisted lasers were used instead of excavating until all dentine was hard. When not attempting to remove all softened dentine, the time required for excavation was shortest, whilst the greatest number bacteria remained.

Not attempting to remove all softened or stainable dentine might reduce the risk of complications. Data regarding self-limiting excavation is insufficient for definitive conclusions. Excavation criteria should be validated against clinically relevant outcomes.

Given current evidence, dentists might not need to attempt excavation until only hard dentin remains in proximity to the pulp. Instead, their choice of excavation criterion or method should be guided by clinical requirements and outcomes.