Failure analysis and survival rate of post and core restorations under cyclic loading

To investigate in a laboratory setting the influence of (i) post material (ii) preparation design and (iii) luting agent on the survival probability of root filled teeth, restored with all-ceramic restorations.

The crowns of 80 extracted single-rooted human teeth were removed, and root canal treatment was performed including canal filling with Gutta-percha without sealer (crown-down-pressureless technique). The root fillings were removed and the root canal enlarged with a reamer up to size 110. Prefabricated zirconia (CeraPost) or glass-fibre-reinforced posts (DentinPost) were luted using either Ketac Cem or Panavia F 2.0. A core build-up was applied (Clearfil Photocore), and the teeth were prepared with or without a 2-mm ferrule design (n=10 per experimental group). The prepared teeth were scanned (Cerec 3D) and crowns fabricated. After luting of the crowns (Ketac Cem), teeth were subjected to thermocycling (×4000, 5-55 °C) and cyclic loading (1.5 million cycles, 90 N). After load cycling, the teeth were immersed in methyleneblue solution for 24 h and subsequently sectioned in three segments for a dye penetration test. Kaplan-Meyer analysis was performed to assess the survival probability followed by a Cox regression analysis (α=5%).

Teeth prepared using the ferrule design as well as the teeth with DentinPosts exhibited a significantly higher survival probability (P<0.05). The luting agent was of minor importance (P>0.05). Most common failure was debonding of posts (CeraPost) and post fracture (DentinPost). The majority of the teeth showed dye penetration after cyclic loading.

Post material and ferrule design were of paramount importance regarding the survival probability of the post and core restorations using pre-fabricated posts. DentinPosts showed superior results versus CeraPosts.