The sealing ability of retrograde root fillings of amalgam plus cavity varnish, EBA cement, glass ionomer cement, light-cured composite resin, dentine bonding agent, and light-cured composite resin plus dentine bonding agent were compared with laterally condensed conventional root fillings. A dye penetration technique was used to assess microleakage and it was concluded that none of the materials produced a perfect seal but that glass ionomer cement, light-cured composite resin, light-cured composite resin plus dentine bonding agent and dentine bonding agent alone produced better seals than conventional laterally condensed gutta-percha, amalgam plus cavity varnish and EBA cement.

A dye penetration technique was used to investigate the sealing ability of amalgam retrograde root fillings in vitro. A factorial design was employed to determine the effect of amalgam fillings of 1, 2 and 4 mm in length with and without prior filling of the canals with gutta-percha. The results were compared with a group of teeth filled with laterally condensed gutta-percha alone and another with laterally condensed gutta-percha plus root resection but with no retrograde filling. It was concluded that increasing the length of the amalgam filling did not improve the seal, that prior filling of the canal did not improve the seal, that retrograde root fillings were no worse than laterally condensed gutta-percha alone and that root resection of teeth filled with laterally condensed gutta-percha without retrograde filling was no worse than those filled with laterally condensed gutta-percha without root resection or retrograde root filling.