Objectives: To test the null hypothesis that perfusion of dentin has no effect on the sealing or bond strength in Class V resin composite restorations using a two-step total-etch (Scotchbond 1), or a one-step self-etch (all-in-one) (Prompt L-Pop) adhesive. Methods: Dentin permeability in Class V cavity preparations cut into extracted human third molars was measured in a fluid transport setup. Resin composites were placed using either Scotchbond 1, or Prompt L-Pop with or without dentin perfusion at 1.47 kPa. Permeability was re-measured, and the reduction in permeability was calculated as a percentage of the initial value. Additionally, microtensile bond strength (muTBS) to dentin was measured. Results: Whereas the ability of Scotchbond 1 to seat dentin was significantly decreased due to dentin perfusion (p = 0.00025), that of Prompt L-Pop was not affected (p = 0.099). Scotchbond 1 produced significantly less dentinal seat than Prompt L-Pop only in the restorations carried out with perfusion (p = 0.0003), in the absence of perfusion, however, the difference was not significant (p = 0.343). Dentin perfusion significantly decreased the muTBS of Scotchbond 1 to dentin (p = 0.001). Due to excessive premature failure during sectioning, muTBS test could not be performed with Prompt L-Pop. Significance: Dentinal fluid flow may have detrimental effects on the sealing ability of dental adhesives depending on the type of adhesive used. Studies aiming to rank adhesives in relation to their sealing ability in vitro should not disregard dentin perfusion. (C) 2003 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
