Purpose: To evaluate the effect of three different liquids used for intrapulpal pressure (IPP) simulation on the microtensile bond strength (µTBS) of three adhesive systems to dentin. Materials and Methods: The occlusal surfaces of sound human molars were ground flat down to mid-dentin depth. The teeth were bonded under 15 mmHg simulated IPP using distilled water, phosphate buffered saline, or human plasma as a simulating liquid. Three adhesive systems were tested: a single-bottle etch-and-rinse adhesive (SingleBond, 3M ESPE), and two single-step self-etching adhesives (G-Bond, GC) and (iBond, Heraeus Kulzer). Resin composite (Tetric EvoCeram, Ivoclar Vivadent) buildups were made in 2 increments, each 2 mm in height. Specimens were stored in artificial saliva under 20 mmHg IPP at 37°C for 24 h prior to testing. µTBS (n = 15) was tested using a universal testing machine, and failure modes were determined. Data were statistically analyzed using ANOVA and Bonferroni post-hoc tests at p = 0.05. Results: With SingleBond adhesive, distilled water showed significantly higher µTBS compared to plasma and phosphate buffered saline. With G-Bond, no significant difference was found between distilled water and phosphate buffered saline, whereas plasma showed significantly lower µTBS values. In contrast, no significant difference was encountered between the three IPP liquids for iBond adhesive. Predominant modes of failure were adhesive and mixed. Conclusion: A difference in intrapulpal pressure simulating liquids influences the bonding of adhesives to dentin. Etch-and-rinse adhesives are more sensitive to intrapulpal simulating liquids than are self-etching adhesives. Adhesives containing protein-coagulating components perform better with plasma perfusion than those lacking such components. Palabras clave: intrapulpal pressure, human plasma, microtensile bond strength, adhesive systems