This study aimed to evaluate the effect of dentin hypersensitivity treatments on immediate and long-term shear bond strength (SBS) of composite restorations. Ninety non-carious extracted human molars were cut to expose dentin, which was embedded in acrylic resin, and randomly divided into three groups (n = 30/group) according to surface treatment: 1) no treatment (C and C*; control); 2) silver diamine fluoride with potassium iodide (SDF/KI and SDF/KI*; Riva Star); and 3) nano-hydroxyapatite (nHAp and nHAp*; PrevDent). The specimens were etched through the etch-and-rinse technique, followed by universal adhesive application and resin composite cylinders (2.38 mm in diameter × 3.5 mm high). The SBS was tested immediately (24 h after the restoration) and after thermocycling (*) (5000 cycles, 5 °C to 55 °C) at a 0.5 mm/min crosshead speed using a universal testing machine. A stereomicroscope was used to evaluate the mode of failure, and representative scanning electron microscopy (SEM) images were also acquired. Data normality was verified, and two-way ANOVA and Tukey's post hoc tests were performed for multiple comparisons (α = 0.05). The control group presented the highest SBS (27.10 MPa), while SDF/KI* had the lowest values (6.87 MPa). nHAp-based desensitizer exhibited higher SBS than SDF/KI for both immediate (22.6 MPa) and thermocycled (19.03 MPa) conditions. No intragroup difference was evidenced between immediate and thermocycled samples for any group. Most specimens for the C and nHAp groups presented mixed failure, while the SDF/KI groups presented comparable adhesive and mixed failures. The SBS of adhesive restorations after the application of desensitizing agents is material dependent, where SDF/KI reduces SBS values below the acceptable minimum bond strength, while the nHAp application meets the minimally required bond strength.