Purpose: To evaluate the extent of crestal bone loss (CBL) at 2, 12, 36, and 60 months in implants placed with the shoulder at the equicrestal level and 2 mm below the alveolar ridge. Materials and Methods: A split-mouth randomized controlled clinical trial was conducted by selecting subjects with a Kennedy Class IV partially edentulous mandible. Two implants of equal length and diameter were inserted, one equicrestal and the other subcrestal, in the lateral incisor sites. Using Rinn centering devices, intraoral periapical radiographs were taken at implant insertion (T0) and at 2 (T1), 12 (T2), 36 (T3), and 60 months (T4). Descriptive statistics and t test were used, with P ≤ .05 considered statistically significant. Twentyfive patients were recruited, with a mean age of 65 ± 9.88 years (range: 42 to 82 years), and none dropped out. A total of 50 implants were inserted, 25 at the crestal level and 25 at the subcrestal level. Results: At the 60-month follow-up, no implant or prosthetic failure was recorded. An average CBL of 0.81 ± 0.40 mm (range: 0.1 to 1.6 mm) was recorded in the crestal implant group, while the subcrestal implants had an average CBL of 0.87 ± 0.41 mm (range: 0.2 to 2 mm); however, the higher CBL in the subcrestal group was not statistically significant (P = .65). Comparing the mean CBL of both groups at the various follow-ups, greater crestal bone resorption was recorded in subcrestal implants between T0 and T1 (0.25 vs 0.1 mm) and between T1 and T2 (0.39 vs 0.23 mm), while in subsequent follow-ups, a greater and statistically significant (P = .01) CBL was recorded in equicrestal implants between T3 and T4 (0.05 vs 0.18 mm). Conclusions: Thus, over time, the extent of CBL seems to be reduced in subcrestal implants, with bone retention above the implant shoulder. Although the position of the implant shoulder relative to the crestal ridge does not affect the CBL, subcrestal placement is recommended in order to reduce the risk of exposing the rough implant surface.