Background It is unknown whether it is possible to measure implant stability with polyetheretherketone (PEEK) healing abutments using electronic percussive testing (EPT). Purpose To investigate the reliability of the percussive test values (PTVs) measured with PEEK healing abutments and to compare them with the PTVs measured with titanium healing abutments. Material and Methods Thirty dental implants were inserted into the fresh pelvis belonging to a cow. Titanium healing abutments (2 and 5 mm), PEEK healing abutments (5 mm), and prepable standard titanium abutments (5 mm) were screwed to the implants, respectively, and PTVs were measured by two examiners using a wireless EPT device. Differences in PTVs between different dental implant components were evaluated using Friedman's test with post hoc Wilcoxon signed-rank test and Bonferroni correction. Inter and intra-observer reliabilities were detected using interclass correlation coefficients (ICCs) Results The mean PTVs obtained using the PEEK healing abutments were significantly higher than the mean PTVs obtained using the other abutments for both examiners (P < .01). The ICCs for intra-observer reliability were detected as poor for PEEK healing abutments; and excellent for the other abutment types for both examiners. The ICCs for the inter-observer reliability between the two examiners were poor (0.25) for PEEK healing abutments, and excellent for the 2 and 5 mm titanium healing abutments and for standard abutments (0.82, 0.84 and 0.93, respectively). Conclusion Within the limitations of this in vitro study, it may be concluded that EPT measurements should be avoided using PEEK healing abutments because of the poor reliability.