Comparison of intraocular pressure measured by Goldman applanation tonometer, noncontact tonometer and trans-palpebral tonometer

To evaluate reliability and agreement of non-contact tonometry (NCT) and trans-palpebral tonometry (TPT) compared with Goldmann-applanation tonometry (GAT). A total of 224 eyes from 112 patients were included between January and April 2025. Intraocular pressure (IOP) was measured using GAT, NCT, and TPT. Comparison among the three measurement groups were performed using the related-samples Friedman test. Post hoc pairwise comparisons were conducted using the Wilcoxon signed-rank test with Bonferroni correction. Reliability was assessed using Cronbach’s alpha and the intraclass correlation coefficient (ICC). Agreement between methods was evaluated using Bland–Altman plots and Spearman’s rank-order correlation. Median IOP were 14 mmHg for GAT, 14 mmHg for NCT and 15 mmHg for TPT. Friedman’s ANOVA showed a statistically significant difference among three methods (χ² = 40.652, p < 0.001). TPT readings differed significantly from both GAT and NCT (p < 0.001). The difference between GAT and NCT was not significant (p = 0.024) because it did not meet the Bonferroni-adjusted threshold (α = 0.017). Cronbach’s alpha was 0.691, indicating acceptable internal consistency. The ICC for average IOP was 0.668 (95% CI: 0.571–0.743, p < 0.001), suggesting moderate reliability. Spearman’s correlation showed moderate-agreement between GAT and NCT (ρ = 0.586, p < 0.001), weak-agreement between TPT and NCT (ρ = 0.315, p < 0.001), and no correlation between GAT and TPT (ρ = 0.090, p = 0.177). Bland-Altman analysis confirmed good-agreement between GAT and NCT, fair-agreement between NCT and TPT, and poor-agreement between GAT and TPT. GAT and NCT showed comparatively better agreement, whereas TPT demonstrated wide limits of agreement with GAT and should not be used interchangeably for clinical decision-making; TPT may be considered only for preliminary screening when contact tonometry is not feasible.