Goldmann tonometry tear film error and partial correction with a shaped applanation surface

Purpose The aim of the study was to quantify the isolated tear film adhesion error in a Goldmann applanation tonometer (GAT) prism and in a correcting applanation tonometry surface (CATS) prism. Methods The separation force of a tonometer prism adhered by a tear film to a simulated cornea was measured to quantify an isolated tear film adhesion force. Acrylic hemispheres (7.8 mm radius) used as corneas were lathed over the apical 3.06 mm diameter to simulate full applanation contact with the prism surface for both GAT and CATS prisms. Tear film separation measurements were completed with both an artificial tear and fluorescein solutions as a fluid bridge. The applanation mire thicknesses were measured and correlated with the tear film separation measurements. Human cadaver eyes were used to validate simulated cornea tear film separation measurement differences between the GAT and CATS prisms. Results The CATS prism tear film adhesion error (2.74±0.21 mmHg) was significantly less than the GAT prism (4.57±0.18 mmHg, p<0.001). Tear film adhesion error was independent of applanation mire thickness (R2=0.09, p=0.04). Fluorescein produces more tear film error than artificial tears (+0.51±0.04 mmHg; p<0.001). Cadaver eye validation indicated the CATS prism's tear film adhesion error (1.40±0.51 mmHg) was significantly less than that of the GAT prism (3.30±0.38 mmHg; p=0.002). Conclusion Measured GAT tear film adhesion error is more than previously predicted. A CATS prism significantly reduced tear film adhesion error bŷ41%. Fluorescein solution increases the tear film adhesion compared to artificial tears, while mire thickness has a negligible effect.