Purpose: The purpose of this study was to compare the reliability of the 'gold standard' Goldmann applanation tonometer (GAT), with that of the ocular response analyser (ORA), and the dynamic contour tonometer (DCT).
Patients and methods: A total of 694 subjects were recruited to participate from the TwinsUK (UK Adult Twin Registry) at St Thomas' Hospital, London. Intraocular pressure (IOP) was measured using GAT, ORA, and the DCT. The agreement between the three methods was assessed using the Bland-Altman method. Repeatability coefficients and coefficient of variation between first and second readings of the same eye were used to assess reliability.
Results: Mean age was 57.5 years (SD, 13.1; range, 16.1-88.5). The mean IOPs, calculated using the mean of two readings from the right eye were as follows: Goldmann (GAT), 14.1+/-2.8 mm Hg; IOPg (ORA), 15.9+/-3.2 mm Hg; IOPcc (ORA), 16.6+/-3.2 mm Hg; and DCT, 16.9+/-2.7 mm Hg. The 95% limits of agreement were for ORA (IOPcc): GAT, -2.07 to 7.18 mm Hg; for DCT: GAT, -0.49 to 6.21 mm Hg; and for DCT: ORA (IOPcc), -3.01 to 4.85 mm Hg. Coefficients of variation for the three tonometers were GAT, 8.3%; ORA, 8.2%; DCT, 6.3%. The repeatability coefficients were 3.4 mm Hg for GAT, 3.57 mm Hg for ORA and 3.09 mm Hg for DCT. GAT and ORA (IOPg) readings showed a positive correlation with central corneal thickness (P<0.005).
Conclusions: This study found similar reliability in all three tonometers. Bland-Altman plots showed the three instruments to have 95% limits of agreement outside the generally accepted limits, which means they are not interchangeable. GAT measurements were found to be significantly lower than the two newer instruments.