[Clinical evaluation of the Pascal dynamic contour tonometer].J Fr Ophtalmol 2007; 30(3):260-70JF
The Pascal dynamic contour tonometer (DCT) was designed to measure IOP independently of corneal properties. This study aimed at 1) assessing the intra- and interindividual variability of DCT IOP measurements, the differences between DCT and applanation tonometry IOP measurements (APL), and their correlations with central corneal thickness (CCT); 2) analyzing the variability of the ocular pulse amplitude (OPA) and its correlations with age, blood pressure (BP), cardiac beat pulse (CP), diagnosis of glaucoma, IOP, and severity of glaucomatous visual field (VF) defects.
Twenty-five normal subjects (25 eyes), 14 patients with ocular hypertension (27 eyes), and 54 glaucomatous patients (104 eyes) were included in this prospective study. In the first 12 normal subjects, three consecutive IOP measurements were taken by three different observers using DCT, directly followed by three measurements with APL by the same observer. In the following 13 subjects, the reverse sequence was followed. In the other group, the IOP measurements (three DCT and three APLs) were taken by the same observer. Only DCT measurements with quality levels 1-3 were considered for analysis.
In the normal group, DCT IOP measurement variability varied between 4.4%-7.3% (intraobserver variation coefficient) and 8% (interobserver variation coefficient). DCT IOP measurement was not influenced by the sequence of measurements or the observer. DCT overestimated IOP by a mean of 2.2 mmHg compared with APL (p<0.001). The 95% limits of agreement for each subject tested with both tonometers ranged from -0.5 mmHg to +6.3 mmHg. IOP APL and DCT measurements were strongly correlated. Both DCT and APL were not correlated with CCT. OPA ranged from 1.2 mmHg to 6.6 mmHg (mean, 3.1+/-1.2 mmHg) and was comparable between the three observers. Intraobserver OPA variability ranged from 7.6% to 9.5%. The interobserver OPA variability coefficient was 8.8%. OPA was only correlated with systolic BP (p<0.05). In glaucomatous patients, the correlation between DCT and APL IOP measurements was highly significant (r=0.860, p<0.001). DCT overestimated IOP by a mean 2 mmHg compared with APL (p<0.001). IOP differences between both tonometers were not influenced by the sequence of measurements. Unlike APL, DCT was not or only slightly influenced by CCT (p=0.07 for DCT; p=0.001 for APL). The mean difference between IOP DCT and APL was larger in thin corneas (<520 microm): 2.8+/-3.1 mmHg versus 0.8+/-2.3 mmHg in thick corneas (580 microm) (p=0.001). OPA was not correlated with age. It was positively correlated with IOP (p<0.001), systolic BP (p=0.047), and MD (mean deviation) (p=0.018). It was negatively correlated with diastolic BP (p=0.003), cardiac frequency (p<0.001), severity of glaucomatous VF defects (p=0.002), and PSD (pattern standard deviation) (p=0.008). It was significantly higher in the OHT subgroup and significantly lower in the NTG subgroup (p<0.05). In both groups, the IOP difference between DCT and APL was not correlated with age (p>0.05).
IOP measurements with the Pascal(R) DCT and APL correlated well and were reproducible. DCT IOP measurement variability was slightly higher than APL with relatively wide 95% limits of agreement. Considering the entire study population, DCT overestimated IOP by a mean 2.0 mmHg compared with APL. DCT was independent of CCT, especially in thin corneas. The DCT does not appear to be clinically advantageous over the Goldmann tonometer in the IOP measurement in thick corneas. Therefore an IOP follow-up by APL tonometry and pachymetry appeared to be mandatory for the interpretation of the true IOP. Interindividual OPA variations were high, as was measurement variability. OPA was correlated with BP, cardiac frequency, IOP, diagnosis of glaucoma, and severity of glaucomatous VF defects. These must be considered in the clinical interpretation of this parameter.