How and where should intraocular pressure be measured after photorefractive keratectomy?

The study by Levy et al.1 confirms that central corneal Goldmann applanation tonometer (GAT) measurements underestimate intraocular pressure (IOP) after photorefractive keratectomy (PRK), especially in steroid-treated eyes. The authors do not report the pre-PRK IOP values or the duration of steroid use at the time of the study. Although the fellow eye served as the control in this study, most patients have bilateral PRK and it is therefore important to measure IOP accurately in the treated eye itself. Since the central cornea is altered by the PRK procedure, the choice of instrument and area of the cornea in which to measure IOP are not clear. Previous studies indicate that the pneumotonometer is reliable in measuring IOP after PRK,2 while noncontact tonometer measurements need a correction factor, based on the degree of myopia treated.3 Factors described to explain underestimation of IOP by GAT include corneal thinning,4 older patient age,5 and corneal flattening.1 Theoretical considerations indicate that Tono-Pen measurements may be less affected by the corneal alterations that occur after PRK.1 Early studies comparing the IOP measured by the Tono-Pen and GAT in patients who did not have corneal refractive surgery indicate that the Tono-Pen overestimated IOP by 0.3 to 0.6 mmHg (P > .05)6 to 1.7 mmHg (P < .0001).7 The authors conclude that these differences are unlikely to be clinically significant. However, since the Tono-Pen works partly on the applanation principle (contact diameter of 1.5 mm), changes in corneal thickness could affect IOP measurements, although less than with the GAT. We performed a study similar to that of Levy et al. in 29 patients (29 eyes) whose mean age was 30.6 years ± 8.7 (SD) (range 21 to 52 years). The measurements were performed between June and December 1998, and the mean duration after PRK was 27.8 ± 3.3 months (range 3 to 56.1 months). Mean preoperative refractive error in the treated eyes was −7.73 ± 4.95 diopters (D) (range −2.00 to −22.00 D). We measured preoperative keratometry (Shin-Nippon), central and temporal peripheral pachymetry (ultrasound pachymeter), and IOP in the central cornea with a GAT attached to a Haag-Streit slitlamp in both eyes. Postoperative measurements in both eyes included keratometry, pachymetry (central and temporal periphery), refraction, and central and peripheral IOPs (Tono-Pen 2 tonometer and GAT). The results of the IOP measurements are shown in Table 1. Postoperative central GAT was less than the preoperative value (1.06 ± 3.65 mmHg; P = .06) and the central GAT in the fellow eye (0.66 ± 2.30 mmHg; P = .06). Intraocular pressure measured by the Tono-Pen was higher than that estimated by GAT in both the central and peripheral cornea. However, the magnitude of the difference between PRK-treated and untreated eyes was significantly different in only the central cornea (P = .03). The underestimation of central IOP by GAT was not related to any of the following parameters in our study—age and sex of the patient, degree of myopia treated, the interval between PRK and IOP measurement, and postoperative corneal pachymetry.Table 1: Central and peripheral IOP after PRK in 29 patients.The lack of a significant change in Tono-Pen and GAT temporal IOP measurements after PRK has been described.8 However, IOP measurements in the thicker peripheral cornea are likely to be different from those in the central cornea. The need to abduct the eye to perform temporal GAT may also result in artifactual changes in IOP. Although the Tono-Pen is less likely to be affected by the corneal changes after PRK, GAT is the most widely used device to measure IOP. We feel that a large-scale prospective study of preoperative and postoperative Tono-Pen and GAT IOP measurements in the central and peripheral cornea would provide a better understanding of the interaction between corneal curvature, thickness, treatment zone size, myopia treated, and the IOP. This would help answer the question of where and how IOP should be measured after PRK.