The Accuracy of Intraocular Pressure Measurements

The Accuracy of Intraocular Pressure Measurements

SURVEY OF OPHTHALMOLOGY VOLUME 51  NUMBER 1  JANUARY–FEBRUARY 2006 CURRENT OPHTHALMOLOGY AND MEDICINE WILLIAM GOOD AND PAUL LAMA, EDITORS The Acc...

48KB Sizes 0 Downloads 12 Views

SURVEY OF OPHTHALMOLOGY

VOLUME 51  NUMBER 1  JANUARY–FEBRUARY 2006

CURRENT OPHTHALMOLOGY AND MEDICINE WILLIAM GOOD AND PAUL LAMA, EDITORS

The Accuracy of Intraocular Pressure Measurements Is the Tono-Pen Accurate for Measuring Intraocular Pressure in Young Children With Congenital Glaucoma? by J Levy, T Lifshitz, S Rosen, Z Tessler, and BZ Biedner. J AAPOS 9:321--5, 2005 Purpose: We sought to compare intraocular pressure (IOP) measurements by Perkins tonometer and TonoPen in young children with primary congenital glaucoma (PCG). Methods: This was a retrospective comparative case series. We reviewed the clinical records of all children with primary congenital glaucoma who underwent examinations under general anesthesia at Soroka University Medical Center between January 1999 and July 2002. Our main outcome measures were IOP with Perkins hand-held tonometer and Tono-Pen tonometer. Results: A total of 28 eyes of 16 children were examined under general anesthesia. The mean IOP was 18 6 6 mm Hg with the Perkins tonometer and 22 6 8 mm Hg with the Tono-Pen. In 18 eyes, IOP was less than 21 mm Hg with the Perkins tonometer; these eyes had already undergone surgical procedures. The other 10 eyes with IOP greater than 21 mm Hg with the Perkins tonometer underwent surgery at the end of the examination under anesthesia. In eyes with IOP greater than 16 mm Hg (Group A, n 5 18), a significant difference (p ! 0.001) was found between the Perkins and Tono-Pen measurements, even although the values were strongly correlated (r 5 0.60). In contrast, in eyes with IOP less than 16 mm Hg (Group B, n 5 10) no statistically significant difference (p 5 0.28) and good correlation (r 5 0.78) were obtained. A difference of 5.8 6 3.8 mm Hg and 0.6 6 1.7 mm Hg between Perkins and Tono-Pen readings, respectively, was found in Groups A and B. Conclusions: Tono-Pen readings disagree with Perkins tonometer measurements for measuring IOP in children with PCG who present with IOP greater than 16 mm Hg and tends to overestimate IOP. A further study with a similar population is necessary to confirm these results.

Comment The authors present results from a method comparison study using two devices for intraocular pressure (IOP) measurement, the gold standard Perkins tonometer and the Tono-Pen. Data was collected only in a diseased population, children with congenital glaucoma of mean age 26 weeks (range 1--112 weeks). Although retrospective, the design of the study was very good. The study demonstrates the importance of statistical methodology and in particular the importance of mean versus difference analysis (also known as Bland-Altman plots) in such studies. The main finding was that despite acceptable correlation between the two methods, there was a clear but unpredictable bias when IOP was greater than 16 mm Hg. This bias was consistently for significantly higher Tono-Pen measurements, on average 5.8 mm Hg (standard deviation 63.8). The authors discuss well the confounding factors within their study design. It is highly recommended that this paper is read in conjunction with another report in the same issue of this journal, that of Simon et al: ‘‘Central Corneal Thickness and Glaucoma in Aphakic and Pseudophakic Children.’’ Future, well-designed prospective studies are needed in this area controlling especially for type and delivery of anesthesia, corneal thickness, and other measurements relevant to the diagnosis of pediatric 86 Ó 2006 by Elsevier Inc. All rights reserved.

0039-6257/06/$--see front matter doi:10.1016/j.survophthal.2005.11.014

CURRENT OPHTHALMOLOGY AND MEDICINE

87

glaucoma. Perhaps a number of algorithms for diagnosing glaucoma in the pediatric population may be prospectively designed and then, after such data is collected, may be tested against final incontrovertible evidence of those developing the disease. Such information should help to increase the efficiency of clinical practice in this area. Sean I. Chen, MD Liverpool, United Kingdom