Comparing Methods of Quantifying Diplopia

Purpose: Quantification of diplopia is important for describing severity of strabismus, measuring change over time, and reporting surgical outcomes. The cervical range of motion (CROM) method has been proposed as a simple, inexpensive alternative to the Goldmann perimeter for quantifying diplopia. The purpose of this study was to compare these 2 techniques and describe advantages and disadvantages. Design: Cohort study. Participants: Seventy-six consecutive patients with binocular diplopia associated with any type of strabismus. Methods: Patients underwent diplopia assessment with the CROM method and Goldmann perimeter; diplopia was scored between 0 and 100 based on previously published scoring systems. Where CROM and Goldmann results were disparate by >20 points, the medical record was reviewed independently by 2 clinicians to determine the most likely reason for the discrepancy. Main Outcome Measures: (1) Measure of agreement between diplopia scores using the CROM and Goldmann methods using the κ test and (2) the reasons for any disagreement between tests of >20 points. Results: Overall agreement between the 2 tests was good (intraclass correlation coefficient, 0.65; 95% confidence interval, 0.50-0.77). In 49 (64%) of 76 tests, the CROM and Goldmann results were within 20 points. Of the 27 (36%) showing a more than 20-point discrepancy, 17 were worse using the Goldmann technique and 10 were worse using the CROM technique. The most frequent reason for worse diplopia using the Goldmann technique was poorer ability to fuse or suppress in the Goldmann apparatus compared with the real-world targets used in free space for the CROM method. Worse diplopia using the CROM method most often was the result of the deviation being present for distance only. In some cases, differences were the result of the greater weighting of primary position using the current scoring system for the CROM method. Conclusions: In most cases, the CROM and Goldmann methods provide equivalent measures of diplopia severity. However, the Goldmann method seems to overestimate diplopia in patients with fragile fusion or tenuous suppression and seems to underestimate diplopia in deviations present for distance only. The CROM method may be more representative of diplopia severity as experienced in everyday life.