Postoperative Diplopia Prediction Test Demonstrates Poor Predictive Value

The postoperative diplopia test (PODT) may not be useful for predicting postoperative diplopia due to excessive test-retest and interobserver variability, according to research published in Journal of the American Association for Pediatric Ophthalmology and Strabismus.  

Researchers recruited 39 participants (mean age, 39.4 years) with constant manifest strabismus and conducted the study in 2 phases. During phase 1, patients were assigned to 1 of 2 groups. Group 1 participants (n=20) underwent the PODT procedure twice by the same observer, while individuals in group 2 (n=19) underwent the PODT twice by 2 different observers. 

Phase 2 involved a retroscopic review of exam notes from phase 1 in which the team assessed the predictive outcome of the PODT. Information including deviation type, surgical procedure, pre and postoperative strabismus angle for near and distance viewing, and PODT results for near and distance viewing was collected during phase 2. 

Both test-retest and interobserver reliability demonstrated high variability, according to the report. Among group 1 participants, 12 showed agreement in the area of suppression for near and distance fixation while 8 exhibited variability. Group 2 participants also demonstrated high variability in both near (n=9) and distance (n=10) fixation.  

Phase 2 analysis revealed that out of 14 participants with predicted diplopia following surgery, only 2 experienced persistent diplopia (3 months postoperatively). However, the predictive value for not experiencing diplopia showed reliability. Every patient predicted to not experience diplopia did not experience it. 

The study authors urge clinicians to use caution when assessing PODT results. “Both the orthoptist and ophthalmologist must be aware of the limited repeatability and predictive value of the PODT,” the researchers explain. “They should not base decisions solely on the results of a positive PODT result, because many patients could be inappropriately denied surgery.”

Study limitations include the use of convenience sampling, small sample size, and loss to follow-up.