Study On The Stereopsis Rebuilding Rule In Postoperative Intermittent Exotropia

Background and ObjectiveStrabismus is the main reason of stereopsis dysfunction. Since intermittent exotropiapatients have binocular vision in the development process, part of the postoperativestereopsis had natural recovery to normal levels with stereotest detection. Consideringthe current stereopsis test methods all have their shortcomings, the accuracy of theresults are questionable. A retrospective analysis was made on the78cases ofpostoperative intermittent exotropia patients to find the time course of near stereopsisrebuilding in intermittent exotropia after surgery. Also we evaluated the sensitivity ofTitmus Stereotest, Randot Stereotest and Frisby Stereotest in stereopsis detection inpostoperative, and is there any difference between the natural recovery stereopsis andnormal stereopsis is our purpose in this study.MethodsCollect78postoperative patients with intermittent exotropia from July2011to June2013. The stereoacuity was examined by Titmus Stereotest, Randot Stereotest andFrisby Stereotest preoperatively and half-month,1month,3months,1yearpostoperatively. The results were statistically analyzed. Selected20adult cases from the78cases which stereopsis returned to normal level as experimental group, randomlyselected20cases of healthy adults as normal control group, both groups were tested byMatlab program which written by Vision research lab to carry on the parallax sensitivitycurve testing. The results of the two groups were analyzed by ROC curve, evaluate thesensitivity of three stereotests according to the area under the ROC curve and the best cutoff value corresponding evaluation parameters. Accuracy between the two groupswas using t-Test, group under the different parallax within the accuracy were usingOne-Way ANOVA method, the mean between multiple pairwise comparisons wereusing LSD method.Results（1）Among the32patients who obtain the central fovea stereopsis,14cases (43.75%)returned to normal in half-month;12cases (37.50%) return to normal in half-month to1month;3cases (9.38%) return to normal in1month to3months;3cases (9.38%)return to normal in the3months to1year.（2）The average stereopsis of strabismus group was slightly lower than the normal.The area under ROC curves of the three methods (Titmus Stereotest, Randot Stereotest,Frisby Stereotest) were0.647±0.078,0.682±0.076and0.750±0.069, promptdiagnosis performance: Frisby Stereotest> Randot Stereotest> Titmus Stereotest.（3）Accuracy under all levels of parallax experimental group were lower than thecontrol group, the result was obvious different(P <0.05). The two group accuracybetween4,8,12,16,22,29-level parallax was not significant (P>0.05).Conclusion（1）The first month after surgery is the sensitive period of near stereopsis recovery,after which the recovery rate decreased significantly. The stereoacuity will be difficultto be completely recovered back to normal if not recovered in3months，then thefunctional vision training cannot be performed.（2）The results recommended to combine the three methods in stereopsis detection among the follow-up of postoperative strabismus to overcome the false positive resultcaused by monocular cue and false negative results caused by the separation ofbinocular vision.（3）Some of the stereopsis natural recovery still exist fusion dysfunction, andstereoacuity is lower than normal people under sophisticated parallax.