J L Mims1, R C Wood. 1. Department of Ophthalmology, University of Texas Health Science Center, Christus Santa Rosa Children's Hospital, San Antonio, Texas, USA.
Abstract
PROBLEM: A series of overcorrections after this surgical sequence. SUBJECTS & METHODS: Forward stepwise multiple regression analysis of the effect of bilateral lateral rectus resections, retrieved and tabulated retrospectively from the clinical charts of 50 children with infantile esotropia who had significant residual non- accommodative esotropia after bilateral medial rectus recessions. Analysis allowed for selection among linear and quadratic forms of two relevant predictor variables: the mm of previous bilateral medial rectus recession and the prism diopters of residual esotropia. RESULTS: We found a correlation between the effect of the second surgery (bilateral LR resections) and the two variables. This relation could be expressed by a surgical dose-response schedule-surface with the formula: [2.39 + 0.26 (mm MR recession) Squared + 0.41 (mm LR Resection) Squared = prism diopters of effect]. A corresponding schedule (table) with these two parameters was developed to facilitate clinical application. To test and prove this schedule, we re-studied 19 of the 50 original cases who had had surgery, fortuitously, +/-0.3 mm of our new surgical dose- response surface-schedule plus 10 new cases performed according to the new surface-schedule. For these 29 cases there was only one "failure" at 6 months postop' (3%). As a control, for the 31 cases of the original 50 whose surgery had not been within +/- 0.3 mm of the new schedule, 7/31 (23%) were "failures", all overcorrections. ("Failure" vs. "success" - there were only two categories of outcome - was defined as an esotropia over 8 PD or exophoria over 6 PD.) CONCLUSION: This difference in results, expressed as "successes" of 28/29 vs 24/31, or 97% vs 77% gave, on statistical analysis, a p=0. 053 of rejecting the null hypothesis, which is considered a valid "trend" or marginally "statistically significant" and definitely "clinically-medically significant". We have adopted the regular application of our new surgical dose- response schedule in these cases and recommend it for all such cases. binocular vision; complications, surgical; dose-response surgical schedule; esotropia, congenital/infantile; lateral rectus muscle, resection; outcomes; overcorrections; strabismus surgery; study, retrospective, clinical; surgery, strabismus.
PROBLEM: A series of overcorrections after this surgical sequence. SUBJECTS & METHODS: Forward stepwise multiple regression analysis of the effect of bilateral lateral rectus resections, retrieved and tabulated retrospectively from the clinical charts of 50 children with infantile esotropia who had significant residual non- accommodative esotropia after bilateral medial rectus recessions. Analysis allowed for selection among linear and quadratic forms of two relevant predictor variables: the mm of previous bilateral medial rectus recession and the prism diopters of residual esotropia. RESULTS: We found a correlation between the effect of the second surgery (bilateral LR resections) and the two variables. This relation could be expressed by a surgical dose-response schedule-surface with the formula: [2.39 + 0.26 (mm MR recession) Squared + 0.41 (mm LR Resection) Squared = prism diopters of effect]. A corresponding schedule (table) with these two parameters was developed to facilitate clinical application. To test and prove this schedule, we re-studied 19 of the 50 original cases who had had surgery, fortuitously, +/-0.3 mm of our new surgical dose- response surface-schedule plus 10 new cases performed according to the new surface-schedule. For these 29 cases there was only one "failure" at 6 months postop' (3%). As a control, for the 31 cases of the original 50 whose surgery had not been within +/- 0.3 mm of the new schedule, 7/31 (23%) were "failures", all overcorrections. ("Failure" vs. "success" - there were only two categories of outcome - was defined as an esotropia over 8 PD or exophoria over 6 PD.) CONCLUSION: This difference in results, expressed as "successes" of 28/29 vs 24/31, or 97% vs 77% gave, on statistical analysis, a p=0. 053 of rejecting the null hypothesis, which is considered a valid "trend" or marginally "statistically significant" and definitely "clinically-medically significant". We have adopted the regular application of our new surgical dose- response schedule in these cases and recommend it for all such cases. binocular vision; complications, surgical; dose-response surgical schedule; esotropia, congenital/infantile; lateral rectus muscle, resection; outcomes; overcorrections; strabismus surgery; study, retrospective, clinical; surgery, strabismus.