Human error in strabismus surgery: quantification with a sensitivity analysis.

Background: Reoperations are frequently necessary in strabismus surgery. The goal of this study was to analyze human-error related factors that introduce variability in the results of strabismus surgery in a systematic fashion.

Methods: We identified the primary factors that influence the outcome of strabismus surgery. For each of the human-error related factors we quantified variation with clinical assessments: measurement of the angle of strabismus, surgical strategy and surgical accuracy. Firstly, six patients were examined by six orthoptists, and accuracy of prism cover tests was assessed. Secondly, a questionnaire with sample cases (10 degrees, 15 degrees and 20 degrees of infantile esotropia) was put to orthoptists, to determine variation in current surgical strategy. Finally, photographs made during surgery were analyzed to assess surgical accuracy. The influence of human-error related factors was related to the influence of inter-patient differences with a mechanical model. The relative contribution of all factors was assessed with a sensitivity analysis, and results were compared to clinical studies.

Results: The surgical trajectory of strabismus surgery could be modeled mathematically. Measurement of angle of strabismus, surgical technique, anatomy and physiology were considered. Variations in the human-error related factors were: (1) the latent angle at distant fixation was measured with a 90% confidence interval of 5 degrees , and comprised 20% of the total variance of the postoperative angle, (2) orthoptists decided for bilateral recessions of, respectively, 7.3 +/- 1.7 mm (total amount of two recessions), 9.1 +/- 1.2 mm and 10.3 +/- 1.3 mm, which comprised 15% of the total variance, and (3) surgical accuracy was estimated at +/-0.5 mm, which comprised 20% of the total variance.

Conclusion: The human error in strabismus surgery could be quantified with a sensitivity analysis. Approximately half of the reoperations in strabismus surgery are caused by inaccuracy in the measurement of the angle of strabismus, variability in surgical strategy and imprecise surgery.