Strabismus Surgery: Do Millimeters Matter?

Who hasn’t longingly hoped for a dose–response curve or table showing prism diopters/millimeter (PD/mm) that would solve all the problems of strabismus surgery? Thinking back to the early years of my career, the first question I asked of skilled surgeons was, “What surgical table do you use?” Subsequently, I realized there were different tables with different surgical doses advocated by different experts. Despite the sometimes wide variation in the tables, the reported results from the different tables were surprisingly similar. But even if I strictly relied on any given table, the surgical outcomes were inconsistent. My initial discouragement prompted me to try to understand why there was so much variability. This raised the intriguing question, “Do numbers really matter?” In 2002, during my fellowship in Charleston, SC, USA, Dr. Ed Wilson told me that during a meeting he had attended, Mexican colleagues referred to American strabismus surgeons as “millimeter surgeons” (Ed Wilson personal communication). Apparently, not all strabismus surgeons rely as heavily on surgical formulas as those in the USA. After hearing that, the question, “Do numbers really matter?” really obsessed me! In 1951, Scobee wrote, “The amount of surgical correction obtained in a patient with esotropia is usually directly proportional to the deviation present before surgery and is not particularly related to the amount of surgery performed as measured in millimeters.” He divided operations into three classes: Class 1: esotropia <50PD, for which he performed one medial rectus recession of 6 mm; Class 2: esotropia >50PD, for which he performed bilateral medial rectus recession of 6 mm; and Class 3: if amblyopia was present, for which he performed unilateral surgery consisting of a medial rectus recession of 6 mm plus a lateral rectus resection of 6–7 mm. He reported that if one patient has 15° of esotropia and another one has 30° of esotropia, surgery will result in 15° and 30° of correction, respectively, even if they receive the same amount of surgery. In 1989, Kushner and coauthors reported that axial length played a role in the response to surgery. They found a significant correlation between the response to strabismus surgery and axial length in patients with esotropia, but the correlation coefficient they found suggested that only 28% of the variance in the response to strabismus surgery can be attributed to axial length. Thus, a surgical formula that takes axial length into account should decrease the variability in response to strabismus surgery, but will not eliminate it. Poor correlation was found between axial length and response to surgery in exotropic patients. However, in 1993 Kushner subsequently found that the response to strabismus surgery (PD of change per millimeter of rectus muscle recession) correlated significantly with the preoperative deviation for esotropic and exotropic patients and that axial length, age, or refractive error were not significant independent predictors. Over a 40-year period, Kushner kept a record of masked measurements obtained at surgery in about 10,000 patients. He noted and recorded the amount of recession actually performed, the planned amount of recession based on his preferred surgical table, and the preand post-operative deviation. He found that the mean response at 1 week after surgery for esotropes was 2.8 PD more in patients in whom the actual recession performed totaled 1.0 mm more than what was planned and 2.7 PD less if the actual surgery recession performed totaled 1.0 mm less than what was planned. For intermittent exotropes, the results were similar. For those patients in whom the actual surgery performed differed from the planned amount by a total of 1.5 mm, the discrepancy was 4.7 PD for esotropes and 5.1 PD for intermittent exotropes. These differences were all significant (P < .05). He conceded that an assessment made at 1 week after surgery does not foretell the final result, but it does reflect the immediate response to surgery before it is contaminated by sensory adaptations. This suggests we should be meticulous with respect to measurements of the pre-operative deviation and the amount of surgery we perform. This leads to a conundrum. On one hand, we have Scobee’s suggestion that millimeters do not matter in determining the response to strabismus surgery. On the other hand, we have Kushner’s observations that, in fact, small changes in the amount of rectus muscle recession have a measurable impact on the response. How could these opposing concepts both be correct? In 2018, in his Costenbader lecture, Steven Archer wrote that the size of the preoperative angle is more important than the amount of surgery performed in predicting the JOURNAL OF BINOCULAR VISION AND OCULAR MOTILITY 2023, VOL. 73, NO. 3, 59–60 https://doi.org/10.1080/2576117X.2023.2208019