The simplified forced torsion traction test
Guyton’s exaggerated traction test is considered invaluable for assessing the tightness of the oblique . Jung and Holmes proposed a new traction test in which the globe is maximally excyclorotated without retroplacement until the first instance of resistance is felt . The authors reported a median maximum excyclorotation of 62.5 degrees after SO disinsertion in six eyes. In the current simplified forced torsion traction test, the eyeball is grasped and retro-pulsed, thereby stretching the oblique muscles and adding slack to the rectus muscles. Compared with the reported SO forced traction test [3,4,6,12,14] used in SO-weakening procedures, the present test is simpler to perform. It is practical to evaluate torsion traction tests by the number of hours required in an operation, especially for inexperienced surgeons. Residual SOOA is likely the most common complication of SO-weakening procedures, which results in a positive exaggerated forced duction test result . If the tenectomy is completed, additional extorsion of one or one-and-a-half clock hours can be achieved; if not, the surgeon needs to inspect and verify whether the tendon fibres have been severed completely.
Minimal disturbance with visualizationunder a microscope
Both the complexity of the function and anatomical variation in the SO lead to challenges in isolating and manipulating the SO during the procedure [15,16]. Without visualization, SO tendon isolation can lead to injury of the vortex vein and orbital fat prolapse, a posterior Tenon’s capsule tear, iatrogenic ptosis, the SR being mistaken for the SO, and failure to hook the SO tendon . Excessively dissecting Tenon’s capsule overlying the tendon may result in postoperative restricted globe elevation and even acquired Brown syndrome . Parks and Helveston reported a preferred nasal-weakening procedure with a temporal conjunctival incision to keep the nasal intermuscular septum intact and reduce the risk for postoperative SO palsy or limited depression [15,18]. In the present study, the nasal approach was used, and a small incision with minimal disturbance was made under an operating microscope. Low-power magnification (4×) is required for SO tendon identification. Even on the downward rotated globe, the distance between the anterior border of the SO and the SR insertion varies and does not remain at 8 mm , which make it difficult to identify the SO. With a sufficient operating field and appropriate depth, this difficulty can be overcome by using different retractors with increasing lengths to 20 mm for individual patients. A sheath was detected to be enveloping the tendon (Figure 2d) in all cases, which is consistent with the findings of Helveston’s study . Blood vessels were noted along the sheath in some cases (Figure 2c)  and were preserved with intrasheath tenectomy under a microscope, which is inconsistent with the findings of previous reports . With a higher magnification, at 6× to 8×, a small incision in the intermuscular septum and sheath was sufficient to explore the tendon fibres while keeping the deeper intermuscular septum intact (Figure 2d; 2e), thereby avoiding the surrounding adhesion.
Intrasheath tenectomy at the nasal border of the SR
SO tenotomy can also be performed from the temporal side of the SR, as the SO is easy to isolate , but the weakening effect is smaller than that with nasal tenotomy [3,4]. Wei et al. reported that bilateral SO posterior tenectomy is effective for treating mild and moderate SOOA-associated A-pattern . The mean SOOA value decreased by 1.85, with a mean preoperative A-pattern deviation reduction of 12.75 PD, which were smaller than the values in the present study. The closer to the trochlea that tenotomy is performed, the more effective the procedure is [1,3,16], but the risk of iatrogenic SO palsy is higher . Furthermore, the ‘frenulum’ between the SO and SR can hinder the SO weakening effect of temporal disinsertion or suspension recession . Debert et al performed 6 mm bilateral SO transection with the complete width from its insertion without frenulum dissection . The mean A-pattern collapse value was 18 PD, and the mean preoperative A-pattern deviation was 21 PD, which was similar to our result. However, four patients had postoperative vertical deviation. Heo et al performed 10 mm SO posterior tenectomy with dissection of the frenulum to the extent possible . The mean A-pattern correction was 17.63 PD, which was similar to that in our study. However, 5 of 75 patients showed mild inferior oblique overaction. Parks and colleagues performed intrasheath tenotomy or tenectomy to achieve a sufficient weakening effect [15,17]. Helveston  also proposed that the fascia around the SO tendon should be left undisturbed to achieve more predictable results. Berke  performed intrasheath tenotomy or tenectomy of the SO tendon near the nasal border of the SR, where the sheath wrapping the tendon has many delicate areolar fibrillae. In the current study, MSOIT at the nasal border of the SR was performed for 130 eyes in 66 A-pattern patients with SOOA who were followed up for a mean of 33.45 months. Mild inferior oblique overaction occurred in only one eye. This treatment improved the mean A-pattern deviation by 19.65 PD, and the mean SOOA value improved by 2.95 (Table 1). The magnitude of correction of A-pattern was significantly correlated with the preoperative severity of A-pattern (Figure 3). The SO tendon was weakened with minimal sheath disturbance and the frenulum left intact. The preserved sheath was still connected to the two cut tendon edges in its original course, causing minimal disturbance to the fascia and allowing connections to form. Due to the temporal section and insertion being left intact, nasal transection resolved the case of SOOA, preserved important function and avoided SO palsy .