Many techniques for preventing LSI have been reported, such as appropriate correction rate of the MT curve [1, 21, 22] and selection of an appropriate UIV [9, 13, 14, 23]; however, this complication remains prevalent in scoliosis patients, reported in at least 25% of cases [1, 2], and how to achieve LSB postoperatively remains unclear [8].
Our study found that the change in T1 tilt after correction surgery was significantly related to the change in the CA, with correlation coefficients greater than 0.98, which suggests a close connection between T1 and the clavicle. Then, the ideal T1 tilt can be predicted before surgery to prevent LSI. However, in clinical work, we do not routinely fuse T1 but instead fuse T2 or below [9, 12]. Therefore, the question of the factors influencing T1 tilt should be speculated. First, the existing preoperative CA and UIV tilt need to be considered [15, 24]. Second, the flexibility between T1 and the UIV needs to be considered. If the selected UIV is T1, the preoperative CA can be directly subtracted from the preoperative T1 tilt, yielding the angle at which the postoperative UIV should be reserved. In the current retrospective study, the change in UIV tilt after correction surgery was significantly related to the change in the CA, while the correlation coefficients were smaller than those for T1, which may be related to the flexibility between UIV and T1 [15].
Traditionally, the ideal orthopedic effect is to correct scoliosis as much as possible and have a good overall balance. However, the goal of leveling the upper thoracic spine does not appear to guarantee clinically balanced shoulders or clavicles [8]. Therefore, how to place the UIV as flat as possible on the premise of ensuring shoulder balance requires good calculation and evaluation. We have found in practice that if the right side of the UIV is high before surgery, if we place the UIV as flat as possible, and the left side needs to be raised, we need to take into account the maximum compensatory ability between T1 and UIV to bend to the left; if the UIV is high on the left side before surgery, in order to put the UIV as flat as possible, the right side needs to be raised, we need to refer to the maximum compensatory ability between T1 and UIV to bend to the right. Then, the ideal UIV tilt angle can be precisely calculated before the operation by determining the preoperative UIV tilt and subtracting the reserve flexibility between the T1 and the UIV and the already existing CA, as described by the formula mentioned above.
The results of the correlation analysis between the difference in the actual and ideal postoperative UIV tilt values and the postoperative CA in the retrospective study confirmed that our speculation was correct. If the UIV is positioned to achieve the ideal postoperative UIV tilt, LSB should be achieved; otherwise, the difference between the actual postoperative UIV tilt and the ideal postoperative UIV tilt should be the value of the postoperative CA.
The current prospective study verified the feasibility of this method in clinical practice. Consistent with the retrospective study, the correlation analysis showed a significant correlation between the difference in the actual and ideal postoperative UIV tilt values and the postoperative CA, as well as between the actual postoperative UIV tilt and the ideal postoperative UIV tilt, which resulted in satisfactory postoperative LSB in all enrolled scoliosis patients. The significant correlation between the intraoperative UIV tilt and the ideal postoperative UIV tilt, as well as between the intraoperative UIV tilt and the actual postoperative UIV tilt, suggests that the improved crossbar method can control the intraoperative UIV tilt well during correction surgery.
The advantage of this method is that the ideal postoperative UIV tilt predicted by the formula is calculated when the maximum compensation is reached between the UIV and T1 vertebra, so the LSB problem is digitalized. The difference between the ideal postoperative UIV tilt and the actual postoperative UIV tilt will be directly reflected by the postoperative CA, so the operation can be accurately performed by adjusting the intraoperative UIV tilt to be consistent with the predicted ideal postoperative UIV tilt using the improved crossbar method to indirectly affect the postoperative T1 tilt, more reliably resulting in LSB after correction surgery.
Previous studies have shown that some of the patients with shoulder imbalance immediately after surgery had improvement in shoulder balance during the 2-year follow-up period, which may be associated with postoperative adding-on or trunk shift phenomenon [3, 25]. Although patients with LSI do not require revision surgery, LSI should be avoided because it may not only affect the patients’ appearance but may also aggravate the postoperative adding-on or trunk shift phenomenon during the follow-up period [3]. Therefore, it is necessary to optimize shoulder balance during correction maneuvers intraoperatively to prevent these problems, and using our method to achieve lateral shoulder balance immediately after surgery may be an option to avoid such problems.