Ponte osteotomies increase risk of intraoperative neuromonitoring alerts in adolescent idiopathic scoliosis surgery

Clinical analysis

Eighty cases were included in the study. The mean patient age was 16.4 ± 3.08 years [Table 1]. Most patients were female (82.5%). The average thoracic Cobb angle was 67.8 ± 17.2°. On bending film, the thoracic Cobb angle was 47.4 ± 16.8. The overall postoperative coronal correction rate was 67.6 ± 11.74. The thoracic kyphosis (T5–T12) was 32.6 ± 14.4. Curve distributions according to the Lenke classification were Type 1 (36.3%), Type 2 (30%), Type 3 (15%), Type 4 (11.3%), and Type 6 (7.1%). Patients’ demographics and spinal deformity characteristics (i.e., including the coronal standing parameters and supine bending Cobb angles for the main thoracic [MT] curve and the thoracic kyphosis [T5–T12]) were studied before and after surgery. Cases were classified using the Lenke classification system. The correction index was calculated using the formula 1-postoperative Cobb/preoperative Cobb. Kyphosis change was calculated using the formula postoperative kyphosis Cobb – preoperative kyphosis Cobb.

Table 1:: Demographic characteristics of 80 subjects.

Out of 80 patients, half had PO. There were no statically significant differences between the two groups in age, sex, Lenke type, preoperative thoracic Cobb, postoperative coronal thoracic Cobb, and thoracic coronal Cobb in the bending view [Table 2].

Table 2:: Comparison of all factors between OSTEO (PO vs. No-PO).

Intraoperative neural monitoring (IOM)

An IOM alert was considered significant, if there was a decrease of more than 50% of the motor or somatosensory signal. IOM alerts related to anesthesia, systemic causes, and technical neurophysiologic causes were not counted. Only cases that did not improve after the above-mentioned causes were counted.

Summary of surgical technique

All procedures were performed on a Jackson table using IONM Bilateral pedicle screws that were placed at all levels using a free hand technique.[7] PO were performed (i.e., from Cobb to Cobb of the MT curve and 4–6 osteotomies are usually required per case). Deformity correction was obtained using 5.5 mm cobalt chromium rods; over-bent on the concave side and under-bent on the convex side. After the first concave rod insertion, distraction from the apex to the end was done and repeated 2–3 times to gain kyphosis. Then, the convex rod was inserted using the cantilevering technique.

Statistical analysis

The two groups with and without PO were compared using the Student’s t-test and Fisher’s exact test, and results were confirmed by the nonparametric Mann–Whitney U-test. Data analysis was done using SPSS version 25.

Advantages of PO accompanying AIS surgery

The 40 patients undergoing PO procedures with AIS surgery demonstrated a higher overall correction rate, a lesser degree of average change in postoperative kyphosis. However, five patients having PO during AIS surgery (12.5%) demonstrated significant intraoperative IOM changes versus none observed, where no PO were performed. Nevertheless, none of the five patients with such IOM changes during their AIS procedures ended up with new postoperative neurological deficits. [Figure 1] showed when a signal alert was considered surgical procedure. Signal alert improved in all patients after reversal of the last surgical maneuver, that is, removal of rod would reverse the signal. Concave rod reinsertion with distraction avoidance was enough to keep the signal (two cases). If this strategy fail, convex rod first insertion is helpful to spare the signal (one case). Otherwise, the case staged while using the temporary rod waiting for 1 week to optimize the patient. The surgery was finished using the same surgical plan without signal alert (two cases) [Figure 1].

Close

Figure 1:: Protocol for the management of surgical signal alert.

Export to PPT

Complication rates for PO used in AIS surgery

Although PO help in the 3D correction of the AIS deformity,[4] serious, complications may happen due to an increase in corrective forces.[8] Shah reported IOM alerts of 8% in a study of 87 patients with a thoracic Cobb angle of 57 ± 9.7 in Lenke types 1–4.[7] In the present study, 80 cases showed that the average thoracic Cobb angle was 67.8 ± 17.2° with the Lenke classification, which were Type 1, 2, 3, 4, and 6. There were no statically significant differences between the two groups in age, sex, Lenke type, preoperative thoracic Cobb, postoperative coronal thoracic Cobb, and thoracic coronal Cobb in the bending view [Table 2]. In a multicenter study of 2210 surgical AIS patients, the incidence of IOM alerts was 9.3% in patients who underwent PO and 4.2% in patients without PO.[3] Our comparative study notes a major difference in IOM alerts between the two groups (PO: 12.5% vs. No-PO: 0%, P = 0.021). Our overall coronal Cobb deformity was 67.8 ± 17.2°; this difference in Cobb angle may explain the higher rate of IOM alerts in our PO group.

Higher correction rate in PO patients undergoing AIS surgery

In a multicenter study of 191 subjects comparing PO and NoPO subjects, Samadani et al. 2015 showed a higher correction rate in the PO group (67.1% vs. 61.8%, P = 0.01). [6] Our patients’ correction rate was 71% ± 10.9 for the PO group and 64.2% ± 11.5 for the No-PO group (P = 0.008). This better correction rate in our patients may be related to our all-levels screw fixation system.

Better kyphosis outcomes with PO in AIS surgery

The kyphosis outcome was significantly better in the PO group (25.2 ± 6 vs. No-PO: 17.5 ± 9.4, P = 0.0001). We had a kyphosis decrease in both groups, which was more obvious in the PO group, where the kyphosis change was higher (-3.8 ± 11.6 vs. No-PO: -18.6 ± 10.1, P = 0.0001). This study showed the importance of good intraoperative neurophysiology monitoring during spinal deformity surgeries, especially when three-dimensional correction of the deformity is the goal.