The rapid correction of severe scoliosis complicated with respiratory impairment can increase the risk of neurologic injury, morbidity, and mortality [10]. Halo traction has been used as an adjunctive method in the treatment of severe spinal deformities, which can reduce the risk of spinal cord injury while effectively obtaining correction of severe spinal deformities in a controlled and safe manner. As a result, less correction is needed during surgery, and aggressive procedures, such as vertebral column resection, might be avoided. Finally, this approach can also help improve pre-operative pulmonary function, allowing a better tolerance for more aggressive procedures [5, 6, 11]. Unlike HGT, HPT can provide powerful traction forces throughout the day to effectively correct spinal deformities. However, traditional HPT has a number of various complications and poor tolerance. As such, the use of HPT has gradually declined [12]. In this study, the pelvic ring of HPT device was a half-ring, and the rods were all placed on the anterolateral side of the truck (Fig. 1). This modified method allowed the patients to sleep in a supine position, wear clothes, and move by themselves while still achieving 24 hours of continuous traction. The gradual traction applied during the pre-operative period might also help in evaluating the neurologic function and estimating the amount of correction that can safely be obtained.
O'Brien et al. first used HPT in the treatment of scoliosis in the 1970s [6], but there are few subsequent literature reports on the use of HPT for the treatment of severe spinal deformities, especially over the past 10 years. In this current study, with the help of HPT, the correction rate after HPT was 45.15% for scoliosis and 47.59% for kyphosis. This outcome was superior to the pre-operative HGT documented in prior reports, which demonstrated 15–38% correction in scoliosis and 17–35% correction in kyphosis [3–5, 11]. In the study by Janus et al., the correction of pre-operative HGT was similar to the final correction after surgery when compared with the initial curve (32% vs. 36% for scoliosis, 24% vs. 27% for kyphosis), but the total correction was significantly lower than that observed in this current study (36% vs. 59.81% for scoliosis, 27% vs. 53.67% for kyphosis) [13]. The correction of pre-operative HGT was similar to Janus in the studies by Koller et al. [9, 14]. Xia et al. reported a significant correction of spinal deformity conferred by pre-operative HGT in conjunction with posterior fusion; however, the correction conferred by pre-operative HGT in this prior study was lower (18.4–23%) [15], which confirms the effectiveness of our modified pre-operative HPT technique for improving severe spinal deformities prior to surgery due to the strong traction force and the subsequently reduced difficulty of the second-stage orthopedic corrective surgery. Moreover, the sagittal correction conferred by pre-operative HPT was superior to the coronal correction (45.15% for scoliosis, 47.59% for kyphosis), which was dissimilar to that reported for HGT [9, 13–15]. This difference might result because the force of HPT was primarily in front of the body in this study, or it might be due to the selection of patients with severe kyphosis. In this study, the sagittal imbalance of the patients became worse, with the center of the C7 vertebral body positioned further behind the sacrum after HPT traction. However, after surgery, the patients obtained sagittal balance. Thus, according to these results, the use of pre-operative HPT of severe spinal deformities had no measurable negative impact on sagittal balance.
Severe spine deformities can reduce the compliance of the respiratory system by affecting the thoracic cage as well as muscular and diaphragmatic function [16, 17]. When Cobb > 100°, respiratory system compliance is decreased to levels comparable to adult respiratory distress syndrome [8]. Therefore, halo traction is recommended to improve the poor respiratory function and to decrease complications in patients with severe spine deformities [9, 11, 18]. However, there are few reports regarding the effect of pre-operative HPT on the respiratory function in patients with severe spine deformities. In our sample, all of the patients had moderate or severe pulmonary impairment prior to traction, and significant improvements in the PFT results were achieved after HPT. These results indicated an increase of 10.03 ± 9.60% in the FVC%, which was similar to previously reported data on pediatric and adolescent populations. The mean change in FVC% after pre-operative HGT ranged from 10–14% [5, 9, 19]. This also confirmed that our pre-operative HPT could significantly improve pulmonary function in patients with severe spine deformities. However, it would still be necessary to use some pulmonary exercises to help improve muscle endurance and strength in these patients [11, 20]. We used feedback-breathing exercises in our patients to help improve endurance and muscle strength.
The length of traction depended on the response of the various curves to the traction and the patient’s systemic, respiratory, and nutritional conditions. Currently, the duration of HGT varies from 2 to 12 weeks, as based on reports in previous studies [5]. However, there is no consensus on the optimal duration of traction for HPT, and there are few studies reporting the relationship between the duration of traction and the correction of HPT. In this current study, the major correction of HPT on scoliosis and kyphosis was obtained in the first 2 weeks, and the correction plateaued at nearly 50% after approximately 4–6 weeks after a rapid period of initial correction. In regard to pulmonary function, Koller et al. reported that prolonged traction might not help further improve the pulmonary function [9]. For optimal correction and reduction in surgical risk, we recommend HPT traction for at least 4 weeks, and appropriately prolonged period of traction according to the patient’s condition. However, the optimal length of traction remains to be determined and requires additional studies.
O'Brien et al. previously described the complications associated with HPT in detail. These included perforation of the intestine during insertion of the pelvic pins, infection of the pelvic pins, injury to the cervical spine, both acute and degenerative nerve injuries, and paraplegia [6]. Different from the HPT method used in the study by O'Brien et al., the pelvic pins used in the our HPT did not need to be drill through the iliac crest (Fig. 5), and there was no perforation of the intestine or injuries of other pelvic organs observed in this study. Strict protocols were used, including daily pin checks and prudent hygiene, and there were no other complications related to the pins, including infection of the pins or pin loosening. As for injury to the cervical spine, O'Brien et al. reported that more than 50% of patients showed some degeneration of the cervical spine when the traction was prolonged over 3 months [6]. Tredwell and O'Brien noted that the incidence of apophyseal joint degeneration was 47.4% in patients with HPT occurring over 9.47 months [21]. In this current study, the average time of traction was only 5.37 weeks, which was substantially shorter than that presented in O'Brien et al. or Tredwel and O'Brien. Additionally, significant degeneration of the cervical spine was not observed in our study; however, our patients often complained of cervical discomfort and trapezial soreness, but these symptoms disappeared after surgery. Neurologic complications, such as cranial nerve injuries and paraplegia, are known to occur during traction, and warning signs must be monitored daily. In our study, two patients (6.7%) developed neurologic complications after HPT traction, thought these occurrences returned to normal after reducing the traction. Moreover, due to the lack of immobilization of the patients receiving the modified HPT, any occurrence of osteoporosis could be avoided [22, 23].
To our knowledge, this was the first study to report on the use of this modified HPT, in which the pelvic ring was changed to a half-ring and the rods were all placed on the anterolateral side of the truck. All of these changes were made to improve patient tolerance and comfort while ensuring traction strength and effectiveness. However, this study has several limitations. The study was a retrospective review of a single cohort without comparison to a control group to confirm the efficacy of the HPT since the standard protocol in our center is surgical treatment after HPT in severe scoliosis cases. Secondly, this study is missing the post-operative data regarding PFT and the results of 1 year and 2 year follow-ups, as the acquisition of post-operative PFT was not a regular protocol and some patients did not achieve a 2 year follow-up. In the future, the results of 1 year and 2 years follow-ups will be assessed to complete the post-operative data set.