Abstract
Study Design
Retrospective cohort study.
Objective
This study aimed to determine whether the outcomes of posterior correction and fusion surgery (PSF) in patients with residual adolescent idiopathic scoliosis (AIS) during adulthood (Lenke type 5 AdIS) are inferior to those in patients with AIS treated during adolescence.
Methods
We retrospectively evaluated 48 patients with type 5 AdIS who underwent PSF after 20 years old. As controls, we selected 67 patients with type 5 AIS who underwent PSF before 18 years old. Both groups were matched based on the preoperative thoracic and thoracolumbar/lumbar Cobb angles, resulting in 33 patients in each group. We compared preoperative and postoperative radiographic parameters and SRS-22 scores between the 2 groups.
Results
Preoperatively, the AdIS group had significantly lower flexibility index (59.0% ± 13.0% vs 66.6% ± 16.7%; P = 0.01). The preoperative SRS-22 scores for pain and mental health domain were significantly worse in the AdIS group (pain, 3.9 ± 0.8 vs 4.3 ± 0.7, P < 0.01; mental health, 3.6 ± 0.8 vs 4.2 ± 0.6, P < 0.01). The AdIS group had significantly more fused vertebrae and longer intraoperative time than the AIS group. Postoperatively, lumber Cobb angle (16.1° ± 5.6° vs 8.6° ± 4.0°, P < 0.01) and correction rate were both less favorable in the AdIS group. Both pain and mental health scores remained significantly worse in the AdIS group (pain: AdIS, 4.3 ± 0.6 vs AIS, 4.7 ± 0.3; P < 0.01; mental health: AdIS, 4.1 ± 0.7 vs AIS, 4.5 ± 0.5; P = 0.02).
Conclusion
Patients with type 5 AdIS had a similar correction of the major curve as those with AIS; however, the surgery was more invasive, with inferior postoperative clinical outcomes, than those with AIS. These findings suggest that the surgical timing should be carefully considered in patients with type 5 AIS to optimize outcomes.
Keywords
Introduction
Adolescent idiopathic scoliosis (AIS) is a common pediatric spinal deformity, and its treatment is approximately determined based on the magnitude, position, and extent of the curve. Patients with mild or moderate AIS are often treated conservatively, whereas those with a Cobb angle > 40° are candidates for corrective surgery.1,2
Curves > 40° tend to progress at a rate of approximately 0.4° to 0.5° per year.3,4 As these patients reach adulthood, symptoms such as low back pain, reduced physical function, psychological distress, and cosmetic concerns often worsen despite completion of skeletal growth.5-7 Consequently, in adult patients with residual AIS (AdIS), even those with thoracolumbar/lumbar curves below 50°, clinically significant back pain and degenerative changes can develop by early adulthood, ultimately necessitating surgical intervention.
Several studies have examined the surgical outcomes of AdIS.8-10 Lonner et al reported that surgical treatment in patients with AdIS was associated with more extensive fusion levels, longer operative times, and higher complication rates than those in patients with AIS. 9 Takahashi et al found that although surgical treatment in patients with AdIS significantly reduces pain and corrects the main Cobb angle, the degree of correction decreases with increasing age. 8 However, these studies either analyzed all Lenke types collectively or focused on thoracic curves such as Lenke types 1 or 2. Few studies have specifically investigated the outcomes of Lenke type 5 AIS in adults (type 5 AdIS).
This gap is important because patients with residual lumbar AIS curves tend to experience greater low back pain and disc degeneration than those with residual thoracic curves. Watanabe et al observed that patients with AdIS who had thoracolumbar curves exhibited significantly worse visual analog scale scores for low back pain compared to those with main thoracic curves. 3 Additionally, Glassman et al reported that the pain subdomain of the SRS-22 score was significantly lower in patients with AdIS with lumbar curves than in those with thoracic curves. 11 Furthermore, Suzuki et al showed that residual lumbar curvature from AIS may accelerate intervertebral disc degeneration in adulthood more than thoracic curves. 12 These findings suggest that untreated lumber curves, even of moderate magnitude, may predispose patients to early degenerative changes and chronic symptoms in their twenties and thirties.
Based on these findings, we hypothesized that the surgical outcomes for patients with type 5 AdIS may be inferior to those treated for the same curve type during adolescence. This study aimed to compare the radiographic findings and clinical outcomes of patients with Lenke type 5 AIS treated surgically in young adulthood with those treated during adolescence to test this hypothesis.
Study Design and Population
We conducted a retrospective cohort study using data from our institution. A total of 48 patients with type 5 AdIS with a mean age of 27.0 years (range, 20 - 40 years) were included in the AdIS group. All patients diagnosed with AIS between the ages of 10 and 19 years who underwent posterior correction and fusion surgery (PSF) after 20 years old, between 2013 and 2023, and were followed up for a minimum of 2 years were enrolled. Patients with congenital vertebral anomalies, neuromuscular conditions, or a history of spinal surgery were excluded from the study.
As controls, we selected 67 patients with Lenke type 5 AIS (3 males and 54 females; mean age, 15.4 years; range, 12-19 years) who underwent PSF between the ages of 12 and 18 years, also between 2013 and 2023, and were followed up for a minimum of 2 years (AIS group). Both the AdIS and AIS groups were matched based on preoperative thoracic and lumbar Cobb angles. A one-to-one matching analysis was conducted between the AdIS and AIS groups using the nearest-neighbor method within a caliper width of 0.25 standard deviations of the logit of the propensity score, resulting in 33 patients being assigned to each group. Then, we compared the preoperative and postoperative radiographic parameters, self-administered questionnaires, and surgical outcomes between the 2 groups.
Surgical Procedure
All of the patients underwent PSF following a short fusion strategy throughout the study period.12,13 Briefly, after segmentally placing the pedicle screws, the scoliosis was corrected using a rod rotation maneuver, which simultaneously corrected the scoliosis and restored lumbar lordosis on the convex side. An underbent concave rod was then placed to derotate the vertebral bodies, and the lower instrumented vertebra (LIV) was placed horizontally. The choice of LIV was determined preoperatively by the surgeon considering the level of the lower-end vertebra, L3/4 lateral translation, and lumbar disc degeneration on magnetic resonance imaging (MRI), particularly at L3/4.
Radiographic Measurements and Intraoperative Findings
Radiographic evaluations were performed using whole-spine standing posteroanterior, lateral, and supine side-bending radiographs. Coronal parameters included measurements of the minor thoracic and main thoracolumbar/lumbar Cobb angle, supine side-bending angle, and L3 and L4 tilt angles. The C7–central sacral vertical line (CSVL) distance was used to assess global coronal balance. The L3 and L4 tilts were defined as the angles between the inferior endplate and the horizontal line. The Flexibility Index was determined by subtracting the side-bending angle from the Cobb angle, dividing the difference by the Cobb angle, and then multiplying the quotient by 100. The sagittal parameters included T2-T5 kyphosis, T5-T12 kyphosis,T10-L2 kyphosis, lumbar lordosis (T12-L5), lower lumber lordosis(L4-S1), pelvic tilt, pelvic incidence, sacral slope, and sagittal vertical axis. The PI–LL mismatch was calculated as the difference between pelvic incidence and lumbar lordosis.
Intraoperative findings, such as operative time, blood loss, number of fused vertebrae, and days of hospitalization, were recorded. The location of the LIV was also noted.
MRI Evaluation
Preoperative MRI assessments were performed using a 1.5-T superconducting imager (Signa Excite HD 1.5T; GE Healthcare, Milwaukee, WI, USA). A spine surgeon (R.S.) performed the evaluations, focusing on T2-weighted sagittal images and using the Pfirrmann grading scale 13 to assess intervertebral disc degeneration.
Clinical Outcomes
The clinical outcomes were assessed using the SRS-22 questionnaire, which was administered preoperatively and at the final follow-up. The results for the 5 domains of the SRS-22 (function, pain, self-image, mental health, and satisfaction) were compared between the groups.
Statistical Analyses
Continuous, discrete, and rating variables were presented as mean ± standard deviation, and categorical variables were expressed as frequency or percentage. Comparisons of each independent variable were performed using Student’s t-test or analysis of variance for continuous variables, chi-square test for categorical variables, and Mann-Whitney U test for ordinal variables. Statistical significance was set at P < 0.05. All statistical analyses were performed using SPSS version 24.0 (IBM Corp., Armonk, NY, USA).
Ethical Approval
This study was approved by the Committee on Ethics and Institutional Review Board (approval number: 20090042), and all participants provided informed consent for inclusion before treatment.
Results
Demographic Characteristics
Clinical Characteristics of all Subjects
Values indicate mean ± standard deviation.
Bold values indicate P < 0.05.
BMI indicates body mass index.
AdIS indicates adolescent idiopathic scoliosis treated in adult.
AIS indicates adolescent idiopathic scoliosis.
Preoperative Radiographic Parameters
Comparison of Preoperative and Postoperative Radiographic Findings and Pfirrmann Grade Between AdIS Group and AIS Group
Values indicate mean ± standard deviation.
Bold values indicate P < 0.05.
PI indicates pelvic index.
LL indicates lumber lordosis.
SVA indicates sagittal vertical axis.
CSVL indicates central sacrum vertical line.
AdIS indicates adolescent idiopathic scoliosis treated in adult.
AIS indicates adolescent idiopathic scoliosis.
Postoperative Radiographic Findings
Comparison of Preoperative and Postoperative SRS-22 Score Between AdIS Group and AIS Group
Values indicate mean ± standard deviation.
Bold values indicate P < 0.05.
Δ represents the difference between preoperative and postoperative SRS-22 scores.
Preoperative MRI Findings
As shown in Table 2, the preoperative MRI findings showed significantly more severe lumbar disc degeneration at L3/4 and 4/5 in the AdIS group than in the AIS group (L3/4, 2.3 ± 0.7 vs 1.8 ± 0.4; P < 0.01; L4/5, 2.4 ± 0.7 vs 1.8 ± 0.4; P < 0.01). Meanwhile, there were no significant difference in the L1/2, L2/3 and L5/S1 ratio between the 2 groups.
Preoperative SRS-22 Score
A comparison of the preoperative subdomain scores of SRS-22 is presented in Table 3. The AdIS group had significantly lower scores in the pain and mental-health subdomains than the AIS group (pain, 3.9 ± 0.8 vs 4.3 ± 0.7; P < 0.01; mental health, 3.6 ± 0.8 vs 4.2 ± 0.6; P < 0.01). No statistically significant differences were observed in the function or self-image subdomains between the 2 groups.
Postoperative SRS-22 Score
Postoperatively, the AdIS group had significantly lower scores in the pain and mental health subdomains of the SRS-22 than the AIS group (pain, 4.3 ± 0.6 vs 4.7 ± 0.3; P < 0.01; mental health, 4.1 ± 0.7 vs 4.5 ± 0.5; P = 0.02). No significant differences were found in the other SRS-22 subdomains between the 2 groups.
Preoperative and Postoperative Changes in SRS-22 Scores
When comparing the changes in SRS-22 scores from the preoperative to postoperative assessments, no statistically significant differences were found between the AdIS and AIS groups across all SRS-22 subdomains.
Surgery-Related Factors
Factors Related Surgery
Values indicate mean ± standard deviation.
Bold values indicate P < 0.05.
LIV indicates lowest instrumented vertebrae.
Representative Cases
A 29-year-old woman in the AdIS group presented with a preoperative lumbar Cobb angle of 45° and bending angle of 22° (Figure 1). Her preoperative SRS-22 scores were as follows: function, 4.8; pain, 4.0; self-image, 4.6; mental health, 3.2; satisfaction, 4.5. Radiographic evaluation revealed a thoracic kyphosis of 12° and lumbar lordosis of 63° prior to surgery. The patient underwent posterior correction and fusion surgery from T9 to L3, resulting in correction of the major curve to a Cobb angle of 10° at the final follow-up. Postoperatively, her SRS-22 scores improved as follows: function, 5.0; pain, 4.6; self-image, 4.4; mental health, 3.8; satisfaction, 5.0. Radiographs obtained after surgery showed a thoracic kyphosis of 13° and lumbar lordosis of 57°. Pre- and postoperative radiographs of Lenke type 5 AdIS patient A 29-year-old woman in the AdIS group A. (A) lumbar Cobb angle of 45° and a bending angle was 22° before surgery. The preoperative SRS-22 scores were as follows: Function 4.8, Pain 4, Self-image 4.6, Mental-health 3.2, Satisfaction 4.5. (B) The thoracic kyphosis was 12° and the lumber lordosis was 63° before surgery. (C) Posterior correction and fusion surgery at T9-L3 resulted in the Cobb angle of the major curve being corrected to 10° at the final follow-up. The postoperative SRS-22 scores were as follows: Function 5, Pain 4.6, Self-image 4.4, Mental-health 3.8, Satisfaction 5. (D) The thoracic kyphosis was 13° and the lumber lordosis was 57°
Discussion
This study compared the radiographic findings and clinical outcomes of patients with Lenke type 5 AIS treated surgically in young adult with those treated during adolescence. The results showed that the flexibility index and lumbar disc degeneration were significantly worse in the AdIS group than in the AIS group, despite having similar and relatively moderate curve magnitudes preoperatively. Additionally, the AdIS group had significantly more fused vertebrae and lower LIV than the AIS group. Both preoperatively and postoperatively, the pain and mental health domains of the SRS-22 were significantly worse in the AdIS group.
Preoperative Radiographic Findings
Despite focusing on a relatively young population (≤40 years) in the present study, patients with AdIS had significantly lower flexibility than patients with AIS, even with comparable lumbar curve magnitudes. This is consistent with the findings of a previous study, 14 showing that reduced flexibility was associated with age and curve size and indicating that stiffness can occur even in younger patients with AdIS.
Sagittal parameters, such as pelvic tilt, sagittal vertical axis, regional thoracic kyphosis (T2–T5, T5–T12, T10–L2), lumber lordosis, L4-S1 lordosis and spinopelvic measures (pelvic tilt, sacral slope, and pelvic incidence–lumbar lordosis mismatch) were similar between the 2 groups.
However, prior studies have documented age-related sagittal deterioration in AdIS patients, particularly in those with thoracolumbar/lumbar curves, with older adult cohorts showing worse sagittal imbalance than adolescents.15,16 The absence of significant sagittal differences in our relatively young cohort suggests that major spinopelvic compensatory changes and degeneration had not yet fully manifested by the third decade of life.
Postoperative Radiographic Findings
Pan et al. reported significantly lower correction rates in patients with AdIS with thoracolumbar/lumbar curve than in patients with AIS (65.6 ± 15.7% vs 79.1 ± 13.2%). 16 In contrast, Lonner et al found no significant difference in correction rates between patients with Lenke type 5 AdIS and those with AIS (54.0 ± 11.4% vs 60.4 ± 36.4%), despite significantly greater surgical invasiveness in the AdIS group. 9 In our study, the thoracolumbar/lumbar Cobb angle correction was significantly smaller in the AdIS group than in the AIS group at final follow-up, in line with Pan et al’s findings. These results suggest that residual Lenke type 5 curves in adults are generally more rigid and less amenable to full correction compared to those in adolescents. Notably, our cohort included relatively young adult patients, yet a significant difference in postoperative alignment and correction rate was still observed between 2 groups. This finding implies that even in relatively young adults, residual lumbar curves originating from adolescent idiopathic scoliosis may exhibit early degenerative changes or structural stiffness that compromise surgical outcomes. However, given our limited sample size in this study, further research with a larger cohort is necessary to confirm the extent to which early degeneration affects correctability in this population.
Pan et al demonstrated that patients with AdIS had significantly larger L3 tilt postoperatively than patients with AIS (9.4° ± 4.1° vs 4.2° ± 3.5°), 16 which is consistent with our finding of a greater residual L3 tilt in the AdIS group (3.9° ± 4.1° vs 1.3° ± 4.0°). One possible explanation is that the reduced flexibility and disc degeneration in the AdIS group made it more difficult to level the LIV during surgery.
Preoperative MRI Findings
A previous study by Suzuki et al demonstrated significantly worse intervertebral disc degeneration at L1/2, L2/3, L3/4, and L4/5 in patients with AdIS than in those with AIS, with no significant difference at L5/S1, 12 which is consistent with our findings. The disc wedging angle, disc inclination angle, and coronal translation distance have all been implicated in disc degeneration, with the coronal translation distance particularly correlated with worsening disc degeneration.12,17 Further studies incorporating these parameters could help elucidate the underlying mechanisms driving disc degeneration in patients with AdIS.
Preoperative SRS-22 Score
The pain and mental health subdomains of the SRS-22 were significantly lower in the AdIS group than in the AIS group both before and after surgery. Previous studies have shown that pain tends to worsen with age in patients with AdIS, even when the degree of curvature remains comparable. Our results are consistent with these findings but do not reflect all patients with type 5 AdIS.
Erwin et al. reported that patients who underwent surgery had significantly lower preoperative SRS-22 scores than those who did not, despite similar ages and curve magnitudes. 5
In our cohort, by design, we included only patients who underwent surgery, meaning that patients with milder symptoms who did not require surgery were excluded. This selection inherently biases our sample toward more severe cases and could lead to overestimation of overall pain severity because the rate at which adolescents with sub-threshold curves go on to seek surgery in later adult life is unknown, representing an inherent uncertainty in the long-term prognosis of moderate AIS. In fact, long-term natural history data suggest that many AIS patients with mild to moderate curves experience only modest pain and disability in adulthood. For example, Farshad et al. followed moderate AIS cases for a mean of 42 years and found they had generally minimal long-term disability. 18 Likewise, Ohashi et al observed that patients with unoperated thoracolumbar/lumbar curves typically reported low back pain in only the mild to moderate range during 25-year followup. 19 We have now acknowledged in our analysis that our surgical series represents a more symptomatic subset of AIS. Further research is necessary to determine whether the pain levels observed in our AdIS group are truly higher than those of all patients with Lenke type 5 curves, or if they simply reflect the pre-selection of patients with greater symptoms for surgery.
Postoperative SRS-22 Score
Lonner et al found that the pain subdomain of the SRS-22 was significantly lower postoperatively in patients with AdIS across all Lenke types. 9 Similarly, our study showed that the pain and mental health scores of the SRS-22 were significantly lower in the AdIS group than in the AIS group after surgery.
One factor that may contribute to this difference is the distal extent of fusion: Chiu et al also noted that patients with Lenke type 5 AIS with LIV at L4 or lower had significantly lower pain scores than those with LIV at L3 or above. 20 In our study, LIV was significantly extended to L4 in the AdIS group, which may partly explain their inferior pain outcomes. In essence, the combination of greater degenerative change and the need for a longer fusion in adult patients can adversely affect postoperative quality-of-life measures.
Surgery-Related Factors
Prior reports analyzed the surgical outcomes of patients with AdIS across various curve types. Watanabe et al. noted that adult idiopathic scoliosis patients (aged ≥30 years) required lower LIV levels and longer operative times due to reduced flexibility. 10 Similarly, Lonner et al observed that patients with AdIS required more extensive fusion levels, had longer operative times, greater blood loss, and longer hospital stays than adolescent patients, despite comparable correction rates. 21 Consistent with these observations, Lavelle et al. have reported that young adult AIS patients tend to have more levels fused, higher intraoperative blood loss, and slightly lower percentage curve corrections relative to adolescents. 22 Their analysis of patients with Lenke type 5 AIS, although involving fewer cases, also found similar correction rates but significantly greater surgical invasiveness in the AdIS group. Furthermore, Pan et al reported that adult Lenke type 5 scoliosis patients had significantly greater preoperative L3/L4 coronal tilt and translation and achieved less correction of the major curve and vertebral rotation postoperatively than adolescents. Our study differs from previous work in that it focused exclusively on Lenke type 5 deformities in a relatively young adult cohort (<40 years), and it included a larger number of type 5 cases without combined surgical interventions. In contrast to previous studies, our results showed that the adult group achieved a significantly smaller magnitude of curve correction than the adolescent group. In addition to having markedly lower preoperative flexibility, the adult patients required more extensive surgical procedures, including fusion to a more caudal vertebra. These findings suggest that when definitive surgery is delayed until adulthood in Lenke type 5 AIS, surgeons are not only faced with greater invasiveness but may also be unable to achieve the same level of correction as in adolescent cases.
Limitations
This study had several limitations. First, the sample size was relatively small, which may introduce selection bias and limit the statistical power to detect small differences. Second, we focused on a young adult cohort (ages 20-40) and did not include patients older than 40 years; as such, our findings may not be generalizable to middle-aged or elderly patients with residual AIS. Third, pain was only assessed using the pain subdomain of the SRS-22. Future studies should incorporate additional outcome measures such as the visual analog scale or Oswestry Disability Index. Fourth, as mentioned earlier, since our study included only patients who eventually underwent surgery, it may not accurately represent the full clinical picture of Lenke type 5 AdIS, including those who were managed nonoperatively.
Fifth, although our data clearly indicate worse clinical outcomes in patients with type 5 AdIS compared to those who had surgery in adolescence, the retrospective design precludes a definitive conclusion that surgery must be performed during adolescence to ensure better long-term results. Previous studies have shown favorable long-term outcomes following surgery during adolescence for Lenke type 5 AIS. Kai et al and Fletcher et al have both reported that patients who underwent surgery during adolescence demonstrated sustained radiographic and clinical benefits over a 10-year follow-up.23,24 While these findings support the benefits of early surgery, a definitive conclusion regarding its superiority requires a direct comparison between the long-term outcomes of those who had surgery in adolescence and the short-term outcomes of those treated in early adulthood.
Prospective longitudinal studies will be needed to confirm whether earlier surgical intervention in Lenke type 5 AIS indeed confers superior outcomes and to further refine the recommended timing of surgery for these patients.
Conclusion
This study compared the surgical outcomes of patients with Lenke type 5 AIS treated in young adulthood with those of patients treated during adolescence. The AdIS group had significantly reduced flexibility with increased lumbar disc degeneration compared with the AIS group. The AdIS group had fewer LIV placements, longer operative times, and longer hospital stays than the AIS group. The preoperative and postoperative pain scores in the AdIS group were significantly lower than those in the AIS group. These results provide valuable insights into the timing of surgery in patients with residual Lenke type 5 AIS.
Footnotes
Ethical Considerations
This study was approved by the Committee on Ethics and Institutional Review Board (approval number: 20090042).
Consent to Participate
All participants provided informed consent for inclusion before treatment.
Author Contributions
RS and KW designed and conceived this study. RS collected data and drafted the manuscript. SS and OT analyzed and interpreted the results. KT, TI, TO, NN, MM and MN revised the manuscript critically for important intellectual content. All authors read and approved the final manuscript.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.