Selective lowest and upper instrumented vertebra for the correction of Lenke type 6C adolescent idiopathic scoliosis

Abstract

Retrospective Study. The selection of the upper instrumented vertebra (UIV) and lowest instrumented vertebra (LIV) in adolescent idiopathic scoliosis (AIS) Lenke type 6C is critical for achieving optimal spinal alignment and preventing post-surgical complications. This study evaluates the radiographic and clinical outcomes of two UIV and LIV selection strategies in patients with AIS Lenke 6C undergoing posterior spinal fusion. A retrospective analysis was conducted on 79 patients treated between 2011 and 2020. Patients were divided into two groups based on LIV and UIV selection: Group A, with LIV at the lower end vertebra (LEV) and UIV at the upper end vertebra (UEV), and Group B, with LIV one level caudal to the LEV (LEV-1) and UIV either one level above or below the UEV (UEV + 1 or UEV-1). Radiographic parameters, including coronal and sagittal balance, and clinical outcomes were compared between the groups. Both groups demonstrated significant improvements in spinal alignment. In Group A, the thoracic and lumbar/thoracolumbar Cobb angles improved from 36.1 ± 6° and 51.3 ± 3° preoperatively to 1.8 ± 0° and 2.8 ± 1° at six months postoperatively, and 1.7 ± 0° and 3.1 ± 2° at the final follow-up. In Group B, the thoracic and lumbar/thoracolumbar Cobb angles improved from 33.6 ± 5° and 51.7 ± 3° preoperatively to 2.6 ± 2° and 3.7 ± 2° at six months postoperatively, and 2.6 ± 2° and 3.7 ± 2° at the final follow-up (P = 0.105). Coronal and sagittal balance parameters showed comparable improvements in both groups. The SRS-22 scores at the final follow-up indicated significant enhancements in all domains, including pain, function, and mental well-being. The selection of UIV and LIV significantly impacts radiographic and clinical outcomes in AIS Lenke type 6C. Both strategies—LIV at LEV with UIV at UEV, and LIV at LEV-1 with UIV at UEV ± 1yielded comparable improvements in spinal alignment and patient-reported outcomes. However, the choice of UIV and LIV should be tailored to individual patient anatomy and surgical goals. This study underscores the importance of careful UIV and LIV selection in optimizing postoperative outcomes for AIS Lenke type 6C patients.

Introduction

Adolescent idiopathic scoliosis (AIS) represents an idiopathic spinal deformity of unknown aetiology. The increasing reliance on surgical interventions has demonstrated the potential to circumvent the complications associated with this deformity¹. Among the various Lenke classifications of AIS deformities, the double major curves, thoracic and thoracolumbar/lumbar curve, Lenke type 6C is a rare case that presents unique challenges in surgical management². Specifically, the selection of the lowest instrumented vertebra (LIV) and upper instrumented vertebra (UIV) in Lenke 6C cases remains a subject of debate, as improper selection can lead to suboptimal outcomes and postoperative complications³.

The LIV selection is particularly critical in Lenke 6C due to the biomechanical complexity of the thoracolumbar/lumbar curve and its impact on spinal balance and mobility. The LIV must balance the need for adequate curve correction with the preservation of lumbar motion segments to maintain functional mobility. However, selecting an LIV that is too proximal may result in the “adding-on” phenomenon, where the curve progresses distally, leading to loss of correction, coronal imbalance, and pelvic obliquity⁴. Conversely, an LIV that is too distal may unnecessarily sacrifice motion segments, increasing the risk of adjacent segment degeneration and long-term disability. Additionally, the timing of surgery relative to the patient’s growth spurt further complicates LIV selection, as overcorrection or under correction can have lasting effects on spinal alignment and posture⁵.

Over the past decades, there have been different criteria for selecting LIV for AIS Lenke 5/6 based on diverse expertise and experience⁶. However, no study entirely focuses on AIS Lenke 6C LIV selection. Few studies have laid the significance of proper manoeuvres for LIV selection during decision-making with the help of new technological advancements^7,8. Proper LIV selection could minimise the sequelae that are synonymous with improper LIV selection.

Similarly, the selection of UIV for Lenke 6C is an area with no substantial criteria. Improper UIV selection could lead to complications such as loss of correction, poor posture, shoulder imbalance, and loss of trunk balance and could further lead to postjunctional kyphosis⁹. The UIV selection for Lenke 6C curves can be influenced by various factors, including shoulder balance, proximal thoracic kyphosis, and the specific characteristics of the patient’s deformity.

Despite advancements in surgical techniques and instrumentation, there is no consensus on the optimal UIV and LIV selection for Lenke 6C, underscoring the need for further research in this area. Therefore, this study aims to assess the clinical efficacy of our UIV and LIV selection criteria for the correction of AIS Lenke 6C.

Method

According to the Declaration of Helsinki, the Institutional Review Board (IRB) approved this study. 79 patients with AIS Lenke 6C who underwent posterior-only instrumentation corrective surgery between 2010 and 2021.

The inclusion criteria for the study were as follows: (1) AIS patients without a history of prior operations and with AIS Lenke 6C scoliosis. (2) All patients underwent posterior corrective surgery with rod de-rotation and posterior instrumentation. (3) Patients had follow-up visits for a minimum of two years. (4) Demographic and radiological data were collected preoperatively and postoperatively, and a whole spine MRI was conducted in all patients to check for any spinal cord anomaly.

The exclusion criteria included: (1) patients who had previously undergone spine surgery for scoliosis or another condition; (2) patients who did not have selective fusion; (3) patients who did not have a minimum follow-up of three years; (4) patients who fit into additional categories of Lenke classification, such as Lenke 1, 2, 3, 4, and 5; (5) patients who had degenerative and neuromuscular scoliosis in addition to the aforementioned types. Patients lacking essential imaging, such as full-spine standing radiographs (anteroposterior and lateral views) at preoperative, immediate postoperative, and minimum three-year follow-up evaluations, were excluded.

Patients were divided into two groups based on instrumentation levels and age to evaluate the impact of LIV and UIV selection on outcomes:

Group A.

1. Age 16–18.

2. Upper instrumented Vertebrae (UIV) = Thoracic curve Upper End Vertebra (UEV).

3. Lowest Instrumented Vertebrae (LIV) = Lumbar curve Lower End Vertebra (LEV).

The LIV selection criteria align with well-established principles, including Lenke’s Lowest Substantial Touched Vertebra (LSTV) and Dubousset’s global balance guidelines^10,11. These methodologies prioritize selecting an LIV that ensures spinal stability, prevents adding-on, and maintains overall sagittal and coronal alignment¹².

Group B.

1. Age 10–15.

2. Upper Instrumented Vertebrae (UIV) = Thoracic curve Upper End Vertebra (UEV).

3. Shoulder Height Imbalance: Higher Left Shoulder UIV = UEV + 1, Higher Right Shoulder UIV = UEV-1.

4. Lowest Instrumented Vertebrae (LIV) = Lumbar curve Lower End Vertebra − 1 (LEV-1).

The rationale for selecting LIV as LEV-1 in Group B was based on the following clinical and radiographic considerations:

1. Lumbar Curve Flexibility: Patients with flexible lumbar curves, as demonstrated on preoperative bending films, were considered for LIV = LEV-1. Specifically, if the disc below the LEV could be corrected to a neutral or near-neutral position on bending films, LIV = LEV-1 was chosen to preserve an additional motion segment.

2. Age and Growth Potential: Younger patients (aged 10–15 years) with significant growth potential were more likely to undergo LIV = LEV-1 to minimize the risk of distal adding-on and preserve lumbar mobility.

3. Distal Disc Characteristics: The opening direction of the disc below the proposed LIV was evaluated on bending films. If the disc demonstrated flexibility and the ability to return to a horizontal position postoperatively, LIV = LEV-1 was selected.

By incorporating these criteria, the selection of LIV = LEV-1 in Group B was tailored to individual patient characteristics, ensuring optimal outcomes while preserving motion segments where appropriate. To minimize bias, we ensured that selection was based on standardised preoperative radiographic parameters rather than subjective surgeon preference.

Procedure

Under general anaesthesia, patients were positioned on a radiolucent table, and the lower and upper instrumented vertebrae (LIV and UIV) were identified using a freehand technique. Pedicle screws were inserted following posterior spinal anatomy, and a concave rod was contoured with a mild kyphotic curve to mitigate rod flattening during derotation. Rods were introduced cephalad to caudad using direct translation and rod rotation into the screw heads. A convex rod was placed in situ to preserve thoracic kyphosis without compressing the primary thoracic curve. Spinal deformity correction was achieved through concave rod derotation and coronal in situ bending, with benders positioned low to simultaneously address scoliosis and hypokyphosis. Compression was applied to screw heads as needed for structural upper thoracic or thoracolumbar/lumbar curves. Fusion was performed using autologous bone graft from facetectomies augmented with tricalcium phosphate and hydroxyapatite. Intraoperative spinal cord monitoring (MEP, SSEP, and electromyography, with or without screw stimulation) was utilized in all cases.

Radiography assessment

Preoperative, postoperative, and follow-up standing posteroanterior (PA) and lateral radiographs of the whole spine were obtained. The following parameters were measured:

Coronal plane measurements

1. Coronal Balance: Coronal balance was assessed by measuring the horizontal distance (in millimetres) between the C7 plumb line (a vertical line drawn downward from the centre of the C7 vertebral body) and the central sacral vertical line (CSVL) (a vertical line drawn upward from the centre of the sacrum). A value of 0 mm indicates perfect coronal balance, while positive or negative values indicate lateral deviation to the right or left, respectively.

2. UIV-CSVL: The UIV-CSVL measurement represents the horizontal distance (in millimetres) between the centre of the UIV and the central sacral vertical line (CSVL). This parameter evaluates the alignment of the upper end of the instrumentation relative to the midline of the spine. A smaller UIV-CSVL distance indicates better coronal alignment.

3. LIV-CSVL: Similarly, the LIV-CSVL measurement represents the horizontal distance (in millimetres) between the centre of the LIV and the CSVL. This parameter assesses the alignment of the lower end of the instrumentation relative to the midline.

4. LIV Tilt and LIVDA (Lower Instrumented Vertebra Disc Angle): LIV tilt was measured as the angle between the inferior endplate of the LIV and the horizontal plane. While LIVDA was measured as the angle between the inferior endplate of the LIV and the superior endplate of the adjacent caudal vertebra.

5. Radiographic Shoulder Height (RSH) and Pelvic Obliquity: RSH was measured as the vertical distance (in millimetres) between the horizontal lines drawn at the apex of each shoulder. Pelvic obliquity was measured as the angle between the horizontal plane and a line connecting the highest points of the iliac crests.

Sagittal plane measurements

1. Thoracic Kyphosis (T5–T12): Measured as the angle between the superior endplate of T5 and the inferior endplate of T12.

2. Lumbar Lordosis (T12–S1): Measured as the angle between the superior endplate of T12 and the superior endplate of S1.

3. Post-Junctional Angle: Measured as the angle between the inferior endplate of the UIV and the superior endplate of the adjacent cranial vertebra.

4. Sagittal Balance: Assessed by measuring the horizontal distance (in millimetres) between the C7 plumb line and the posterior superior corner of S1. A value of 0 mm indicates perfect sagittal balance, while positive or negative values indicate anterior or posterior deviation, respectively.

Clinical measurements

• The rib hump was measured using a scoliometer, both preoperatively, postoperatively, and at the final follow-up, to assess changes in trunk asymmetry over time.

Questionnaire

The SRS-22 questionnaire was translated and mailed to all patients prior to their last follow-up consultation. At the last follow-up, SRS-22 questionnaires were provided. Depending on the patient’s abilities, the SRS-22 questionnaire was completed at home by the patient or carers and returned at follow-up appointments.

Statistics evaluation

All analyses were performed using SPSS version 21.0 for Windows (SPSS Inc., Chicago, IL, USA). Descriptive statistics (means, standard deviations, and ranges) summarised the data. For longitudinal assessment of SRS-22 domains, two-way repeated measures ANOVA (Group × Time) with post-hoc paired t-tests (Bonferroni-corrected) were applied, while independent samples t-tests compared baseline characteristics between groups. Pearson correlation evaluated radiographic parameter relationships. A p-value < 0.05 defined statistical significance.

Results

The study involved 79 eligible patients who met the inclusion criteria (51 females and 28 males), divided into two groups (group A, and B) according to aforementioned selection criteria with a mean follow-up length of 39.44 ± 9 months for group A and 39.48 ± 10 months for group B. In Group A, the UIV level is at the UEV level, which is either T4 or T5 depending on the thoracic curve’s UEV, and the LIV is at L4. In Group B, the UIV is at T3 or T5, and the LIV is at L3. The average age in group A was 16.3 ± 1years and B 13.5 ± 1years. The average Risser sign of the enrolled patients was 3.1 ± 0 in both groups. The average operation time was 254 ± 37 min in group A and 256 ± 22 min in group B. Furthermore, the average blood loss in group A was 475 ± 53 ml; in group B, it was 468 ± 90 ml. The fusion segment was at the average of 11.08 ± 0 and 11.03 ± 0 instrumentation levels for both groups, respectively. Table 1 summarises the demographic and clinical features of the patients.

Table 1. Patient’s characteristics and surgical demography.

	A	B	P value
Age (yrs)	16.3 ± 1	13.5 ± 1	0.000
Sex (M/F)	13/26	15/25
Weight (kg)	54.5 ± 5	53.0 ± 5	0.276
Height (cm)	163.5 ± 4	161.5 ± 5	0.370
Risser	3.1 ± 0	3.1 ± 0
Follow-up (months)	39.44 ± 9	39.48 ± 10	0.986
Blood loss (ml)	475 ± 53	468 ± 90	0.649
Surgical time (mns)	254 ± 37	256 ± 22	0.733
Fused segment	11.08 ± 0	11.03 ± 0	0.797

Table 2 displays the radiographic parameters. The average Cobb angle of the thoracic and lumbar/thoracolumbar curves for group A was corrected from 36.1 ± 6° and 51.3 ± 3° to 1.8 ± 0°, 2.8 ± 1° (six months post-operative) and 1.7 ± 0°, 3.1 ± 2° (last follow-up) and group B average Cobb angle of the thoracic and lumbar/thoracolumbar curves were corrected from 33.6 ± 5° and 51.7 ± 3° to 2.6 ± 2° and 3.7 ± 2° six-month post-op and last follow-up respectively, P = 0.105. Preoperatively, the average C7-CSVL in both groups (A and B) were 6.6 ± 17and 8.1 ± 21 mm, P = 0.720. However, there was significant improvement at six months, and the last follow-up did not show statistical significance in both groups, with P = 0.190.

Table 2. Comparison of radiography Parameters.

	A	B	P -Value
Preoperative TK (0)	20.8 ± 9	18.9 ± 11	0.413
Postoperative	21.4 ± 6	22.9 ± 15	0.585
Last follow-up	24.6 ± 5	26.3 ± 6	0.221
Preoperative LL (0)	50.2 ± 12	48.4 ± 10	0.493
Postoperative	49.1 ± 22	53.6 ± 11	0.262
Last follow-up	54.9 ± 12	54.1 ± 11	0.762
Preoperative. C. alignment (mm)	6.6 ± 17	8.1 ± 21	0.720
Postoperative	6.4 ± 10	2.9 ± 10	0.146
Last follow-up	5.0 ± 11	1.7 ± 11	0.190
Preoperative SVA (mm)	14.2 ± 18	11.9 ± 12	0.527
Postoperative	19.0 ± 25	17.9 ± 19	0.837
Last follow-up	16.5 ± 24	14.2 ± 16	0.627
Thoracic curve (0)
Preoperative.	36.1 ± 6	33.6 ± 5	0.074
Postoperative	1.8 ± 0	2.6 ± 2	0.040
Last follow-up	1.7 ± 0	2.2 ± 1	0.074
Lumbar curve (0)
Preoperative	51.3 ± 3	51.7 ± 3	0.629
Postoperative	2.8 ± 1	3.7 ± 2	0.082
Last follow-up	3.1 ± 2	4.3 ± 3	0.105
RSH (mm)
Preoperative	1.6 ± 1	2.0 ± 1	0.340
Postoperative	1.5 ± 1	1.5 ± 1	0.915
Last follow-up	1.4 ± 1	1.5 ± 1	0.809
LIV tilt(0)
Preoperative	2.8 ± 2	2.6 ± 1	0.650
Postoperative	1.1 ± 1	1.2 ± 1	0.623
UIV tilt (0)
Preoperative	4.1 ± 1	3.6 ± 1	0.244
Postoperative	0.9 ± 0	0.8 ± 0	0.754
Follow-up	1.0 ± 0	1.0 ± 0	0.348
Preoperative Flexibility
Thoracic	37.7 ± 6	37.2 ± 4	0.705
Lumbar	56.0 ± 4	55.9 ± 4	0.924
Adding on (0)
Preoperative	5.0 ± 3	4.6 ± 3	0.573
Postoperative	1.3 ± 0	1.4 ± 1	0.838
Last follow-up	1.5 ± 1	1.6 ± 1	0.710
PJA (0)
Preoperative	2.6 ± 1	2.8 ± 1	0.514
Postoperative	3.5 ± 2	3.3 ± 2	0.563
Last follow-up	4.3 ± 2	5.2 ± 3	0.166
LIVDA (0)
Preoperative	3.8 ± 2	3.0 ± 1	0.092
Postoperative	1.3 ± 0	1.1 ± 0	0.328
Last follow-up	1.3 ± 1	1.3 ± 1	0.883
PO (0)
Preoperative	1.7 ± 1	2.2 ± 1	0.126
Postoperative	1.4 ± 1	1.7 ± 1	0.488
Last follow-up	1.7 ± 1	1.8 ± 1	0.779
UIV-CSVL Preoperative (mm)	27.9 ± 7	27.8 ± 8	0.965
Postoperative	9.9 ± 2	10.1 ± 2	0.776
Last Follow-up	4.5 ± 2	3.8 ± 2	0.226
LIV-CSVL Preoperative (mm)	13.2 ± 6	13.4 ± 6	0.925
Postoperative	6.4 ± 3	6.1 ± 3	0.749
Last Follow-up	4.5 ± 3	4.4 ± 3	0.851

Mean ± standard deviation (SD) for all numerical values.

The average RSH decreased from 1.6 mm to 2.0 mm (groups A and B) to 1.5 mm and 1.5 mm at six months post-operatively. At the last follow-up, the average RSH was 1.4 mm for group A and 1.5 mm for group B, with no significant differences between the groups (P = 0.809). Both groups exhibited no significant changes in pelvic obliquity from the preoperative to the last post-operative follow-up.

In terms of sagittal alignment metrics, thoracic kyphosis considerably changed from 20.8 ± 9° for group A and 18.9 ± 11° for group B to 21.4 ± 6° for group A and 22.9 ± 15° for group B at six months after surgery, and there were significantly increased at last follow-up to 24.6 ± 5° for group A and 26.3 ± 6° for group B (P = 0.221). There was no statistically significant difference in lumbar lordosis or SVA before and after surgery.

There was a significant decrease in LIV tilt from 2.8 ± 2° for group A and 2.6 ± 1° for group B preoperatively to 1.1 ± 1° for group A and 1.2 ± 1° at six months post-operative (P = 0.632). The LIV-CSVL preoperative distance of 13.2 ± 6 and 13.4 ± 6 in both groups decreased significantly to 4.5 ± 3 mm and 4.4 ± 3 mm, with no statistical significance (P = 0.851) at the last follow-up. The UIV-CSVL also showed statistically insignificant values of 4.5 ± 2 mm and 3.8 ± 2 mm at the last follow-up in both groups, compared to the preoperative gap of 27.9 ± 7 mm and 27.8 ± 8 mm in both groups, respectively. Also, there was a noteworthy decrease in the lowest instrumented vertebrae disc angle (LIVDA) from 3.8 ± 2° and 3.0 ± 1° to 1.3 ± 1° and 1.3 ± 1° in groups A and B, respectively, at the last follow-up (P = 0.883). Similarly, there was a significant decrease in the UIV tilt in both groups, with no significant differences.

There was a slight increase in PJA in both groups (group A and B) with an increase from 2.8 ± 1° and 2.6 ± 1° (preoperative) to 4.3 ± 2° and 5.2 ± 3° (last follow-up, P = 0.166). The adding-on recorded a significant decrease in both groups A and B from 5.0 ± 3° and 4.6 ± 3° preoperatively to 1.5 ± 1° and 1.6 ± 1° (last follow-up, P = 0.710).

The thoracic rib hump was measured pre, post and last follow-up; there were significant differences, with Fig. 1 showing the outcome.

Fig. 1 [Images not available. See PDF.]

Showing the Rib hump levels at pre, post and at the last follow-up visitation.

The correlation analysis revealed significant associations between preoperative and last follow-up data. Specifically, the preoperative C7-CSVL showed a positive correlation with the last follow-up UIV-tilt (r = 0.364, P = 0.001). Furthermore, the preoperative UIV tilt exhibited a positive correlation with postoperative adding-on, LIV-tilt, and last follow-up Cobb’s angle (r = 0.290, r = 0.364, and r = 0.293, P = 0.010, 0.001, and 0.009). Additionally, the preoperative UIV-CSVL was positively correlated with the postoperative and last follow-up UIV-CSVL (r = 0.437, r = 0.433, P = 0.000), as well as with preoperative to the last follow-up LIV-CSVL (r = 0.865, r = 0.506, and r = 321, P = 0.000, 0.000, 0.004). Furthermore, the last follow-up Adding-on was positively correlated with the preoperative UIV-tilt, LIV-tilt, and postoperative Cobb’s angle, SVA, PO, LIVDA, and LIV-tilt (r = 0.450, r = 0.594, and r = 0.460, r = 0.398, r = 0.379, r = 0.424 and r = 0.830, P = 0.000 respectively), as well as with the last follow-up Cobb’s angle, SVA, PO, and LIVDA (r = 0.613, r = 0.359, r = 0.414 and r = 0.599, P = 0.000, respectively).

Functional outcome

All participating patients completed the SRS-22 questionnaires. Table 3 shows improvements in all SRS-22 domains at follow-up compared to preoperative scores, including function, pain, self-image, mental health, and satisfaction.

Table 3. Comprehensive longitudinal analysis of SRS-22 domains (Group A: n = 39; group B: n = 40).

Domain	Pre-op Mean ± SD	Post-op Mean ± SD	Follow-up Mean ± SD	Time Effect (ANOVA, p)	Group Effect (ANOVA, p)	Group×Time (ANOVA, p)	Pre→Follow-up (Mean ± SD)	Paired t-test (p)
Function	4.48 ± 0.23	4.48 ± 0.41	4.53 ± 0.25	F = 1.50, p = 0.23	F = 0.12, p = 0.73	F = 0.45, p = 0.64	+ 0.05 ± 0.42	p = 0.65
Pain	4.52 ± 0.32	4.32 ± 0.71	4.42 ± 0.38	F = 4.20, p = 0.02	F = 0.85, p = 0.36	F = 1.20, p = 0.31	−0.10 ± 0.51	p = 0.40
Self-Image	3.12 ± 0.35	4.28 ± 0.52	4.37 ± 0.29	F = 15.3, p < 0.001	F = 0.30, p = 0.59	F = 0.60, p = 0.55	+ 1.25 ± 0.38	p < 0.001
Mental Health	3.75 ± 0.12	4.32 ± 0.42	4.48 ± 0.15	F = 1.80, p = 0.17	F = 0.05, p = 0.83	F = 0.20, p = 0.82	+ 0.73 ± 0.28	p = 0.01
Satisfaction	3.13 ± 0.25	4.41 ± 0.32	4.54 ± 0.28	F = 12.5, p < 0.001	F = 0.22, p = 0.64	F = 0.10, p = 0.91	+ 1.41 ± 0.30	p < 0.001

Mean ± standard deviation (SD) for all numerical values.

All 79 patients had no wound infections, neurologic problems, pseudarthrosis, paralysis, revision procedures, or other issues. Additionally, no intraoperative or postoperative complications were documented during the follow-up visit.

Discussion

Proper LIV selection is crucial for achieving maximum correction while retaining lumbar mobility in treating AIS patients with structural thoracic and TL/L curves³. Furthermore, UIV selection is vital for achieving appropriate body and posture balance, particularly in patients with AIS Lenke 6C with double major curves. Improper UIV selection could lead to poor posture, shoulder imbalance, and body image disturbance. Body image disturbance in AIS could lead to chronic discontent, concern, and distress over outer appearance that interferes with social relationships and activities and causes psychosocial suffering¹³. AIS often exhibits more significant body image disturbance than healthy controls, which has been linked to psychological problems such as isolation, denial, discomfort, and depression^14,15.

AIS Lenke 6C is one of the rare forms of Lenke’s classification with the double major curve that has historically been managed based on the criteria for AIS Lenke 5^4,16. The selection of the LIV in AIS Lenke 5 and 6 has been a topic of debate, with various authors reporting different selection criteria^5,6,14,17. Lenke’s LSTV criteria emphasize selecting the LIV as the last vertebra substantially touched by the CSVL on standing radiographs^18,19. This approach aims to achieve optimal coronal balance while minimizing the risk of distal adding-on. In our study, Group A aligned closely with this principle, as the LEV is often the last vertebra touched by the CSVL in Lenke 6C curves. The results showed excellent coronal balance and low rates of adding-on in this group, supporting the validity of Lenke’s criteria for LIV selection. However, in Group B (LIV = LEV-1), the LIV was intentionally selected one level caudal to the LEV to evaluate the impact of extending fusion. While this approach also yielded satisfactory outcomes, it raises questions about the potential overcorrection of the lumbar curve and the unnecessary sacrifice of motion segments, which could increase the risk of adjacent segment degeneration in the long term. On the other hand, Dubousset’s principles for LIV selection focus on the opening direction of the disc below the LIV on bending films and the ability of this disc to return to a horizontal position postoperatively²⁰. According to these principles, the LIV should be selected at a level where the disc below is flexible and can be corrected to a neutral or near-neutral position. In our study, preoperative bending films were used to assess disc flexibility, but the decision to extend fusion to LEV-1 in Group B was primarily based on age and growth potential rather than disc characteristics.

Moreover, some authors have suggested LIV selection based on their experience, including considerations such as LIV translation 28 mm and 25⁰ and the vertebra’s position relative to the CSVL line⁷. However, controversy and unsatisfactory outcomes have surrounded these selection criteria, with reported correction and unsatisfactory rates of 42%¹². There have been instances of an 87% correction rate with an unsatisfactory rate of 8.7%⁷. Therefore, the proper selection of LIV and UIV is essential for achieving optimal correction and avoiding complications in AIS patients with different curve patterns.

A recent study employed a proper de-rotation technique for the LIV selection following Nash-Moe grade II vertebrae rotation²¹. Despite the correction rate in their study, complications such as adding-on and coronal decompensation were observed. Howbeit, due to different guidelines and preferences stemming from experiences, there is still no consensus yet agreed to identify the LIV designated level in a patient with AIS Lenke 6C without complications. Improper LIV selection is known to potentially lead to lower back pain, adjacent vertebrae disc degeneration, lumbar function loss and adding-on^{3,22, 23–24}. Therefore, our study aimed to assess the clinical efficiency of selective LIV and UIV for AIS Lenke 6C.

One of the strengths of our study was the utilisation of different levels of LIV selection in both groups and the inclusion of age, which we believe could be a significant factor in determining the appropriate LIV level to preserve more segments for mobility and flexibility. Contra to age, the inclusion of gender is also an integral part of LIV selection due to differences in growth spurts between females and males.

Female patients typically experience growth spurts between the ages of 10–15, while male patient’s growth spurts can extend to around 16-17-year-old. Therefore, the selection of LIV to preserve more lumbar segments can also be considered based on gender. In our study, the B group’s average age was 13 years old, ranging between 10 and 15, compared to group A, with an average of 16 years old Fig. 2.

Fig. 2 [Images not available. See PDF.]

A 16-year-old female patient with Lenke 6C correction fusion demonstrated great coronal balance and correction rate maintained at 4 years.

Our study criteria for LIV selections included LEV = LIV, LEV-1 = LIV and the identification of the most tilted vertebrae that could return to a normal position under bending lateral radiography images. Withal, our study’s selective LIV fusion level aimed to preserve more segments for mobility while avoiding a series of complications reported in previous studies.

In patients between the ages of 10–15 years old in group B, LIV selection criteria are at LEV-1 Fig. 3. Our studies showed remarkable correction compared to earlier studies without reported postsurgical complications^4,8,17,21. The average age of our study in group B is 14 years old, with good post-operative outcomes on follow-up. The aim for LIV selection in younger patients is due to proper de-rotation compared to older patients with rigid de-rotation. In group A, with older patients, the LIV selection level is quite lower due to rigidity in derotation when the patient is attaining maturity. Furthermore, there is a substantial reduction in adjacent intervertebral disc degeneration in younger patients after fusion.

Fig. 3 [Images not available. See PDF.]

A 14-year-old female patient with Lenke 6C correction fusion demonstrated great coronal balance and correction rate maintained at 3 years.

From the patient’s point of view, improvement of the thoracic rib hump, reduction of waistline asymmetry, shoulder height disparity, and impaired truncal balance are expected outcomes of AIS treatment. According to specific reports, such a physical look causes the patient psychological discomfort related to how severe the deformity is¹⁴. The patient’s opinion of their deformity is primarily influenced by the asymmetry of their trunk or waist, which is the most subjective of these physical characteristics^14,15. Therefore, the selection of UIV is essential, and if there is improper selection of UIV in patients with AIS Lenke 6C, this could lead to postsurgical psychological discomfort.

As it has never been a criterion for the selection of UIV in patients with AIS Lenke 6C, a recent study on short-level instrumentation in Lenke 5 and 6 reported an increase in the main thoracic curve³. Contrary to their report, to minimise instrumentation level, selective UIV instrumentation level is essential for the correction of the main thoracic curve to aid good posture, shoulder balance and postsurgical psychological comfort for patients. In our study, our UIV selection criteria are either between UEV in group A compared to group B with UIV at UEV + 1 or −1 in case of preoperative right or left shoulder imbalance and thoracic rib hump. At the last follow-up, both groups reported a better Cobb angle correction of an average mean of 3.1 ± 2° for group A and 4.3 ± 3° for group B compared to previous studies^2,21.

It is essential to thoroughly assess the coronary balance and alignment before and after surgery, particularly when choosing different levels of LIV and UIV in both groups. Upon conducting UIV-CSVL and LIV-CSVL assessments in both groups, we observed a significant improvement during the last follow-up visit. However, it is worth noting that there was no statistically significant difference between the two groups. Furthermore, the rib hump correction shows a significant outcome on follow-up visitation compared to the preoperative level.

In patients with AIS, the ultimate therapeutic goal is not to correct radiological measurements but to improve their look and spinal balance while shortening the fusion level to reduce operating time and accompanying perioperative problems. Nonetheless, the radiography outcome of our study was outstanding; the coronal balance, sagittal balance, kyphosis, and lumbar lordosis improved at the last follow-up, and there were no substantial differences in RSH level at the last follow-up visitation.

Considering the relevance of shoulder balance and trunk or waist asymmetry on the patient’s perception of their deformity, we believe that selective UIV and LIV are viable treatment options for AIS Lenke 6C curves. The modern advancement of surface topography technology, which provides consistent and reproducible measurements of cosmetic indices and the patient-reported scoring system, may help future research.

The clinical outcome using the SRS-22 questionnaire was remarkable in both groups; there were improvements in all aspects of the clinical outcome, and no postsurgical complications such as adding-on or shoulder imbalance and postjunctional kyphosis were reported.

Furthermore, our study demonstrated some parameters for predicting postsurgical complications. The presented parameter could further aid the surgeon’s preparation for the selection of instrumentation level in patients with AIS Lenke 6C. This is the first study that comprehensively evaluated the efficiency of selective UIV and LIV instrumentation levels in AIS Lenke 6C without associating with other Lenke classifications.

The current study has several important limitations that must be acknowledged. Firstly, it was retrospective in design, which inherently introduces biases, including selection and information bias, due to the reliance on historical clinical and radiological data collection. Secondly, all surgical procedures were conducted by a single senior surgeon at one facility, limiting the generalizability of the results and potentially introducing operator-dependent biases. Additionally, the rationale for lowest instrumented vertebra (LIV) selection in Group B was individualized based on patient-specific characteristics, reflecting surgeons’ preferences rather than standardized criteria. This individualized approach introduces selection bias, which must be explicitly recognized as it may influence outcomes significantly.

Furthermore, there were notable differences in baseline characteristics between the two groups, such as upper instrumented vertebra (UIV) levels and age distribution. These differences may confound the results and limit the ability to draw definitive conclusions about causal relationships. The association between these variables and the outcomes was not thoroughly addressed with advanced statistical adjustments, and we acknowledge this as a limitation of our methodological approach.

Moreover, the study included a relatively small sample size and a limited follow-up period, constraining the robustness of the statistical analyses and long-term outcome assessment. Therefore, future prospective studies with standardized LIV selection criteria, larger patient cohorts, longer follow-up periods, and multi-centre collaboration would provide stronger evidence and greater clarity to substantiate and generalize our findings.

Conclusion

The current study assessed the clinical effectiveness of our UIV and LIV selection levels for correcting AIS Lenke 6C. The clinical and radiography outcomes in this study highlighted that different selections of UIV and LIV levels could be considered for better postsurgical outcomes of patients. Our study demonstrated outstanding correction of Cobb’s angle, coronal balance, sagittal, RSH, rib hump, and pelvic oblique balance without intraoperative or postsurgical complications on the last follow-up visitation. Howbeit, in the UIV and LIV selection criteria, shoulder balance, rib hump and age should be considered for better selection to minimise postoperative complications and promote outstanding body posture and mental well-being.

Acknowledgements

Acknowledgements: None.

Author contributions

Author’s contributions: All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work as a whole, and approve this version to be published.EA was responsible for the concept and design of the overall study. Data were retrieved and assembled by EA, ZHQ, DA and GZ. All authors contributed to the writing and final approval of the manuscript.

Funding

None.

Data availability

Availability of data and materials: Data is provided within the manuscript.

Declarations

Competing interests

The authors declare no competing interests.

Ethical statement

All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work as a whole, and approve this version to be published. The ethical committee of Xiangya Hospital, Central South University approved the study, and informed consent was obtained from all subjects and/or legal guardian(s). Furthermore, all methods were performed per our hospital’s relevant guidelines and regulations. Ethics approval number 21017033559.

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Karol, L. A., Virostek, D., Felton, K., Jo, C. & Butler, L. The Effect of the Risser Stage on Bracing Outcome in Adolescent Idiopathic Scoliosis Journal of Bone and Joint Surgery 98(15), 1253–1259 (2016).

2. Mattila, M; Jalanko, T; Helenius, I. En bloc. 2013;38(18):1576–83. En bloc lateral mass resection in spinal tumor surgery. Spine (Phila Pa. 1976); 2013; 38, 18 pp. 1576-1583. [DOI: https://dx.doi.org/10.1097/BRS.0b013e3182997092] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/23680831]

3. Li, Z et al. The radiographic parameter risk factors of rapid curve progression in Lenke 5 and 6 adolescent idiopathic scoliosis. Med. (United States); 2017; 96, 52 pp. 1-7.1:CAS:528:DC%2BC2sXhsVSlt7jE

4. Yanik, H. S., Ketenci, I. E. & Erdem, S. Lowest instrumented vertebrae selection in posterior fusion of Lenke 3 C/6 C adolescent idiopathic scoliosis: L3 versus L4, when LEV is L4. Arch Orthop Trauma Surg [Internet]. 2023;(0123456789). Available from: https://doi.org/10.1007/s00402-023-04872-4

5. Li, J. et al. Selective Fusion in Lenke 5 Adolescent Idiopathic Scoliosis. World Neurosurg [Internet]. ;118:e784–91. (2018). Available from: https://doi.org/10.1016/j.wneu.2018.07.052

6. Ilharreborde, B et al. Selective versus hyperselective posterior fusions in Lenke 5 adolescent idiopathic scoliosis: comparison of radiological and clinical outcomes. Eur. Spine J.; 2017; 26, 6 pp. 1739-1747.1:STN:280:DC%2BC1cvlvVKltQ%3D%3D [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28389887]

7. Shao, X. et al. How to select the lowest instrumented vertebra in Lenke 5/6 adolescent idiopathic scoliosis patients with derotation technique. Eur Spine J [Internet]. ;31(4):996–1005. (2022). Available from: https://doi.org/10.1007/s00586-021-07040-7

8. Chang, S. Y. et al. Clinical outcomes of selective fusion for the thoracolumbar-lumbar curve in patients with Lenke type 6C adolescent idiopathic scoliosis: A preliminary study. Journal of Pediatric Orthopaedics Part B 30(3), 211–217 (2021).

9. Oba, H et al. Upper instrumented vertebra to the right of the lowest instrumented vertebra as a predictor of an increase in the main thoracic curve after selective posterior fusion for the thoracolumbar/lumbar curve in Lenke type 5 C adolescent idiopathic scoliosis: mul. J. Neurosurg. Spine; 2019; 31, 6 pp. 857-864. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31443081]

10. Beauchamp, EC; Lenke, LG; Cerpa, M; Newton, PO; Kelly, MP. Selecting the touched vertebra as the lowest instrumented vertebra in Lenke type 1 and 2 adolescent idiopathic scoliosis curves. Spine Deform; 2020; 8, 2 pp. 1966-1973. [DOI: https://dx.doi.org/10.1007/s12345-020-01011-z]

11. Yang, M et al. Prevalence, risk factors, and characteristics of the adding-On phenomenon in idiopathic scoliosis after correction surgery. Spine (Phila Pa. 1976); 2018; 43, 11 pp. 780-790. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28937532]

12. Deml, MC. Commentary on selection of optimal lower instrumented vertebra for adolescent idiopathic scoliosis surgery. Neurospine; 2023; 20, 3 pp. 808-809. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37798974][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10562250]

13. Kuklo, T. R. et al. Correlation of radiographic, clinical, and patient assessment of shoulder balance following fusion versus nonfusion of the proximal thoracic curve in adolescent idiopathic scoliosis. Spine 45(18), 2013–2020 (2020).

14. Lonner, B. S. et al. Body Image Disturbance Improvement After Operative Correction of Adolescent Idiopathic Scoliosis. Spine Deform [Internet]. ;7(5):741–5. (2019). Available from: https://doi.org/10.1016/j.jspd.2018.12.005

15. Lafond, D; Descarreaux, M; Normand, MC; Harrison, DE. Photographic analysis of human posture: A literature review. J. Chiropr. Med.; 2014; 13, 2 pp. 56-61. [DOI: https://dx.doi.org/10.1016/j.jcm.2014.03.002]

16. Chiu, CK et al. Does age at surgery affect outcomes in adolescent idiopathic scoliosis patients?. Asian Spine J.; 2019; 13, 5 pp. 785-792. [DOI: https://dx.doi.org/10.31616/asj.2018.0322]

17. Chen, K. et al. Immediate postoperative coronal imbalance in Lenke 5 and Lenke 6 adolescent idiopathic scoliosis: Is it predictable? Eur Spine J [Internet]. ;28(9):2042–52. (2019). Available from: https://doi.org/10.1007/s00586-019-06019-9

18. Alonge, E., Zhang, G., Zhang, H. Q. & Guo, C. Comparison between the lowest instrumented vertebrae L3 with the use of direct vertebrae rotation (DVR) and the lowest instrumented vertebrae L4 for non-DVR in adolescents with idiopathic scoliosis Lenke 5 C/6 C: when LEV is L4. J Orthop Surg Res [Internet]. ;19(1). (2024). Available from: https://doi.org/10.1186/s13018-024-04961-z

19. Yamauchi, I et al. Relationship between lumbosacral transitional vertebra and postoperative outcomes of patients with Lenke 5 C adolescent idiopathic scoliosis: a minimum 5-year follow-up study. Eur. Spine J. Off Publ Eur. Spine Soc. Eur. Spinal Deform Soc. Eur. Sect. Cerv. Spine Res. Soc.; 2023; 32, 6 pp. 2221-2227.

20. Yang, X et al. Pre-and postoperative spinopelvic sagittal balance in adolescent patients with Lenke type 5 idiopathic scoliosis. Spine (Phila Pa. 1976); 2015; 40, 2 pp. 102-108. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25569527]

21. Sudo, H; Abumi, K; Ito, M; Kotani, Y; Minami, A. The impact of direct vertebral body derotation on the lumbar prominence in Lenke type 5 C curves. Spine (Phila Pa. 1976); 2012; 37, 21 pp. E1299-E1305. [DOI: https://dx.doi.org/10.1097/BRS.0b013e3182667c8d]

22. Direito-Santos, B et al. Long-Term Follow-Up of anterior spinal fusion for thoracolumbar/lumbar curves in adolescent idiopathic scoliosis. Spine (Phila Pa. 1976); 2019; 44, 16 pp. 1137-1143. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30896585]

23. O’Donnell, C. et al. Anterior Spinal Fusion and Posterior Spinal Fusion Both Effectively Treat Lenke Type 5 Curves in Adolescent Idiopathic Scoliosis: A Multicenter Study*. Spine Deform [Internet]. ;6(3):231–40. (2018). Available from: https://doi.org/10.1016/j.jspd.2017.09.054

24. Zuckerman, S. L. et al. Surgeons lack of agreement on determining preoperative radiographic and clinical shoulder balance in adolescent and adult idiopathic scoliosis patients. Eur Spine J [Internet]. ;30(3):661–7. (2021). Available from: https://doi.org/10.1007/s00586-020-06702-2