The Influence of Relative Curve Correction and Upper Instrumented Vertebra (UIV) Tilt Angle on Post-operative Shoulder Balance Following Posterior Spinal Fusion (PSF) in Lenke Type 1 and 2 Adolescent Idiopathic Scoliosis (AIS) Patients

Abstract

Study Design

Retrospective study.

Objective

To investigate the relative curve correction of Proximal Thoracic (PT) and Main Thoracic (MT) scoliosis curves and their relationship with post-operative UIV tilt angle as well as post-operative shoulder imbalance.

Methods

151 AIS patients with Lenke type 1 and 2 curves who underwent PSF were reviewed. Relative PT/MT Correction Rate (RCR), Relative PT/MT Residual Cobb angle (RRCA) and Post-operative UIV Tilt Angle were investigated for their association with post-operative T1 tilt (medial shoulder balance), Cervical Axis (neck balance) and Clavicle Angle (Lateral Shoulder Balance).

Results

There were 107 patients with Lenke 1 curves and 44 patients with Lenke 2 curves. Overall, the prevalence of medial shoulder, neck and lateral shoulder imbalance was 27.8%, 23.2% and 9.9%, respectively. There was a significant association between RRCA and medial shoulder, lateral shoulder and neck balance but RCR had a significant association with lateral shoulder and neck balance. There was a significant and strong correlation (r = 0.708) between post-operative UIV tilt angle and post-operative T1 Tilt. Unlike RCR, RRCA had shown a statistically significant correlation to post-operative UIV tilt angle with an r-value of 0.406.

Conclusions

RRCA had shown a statistically significant correlation with the post-operative UIV tilt angle, and post-operative T1 tilt ie, medial shoulder balance. However, the post-operative UIV tilt angle showed the strongest strength of correlation with the post-operative T1 tilt (r = 0.708).

Keywords

adolescent idiopathic scoliosis relative correction rate relative residual cobb angle UIV tilt angle shoulder balance

Introduction

Shoulder balance plays a significant role in the overall cosmetic appearance among patients with adolescent idiopathic scoliosis (AIS).¹ Post-operative shoulder imbalance (PSI) is common following posterior spinal fusion (PSF) surgery.² The incidence of PSI among scoliosis patients post-operatively was reported to be up to 25%.³ Shoulder imbalance which comprises the medial shoulder, lateral shoulder as well as neck are represented by radiographic parameters of TI Tilt, Clavicle Angle and Cervical Axis, respectively.^3-8

Recently, 2 important risk factors of PSI have been reported, ie, relative curve correction of the proximal thoracic curve/main thoracic curve (PTC/MTC) and the upper instrumented vertebra (UIV) tilt angle. Lee et al had emphasized that PSI was associated with higher Risser grade, larger post-operative proximal wedge angle (PWA) and greater post-operative PTC/MTC ratio⁴ in Lenke 2 curves. Yuan et al also highlighted the importance of harmonizing the correction rate ratio of the PTC, MTC and LC in their study.⁶ As for the UIV tilt angle, this radiological parameter is an independent variable that significantly correlates with post-operative T1 Tilt and Cervical Axis measurement.⁹ Patients with positive post-operative UIV tilt angle were 14.9 times more likely to experience positive medial shoulder imbalance (T1 tilt angle ≥+4°) and 3.3 times more likely to experience positive neck imbalance (cervical axis ≥+4°).¹⁰ Therefore, achieving the optimal UIV tilt angle may prevent neck tilt with medial shoulder imbalance due to trapezial prominence but not lateral shoulder imbalance.¹¹

To date, the correlation between these 2 factors ie, Relative Curve Correction (RCC) and Post-operative UIV Tilt Angle is yet to be fully investigated. Therefore, the purpose of this study is to investigate the RCC and its relationship with Post-operative UIV Tilt Angle as well as post-operative shoulder imbalance.

Methodology

Statement

Informed consent was not required for this study as it is a retrospective study. The study was deemed exempt from the requirement for informed consent by the institutional Medical Research Ethics Committee (MREC).

Study Design

This was a retrospective study that was carried out in a single tertiary academic centre. 151 AIS patients with Lenke type 1 and 2 curves who underwent PSF between 2019 to 2021 were reviewed. The institutional ethical approval was obtained prior to the commencement of the study (MREC ID NO: 2023519-12466). The inclusion criteria included AIS patients with Lenke 1 and 2 curves, patients with sufficient clinical and radiological parameters with at least 24 months follow-up post-operatively. Patients with non-idiopathic scoliosis, juvenile scoliosis and congenital scoliosis and patients who underwent anterior spinal fusion surgery, two-staged surgery or revision surgery were excluded from the study.

The null hypothesis proposed that there was no significant association between relative curve correction, post-operative UIV tilt angle and post-operative medial, neck and lateral shoulder balance.

The objectives of this study were to investigate (comparing between Lenke 1 and Lenke 2):

1. The prevalence of radiological medial shoulder, neck and lateral shoulder imbalance at the final follow-up.

2. The association between medial shoulder, neck and lateral shoulder imbalance with Relative Curve Correction (RCC) and Post-operative UIV Tilt Angle.

3. The association between RCC and Post-operative UIV Tilt Angle.

Definition of Shoulder Balance Parameters

PSI is divided into 2 components which are lateral and medial shoulder balance. Medial shoulder balance is represented by T1 Tilt whereas lateral shoulder balance is represented by Clavicle Angle.¹² The Cervical Axis represents the neck tilt.^8,13 The parameters^14-17 were defined and illustrated as shown below (Figure 1):

Figure 1.

Illustration of T1 Tilt, Clavicle Angle, Cervical Axis and UIV tilt angle measurements.

T1 Tilt

Angle (in degree) is formed between a horizontal line and a line drawn along the parallel to the superior end plate of the T1 vertebrae. This parameter is directional. When the T1 Tilt is tilted towards the right, it is regarded as a positive (+) value and vice versa.

Clavicle Angle

Angle (in degree) is formed by the intersection of a horizontal line and a line connecting the highest 2 points of each clavicle. This parameter is directional. When the Clavicle Angle is tilted towards the right, it is regarded as a positive (+) value and vice versa.

Cervical Axis

Angle (in degree) between the longitudinal axis of cervical spine (centre of C2 odontoid process and centre of C7) and the vertical axis. This parameter is directional. When the Cervical Axis is tilted towards the right, it is regarded as a positive (+) value and vice versa.

Upper Instrumented Vertebra (UIV) Tilt Angle

Angle (in degree) is formed by the intersection between a horizontal line and a parallel line drawn to the upper end plate of the UIV. This parameter is directional. When the UIV tilt angle is tilted towards the right, it is regarded as a positive (+) value and vice versa¹⁰

Patients were categorized into balanced and imbalanced shoulder based on the following criteria as illustrated in Table 1. Patients with −3° ≤ T1 Tilt ≤3° , −2° ≤ Clavicle Angle ≤2° and −3° ≤ Cervical Axis ≤3° were considered as having balanced shoulder whereas values out of these ranges would be described as imbalanced shoulder.¹⁴ (Table 1)

Table 1.

Range of Balanced and Imbalanced T1 Tilt, Clavicle Angle and Cervical Axis.

Parameters	Range
Parameters	Balanced	Imbalanced
T1 tilt	−3° ≤ T1 Tilt ≤3°	T1 tilt < −3°, T1 tilt >3°
Clavicle angle	−2° ≤ Cla-A ≤2°	Cla A < −2° , Cla-A >2°
Cervical axis	−3° ≤ CA ≤3°	CA < −3°, CA >3°

Relative curve correction was assessed using these radiological parameters:

(1). Relative PT/MT Correction Rate (RCR): This parameter is the ratio of the Correction Rate (CR) of the PT curve to the MT curve. This parameter is reported as a percentage whereby a value of 100% indicates the RCR is equal.

R C R (%) = [\frac{\frac{(p r e o p P T C o b b - p o s t o p P T C o b b)}{p r e o p P T C o b b}}{\frac{(p r e o p M T C o b b - p o s t o p M T C o b b)}{p r e o p M T C o b b}}] x 100

(2). Relative PT/MT Residual Cobb angle (RRCA): This parameter is the ratio between the post-operative Cobb Angle of the PT to the MT curve. This parameter is reported as a percentage whereby a value of 100% indicates the RRCA is equal.

R R C A (%) = [\frac{p o s t o p P T C o b b}{p o s t o p M T C o b b}] x 100

Data Collection

Participating surgeons measured the PT Cobb angle, MT Cobb angle, PT side bending Cobb angle, MT side bending Cobb angle, UIV tilt angle, T1 Tilt, Clavicle Angle as well as Cervical Axis pre-operatively and post-operatively. Then, the PT correction rate, MT correction rate and RCR and RRCA percentages were calculated. RCR, RRCA and UIV tilt angle were correlated with PSI parameters while RCR and RRCA were correlated with UIV tilt angle.

Surgical Strategy

All patients underwent PSF using pedicle screws construct. Alternate level pedicle screw placement was implemented. Correction maneuver consisted of rod translation technique and direct vertebral derotation. Radical facetectomies were performed to increase spinal flexibility prior to correction as well as to augment the fusion process. Autogenous bone grafts obtained from the spinous processes, laminae, facet joints and transverse processes were processed and distributed over the fusion bed. The UIV in Lenke 1 curves were at T3, T4, or T5 when the right medial shoulder was high or when the medial shoulder was balanced. Fusion to T2 was performed when the left medial shoulder was high. In Lenke 2 curves, the UIV were T1 or T2 when the left medial shoulder was high and T3 when it was level or the right medial shoulder was high. The optimal UIV tilt angle can be calculated from the Supervised Supine Side Bending radiographs preoperatively (11) and confirmed during surgery using the crossbar method (17).

Statistical Analysis

SPSS Version 27.0 (SPSS Statistics for Windows, IBM Corp, Armonk, NY, USA) was utilised for data analysis. Pearson chi-squared test and Fisher’s exact test were used to analyse the categorical parameters. Shapiro-Wilk tests of normality were run for continuous parameters. Normally distributed parameters were analysed using Independent sample t-test and reported in mean and standard deviation or percentages. Non-normally distributed parameters were analysed by Mann-Whitney U Test non-parametric test and reported in median and interquartile range. 0.05 was the cut-off point of statistical significance applied. Pearson correlation analysis was used to determine the correlations and relationships between variables.

Results

There were 151 patients in total with 107 patients with Lenke type 1 curves and 44 patients with Lenke type 2 curves. The median age was 15.0 years. There were 21 males (13.9%) and 130 females (86.1%). When Lenke 1 and Lenke 2 curves were compared, the pre-operative PT Cobb angle, MT Cobb angle, PT side bending Cobb angle, MT side bending Cobb angle, PT flexibility rate and MT flexibility rate had shown significant differences. Furthermore, patients with Lenke 2 curves had more severe medial shoulder imbalance (median T1 Tilt of 3.5°) and neck imbalance (mean Cervical Axis of 3.4 ± 3.6°) pre-operatively. (Table 2)

Table 2.

Patient Demographics and Pre-operative Parameters.

Parameters	Lenke 1 (n = 107)	Lenke 2 (n = 44)	Overall (n = 151)	P value
Age (years)	15.0 (5.0)^a	16.0 (6.0)^a	15.0 (5.0)^a	0.514
Gender (n (%))	12 M (11.2%)	9 M (20.5%)	21 M (13.9%)	0.183
Gender (n (%))	95 F (88.8%)	35 F (79.5%)	130 F (86.1%)	0.183
PT cobb angle (°)	29.0 ± 7.7^b	44.5 (12.0)^a	33.0 (14.0)^a	0.000
MT cobb angle (°)	59.0 (17.0)^a	76.5 (24.0)^a	63.0 (22.0)^a	<0.001
PT side bending cobb angle (°)	16.0 (9.0)^a	29.5 (9.0)^a	19.0 (13.0)^a	0.000
MT side bending cobb angle (°)	23.0 (16.0)^a	38.5 (20.0)^a	27.0 (21.0)^a	<0.001
PT flexibility rate (%)	47.7 (19.4)^a	29.9±10.4^b	42.9 (19.9)^a	<0.001
MT flexibility rate (%)	60.4 ± 15.1^b	48.5 ± 13.3^b	56.9 ± 15.5^b	<0.001
T1 tilt (°)	0.0 (8.0)^a	3.5 (7.0)^a	1.0 (8.0)^a	<0.001
Clavicle angle (°)	−1.0 (3.0)^a	−0.4 ± 2.7^b	−1.0 (4.0)^a	0.026
Cervical axis (°)	1.5 ± 3.6^b	3.4 ± 3.6^b	2.0 (5.0)^a	0.004

^aNon-normally distributed data is analyzed using Mann-Whitney U Test and reported in median (interquartile range).

^bNormally distributed parameter was analyzed using Paired t test for statistical significance and reported in mean ± SD.

The post-operative PT and MT Cobb angles had shown statistically significant differences when comparing Lenke 1 and Lenke 2 curves. The mean post-operative PT Cobb angle was 16.0 ± 6.0° and 27.1 ± 8.2° in Lenke 1 and Lenke 2, respectively, whereas the mean post-operative MT Cobb angle was 20.0° and 30.0°, respectively. Unlike PT correction rate, MT correction rate was significantly different between Lenke 1 and Lenke 2 curves (64.6 ± 9.9% vs 60.4 ± 8.7%, respectively, P = 0.016). Besides that, post-operative T1 Tilt was also statistically significant between both groups (1.0° and 3.3°, respectively, P < 0.001) (Table 3).

Table 3.

Patient Post-operative Parameters.

Parameters	Lenke 1 (n = 107)	Lenke 2 (n = 44)	Overall (n = 151)	P value
PT cobb angle (°)	16.0 ± 6.0^a	27.1 ± 8.2^a	18.0 (11.0)^b	<0.001
MT cobb angle (°)	20.0 (9.0)^b	30.0 (16.0)^b	22.0 (13.0)^b	<0.001
PT correction rate (%)	44.8 ± 16.1^a	41.4 ± 14.7^a	43.8 ± 15.8^a	0.234
MT correction rate (%)	64.6 ± 9.9^a	60.4 ± 8.7^a	63.4 ± 9.7^a	0.016
PT SBCI	0.9 (0.5)^b	1.5 (0.7)^b	1.0 (0.6)^b	<0.001
MT SBCI	1.1 (0.3)^b	1.3 (0.5)^b	1.1 (0.4)^b	<0.001
T1 tilt (°)	1.0 (4.0)^b	3.3 ± 2.7^a	2.0 (3.0)^b	<0.001
Clavicle angle (°)	0.0 (3.0)^b	−0.5 (2.0)^b	0.0 (2.0)^b	0.061
Cervical axis (°)	0.0 (3.0)^b	0.0 (3.0)^b	0.0 (3.0)^b	0.649

^bNon-normally distributed data is analyzed using Mann-Whitney U Test and reported in median (interquartile range).

^aNormally distributed parameter was analyzed using Paired t test for statistical significance and reported in mean ± SD.

Overall, the prevalence of medial shoulder, neck and lateral shoulder imbalance was 27.8%, 23.2% and 9.9%, respectively. 24 (22.4%) patients with Lenke 1 curves had medial shoulder imbalance compared to 18 (40.9%) patients with Lenke 2 curves. In terms of lateral shoulder and neck balance, the prevalence of radiological imbalance was comparable between the 2 groups. There were 15 patients (9.9%) with lateral shoulder imbalance whereas 35 patients (23.2%) had neck imbalance post-operatively (Table 4).

Table 4.

Post-operative Medial Shoulder (T1 Tilt), Lateral Shoulder (Clavicle Angle) and Neck (Cervical Axis) Balance.

Parameters	Status	Lenke 1 (n = 107)	Lenke 2 (n = 44)	Overall (n = 151)	P value
T1 tilt [median (IQR)]^a	Balanced	83 (77.6%)	26 (59.1%)	109 (72.2%)	0.069
T1 tilt [median (IQR)]^a	Imbalanced	24 (22.4%)	18 (40.9%)	42 (27.8%)	0.004
Clavicle angle [median (IQR)]^a	Balanced	96 (89.7%)	40 (90.9%)	136 (90.1%)	0.210
Clavicle angle [median (IQR)]^a	Imbalanced	11 (10.3%)	4 (9.1%)	15 (9.9%)	0.078
Cervical Axis [median (IQR)]^a	Balanced	86 (80.4%)	30 (68.2%)	116 (76.8%)	0.938
Cervical Axis [median (IQR)]^a	Imbalanced	21 (19.6%)	14 (31.8%)	35 (23.2%)	0.198

^aNon-normally distributed data is analysed using Mann-Whitney U Test and reported in median (interquartile range).

For the whole cohort, the mean Relative Correction Rate (RCR) had a significant association with lateral shoulder and neck balance. The RCR for patients with lateral shoulder imbalance was averaged at 60.3% while those with neck imbalance was averaged at 62.2%. There was a significant association between RRCA and medial shoulder, lateral shoulder and neck balance. The RRCA in those with balanced medial shoulder was 75.0% compared to 92.1% among patients with medial shoulder imbalance; 77.8% in balanced lateral shoulder compared to 97.5% among those with lateral shoulder imbalance; 76.0% in patients with balanced neck and 83.8% in patients with neck imbalance. Post-operative UIV tilt angle had a significant association with medial shoulder imbalance (median of −4.0° vs −1.5° comparing patients with balanced and imbalanced medial shoulder) (Table 5).

Table 5.

RCR and RRCA in the Balanced and Imbalanced Medial Shoulder (T1 Tilt), Lateral Shoulder (Clavicle Angle) and Neck (Cervical Axis).

Parameter	Medial shoulder (T1 Tilt)			Lateral shoulder (Clavicle Angle)			Neck (Cervical Axis)
Status	Balanced	Imbalanced	P value	Balanced	Imbalanced	P value	Balanced	Imbalanced	P value
RCR (%) [median (IQR)]^a	73.8 (26.5)	68.0 (19.5)	0.150	71.7 (26.1)	60.3 (18.5)	0.021	72.6 (25.6)	62.2 (22.1)	0.021
RRCA (%) [median (IQR)]^a	75.0 (24.2)	92.1 (30.4)	0.001	77.8 (30.6)	97.5 (29.4)	0.016	76.0 (32.4)	83.8 (31.3)	0.037
Postoperative UIV tilt angle (°) [median (IQR)]^a	−4.0 (3.0)	−1.5 (9.0)	0.017	−3.0 (4.0)	−1.0 (7.0)	0.078	−3.0 (4.0)	−3.0 (6.0)	0.515

^aNon-normally distributed data is analyzed using Mann-Whitney U Test and reported in median (interquartile range).

Post-operative T1 Tilt had a significant correlation (P < 0.001) with RRCA with a moderate strength of correlation (R = 0.628) (Figure 2). RCR was found to have a significant correlation with post-operative T1 Tilt (P = 0.009) and post-operative Cervical Axis (P < 0.001). However, all parameters had shown a weak correlation to RCR (Figure 3). There was a significant and strong correlation (r = 0.708) between post-operative UIV tilt angle and post-operative T1 Tilt (P < 0.001). (Figure 4). Unlike Relative Correction Rate (RCR), Relative Residual Cobb Angle (RRCA) had shown a statistically significant correlation (P < 0.001) to post-operative UIV tilt angle with an r-value of 0.406 (Figure 5).

Figure 2.

Correlation between Relative Residual Cobb Angle (RRCA) and post-operative T1 Tilt, Clavicle Angle and Cervical Axis in Lenke 1 and Lenke 2 patients.

Figure 3.

Correlation between Relative Correction Rate (RCR) and post-operative TI Tilt, Clavicle Angle and Cervical Axis in Lenke 1 and Lenke 2 patients.

Figure 4.

Correlation between post-operative UIV tilt angle vs post-operative T1 Tilt, Clavicle Angle and Cervical Axis in Lenke 1 and Lenke 2 patients.

Figure 5.

Correlation between Relative Residual Cobb Angle (RRCA) and Relative Correction Rate (RCR) vs post-operative UIV tilt angle in Lenke 1 and Lenke 2 patients.

The correlation analysis among the 3 factors (T1 tilt, clavicle angle, and cervical axis) showed no significant difference: T1 tilt and clavicle angle (r = 0.140, P = 0.087), T1 tilt and cervical axis (r = 0.114, P = 0.162), and cervical axis and clavicle angle (r = 0.132, P = 0.105).

The ROC curve analysis for factors that had a significant association with 3 parameters; T1 tilt, clavicle angle and cervical axis which were summarized in Table 6. For T1 tilt, the cut-off for RCCA for T1 tilt which was 84.7%, whereas, the cut-off for postop UIV for T1 tilt was −1.5°. In contrast, for clavicle Angle, the cut-off for RCR was 61.6% compared to the cut-off for RRCA which was 87.7%. For cervical axis, the cut-off for RCR was 66.8%, whereas the cut-off for RRCA was 85.3%.

Table 6.

ROC Analysis of RCR and RRCA for Medial Shoulder (T1 Tilt), Lateral Shoulder (Clavicle Angle) and Neck (Cervical Axis) Balance.

Parameter	Cutoff point (Threshold)	Sensitivity	1-Specificity	Specificity	Youden’s J
RRCA/T1 tilt	84.7	0.333	0.619	0.381	−0.286
Postoperative UIV/T1 tilt	−1.5	0.375	0.381	0.619	−0.006
RCR/clavicle angle	61.6	0.684	0.333	0.667	0.351
RRCA/clavicle angle	87.7	0.309	0.6	0.4	−0.291
RCR/cervical axis	66.8	0.629	0.457	0.543	0.172
RRCA/cervical axis	85.3	0.379	0.486	0.514	−0.107

Youden’s J is calculated using the formula: Sensitivity + Specificity - 1.

However, when we evaluated the discriminative value of these cut-off values (Table 7), there were only 2 parameters which had the acceptable discrimination which is the cut-off for RCR for clavicle angle which was 61.6% with area under the curve of 0.681 (Figure 6), and the cut-off for RCR for cervical axis which had cut-off point of 66.8% and area under the curve of 0.629 (Figure 7).

Table 7.

Cut-Off Point of RCR and RRCA for Medial Shoulder (T1 Tilt), Lateral Shoulder (Clavicle Angle) and Neck (Cervical Axis) Balance.

Parameter	Cut-off point (Threshold)	AUC	Discrimination
RRCA/T1 tilt	84.7	0.329	No discrimination
Postoperative UIV/T1 tilt	−1.5	0.375	No discrimination
RCR/clavicle angle	61.6	0.681	Acceptable discrimination
RRCA/clavicle angle	87.7	0.310	No discrimination
RCR/cervical axis	66.8	0.629	Acceptable discrimination
RRCA/cervical axis	85.3	0.383	No discrimination

AUC = Area Under Curve.

Figure 6.

Receiver-operating-characteristic (ROC) of RCR/clavicle angle analysis showed acceptable discrimination of area under the curve (AUC) of 0.681 (P = 0.68). The cut-off value 61.6% had a sensitivity of 0.684 and specificity of 0.667.

Figure 7.

Receiver-operating-characteristic (ROC) of RCR/cervical axis analysis showed acceptable discrimination of area under the curve (AUC) of 0.629 (P = 0.50). The cut-off value 66.8% had a sensitivity of 0.629 and specificity of 0.543.

Discussion

Despite various recommendations and strategies to achieve post-operative shoulder balance, PSI is still a common occurrence among AIS patients.^18-21 The occurrence of PSI post-PSF surgery is 1 of the factors that contributes to poor satisfaction among AIS patients, especially in those with left shoulder elevation pre-operatively.^22,23 Bram et al revealed that severe PSI could be predicted by pre-operative PT Cobb angle >34.5° in which 44.4% of patients would develop severe PSI after 2 years.²³ Earlier studies had focused mainly on UIV selection and advocated the fusion of the proximal thoracic curve to reduce the incidence of PSI.^24,25

However, recent studies had shown that PSI might stem from over-correction of the MT curve and under-correction of the PT curve. Sielatycki et al reported that 59% of patients developed shoulder imbalance when the correction of the MT curve was more than 54% while the correction of the PT curve was less than 52%.²⁶ Berlin et al had also concluded that MT correction of more than 64.9% and lumbar correction of more than 93.2% had resulted in PSI in 51.9% of AIS patients with Lenke 2 curves.²⁷ This was further supported by Landrum et al who had proposed the idea of Thoracic Curve Correction Ratio (TCCR) which was the PT/MT correction rate divided by pre-operative PT/MT Cobb angle.²⁸ TCCR had a significant correlation with post-operative T1 Tilt.²⁸ In a retrospective study involving 104 Lenke 1 patients, Banno et al found that PSI was associated with a higher MT correction.²⁹

Therefore, we postulated that relative correction of the PT curve to the MT curve would ultimately affect the position of the upper instrumented vertebrae, and this could be associated with the UIV tilt angle. The concept of UIV tilt angle was introduced in 2016.¹¹ Patients were reported to have lower risk of clinical neck tilt with medial shoulder imbalance if their UIV tilt angle was within their UIV’s reserve motion.¹¹ Besides that, they also discovered that the difference in UIV tilt angle (UIVDiff) had a strong correlation with medial shoulder balance (represented by T1 Tilt) and a moderate correlation with neck balance (represented by Cervical Axis).¹¹ Subsequent published literature had shown that post-operative UIV Tilt Angle was an independent variable that significantly correlated with post-operative T1 Tilt (r = 0.76) and Cervical Axis measurement (r = 0.42).⁹

Patients with positive post-operative UIV tilt angle were also at higher risk (14.9x) of having medial shoulder imbalance.¹⁰ Among 136 AIS patients with Lenke 1 and 2 curves, there were 51.6% of patients with positive post-operative UIV tilt angle and medial shoulder imbalance.¹⁰ Zhang et al introduced a formula to predict the ideal post-operative UIV tilt: ideal post-operative UIV tilt = pre-operative UIV tilt - the flexibility between T1 and UIV – pre-operative CA.³⁰ The importance of UIV tilt angle was also investigated by Deng et al who reported that 69.2% of Lenke 2 AIS patients suffered from medial shoulder imbalance post-operatively even though lateral shoulder balance had been achieved.³¹ They concluded that the significant predictors for medial shoulder imbalance were post-operative UIV tilt angle (P < 0.001), PT correction rate (P < 0.001), ratio of PTC/MTC (P < 0.001) and T2 vertebra rotation ratio (P = 0.008).³¹ To date, no studies have been carried out to investigate the association between PT/MT relative curve correction and post-operative UIV tilt angle.

In this study, the relative curve correction of the PT/MT segment based on the Relative Correction Rate (RCR) and the Relative Residual Cobb angle (RRCA) was assessed. Higher RCR would indicate a higher correction rate of PT curve than MT curve whereas higher RRCA would indicate a higher post-operative PT Cobb angle than MT Cobb angle. The results of this study showed that RRCA had a stronger correlation with medial shoulder balance compared to RCR. UIV tilt angle and medial shoulder balance had the strongest correlation (r = 0.708). There was also a significant correlation between RRCA and UIV tilt angle (r = 0.406) but not with RCR. This is the first study to elucidate the association between RRCA, RCR and post-operative UIV tilt angle. From a practical point of view, surgeons must be cognizant of the importance of PT segment correction as this segment is inherently stiffer than the MT segment.^32-35

Therefore, relative under-correction of the PT segment in relation to the MT segment is common and this could lead to medial PSI. RRCA illustrates a concept whereby PT correction should be optimised to avoid PSI. However, the post-operative UIV tilt angle offers the surgeons an objective target and this can be determined pre-operatively. The formula to calculate the optimal UIV tilt angle based on the supervised supine side bending films; Optimal UIV tilt angle = - ((RSB angle + LSB angle) / 2).¹¹ Furthermore, this target can be confirmed intra-operatively using the crossbar method.³⁶ As it is a highly objective measurement, understanding this radiological parameter is imperative to prevent medial PSI.

This study however has several limitations. The assessment of the medial, neck and lateral shoulder balance are represented by radiological parameters. Additional clinical assessment could strengthen the study despite these radiological parameters had been previously validated.^7,16,37-39 The final assessment of the radiological outcome was carried out at minimum 2 years follow-up. Longer follow-up duration could reflect the final outcome more accurately. Although the concept of RRCA is important, it is difficult to predict it pre-operatively and to quantify and execute during the surgery. More studies would be needed to validate the concept of RRCA and its association with PSI.

Conclusion

In conclusion, the prevalence of post-operative medial shoulder, neck and lateral shoulder imbalance was 27.8%, 23.2% and 9.9%, respectively. RRCA and post-operative UIV tilt angle had shown a significant correlation with medial shoulder balance, but post-operative UIV tilt angle had shown the strongest strength of correlation (r = 0.708). RRCA had shown a statistically significant correlation with post-operative UIV tilt angle with a r-value of 0.406.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Chris Yin Wei Chan

Chee Kidd Chiu

Mun Keong Kwan

References

Yagi

Takemitsu

Machida

. Clavicle chest cage angle difference (CCAD): a novel predictor of postoperative shoulder imbalance in patients with adolescent idiopathic scoliosis. Spine. 2013;38:E705-E712. doi:10.1097/BRS.0b013e31828f0bcd

Yang

Zhao

Yang

. Postoperative shoulder imbalance in adolescent idiopathic scoliosis: risk factors and predictive index. Eur Spine J. 2019;28:1331-1341. doi:10.1007/s00586-019-05933-2

Zhang

Feng

Yang

. Incidence and risk factors for postoperative shoulder imbalance in scoliosis: a systematic review and meta-analysis. Eur Spine J. 2018;27:358-369. doi:10.1007/s00586-017-5289-y

Lee

Hwang

Lim

Lee

Cho

. A retrospective study to reveal factors associated with postoperative shoulder imbalance in patients with adolescent idiopathic scoliosis with double thoracic curve. J Neurosurg Pediatr. 2016;25:744-752. doi:10.3171/2016.6.Peds16162

Matsumoto

Watanabe

Kawakami

, et al. Postoperative shoulder imbalance in Lenke Type 1A adolescent idiopathic scoliosis and related factors. BMC Muscoskelet Disord. 2014;15:366. doi:10.1186/1471-2474-15-366

Yuan

Fan

Hai

Zang

. What is the impact of scoliotic correction on postoperative shoulder imbalance in severe and rigid scoliosis. BMC Muscoskelet Disord. 2021;22:868. doi:10.1186/s12891-021-04763-y

Bagó

Carrera

March

Villanueva

. Four radiological measures to estimate shoulder balance in scoliosis. J Pediatr Orthop B. 1996;5:31-34. doi:10.1097/01202412-199605010-00006

Zhang

Wang

, et al. Assessment of neck imbalance in adolescent idiopathic scoliosis patients: a cross-section study based on body image of 115 patients with main or double thoracic curve. Orthop Surg. 2023;15:572-578. doi:10.1111/os.13583

Chan

CYW

Chiu

Ler

, et al. Upper instrumented vertebrae (UIV) tilt angle is an important postoperative radiological parameter that correlates with postoperative neck and medial shoulder imbalance. Spine. 2018;43:E1143-E1151. doi:10.1097/brs.0000000000002645

10.

Kwan

Lee

Ch’ng

Chung

Chiu

Chan

CYW

. A positive (+ve) postoperative upper instrumented vertebra tilt angle (≥0°) significantly increases the risk of medial shoulder and neck imbalance in Lenke 1 and 2 adolescent idiopathic scoliosis patients. Spine. 2020;45:E694-E703. doi:10.1097/brs.0000000000003407

11.

Kwan

Chan

. Is there an optimal upper instrumented vertebra (UIV) tilt angle to prevent post-operative shoulder imbalance and neck tilt in Lenke 1 and 2 adolescent idiopathic scoliosis (AIS) patients? Eur Spine J. 2016;25:3065-3074. doi:10.1007/s00586-016-4529-x

12.

Ono

Bastrom

Newton

. Defining 2 components of shoulder imbalance: clavicle tilt and trapezial prominence. Spine. 2012;37:E1511-E1516. doi:10.1097/BRS.0b013e31826e2bbb

13.

Lee

Ch’ng

Wong

, et al. Patients’ perception and satisfaction on neck and shoulder imbalance in adolescent idiopathic scoliosis. Glob Spine J. 2023;13:752-763. doi:10.1177/21925682211007795

14.

Chung

Chiu

Goh

Chan

CYW

Kwan

. How common is medial and lateral shoulder discordance in Lenke 1 and 2 curves? a preoperative analysis of medial and lateral shoulder balance among 151 Lenke 1 and 2 adolescent idiopathic scoliosis patients. Spine. 2019;44:E480-E486. doi:10.1097/brs.0000000000002890

15.

Pesenti

Blondel

Peltier

Choufani

Bollini

Jouve

. Interest of T1 parameters for sagittal alignment evaluation of adolescent idiopathic scoliosis patients. Eur Spine J. 2016;25:424-429. doi:10.1007/s00586-015-4244-z

16.

Kuklo

Lenke

Graham

, 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. 2002;27:2013-2020. doi:10.1097/00007632-200209150-00009

17.

Kwan

Wong

Lee

Chan

. Is neck tilt and shoulder imbalance the same phenomenon? A prospective analysis of 89 adolescent idiopathic scoliosis patients (Lenke type 1 and 2). Eur Spine J. 2016;25:401-408. doi:10.1007/s00586-015-4016-9

18.

Rose

Lenke

. Classification of operative adolescent idiopathic scoliosis: treatment guidelines. Orthop Clin N Am. 2007;38:521-529. doi:10.1016/j.ocl.2007.06.001

19.

Ilharreborde

Even

Lefevre

, et al. How to determine the upper level of instrumentation in Lenke types 1 and 2 adolescent idiopathic scoliosis: a prospective study of 132 patients. J Pediatr Orthop. 2008;28:733-739. doi:10.1097/BPO.0b013e318185a36b

20.

Bjerke

Cheung

Shifflett

Iyer

Derman

Cunningham

. Do current recommendations for upper instrumented vertebra predict shoulder imbalance? An attempted validation of level selection for adolescent idiopathic scoliosis. HSS J. 2015;11:216-222. doi:10.1007/s11420-015-9451-y

21.

Trobisch

Ducoffe

Lonner

Errico

. Choosing fusion levels in adolescent idiopathic scoliosis. J Am Acad Orthop Surg. 2013;21:519-528. doi:10.5435/jaaos-21-09-519

22.

Menon

Pillay

Tahasildar

. Post-operative shoulder imbalance in adolescent idiopathic scoliosis: a study of clinical photographs. Scoliosis. 2015;10:31. doi:10.1186/s13013-015-0055-6

23.

Bram

Mehta

Flynn

, et al. Sinister! The high pre-op left shoulder is less likely to be radiographically balanced at 2 years post-op. Spine Deform. 2021;9:451-460. doi:10.1007/s43390-020-00236-9

24.

Suk

Kim

Lee

, et al. Indications of proximal thoracic curve fusion in thoracic adolescent idiopathic scoliosis: recognition and treatment of double thoracic curve pattern in adolescent idiopathic scoliosis treated with segmental instrumentation. Spine. 2000;25:2342-2349. doi:10.1097/00007632-200009150-00012

25.

Jiang

Qian

Qiu

Wang

Zhu

. Full fusion of proximal thoracic curve helps to prevent postoperative cervical tilt in Lenke type 2 adolescent idiopathic scoliosis patients with right-elevated shoulder. BMC Muscoskelet Disord. 2017;18:362. doi:10.1186/s12891-017-1730-y

26.

Sielatycki

Cerpa

Beauchamp

, et al. The amount of relative curve correction is more important than upper instrumented vertebra selection for ensuring postoperative shoulder balance in Lenke type 1 and type 2 adolescent idiopathic scoliosis. Spine. 2019;44:E1031-E1037. doi:10.1097/brs.0000000000003088

27.

Berlin

Quante

Thomsen

Koeszegvary

Pecsi

Halm

. How can postoperative shoulder imbalance be prevented in adolescent idiopathic scoliosis type 2? Acta Orthop Belg. 2022;88:457-466. doi:10.52628/88.3.9466

28.

Landrum

Milby

Yaszay

, et al. Thoracic curve correction ratio: an objective measure to guide against overcorrection of a main thoracic curve in the setting of a structural proximal thoracic curve. J Clin Med. 2022;11:1545. doi:10.3390/jcm11061545

29.

Banno

Yamato

Hasegawa

, et al. Preoperative less right shoulder elevation had a higher risk of postoperative shoulder imbalance when main thoracic curve shows higher correction regardless of the upper instrumented vertebra level for patients with adolescent idiopathic scoliosis Lenke type 1. Asian Spine J. 2023;17:166-175. doi:10.31616/asj.2022.0020

30.

Zhang

, et al. A novel method for predicting ideal postoperative upper instrumented vertebra tilt to prevent lateral shoulder imbalance after scoliosis correction surgery. J Personalized Med. 2023;13:393. doi:10.3390/jpm13030393

31.

Deng

Wang

, et al. Incidence and risk factors of postoperative medial shoulder imbalance in Lenke Type 2 adolescent idiopathic scoliosis with lateral shoulder balance. BMC Muscoskelet Disord. 2022;23:947. doi:10.1186/s12891-022-05882-w

32.

Kuklo

Lenke

Won

, et al. Spontaneous proximal thoracic curve correction after isolated fusion of the main thoracic curve in adolescent idiopathic scoliosis. Spine. 2001;26:1966-1975. doi:10.1097/00007632-200109150-00006

33.

Kirk

Kuklo

Polly

Jr . Traction versus side-bending radiographs: is the proximal thoracic curve the stiffer curve in double thoracic curves? Am J Orthoped. 2003;32:284-288.

34.

Lenke

Bridwell

O'Brien

Baldus

Blanke

. Recognition and treatment of the proximal thoracic curve in adolescent idiopathic scoliosis treated with Cotrel-Dubousset instrumentation. Spine. 1994;19:1589-1597. doi:10.1097/00007632-199407001-00007

35.

Chan

CYW

Chiu

Kwan

. Assessing the flexibility of the proximal thoracic segments above the “potential upper instrumented vertebra” using the cervical supine side bending radiographs in Lenke 1 and 2 curves for adolescent idiopathic scoliosis patients. Spine. 2016;41:E973-e980. doi:10.1097/brs.0000000000001516

36.

Kwan

Chiu

Goh

, et al. The reliability of intraoperative crossbar technique in determining the upper instrumented vertebra (UIV) tilt angle for adolescent idiopathic scoliosis (AIS) undergoing posterior spinal fusion. Clin Spine Surg. 2019;32:256-262. doi:10.1097/bsd.0000000000000769

37.

Akel

Pekmezci

Hayran

, et al. Evaluation of shoulder balance in the normal adolescent population and its correlation with radiological parameters. Eur Spine J. 2008;17:348-354. doi:10.1007/s00586-007-0546-0

38.

Drummond

Ranallo

Lonstein

Brooks

Cameron

. Radiation hazards in scoliosis management. Spine. 1983;8:741-748. doi:10.1097/00007632-198310000-00010

39.

Qiu

, et al. Discrepancy between radiographic shoulder balance and cosmetic shoulder balance in adolescent idiopathic scoliosis patients with double thoracic curve. Eur Spine J. 2009;18:45-51. doi:10.1007/s00586-008-0833-4