Effect of Sacral Slanting on Coronal Balance and Residual Curve After Lumbar Curve Correction in Adolescent Idiopathic Scoliosis: A 5-Year Follow-up Study

Abstract

Study Design

Retrospective cohort study.

Objectives

Sacral slanting is defined as the angle between the horizontal line and the upper endplate of the sacrum in the coronal plane. This study evaluated changes in disc wedge angle (DWA) below the lowest instrumented vertebra (LIV) and coronal balance over a 5-year follow-up in patients with adolescent idiopathic scoliosis (AIS) with structural lumbar curve exhibiting sacral slanting.

Methods

The study included 118 patients with AIS and structural lumbar curves who underwent posterior fusion to L3 or L4, with a minimum 5-year follow-up. Radiological parameters were evaluated preoperatively and at 5 years and compared between LIV-L3 and LIV-L4 groups, with subgroup analyses for high (≥5°) vs low (<5°) sacral slanting within each group.

Results

Preoperative radiographs showed no significant differences between groups, except for the Cobb angle. In the LIV-L3 group, patients with sacral slanting ≥5° presented greater DWA at 5 years, whereas those with slanting <5° demonstrated a gradual decrease in DWA. In the LIV-L4 group, patients with sacral slanting ≥5° showed a progressive increase in C7-cervical sagittal vertical line, which was significantly greater at 5 years. Scoliosis Research Society-22 scores were similar between high and low slanting groups at the 5-year follow-up.

Conclusions

In patients with AIS and structural lumbar curve, increased preoperative sacral slanting was associated with an increased DWA below LIV when selecting LIV-L3. However, this increase does not progress over time. Conversely, selecting LIV at L4 in patients with high sacral slanting was associated with progressive coronal decompensation.

Keywords

adolescent idiopathic scoliosis structural lumbar curve sacral slanting disc wedge angle coronal decompensation

Introduction

Sacral slanting is commonly found in patients with adolescent idiopathic scoliosis (AIS).¹ It is defined as the angle between the horizontal line and the upper end plate of the sacrum in the coronal plane (Figure 1). Because the sacrum serves as the foundation of the whole spine and the lever arm length from the sacrum to T1 is long, sacral slanting can affect shoulder height balance² and contribute to coronal decompensation.¹ Given that AIS involves three-dimensional deformities, sacral slanting holds clinical importance at the distal fusion level. This corresponds to the concept of shoulder height difference at the proximal fusion level.^3-5

Figure 1.

Measurement of sacral slanting in patients with adolescent idiopathic scoliosis

During correction of a structural lumbar curve, distal adding-on may occur, in which the disc wedge angle (DWA) below the lowest instrumented vertebra (LIV) is widened. Distal adding-on is defined as a postoperative increase in DWA of >5° below the LIV.⁶ It refers to a postoperative loss of correction, characterized by progressive deviation and rotation of the lumbar spine, along with wedging of the disc space distal to the LIV. Observed in approximately 2-13% of patients, distal adding-on may necessitate revision surgery^7-9 and may have detrimental effects on the long-term health of the lumbar spine.

Sacral slanting is a key factor in determining the distal fusion level when planning the correction of lumbar curves. In the correction of AIS, the decision of whether to stop at L3 or extend down to L4 is critical, as preserving mobile segments is essential. According to Lee et al,¹ when sacral slanting is high and the fusion extends to L3, three compensatory segments remain, which may lead to a widening of the DWA below the LIV. Conversely, extending fusion to L4 leaves only two mobile segments, potentially resulting in coronal decompensation rather than widening of the DWA below the LIV. Lee et al proposed this mechanism based on the results of their 2-year follow-up study.

However, it remains unclear whether widening of the DWA below the LIV or coronal decompensation has a more detrimental impact. Furthermore, knowledge regarding long-term follow-up data exceeding 5 years and clinical outcomes related to the sacral slanting is currently lacking. Therefore, we investigated whether widening of the DWA below the LIV and coronal decompensation progressed over time. The purpose of the present study was to evaluate the long-term effects on the distal remaining segments below the LIV and coronal decompensation in patients with AIS with sacral slanting following fusion of the lumbar curve over a 5-year follow-up period.

Materials and Methods

Study Design and Population

The study protocol was approved by the relevant institutional review board (no. 2024-1138). The requirement for informed consent was waived owing to the retrospective nature of the study. We analyzed patients who underwent deformity correction and posterior fusion using a pedicle screw for AIS exhibiting a structural lumbar curve¹⁰ between 2010 and 2017. The inclusion criteria were the LIV of L3 or L4, follow-up data of >5 years, and age between 12 and 18 years at the time of the operation. Patients who had undergone revisional surgery, had a leg length discrepancy >2 cm on standing Bell-Thompson radiographs, and who did not undergo regular follow-up for 5 years were excluded from the analysis. At our institution, routine follow-up radiographs for patients with AIS were obtained at 1 month and then at 1, 2, and 5 years postoperatively. All major structural curves were determined based on the largest Cobb angle measurement, and minor structural curves identified as Cobb angle ≥25° on preoperative coronal bending radiographs were included in the fusion levels. The uppermost instrumented vertebra (UIV) was selected as the proximal end vertebra or one level proximal to the end vertebra. We typically chose L3 as the LIV when the last touching vertebra was L3, whereas L4 was selected when the last touching vertebra was L4 or L5.^11-13

Patients were categorized into the LIV-L3 and LIV-L4 groups for comparative analysis. As in a previous study,¹ each group was further subdivided into the high slanting (presacral slanting ≥5°) and low slanting (presacral slanting <5°) groups for subgroup analyses. We analyzed the changes in radiological variables and clinical outcomes over a 5-year follow-up. Additionally, risk factors associated with the adding-on phenomenon and coronal decompensation were examined.

Variables

Radiographical assessments were conducted preoperatively and postoperatively after 1 and 6 months, and at 1, 2, and 5 years using whole spine radiographs. Radiographic measurements were performed by two experienced spine fellowship-trained surgeons with over 5 years of experience. Sacral slanting was defined as the angle between the horizontal line and the upper endplate of the sacrum in the coronal plane. The scoliosis angle was assessed using the Cobb angle of the proximal thoracic, main thoracic, and thoracolumbar curves. DWA below the LIV was defined as the angle between the lower end plate of the LIV and the upper end plate of the vertebra below the LIV. LIV+1 tilt was defined as the tilted angle of the upper endplate of the vertebra just inferior to the LIV. The C7–central sacral vertical line (CSVL) coronal offset (C7–CSVL) was defined as the horizontal distance between the C7 plumb line and the CSVL on the coronal plane. Spinopelvic parameters including pelvic obliquity, pelvic incidence (PI), sacral slope (SS), and pelvic tilt (PT) were measured. In addition, sagittal parameters were assessed as follows: thoracic kyphosis, defined as the Cobb angle between the upper endplate of T1 and the lower endplate of T12; lumbar lordosis, defined as the Cobb angle between the upper endplates of L1 and S1; and the sagittal vertical axis (SVA), defined as the horizontal distance between the posterosuperior corner of S1 and the vertical line drawn from the center of the C7 vertebral body.

The adding-on phenomenon was defined as an increase in DWA of >5° below the LIV compared with preoperative measurements.^6,14 Coronal decompensation was defined as an increase of ≥15 mm in C7-CSVL.^15,16 Scoliosis Research Society-22 (SRS-22) questionnaire data were reviewed preoperatively and at the 5-year follow-up. Patients completed paper-based questionnaires during their outpatient visits at each corresponding time point.

Statistical Analysis

All data are expressed as mean ± standard deviation or percentile values. The demographic and radiographical results were analyzed by categorizing the patients into two groups. Normality was assessed using the Shapiro–Wilk test. Continuous variables were compared using the independent t-test when data were normally distributed, and nonparametric analyses (the Mann–Whitney U test for two-group comparisons) were performed when the assumption of normality was not satisfied. Categorical variables were analyzed using the chi-squared test. The paired t-test was used to analyze differences in the same variable over time. Furthermore, multivariate logistic regression analyses were conducted to identify the factors associated with the adding-on phenomenon and coronal decompensation. Two orthopedic surgeons performed radiographic assessments. The intraclass correlation coefficient (ICC) was used to determine interobserver reliability. The ICC for interobserver reliability was 0.849 for the measurement of sacral slanting, 0.900 for DWA below the LIV, and 0.972 for C7-CSVL, as well as 0.872 for PI, 0.861 for SS, 0.889 for SVA, 0.835 for TK, and 0.903 for LL. A P-value <.05 was considered to be statistically significant. All statistical analyses were conducted using SPSS version 28.0 (SPSS Inc, Chicago, IL, USA).

Results

Study Population and Characteristics

The study enrolled 182 patients with AIS of Lenke types 3, 4, 5, and 6 with a lumbar modifier C, all of whom had an LIV at L3 or L4. Consequently, 118 patients who met the inclusion criteria were selected for the final analysis (follow-up rate, 64.8% [118/182]). Among the final cohort, 67 patients (56.8%) were classified into the high slanting group and 51 patients (43.2%) into the low slanting group. The LIV-L3 group included 82 (69.5% [82/118]) patients, and the LIV-L4 group comprised 36 (30.5% [36/118]) patients (Figure 2). None of the patients required revision surgery during follow-up. Notably, implant failure occurred in one patient (0.8% [1/118]) in the high slanting group, in whom left rod dislodgement occurred 2 months after a T3-L3 deformity correction, leading to the adding-on phenomenon.

Figure 2.

Patient selection process

No difference in demographics and clinical outcomes of patients was observed between the LIV-L3 and LIV-L4 groups. Regarding radiological outcomes, the Cobb angles of the proximal, main thoracic, and lumbar curves were significantly greater in the L4 group (Table 1). In addition, there were no significant differences in preoperative demographic factors or coronal radiologic parameters between the patients lost to follow-up and those included in the analysis (Table 2).

Table 1.

Baseline Characteristics of Patients According to LIV Difference

Characteristic	Overall (n = 118)	LIV-L3 (n = 82)	LIV-L4 (n = 36)	P-value (L3 vs L4)
Sex
Male	15 (12.7%)	11 (13.1%)	4 (11.1%)	.729
Female	103 (87.3%)	71 (86.6%)	32 (88.9%)	.729
Age, yr	15.3 $\pm 3.5$	$15.4 \pm$ 3.6	15.2 $\pm 3.1$	.788
Body mass index	19.8 $\pm 2.6$	$20.1 \pm 2.5$	19.2 $\pm 2.8$	.134
Follow-up period (months)	$82.1 \pm 34.9$	81.1 $\pm 27.8$	84.3 $\pm 47.7$	.653
Risser stage	3.58 $\pm 1.5$	3.6 $\pm 1.4$	3.5 $\pm 1.5$	.976
Lenke classification
3	41 (34.5%)	27 (32.9%)	14 (38.9%)	.727
4	7 (5.9%)	5 (6.1%)	2 (5.6%)
5	48 (40.3%)	36 (43.9%)	12 (33.3%)
6	22 (18.5%)	14 (17.1%)	8 (22.2%)
Thoracic kyphosis	$21.1 \pm$ 10.7	20.4 $\pm 10.1$	22.4 $\pm 11.8$	.349
Lumbar lordosis	$47.6 \pm 11.8$	47.5 $\pm 11.7$	47.9 $\pm 12.3$	.843
Proximal thoracic	18.0 $\pm 13.8$	15.3 $\pm 9.7$	25.1 $\pm 19.8$	.032*
Main thoracic	47.2 $\pm 17.4$	$44.2 \pm 14.8$	53.7 $\pm 20.9$	.017*
Lumbar curve	$52.9 \pm 11.7$	51.0 $\pm 10.1$	57.5 $\pm 13.8$	.016*
Sacral slanting	$5.4 \pm 2.9$	5.1 $\pm 2.8$	6.1 $\pm 2.9$	.087
Pelvic obliquity
C7-CSVL	$14.3 \pm 9.0$	14.1 $\pm 8.9$	14.6 $\pm 9.3$	.819
DWA below LIV	$6.0 \pm$ 4.4	4.8 $\pm 3.4$	7.1 $\pm 5.5$	.250
LIV tilt	$21.0 \pm 9.5$	21.3 ± 9.2	20.3 ± 10.3	.602
LIV +1 tilt	$17.4 \pm 7.0$	17.4 $\pm 6.7$	13.8 $\pm 5.5$	.178

*: P-value <.05; LIV, lowest instrumented vertebra; CSVL, central sacral vertical line; DWA, disc wedge angle.

Table 2.

Baseline Characteristics of Patients: a Comparison Between Those Included in the Final Analysis and Those Lost to Follow-up

Characteristic	Overall(n = 118)	Patients lost to follow-up(n = 64)	P-value
Sex
Male	15 (12.7%)	7 (10.9%)	.563
Female	103 (87.3%)	57 (90.1%)	.563
Age, yr	15.3 $\pm 3.5$	15.7 $\pm$ 4.8	.447
Body mass index	19.8 $\pm 2.6$	19.6 $\pm$ 2.5	.613
Follow-up period (months)	$82.1 \pm 34.9$	50.7 $\pm$ 11.7	.001*
Risser stage	3.58 $\pm 1.5$	3.4 $\pm$ 1.5	.358
Lenke classification
3	41 (34.5%)	26 (40.6%)	.126
4	7 (5.9%)	2 (3.1%)
5	48 (40.3%)	22 (34.3%)
6	22 (18.5%)	14 (21.8%)
Thoracic kyphosis	$21.1 \pm$ 10.7	24.6 $\pm$ 9.4	.035*
Lumbar lordosis	$47.6 \pm 11.8$	46.7 $\pm$ 10.3	.573
Proximal thoracic	18.0 $\pm 13.8$	18.8 $\pm$ 9.5	.705
Main thoracic	47.2 $\pm 17.4$	46.5 $\pm$ 16.9	.802
Lumbar curve	$52.9 \pm 11.7$	50.5 $\pm$ 10.9	.568
Sacral slanting	$5.4 \pm 2.9$	4.5 $\pm$ 3.1	.128
LIV level
L3	82 (69.4%)	49 (76.5%)	.124
L4	36 (30.6%)	15 (23.5%)	.124
C7-CSVL	$14.3 \pm 9.0$	12.6 $\pm$ 10.2	.213
DWA below LIV	$6.0 \pm$ 4.4	5.1 $\pm$ 3.8	.084
LIV tilt	$21.0 \pm 9.5$	22.1 $\pm$ 8.5	.150
LIV +1 tilt	$17.4 \pm 7.0$	18.3 $\pm$ 5.6	.278

*: P-value <.05; LIV, lowest instrumented vertebra; CSVL, central sacral vertical line; DWA, disc wedge angle.

Comparison Between the High Slanting and Low Slanting Groups in the LIV-L3 and LIV-L4 Cohorts

The LIV-L3 group comprised 82 patients, of whom 43 (52.4%) were included in the high slanting group and 39 (47.6%) in the low slanting group. The high slanting group exhibited significantly greater lumbar curves than the low slanting group both preoperatively and at all time points postoperatively. No difference in C7-CSVL was noted between the two groups. At the 5-year postoperative mark, DWA below the LIV was significantly smaller in the low slanting group than in the high slanting group, showing a tendency to decrease gradually over the follow-up period (Table 3, Figure 3). The high slanting group demonstrated greater PI and PT preoperatively, and significantly greater PI, SS, and PT at the 5-year follow-up (Table 4).

Table 3.

Comparative Analysis Between High and Low Sacral Slanting in the LIV-L3 Group

	Preop	Postop1 month	Postop1 year	Postop2 year	Postop5 year	P-value (pre vs 2 y)	P-value (2 y vs 5 y)
High sacral slanting N = 43
Proximal thoracic curve	$14.6 \pm 9.8$	$8.6 \pm 5.4$	8.4 $\pm 5.4$	7.1 $\pm 5.6$	$4.9 \pm 5.4$	.000	.182
Main thoracic curve	$44.8 \pm 16.4$	$10.5 \pm 5.8$	$12.5 \pm 6.7$	$12.2 \pm 6.5$	11.8 $\pm 7.3$	.000	.914
Lumbar curve	$53.6 \pm 10.4$ *	$14.1 \pm 8.5$ *	$15.9 \pm 8.9$ *	$16.4 \pm 9.0$ *	16.0 $\pm 8.7$ *	.000	.328
Sacral slanting	$7.1 \pm$ 2.2*	$5.4 \pm 2.5$ *	$5.8 \pm 2.8$ *	$5.4 \pm$ 2.0*	$5.5 \pm$ 2.5*	.000	.506
C7-CSVL	12.8 $\pm 8.7$	$14.4 \pm 10.3$	8.9 $\pm 7.1$	9.1 $\pm 7.2$	$7.7 \pm$ 7.1	.032	.224
LIV tilt	$21.2 \pm 10.2$	$6.5 \pm 3.8$	$6.1 \pm 4.4$	$5.8 \pm 4.9$	$6.0 \pm 4.9$	.000	.807
LIV +1 tilt	$17.1 \pm 7.4$	$9.7 \pm 3.6$ *	$9.9 \pm 4.4$ *	$9.3 \pm 4.1$	$9.0 \pm 4.2$	.000	.096
DWA below LIV	$4.7 \pm 3.5$	$4.1 \pm 2.6$ *	$4.3 \pm 3.0$	$4.1 \pm 2.8$	$3.8 \pm 2.4$ *	.169	.384
Low sacral slanting N = 39
Proximal thoracic curve	$16.0 \pm 9.6$	$7.7 \pm 7.4$	8.1 $\pm 6.6$	7.6 $\pm 6.4$	5.5 $\pm 6.3$	.000	.546
Main thoracic curve	$43.6 \pm 12.9$	$9.7 \pm 5.1$	$10.8 \pm 4.4$	$11.0 \pm 5.2$	$10.7 \pm 5.3$	.000	.538
Lumbar curve	$48.2 \pm 9.1$ *	10.6 $\pm 6.9$ *	$11.7 \pm 7.2$ *	$11.8 \pm 7.9$ *	11.5 $\pm 7.8$ *	.000	.712
Sacral slanting	$2.8 \pm 1.4$ *	$2.6 \pm 2.0$ *	$2.4 \pm 1.6$ *	$2.8 \pm$ 2.0*	$2.5 \pm 2.3$ *	.910	.146
C7-CSVL	15.6 ± 8.9	$11.8 \pm 8.9$	$7.3 \pm 6.5$	$7.9 \pm 6.8$	$8.9 \pm$ 6.7	.000	.372
LIV tilt	$21.4 \pm 8.3$	$5.6 \pm 3.4$	$5.1 \pm 3.2$	$5.6 \pm 3.4$	$5.7 \pm 3.6$	.000	.560
LIV +1 tilt	$16.2 \pm 5.8$	$8.1 \pm 3.0$ *	$8.1 \pm 3.5$ *	8.2 $\pm 4.3$	$7.5 \pm 4.3$	.000	.016
DWA below LIV	$5.0 \pm 3.2$	$3.1 \pm 2.5$ *	$3.2 \pm 2.8$	$3.1 \pm 3.0$	$2.5 \pm 2.0$ *	.010	.039

*: P-value <.05, comparing High and Low sacral group; LIV, lowest instrumented vertebra; CSVL, central sacral vertical line; DWA, disc wedge angle.

Figure 3.

Subgroup analyses comparing patients in the high slanting (pre-sacral slanting ≥5°) and low slanting groups of the LIV L3 and LIV L4 group. (A) In the LIV L3 group, the mean distal wedge angle in the high slanting group changed from 4.7° preoperatively to 4.1°, 4.3°, 4.1°, and 3.8° at 1 month and at 1, 2, and 5 years postoperatively, respectively, whereas in the low slanting group the mean distal wedge angle changed from 5.0° to 3.1°, 3.2°, 3.1°, and 2.5° at the corresponding time points. (B) In the LIV L4 group, the mean C7–CSVL in the high slanting group changed from 15.2 mm preoperatively to 21.9, 12.8, 9.9, and 13.3 mm, whereas in the low slanting group the mean C7–CSVL changed from 13.2 mm to 17.4, 7.7, 7.3, and 7.0 mm at 1 month and at 1, 2, and 5 years postoperatively, respectively. LIV, lowest instrumented vertebra; CSVL, central sacral vertical line. ∗Significant difference between the two groups; †Significant difference from the prior assessment; ‡ Significant difference from the preoperative assessment

Table 4.

Sagittal Parameter Analysis Within the LIV-L3 Group Comparing Patients With High and Low Sacral Slanting

	Preoperative	1 month follow-up	5-year follow-up	P-value (pre vs 5 years)
LIV L3 (n = 82)
High slanting group (n = 43)
PI (°)	53.8 ± 9.7*	53.2 ± 8.5*	54.3 ± 9.2*	.431
SS (°)	41.3 ± 8.8	42.9 ± 7.9*	42.3 ± 6.1*	.490
PT (°)	12.5 ± 6.3*	10.3 ± 9.2	12.0 ± 7.4*	.711
SVA (mm)	−1.0 ± 23.3	−3.3 ± 20.9	5.1 ± 22.5	.391
TK (°)	22.1 ± 10.6	26.6 ± 10.9	35.8 ± 11.0	.000^†
LL (°)	47.0 ± 9.7	52.6 ± 11.7	54.5 ± 24.4	.154
Low slanting group (n = 39)
PI (°)	44.5 ± 7.7*	44.6 ± 7.4*	45.2 ± 6.7*	.159
SS (°)	37.8 ± 7.2	38.2 ± 7.6*	38.0 ± 6.2*	.845
PT (°)	6.7 ± 6.2*	6.5 ± 5.3	7.2 ± 6.5*	.595
SVA (mm)	6.8 ± 29.0	−9.3 ± 22.6	4.7 ± 17.8	.873
TK (°)	18.4 ± 9.3	26.7 ± 8.6	33.9 ± 8.7	.001^†
LL (°)	43.0 ± 12.8	49.2 ± 9.4	53.8 ± 9.8	.001^†

*: P-value <.05, comparing high and low sacral group; †: P value <.05; PI, pelvic incidence; SS, sacral slope; PT, pelvic tilt; SVA, sagittal vertical axis; TK, thoracic kyphosis; LL, lumbar lordosis.

Among the 36 patients in the LIV-L4 group, 24 (66.6%) were included in the high slanting group and 12 (33.4%) in the low slanting group. The postoperative LIV tilt was significantly greater in the high slanting group than in the low slanting group, and it further increased over time. Furthermore, C7-CSVL was significantly larger at 5 years postoperatively, showing a tendency to further increase from the 2- to 5-year postoperative marks. No significant difference in DWA below the LIV was observed between the two groups (Table 5, Figure 3). Moreover, the high slanting group exhibited significantly greater PI and SS both preoperatively and at the 5-year follow-up; however, no significant difference in SVA was observed between the two groups (Table 6).

Table 5.

Comparative Analysis Between High and Low Sacral Slanting in the LIV-L4 Group

	Preop	Postop1 month	Postop1 year	Postop2 year	Postop5 year	P-value (pre vs 2 y)	P-value (2 y vs 5 y)
High sacral slanting n = 24
Proximal thoracic curve	23.8 ± 16.3	8.2 $\pm 9.1$	10.2 $\pm 11.6$	10.8 $\pm 7.6$	4.4 $\pm$ 10.4	.033	.666
Main thoracic curve	53.6 ± 21.4	12.5 $\pm 8.0$	$14.5 \pm 9.0$	$17.7 \pm 8.9$	16.2 $\pm 10.9$	.000	.669
Lumbar curve	58.8 ± 13.8	$12.6 \pm 7.3$	$13.9 \pm 7.1$	$17.1 \pm 8.9$ *	$15.2 \pm 10.8$	.000	.950
Sacral slanting	7.5 ± 2.5*	$4.9 \pm 2.3$	$5.0 \pm 2.5$ *	$5.2 \pm 2.3$ *	$5.3 \pm 2.4$ *	.000	.767
C7-CSVL	15.2 ± 9.4	$21.9 \pm 16.2$	$12.8 \pm 16.9$	$9.9 \pm 8.6$	$13.3 \pm 8.4$ *	.051	.035
LIV tilt	23.3 ± 9.6*	$7.0 \pm 3.5$	$7.0 \pm 4.1$ *	$6.8 \pm 4.3$ *	7.2 $\pm 4.4$ *	.000	.038
LIV +1 tilt	$14.8 \pm 5.3$	$6.7 \pm 3.1$	$7.2 \pm 3.5$ *	$7.2 \pm 2.6$	$7.3 \pm 2.7$ *	.000	.767
DWA below LIV	8.1 $\pm 5.9$	1.9 $\pm 1.5$	$2.1 \pm 1.3$	$2.9 \pm 2.7$	2.5 $\pm 2.9$	.001	.578
Low sacral slanting n = 12
Proximal thoracic curve	27.1 $\pm 25.3$	$11.6 \pm 7.5$	$15.5 \pm 0.7$	$14.6 \pm 14.5$	$7.9 \pm 12.2$	.084	.142
Main thoracic curve	53.9 $\pm 20.9$	$15.8 \pm 6.2$	$16.7 \pm 4.1$	$13.8 \pm 5.6$	$14.4 \pm 7.2$	.000	.764
Lumbar curve	$54.5 \pm 13.9$	$13.4 \pm 8.7$	$13 \pm 8.5$	$10.2 \pm 7.6$	$11.1 \pm 8.8$	.000	.901
Sacral slanting	$3.2 \pm 1.2$ *	$3.4 \pm 2.7$	$2.1 \pm 1.4 *$	$2.5 \pm 1.9$	$3.5 \pm 1.7$ *	.107	.066
C7-CSVL	13.2 $\pm 9.2$	$17.4 \pm 10.9$	$7.7 \pm 5.4$	7.3 $\pm 6.3$	$7.0 \pm 4.8$ *	.178	.940
LIV tilt	$18.2 \pm 9.0$ *	$5.1 \pm 2.8$	$3.9 \pm 3.2$ *	$3.8 \pm 3.2$ *	$3.8 \pm 2.7$ *	.005	.899
LIV +1 tilt	$11 \pm 3.3$	$5.3 \pm 3.3$	$3.9 \pm 3.8$ *	$5.2 \pm 3.8$	$4.4 \pm 2.9$ *	.020	.390
DWA below LIV	$6.7 \pm 4.4$	$2.1 \pm 1.5$	$3.3 \pm 2.4$	2.4 $\pm 2.6$	2.9 $\pm 2.1$	.026	.722

*: P-value <.05, comparing high and low sacral group; LIV, lowest instrumented vertebra; CSVL, central sacral vertical line; DWA, disc wedge angle.

Table 6.

Sagittal Parameters Analyses Within the LIV-L4 Group Comparing Patients in High and Low Sacral Slanting

	Preoperative	1 month follow-up	5-year follow-up	P-value (pre vs 5 years)
LIV L4 (n = 36)
High slanting group (n = 24)
PI (°)	52.6 ± 9.9*	51.4 ± 8.4*	52.4 ± 8.1*	.832
SS (°)	43.8 ± 10.8*	41.8 ± 9.6*	43.6 ± 9.9*	.797
PT (°)	8.7 ± 6.2	9.6 ± 8.3	8.8 ± 6.6	.958
SVA (mm)	2.4 ± 22.7	−5.3 ± 20.8	−2.3 ± 26.8	.433
TK (°)	21.7 ± 11.0	26.7 ± 7.3	31.9 ± 7.3	.001^†
LL (°)	50.7 ± 10.4	52.3 ± 12.8	57.1 ± 10.9	.015^†
Low slanting group (n = 12)
PI (°)	45.2 ± 6.1*	44.7 ± 5.1*	46.8 ± 5.9*	.286
SS (°)	36.0 ± 7.5*	34.8 ± 6.8*	35.9 ± 6.4*	.919
PT (°)	9.2 ± 4.2	9.9 ± 6.6	10.9 ± 6.2	.188
SVA (mm)	−3.0 ± 21.0	2.1 ± 22.9	3.5 ± 24.1	.065
TK (°)	23.9 ± 13.7	29.7 ± 10.5	37.0 ± 11.2	.001^†
LL (°)	45.8 ± 11.3	47.7 ± 7.1	54.0 ± 8.3	.149

An adding-on phenomenon was observed in four patients in the LIV-L3 group, with three (7% [3/43]) in the high slanting group and one (2.6% [1/39]) in the low slanting group. Coronal decompensation occurred in 12 patients in the LIV-L4 group, with 11 (45.8% [11/24]) in the high slanting group and 1 (8.3% [1/12]) in the low slanting group (P = .026) (Table 7).

Table 7.

Incidence of the Adding-on Phenomenon and Coronal Decompensation

	LIV L3			LIV L4
	High sacral slanting	Low sacral slanting	P-value	High sacral slanting	Low sacral slanting	P-value
Adding-on	3/43 (7.0%)	1/39 (2.6%)	.354	2/24(8.3%)	1/12 (8.3%)	.999
Coronal decompensation	10/43 (23.3%)	8/39 (20.5%)	.764	11/24(45.8%)	1/12 (8.3%)	.024*

*: P-value <.05 by the chi-squared test.

No significant difference in SRS-22 scores was observed between the high and low slanting groups preoperatively or at 5 years postoperatively. Both groups demonstrated improvements in five domains of the SRS-22 questionnaire (function, pain, self-image, mental health, and satisfaction) at 5 years postoperatively (Table 8).

Table 8.

SRS-22 Scores in the High and Low Slanting Groups

	High slanting group	Low slanting group	P-value
Preoperative
Function	1.45 $\pm 1.20$	1.71 $\pm 1.14$	.415
Pain	$1.38 \pm$ 1.18	1.65 $\pm 1.11$	.386
Self-image	$2.29 \pm 0.81$	$2.39 \pm 0.79$	.655
Mental health	1.88 $\pm 0.77$	$1.89 \pm$ 0.96	.975
Satisfaction	$1.80 \pm 0.76$	1.95 $\pm 0.77$	.489
Postoperative 5-year
Function	$2.40 \pm 0.66$ ^†	$2.62 \pm$ 0.68^†	.131
Pain	$2.02 \pm 0.45$ ^†	$2.09 \pm 0.40$	.503
Self-image	$2.72 \pm 0.49$	$2.79 \pm 0.51$	.495
Mental health	$2.60 \pm 0.55$ ^†	$2.74 \pm 0.52$ ^†	.221
Satisfaction	$2.50 \pm$ 0.40^†	$2.61 \pm 0.42$ ^†	.212

SRS-22, Scoliosis Research Society-22 questionnaire; †: P-value <.05, Paired t test was used for the comparison between preoperative and postoperative 5-year results.

Factors Associated With the Adding-on Phenomenon and Coronal Decompensation at 5 Years Postoperatively

In the multivariate logistic regression analysis conducted to identify preoperative risk factors, sacral slanting was identified as a risk factor for the adding-on phenomenon (odds ratio [OR] (95% confidence interval [CI]) = 1.350 (1.028-1.774); P = .031), along with the DWA below the LIV (OR [95% CI] 0.593 [0.378-0.929]; P = .023) (Table 9). Additionally, sacral slanting was recognized as a risk factor for coronal decompensation (OR [95% CI] 1.209 [1.010-1.447]; P = .038) (Table 10).

Table 9.

Multivariate Logistic Regression Analysis Demonstrating Preoperative Factors Associated With the Adding-on Phenomenon

Variable	Univariate analysis		Multivariate analysis
Variable	Odds ratio (95% CI)	P-value	Odds ratio (95% CI)	P-value
Age	1.053 (0.869-1.277)	.597	1.085 (0.847-1.391)	.518
Risser stage	1.042 (0.580-1.873)	.890	1.056 (0.497-2.240)	.888
LIV L3/L4	1.773 (0.376-8.363)	.469	3.757 (0.630-22.399)	.146
Proximal thoracic curve	1.033 (0.989-1.080)	.141
Main thoracic curve	1.018 (0.976-1.061)	.412
Lumbar curve	1.023 (0.960-1.091)	.481	1.070 (0.966-1.185)	.192
Sacral slanting	1.288 (1.015-1.635)	.037*	1.350 (1.028-1.774)	.031*
C7-CSVL	0.993 (0.911-1.082)	.873	0.977 (0.879-1.086)	.669
DWA below LIV	0.623 (0.413-0.939)	.024*	0.593 (0.378-0.929)	.023*
LIV tilt	1.000 (0.922-1.084)	.994
LIV +1 tilt	1.068 (0.960-1.188)	.225

*P < .05. CI, confidence interval; LIV, lowest instrumented vertebra; CSVL, central sacral vertical line; DWA, disc wedge angle.

Table 10.

Multivariate Logistic Regression Analysis Demonstrating Preoperative Factors Associated With Coronal Decompensation

Variables	Univariate analysis		Multivariate analysis
Variables	Odds ratio (95% CI)	P-value	Odds ratio (95% CI)	P-value
Age	1.000 (1.000-1.000)	.760	0.969 (0.805-1.167)	.742
Risser stage	1.135 (0.794-1.621)	.488	1.179 (0.804-1.729)	.400
LIV L3/L4	1.778 (0.746-4.235)	.194	1.940 (0.685-5.496)	.213
Proximal thoracic curve	0.991 (0.953-1.031)	.655
Main thoracic curve	0.995 (0.971-1.020)	.692
Lumbar curve	1.000 (0.965-1.036)	.996	0.984 (0.929-1.042)	.579
Sacral slanting	1.183 (1.024-1.366)	.022*	1.209 (1.010-1.447)	.038*
C7-CSVL	1.018 (0.973-1.066)	.445	1.012 (0.954-1.073)	.693
DWA below LIV	0.985 (0.896-1.082)	.749	0.919 (0.816-1.035)	.163
LIV tilt	0.999(0.956-1.044)	.969
LIV +1 tilt	0.993 (0.936-1.054)	.828

*P < .05. CI, confidence interval; LIV, lowest instrumented vertebra; CSVL, central sacral vertical line; DWA, disc wedge angle.

Representative Cases of High Slanting in the LIV-L3 and LIV-L4 Cohorts

A 13-year-old female patient who underwent fusion from T4 to L3 exhibited a stationary course of the DWA below the LIV over time (Figure 4A). In contrast, a 14-year-old female patient who underwent fusion from T5 to L4 showed an increase in C7-CSVL over time (Figure 4B).

Figure 4.

Representative cases of high sacral slanting in the LIV L3 and LIV L4 cohorts. (A) A 13-year-old female patient who underwent fusion extending from T4 to L3 exhibited a stationary course of the distal wedge angle below the LIV over time (B) A 14-year-old female patient who underwent fusion extending from T5 to L4 exhibited an increase in C7-CSVL over time. LIV, lowest instrumented vertebra; CSVL, central sacral vertical line

Discussion

In the present study, we elucidated the impact of sacral slanting in patients with AIS with structural lumbar curves, focusing on patients with LIV-L3 or LIV-L4. In patients with LIV-L3, when sacral slanting was high, the DWA below the LIV was greater than that in the low slanting group at 5 years, but it did not show a tendency to increase over time. Conversely, in the low slanting group, the DWA below the LIV decreased over time. For patients with LIV-L4, those with high slanting exhibited an increasing trend in C7-CSVL over the 5-year postoperative period. Additionally, sacral slanting was identified as a common risk factor for both the adding-on phenomenon and coronal decompensation, whereas no differences were observed in clinical outcomes related to sacral slanting at 5-year postoperatively.

Sacral slanting in patients with AIS is a distinctive phenomenon primarily observed in individuals with considerable lumbar curves, especially those identified as Lenke types 5 and 6 AIS with a lumbar modifier C.¹ According to Joo et al,⁴ sacral slanting typically occurs to the left direction in the L4-left tilt and to the right in the L4-right tilt in Lenke type 1A AIS. In this study, patients with structural lumbar curves classified as Lenke types 3, 4, 5, and 6 AIS were examined, noting sacral slanting and L4 tilt in the same direction for all patients; therefore, the direction of sacral slanting was not taken into consideration. As the sacrum serves as the foundation of the spine, tilting of the sacrum prompts the remaining mobile segments to compensate to align the C7 plumb line with CSVL.¹⁷ Several studies have suggested that lumbar curve magnitude and pelvic obliquity are correlated,^18,19 and it is possible that sacral slanting may develop secondarily as a consequence of the lumbar curve. Sacral slanting may also be associated with pelvic obliquity. However, sacral slanting is not always directly proportional to lumbar curve magnitude, and differences between sacral slanting and pelvic obliquity have frequently been observed.¹⁹ These findings indicate that sacral slanting cannot be regarded merely as a reflection of the lumbar curve. In our cohort, sacral slanting showed a significant association with the adding-on phenomenon and coronal decompensation, whereas lumbar curve magnitude did not. This suggests that sacral slanting has a distinct role beyond its correlation with the lumbar curve. Regardless of its underlying cause, sacral slanting represents a tilting of the sacral foundation, which is of critical importance in maintaining coronal balance and supporting the distal residual segments.

When fusion is extended to L3, three segments remain in the mobile area, including the L3-4, L4-5, and L5-S1 discs. These segments encounter forces in the opposite direction of the sacral slanting, with greater stress occurring at the uppermost segment due to the rigidity of the implants.²⁰ Therefore, the high slanting group had a larger DWA below the LIV compared with the low slanting group. However, the DWA did not increase over time because the presence of the other two segments helped distribute the stress, resulting in the stabilization of the DWA below the LIV over time. This also accounted for the gradual decrease in the DWA below the LIV that was observed in the low slanting group within the LIV-L3 cohort. The adding-on phenomenon was noted in seven patients (5.9% [7/118]). In the high slanting group, the phenomenon was observed in 7.0% (3/43) of the patients, and in the low slanting group, it presented in 2.6% (1/39) of the patients within the LIV-L3 cohort, indicating that it occurs more frequently with greater slanting, although this difference was not statistically significant. In a study on the adding-on phenomenon in Lenke type 1A AIS, Joo et al.⁴ revealed that a larger lumbar curve, younger age, and larger sacral slanting angle were associated with a higher probability of the adding-on phenomenon, and the authors identified the preoperative lumbar curve as an independent risk factor. In our study, sacral slanting was identified as a risk factor for the adding-on phenomenon, as we focused exclusively on patients with structural lumbar curves, who exhibited greater preoperative sacral slanting (5.4 ± 2.9°), resulting in a more significant impact.

When the LIV was at L4, C7-CSVL increased immediately, regardless of sacral slanting, but showed a decrease at 1 year postoperatively. Subsequently, from 2 years postoperatively, C7-CSVL exhibited an increasing trend in the high slanting group (Figure 3). This suggests that the structural nature of the lumbar curve contributes to coronal imbalance, with only two compensatory segments L4–L5 and L5–S1 remaining to maintain coronal alignment. In cases of coronal decompensation, it occurred significantly more often in the LIV-L4 high slanting group, with an incidence rate of 45.5% (11/24). Additionally, logistic analysis revealed sacral slanting as a factor influencing C7-CSVL. According to Gomez et al,²¹ insufficient correction of the major curve can lead to coronal decompensation in patients with AIS 2 years postoperatively. In our study, the correction rate of the major lumbar curve did not have a significant impact on coronal decompensation. However, because we focused solely on patients with structural lumbar curves, our findings may differ from the studies involving various Lenke types.

Coronal decompensation can occur when compensatory curves are not functional. According to Winter et al, overcorrection, defined as correction exceeding the flexibility of the main curve, can lead to problems with compensatory curves.^7,22 They emphasized that overcorrection of the primary thoracic or thoracolumbar/lumbar curve is the most significant factor related to coronal decompensation. To achieve a well-balanced spine, surgeons refrain from performing excessive correction beyond the preoperative flexibility and avoid instrumentation within the transitional mobile segments that connect the major and secondary curves.⁷ In this study, a higher incidence of coronal decompensation was observed in patients with considerable sacral slanting in LIV-L4, all of whom underwent fusion with the UIV above T5, including the main thoracic curve. This finding suggests that in patients with extensive proximal fusion, selecting L4 as the LIV may exacerbate coronal decompensation, particularly when the sacrum is tilted, owing to the limited distal compensatory segments rather than due to overcorrection of the main curve.

Although no significant difference in SRS-22 scores was observed between the high and low sacral slanting groups at the 5-year follow-up, this finding should be interpreted with caution. Radiographic differences, such as increased DWA or coronal decompensation, may not immediately translate into patient-reported outcomes, particularly given the ceiling effect of the SRS-22 questionnaire^23,24 and the relatively young age of the study population. Compensatory mechanisms may also allow patients to maintain satisfactory function and quality of life despite measurable radiographic imbalance. Nevertheless, the long-term biomechanical implications of sacral slanting remain important, as subtle radiographic differences observed in adolescence may predispose patients to future degenerative changes. Longer-term follow-up will be required to clarify the clinical significance of these findings.

Our study has some limitations. First, it was a retrospective study with a relatively small sample size and a substantial loss to follow-up, which limits the generalizability of the findings. Therefore, a well-designed study with a larger number of cases is needed. Second, longitudinal comparisons across multiple postoperative timepoints were performed using repeated paired t-tests rather than a formal repeated-measures model, which may increase the risk of type I error. In addition, the relatively small number of adding-on and coronal decompensation events may have affected the stability and precision of the estimated odds ratios in the multivariable analyses. Third, differences in preoperative radiographic parameters between the LIV-L3 and LIV-L4 groups may have introduced selection bias. Because patients requiring fusion to L4 generally present with severe preoperative deformities, residual selection bias could not be fully eliminated despite multivariate adjustment. Fourth, although the 5° cutoff used to categorize sacral slanting was referenced from previous literature,¹ this threshold remains somewhat arbitrary as sacral slanting is essentially a continuous variable. To mitigate this limitation, we performed additional sensitivity analyses using alternative thresholds (3°, 5°, and 7°). Across all cutoffs, the results were consistent, which supported the robustness and stability of our findings (Supplemental Table 1). Despite these limitations, we conducted a long-term follow-up over 5 years, specifically focused on patients with structural lumbar curves. The strength of our study lies in the analysis of the fate of distal segments below the LIV and residual coronal balance related to sacral slanting by categorizing the patients into the LIV-L3 and LIV-L4 groups.

Conclusion

In patients with AIS with structural lumbar curves, extending fusion to L3 resulted in an increase in the DWA below the LIV in cases of high sacral slanting; however, it did not show a continuous increase over time. In contrast, selecting LIV-L4 in patients with high sacral slanting led to coronal decompensation, which tended to increase over time. Sacral slanting was identified as an independent risk factor for the adding-on phenomenon and coronal decompensation; however, it had no significant impact on clinical outcomes at the 5-year follow-up. Surgeons may benefit from considering these factors when selecting the LIV in patients with AIS with high sacral slanting.

Supplemental Material

Supplemental Material - Effect of Sacral Slanting on Coronal Balance and Residual Curve After Lumbar Curve Correction in Adolescent Idiopathic Scoliosis: A 5-Year Follow-up Study

Supplemental Material for Effect of Sacral Slanting on Coronal Balance and Residual Curve After Lumbar Curve Correction in Adolescent Idiopathic Scoliosis: A 5-Year Follow-up Study by Chang Ju Hwang, Jae Hwan Cho, Sehan Park, Dong-Ho Lee, Taehwan Ahn, , and Gumin Jeong in Global Spine Journal

Footnotes

ORCID iDs

Chang Ju Hwang

Jae Hwan Cho

Sehan Park

Dong-Ho Lee

Gumin Jeong

Ethical Considerations

The study protocol was approved by the institutional review board of Asan Medical Center (no. 2024-1138). The requirement for informed consent was waived owing to the retrospective nature of the study.

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.

Supplemental Material

Supplemental material for this article is available online.

References

Lee

J-K

Kim

, et al. The clinical importance of sacral slanting in patients with adolescent idiopathic scoliosis undergoing surgery. Spine J. 2015;15(5):834-840.

Hwang

Lee

, et al.

Does sacral slanting affect postoperative shoulder balance in patients with lenke type 2A adolescent idiopathic scoliosis?

Neurospine. 2024;21(1):286-292. doi:10.14245/ns.2347072.536

Han

Yang

Wen

, et al. Morphology and deformity of the shoulder and pelvis in the entire spine radiographs of adolescent idiopathic scoliosis. Quant Imag Med Surg. 2023;13(5):3266-3278. doi:10.21037/qims-22-656

Joo

Hwang

Cho

, et al.

Does sacral slanting affect distal adding-on in lenke type 1A adolescent idiopathic scoliosis?

Spine. 2018;43(17):E990-E997.

Higuchi

Ikegami

Oba

, et al. Postoperative residual coronal decompensation inhibits self-image improvement in adolescent patients with idiopathic scoliosis. Asian Spine J. 2023;17(1):149-155. doi:10.31616/asj.2021.0429

Wang

Hansen

Høy

Bünger

. Distal adding-on phenomenon in lenke 1A scoliosis: risk factor identification and treatment strategy comparison. Spine. 2011;36(14):1113-1122.

Thompson

Transfeldt

Bradford

Ogilvie

Boachie-Adjei

. Decompensation after cotrel-dubousset instrumentation of idiopathic scoliosis. Spine. 1990;15(9):927-931.

Matsumoto

Watanabe

Hosogane

, et al. Postoperative distal adding-on and related factors in lenke type 1A curve. Spine. 2013;38(9):737-744.

Suk

Lee

Chung

Kim

. Selective thoracic fusion with segmental pedicle screw fixation in the treatment of thoracic idiopathic scoliosis: more than 5-Year Follow-Up. Spine. 2005;30(14):1602-1609. doi:10.1097/01.brs.0000169452.50705.61

10.

Lenke

Edwards

CCI

Bridwell

. The lenke classification of adolescent idiopathic scoliosis: how it organizes curve patterns as a template to perform selective fusions of the spine. Spine. 2003;28(20S):S199-S207. doi:10.1097/01.Brs.0000092216.16155.33

11.

Lee

J-K

Hwang

Lee

D-H

Kim

Cho

. Is it enough to stop distal fusion at L3 in adolescent idiopathic scoliosis with major thoracolumbar/lumbar curves? Eur Spine J. 2016;25:3256-3264.

12.

Lee

Hwang

Lee

D-H

Cho

. Five major controversial issues about fusion level selection in corrective surgery for adolescent idiopathic scoliosis: a narrative review. Spine J. 2017;17(7):1033-1044.

13.

Zhuang

Zhang

Wang

Yang

Lin

. How to select the lowest instrumented vertebra in lenke type 5 adolescent idiopathic scoliosis patients? Spine J. 2021;21(1):141-149.

14.

Yang

Zhao

Yin

, et al. Prevalence, risk factors, and characteristics of the “Adding-On” phenomenon in idiopathic scoliosis after correction surgery: a systematic review and meta-analysis. Spine. 2018;43(11):780-790. doi:10.1097/brs.0000000000002423

15.

Márkus

Schlégl

ÁT

Burkus

, et al. The effect of coronal decompensation on the biomechanical parameters in lower limbs in adolescent idiopathic scoliosis. Orthop Traumatol: Surgery & Research. 2018;104(5):609-616. doi:10.1016/j.otsr.2018.06.002

16.

Qiu

X-S

Wang

Z-W

Qiu

, et al. Preoperative pelvic axial rotation: a possible predictor for postoperative coronal decompensation in thoracolumbar/lumbar adolescent idiopathic scoliosis. Eur Spine J. 2013;22(6):1264-1272. doi:10.1007/s00586-013-2695-7

17.

Mac-Thiong

J-M

Transfeldt

Mehbod

, et al.

Can C7 plumbline and gravity line predict health related quality of life in adult scoliosis?

Spine. 2009;34(15):E519-E527. doi:10.1097/BRS.0b013e3181a9c7ad

18.

Banno

Yamato

Hasegawa

, et al. Impact of pelvic obliquity on coronal alignment in patients with adolescent idiopathic scoliosis. Spine Deform. 2020;8(6):1269-1278. doi:10.1007/s43390-020-00145-x

19.

Cho

Lee

Joo

Park

Hwang

Lee

. Association between sacral slanting and adjacent structures in patients with adolescent idiopathic scoliosis. Clin Orthop Surg. 2017;9(1):57-62. doi:10.4055/cios.2017.9.1.57

20.

Han

Hyun

S-J

Kim

K-J

Jahng

T-A

Lee

Rhim

S-C

. Rod stiffness as a risk factor of proximal junctional kyphosis after adult spinal deformity surgery: comparative study between cobalt chrome multiple-rod constructs and titanium alloy two-rod constructs. Spine J. 2017;17(7):962-968. doi:10.1016/j.spinee.2017.02.005

21.

Gomez

Matsumoto

Colacchio

, et al. Risk factors for coronal decompensation after posterior spinal instrumentation and fusion in adolescent idiopathic scoliosis. Spine Deformity. 2014;2(5):380-385. doi:10.1016/j.jspd.2014.05.001

22.

Winter

Lonstein

Denis

. How much correction is enough? Spine. 2007;32(24):2641-2643. doi:10.1097/BRS.0b013e31815a5207

23.

Stone

Ames

Pellise

Newton

Upasani

Kelly

Harms Study Group . Scoliosis research Society-22r and ceiling effects: limited capabilities for precision medicine with adolescent idiopathic scoliosis. Spine. 2025;50(1):34-39. doi:10.1097/brs.0000000000004993

24.

Bastrom

Bartley

Marks

Yaszay

Newton

Harms Study Group . Postoperative perfection: ceiling effects and lack of discrimination with both SRS-22 and -24 outcomes instruments in patients with adolescent idiopathic scoliosis. Spine. 2015;40(24):E1323-E1329. doi:10.1097/brs.0000000000001082

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB