Personalized Spine Surgery in Adult Deformity: Reoperation Rates and Mechanical Complications Following Customized Planning and Interbody Implant Use

Abstract

Study Design

Retrospective cohort study.

Objectives

Prior studies have shown that adult spinal deformity (ASD) patients undergoing revision surgery due to mechanical complications had less radiographic improvement and worsening patient-reported outcomes scores. The combination of customized 3D planning and personalized implants has been shown to contribute to improved achievement of alignment goals. This study aimed to determine whether such improved correction also results in a correspondingly lower revision surgery rate due to mechanical complications.

Methods

Pre- and postoperative radiographic alignment measures, including lumbar lordosis (LL), distal lumbar lordosis (DLL), pelvic incidence (PI) – LL mismatch, and L1 pelvic angle (L1PA), as well as data on mechanical complications leading to reoperation were collected on 115 ASD patients with personalized interbody implants and minimum 2-year follow-up. This cohort was statistically compared to a multicenter dataset (ISSG) of 997 ASD patients treated using stock devices and using the same reoperation classifications.

Results

Postoperatively achieved alignment measures correlated significantly with their respective preoperative alignment goals, with the following average offsets from plan: 2.4° LL, −0.2° DLL, −2.4° PI-LL, 0.3° L1PA. Compared to the ISSG cohort that utilized stock interbody implants, the cohort utilizing 3D preoperative planning and personalized interbody implants resulted in significantly fewer revisions for mechanical complications up to 2 years postoperatively: 5/115 (4.3%) vs 166/997 (16.6%), P < 0.001.

Conclusions

These findings suggest that achieving planned alignment targets with personalized interbody devices is associated with reduced revision surgery for mechanical complications, a result which may have positive implications for improved patient outcomes and reduced cost.

Keywords

revision sagittal alignment ALIF LLIF TLIF PJK comparison

Introduction

Despite advances in spinal surgery techniques, reoperation rates following correction of adult spinal deformity (ASD) remain high and continue to pose a significant burden for patients and the healthcare system.^1-4 Lafage et al (2024) reported that only one-third of ASD patients remained complication-free 2 years postoperatively, with nearly 17% requiring reoperation due to mechanical or implant-related complications.⁴ Similarly, it is well appreciated that ASD patients that undergo surgery may be predisposed to mechanical complications when optimal postoperative sagittal alignment is not achieved.^5-7

To improve outcomes, surgeons have employed various classifications and strategies to define alignment targets, including age-adjusted parameters, software-assisted planning, and more recently, normative values of the spinopelvic angles.^8-11 However, translating these preoperative goals into the intraoperative setting remains a persistent challenge. Smith et al (2024), using the International Spine Study Group (ISSG) dataset, identified that even among experienced senior spine surgeons, nearly two-thirds of 266 complex adult deformity surgery cases failed to meet all planned sagittal alignment targets when using conventional instrumentation, highlighting a systemic limitation in current surgical execution.¹² Even with advanced planning, the use of stock (ie, not patient-specific) implants, particularly interbody implants, introduces variability due to their generic shape, which often does not match the patient’s endplate anatomy.¹³ This can lead to unpredictable or inconsistent alignment outcomes, despite surgical intent, such as was found in the Smith et al. ISSG study,¹² and corresponding mechanical failures, as found in the Lafage et al. ISSG study.⁴ Specifically, the Lafage et al (2024) study of 997 ASD patients treated across 20 institutions reported that reoperations due to implant failure and/or radiographic complications occurred in 166 (16.6%) patients within 2 years postoperatively.⁴

Personalized interbody (PIB) implants, which are designed from a 3D patient-specific surgical plan to achieve individualized alignment goals as well as match the anatomy of the patient’s vertebral endplates, received “Breakthrough Device” designation from the US Food and Drug Administration in 2020, and have been in clinical use since 2021. Prior to clinical use, early studies of this concept employing cadaveric testing, biomechanical studies, and finite element models provided findings that patient-specific interbody devices could achieve planned correction more reliably,¹⁴ decrease stress on posterior rods,¹⁵ significantly increase contact area on the endplate,¹⁶ reduce endplate stress concentration,¹⁶ and reduce the stress increase in adjacent discs and facets.¹⁷ Clinical data has further shown that personalized implants achieve intervertebral and regional lordosis targets with greater accuracy.^13,18,19 By enhancing the predictability of alignment correction, the use of PIB implants may reduce the incidence of revision surgeries related to mechanical complications such as junctional failure, rod fracture, and pseudoarthrosis. The goal of the present study was to evaluate whether the use of personalized 3D preoperative planning, in combination with patient-specific interbody implants, results in successfully achieving planned alignment goals and reduces mechanical complications and reoperation rates at 2 years postoperatively in patients treated for ASD, compared to reoperation rates in the previously cited ISSG study of similar ASD patients.⁴

Methods

Study Design and Patient Populations

This study is an institutional review board-exempt multicenter retrospective chart review of patients diagnosed with ASD who underwent surgical correction using 3D planning & visualization software combined with custom-made anatomically designed interbody devices to match the surgeon-approved personalized plan (aprevo®, Carlsmed, Carlsbad, CA). The current PIB cohort represents all consecutive cases from 10 institutions across the United States, with surgeries performed between February 2021 and March 2023 meeting the specified inclusion criteria. No cases were excluded as learning-curve cases.

Patients were included if they were at least 18 years old when surgically treated for ASD with one or more anatomically designed PIB devices; met the same radiographic criteria for ASD as employed in the ISSG study by Lafage et al (ie, having at least one of the following: scoliosis ≥20°, sagittal vertical axis (SVA) ≥ 5 cm, pelvic tilt (PT) ≥ 25°, or thoracic kyphosis ≥60°)⁴; and had a documented clinical follow-up of at least 2 years postoperatively. In both this and the ISSG cohort, decisions regarding surgical indications, clinical and radiographic evaluations, surgical procedures, and instrumentation were at the discretion of the operating surgeon, with the assumption that the chosen surgical strategy was deemed by the surgeon to provide the greatest chance of a successful outcome based on the experience of the operating surgeon.

All patients in the PIB cohort underwent comprehensive 3D preoperative planning prior to surgical treatment. This planning process incorporated the treating surgeon’s alignment goals and was used to define the final lumbar alignment, including the design specifications for the patient-specific interbody implants. The implants were fabricated to match the superior and inferior endplate morphology at each interbody fusion level, with the intent to improve alignment predictability and reduce variability often seen with stock cages.

Data Collection

Data collected from patient records included demographic variables such as age, sex, and history of prior spinal instrumentation. Preoperative imaging involved standing anteroposterior and lateral spine radiographs and computed tomography (CT) scans used for personalized alignment planning and implant design. Postoperative alignment was assessed using each patient’s latest available standing radiograph, obtained between 6 weeks and 2 years postoperative.

Radiographic measurements were performed independently by 2 readers -- a senior radiologist and a senior spine surgeon -- using validated DICOM viewer software (Microdicom® Ltd, Sofia, Bulgaria). In cases of discrepancies between readers, measurements were jointly reviewed to reach a consensus. This dual-review process ensured consistency and reduced interobserver variability across the multicenter dataset. Radiographic alignment parameters measured included pelvic incidence (PI), lumbar lordosis (LL_L1-S1), distal lumbar lordosis (DLL_L4-S1), intervertebral lordosis (IVL) and coronal angle (IVCA), intervertebral posterior disc height (IVPH), and L1 pelvic angle (L1PA).

Reoperation data were collected through a standardized questionnaire completed by the operating surgeon at each institution. This form captured whether the patient underwent revision surgery, the date of the revision, and the primary cause. To ensure consistent classification, a panel of experienced spine surgeons reviewed all reported revisions and adjudicated whether each was due to a mechanical complication occurring within the 2-year postoperative follow-up. The surgeon panel classified all revision surgeries as due to mechanical complications according to the ISSG-AO Spine Complication Classification System²⁰ if they were attributable to (1) implant failure (eg, interbody dislodgement, loose dislodgement, rod breakage, rod dislodgement, screw breakage, or screw loosening) or (2) radiographic failure (eg, adjacent segment degeneration, coronal imbalance, distal junctional kyphosis (DJK), proximal junctional kyphosis (PJK), pseudoarthrosis, or sagittal imbalance).

Data Assessment and Statistical Analysis

Statistical analysis was performed using Python 3.11 or IBM SPSS v30.0. Descriptive statistics were reported as mean and standard deviation (SD) for continuous variables and frequencies with percentages for categorical variables. The Pearson chi-square test was used to compare categorical variables. Continuous data were assessed for normality using the Shapiro-Wilk test. The effect of levels treated (one vs two or more levels) was assessed using an independent t-test for data with equal variances and Welch’s t-test for data with unequal variances. For comparison of preoperative, planned, and postoperative data, normally distributed continuous data were compared using repeated measures ANOVA followed by planned comparisons of postoperative to preoperative data as well as postoperative to planned data. For non-normal data, preoperative, planned, and postoperative data were compared using a generalized linear mixed model followed by pairwise comparisons. For intervertebral measures, only treated levels were compared. Bland-Altman plots were drawn for each parameter. Comparison with Lafage et al proportions⁴ was performed with a two-proportion z-test. All tests were two-tailed, with a significance level of α = 0.05.

Results

Patient Population

Baseline demographic and operative characteristics for the 115 patients who met inclusion criteria are summarized in Table 1. Differences in baseline characteristics between this cohort and the cohort reported by Lafage et al are indicated. Of note, the Lafage cohort had a larger number of females, whereas the PIB cohort was statistically significantly older. The potential balancing effect of these differences is unknown as Lafage et al did not report revision data specific to these characteristics. Of the average 1.8 PIB devices used per patient (n = 212), 69% were implanted at the distal levels (L4-S1). The median clinical follow-up was 34 months (IQR 30-39 months), vs that of the Lafage et al study⁴ (25 months, IQR 24-29 months).

Table 1.

Demographic and Treatment Details

Demographic parameter	Current study (n = 115)	Lafage et al (n = 997)⁴	P-value
Sex, number female (%)	73 (63)	761 (76.3)	0.003
Mean age, years (SD)	68.8 (8.1)	60.4 (14.3)	<0.001
Prior spinal fusion surgery, number (%)	43 (37.4)	377 (37.9)^a	0.917
SRS classification²¹
Pre-op coronal deformity type, %
N (no coronal curve; all <30°)	37.7%	40.0%	0.635
T (thoracic-only curve, lumbar <30°)	3.5%	4.3%	0.725
L (lumbar-only curve, thoracic <30°)	46.5%	32.2%	0.002
D (double curve, both >30°)	12.3%	23.5%	0.007
Pre-op PI-LL modifier, %
0 (PI-LL <10°; non-pathologic)	38.4%	35.7%	0.568
+ (10°< PI-LL <20°; moderate deformity)	27.7%	22.4%	0.201
++ (PI-LL >20°; marked deformity)	33.9%	41.9%	0.099
Pre-op PT modifier, %
0 (PT <20°; non-pathologic)	14.5%	34.0%	<0.001
+ (20°< PT <30°; moderate deformity)	41.8%	36.8%	0.304
++ (PT >30°; marked deformity)	43.6%	29.2%	0.002
Pre-op SVA modifier, %
0 (SVA <4 cm; non-pathologic)	26.7%	39.6%	0.011
+ (4 cm< SVA <9.5 cm; moderate deformity)	41.6%	29.0%	0.009
++ (SVA >9.5 cm; marked deformity)	31.7%	30.4%	0.787
Median levels fused, number (IQR)	8 (8-13)	9 (8-14)
Mean PIB devices used per patient, number (SD)	1.8 (0.9)	N/A
Posterior fusion, number (%)	115 (100)	986 (98.9)	0.258
ALIF, number of patients (%)	54 (47)	242 (24)	<0.001
LLIF, number of patients (%)	29 (25)	148 (15)	0.004
TLIF, number of patients (%)	44 (38)	306 (31)	0.098

SD = standard deviation; SRS = Scoliosis Research Society; PI-LL = pelvic incidence – lumbar lordosis mismatch; PT = pelvic tilt; SVA = sagittal vertical axis; IQR = interquartile range (25^th and 75^th percentile); PIB = personalized interbody; N/A = not applicable; ALIF = anterior lumbar interbody fusion; LLIF = lateral lumbar interbody fusion; TLIF = transforaminal lumbar interbody fusion.

^aThe authors in Lafage et al⁴ report 503 (50.5%) having an unspecific prior spine surgery. A portion of this cohort was previously reported by Smith et al² showing that 75% of prior spine surgeries were spinal fusion surgeries. This proportion is applied to the larger cohort for comparison to the number of patients in this series having a prior spinal fusion surgery.

Radiographic Alignment Correction

Complete baseline, planned, and postoperatively achieved radiographic parameters, including segmental averages by interbody procedure and disc level are shown in Tables 2 and 3. Bland-Altman plots for each parameter showed good agreement between postoperatively achieved and planned LL, DLL_L4-S1, PI-LL, and L1PA (Figure 1). The bias of LL, DLL_L4-S1, and PI-LL were minor at 2.4° (±9.3°), −0.2° (±7.6°), and −2.4 (±9.3°), respectively. Additionally, postoperative L1PA was achieved to within a bias of 0.3° (±5.6°) of the ideal value defined by Hills et al in a normative population.²² Focally, IVL was achieved with a mean offset of 0.5° (±4.0°) from the preoperative plan. IVCA and IVPH were achieved with mean offsets of 0.7° (±2.9°) and 0.2 mm (±4.0 mm), respectively. The percentage of patients meeting literature-established ‘normal’ alignment²¹ postoperatively were: PI-LL <10 = 49.1% (versus 63.0%,⁴ P = 0.0043), PT <20 = 44.5% (versus 40.3%,⁴ P = 0.3970); SVA <4.5 cm = 59.2% (versus 54.9%,⁴ P = 0.4158) (Table 4). Segmental alignment offsets stratified by vertebral level are presented in Table 5. The planned LL was more reliably achieved in patients in whom two or more PIB devices were used compared to those with only one (P = 0.019; Figure 2). An example of a 3D preoperative plan and achieved alignment is shown in Figure 3, while a detailed case report is shown in Figure 4.

Table 2.

Baseline, Planned, and Postoperative Radiographic Alignment Measures

Radiographic parameter	Baseline (mean ± SD)	Plan (mean ± SD)	Postop (mean ± SD)	P-value (Postop vs Preop)	P-value (Postop vs Plan)
LL_L1-S1 (°)	40.2 ± 16.9	46.5 ± 15.8	49.3 ± 12.7	<0.001	0.007
DLL_L4-S1(°)	34.1 ± 11.4	37.8 ± 10.8	37.8 ± 10.3	<0.001	0.733
\|PI-LL\| (°)	17.1 ± 13.5	13.0 ± 11.5	12.1 ± 9.0	<0.001	0.237
L1PA (°)	12.0 ± 8.5	7.4 ± 6.8	7.6 ± 6.1	<0.001	0.576
IVL (°)	8.6 ± 6.8	11.1 ± 6.0	11.6 ± 5.6	<0.001	0.671
IVL (°) - ALIF	10.4 ± 6.9	14.9 ± 5.2	14.4 ± 5.2	<0.001	0.281
IVL (°) - LLIF	3.5 ± 3.9	5.4 ± 4.0	7.1 ± 3.5	<0.001	0.003
IVL (°) - TLIF	11.0 ± 6.0	11.5 ± 4.0	12.0 ± 4.7	0.113	0.202
IVCA (°)	0.2 ± 5.4	−0.3 ± 3.1	0.4 ± 3.1	0.877	0.218
IVPH (mm)	2.6 ± 1.9	6.8 ± 3.5	7.0 ± 1.9	<0.001	0.779

SD = standard deviation; LL = lumbar lordosis; DLL = distal lumbar lordosis; PI-LL = pelvic incidence – lumbar lordosis mismatch; L1PA = L1 pelvic angle; IVL = intervertebral lordosis; ALIF = anterior lumbar interbody fusion; LLIF = lateral lumbar interbody fusion; TLIF = transforaminal lumbar interbody fusion; IVCA = intervertebral coronal angle; IVPH = intervertebral posterior height.

Table 3.

Baseline, Planned, and Postoperative Radiographic Alignment Measures, by Level Treated

Radiographic parameter	Baseline (mean±SD)	Plan (mean±SD)	Postop (mean±SD)	P-value (Postop vs Preop)	P-value (Postop vs Plan)
IVL (°) - L1/L2	2.0±2.6	5.6±3.6	6.6±3.3	<0.001	0.252
IVL (°) - L2/L3	3.2±4.1	5.1±4.4	7.1±3.5	<0.001	0.032
IVL (°) - L3/L4	4.4±3.9	6.5±4.1	8.1±3.5	<0.001	0.069
IVL (°) - L4/L5	8.2±4.8	11.3±4.0	10.4±3.5	0.002	0.078
IVL (°) - L5/S1	13.1±6.5	15.2±4.9	15.9±4.9	<0.001	0.239
IVCA (°) - L1/L2	−2.1±4.4	−2.2±3.9	0.2±4.5	0.156	0.083
IVCA (°) - L2/L3	0.2±6.4	−0.6±3.8	0.2±3.9	0.992	0.059
IVCA (°) - L3/L4	4.1±4.3	1.0±2.5	1.7±1.7	0.034	0.193
IVCA (°) - L4/L5	0.1±6.2	−0.6±2.8	−0.2±3.1	0.588	0.347
IVCA (°) - L5/S1	−0.6±3.9	0.1±2.5	0.5±2.2	0.035	0.113
IVPH (mm) - L1/L2	2.2±1.6	7.5±1.2	6.7±1.8	<0.001	0.152
IVPH (mm) - L2/L3	1.8±1.8	7.3±1.3	6.9±2.2	<0.001	0.447
IVPH (mm) - L3/L4	1.9±1.4	7.1±1.4	7.1±1.9	<0.001	0.861
IVPH (mm) - L4/L5	2.4±1.5	6.8±1.6	7.0±2.0	<0.001	0.477
IVPH (mm) - L5/S1	3.1±2.1	6.4±5.1	6.9±1.7	<0.001	0.456

SD = standard deviation; IVL = intervertebral lordosis; IVCA = intervertebral coronal angle; IVPH = intervertebral posterior height.

Figure 1.

Bland-Altman plots illustrating the relationship of achieved and planned (goal) lumbar lordosis (LL) [top-left], distal lumbar lordosis (DLL_L4-S1) [top-right], pelvic incidence – lumbar lordosis mismatch (PI-LL) [bottom-left], and L1 pelvic angle (L1PA) [bottom-right], respectively. The mean of each pair of measures is plotted on the X axis and the difference between each pair of measures is plotted on the Y axis. Gray dashed lines indicate the bias or mean of differences. Red dashed lines indicate the 95% limits of agreement

Table 4.

Percentage of Patients Meeting Target Postoperative Radiographic Alignment Thresholds (per SRS Classification²¹)

Radiographic parameter	Current study	Lafage et al⁴	P-value
Post-op coronal deformity type, %
N (no coronal curve; all <30°)	88.6%	73.0%	<0.001
T (thoracic-only curve, lumbar <30°)	2.6%	4.0%	0.464
L (lumbar-only curve, thoracic <30°)	6.1%	8.8%	0.330
D (double curve, both >30°)	2.6%	7.0%	0.073
Post-op PI-LL modifier, %
0 (PI-LL <10°; non-pathologic)	49.1%	63.0%	0.004
+ (10°< PI-LL <20°; moderate deformity)	34.8%	17.0%	<0.001
++ (PI-LL >20°; marked deformity)	16.1%	12.8%	0.329
Post-op PT modifier, %
0 (PT <20°; non-pathologic)	44.5%	40.3%	0.397
+ (20°< PT <30°; moderate deformity)	45.5%	33.8%	0.015
++ (PT >30°; marked deformity)	10.0%	18.7%	0.024
Post-op SVA modifier, %
0 (SVA <4 cm; non-pathologic)	59.2%	54.9%	0.416
+ (4 cm< SVA <9.5 cm; moderate deformity)	33.7%	26.2%	0.112
++ (SVA >9.5 cm; marked deformity)	7.1%	11.6%	0.179

PI-LL = pelvic incidence – lumbar lordosis mismatch; PT = pelvic tilt; SVA = sagittal vertical axis.

Table 5.

Segmental Corrections Achieved Versus Planned

Level	n, %	IVL (°)	IVCA (°)	IVPH (mm)
L1/L2	17, 8	1.0 ± 3.4	2.4 ± 5.2	−0.8 ± 2.1
L2/L3	26, 12	1.9 ± 4.2	0.8 ± 2.0	−0.4 ± 2.4
L3/L4	23, 11	1.6 ± 3.9	0.8 ± 2.5	0.1 ± 2.6
L4/L5	62, 29	−0.8 ± 3.5	0.4 ± 2.9	0.2 ± 2.2
L5/S1	84, 40	0.5 ± 4.0	0.4 ± 2.5	0.5 ± 5.5

n = number; IVL = intervertebral lordosis; IVCA = intervertebral coronal angle; IVPH = intervertebral posterior height.

Figure 2.

Graphs illustrating the average offsets of achieved and planned (goal) lumbar lordosis (LL_L1-S1) [left] and distal lordosis (DLL_L4-S1) [right] when patients were treated with one or more levels of personalized interbody implants

Figure 3.

Patient example illustrating achieved vs planned alignment for lumbar lordosis (LL), lordosis from L4-S1 (DLL_L4-S1), pelvic incidence (PI) - LL mismatch (PI-LL), and L1PA compared to ideal calculated as (L1PA = 0.5 x PI-21).²² Patient is a 64-year-old woman with global sagittal malalignment (adult spinal deformity with fractional curve at lumbosacral junction). Shown are antero-posterior (AP) and lateral preoperative and postoperative radiographs (A); the 3D preoperative planning models, antero-posterior (AP) and lateral 3D reconstruction views (B) of preoperative (blue) and planned (gray) alignment; (C) details of the intervertebral asymmetry in the levels treated preoperative and restored in the planning after placement of the personalized interbody implant, restoring the alignment and providing indirect decompressions in the foramen; (D) the asymmetrical morphology of the ALIF implants used in the case to fill the intervertebral space and provide the planned alignment; and (E) the alignment parameters measured compared to the planned and ideal values achieved for LL, DLL_L4-S1, PI-LL, and L1PA

Figure 4.

Example of an individual postoperative case. Postoperative data, combined with preoperative, planned parameters and implant specifications, are stored in a centralized database to generate feedback and enhance the 3D AI-enabled planning platform. This process supports continuous refinement of the surgical plan and alignment targets based on the patient’s specific anatomy, preoperative condition, and the alignment outcomes achieved by each individual surgeon to better adjust their preoperative plans

Revisions Required for Mechanical Complications

Between postoperative day-one (POD1) and 2 years follow-up, five PIB cases required revision for mechanical complications. One case involved the fracture of a sacroiliac screw between postoperative weeks 4 and 12, and four cases experienced proximal junctional kyphosis (PJK), one prior to postoperative 6 weeks and three between postoperative 6 weeks and 2 years. Neither demographics (age nor sex) nor treatment details (number or location of interbody fusion levels or approach) were factors in the incidence of mechanical complications leading to revision in this cohort. Mean age was 62 years in those with revision and 69 years in those without (Mann-Whitney U Test P = 0.102); 60% of those with revision were female vs 63% of those without (Chi-square P = 0.869). The five revision cases comprised one 1-level, two 2-level, and two 3-level interbody fusion (IBF) cases, with four total levels of ALIF, two of LLIF, and five of TLIF (P > 0.05). In the Lafage study, 22 of the 166 revisions performed for mechanical complications were performed between POD1 and postoperative 6 weeks, while 144 occurred between postoperative 6 weeks and 2 years.⁴ Compared to the ISSG cohort,⁴ utilization of 3D preoperative planning and PIB implants was associated with significantly fewer revisions for mechanical complications up to 2 years postoperatively: 5/115 (4.3%) vs 166/997 (16.6%), P < 0.001 (Tables 6 and 7).

Table 6.

Comparative Incidence of Revisions due to Mechanical Complications up to Two Postoperative Years

	Current study (n = 115)	Lafage et al (n = 997)⁴	P-value
Implant failure^a, number (%)	1 (0.9%)	65 (6.5%)	0.016
Radiographic imbalance^b, number (%)	4 (3.5%)	101 (10.1%)	0.022
Total, number (%)	5 (4.3%)	166 (16.6%)	<0.001

^aImplant failure included interbody dislodgement, loose dislodgement, rod breakage, rod dislodgement, screw breakage, and screw loosening.²⁰

^bRadiographic imbalance included: adjacent segment degeneration, coronal imbalance, distal junctional kyphosis, proximal junctional kyphosis, pseudoarthrosis, and sagittal imbalance.²⁰

Table 7.

Comparative Incidence in Revisions due to Mechanical Complications, by Postoperative Time Period of the Complication

	Current study (n = 115)	Lafage et al (n = 997)⁴	Difference in incidence (95%CI)
POD1 to 6 weeks, number (%)	1 (0.9%)	22 (2.2%)	−1.3% (−0.033, 0.006)
6 weeks to 1 year, number (%)	4 (3.5%)	87 (8.7%)	−5.2% (−0.090, −0.015)
1 year to 2 years, number (%)	0 (0%)	57 (5.7%)	−5.7% (−0.072, −0.043)
Overall, number (%)	5 (4.3%)	166 (16.6%)	−12.3% (−0.167, −0.079)

CI = confidence interval; POD1 = postoperative day one; Difference in incidence between POD1 and 1-year follow-up is −6.6% (−0.108, −0.024). Difference in incidence between 6 weeks and 2-year follow-up is −10.9% (−0.150, −0.070).

On the basis of Kaplan-Meier analysis, the percentage of patients remaining free of mechanical complications leading to revision was 100%, 98.3 ± 1.2%, 95.7 ± 1.9%, and 95.7 ± 1.9% at POD-30, 6 months, 1 year, and 2 years, respectively (Figure 5).

Figure 5.

Kaplan-Meier survival plot showing cumulative proportion of patients free from mechanical complications that required revision at any time period between postoperative day one and 2-year postoperative follow-up. The line represents the survival rate and the shading the standard error of the estimate

Discussion

The importance of achieving planned spinal alignment in ASD surgery has been increasingly emphasized, as it directly affects both clinical and radiographic outcomes, including mechanical complication rates and long-term reoperation risks.^5-7,23-27 Diebo et al (2024) reported that residual malalignment postoperatively is associated with worse health outcomes and higher revision rates in ASD patients.²³ A retrospective cohort study by Krol et al (2021) showed that 317/762 (42%) of patients exhibited radiographic complications, which negatively impacted Oswestry disability index (ODI) and Scoliosis Research Society (SRS) scores more than any other complication. They concluded that the most detrimental contributors to poor long-term outcomes were almost exclusively related to poor radiographic correction, loss of correction postoperatively, and mechanical failure.²⁴ A multicenter study of 891 ASD revision cases by Passias et al. found that the primary reason for revision was mechanical complications, and that patients requiring revision due to radiographic malalignment were 93.0% less likely to reach minimum clinically important difference (MCID), indicating persistent disability and functional limitations.²⁵ A 2022 study to investigate the prevalence of decisional regret in ASD patients reported that one-in-five regretted their decision to undergo spinal deformity surgery, and almost twice as many patients in the high-decisional-regret group experienced a postoperative complication compared with the low-decisional-regret group.²⁶ A study by Du et al (2024) showed that in multivariate analysis, only revision surgery was independently associated with increase in risk for medium–high decisional regret (P = 0.041). Remarkably, among patients receiving a lumbar fusion for degenerative conditions, a significantly higher number of patients undergoing revision fusion (29.4%) exhibited high regret compared to the number of primary fusion patients exhibiting regret (5.6%), (P = 0.026).²⁷ Although the current study design and associated consents did not allow for the inclusion of patient-reported outcomes measures (PROMs), these prior studies highlight the potential clinical value of improved postoperative alignment and sparing of patients from mechanical complications that lead to revision.

Despite advances in surgical planning and techniques, mechanical complication rates remain a significant clinical and economic burden following ASD correction. In addition to the negative affect on patient-reported clinical outcomes as discussed above,^5-7,23-27 Theologis et al. reported that revision operations for proximal junctional failure after long thoracolumbar fusions for ASD are associated with an average direct cost of more than $55,000 per case.²⁸ And McCarthy et al found that the incidence of readmissions increased the average cost of ASD surgery by more than 70%.²⁹ Although this study did not analyze cost and cannot infer cost effectiveness from its results, there’s little debate that avoiding the need for reoperation intuitively reduces costs compared to similar cases where reoperation is required.

The current study compared a cohort of ASD patients using PIB devices to a historical literature control using the same radiographic inclusion criteria. To the authors’ knowledge at the time of analysis, the Lafage et al (2024) study represented the largest reported cohort with the most robust and detailed account of complications in this challenging patient population, with surgical treatment by some of the most experienced spine deformity surgeons in the US.⁴ The data reported by Lafage et al (2024) included both the chronology and details of complications and revisions for 4 time points: intraoperative, zero to 6 weeks, 6 weeks to 1 year, and 1 year to 2 years postoperative.⁴ In the full 2-year postoperative period, the all-cause revision rate was 24.0%, with 69% of these revisions, or 166/997 (16.6% of the total study cohort) attributable to mechanical complications.⁴ The depth of detail in the Lafage et al (2024) study provides a basis for comparison to the present study, in which the same methodology is applied, and the cohorts are similar. It also was a more challenging comparator given its relatively low reported rates of complication compared to the published literature on the topic. By comparison, other smaller, single-center studies have shown similar or higher revisions rates due to mechanical complications at 2 years. Bari et al. (2020) studied 233 ASD patients from a single institution having a mean 36-month follow-up (range 22-50 months) and estimated the cumulative incidence of revision surgery due to mechanical failure at 2 years following surgery to be 28%.³⁰ Teles et al. (2022) reported a 32.1% revision rate for mechanical complications in a single-institution cohort of 84 ASD patients, concluding that causative factors are likely multifactorial, with elevation of postoperative SRS-Schwab Classification sagittal modifiers of PI-LL mismatch, PT, and SVA as positive predictors of mechanical complications.⁵ The current study’s results showed the PIB cohort to have a higher rate of moderate postoperative mismatch and PT, but the Lafage et al cohort had a higher rate of marked postoperative PT. Postoperative SVA modifiers were similar between cohorts. These results may reflect a correction strategy in the PIB series focused on achieving physiologic rather than absolute alignment targets, representing an intentional effort to avoid overcorrection in older spines with more severe sagittal malalignment (higher preoperative PT and SVA). A more recent retrospective study by Compagnone et al. (2025), which applied inclusion criteria similar to those used in the current analysis but at a single institution, reported 23% revision rate at 2-year follow-up, with the nature of mechanical complications requiring revision being junctional failure (14.5%) and implant failure (8.5%).³¹ The authors emphasized that to reduce complication rates, surgeons should focus on restoring appropriate maximum lumbar lordosis and ensuring proper lordosis distribution supported by adequate preoperative surgical planning and intraoperative reconciliation to the plan.³¹ The statistically significant lower rate in the current study suggests that personalized planning combined with anatomy-matched implants may contribute to overcoming the persistent challenge of mechanical complications.

Several predictive models have been developed to analyze a multitude of variables, including spinopelvic parameters, to identify risk factors that could lead to reoperation following ASD surgery.^32-35 Among those, a growing body of research has shown that not only the overall magnitude of lordosis but also its regional distribution, specifically between the upper (L1-L4) and lower (L4-S1) lumbar segments plays a significant role in minimizing postoperative complications and the need for revision surgeries.^4,23,36-38 Establishing a biomechanically stable lumbosacral foundation and optimal DLL alignment has been shown to significantly reduce mechanical complications.²³ This reinforces the concept that both global and segmental regional alignment should be carefully planned and executed.

Achieving the necessary alignment to address a patient’s condition and improve the chances of a successful outcome requires careful planning and precise execution of the surgery. In a 2024 study published by Smith et al¹² a group of experienced surgeon investigators of the ISSG measured planned vs achieved alignment in 266 complex adult spinal deformity patients having a mean number of 14 levels fused. The study showed that at 6 weeks postoperative only 31.5% of cases achieved the targeted PI-LL mismatch within 5°, with a mean overcorrection of 4.6°. A similar percentage of cases (30.4%) missed the planned PI-LL mismatch by > 15°. This data demonstrates that even highly experienced surgeons are frequently unable to correct alignment when stock cages are used. The authors noted that, although patient-specific rods may help to lock in desired alignment, deformity correction primarily occurs through osteotomies and through release and realignment of the disc spaces, where interbody fusion devices are placed.

Smith et al (2025)¹⁸ subsequently assessed whether PIB devices are associated with improved rates of achieving goal alignment following adult spinal deformity (ASD) surgery. The authors assessed the effectiveness of personalized interbody devices in achieving goal alignment following ASD surgery based on a multicenter cohort of 65 patients and compared the rates of achieving goal alignment to their previously published study using stock interbody implants, using the same radiographic inclusion and same core lab for measurements. Data describing achieved vs planned PI-LL mismatch was compared to a radiographically matched patient group receiving personalized interbody devices. The comparison showed that the percentage of cases achieving the targeted PI-LL mismatch within 5° increased by 41.6% (44.6% vs 31.5%) vs the cases using stock devices (P = 0.046) and there was a 50% decrease in cases in which the planned PI-LL mismatch was missed by > 15° vs in the ISSG study (P = 0.012).

A recent large cohort study has shown that over-correction of proximal lumbar lordosis increases the risk of PJK, while under-correction of distal lordosis raises the rate of revision for implant failure.²³ Lafage et al. (2018) conducted a propensity matched analysis of 312 PJK patients, finding that lordosis increases concentrated at the upper lumbar levels (L1–L3) significantly increased the risk of PJK, while restoration of lordosis at the lower lumbar spine (L4–S1) reduced the risk of PJK.³⁹ They concluded that creating excess proximal lumbar lordosis does not compensate for hypo-correction in the distal lumbar region, and restoration of distal lumbar lordosis may result in a decreased risk of developing PJK. In the present study’s cohort, the mean L4–S1 lordosis increased significantly from 34.1° (±11.4°) preoperatively to 37.8° (±10.3°) postoperatively (P < 0.001), closely matching the planned goal of 37.8° (±10.8°), with a mean offset of −0.2° (P = 0.733). These findings suggest that the use of 3D preoperative planning and PIB devices facilitated a more harmonious and precise restoration of the distal lumbar lordosis, potentially contributing to a significant reduction in both radiographic imbalance (malalignment) and implant failure translating to a 74% reduction in revisions due to mechanical complications compared to the Lafage cohort.⁴

Even though a single vertebral-pelvic angle alone may not fully capture clinical benefit, L1PA was measured against the ideal value proposed by Hills et al²² to evaluate its achievement and potential contribution to reducing reoperation rates. The fact that it closely matches the ideal value in this cohort suggests that it may play a complementary role, alongside the other alignment parameters included in the planning process, in reducing the incidence of mechanical complications requiring revision surgery.

Regarding intervertebral alignment, there was a statistically significant increase in IVL from baseline to postoperative measurements, indicating effective segmental correction. At the levels where anterior lumbar interbody fusion was performed, a significant gain in intervertebral lordosis was achieved, and the difference between the planned and achieved alignment was not statistically significant (P = 0.281), indicating close accuracy to the plan. The lateral lumbar fusions also showed an improvement in lordosis, with an offset of 1.7° compared to the plan (P = 0.003), indicating a slight tendency towards over-correction. The TLIFs did not show a significant increase in lordosis in this cohort; they tended to only slightly increase or maintain intervertebral lordosis.

The implantation of two or more PIB devices per patient was associated with better achievement of lumbar lordosis correction. Patients treated with multiple PIB implants demonstrated a mean offset of only 0.5° between achieved and planned lumbar lordosis, compared to a 5° offset when a single interbody device was used (P = 0.009). This is similar to a finding by Smith, et al (2023) in which planned vs achieved PI-LL mismatch demonstrated a trend toward a closer achievement of goal when two or more levels of PIB devices were used.¹⁸ That study may not have been sufficiently powered to demonstrate statistical significance, whereas the larger cohort of personalized implant patients in the present study resulted in a statistically significant finding. Conversely, multiple studies have found that in both one- and two-level lumbar interbody fusion procedures, changes in segmental lumbar lordosis were not significantly different between standard (6°) and hyperlordotic (20°) cages, demonstrating the unpredictability of alignment obtained when stock devices are used.^40,41

As these studies indicate, multilevel anterior support can effectively achieve significant changes in lumbar alignment, potentially reducing the need for more aggressive focal correction techniques such as osteotomies, thereby contributing to a more gradual distribution of lordosis.^13,23 However, achieving optimal anterior column alignment with the use of stock interbody devices remains a challenge. Oikonomidis et al. showed that among 138 patients, a flat non-lordotic cage produced the predicted no change in lordosis (average pre-op 27°and post-op 26°), however a 10° lordotic cage produced an average additional lordosis of only 1°.⁴² Lovecchio et al. showed that a 10° cage produced an incremental lordosis of only 1.6°, and among 57 patients, a 20° cage produced only 3.4° of added lordosis.⁴⁰ Mathew et al. evaluated 53 patients who underwent fusion using 6° cages, showing an average incremental lordosis of 10.9° while the patients receiving 20° cages demonstrated an average incremental lordosis of only 8.6°.⁴¹

Vertebral endplate variability is one potential explanation for the significant discrepancy between the lordotic angle of stock cages and the intervertebral lordosis they create. Stock devices, which are essentially flat and uniform on the upper and lower surfaces, do not achieve a precise fit against the irregular bony topography of vertebral endplate surfaces. Vertebral endplate abnormalities are common in lumbar fusion patients, especially the elderly. In a study of 1564 endplates in 133 subjects with Modic changes on MRI, 27.8% of all endplates exhibited defects, with 31% of L4-L5 and 49% of L5-S1 endplates exhibiting a defect.⁴³

Sadrameli et al. conducted a retrospective study of 217 patients with spinal deformity or degenerative conditions undergoing IBF with PIB devices.¹³ The desired intervertebral lordosis angle (IVL) was prescribed into the device design for each personalized device (IVL goal). Among the 365 treated levels, IVL offset was 1.1° ± 4.4° (mean ± SD). The targeted intervertebral lordosis was achieved within 5° of plan in 299/365 levels (81.9%): 85.9% of LLIFs, 82.6% of TLIFs, and 78.6% of ALIFs. Only ten levels (2.7%) missed the planned IVL by > 10°. Interestingly, among 41 TLIF levels at L4-5, an average increase in IVL lordosis of 3.8° was observed, representing a strong association to the planned lordosis increase of 4° and among 29 TLIF levels at L5-S1, an average increase in IVL lordosis of 2.6° was observed against a planned lordosis increase of 4.4°. These results are a stark contrast to the previously reported unpredictability of alignment achieved with stock devices.^40-42

In addition to the potential to improve segmental correction, studies specific to evaluating customized endplate fit have demonstrated multiple attributes specific to improving the interbody fusion/mechanical environment. For example, a customized endplate fit has been shown to provide greater contact area between the interbody device and the endplate, which distributes the stress more evenly, significantly reducing stress on the endplates^15,44,45 as well as the supra- and infra-adjacent disc spaces,¹⁷ and decreasing stress in the posterior rods.¹⁵ The contribution of interbody device endplate contact area in fact outweighs the contribution of cage material to preventing subsidence.^15,44,46 Taken together, favorable endplate dynamics and precise segmental alignment correction may help explain a lower incidence of mechanical complications leading to revision when PIB devices are used.

The combination of a personalized 3D surgical plan and custom-made anatomically designed interbody devices to achieve the plan appears to be a significant step toward reliably achieving patient-specific alignment goals, which may explain the reduced rate of reoperations due to mechanical failures in the current study. In addition, cases using PIB devices may be postoperatively analyzed to assess planned vs achieved alignment, with postoperative data integrated into an AI-enabled 3D planning platform to further improve the “surgeon-specific” aspect of surgical planning. This continuous learning loop allows surgeons to analyze their own outcomes and make informed adjustments to optimize future plans, tailoring them to their individual techniques and the specific implant types used, correlating to the final achieved alignment (Figure 4).

Limitations

As a retrospective study with literature-based control, the study design is an inarguably imperfect comparison. Potential confounders exist, including the risk of biases related to patient selection, surgeon experience, and treatment strategy. Propensity-score matching across the 2 cohorts was not possible due to lack of access to the historical control’s raw data. However, the current study sought to mitigate some of the potential biases by matching the radiographic inclusion criteria of the Lafage et al study⁴ and including all consecutive cases meeting the inclusion criteria (ie, no learning-curve cases were excluded). The comparison was further benefitted by the majority case contribution by coincident surgeons in both studies. Additionally, utilization of personalized surgery digital planning software in the current study likely helped to provide further uniformity in applying current evidence-based guidelines for surgical alignment correction. Given the shared radiographic inclusion criteria, the 2 cohorts were shown to be largely comparable in overall deformity severity. Preoperatively, both groups exhibited similar coronal alignment distributions, with minor differences unlikely to represent clinically meaningful discrepancies in baseline severity.^47,48 In the sagittal plane, the 2 cohorts were similar in baseline deformity severity, with the PIB series displaying only a slightly greater degree of sagittal compensation, reflected by higher PT and SVA modifiers, which is consistent with these surgeon-authors’ tendency to select personalized implant constructs for more complex cases.

There were minor differences in demographics between cohorts, including older age (average 60.4 vs 68.8 years) and more females (76% vs 63%) in the PIB cohort. The published literature on the topic of age as a factor in mechanical complications leading to revision is inconclusive. While some studies report that younger age may lead to higher risk of revisions,³⁵ others show no effect of age,^32,49,50 and yet others report that older age increases risk.^33,51,52 Indeed, older age is associated with reduced bone density, higher frailty, and increased risk of pseudarthrosis or mechanical failure. Moreover, age was not reported in the Lafage et al study to be a factor in their outcomes and was determined to be insignificant in the revision rate in the current study’s PIB cohort. Also, while both cohorts reflect the female predominance typical of ASD, the modest difference in sex distribution between cohorts is unlikely to meaningfully affect mechanical complication risk, as the published literature on the topic is similarly inconclusive. Varshneya et al. (2022) showed a lower risk of revision surgery,⁵³ while Yagi et al (2019) showed a 2.9 odds ratio increased risk for males.³³ Again, the Lafage et al study did not identify sex as a confounding factor in their results, while the current study’s results show that patient sex was not a confounding factor in the PIB cohort. Yet, despite including an older and potentially higher-risk population, the current study’s PIB cohort did not demonstrate higher revision rates, which lends support to the argument that the observed differences in outcomes are not driven by demographic advantage.

In terms of treatment strategy and treatment execution details, the 2 cohorts had a similar median number of levels of fusion but differed in the use of interbody fusion overall (100% in the PIB cohort vs 64% in the Lafage et al cohort), and in the distribution of IBF approaches (more ALIF and LLIF procedures in the PIB group). While these variables proved to be insignificant to the primary outcome in the current study’s cohort, it is worth noting that the increased use of IBF may be protective – particularly against pseudarthrosis and rod fracture,³ and anterolateral approaches, in particular, are lordosis-restoring,^54,55 factors which may influence the personalized surgical planning of these cases requiring customized segmental corrections. Despite this potential advantage, however, current utilization data shows that the use of IBF devices in lumbar fusions is not outpacing posterior-only lumbar fusion procedures as an evolution of best practices;⁵⁶ rather, surgeons continue to choose the strategy that provides the greatest chance of a successful outcome based on their experience.

Indeed, application of best practices is expected to evolve as evidence mounts and new technologies emerge. Precisely because of these differences, the current comparison offers some insight into the potential benefits of the collective planning and execution strategy. Not all treatment details were available for detailed comparison, however. It is important to note that the posterior procedures in the present study’s cohort involved a mixture of personalized and stock rods. While PIB devices provided anterior column support and alignment, the variability introduced by posterior fixation could affect the overall precision in achieving global alignment goals. Further investigations comparing fully personalized spine constructs (personalized rods and personalized interbody implants) to constructs using PIB implants and manually bent rods are warranted. Such studies could determine whether the personalization of the entire construct enhances overall alignment predictability and long-term durability. Similarly, neither this study nor the Lafage et al study differentiated data on mechanical complications based on proximal transition strategies, such as the use of posterior tethering, hook constructs, or prophylactic vertebroplasty, or supplemental rod utilization. These strategies have been shown to influence the risk of PJK, PJF, and rod fractures, although the findings continue to be debated as heterogenous and conflicting.^57-59 The absence of this data is an important limitation. Future studies involving larger cohorts will allow for an evaluation of subgroups to assess the respective contributions of proximal transition techniques and achieved distal lordosis goals through PIB correction toward reducing complications due to proximal junctional failure.

Future studies may also benefit from propensity-score matching to better address some of the above differences in demographics, deformity severity, and treatment details to strengthen the understanding of how optimizing spinal alignment influences the occurrence of mechanical complications and the need for reoperation. Lack of access to the raw data of the historical cohort prohibited such matching in this retrospective study. However, it should be noted that propensity-score matching would match cohorts based on known or presumed covariates, while the current study found no confounders for the incidence of revisions due to mechanical complications. While all-cause revision was not the focus of this study, matching over broader clinical variables such as diagnosis and preoperative complaints/dysfunction may be useful to evaluate the incidence of all revisions such as might be required to address poor clinical outcome/unresolved symptoms. Moreover, matching the control study’s inclusion criteria resulted in demonstrably similar cohorts with respect to deformity severity, lending reliability to the comparison for this study’s primary outcomes of postoperative alignment achievement and the incidence of mechanical complications leading to revision. While the comparative outcomes suggest a positive correlation between improved postoperative alignment and reduced mechanical complications leading to revision when PIB devices are used, given the above limitations, causality between the use of PIB devices and reduced reoperation rates cannot be established in this retrospective design.

Conclusion

Recent multicenter analyses have demonstrated that 3D preoperative planning combined with personalized anatomically designed interbody devices contribute to more reliably achieving alignment goals, particularly in complex ASD surgeries. The current study further supports this finding, with minimal offset between planned and achieved alignment targets, and a significantly reduced reoperation rate due to mechanical complications compared to previously published results from a large database of ASD cases performed by highly experienced surgeons using stock implants. These findings suggest that, as spine surgery technologies evolve, personalized 3D preoperative planning and patient-specific interbody implants offer a promising strategy to address the alignment-related shortcomings of traditional instrumentation, which may improve the clinical outlook and reduce the economic burden of revisions in these challenging cases.

Footnotes

Acknowledgments

The authors would like to thank Michele Temple-Wong, PhD and Kannika Dailey at Carlsmed for their statistical support.

ORCID iDs

Justin S. Smith

Kelli M. Howell

Jeffrey P. Mullin

Ethical Considerations

This analysis utilized secondary research consisting of deidentified data for which consent is not required and was therefore exempt from institutional review board review under 45 CFR §46.104(d)(4)(ii). No direct patient involvement occurred.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by Carlsmed.

Declaration of Conflicting Interests

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: One or more authors declare potential competing financial interests or personal relationships as specified on required ICMJE disclosure forms.

References

Scheer

Tang

Smith

, et al. Reoperation rates and impact on outcome in a large, prospective, multicenter, adult spinal deformity database: clinical article. J Neurosurg Spine. 2013;19(4):464-470.

Smith

Klineberg

Lafage

, et al. Prospective multicenter assessment of perioperative and minimum 2-year postoperative complication rates associated with adult spinal deformity surgery. J Neurosurg Spine. 2016;25(1):1-14.

Burke

Scheer

Lau

, et al. Failure in adult spinal deformity surgery: a comprehensive review of current rates, mechanisms, and prevention strategies. Spine. 2022;47(19):1337-1350.

Lafage

Bass

Klineberg

, et al. Complication rates following adult spinal deformity surgery: evaluation of the category of complication and chronology. Spine. 2024;49(12):829-839.

Teles

Aldebeyan

Aoude

, et al.

Mechanical complications in adult spinal deformity surgery: can spinal alignment explain everything?

Spine. 2022;47(1):E1-E9.

Lafage

Passias

Sheikh Alshabab

, et al. Patterns of lumbar spine malalignment leading to revision surgery for proximal junctional kyphosis: a cluster analysis of over-versus under-correction. Glob Spine J. 2023;13(7):1737-1744.

Quarto

Zanirato

Vitali

, et al. Adult spinal deformity correction surgery using age-adjusted alignment thresholds: clinical outcomes and mechanical complication rates. A systematic review of the literature. Eur Spine J. 2024;33(2):553-562.

Yamato

Hasegawa

, et al. The validation study of preoperative surgical planning for corrective target in adult spinal deformity surgery with 5-year follow-up for mechanical complications. Eur Spine J. 2022;31(12):3662-3672.

Hills

Weisenthal

Wanner

, et al. A patient-specific approach to alignment and proximal junctional kyphosis risk assessment in adult spinal deformity surgery: development and validation of a predictive tool. Clin Spine Surg. 2022;35(6):256-263.

10.

Park

Lee

Park

Jeon

. A validation study of four preoperative surgical planning tools for adult spinal deformity surgery in proximal junctional kyphosis and clinical outcomes. Neurosurgery. 2023;93(3):706-716.

11.

Khalifé

Skalli

Vergari

, et al. Redefining physiological whole-body alignment according to pelvic incidence: normative values and prediction models. Eur Spine J 2025 Apr 22;1-2.

12.

Smith

Elias

Sursal

, et al.

How good are surgeons at achieving their preoperative goal sagittal alignment following adult deformity surgery?

Glob Spine J. 2024;14(7):1924-1936.

13.

Sadrameli

Blaskiewicz

Asghar

, et al. Predictability in achieving target intervertebral lordosis using personalized interbody implants. Int J Spine Surg. 2024;18(S1):S16-S23.

14.

Esterberg

Casey

. Comparison between patient specific interbody devices versus stock devices to achieve the planned correction in the treatment of adult spinal deformity. In: NASS Annual Meeting, Innovative Technologies Session; 2020.

15.

Chatham

Patel

Yakacki

Carpenter

. Interbody spacer material properties and design conformity for reducing subsidence during lumbar interbody fusion. J Biomech Eng. 2017;139(5):051005.

16.

Patel

. Does Patient-Specific Implant Design Reduce Subsidence Risk in Lumbar Interbody Fusion? A Bottom up Analysis of Methods to Reduce Vertebral Endplate Stress. Doctoral dissertation. Denver: University of Colorado; 2018.

17.

Zhang

Yin

, et al. Can an endplate-conformed cervical cage provide a better biomechanical environment than a typical non-conformed cage? A finite element model and cadaver study. Orthop Surg. 2016;8:367-376.

18.

Smith

Mundis

Osorio

, et al. Analysis of personalized interbody implants in the surgical treatment of adult spinal deformity. Glob Spine J. 2025;15(2):930-939.

19.

Mullin

Asghar

Patel

, et al. Changes in alignment at untreated vertebral levels following short-segment fusion using personalized interbody cages: leveraging personalized medicine to reduce the risk of reoperation. Int J Spine Surg. 2024;18(S1):S32-S40.

20.

Klineberg

Wick

Lafage

, et al. Development and validation of a multidomain surgical complication classification system for adult spinal deformity. Spine. 2021;46(4):E267-E273.

21.

Schwab

Ungar

Blondel

, et al. Scoliosis research society—Schwab adult spinal deformity classification: a validation study. Spine. 2012;37(12):1077-1082.

22.

Hills

Lenke

Sardar

, et al. The T4-L1-hip axis: defining a normal sagittal spinal alignment. Spine. 2022;47(19):1399-1406.

23.

Diebo

Balmaceno-Criss

Lafage

, et al.

Lumbar lordosis redistribution and segmental correction in adult spinal deformity: does it matter?

Spine. 2024;49(17):1187-1194.

24.

Krol

Passias

Lafage

, et al. 138. Radiographic malalignment has a far greater impact on clinical outcomes than perioperative and postoperative complications in ASD surgery. Spine J. 2021;21(9):S69.

25.

Passias

Dave

Smith

, et al. Cause and effect of revisions in adult spinal deformity surgery: a multicenter study on outcomes based on etiology. Spine J. 2025;25(6):1229-1235.

26.

Adogwa

Caruso

Eldridge

, et al. Decisional regret among older adults undergoing corrective surgery for adult spinal deformity: a single institutional study. Spine. 2022;47(8):E337-E346.

27.

Lovecchio

Kazarian

, et al. Decisional regret following corrective adult spinal deformity surgery: a single institution study of incidence and risk factors. Spine Deform. 2024;12(3):775-783.

28.

Theologis

Miller

Callahan

, et al. Economic impact of revision surgery for proximal junctional failure after adult spinal deformity surgery: a cost analysis of 57 operations in a 10-year experience at a major deformity center. Spine. 2016;41(16):E964-E972.

29.

McCarthy

Hostin

Ames

, et al. Total hospital costs of surgical treatment for adult spinal deformity: an extended follow-up study. Spine J. 2014;14(10):2326-2333.

30.

Bari

Hansen

Gehrchen

. Surgical correction of adult spinal deformity in accordance to the Roussouly classification: effect on postoperative mechanical complications. Spine Deform. 2020;8(5):1027-1037.

31.

Compagnone

Cecchinato

Pezzi

, et al. How to reduce the risk of mechanical failures in adult deformity surgery: comparing GAP score and Roussouly type restoration. Glob Spine J 2025;15(8):3664-3678.

32.

Scheer

Smith

Schwab

, et al. Development of a preoperative predictive model for major complications following adult spinal deformity surgery. J Neurosurg Spine. 2017;26(6):736-743.

33.

Yagi

Hosogane

Fujita

, et al. Predictive model for major complications 2 years after corrective spine surgery for adult spinal deformity. Eur Spine J. 2019;28:180-187.

34.

Joshi

Lau

Scheer

, et al. State-of-the-art reviews predictive modeling in adult spinal deformity: applications of advanced analytics. Spine Deform. 2021;9(5):1223-1239.

35.

Ryu

, et al. Risk factors for unplanned reoperation after corrective surgery for adult spinal deformity. Bone Joint Res. 2023;12(4):245-255.

36.

Bari

Ohrt-Nissen

Hansen

Dahl

Gehrchen

. Ability of the global alignment and proportion score to predict mechanical failure following adult spinal deformity surgery—validation in 149 patients with two-year follow-up. Spine Deform. 2019;7(2):331-337.

37.

Bari

Heegaard

Bech-Azeddine

Dahl

Gehrchen

. Lordosis distribution index in short-segment lumbar spine fusion–can ideal lordosis reduce revision surgery and iatrogenic deformity? Neurospine. 2021;18(3):543-553.

38.

Nilssen

Narendran

Skaggs

, et al. Long-term reoperation risk of thoracic to pelvis instrumentation for spinal deformity: a longitudinal study of 7,062 patients. Eur Spine J. 2025;34(3):1034-1041.

39.

Lafage

Obeid

Liabaud

, et al. Location of correction within the lumbar spine impacts acute adjacent-segment kyphosis. J Neurosurg Spine. 2018;30(1):69-77.

40.

Lovecchio

Vaishnav

Steinhaus

, et al.

Does interbody cage lordosis impact actual segmental lordosis achieved in minimally invasive lumbar spine fusion?

Neurosurg Focus. 2020;49(3):E17.

41.

Mathew

Cerpa

Lee

, et al. Comparing hyperlordotic and standard lordotic cages for achieving segmental lumbar lordosis during transforaminal lumbar interbody fusion in adult spinal deformity surgery. J Spine Surg. 2021;7(3):318-325.

42.

Oikonomidis

Heck

Bantle

, et al. Impact of lordotic cages in the restoration of spinopelvic parameters after dorsal lumbar interbody fusion: a retrospective case control study. Int Orthop. 2020;44(12):2665-2672.

43.

Feng

Liu

Yang

Battié

Wang

. Lumbar vertebral endplate defects on magnetic resonance images: classification, distribution patterns, and associations with modic changes and disc degeneration. Spine. 2018;43(13):919-927.

44.

Patel

. Does Patient-Specific Implant Design Reduce Subsidence Risk in Lumbar Interbody Fusion? A Bottom Up Analysis of Methods to Reduce Vertebral Endplate Stress. Doctoral dissertation. Denver: University of Colorado; 2018.

45.

Wang

Kang

Shi

, et al. Investigation into factors affecting the mechanical behaviours of a patient-specific vertebral body replacement. Proc Inst Mech Eng H. 2018;232(4):378-387.

46.

Suh

Puttlitz

Lewis

Bal

McGilvray

. The effect of cervical interbody cage morphology, material composition, and substrate density on cage subsidence. J Am Acad Orthop Surg. 2017;25(2):160-168.

47.

Glassman

Berven

Bridwell

Horton

Dimar

. Correlation of radiographic parameters and clinical symptoms in adult scoliosis. Spine. 2005;30(6):682-688.

48.

Daubs

Lenke

Bridwell

, et al.

Does correction of preoperative coronal imbalance make a difference in outcomes of adult patients with deformity?

Spine. 2013;38(6):476-483.

49.

Puvanesarajah

Shen

Cancienne

, et al. Risk factors for revision surgery following primary adult spinal deformity surgery in patients 65 years and older. J Neurosurg Spine. 2016;25(4):486-493.

50.

Chan

Hai

Liu

Zhang

. Risk factors of unintended return to the operating room in adult spinal deformity. J Orthop Surg Res. 2021;16(1):240.

51.

Pitter

Lindberg-Larsen

Pedersen

Dahl

Gehrchen

. Revision risk after primary adult spinal deformity surgery: a nationwide study with two-year follow-up. Spine Deform. 2019;7(4):619-626.

52.

Barone

Giudici

Martinelli

, et al. Mechanical complications in adult spine deformity surgery: retrospective evaluation of incidence, clinical impact and risk factors in a single-center large series. J Clin Med. 2021;10(9):1811.

53.

Varshneya

Stienen

Medress

, et al. Risk factors for revision surgery after primary adult thoracolumbar deformity surgery. Clin Spine Surg. 2022;35(1):E94-E98.

54.

Bentz

Diaz

Pallotta

Lambrechts

. Comparison and clinical applicability analysis of all-posterior, combined anterior-posterior, and combined lateral-posterior approaches in adult spinal deformities: a narrative review. AME Surg J. 2025;5:16.

55.

Lechtholz-Zey

Ayad

Gettleman

, et al. Changes in segmental and lumbar lordosis following lumbar interbody fusion: a systematic review and meta-analysis. Clin Spine Surg. 2025;38(6):294-303.

56.

Orthopedic Network News . 2024 Spinal Surgery Update. Volume 35, Number 4, November 2024. Accessed at.https://onn-app.curvolabs.com/download/newsletter/7ac2b3df-8887-4505-983d-6b829540705f

57.

Shlobin

Scheer

Tan

. State of the evidence for proximal junctional kyphosis prevention in adult spinal deformity surgery: a systematic review of current literature. World Neurosurg. 2022;161:179-189.e1.

58.

Vercoulen

Doodkorte

Roth

de Bie

Willems

. Instrumentation techniques to prevent proximal junctional kyphosis and proximal junctional failure in adult spinal deformity correction: a systematic review of clinical studies. Glob Spine J. 2022;12(6):1282-1296.

59.

Sursal

Kim

Sardi

Yen

Smith

. Use of tethers for proximal junctional kyphosis prophylaxis in adult spinal deformity surgery: a review of current clinical evidence. Int J Spine Surg. 2023;17(S2):S26-S37.