Machine Learning-Guided LIV Selection for Adolescent Idiopathic Scoliosis

NCT07723053 | Not Yet Recruiting DMC

• 1 other identifier: NDTH-20260578
• Study Type: interventional
• Target: 300
• Locations: 1 country
• Sites: 1

Brief Summary

This study will evaluate whether a machine learning-based decision support model, called the Drum Tower Rule, can help surgeons select the lowest instrumented vertebra during corrective surgery for adolescent idiopathic scoliosis. Patients with Lenke type 1 or Lenke type 5 adolescent idiopathic scoliosis who are scheduled for posterior spinal fusion will be randomly assigned to one of two groups. In the model-guided group, surgeons will receive the model-predicted risk of postoperative distal adding-on and a recommendation for lowest instrumented vertebra selection. In the conventional-experience group, surgeons will select the lowest instrumented vertebra according to routine clinical experience and existing surgical principles, without access to the model output. All patients will receive standard posterior spinal fusion. The main outcome is the incidence of distal adding-on at 24 months after surgery, assessed by blinded radiographic reviewers.

Conditions

Adolescent Idiopathic Scoliosis; Lenke Type 1 Adolescent Idiopathic Scoliosis; Lenke Type 5 Adolescent Idiopathic Scoliosis

Keywords

Lowest Instrumented Vertebra; Distal Adding-on; Machine Learning; Surgical Decision Support; Posterior Spinal Fusion; Lenke Classification; Drum Tower Rule

Outcome Measures

Primary Outcomes (1)

• Incidence of Distal Adding-on

  Distal adding-on is defined as an increase of more than 5 degrees in the disc angle below the lowest instrumented vertebra compared with the immediate postoperative radiograph, or an increase of more than 5 mm in the translation of the vertebra below the lowest instrumented vertebra relative to the central sacral vertical line. The outcome will be assessed by blinded radiographic reviewers.

  24 months after surgery

Secondary Outcomes (1)

• Scoliosis Research Society-22 Score

  24 months after surgery

Other Outcomes (1)

• Incidence of Surgery-Related Complications

  From surgery to 24 months after surgery

Study Arms (2)

Model-Guided Group

EXPERIMENTAL

Participants in this group will undergo standard posterior spinal fusion. Before surgery, preoperative clinical and radiographic variables will be entered into the locked Drum Tower Rule machine learning model. The model will generate a predicted risk of distal adding-on and a recommendation for lowest instrumented vertebra selection. The surgeon will make the final decision after considering the model output and clinical judgment.

Other: Drum Tower Rule Machine Learning-Guided Decision Support; Procedure: Posterior Spinal Fusion

Conventional-Experience Group

ACTIVE COMPARATOR

Participants in this group will undergo standard posterior spinal fusion. The surgeon will select the lowest instrumented vertebra according to routine clinical experience and existing surgical principles. The Drum Tower Rule model output will not be provided to the surgeon for participants in this group.

Other: Conventional LIV Decision-Making; Procedure: Posterior Spinal Fusion

Interventions

Drum Tower Rule Machine Learning-Guided Decision Support OTHER

A locked machine learning-based decision support model will be used before surgery to estimate the risk of postoperative distal adding-on and provide a recommendation for lowest instrumented vertebra selection. The model output will be available to surgeons in the model-guided group only.

Model-Guided Group

Conventional LIV Decision-Making OTHER

The lowest instrumented vertebra will be selected by the surgeon according to routine clinical experience and existing surgical principles, without access to the Drum Tower Rule model output.

Conventional-Experience Group

Posterior Spinal Fusion PROCEDURE

All participants will undergo standard posterior spinal fusion using an all-pedicle screw instrumentation system.

Conventional-Experience Group; Model-Guided Group

Eligibility Criteria

• Age: 10 Years - 18 Years
• Sex: all
• Healthy Volunteers: No
• Age Groups: Child (0-17), Adult (18-64)

You may qualify if:

• Diagnosis of adolescent idiopathic scoliosis classified as Lenke type 1A or Lenke type 5C.
• Age 10 to 18 years, inclusive.
• Scheduled to undergo posterior spinal fusion using an all-pedicle screw instrumentation system.
• Planned selective thoracic fusion or selective lumbar fusion, with a clinical need for lowest instrumented vertebra decision-making.
• Availability of required preoperative standing full-spine radiographs and bending radiographs.
• Ability and willingness to complete the planned 24-month postoperative follow-up.
• Written informed consent provided by the participant and legal guardian.

You may not qualify if:

• \- Congenital scoliosis, neuromuscular scoliosis, syndromic scoliosis, or other non-idiopathic scoliosis.
• History of spinal trauma, spinal tumor, spinal tuberculosis, or spinal infection.
• Previous spinal surgery.
• Severe sagittal spinal deformity, such as Scheuermann disease, for which the study model is not applicable.
• Neurological symptoms or signs.
• Leg length discrepancy greater than 10 mm.
• The surgeon determines that there is no clinical equipoise for lowest instrumented vertebra selection because one option is clearly contraindicated for safety or anatomical reasons.
• Inability to complete follow-up or required study assessments.

Sponsors & Collaborators

• The Affiliated Nanjing Drum Tower Hospital of Nanjing University Medical School: lead

Study Sites (1)

Nanjing Drum Tower Hospital, The Affiliated Hospital of Nanjing University Medical School

Nanjing, Jiangsu, 210008, China

Location

Related Publications (2)

• Weinstein SL, Dolan LA, Cheng JC, Danielsson A, Morcuende JA. Adolescent idiopathic scoliosis. Lancet. 2008 May 3;371(9623):1527-37. doi: 10.1016/S0140-6736(08)60658-3.

  PMID: 18456103 BACKGROUND

• Lenke LG, Betz RR, Harms J, Bridwell KH, Clements DH, Lowe TG, Blanke K. Adolescent idiopathic scoliosis: a new classification to determine extent of spinal arthrodesis. J Bone Joint Surg Am. 2001 Aug;83(8):1169-81.

  PMID: 11507125 BACKGROUND

Central Study Contacts

Xiaodong Qin, MD, PhD

CONTACT

025-68182923; gyethics@163.com

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: OUTCOMES ASSESSOR
• Masking Details: Surgeons cannot be masked because the model output is provided only in the model-guided group. Participants will not be actively informed of the assigned decision-support strategy, but participant masking is not considered the primary masking method. Radiographic outcome assessors who determine distal adding-on and other imaging outcomes will be blinded to treatment allocation.
• Purpose: TREATMENT
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Model Details: Participants will be assigned in a 1:1 ratio to either the model-guided group or the conventional-experience group. Randomization will be stratified by Lenke classification. Both groups will undergo standard posterior spinal fusion; the difference between groups is the method used to support lowest instrumented vertebra selection.

Study Record Dates

• First Submitted: July 19, 2026
• First Posted: July 23, 2026
• Study Start: July 1, 2026
• Primary Completion (Estimated): June 1, 2027
• Study Completion (Estimated): September 1, 2030
• Last Updated: July 23, 2026

Record last verified: 2026-05

Data Sharing

• IPD Sharing: Will not share

Locations

CN (1)