1
|
Shi W, Giuste FO, Zhu Y, Tamo BJ, Nnamdi MC, Hornback A, Carpenter AM, Hilton C, Iwinski HJ, Wattenbarger JM, Wang MD. Predicting pediatric patient rehabilitation outcomes after spinal deformity surgery with artificial intelligence. COMMUNICATIONS MEDICINE 2025; 5:1. [PMID: 39747461 PMCID: PMC11697361 DOI: 10.1038/s43856-024-00726-1] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/25/2024] [Accepted: 12/20/2024] [Indexed: 01/04/2025] Open
Abstract
BACKGROUND Adolescent idiopathic scoliosis (AIS) is the most common type of scoliosis, affecting 1-4% of adolescents. The Scoliosis Research Society-22R (SRS-22R), a health-related quality-of-life instrument for AIS, has allowed orthopedists to measure subjective patient outcomes before and after corrective surgery beyond objective radiographic measurements. However, research has revealed that there is no significant correlation between the correction rate in major radiographic parameters and improvements in patient-reported outcomes (PROs), making it difficult to incorporate PROs into personalized surgical planning. METHODS The objective of this study is to develop an artificial intelligence (AI)-enabled surgical planning and counseling support system for post-operative patient rehabilitation outcomes prediction in order to facilitate personalized AIS patient care. A unique multi-site cohort of 455 pediatric patients undergoing spinal fusion surgery at two Shriners Children's hospitals from 2010 is investigated in our analysis. In total, 171 pre-operative clinical features are used to train six machine-learning models for post-operative outcomes prediction. We further employ explainability analysis to quantify the contribution of pre-operative radiographic and questionnaire parameters in predicting patient surgical outcomes. Moreover, we enable responsible AI by calibrating model confidence for human intervention and mitigating health disparities for algorithm fairness. RESULTS The best prediction model achieves an area under receiver operating curve (AUROC) performance of 0.86, 0.85, and 0.83 for individual SRS-22R question response prediction over three-time horizons from pre-operation to 6-month, 1-year, and 2-year post-operation, respectively. Additionally, we demonstrate the efficacy of our proposed prediction method to predict other patient rehabilitation outcomes based on minimal clinically important differences (MCID) and correction rates across all three-time horizons. CONCLUSIONS Based on the relationship analysis, we suggest additional attention to sagittal parameters (e.g., lordosis, sagittal vertical axis) and patient self-image beyond major Cobb angles to improve surgical decision-making for AIS patients. In the age of personalized medicine, the proposed responsible AI-enabled clinical decision-support system may facilitate pre-operative counseling and shared decision-making within real-world clinical settings.
Collapse
Affiliation(s)
- Wenqi Shi
- School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, 30322, USA.
- University of Texas Southwestern Medical Center, Dallas, TX, 75390, USA.
| | - Felipe O Giuste
- The Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, Atlanta, GA, 30322, USA
| | - Yuanda Zhu
- School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, 30322, USA
| | - Ben J Tamo
- School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, 30322, USA
| | - Micky C Nnamdi
- School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, 30322, USA
| | - Andrew Hornback
- The Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, Atlanta, GA, 30322, USA
| | | | | | | | | | - May D Wang
- School of Electrical and Computer Engineering, Georgia Institute of Technology, Atlanta, GA, 30322, USA.
- The Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, Atlanta, GA, 30322, USA.
| |
Collapse
|
2
|
Goodwin ML, Buchowski JM, Sciubba DM. Why X-rays? The importance of radiographs in spine surgery. Spine J 2022; 22:1759-1767. [PMID: 35908587 DOI: 10.1016/j.spinee.2022.07.102] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 05/26/2022] [Revised: 07/22/2022] [Accepted: 07/23/2022] [Indexed: 02/03/2023]
Abstract
Despite the advances made in high-resolution spinal imaging, plain films (radiographs or x-rays) remain a cornerstone of evaluating and caring for spine patients in the preoperative, intraoperative, and postoperative settings. Although often undervalued when compared with more advanced imaging such as MRIs or CT scans, plain films provide surgeons invaluable information that other imaging modalities oftentimes cannot. In addition to their use during surgery for localization or evaluation of hardware placement, x-rays provide an overall image of a patient's spine, are useful in evaluating hardware complications, allow detailed assessment of alignment and stability and allow for repeated images in clinic during follow-up. Plain films continue to provide critical information that cannot be obtained with other imaging modalities, and they remain central to providing optimal care for spine patients.
Collapse
Affiliation(s)
- Matthew L Goodwin
- Department of Orthopedic Surgery, Washington University School of Medicine, 660 S Euclid Ave, St. Louis, MO 63110 USA.
| | - Jacob M Buchowski
- Department of Orthopedic Surgery, Washington University School of Medicine, 660 S Euclid Ave, St. Louis, MO 63110 USA
| | - Daniel M Sciubba
- Department of Neurosurgery, Northwell Health Chair of Neurosurgery at North Shore University Hospital and Long Island Jewish Medical Center, NY, USA; Institute for Neurology and Neurosurgery, Northwell Health and Chair of Neurosciences, Donald and Barbara Zucker School of Medicine at Hofstra Northwell, NY, USA
| |
Collapse
|
3
|
No added value of 2-year radiographic follow-up of fusion surgery for adolescent idiopathic scoliosis. EUROPEAN SPINE JOURNAL : OFFICIAL PUBLICATION OF THE EUROPEAN SPINE SOCIETY, THE EUROPEAN SPINAL DEFORMITY SOCIETY, AND THE EUROPEAN SECTION OF THE CERVICAL SPINE RESEARCH SOCIETY 2021; 30:759-767. [PMID: 33392756 DOI: 10.1007/s00586-020-06696-x] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Subscribe] [Scholar Register] [Received: 07/04/2020] [Revised: 10/25/2020] [Accepted: 12/09/2020] [Indexed: 10/22/2022]
Abstract
PURPOSE For fusion surgery in adolescent idiopathic scoliosis (AIS) consensus exists that a 2-year radiographic follow-up assessment is needed. This standard lacks empirical evidence. The purpose of this study was to investigate the radiographic follow-up after corrective surgery in AIS, from pre-until 2 years postoperative. METHODS In this historical cohort study, 63 patients surgically treated for AIS, age ≤ 25 years, with 2-year radiographic follow-up, were enrolled. The primary outcome measure was the major Cobb angle. Secondary outcomes were coronal and sagittal spino-pelvic parameters, including proximal junction kyphosis (PJK) and distal adding-on. Change over time was analyzed using a repeated measures ANOVA. RESULTS The major curve Cobb angle showed a statistically significant change for pre- to 1 year postoperative, but not for 1- to 2-year follow-up. Seven out of 63 patients did show a change exceeding the error of measurement (5°) from 1- to 2-year follow-up (range -8° to +7°), of whom 2 patients showed curve progression and 5 showed improvement. PJK or distal adding-on was not observed. CONCLUSIONS No statistically significant changes in major curve Cobb angle were found during postsurgical follow-up, or in adjacent non-fused segments. The findings of this study are not supportive for routine radiographs 2 years after fusion surgery in AIS patients.
Collapse
|
4
|
Lechtig A, Barghi A, Wolf BT, Weaver M, Wixted JJ, Rodriguez EK, Appleton PT. The utility of post-operative hip radiographs in patients treated with hip hemiarthroplasty for femoral neck fractures. Injury 2019; 50:1448-1451. [PMID: 31320108 DOI: 10.1016/j.injury.2019.07.005] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/03/2019] [Accepted: 07/07/2019] [Indexed: 02/02/2023]
Abstract
BACKGROUND Patients treated with hip hemiarthroplasty for low energy femoral neck fractures routinely undergo hip radiographs at each postoperative clinic visit regardless of history and physical findings. No studies to date have evaluated the effectiveness of this accepted practice. The goal of this study was to identify the postoperative utility of both history and physical (H/P) and hip radiographs in the treatment course of patients treated with hip hemiarthroplasty for low energy femoral neck fractures. METHODS A retrospective chart review was performed on consecutive patients treated with hip hemiarthroplasty for low energy femoral neck fractures. An abnormal H/P and hip radiographs as well as a change in treatment course were recorded at each clinic or emergency department visit. RESULTS Five hundred and eighty-three patients met inclusion criteria, consisting of 1177 clinic and 50 emergency department (ED) visits. An abnormal radiograph in the presence of a normal H/P did not lead to a change in treatment course. An abnormal H/P alone changed treatment course in 28 (3%) clinic visits and 18 (36%) ED visits. An abnormal H/P and the presence of an abnormal hip radiograph changed the treatment course in 23 (2%) clinic visits and 18 (36%) ED visits. In only one case - 0.3% of abnormal radiographs or 0.08% of 1177 clinic visits - did an abnormal hip radiograph change treatment course in the setting of an abnormal H/P within 6 months from surgery. The average cost of a series of hip and pelvis radiographs was $242. CONCLUSIONS Abnormal radiographs do not change treatment course in the presence of a normal H/P. Hip radiographs obtained in clinic within 6 months of surgery rarely lead to a change in treatment course and thereby are a source of excess cost and radiation exposure to the patient.
Collapse
Affiliation(s)
- Aron Lechtig
- Beth Israel Deaconess Medical Center, 330 Brookline Ave, Boston, MA, 02215, United States; Harvard Medical School Orthopedic Trauma Initiative, Boston, MA, 02215, United States
| | - Ameen Barghi
- Harvard Medical School, Boston, MA, 02115, United States; Harvard Medical School Orthopedic Trauma Initiative, Boston, MA, 02215, United States.
| | - Bryce T Wolf
- Maine Ortho, 1601 Congress Street, Portland, ME, 04102, United States; Harvard Medical School Orthopedic Trauma Initiative, Boston, MA, 02215, United States
| | - Michael Weaver
- Brigham and Woman's Hospital, Boston, MA, 02215, United States; Harvard Medical School Orthopedic Trauma Initiative, Boston, MA, 02215, United States
| | - John J Wixted
- Beth Israel Deaconess Medical Center, 330 Brookline Ave, Boston, MA, 02215, United States; Harvard Medical School Orthopedic Trauma Initiative, Boston, MA, 02215, United States
| | - Edward K Rodriguez
- Beth Israel Deaconess Medical Center, 330 Brookline Ave, Boston, MA, 02215, United States; Harvard Medical School Orthopedic Trauma Initiative, Boston, MA, 02215, United States
| | - Paul T Appleton
- Beth Israel Deaconess Medical Center, 330 Brookline Ave, Boston, MA, 02215, United States; Harvard Medical School Orthopedic Trauma Initiative, Boston, MA, 02215, United States
| |
Collapse
|
5
|
What's New in Pediatric Orthopaedic Quality, Safety, and Value? A Systematic Review With Results of the 2016 POSNA Quality, Safety, and Value Initiative (QSVI) Challenge. J Pediatr Orthop 2018; 38:e646-e651. [PMID: 30074591 DOI: 10.1097/bpo.0000000000001241] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.4] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 02/07/2023]
Abstract
BACKGROUND Enhancing the safety, quality, and value of care provided is a point of emphasis for modern health care systems. We performed a review of recent literature to highlight those efforts relevant to pediatric musculoskeletal care. METHODS We searched the PubMed database for all papers related to quality improvement, patient safety, and/or value in pediatric orthopaedics published from October 1, 2012 to October 31, 2017, yielding 193 papers. RESULTS A total of 36 papers were selected for review based upon new findings. Papers were selected based on significant contributions in the following categories: casting safety, antibiotic stewardship/infection prevention, perioperative care pathways, blood conservation, venous thromboembolic disease prevention, and imaging safety/appropriateness. CONCLUSIONS There have been numerous advances in safety, quality, and value in pediatric orthopaedic care. Quality improvement efforts emphasizing provider education and safety monitoring can lead to a decrease in cast-related complications. Perioperative care pathways and bundles are associated with a decrease risk of surgical site infection and decreased length of stay in pediatric spinal deformity surgery. Increased scrutiny has been placed on the value of routine follow-up radiographs in pediatric fracture and spinal deformity care. LEVEL OF EVIDENCE Level 4-literature review.
Collapse
|
6
|
Pannell WC, Alluri RK, Sivasundaram L, Heckmann N, Ghiassi A. Utility of Postoperative Imaging in Radial Shaft Fractures. Hand (N Y) 2016; 11:184-7. [PMID: 27390560 PMCID: PMC4920538 DOI: 10.1177/1558944715627629] [Citation(s) in RCA: 8] [Impact Index Per Article: 0.9] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/17/2022]
Abstract
BACKGROUND Postoperatively, radial shaft fractures are often followed clinically with serial radiographs to assess for fracture healing. Currently, there is no standard of care regarding postoperative imaging for these injuries. The purpose of this study is to determine whether imaging influences management decisions. METHODS Patients who presented to a level I trauma center between 2009 and 2014 with an operatively treated radial shaft fracture were retrospectively screened for inclusion in our study. Patients with ipsilateral ulna or radius fractures, or with inadequate imaging or inadequate follow-up, were excluded. Four blinded, board-certified, orthopedic surgeons reviewed the postoperative films twice for each patient and stated whether the imaging would influence management decisions. Images were separated into 3 groups based on time from surgery: 0 to 4 weeks, 4 to 8 weeks, and greater than 8 weeks. The number of times imaging influenced these hypothetical management decisions was recorded. Interobserver and intraobserver agreements were calculated using Fleiss's and Cohen's kappa coefficients, respectively. RESULTS One hundred eighteen patients underwent operative fixation for an isolated radial shaft fracture, of whom 38 met inclusion criteria. Imaging from 0 to 4 weeks, 4 to 8 weeks, and greater than 8 weeks postoperatively resulted in a change of management in 0%, 32%, and 16% of patients, respectively. After 4 weeks, changes were primarily for immobilization and activity-level modification. Intraobserver and interobserver agreement kappa coefficients were 0.761 and 0.563, respectively. CONCLUSIONS Films obtained within 4 weeks of surgery for radial shaft fractures are unlikely to change postoperative management and may not be warranted during routine postoperative follow-up.
Collapse
Affiliation(s)
- William C. Pannell
- Keck School of Medicine of USC, Los Angeles, CA, USA,William C. Pannell, Department of Orthopaedic Surgery, Keck School of Medicine of USC, 1200 N. State Street, GNH 3900, Los Angeles, CA, USA.
| | - Ram K. Alluri
- Keck School of Medicine of USC, Los Angeles, CA, USA
| | | | | | | |
Collapse
|
7
|
Werner BC, Burrus MT, Kew ME, Dempsey IJ, Gwathmey FW, Miller MD, Diduch DR. Limited utility of routine early postoperative radiography after primary ACL reconstruction. Knee 2016; 23:237-40. [PMID: 26791681 DOI: 10.1016/j.knee.2015.09.006] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.6] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 03/16/2015] [Revised: 07/02/2015] [Accepted: 09/02/2015] [Indexed: 02/02/2023]
Abstract
BACKGROUND Given the overall success of anterior cruciate ligament (ACL) reconstruction and the infrequent occurrence of complications detectable on radiographs, the clinical utility and cost-effectiveness of routine radiographs in the early postoperative setting is questionable. METHODS Nine hundred thirty-three consecutive adult patients undergoing uncomplicated ACL reconstruction at a single institution were retrospectively reviewed to determine whether a postoperative knee radiograph was obtained within the first three months postoperatively. Images, reports and clinical notes were reviewed to determine if any clinical management change occurred due to x-ray findings. Radiograph charges, including imaging, technical and professional charges were calculated. RESULTS Five hundred ninety-nine of 933 primary ACL reconstruction patients (64.8%) had postoperative knee radiography at an average of 6.3±3.5 weeks postoperatively. A musculoskeletal radiologist read 97.7% of x-rays as normal. In the associated visit note, 70.3% of x-ray results were documented. Only 14.1% of patients with a postoperative x-ray had subsequent imaging. There were no significant management changes based on the routine postoperative radiographs using the defined criteria. A total of $336,683 ($562 per patient) was billed to patients for postoperative radiographs. CONCLUSIONS Routine early postoperative radiography after primary ACL reconstruction is of questionable utility. The significant per-patient expense is not balanced by the low yield of clinically meaningful data, as nearly all radiographs in the present series were normal and none resulted in significant changes in postoperative clinical management. These results suggest that routine radiographs after uncomplicated ACL reconstruction may be unnecessary although larger, multicenter studies are necessary to confirm these findings. LEVEL OF EVIDENCE Level IV, retrospective case series.
Collapse
Affiliation(s)
- Brian C Werner
- Department of Orthopaedic Surgery, University of Virginia Health System, PO Box 800159, Charlottesville, VA 22908, USA.
| | - M Tyrrell Burrus
- Department of Orthopaedic Surgery, University of Virginia Health System, PO Box 800159, Charlottesville, VA 22908, USA
| | - Michelle E Kew
- Department of Orthopaedic Surgery, University of Virginia Health System, PO Box 800159, Charlottesville, VA 22908, USA
| | - Ian J Dempsey
- Department of Orthopaedic Surgery, University of Virginia Health System, PO Box 800159, Charlottesville, VA 22908, USA
| | - F Winston Gwathmey
- Department of Orthopaedic Surgery, University of Virginia Health System, PO Box 800159, Charlottesville, VA 22908, USA
| | - Mark D Miller
- Department of Orthopaedic Surgery, University of Virginia Health System, PO Box 800159, Charlottesville, VA 22908, USA
| | - David R Diduch
- Department of Orthopaedic Surgery, University of Virginia Health System, PO Box 800159, Charlottesville, VA 22908, USA
| |
Collapse
|