What Factors Predict the Risk of Proximal Junctional Failure in the Long Term, Demographic, Surgical, or Radiographic?: Results From a Time-dependent ROC Curve.

Abstract

Study design

Retrospective review of prospective multicenter database.

Objective

To identify an optimal set of factors predicting the risk of proximal junctional failure (PJF) while taking the time dependency of PJF and those factors into account.

Summary of background data

Surgical correction of adult spinal deformity (ASD) can be complex and therefore, may come with high revision rates due to PJF.

Methods

Seven hundred sixty-three operative ASD patients with a minimum of 1-year follow-up were included. PJF was defined as any type of proximal junctional kyphosis (PJK) requiring revision surgery. Time-dependent ROC curves were estimated with corresponding Cox proportional hazard models. The predictive abilities of demographic, surgical, radiographic parameters, and their possible combinations were assessed sequentially. The area under the curve (AUC) was used to evaluate models' performance.

Results

PJF occurred in 42 patients (6%), with a median time to revision of approximately 1 year. Larger preoperative pelvic tilt (PT) (hazard ratio [HR]=1.044, P = 0.034) significantly increased the risk of PJF. With respect to changes in the radiographic parameters at 6-week postsurgery, larger differences in pelvic incidence-lumbar lordosis (PI-LL) mismatch (HR = 0.924, P = 0.002) decreased risk of PJF. The combination of demographic, surgical, and radiographic parameters has the best predictive ability for the occurrence of PJF (AUC = 0.863), followed by demographic along with radiographic parameters (AUC = 0.859). Both models' predictive ability was preserved over time.

Conclusions

Over correction increased the risk of PJF. Radiographic along with demographic parameters have shown the approximately equivalent predictive ability for PJF over time as with the addition of surgical parameters. Radiographic rather than surgical factors may be of particular importance in predicting the development of PJF over time. These results set the groundwork for risk stratification and corresponding prophylactic interventions for patients undergoing ASD surgery.

Level of evidence

4.

Department

Description

Provenance

Citation

Published Version (Please cite this version)

10.1097/brs.0000000000002955

Publication Info

Yang, Jingyan, Marc Khalifé, Renaud Lafage, Han Jo Kim, Justin Smith, Christopher I Shaffrey, Douglas C Burton, Christopher P Ames, et al. (2019). What Factors Predict the Risk of Proximal Junctional Failure in the Long Term, Demographic, Surgical, or Radiographic?: Results From a Time-dependent ROC Curve. Spine, 44(11). pp. 777–784. 10.1097/brs.0000000000002955 Retrieved from https://hdl.handle.net/10161/28201.

This is constructed from limited available data and may be imprecise. To cite this article, please review & use the official citation provided by the journal.

Scholars@Duke

Shaffrey

Christopher Ignatius Shaffrey

Professor of Orthopaedic Surgery

I have more than 25 years of experience treating patients of all ages with spinal disorders. I have had an interest in the management of spinal disorders since starting my medical education. I performed residencies in both orthopaedic surgery and neurosurgery to gain a comprehensive understanding of the entire range of spinal disorders. My goal has been to find innovative ways to manage the range of spinal conditions, straightforward to complex. I have a focus on managing patients with complex spinal disorders. My patient evaluation and management philosophy is to provide engaged, compassionate care that focuses on providing the simplest and least aggressive treatment option for a particular condition. In many cases, non-operative treatment options exist to improve a patient’s symptoms. I have been actively engaged in clinical research to find the best ways to manage spinal disorders in order to achieve better results with fewer complications.


Unless otherwise indicated, scholarly articles published by Duke faculty members are made available here with a CC-BY-NC (Creative Commons Attribution Non-Commercial) license, as enabled by the Duke Open Access Policy. If you wish to use the materials in ways not already permitted under CC-BY-NC, please consult the copyright owner. Other materials are made available here through the author’s grant of a non-exclusive license to make their work openly accessible.