Predicting the Occurrence of Postoperative Distal Junctional Kyphosis in Cervical Deformity Patients

Peter G. Passias, Samantha R. Horn, Cheongeun Oh, Renaud Lafage, Virginie Lafage, Justin S. Smith, Breton Line, Themistocles S. Protopsaltis, Mitsuru Yagi, Cole A. Bortz, Frank A. Segreto, Haddy Alas, Bassel G. Diebo, Daniel M. Sciubba, Michael P. Kelly, Alan H. Daniels, Eric O. Klineberg, Douglas C. Burton, Robert A. Hart, Frank J. SchwabShay Bess, Christopher I. Shaffrey, Christopher P. Ames

Research output: Contribution to journalArticle

Abstract

BACKGROUND: Distal junctional kyphosis (DJK) development after cervical deformity (CD)-corrective surgery is a growing concern for surgeons and patients. Few studies have investigated risk factors that predict the occurrence of DJK. OBJECTIVE: To predict DJK development after CD surgery using predictive modeling. METHODS: CD criteria was at least one of the following: C2-C7 Coronal/Cobb > 10°, C2-7 sagittal vertical axis (cSVA) > 4 cm, chin-brow vertical angle > 25°. DJK was defined as the development of an angle <-10° from the end of fusion construct to the second distal vertebra, and change in this angle by <-10° from baseline to postoperative. Baseline demographic, clinical, and surgical information were used to predict the occurrence of DJK using generalized linear modeling both as one overall model and as submodels using baseline demographic and clinical predictors or surgical predictors. RESULTS: One hundred seventeen CD patients were included. At any postoperative visit up to 1 yr, 23.1% of CD patients developed DJK. DJK was predicted with high accuracy using a combination of baseline demographic, clinical, and surgical factors by the following factors: preoperative neurological deficit, use of transition rod, C2-C7 lordosis (CL)<-12°, T1 slope minus CL > 31°, and cSVA > 54 mm. In the model using only baseline demographic/clinical predictors of DJK, presence of comorbidities, presence of baseline neurological deficit, and high preoperative C2-T3 angle were included in the final model (area under the curve = 87%). The final model using only surgical predictors for DJK included combined approach, posterior upper instrumented vertebrae below C4, use of transition rod, lack of anterior corpectomy, more than 3 posterior osteotomies, and performance of a 3-column osteotomy. CONCLUSION: Preoperative assessment and consideration should be given to these factors that are predictive of DJK to mitigate poor outcomes.

Original languageEnglish
Pages (from-to)E38-E46
JournalNeurosurgery
Volume86
Issue number1
DOIs
Publication statusPublished - 2020 Jan 1

Fingerprint

Kyphosis
Osteotomy
Chin
Area Under Curve
Comorbidity
Spine
Demography

Keywords

  • Cervical deformity
  • Complications
  • Distal junctional kyphosis
  • Predictive analytics

ASJC Scopus subject areas

  • Surgery
  • Clinical Neurology

Cite this

Passias, P. G., Horn, S. R., Oh, C., Lafage, R., Lafage, V., Smith, J. S., ... Ames, C. P. (2020). Predicting the Occurrence of Postoperative Distal Junctional Kyphosis in Cervical Deformity Patients. Neurosurgery, 86(1), E38-E46. https://doi.org/10.1093/neuros/nyz347

Predicting the Occurrence of Postoperative Distal Junctional Kyphosis in Cervical Deformity Patients. / Passias, Peter G.; Horn, Samantha R.; Oh, Cheongeun; Lafage, Renaud; Lafage, Virginie; Smith, Justin S.; Line, Breton; Protopsaltis, Themistocles S.; Yagi, Mitsuru; Bortz, Cole A.; Segreto, Frank A.; Alas, Haddy; Diebo, Bassel G.; Sciubba, Daniel M.; Kelly, Michael P.; Daniels, Alan H.; Klineberg, Eric O.; Burton, Douglas C.; Hart, Robert A.; Schwab, Frank J.; Bess, Shay; Shaffrey, Christopher I.; Ames, Christopher P.

In: Neurosurgery, Vol. 86, No. 1, 01.01.2020, p. E38-E46.

Research output: Contribution to journalArticle

Passias, PG, Horn, SR, Oh, C, Lafage, R, Lafage, V, Smith, JS, Line, B, Protopsaltis, TS, Yagi, M, Bortz, CA, Segreto, FA, Alas, H, Diebo, BG, Sciubba, DM, Kelly, MP, Daniels, AH, Klineberg, EO, Burton, DC, Hart, RA, Schwab, FJ, Bess, S, Shaffrey, CI & Ames, CP 2020, 'Predicting the Occurrence of Postoperative Distal Junctional Kyphosis in Cervical Deformity Patients', Neurosurgery, vol. 86, no. 1, pp. E38-E46. https://doi.org/10.1093/neuros/nyz347
Passias, Peter G. ; Horn, Samantha R. ; Oh, Cheongeun ; Lafage, Renaud ; Lafage, Virginie ; Smith, Justin S. ; Line, Breton ; Protopsaltis, Themistocles S. ; Yagi, Mitsuru ; Bortz, Cole A. ; Segreto, Frank A. ; Alas, Haddy ; Diebo, Bassel G. ; Sciubba, Daniel M. ; Kelly, Michael P. ; Daniels, Alan H. ; Klineberg, Eric O. ; Burton, Douglas C. ; Hart, Robert A. ; Schwab, Frank J. ; Bess, Shay ; Shaffrey, Christopher I. ; Ames, Christopher P. / Predicting the Occurrence of Postoperative Distal Junctional Kyphosis in Cervical Deformity Patients. In: Neurosurgery. 2020 ; Vol. 86, No. 1. pp. E38-E46.
@article{3655c2792c194cef93bfbcba53857cf0,
title = "Predicting the Occurrence of Postoperative Distal Junctional Kyphosis in Cervical Deformity Patients",
abstract = "BACKGROUND: Distal junctional kyphosis (DJK) development after cervical deformity (CD)-corrective surgery is a growing concern for surgeons and patients. Few studies have investigated risk factors that predict the occurrence of DJK. OBJECTIVE: To predict DJK development after CD surgery using predictive modeling. METHODS: CD criteria was at least one of the following: C2-C7 Coronal/Cobb > 10°, C2-7 sagittal vertical axis (cSVA) > 4 cm, chin-brow vertical angle > 25°. DJK was defined as the development of an angle <-10° from the end of fusion construct to the second distal vertebra, and change in this angle by <-10° from baseline to postoperative. Baseline demographic, clinical, and surgical information were used to predict the occurrence of DJK using generalized linear modeling both as one overall model and as submodels using baseline demographic and clinical predictors or surgical predictors. RESULTS: One hundred seventeen CD patients were included. At any postoperative visit up to 1 yr, 23.1{\%} of CD patients developed DJK. DJK was predicted with high accuracy using a combination of baseline demographic, clinical, and surgical factors by the following factors: preoperative neurological deficit, use of transition rod, C2-C7 lordosis (CL)<-12°, T1 slope minus CL > 31°, and cSVA > 54 mm. In the model using only baseline demographic/clinical predictors of DJK, presence of comorbidities, presence of baseline neurological deficit, and high preoperative C2-T3 angle were included in the final model (area under the curve = 87{\%}). The final model using only surgical predictors for DJK included combined approach, posterior upper instrumented vertebrae below C4, use of transition rod, lack of anterior corpectomy, more than 3 posterior osteotomies, and performance of a 3-column osteotomy. CONCLUSION: Preoperative assessment and consideration should be given to these factors that are predictive of DJK to mitigate poor outcomes.",
keywords = "Cervical deformity, Complications, Distal junctional kyphosis, Predictive analytics",
author = "Passias, {Peter G.} and Horn, {Samantha R.} and Cheongeun Oh and Renaud Lafage and Virginie Lafage and Smith, {Justin S.} and Breton Line and Protopsaltis, {Themistocles S.} and Mitsuru Yagi and Bortz, {Cole A.} and Segreto, {Frank A.} and Haddy Alas and Diebo, {Bassel G.} and Sciubba, {Daniel M.} and Kelly, {Michael P.} and Daniels, {Alan H.} and Klineberg, {Eric O.} and Burton, {Douglas C.} and Hart, {Robert A.} and Schwab, {Frank J.} and Shay Bess and Shaffrey, {Christopher I.} and Ames, {Christopher P.}",
year = "2020",
month = "1",
day = "1",
doi = "10.1093/neuros/nyz347",
language = "English",
volume = "86",
pages = "E38--E46",
journal = "Neurosurgery",
issn = "0148-396X",
publisher = "Lippincott Williams and Wilkins",
number = "1",

}

TY - JOUR

T1 - Predicting the Occurrence of Postoperative Distal Junctional Kyphosis in Cervical Deformity Patients

AU - Passias, Peter G.

AU - Horn, Samantha R.

AU - Oh, Cheongeun

AU - Lafage, Renaud

AU - Lafage, Virginie

AU - Smith, Justin S.

AU - Line, Breton

AU - Protopsaltis, Themistocles S.

AU - Yagi, Mitsuru

AU - Bortz, Cole A.

AU - Segreto, Frank A.

AU - Alas, Haddy

AU - Diebo, Bassel G.

AU - Sciubba, Daniel M.

AU - Kelly, Michael P.

AU - Daniels, Alan H.

AU - Klineberg, Eric O.

AU - Burton, Douglas C.

AU - Hart, Robert A.

AU - Schwab, Frank J.

AU - Bess, Shay

AU - Shaffrey, Christopher I.

AU - Ames, Christopher P.

PY - 2020/1/1

Y1 - 2020/1/1

N2 - BACKGROUND: Distal junctional kyphosis (DJK) development after cervical deformity (CD)-corrective surgery is a growing concern for surgeons and patients. Few studies have investigated risk factors that predict the occurrence of DJK. OBJECTIVE: To predict DJK development after CD surgery using predictive modeling. METHODS: CD criteria was at least one of the following: C2-C7 Coronal/Cobb > 10°, C2-7 sagittal vertical axis (cSVA) > 4 cm, chin-brow vertical angle > 25°. DJK was defined as the development of an angle <-10° from the end of fusion construct to the second distal vertebra, and change in this angle by <-10° from baseline to postoperative. Baseline demographic, clinical, and surgical information were used to predict the occurrence of DJK using generalized linear modeling both as one overall model and as submodels using baseline demographic and clinical predictors or surgical predictors. RESULTS: One hundred seventeen CD patients were included. At any postoperative visit up to 1 yr, 23.1% of CD patients developed DJK. DJK was predicted with high accuracy using a combination of baseline demographic, clinical, and surgical factors by the following factors: preoperative neurological deficit, use of transition rod, C2-C7 lordosis (CL)<-12°, T1 slope minus CL > 31°, and cSVA > 54 mm. In the model using only baseline demographic/clinical predictors of DJK, presence of comorbidities, presence of baseline neurological deficit, and high preoperative C2-T3 angle were included in the final model (area under the curve = 87%). The final model using only surgical predictors for DJK included combined approach, posterior upper instrumented vertebrae below C4, use of transition rod, lack of anterior corpectomy, more than 3 posterior osteotomies, and performance of a 3-column osteotomy. CONCLUSION: Preoperative assessment and consideration should be given to these factors that are predictive of DJK to mitigate poor outcomes.

AB - BACKGROUND: Distal junctional kyphosis (DJK) development after cervical deformity (CD)-corrective surgery is a growing concern for surgeons and patients. Few studies have investigated risk factors that predict the occurrence of DJK. OBJECTIVE: To predict DJK development after CD surgery using predictive modeling. METHODS: CD criteria was at least one of the following: C2-C7 Coronal/Cobb > 10°, C2-7 sagittal vertical axis (cSVA) > 4 cm, chin-brow vertical angle > 25°. DJK was defined as the development of an angle <-10° from the end of fusion construct to the second distal vertebra, and change in this angle by <-10° from baseline to postoperative. Baseline demographic, clinical, and surgical information were used to predict the occurrence of DJK using generalized linear modeling both as one overall model and as submodels using baseline demographic and clinical predictors or surgical predictors. RESULTS: One hundred seventeen CD patients were included. At any postoperative visit up to 1 yr, 23.1% of CD patients developed DJK. DJK was predicted with high accuracy using a combination of baseline demographic, clinical, and surgical factors by the following factors: preoperative neurological deficit, use of transition rod, C2-C7 lordosis (CL)<-12°, T1 slope minus CL > 31°, and cSVA > 54 mm. In the model using only baseline demographic/clinical predictors of DJK, presence of comorbidities, presence of baseline neurological deficit, and high preoperative C2-T3 angle were included in the final model (area under the curve = 87%). The final model using only surgical predictors for DJK included combined approach, posterior upper instrumented vertebrae below C4, use of transition rod, lack of anterior corpectomy, more than 3 posterior osteotomies, and performance of a 3-column osteotomy. CONCLUSION: Preoperative assessment and consideration should be given to these factors that are predictive of DJK to mitigate poor outcomes.

KW - Cervical deformity

KW - Complications

KW - Distal junctional kyphosis

KW - Predictive analytics

UR - http://www.scopus.com/inward/record.url?scp=85076558276&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85076558276&partnerID=8YFLogxK

U2 - 10.1093/neuros/nyz347

DO - 10.1093/neuros/nyz347

M3 - Article

C2 - 31838540

AN - SCOPUS:85076558276

VL - 86

SP - E38-E46

JO - Neurosurgery

JF - Neurosurgery

SN - 0148-396X

IS - 1

ER -