Predicting the Magnitude of Distal Junctional Kyphosis Following Cervical Deformity Correction.

Abstract

Retrospective review of a cervical deformity database. This study aimed to develop a model that can predict the postoperative distal junctional kyphosis angle (DJKA) using preoperative and postoperative radiographic measurements. Distal junctional kyphosis (DJK) is a complication following cervical deformity correction that can reduce of patient quality of life and functional status. Although researchers have identified the risk factors for DJK, no model has been proposed to predict the magnitude of DJK. The DJKA was defined as the Cobb angle from the lower instrumented vertebra (LIV) to LIV-2 with traditional DJK having a DJKA change >10°. Models were trained using 66.6% of the randomly selected patients and validated in the remaining 33.3%. Preoperative and postoperative radiographic parameters associated with DJK were identified and ranked using a conditional variable importance table. Linear regression models were developed using the factors most strongly associated with postoperative DJKA. A total of 131 patients were included with a mean follow-up duration of 14±8 months. The mean postoperative DJKA was 14.6±14° and occurred in 35% of the patients. No significant differences between the training and validation cohort were observed. The variables most associated with postoperative DJK were: preoperative DJKA (DJKApre), postoperative C2-LIV, and change in cervical lordosis (∆CL). The model identified the following equation as predictive of DJKA: DJKA=9.365+(0.123×∆CL)-(0.315×∆C2-LIV)-(0.054×DJKApre). The predicted and actual postoperative DJKA values were highly correlated ( R =0.871, R2 =0.759, P <0.001). The variables that most increased the DJKA were the preoperative DJKA, postoperative alignment within the construct, and change in cervical lordosis. Future studies can build upon the model developed to be applied in a clinical setting when planning for cervical deformity correction.