MRI-based endplate bone quality scoring system in cervical OPLL: a novel measuring method predicting titanium mesh cage subsidence following ACCF.

PURPOSE

What are the optimal image magnification ratios for EBQ score measurement and the accuracy of this score in predicting titanium mesh cage (TMC) subsidence in patients with cervical ossification of the posterior longitudinal ligament (OPLL) after anterior cervical corpectomy and fusion (ACCF)?

METHODS

107 cervical OPLL patients who had undergone single-level ACCF were included. Patients were stratified into subsidence and non-subsidence groups, paired by propensity score matching, and the predictive values of EBQ and VBQ scores for TMC subsidence were evaluated via receiver operating curve (ROC) and the area under the curve (AUC).

RESULTS

On radiographs, 75% and 100% magnifications showed better reliability than 25% and 50%, thus 75% was used. Higher EBQ, Global VBQ, U-L VBQ scores and older age are linked to higher TMC subsidence risk. In the cohort of matched patients, An EBQ threshold of 2.44 demonstrated predictive efficacy for TMC subsidence (AUC = 0.752, 78.8% sensitivity, 69.7% specificity). A U-L VBQ score threshold of 2.57 and Global VBQ score threshold of 2.56 both demonstrated good prediction of TMC subsidence. The former achieved 84.8% sensitivity and 57.6% specificity, while the latter showed 86.3% sensitivity with 52.4% specificity. The two matched groups had similar improvements in patient-reported outcomes including VAS score, NDI and mJOA score.

CONCLUSION

Higher EBQ score was an independent factor for TMC subsidence. Use of EBQ score of 2.44 as the cutoff value on preoperative MRIs magnified 75% yielded better reliability. Regarding the two scores, the EBQ score emerges as a more effective predictor.