Nomogram for predicting inadequate treatment response in vertebral osteomyelitis

Area of Science:

• Orthopedics
• Infectious Diseases
• Medical Informatics

Background:

• Vertebral osteomyelitis (VO) often leads to inadequate treatment response (ITR), causing prolonged hospitalization and recurrence.
• Currently, no validated tool exists for early prediction of ITR in VO patients.
• ITR is defined as progression despite appropriate antimicrobial therapy for at least four weeks.

Purpose of the Study:

• To develop and internally validate a nomogram for predicting ITR probability in VO patients.
• The nomogram aims to provide individualized risk assessment for VO patients.
• Facilitating early therapeutic intensification is a key objective.

Main Methods:

• A retrospective study included 219 VO patients.
• Predictors for ITR were identified using LASSO and multivariate logistic regression.
• Nomogram performance was evaluated using AUC, calibration curves, and decision curve analysis (DCA).

Main Results:

• The developed nomogram demonstrated excellent discrimination with an AUC of 0.909.
• Six independent predictors of ITR were identified: intraspinal abscess, paraspinal abscess, spinal implant, pneumonia, low prealbumin, and multiple-site infections.
• The nomogram showed high sensitivity (76.5%) and specificity (91%) at the optimal cutoff.

Conclusions:

• The study presents the first validated nomogram for predicting ITR in VO.
• This tool enables individualized risk assessment, supporting early therapeutic intensification.
• The nomogram has the potential to improve clinical outcomes for VO patients.