JoVE - Journal of Visualized Experiments
JoVE Visualize
Contact Us

Microbiome-Augmented Model for Predicting Knee Osteoarthritis Progression Based on Gut Microbiota and Kellgren-Lawrence Classification

  • 0Department of Orthopedics, The Affiliated Taizhou People's Hospital of Nanjing Medical University, Taizhou, CHN.
Cureus +

|

December 12, 2024

|

December 12, 2024

View abstract on PubMed

Summary

This summary is machine-generated.

A new microbiome-augmented model accurately predicts knee osteoarthritis (OA) progression using gut microbiota data. This tool aids in identifying patients at risk for rapid disease progression, enabling early intervention.

Area Of Science

  • Orthopedics and Rheumatology
  • Microbiome Research
  • Medical Data Science

Background

  • Knee osteoarthritis (OA) is a prevalent degenerative condition with variable progression rates.
  • The gut-joint axis, a bidirectional relationship, significantly influences OA onset and advancement.
  • Predicting OA disease progression is crucial for timely and effective patient management.

Purpose Of The Study

  • To develop and validate a predictive model for knee OA disease progression.
  • To assess the utility of gut microbiota data in predicting OA progression.
  • To establish a model that identifies patients at risk of rapid knee OA deterioration.

Main Methods

  • A prospective cohort study included 270 knee OA patients diagnosed via X-ray and CT scans.
  • Fecal samples were analyzed for gut microbiota composition, alongside general patient information.
  • Least Absolute Shrinkage and Selection Operator (LASSO) regression and microbiome-augmented models, including Light Gradient Boosting Machine (LGBM), were used for prediction and validated using accuracy, sensitivity, specificity, PPV, NPV, and AUC.

Main Results

  • The microbiome-augmented model using LGBM demonstrated superior performance in predicting knee OA progression based on the Kellgren-Lawrence classification.
  • In the test set, the LGBM model achieved an AUC of 0.876, accuracy of 0.830, sensitivity of 0.759, and specificity of 0.917.
  • Lower Shannon index values were observed in patients with Grade I Kellgren-Lawrence classification after one year compared to those with Grade II/III.

Conclusions

  • The LGBM-based microbiome-augmented model effectively predicts knee OA progression one year in advance.
  • This model can identify patients susceptible to rapid OA progression, facilitating personalized treatment strategies.
  • The findings highlight the significant role of the gut-joint axis in knee OA pathogenesis and progression.

Keywords: