Cross-sectional validation of a preoperative multidimensional assessment tool for third molar extraction using...

Area of Science:

• Oral Surgery
• Dental Diagnostics

Background:

• Third molar extractions present variable surgical complexity.
• Accurate preoperative difficulty assessment is crucial for surgical planning and patient outcomes.

Purpose of the Study:

• To evaluate a preoperative difficulty assessment form for third molar extractions.
• To determine the form's correlation with surgery time, perceived difficulty, surgical technique, and postoperative complications.
• To explore the utility of machine learning models in predicting surgical outcomes.

Main Methods:

• A cross-sectional study involving 205 patients undergoing third molar extraction.
• Utilized a validated preoperative difficulty evaluation form assessing clinical, radiological, and surgical variables.
• Employed descriptive, bivariate analyses, and machine learning models for outcome prediction.

Main Results:

• The form's global score significantly correlated with surgery time (rho=0.640, P<0.001) and perceived difficulty (rho=0.395, P<0.001).
• Higher scores were associated with specific surgical techniques (P<0.001) and increased postoperative complications (P=0.012).
• Machine learning models demonstrated superior performance in predicting surgery time compared to perceived difficulty.

Conclusions:

• The preoperative assessment form is a reliable tool for estimating third molar extraction difficulty.
• The form's global score effectively predicts surgery duration, complexity, and potential complications.
• Machine learning offers a promising avenue for enhancing predictive accuracy in oral surgery planning.