Nomogram for predicting malignancy in thyroid nodules using clinical, biochemical, ultrasonographic, and cytologic...

Area of Science:

• Endocrinology
• Oncology
• Radiology

Background:

• Thyroid nodules are frequently found incidentally, posing management challenges, especially with atypical or suspicious cytology.
• Accurate prediction of malignancy risk is crucial for effective patient management.
• Current prediction relies on a combination of clinical, biochemical, ultrasonographic, and cytologic features.

Purpose of the Study:

• To develop and validate a nomogram for precise prediction of malignancy risk in thyroid nodules.
• To integrate key diagnostic features into a single predictive tool.

Main Methods:

• Retrospective review of 158 patients with 190 thyroid nodules undergoing ultrasonography-guided fine needle aspiration and resection (2007-2008).
• Analysis of clinical records, biochemical profiles, pathology reports, ultrasonography images, and cytology slides.
• Application of multivariate logistic regression to identify and quantify predictive variables.

Main Results:

• Eight key variables were identified: thyroid-stimulating hormone (biochemical); nodule shape, echo texture, and vascularity (ultrasonography); and nuclear grooves, pseudoinclusions, cellularity, and colloid presence (cytology).
• The developed nomogram demonstrated excellent predictive accuracy, achieving a concordance index of 91%.
• The nomogram effectively integrates diverse data points for risk stratification.

Conclusions:

• A novel nomogram accurately quantifies the risk of malignancy in thyroid nodules.
• This tool aids clinicians in managing thyroid nodules by providing a precise malignancy risk assessment.
• The nomogram leverages biochemical, ultrasonographic, and cytologic features for improved diagnostic confidence.