Proposed ultrasound algorithm to differentiate thyroglossal duct and dermoid cysts

Area of Science:

• Pediatric Radiology
• Diagnostic Imaging
• Head and Neck Surgery

Background:

• Differentiating thyroglossal duct cysts (TGDC) from dermoid cysts (DC) pre-operatively is challenging.
• The established SIST score has limitations in specificity for differentiating these midline neck masses.

Purpose of the Study:

• To assess the reproducibility of the SIST score.
• To determine inter-observer agreement in ultrasound (US) differentiation of TGDC and DC.
• To develop an improved pre-operative US-based algorithm for differentiating TGDC from DC.

Main Methods:

• Retrospective chart review of pediatric patients with midline neck masses.
• Two blinded pediatric radiologists evaluated US images for specific cyst characteristics.
• Analysis of predictive power of US features and development of a novel 4S algorithm (Septations, depth to Straps, Shape, Solid parts).

Main Results:

• The SIST score achieved 67% diagnostic accuracy; individual radiologists achieved 96% and 86%.
• The 4S algorithm, utilizing sequential filtering of key US characteristics, correctly categorized 100% of cysts.
• Inter-observer agreement for pathologic diagnosis was substantial (K=0.66).

Conclusions:

• The novel 4S algorithm significantly improves pre-operative differentiation of TGDC from DC compared to the SIST score.
• Key US characteristics include septations, depth to strap muscles, shape, and solid components.
• The 4S algorithm offers a more definitive diagnostic approach for pediatric midline neck masses.