Visceral Fat Quantified by a Fully Automated Deep-Learning Algorithm and Risk of Incident and Recurrent...

Area of Science:

• Gastroenterology
• Radiology
• Epidemiology

Background:

• Obesity is a known risk factor for diverticulitis.
• Visceral fat area offers a more precise measure of abdominal adiposity than general obesity.
• The specific association between visceral fat and diverticulitis risk remains under investigation.

Purpose of the Study:

• To determine the association between visceral fat area and the risk of developing new (incident) and recurring (recurrent) diverticulitis.
• To evaluate the risk of complicated diverticulitis and hospitalizations due to recurrence based on visceral fat levels.

Main Methods:

• Retrospective cohort study involving 6,654 patients from the Mass General Brigham Biobank.
• Visceral fat area, subcutaneous fat area, and skeletal muscle area were quantified using a deep-learning model on abdominal CT scans.
• Risk of diverticulitis was assessed based on quartiles of visceral fat area z-scores, controlling for age, sex, and race.

Main Results:

• A higher visceral fat area z-score significantly increased the risk of incident diverticulitis (HR 2.09), complicated diverticulitis (HR 2.56), and recurrent diverticulitis requiring hospitalization (HR 2.76).
• The association between visceral fat and diverticulitis risk was consistent across different BMI strata.
• Subcutaneous fat area and skeletal muscle area showed no significant association with diverticulitis risk.

Conclusions:

• Increased visceral fat area, as measured by CT scans, is significantly associated with a higher risk of both incident and recurrent diverticulitis.
• These findings enhance understanding of diverticulitis pathophysiology.
• The results suggest potential implications for developing targeted preventive strategies for diverticulitis.