Sex-Based Differences in Imaging-Derived Body Composition and Their Association with Clinical Malnutrition in Abdominal Surgery Patients.

Among 1143 abdominal surgery patients, 20.2% had clinical malnutrition, with prevalence varying substantially by procedure type.

• Study was cross-sectional, conducted at a single institution from 2018-2021.
• Patients underwent preoperative abdominal CT scan and elective abdominal surgery.
• Malnutrition prevalence ranged from 3.5% to 38.2% depending on procedure type.
• Clinical malnutrition was diagnosed by registered dietitians using standardized criteria.

A deep learning algorithm was used to quantify five muscle groups and two fat depots from preoperative CT scans.

• Five muscle groups were quantified from CT scans.
• Two fat depots were quantified from CT scans.
• Sex-specific associations between imaging features and malnutrition were evaluated using logistic regression.
• This approach enabled comprehensive body composition assessment beyond standard single-slice measurements.

In males, malnutrition was associated with decreased psoas volume, decreased quadratus lumborum volume, and reduced erector spinae attenuation.

• Decreased psoas volume was associated with malnutrition in males (OR: 0.58, 95% CI [0.41-0.82]).
• Decreased quadratus lumborum volume was associated with malnutrition in males (OR: 0.52, 95% CI [0.35-0.77]).
• Reduced erector spinae attenuation was associated with malnutrition in males (OR: 0.58, 95% CI [0.41-0.82]).

In females, both decreased psoas volume and decreased psoas attenuation were associated with malnutrition.

• Decreased psoas volume was associated with malnutrition in females (OR: 0.56, 95% CI [0.41-0.77]).
• Decreased psoas attenuation (myosteatosis) was associated with malnutrition in females (OR: 0.59, 95% CI [0.44-0.79]).
• Psoas attenuation as a marker of myosteatosis was not significantly associated with malnutrition in males, representing a sex-specific finding.

Myosteatosis was associated with malnutrition only in females, not in males.

• Muscle attenuation on CT, a measure of myosteatosis (fat infiltration into muscle), showed sex-specific associations with malnutrition.
• In females, reduced psoas attenuation was associated with malnutrition (OR: 0.59, 95% CI [0.44-0.79]).
• In males, erector spinae attenuation but not psoas attenuation was associated with malnutrition.
• This finding suggests sex-based differences in the pattern of muscle fat infiltration related to malnutrition.

Increased subcutaneous fat attenuation was associated with malnutrition in both sexes, with a stronger association in females than males.

• In males, increased subcutaneous fat attenuation was associated with malnutrition (OR: 1.58, 95% CI [1.22-2.04]).
• In females, increased subcutaneous fat attenuation was associated with malnutrition (OR: 1.96, 95% CI [1.54-2.50]).
• Higher fat attenuation on CT may indicate reduced lipid content or altered fat composition associated with malnutrition.

Increased visceral fat attenuation was associated with malnutrition in both sexes, with a stronger effect in females.

• In males, increased visceral fat attenuation was associated with malnutrition (OR: 1.43, 95% CI [1.07-1.90]).
• In females, increased visceral fat attenuation was associated with malnutrition (OR: 1.68, 95% CI [1.29-2.20]).
• Both subcutaneous and visceral fat attenuation showed positive associations with malnutrition across both sexes.

Reduced muscle volume was associated with malnutrition in both sexes.

• Both males and females showed associations between reduced muscle volume and clinical malnutrition.
• Specific muscles showing volume associations differed by sex: psoas and quadratus lumborum in males, psoas in females.
• This finding supports muscle volume as a cross-sex marker of malnutrition in abdominal surgery patients.