RSNA 2014 

Abstract Archives of the RSNA, 2014


PDS236

Clinical Findings from Semi-Automated Adipose Volume Measurements from Routine Body CT

Scientific Posters

Presented on December 2, 2014
Presented as part of PDS-TUA: Pediatric Tuesday Poster Discussions

Participants

Chunzhe Duan MS, BS, Abstract Co-Author: Nothing to Disclose
Nabile M. Safdar MD, Abstract Co-Author: Shareholder, Montage Healthcare Solutions, Inc
Evan P Nadler MD, Abstract Co-Author: Nothing to Disclose
Jiamin Liu PhD, Abstract Co-Author: Nothing to Disclose
Ronald M. Summers MD, PhD, Abstract Co-Author: Royalties, iCAD, Inc Research funded, iCAD, Inc Stockholder, Johnson & Johnson Grant, Viatronix, Inc
Marius George Linguraru DPhil, MS, Presenter: Nothing to Disclose

PURPOSE

To investigate the associations between adipose volumes, body mass index (BMI), and clinical parameters (i.e., insulin resistance (IR), hyperglycemia (HG), hypertension (HTN), blood pressure (BP) and blood glucose (BG)) in pediatric patients.

METHOD AND MATERIALS

This retrospective study was IRB approved. We collected axial contrast-enhanced CT images of 235 patients (96 boys and 139 girls) aged 13 to 20 years. The patients’ ethnic breakdown was: 47 Caucasian, 131 African-American, 40 Hispanic/Latino and 17 other race/ethnicity. There were 35 morbidly obese (BMI>40.0) and equal bins of 50 cases of under-weight (BMI<18.5), normal (18.5≤BMI<25.0), over-weight (25.0≤BMI<30.0) and obese (30.0≤BMI<40.0) cases. Adipose tissue volume measurements were performed using previously validated automatic software followed by manual corrections when necessary. We assessed the volume of visceral fat (VAT), subcutaneous fat (SAT), total fat (TAT) and body volume in the region of vertebrae L2 to L4. Percentage of fat in this region (%VAT, %SAT, %TAT) and maximum relative fat volume (MRVAT, MRSAT, MRTAT) in each axial slice were calculated. Spearman’s rank correlation and generalized linear models (GLM) adjusted for age, sex and race/ethnicity were used for data analysis.

RESULTS

All adipose volumes and BMI were significantly correlated (p<0.001 for all). Adipose volume variables from VAT, SAT and TAT, and BMI were positive correlated with systolic BP and mean BP (p<0.05 for all). Additionally, adipose variables from VAT were positive correlated with BG (p<0.05 for all), as well as MRSAT and HG (p=0.048). GLM revealed that BMI was associated with BG (p=0.01), high BG (p=0.005) and systolic and mean BP (p<0.001). In the GLM of adipose variables, VAT, SAT and TAT were associated with high BG, systolic BP, mean BP, high BG, HTN, IR and HG(p<0.05 for all).

CONCLUSION

While both BMI and adipose volume measures was associated with BG, high BG, and systolic and mean BP using GLM in our sample, volumetric measures of adipose tissue from abdominal CT were better indicators of HTN, HG and IR than BMI alone.

CLINICAL RELEVANCE/APPLICATION

Volumetric measures of adipose tissue from imaging studies, such as CT, may identify risk factors for clinical conditions associated with obesity with higher precision than the popular BMI.

Cite This Abstract

Duan, C, Safdar, N, Nadler, E, Liu, J, Summers, R, Linguraru, M, Clinical Findings from Semi-Automated Adipose Volume Measurements from Routine Body CT.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14045605.html