Objective: To determine the level of a single transverse scan of intra-abdominal fat between L1 and L5 vertebrae that best predicts intra-abdominal fat volumes.
Subjects: Sixteen male and seven female patients with non-insulin-dependent diabetes mellitus, aged 44-74 y.
Outcome measures: Volumes and areas from single scans of intra-abdominal fat measured by magnetic resonance imaging with a 1.5 Tesla magnetic field strength.
Results: Intra-abdominal fat volumes and were calculated from fat areas from eight cross-sectional transverse single scans (nine scans in eight men) of 20 mm thickness. Men and women, respectively, had mean body mass index (BMI) of 27.9 (s.d. 3.0) and 31.6 (s.d. 4.7) kg/m2, and intra-abdominal fat of 2.3 (s.d. 0.5) and 2.5 (s.d. 0.6) kg. Intra-abdominal fat area of the fourth scan (in the direction of L1 to L5) gave the highest prediction of total intra-abdominal fat both in men (r = 0.959, P < 0.001) and in women (r = 0.973, P < 0.001). The intra-abdominal fat area of the third scan gave almost as good a prediction. These third and fourth scans corresponded to L2 and L3 vertebrae. The intra-abdominal fat areas from the sixth and seventh scans, corresponded to the frequently used L4-L5 and had lower correlations with intra-abdominal fat. There were no gender differences in the prediction of volumes from areas of intra-abdominal fat. Intra-abdominal fat areas of the fourth scan explained 93% of variance (SEE = 0.14 kg) of total of intra-abdominal fat for both genders: intra-abdominal fat (kg) = 0.0108 x intra-abdominal fat area of the fourth scan (cm2) + 0.244.
Conclusions: In large studies of intra-abdominal fat, using magnetic resonance imaging (MRI) or computerised tomography scanning, a single intra-abdominal fat area at the intervertebral disc between L-2 and L-3 vertebrae offers a cheaper, faster and safer method, with high prediction of total intra-abdominal fat volumes and masses.