Comparison of ten predictive equations for estimating lean body mass with dual-energy X-ray absorptiometry in older patients

Tanuj Puri, Glen M. Blake

Research output: Contribution to journalArticlepeer-review

Abstract

Objectives: White fat contributes to body weight (BW) but accumulates very little [18F]fluorodeoxyglucose ([18F]FDG) in the fasting state. As a result, higher standardised uptake values normalised to BW (SUV) are observed in non-fatty tissue in obese patients compared to those in non-obese patients. Therefore, SUV normalised to lean body mass (SUL) that makes tumour uptake values less dependent on patients' body habitus is considered more appropriate. This study aimed to assess ten mathematical equations to predict lean body mass (LBM) by comparison with dual-energy X-ray absorptiometry (DXA) as the reference method. Methods: DXA-based LBM was compared with ten equation-based estimates of LBM in terms of the slope, bias and 95% limits of agreement (LOA) of Bland-Altman plots, and Pearson correlation coefficients (r). Data from 747 men and 811 women aged 60-65 years were included. Results: Gallagher's equation was optimal in males (slope = 0.13, bias = -2.4 kg, LOA = 12.8 kg and r = 0.900) while Janmahasatian's equation was optimal in females (slope = 0.14, bias = -0.9 kg, LOA = 10.7 kg and r = 0.876). Janmahasatian's equation performed slightly better than Gallagher's in the pooled male and female data (slope = 0.00, bias = -1.6 kg, LOA = 12.3 kg and r = 0.959). Conclusions: The Gallagher and Janmahasatian equations were optimal and almost indistinguishable in predicting LBM in subjects aged 60-65 years.

Original languageEnglish
JournalThe British journal of radiology
Volume95
Issue number1133
Early online date10 Feb 2022
DOIs
Publication statusPublished - 1 May 2022

Research Beacons, Institutes and Platforms

  • Manchester Cancer Research Centre

Fingerprint

Dive into the research topics of 'Comparison of ten predictive equations for estimating lean body mass with dual-energy X-ray absorptiometry in older patients'. Together they form a unique fingerprint.

Cite this