Improved 4-compartment body-composition model for a clinically accessible measure of total body protein

Published on Mar 1, 2013in The American Journal of Clinical Nutrition6.57
· DOI :10.3945/ajcn.112.048074
Joseph Wilson5
Estimated H-index: 5
(UC: University of California),
Boyd Josef Gimnicher Strauss38
Estimated H-index: 38
+ 2 AuthorsJohn A. Shepherd47
Estimated H-index: 47
(UC: University of California)
Background: Muscle wasting is a consequence of many primary conditions including sarcopenia, cachexia, osteoporosis, HIV/AIDS, and chronic kidney disease. Unfortunately, there is not a clinically accessible method to measure total body protein, which is the functional mass of muscle. Objective: We sought to derive a simple method to measure total body protein by using dual-energy X-ray absorptiometry (DXA) and bioimpedance analysis (BIA). Design: We retrospectively analyzed a clinical convenience sample of individuals with numerous metabolic conditions from the Monash Medical Centre, Melbourne, Australia, who had a concurrent protein measure by using neutron activation analysis-derived protein (NAA-TBPro), water measure by using BIA, and whole-body DXA scan. The study was split into calibration and validation data sets by using simple random sampling stratified by sex, BMI category, and age decade. We generated a protein estimate directcalibration protein (DC-TBPro) derived from BIAwater, bone mass, and body volume. We compared NAA-TBPro with DC-TBPro and 2 protein estimates from the literature, one that used the DC-TBPro equation with fixed coefficients [4-compartment Lohman method for analysis of total body protein (4CL-TBPro)] and another that used fat-free mass, age, and sex [Wang equation‐derived protein (W-TBPro)]. Results: A total of 187 participants [119 women; mean (6SD) age: 37.0 6 15.4 y; mean (6SD) BMI (in kg/m 2 ) 24.5 6 7.7] were included. When plotted against NAA-TBPro, DC-TBPro had the highest correlation [coefficient of determination (R 2 ) = 0.87], lowest root mean squared error (RMSE; 0.87 kg), and fewest outliers compared with 4CL-TBPro (R 2 = 0.75; RMSE = 1.22 kg) and W-TBPro (R 2 = 0.80; RMSE = 1.10 kg). Conclusions: A simple method to measure total body protein by using a DXA system and BIA unit was developed and compared with NAA as proof of principle. With additional validation, this method could provide a clinically useful way to monitor musclewasting conditions. Am J Clin Nutr doi: 10.3945/ajcn.112. 048074.
  • References (31)
  • Citations (23)
#1Joseph Wilson (University of California, Berkeley)H-Index: 5
#2Bo Fan (UCSF: University of California, San Francisco)H-Index: 29
Last.John A. Shepherd (University of California, Berkeley)H-Index: 47
view all 3 authors...
#1Winston W. K. Koo (UTHSC: University of Tennessee Health Science Center)H-Index: 11
#2Larry R. Massom (UTHSC: University of Tennessee Health Science Center)H-Index: 4
Last.Jocelyn Walters (UTHSC: University of Tennessee Health Science Center)H-Index: 10
view all 3 authors...
Cited By23
#1Grant M. Tinsley (TTU: Texas Tech University)H-Index: 6
#2M. Lane Moore (TTU: Texas Tech University)H-Index: 1
Last.Megan R. Cruz (TTU: Texas Tech University)H-Index: 1
view all 10 authors...
#1Zachary Clayton (SDSU: San Diego State University)H-Index: 1
#2Zachary S. Clayton (SDSU: San Diego State University)H-Index: 1
Last.Mark Kern (SDSU: San Diego State University)H-Index: 19
view all 7 authors...
View next paperReliability of body-fat estimations from a four-compartment model by using density, body water, and bone mineral measurements.