Moderate protein intake: Effects on body composition and metabolic risk factors and comparison of body composition assessment methods

Anne Keating Mahon, Purdue University

Abstract

Higher-protein (HP) diets are frequently used to enhance weight loss. Short-term HP studies lasting less than 6 months report inconsistent findings, while studies lasting 12-18 months consistently report no enhanced effect. Its effects on body weight, body composition, and cardiovascular and metabolic risk factors solely in overweight postmenopausal women have not been studied. In addition, less is known regarding the effects of protein source. Accordingly, 54 overweight postmenopausal women were instructed to consume 1 of 3 energy-restricted (ER), isocaloric diets for 9 weeks. The HP diets provided 26% protein with either beef or chicken as the primary protein source. The lower protein (LP), lacto-ovo vegetarian diet provided 16% protein. A control group followed their habitual diet. ER groups lost 9% body weight, 13% fat mass (FM), and 5% fat-free mass (FFM). The amount of decrease was not different among ER groups for FM and FFM. Weight loss was greater for women who consumed the HP-chicken vs. LP diet. ER decreased total and LDL cholesterol, but was not different among ER groups. Other CVD and metabolic biomarkers did not change. A subset of 27 women from the ER groups underwent body composition assessment utilizing hydrostatic weighing (HW), air-displacement plethysmography (ADP), dual-energy x-ray absorptiometry (DXA), and measurement of total body water using deuterium oxide isotope dilution (D2O). The four-component (4C) model with body volume from HW (4CHW) was the criterion method. 4C model with body volume from ADP (4CADP) was also compared. No group method differences were detected for changes (Δ) in percent body fat (%BF), ΔFM, and ΔFFM among 4CHW, HW, ADP, DXA, D2O, and 4CADP. Results between methods for individual women differed widely. Compared to 4CHW, 4CADP overestimated Δ%BF by 1.4 %, ΔFM by 0.6 kg and underestimated ΔFFM by 1.2 kg. Regression models were acceptable for %BF (4CADP, 2CHW, and 2C D2O), FM and FFM (4CADP, 3CDXA, 2CHW, and 2CD2O), but not for other methods. Larger studies are needed to develop population-specific regression equations to relate 2C- and 3C-models to a 4C-model estimate of Δ in body composition. Lastly, ER groups lost weight, but no diet effect was demonstrated.

Degree

Ph.D.

Advisors

Campbell, Purdue University.

Subject Area

Nutrition

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS