Poster Presentation Australian and New Zealand Obesity Society Annual Scientific Conference 2024

The role of bioelectrical impedance analysis in predicting 12-month weight loss in people with class 3 obesity (#272)

Mina Toumas 1 , Ritesh Chimoriya 2 , Pamela A Reyes 2 , Kathy Grudzinskas 1 , Lisa Sengul 1 , Nick Kormas 1 , Milan Piya 1 2
  1. South Western Sydney Metabolic Rehabilitation and Bariatric Program, Camden Hospital, SWSLHD, Camden, NSW, Australia
  2. School of Medicine, Western Sydney University, Campbelltown, NSW, Australia

 

 

Aim:

This study aimed to assess the use of body fat percentage (BF%), fat free mass (FFM) and bioimpedance as a predictor of weight loss in people with class 3 obesity attending a non-surgical multidisciplinary metabolic program for 12 months.

 

Methods:

All adults enrolling in a public hospital-based outpatient metabolic program in Sydney between April 2019 and March 2023 had their bioelectrical impedance analysis (BIA) performed at baseline (Marsden MBF-6010 Body Composition Scale) to collect data on BF%, FFM and bioimpedance. Exclusions were people unable to stand on the BIA scales, those with lymphoedema bandaging or open foot wounds. Weight and anthropometric data were collected at baseline and 12-months. Hierarchical multiple regression analysis was used to predict 12-month weight loss using various indicators of bioimpedance. 

 

Results:

There were 140 people with baseline BIA data; 70.7% female (n=99), and 42.9% had type 2 diabetes (n=60). Baseline mean weight was 147.4±29.5kg, BMI 53.8±10.4kg/m2, BF% 45.4±8.7%, FFM 81.56±23.5kg and bioimpedance 348.3±83.5 Ohm. Mean weight loss at 12 months was 7.5±9.7kg (5.4±6.1% total body weight). The hierarchical multiple regression analysis predicting 12-month weight loss involved three steps. In Step 1, BF% significantly predicted weight loss (B=1.460, t=4.337; p<.001), accounting for 13% of the variance (R2=0.130). Step 2 added FFM, which did not significantly improve the model (ΔR2=0.000, B=0.007, t =0.065), leaving R2 unchanged. In Step 3, bioimpedance was added, resulting in a small but significant increase in explained variance (ΔR2=0.009), although bioimpedance itself was not a significant predictor (B=−0.035, t=−1.124). Overall, BF% remained the only significant predictor throughout the analysis. 

 

Conclusion: 

In people with class 3 obesity, 12-month weight loss in a non-surgical public metabolic program was best predicted by BF%. Routine measurement of BIA at baseline may help inform 12-month weight loss, although larger studies are needed to validate these results.