This study looked at whether early changes in resting energy expenditure (REE) and respiratory quotient (RQ) are correlated with later weight changes in patients with type 2 diabetes (T2D) being treated with insulin or GLP-1 analogues, or diet.

A total of 67 patients (age: 57±9 years; BMI: 33.7±5.0kg/m²; HbA_1c: 9.9±1.5%) began taking an insulin analogue at bedtime (INS, n=28; initial dose: 0.2 IU/kg) or a GLP-1 analogue (GLP-1, n=23), or only a dietary intervention (diet, n=16; restricted carbohydrates and calories). Their respiratory exchanges were monitored on days 0, 1 and 2 before breakfast.

Two days after starting the bedtime insulin analogue, fasting glycaemia improved (INS: −65±41mg/dL; GLP-1: −29±48mg/dL; diet: −31±46mg/dL; P<0.05), REE decreased (INS: −162±241kcal/24h; GLP-1: 0±141kcal/24h; diet: −41±154kcal/24h; P<0.05) and RQ increased (from 0.76±0.04 to 0.80±0.04; P<0.01), whereas only RQ decreased with diet (from 0.79±0.05 to 0.76±0.04; P<0.05) and remained unchanged with GLP-1 (P<0.005 for ΔRQ across treatments). Only 33 patients attended the scheduled examination three months later. HbA_1c improved (INS, n=16: −1.7±1.4%; GLP-1, n=12: −2.1±1.4%; diet, n=5: −1.7±2.8%; NS), while weight changes differed (INS: +1.5±4.3kg; GLP-1: −2.8±2.8kg; diet: −2.2±2.7kg; P<0.005). After three months, weight changes correlated with early changes in REE (r=−0.37, P<0.05) and RQ (r=+0.43, P<0.01), and remained correlated when both changes were included in a multivariate regression analysis (r=0.58, P<0.005).

In poorly controlled patients with T2D and two days after the introduction of a bedtime insulin analogue, REE decreased by −9% while RQ increased by +5%, pointing to a reduction of lipid oxidation. These changes were predictive of later weight gain.