The association of dietary macronutrients with anthropometric changes, using iso-energetic substitution models: Tehran lipid and glucose study

Obesity is a growing epidemic of the twenty-first century in developed and developing countries [1]. It is associated with an increased risk of serious health problems including cardiovascular diseases, diabetes and several cancers [2]. The appropriate method for its prevention and treatment is widely acknowledged. Weight gain often occurs gradually over a lifetime due to imbalance of energy intake and expenditure [3]. In addition, macronutrient composition has also been suggested as an important determinant in the obesity epidemic [4]; carbohydrate, fat and protein intake are simultaneously considered as macronutrient balance or overall diet, and in an iso-energetic setting the differences in one dietary macronutrient reflect substitutions for other macronutrients. Analyzing the association between macronutrient intakes and weight change in statistical methods which account that one macronutrient is substituted with the other, may help clarify the complex relationships. Several previous randomized clinical trials [5‐7] have shown that increase in protein percentage in iso-caloric diets had advantages in terms of adiposity, blood lipid profiles, cardiovascular risk markers and insulin levels; however, in short term controlled trials, high protein, moderate or low carbohydrate diets result in more weight loss than traditional low fat, high carbohydrate diets [7]. Long term intervention studies indicate equivalent weight loss with the low fat (< 30% of energy from fat) and low carbohydrate (< 50% of energy from carbohydrate) approaches [6]. Another study on healthy individuals showed that higher protein intake was associated with weight gain [8]. Moreover, one study showed differences in animal and plant protein, with a direct association of animal protein with the risk of overweight and obesity, and an inverse association for plant protein [9]. Therefore, the consequences of optimal macronutrient composition of the usual diet still remain controversial in observational studies. In addition, the debate on protein sources is ongoing, addressing the nutritional quality of dietary proteins and their amino acid composition. Hence, the aim of the current study was to evaluate the substitution dietary macronutrient intakes with each other in relation to anthropometric changes in a group of Tehranian adults.

Characteristics of the population based on sex and percentage of energy from each macronutrient are presented in Table 1. Compared to subjects in the first tertile of energy from carbohydrates, subjects in the third tertile were older, had a lower frequency of current smokers, and lower change in weight and WC during follow-up. Men in the third tertile of energy from carbohydrate had a higher frequency of participants with university degrees, compared to the first tertile; while women in the mentioned tertile, this frequency was lower, compared to the first tertile. Compared to subjects in the first tertile of energy from protein, subjects in the third tertile had a higher education level, higher BMI and WC at baseline and reported a lower energy intake. Women in the third tertile of energy from protein were older compared to the first tertile. Compared to subjects in the first tertile of energy from fat, subjects in the third tertile were younger, had a higher weight change during follow-up and reported higher energy intake; they also had a higher percentage of smokers. In men, fewer subjects in the third tertile of energy from fat had university degrees, compared to the first tertile. Also, WC change increased according to the tertiles of energy from fat in women.

Adjusted weight change for the iso-energetic replacement of 5% of energy from one macronutrient by another macronutrient is presented in Table 2. A 1 unit higher proportion of plant protein at the expense of animal protein (β = − 0.84), non-starch carbohydrates (β = − 0.86), saturated fat (β = − 0.76), mono-unsaturated fat (β = − 0.76), and poly-unsaturated fat (β = − 0.86) was associated with weight loss (P < 0.05). A 1 unit higher proportion of plant protein at the expense of animal protein, non-starch carbohydrates and saturated and poly-unsaturated fatty acids was associated with decreased risk of weight attainment (Table 3). The substitution of other macronutrients was not consistently associated with weight change.

A 5 unit higher proportion of plant protein at the expense of animal protein (β = − 1.60), non-starch carbohydrates (β = − 1.60), saturated fat (β = − 1.54), mono-unsaturated fat (β = − 1.57), and poly-unsaturated fat (β = − 1.61) was associated with WC loss (P < 0.05) (Table 4).

Odds ratio (OR) of increase in WC according to an iso-energetic substitution of macronutrients is shown in Table 5. A 1 unit higher proportion of plant protein at the expense of animal protein, non-starch carbohydrates or fatty acids was associated with a lower risk of WC increase in men and women (Table 5).

In our study, the iso-caloric substitution of carbohydrate with protein was not associated with weight change, whereas the substitution of animal protein, carbohydrate and fatty acids with plant protein was associated with weight loss and reduced WC. This finding can partly justify the contradiction between studies regarding the effect of protein intake on obesity, it seems that the type of protein)animal versus plant protein) is more effective in weight gain than its amount. These results are similar to the findings of another longitudinal study which reported an increase in the risk of obesity with higher animal protein intake, during a 7-year follow up [20]. In contradiction with our findings, two cross sectional studies proposed the use of more animal protein as an effective way for weight loss [21, 22]. Differences in the study design can partly justify this contradiction. Differences in quality, quantity and type of animal protein consumed in different cultures could be another justification for this controversy over the effect of animal protein intake on obesity indices.

Several mechanisms have been proposed for the beneficial effect of plant protein intake on weight control. First, high intake of fiber from plant protein may decrease digestibility, thus, leading to weight loss [23]. Second, plant protein intake increases lean body mass and prevents sarcopenia, leading to an increase in both basal metabolism and physical activity, which in turn reduces the risk of obesity [24]; finally, consumption of legumes increases the expression of proteins involved in fatty acid oxidation, which induce weight loss [25].

In our study, the iso-caloric substitution of fat with carbohydrate decreased the risk of obesity in men. No studies have examined the effects of fat substitution with carbohydrate on obesity, separately in men and women; although, contrary to our results, in two clinical trials, obese subjects lost more weight during 6 months of a carbohydrate-restricted diet than a fat-restricted diet [5, 26]. Diets high in carbohydrate are particularly fattening because of their propensity to elevate insulin secretion, thereby directing fat toward storage in adipose tissue [27]. However, Hall and Guo claimed that low-fat diets increase both energy expenditure (26 kcal/day; P < 0.0001) and fat loss (16 g/day; P < 0.0001) [28], that can effectively decrease weight gain in long term which partly justifies our results.

In our study, replacing all three types of fatty acids (SFA, MUFA and PUFA) with carbohydrate contributed to the prevention of weight gain; however, in another cohort study, increasing the consumption of PUFAs and plant-based MUFA at the expense of carbohydrates was associated with less weight gain; and increasing consumption of SFA and trans-fat was associated with greater weight gain [29]. Undoubtedly, the intakes of different types of fatty acids and its major food sources are different in societies, and this difference can justify these conflicting results.

The effect of iso-caloric substitution of dietary macronutrients on weight or waist change is different by sex, which may be because weight change in women is more dependent on calorie consumption, insulin sensitivity [15] and hormone levels [30].

The prospective design of the study, using a valid and reliable FFQ, and analyzing data using iso-caloric substitution models are strengths of the present study. In the substitution model, the effect of substituting one nutrient intake with another was investigated while energy intake in the diet was kept constant [31]. Weight changes occur gradually at the population level, therefore our 3.6 year assessment period is consistent with the time course of weight change in response to a change in diet [32].

Some limitations should also be mentioned; first, despite using a validated FFQ in this study, FFQs, like other tools of dietary assessments, are subject to measurement errors. Second, the study participants are not representative of the whole Iranian population, which limits the generalizability of our findings to other Iranian and non-Iranian populations that have different proportions of macronutrients in their habitual diet compared to our study participants; third, the use of a substitution model is justified when studying the health effects of different macronutrients in iso-energetic conditions, this approach is only a mathematical model for dietary intakes and not a real-life condition. Finally, we did not have enough information about family income; therefore, in our study, it was not possible to investigate the effect of family income on macronutrient intakes.

our findings demonstrated that the proportions of macronutrient intakes in habitual diets may be associated with obesity indices; higher proportions of plant protein in diet was associated with lower increase in BMI and WC, and decreased the risk of obesity during 3.6 years of follow-up, especially in men.