Developmental Adaptation to an Obesogenic Environment

Differences in birth weight between children with kwashiorkor (K) and marasmus (M) suggest that intrauterine factors influence the development of these two syndromes of severe acute malnutrition (SAM). Such an effect could play a role in setting specific targets for dietary protein that has been proposed to influence total energy intake (TEI), body weight and risk of obesity.

We recruited 63 adult survivors of SAM, aged 17 to 46 years, who in childhood had been treated at the Tropical Metabolism Research Unit (TMRU) Metabolic Ward. Data on birth weight, age, weight and length at admission were abstracted from TMRU records. Subjects provided written informed consent. The study was approved by the Faculty of Medical Sciences Ethics Committee, University of the West Indies.

Measurements were conducted in two phases. All subjects were seen in single-sex pairs and stayed in a dedicated metabolic suite for 9 consecutive days. During phase 1 (days 1 to 3- choice experiment) they ate freely at each mealtime from menus comprising a combination of foods containing different percentages of energy as protein (PEP), set at 10%, 15% or 25% (26). During phase 2 (days 4 to 8), pairs were randomized to one of three groups each of which received only foods that contained 10%, 15% or 25% PEP. The participants were blinded to the treatment. Participants were taken for a 1-hour supervised walk each day at 4 pm.

Thirty-one local recipes of 10 sweet and 21 savoury foods were selected. Each was modified into three recipes containing 10, 15 or 25% energy as protein through the addition of food ingredients, a protein mix and/or maltodextrin (Ross Nutrition). Carbohydrate was adjusted to be 60, 55 or 45% energy and dietary fat was kept constant at 30%. The PEP versions of each food/recipe were taste tested for the ability to determine the protein concentration of any dish due to appearance, smell or texture as well as for pleasantness. If taste testers were able to detect any difference, the recipes were adjusted and retested until no difference was detected. Up to 11 foods were provided on each day during the 8-day period, giving participants both variety and choice at all times as indicated in table 1. During the first three days (phase 1), three food items other than fruit and vegetable salads and tea were offered at breakfast, lunch and dinner to all the participants. These three food items at each of these meal times included foods containing 10%, 15% or 25% protein energy. From days 4 to 8 (phase 2), these same daily food types were repeated every three days but the foods contained 10 PEP in one group, in another group they all contained 15 PEP, and in the third group they all contained 25 PEP protein.

Breakfast was provided at 8 am, lunch at 12:30 pm and dinner at 6 pm. During both phases snack items were made freely available at all times. The amount eaten was determined by weighing each food item before consumption and any not eaten to the nearest gram. The foods were served in weighed quantities on tared containers. Plates were of a single design and neutral (white) color. The same size, style and color plates were used for the 10, 15 and 25 PEP version of each food. Participants were offered optional foods including 100g fruit salad, and decaffeinated tea sweetened with a measured amount of brown sugar with breakfast, and 100g vegetable salad with lunch and dinner. A three-day food diary completed prior to the 8 day ad libitum test period and foods consumed during the test period were analysed for total energy, protein, carbohydrate and fat content.

The weight of the subjects without shoes and in light clothing was measured daily to the nearest 0.1 kg; height was measured to the nearest 0.1 cm. Fat and fat-free mass were obtained from DEXA measurements. Weight gain was calculated from day 1 to 4 in phase 1 and from day 4 to 8 in phase 2.