Whilst there is an increasing prevalence of overweight and obesity worldwide, malnutrition remains common. In addition, malnutrition, overweight, and infections often interact. It is well established that malnutrition in pregnancy, resulting in an infant born with low birth weight, can increase the risk of diseases such as diabetes, heart disease and cancer in adulthood. However, the consequences of malnutrition after birth are much less studied. Severe acute malnutrition in childhood, indicated by extreme thinness, remains common in Africa and Asia. In addition, substantial numbers of adult patients with tuberculosis or HIV, diseases which are common in Africa and Asia, may become malnourished. We are interested in diabetes, which in Africa and Asia affects people at younger age and lower weight than in Europe. There is evidence from epidemiological studies that severe malnutrition in childhood and possibly in adulthood increases the risk of later diabetes but the evidence is piecemeal and there is little information as to the mechanisms involved. It is thus difficult to determine what treatments or preventative strategies are appropriate. We wish to focus on the pancreas which is a key organ in digestion and metabolic processes, especially in relation to diabetes. We will investigate pancreas size, microscopic structure, hormone and digestive enzyme production, and the body's response to these hormones among groups of people in Tanzania, Zambia, India and the Philippines. These groups have participated in the research team's previous studies of malnutrition and were malnourished before birth, as children, or as adults. They now live in places with a wide range of access to foods high in fat and sugar which could affect their risk of diabetes. We will use modern clinical methods to compare their pancreas function to that of never-malnourished controls at each site. We will use advanced statistical methods to understand the links between early malnutrition and later diabetes, taking into account the factors often associated with diabetes such as age, current overweight and infection. The project will have a substantial training component so that staff at all sites can be trained in assessment methods for nutritional status including body fat and lean content, diabetes, and pancreas function and in statistical methods. We will work with local clinicians and patient support groups to ensure that results of the project are taken up and used locally. We also plan to conduct workshops with the child participants to help them understand aspects of the science in which they are involved. Even if we find no important link between early malnutrition and later diabetes, the research will lead to improved understanding of the long-term consequences of malnutrition and the presentation and underlying metabolism of diabetes in Africa and Asia. Thus, the project will lead to improved health care for both malnourished and diabetic people.

Overall hypothesis: Severe malnutrition (MALN) at any age has medium- and long-term detrimental effects on endocrine and exocrine pancreatic function and structure. Specific objective 1 To determine if MALN earlier in life is associated with abnormalities of pancreas structure and function. Hypothesis 1: Prior MALN is associated with later abnormal pancreatic structure and endocrine and exocrine function. Approach: We will follow up cohorts from our previous research in Zambia, Tanzania, India and the Philippines in which there are participants, all now adolescents or adults, who suffered from MALN at different stages earlier in life as well as age- and sex-matched comparison groups who have never been malnourished. We will assess in those with and without prior MALN: a. blood glucose and insulin during an oral glucose tolerance test (OGTT) b. haemoglobin A1c (HbA1c) c. faecal elastase and plasma lipase as exocrine pancreas markers d. pancreas size and microarchitecture using magnetic resonance imaging (MRI) Specific objective 2 To investigate whether pancreatic abnormalities in participants with prior MALN and related diabetes are compatible with a previously prescribed entity of fibro-calculous diabetes. Hypothesis 2: Prior MALN is associated with pancreatic calcifications and is common in those with diabetes. Approach: We will conduct abdominal X-rays to investigate pancreatic calcification among all study participants identified to have diabetes as well as participants with and without prior MALN who underwent MRI imaging. Specific objective 3 To investigate the relative importance of decreased insulin production or increased insulin resistance in MALN with concurrent infections in African and Asian adult populations. Hypothesis 3: The abnormal glucose regulation seen after MALN and infection is associated with relative insulin deficiency with or without insulin resistance. Approach: This hypothesis will be investigated within the Tanzanian and Filipino adult cohorts both of whom had MALN and infections. Based on results of Objective 1, we will select participants from 4 groups: a) prior MALN and no current glucose dysregulation (glucose at 2 hours in a standard OGTT less than 7.8 mmol/L); b) prior MALN and current glucose dysregulation (glucose at 2 hours in a standard OGTT greater than 7.8 mmol/L); c) no prior MALN and no current glucose dysregulation; d) no prior MALN and current glucose dysregulation. We will: a. measure in blood samples collected every 30 minutes during the OGTT (max 120min): glucose, insulin, the incretins gastric inhibitory polypeptide and glucagon-like peptide-1, C-peptide and glucagon, and proinsulin at baseline and 120 minutes; b. conduct intravenous glucose tolerance tests and measure in blood samples collected at 5, 30, 60, 90 and 120 minutes: glucose and insulin (all time points) and C-peptide, glucagon and proinsulin at baseline, 5 and 120 minutes. Specific objective 4 To investigate whether a prior abnormal proinsulin / insulin ratio is associated with diabetes in adults infected with HIV or previously with tuberculosis. Hypothesis 4: An abnormal proinsulin/insulin ratio is associated with later development of diabetes in adults who had MALN and infection. Approach: For participants from the CICADA, Tanzania, cohort we have fasting plasma samples stored from prior follow-ups up to 10 years earlier. Glucose has already been measured in these samples as well as insulin in some samples. We will measure insulin in the remaining stored samples and proinsulin in all to investigate whether prior proinsulin / insulin ratio is associated with subsequent diabetes.