BSPED2015 ORAL COMMUNICATIONS Oral Communications 7 (6 abstracts)
1Sheffield University, Sheffield, UK; 2Sheffield Childrens Hospital, Sheffield, UK.
Introduction: Individuals with cystic fibrosis (CF) frequently exhibit altered insulin and glucose metabolism and many develop cystic fibrosis related diabetes (CFRD). Lung function is influenced by glucose metabolism with changes in glucose metabolism resulting in deterioration in lung function. Recommendations suggest CF patients should have an OGTT annually to screen for the development of CFRD. We examined the OGTT profiles to ascertain whether simpler fasting measures of insulin resistance and beta cell function derived from the OGTT could predict future lung function.
Methods: Data on insulin, glucose and C-peptide were collated over a period of 3 consecutive years for 81 patients with CF (aged 917) during their annual OGTT. This was correlated with lung function (FEV1 and FVC). A number of surrogates of insulin resistance and beta cell function such as HOMA score, QUICKI and fasting insulin:glucose ratios were examined together with markers of lung function using classification tree modelling to establish whether these surrogates could predicted future changes in lung function.
Results: Patients with CFRD and impaired glucose tolerance (IGT) showed significantly later peaks of both insulin and glucose following an oral glucose load compared to CF patients with normal glucose metabolism and the general population. Insulin and glucose both peaked late at 90 min in those with CFRD and IGT compared to a peak glucose at 3060 min and peak insulin level at 60 min in those with CF and normal insulin:glucose handling.
A classification tree model incorporating data of fasting C-peptide, fasting insulin and glucose and 30 min insulin and glucose was 78.4% accurate in predicting a 10% change in FEV1 a year later.
Conclusion: CF patients whose insulin reserve and glucose handling is declining surprisingly show very delayed peaks in absorption of glucose and in production of insulin. This delay predicts declining lung function. Classification tree models offer a potentially useful tool by which to identify patients most at risk of future declines in lung function though they are not as yet refined enough to replace screening by OGTT.