173 Validation of near patient testing blood capillary glucose for the oral glucose tolerance test in cystic fibrosis

Introduction Early diagnosis and treatment of CFRD is important to slow pulmonary function decline and improve outcomes. Oral glucose tolerance test (OGTT) using venous plasma samples (VPS) is the accepted method for diagnosis recommended by WHO. Near patient testing (NPT) by blood capillary sampling (BCS) provides real time results but this technique has not been validated in CF. Aim To validate the accuracy of NPT for an OGTT by BCS using HemoCue Glucose 201 RT™ (HG) by comparing this with standard laboratory VPS. Methods Prospective study of clinically stable patients undergoing OGTT (June 2013 to April 2014). All patients had simultaneous VPS and BCS at 0 min (fasting) and 120 min after a glucose 75 g solution drink. BCS was analysed immediately using HG (daily quality control, QC). VPS were analysed in an accredited laboratory by standard methods. Results 70 patients were recruited. QC for BCS and reproducibility from control material was excellent. WHO criteria for diabetes were used: Lab (VPS) at 120 min – normal glucose tolerance (NGT), n=53 (76%); impaired (IGT), n=9 (13%); and, diabetic (DGT), n=8 (11%). BCS at 120 min – NGT, n=54 (77%); IGT, n=9 (13%); and DGT, n=7 (10%). Sensitivity and specificity of BCS for DGT was 87.5% and 98.4%, respectively. The correlation coefficient(R 2 ) of the different sampling techniques was 0.47 and 0.95 for fasting and 120 min samples, respectively. Conclusion There was a consistent discrepancy between absolute values for BCS and VPS, particularly for fasting samples. Although this had limited impact on diagnostic categories, the sensitivity may not be acceptable for clinical implementation; this needs further evaluation.