Article Text
Abstract
Background Screening for cystic fibrosis related diabetes (CFRD) is recommended as part of the cystic fibrosis annual review. For children over 10 an oral glucose tolerance test (OGTT) is recommended, but this is time consuming and adds to the burden of assessments and investigations of the annual review. As part of a proposed wider pragmatic study to evaluate the feasibility of home OGTT testing in this patient group, we assessed the level of agreement between blood glucose concentration measured by near-patient testing using the Accucheck glucometer and that measured in the laboratory using standard methods (True Lab Glucose—TLG).
Methods 26 children with CF aged 1 yr–16 yrs undergoing annual review were studied. Blood taken at the time of venepuncture was tested for glucose concentration using an Accucheck glucometer, and results compared with TLG. For children under 10 a single, random blood glucose was tested using each method. For those aged 10 or over samples taken after an overnight fast, and 2 h after a standard OGTT oral glucose load were tested using each method. Bland and Altman statistics were used to assess limits of agreement between methods of glucose measurement.
Results 31 pairs of glucose measurements were obtained. TLG measurements ranged from 4.1 to 10.6 mmol/l. Abstract P241 Figure 1 shows the Accucheck glucose measurement plotted against the TLG for each sample, and the line of equality. The agreement between Accucheck and TLG was investigated using the method of Bland and Altman and showed that the Accucheck gave results a mean (SD) of 0.26 (0.84) mmol/l higher than values obtained on TLG.
Conclusions There is a clinically acceptable agreement between blood glucose measured by Accucheck compared with TLG. Just one child with a TLG result that might have prompted further investigation could have been missed, had Accucheck been the sole method used. This suggests that blood glucose measurement in the context of OGTT screening for CFRD could perhaps be performed using near-patient testing in an out of hospital setting (eg, by parents at home) without compromising test validity. Further research is needed to establish if this is the case.