Search

Introduction: We compared test performance and cost per case for strategies detecting diabetes on the oral glucose tolerance test (OGTT) using either (a) glycated haemoglobin (HbA1c) ≥ 6.5% (48 mmol/mol) or (b) two HbA1c thresholds where the first cut-point 'rules out' and the second 'rules in' diabetes. HbA1c values in between the thresholds require confirmatory glucose testing for diagnosis., Materials and Methods: We conducted an analysis of adults aged 40-75 years from the Leicester Ethnic Atherosclerosis and Diabetes Risk (LEADER) cohort (Leicester, UK), from 2002 to 2008, who underwent oral glucose tolerance testing (OGTT) and HbA1c testing., Results: From 8696 individuals (mean age 57.3 years, 73% white Europeans (WE) and 27% South Asians (SA)), HbA1c ≥ 6.5% produced sensitivity of 62.1% for detecting diabetes in WE and 78.9% in SA. Using two selected thresholds, HbA1c ≤ 5.8% (rule-in, 40 mmol/mol) and HbA1c ≥ 6.8% (rule-out, 51 mmol/mol) produced high sensitivity/specificity (> 91.0%) for detecting diabetes, however, 28.8% of the cohort with HbA1c 5.9%-6.7% required a subsequent glucose test. The two cut-point threshold produced a lower cost per case of diabetes detected in WE, compared to HbA1c ≥ 6.5% of £38.53 (1.89 to 86.81) per case, but was more expensive in SA by £84.50 (69.72 to 100.92) per case. Using a risk score to determine HbA1c testing, the same costs per case became £63.33 (23.33 to 113.26) in WE and £69.21 (55.60 to 82.41) in SA., Conclusion: Using a two-threshold strategy may have some benefits over a single cut-point; however, 28.8% of individuals required two blood tests.

There is still a debate as to whether or not glucose variability contributes to diabetes complication risk. There is much in vitro laboratory evidence that glycaemic instability gives rise to increased production of reactive oxygen species and has a detrimental effect on endothelial dysfunction. While some in vivo studies have also shown similar findings in patients with or without diabetes, others have been unable to confirm any association. Furthermore, clinical studies which have sought to translate this possible risk into an increased likelihood of developing micro- or macrovascular complications have so far not demonstrated an effect. However, few of these trials were specifically designed to establish any influence of glucose fluctuations. This issue is now one of the largest remaining unanswered questions in diabetes. While this article focuses on the data which do not support a role for glucose variability in the development of complications, it also highlights the need for further studies to be performed which will definitively resolve the matter.

Aims/hypothesis: It is not known whether mean blood glucose (MBG) predicts the risk of macrovascular complications in diabetes any differently from HbA1c. In this study we therefore analysed data from the Diabetes Control and Complications Trial (DCCT) to assess the relationship between MBG, HbA1c and glucose variability with regard to the risk of cardiovascular disease in patients with type 1 diabetes., Methods: Pre- and postprandial seven-point glucose profiles were collected quarterly during the DCCT in 1441 individuals. The relationship between time to first cardiovascular event and MBG, HbA1c and daily SD of blood glucose was assessed by Cox regression after adjusting for the known risk factors of macrovascular disease and the treatment groups of the patients., Results: Cox regression showed MBG to be predictive of a cardiovascular event (p=0.019), but not HbA1c (p=0.858). A rise of 1 mmol/l in MBG was associated with an 11% rise in cardiovascular risk. MBG remained highly predictive (p=0.015) even after adjustment for HbA1c values and glucose variability., Conclusions/interpretation: This study has shown that during the DCCT MBG was a better predictor of the macrovascular complications of type 1 diabetes than HbA1c. It indicates that the cardiovascular risk associated with hyperglycaemia appeared within the time period of the study and that blood glucose rather than HbA1c may be the preferred means of assessing this risk.

Introduction: There is little evidence that the addition of interpretative comments to biochemistry reports can influence outcome for patients. Interpretative comments on thyroid function test (TFT) requests were introduced in Hull in August 1999, providing the opportunity to determine whether feedback on hypothyroid patients taking thyroxine could lead to a reduction in the proportion whose thyroxine was inadequately replaced., Patients and Methods: The study comprised 15 584 TFT requests, made from 1 August 1999 to 30 August 2002 by general practitioners (GPs), for 8281 patients taking thyroxine. Under-replacement of thyroxine, defined as a TSH concentration above the upper reference limit (i.e. 4.7 mU/L), was usually commented on in the biochemical report., Results: In the first, second and third years following introduction of interpretative comments, the proportions of samples with a TSH concentration of >4.7 mU/L were 21.3%, 17.6% and 16.6%, respectively (chi(2)(trend) = 43.1, P <0.0001). The proportion with a TSH concentration of <0.1 mU/L showed a more modest change, from 12.5% in year 1 to 14.0% and 14.8% in years 2 and 3, respectively (chi(2)(trend) = 22.3, P <0.0001)., Conclusion: This study shows that in the three years following the introduction of interpretative comments there was a 22% reduction in the number of GPs' samples indicating thyroxine under-replacement. It seems likely that these data provide evidence that comments can indeed influence the biochemical outcome of patients.

To assess the inherent potential for detecting mild to moderate reductions in glomerular filtration rate, this study determined the biological variability of serum cystatin C and creatinine in 12 healthy subjects. After accounting for analytical variation, interindividual variance accounted for 93% and intraindividual variance accounted for 7% of serum creatinine biological variation. As such, to lie outside the assay reference interval, some subjects must exceed 13 SD from their usual mean value, whereas in others, a change of only 2 SD would be sufficient. For cystatin C, interindividual variation explained 25% and intraindividual variance explained 75% of biological variability. Therefore, the upper limit of the population reference interval for cystatin C is seldom more than 3-4 SD from the mean value of any healthy individual. The critical difference for sequential values significant at P < or = 0.05 was calculated as 37% for serum cystatin C and 14% for serum creatinine. We conclude that cystatin C is potentially a better marker for detecting impaired renal function than serum creatinine, but serum creatinine is probably still the better marker for detecting temporal changes of renal function in individuals with established renal disease.

The measurement of glycated haemoglobin and serum fructosamine to assess the recent glycaemic control of diabetic patients has become well established. Likewise, the monitoring of blood glucose using glucose test strips and meters has become popular in both the community and in the hospital inpatient environment. However, despite improvements in the methods of analysis, clinically inaccurate assessments of glycaemia can still occur. Specific problems such as the lack of standardization in assays are in the process of being resolved, but inherent difficulties associated with these measures remain. Clinicians should be aware that these tests still need to be interpreted in conjunction with clinical prudence.

To investigate the discrepancy in the assessment of glycaemic control using glycated haemoglobin (HbA1C) and glycated proteins (fructosamine), the effect of age on these variables was measured in non-diabetic individuals. In 232 non-diabetics, there was a linear relationship between HbA1C and age (r = 0.49, p < 0.0001). Mean HbA1C rose from 3.82% to 4.44% between the ages of 20 and 70. Consequently, when Type 2 diabetic patient samples (n = 128, median age 63 years) were classified according to European guidelines into good or poor glycaemic control using both an age-matched (n = 101) and a younger (n = 108, median age 37 years) non-diabetic reference population, fewer patients were in good control (14% vs. 25%) and more in poor control (73% vs. 53%) when the younger reference population was used (both p < 0.05). In a subgroup of 126 non-diabetic subjects, HbA1C rose with age (r = 0.48), but serum fructosamine and fasting glucose did not (r = 0.07, r = 0.009, respectively, p = NS). Age-associated differences in non-diabetic HbA1C values may affect the assessment of glycaemic control in diabetic patients. It may also partly explain discrepancies found when comparing fructosamine with HbA1C as a measure of glucose control. Age-related HbA1C reference intervals may therefore be required for the treatment of patients and the accurate auditing of clinic performance.

A study was performed to assess the effect of varying degrees of sample haemolysis on the measurement of blood glucose by the Accutrend, Companion 2, ExacTech, Glucometer II, Glucometer 4, One Touch II, and Reflolux II blood glucose meters. Fresh venous blood was sonicated to induce complete haemolysis and then added in increasing proportions to homologous untreated blood to obtain nine samples with free haemoglobin concentrations up to 50 g l-1. The Accutrend meter showed the only significant (p < 0.05) linear relationship to degree of haemolysis (r = 0.988, p < 0.0001). For every 7% of red cells lysed, the Accutrend value increased by 15%. All other meters gave results which were within 15% of the non-haemolysed value. However, extreme (100%) haemolysis not only affected the Accutrend (glucose value 108% greater than reference) but also the ExacTech (+98%), the Glucometer II (-32%), and the Companion 2 (-41%). Thus, unwitting use of a haemolysed sample to measure whole blood glucose may, with the Accutrend in particular, lead to erroneous results.

A 6-week clinical evaluation of the One Touch II blood glucose meter was performed in a medical ward specializing in diabetic patients and compared to the existing Glucometer II system. Nursing staff trained in both methods performed 267 capillary blood glucose measurements using the two meters. A reference capillary blood glucose was performed simultaneously on 129 of these occasions using a Yellow Springs Instrument analyser. The One Touch II correlated better with the reference (r = 0.97) than the Glucometer II (r = 0.86) and was equally close or closer to the Yellow Springs Instrument on 82% of occasions. A greater proportion of samples measured by the One Touch II were within 15% of the reference value (83% vs 66%). There were also 17 cases where the One Touch II was able to give a measurement which was outwith the analytical range of the Glucometer II. A questionnaire revealed the nursing staff unanimously preferred using the One Touch II to their present meter. It is concluded that in a ward setting the One Touch II meter showed improved accuracy, usefulness in its wider analytical range, and a greater user preference in comparison to the Glucometer II.

Non-enzymatic glycation of low-density lipoprotein (LDL), and the predominance of small dense LDL particles may together contribute to the increased risk of atherosclerosis in diabetes. We aimed to establish whether the size of LDL particles is related to plasma triglyceride concentration, and to the extent of LDL glycation in type 2 diabetic patients. Sixteen men with type 2 diabetes and 16 age matched non-diabetic controls were studied. LDL size was measured by rapid density gradient ultracentrifugation, and LDL glycation by affinity chromatography. Modal LDL density correlated with plasma triglyceride concentrations in both diabetic and control groups (r = 0.86, P < 0.0001, and r = 0.76, P < 0.0008, respectively). There was no significant difference in these variables between the groups. LDL modal density showed no correlation with HbA1, serum fructosamine or plasma glucose in either group. In the diabetic group the degree of LDL glycation correlated with serum fructosamine (r = 0.74, P < 0.001), HbA1 (r = 0.65, P < 0.008), and with plasma glucose (r = 0.64, P < 0.008). Our results suggest that, in well controlled type 2 diabetic patients LDL size is independent of short-term glycaemic control but can be predicted by plasma triglyceride concentrations.

The Ames DCA 2000 is a benchtop analyser that measures HbA1c by an agglutination inhibition immunoassay using a monoclonal antibody. Laboratory and nursing staff measured HbA1c on-site in 78 patients with Type 1 diabetes at the outpatient clinic. Significant correlations were noted with both the Corning Glytrac total HbA1 assay (r = 0.89) and the Novoclone assay for HbA1c (r = 0.95). Mean within-assay CV was 1.6% and 3.0% at HbA1c of 5.4% and 13.0%, respectively, while between-assay CVs were 4.2% and 3.8%. These results are as good as our routine laboratory method based on the Corning HbA1 assay. Locally derived reference population data for HbA1c were produced and patients were assigned to categories of good, acceptable, and poor glycaemic control using conventional recommendations for Type 2 diabetes. There was poor agreement between the methods, with only 22% of patients achieving good/acceptable control using the DCA 2000, while 46% of patients had an HbA1 in this range. It appears that the convention for derivation of control limits for HbA1 does not hold for this HbA1c assay.

The aim of the study was to define clinical interpretation of the parallel measurements of serum fructosamine and HbA1 in diabetic patients. We studied 14 type 2 diabetic patients over a 16-wk period. The cross-sectional analysis showed no correlation between serum fructosamine and HbA1 concentrations during the period of changing glycaemic control. The correlations, however, became significant (P less than 0.05) at 12 (r = 0.60) and 16 (r = 0.87) weeks, i.e. after glycaemia had stabilised. Longitudinal analysis of individual patients' data over the 16-wk period showed a significant correlation between serum fructosamine and HbA1 (r = 0.55 to r = 0.94) which was present in 8 out of 14 patients. The changes in fructosamine concentration preceded those observed in HbA1. The ratio of fructosamine/HbA1 predicted the changes in HbA1 over the following month (r = 0.54, P less than 0.001). Thus, we demonstrated that the parallel measurement of fructosamine and HbA1 provides information on future trends in HbA1 concentration in diabetic patients.