Glycated haemoglobin threshold for dysglycaemia screening, and application to metabolic syndrome diagnosis in HIV-infected Africans

Abstract

BackgroundGlycated haemoglobin (HbA1c) test has been increasingly promoted as an alternative to fasting plasma glucose (FPG) or oral glucose tolerance test (OGTT) to diagnose dysglycaemia but its performance in HIV-infected Africans has yet to be established. This study aimed to assess the diagnostic accuracy of HbA1c for dysglycaemia including FPG-defined and OGTT-defined dysglycaemia, and OGTT-defined diabetes in HIV-infected Africans, and the effect of HbA1c-predicted dysglycaemia on Joint Interim Statement (JIS)-based prevalent metabolic syndrome (MS).MethodsA cross-sectional study included HIV-positive patients recruited across public healthcare facilities in the Western Cape. The recommended HbA1c cut-points were tested alongside the optimal cut-points obtained from receiver operating characteristic curve analyses, while the agreement between the MS criteria were assessed using kappa statistic.Results748 participants (157 men), median age 38 years, 93% on anti-retroviral drugs were included. The optimal HbA1c cut-points of 5.75% (39.3 mmol/mol) showed 54% sensitivity, 84% specificity for FPG-defined dysglycaemia, and 52% sensitivity, 85% specificity for OGTT-defined dysglycaemia. The HbA1c value of 5.85% (40.4 mmol/mol) (63% sensitivity, 99% specificity) was optimal for diabetes. The internationally advocated cut-point of 6.5% (48 mmol/mol) had 37% sensitivity and 99% specificity for diabetes, while HbA1c ≥5.7% (≥39 mmol/mol) yielded similar performance with the study-specific cut-point for any dysglycaemia. MS prevalence by the JIS criteria (28.2%) increased to 29.7% when using HbA1c ≥5.75% (≥39.3 mmol/mol) and to 32.9% with HbA1c ≥5.7% (≥39 mmol/mol); agreement between the original and modified criteria was generally good.ConclusionsThis study agrees with the internationally recommended HbA1c cut-point for detecting dysglycaemia, but not for diabetes in HIV-infected Africans. In line with previous studies in general African populations, our findings suggest that similar factors interfere with HbA1c values regardless of HIV infection status. Replacing FPG-based with HbA1c-predicted dysglycaemia in the JIS criteria to diagnose MS is feasible in HIV-infected Africans.

RSSDI-ESI clinical practice recommendations for the management of type 2 diabetes mellitus 2020

“…Both type 1 and type 2 Diabetes mellitus (DM) are diagnosed based on the plasma glucose criteria, either the fasting plasma glucose (FPG) levels or the 2-h plasma post-prandial glucose (2-h PPG) levels during a 75-g oral glucose tolerance test (OGTT), or the newer glycosylated haemoglobin (HbA1c) criteria which reflects the average plasma glucose concentration over the previous 8–12 weeks. [ 1 2 ] The International Expert Committee Report recommend a cut-point of ≥6.5% for HbA1c for diagnosing diabetes as an alternative to fasting plasma glucose (FPG ≥7.0 mmol/L). [ 3 ] HbA1c testing has some substantial advantages over FPG and OGTT, such as convenience, pre-analytical stability, and less day-to-day fluctuations due to stress and illness.…”

“…[ 3 ] Additionally, HbA1c has been recognized as marker to assess secondary vascular complications due to metabolic derailments in susceptible individuals. [ 2 4 5 ] However, given ethnic differences in sensitivity and specificity of HbA1c population-specific cut-offs might be necessary. [ 6 7 ] Moreover, measuring HbA1c is expensive as compared to FPG assessments and standardization of measurement techniques and laboratories are poorly practiced across the country.…”

“…Thus, other statins should probably be preferred to atorvastatin in patients with diabetes/MS. [ 574 ] In addition ADA recommend that, either high intensity or moderate intensity statin therapy should be used together with lifestyle intervention according to patient age and ASCVD risk factors[ 2 534 ] The details have been given in Annexure 6 . The Lipid Association of India expert consensus statement 2016 revealed that statin therapy is highly effective in lowering NHDL-C, LDL-C, apolipoprotein B, and remnant cholesterol, besides being remarkably safe.…”

“…The risk of developing infectious diseases due to diabetes is now being considered an important complication of diabetes. [ 1 2 ] Diabetes increases the risk of infection by two to three times in comparison to the non-diabetic population. The morbidity and mortality associated with infectious diseases such as influenza, pneumonia and hepatitis, which is usually preventable by appropriate vaccination, also appears to be very high in diabetes subjects.…”

Screening and diagnosis of gestational diabetes mellitus in South Africa: What we know so far

“…However, it is affected by factors such as ethnicity, anaemia, haemodilation or other blood disorders that hamper its accuracy as a diagnostic tool for GDM. [31][32][33] Therefore, although these tests are convenient, fast, simple, inexpensive and can be done at point of care, the results are inconsistent, with low sensitivity and specificity, and have been unsuccessful to date. Other novel strategies investigated in the SA population include the 'breakfast test'-a non-standardised glucose load administered to pregnant women -instead of the OGTT.…”