THU337 Fructosamine As A Predictor Of Diabetic Microvascular Disease In A Population With High Prevalence Of Red Cell Disorders

Disclosure: A.J. Buckley: None. T. Ashraf: None. M. Iqbal: None. Background: Fructosamine, glucose bound to circulating proteins, is used to evaluate glycaemic control where HbA1c may be unreliable due to variant haemoglobin or altered erythrocyte lifespan. In previous reports fructosamine predicted prevalent diabetic retinopathy, incident cardiovascular disease, and incident CKD3, in populations where it was highly correlated with HbA1c (r=0.80). Consequently in clinical practice fructosamine is converted to estimated HbA1c for interpretation. Use of fructosamine to estimate microvascular risk in populations where it correlates poorly with HbA1c remains unexplored. Aim: To provide evidence-based estimates for risk of incident retinopathy and early diabetic nephropathy according to a single fructosamine reading. Methods: The study population was 1571, predominantly Arab Emirati, individuals managed at Imperial College London Centre Abu Dhabi and Al Ain (60 prediabetes, 161 T1DM, 1350 T2DM), where a clinician suspected that HbA1c (Variant II ion-exchange HLPC, Bio-Rad), was inaccurate. Fructosamine was measured using the Cobas c501 module (Roche) and divided into strata of <250, 250-299, 300-349, 350-399 and ≥400 µmol/L for analysis. Routine yearly digital retinal screening and urinary albumin:creatinine ratio, and results of haematological investigations, were retrieved from electronic records. Results: Fructosamine was weakly associated with HbA1c (r2 = 0.353, p < 0.0001), with both slope and intercept for the relationship significantly varying according to coexisting haemoglobin variant, G6PD deficiency, iron deficiency, or renal impairment. Adjusted for age, systolic BP and smoking status, all fructosamine strata ≥ 300 µmol/L were significantly associated with stepwise increased hazard of new incident retinopathy for each stratum (300-349 µmol/L 2.1 (1.29-3.30) p=0.003; ≥400 µmol/L 3.3 (1.89-5.6) p <0.001), compared with individuals with fructosamine < 250 µmol/L, over mean follow-up of 2.4 ± 1.6 years. Adjusting for age and systolic BP, fructosamine strata ≥ 300 µmol/L were similarly associated with stepwise significant increased hazard of new incident microalbuminuria (300-349 µmol/L 1.40 (1.11-1.8) p=0.005; ≥ 400 µmol/L 2.3 (1.68-3.00) p <0.001). In 746 individuals with HbA1c <7.0% at inclusion, fructosamine strata ≥300 µmol/L remained significantly associated with increased hazard of retinopathy (300-349 µmol/L 2.10 (1.21-1.3.70) p=0.008; ≥ 400 µmol/L 3.6 (1.50-8.80) p=0.004) but not microalbuminuria. Conclusions: Fructosamine ≥ 300 µmol/L is associated with significantly increased hazard of microvascular complications of diabetes in a population with high incidence of haemoglobin variants and G6DP deficiency, even among individuals apparently achieving conventional HbA1c targets. Routine measurement of at least one fructosamine level to refine treatment targets is supported in such populations. Presentation: Thursday, June 15, 2023