Audit of Microalbumin Excretion in Children with Type I Diabetes

OBJECTIVE To investigate prevalence, persistence and clinical correlates of increased microalbumin excretion in random urine samples collected in a paediatric diabetes clinic. METHOD Random urine samples were collected annually in patients >10 years attending the diabetes clinic in the Royal Hospital for Sick Children, Edinburgh. Albumin excretion is expressed as albumin:creatinine ratio (ACR) and classified as normal (10 mg/mmol), or macroalbuminuria (>47 mg/mmol in females, >35 mg/mmol in males). We analyzed retrospectively results on 421 urine samples collected from 217 patients (109 males), of a median age of 12.3 years (94% 10-16 years) over 3 years. For each sample, the corresponding mean HbA1c over the previous year was calculated. RESULTS Prevalence of micro- and macro-albuminuria in individual samples was 1% and 0.5% respectively. ACR was equivocal in 10.1% and 4.7% in samples from females and males respectively (p=0.03). HbA1c showed borderline significant differences across ACR groups (p=0.06). Equivocal ACR excretion was associated with slightly higher mean HbA1c (9.5±1.3%) compared to normal albuminuria (9.0±1.1%, p3.5 mg/mmol. The 14-16 years age group patients were most likely to have ACR >3.5 mg/mmol (p=0.05). CONCLUSIONS Female sex and increasing age, but not HbA1c, were independently associated with increased ACR. A robust mechanism for collection of repeat early morning urine samples from patients with increased ACR in random urine samples, and follow-up of those patients who have persistently high microalbumin excretion are important. It is also important to confirm the usefulness of ACR measurements in random urine samples as a marker of incipent nephropathy.