Women with type 1 diabetes are at increased risk of preeclampsia, and microalbuminuria (urinary albumin excretion of 30–300 mg/24 hours) has proved to be a reliable marker for preeclampsia in these women. Because collecting a 24-hour urine sample is cumbersome and time-consuming, an alternative test would be welcome. One approach might be to estimate the albumin-to-creatinine ratio in random urine samples. Whether this is feasible was determined in 119 women with type 1 diabetes who were seen before 14 weeks gestation and were asked to provide 2 24-hour urine collections and 2 random samples during gestational weeks 7 and 22. Albumin was estimated by an enzyme-linked immunosorbent assay. Microalbuminuria was defined as an albumin to creatinine ratio of 2.5 mg/mmol or more, and nephropathy was defined as a ratio exceeding 25 mg/mmol (210 μg/mg creatinine).
Normoalbuminuria was documented in 103 women, whereas 7 had microalbuminuria and 9 had macroalbuminuria/nephropathy. There were no significant group differences in age, duration of diabetes, or HbA1c. Median urinary albumin excretion was 8, 66, and 677 mg/24 hours, respectively, in the 3 groups. The 24-hour urinary albumin excretion correlated positively with the albumin-to-creatinine ratio in random urine samples (P < .001). All but one of 16 women excreting more than 30 mg albumin in 24 hours had a ratio exceeding 2.5 mg/mmol, a sensitivity of 94%. However, all 9 women excreting more than 300 mg albumin in 24 hours had a ratio above 25.0 mg/mmol for a sensitivity of 100%. All 103 women with normoalbuminuria had a ratio less than 2.5 mg/mmol for a specificity of 100%. Positive and negative predictive values for a low ratio were 100% and 99%, respectively. The day-to-day coefficient of variation was 40% for 24-hour urine collections and 49% for random urine samples.
The investigators believe that measuring the albumin-to-creatinine ratio in 2 random urine samples is a very sensitive and specific means of screening for micro- and macroalbuminuria in pregnant women with type 1 diabetes. It is an acceptable alternative to collecting 24-hour urine samples.