A study examining the bias of albumin and albumin/creatinine ratio measurements in urine.

BACKGROUND: The objective of the study was to examine the bias of albumin and albumin/creatinine (ACR) measurements in urine. METHODS: Pools of normal human urine were augmented with purified human serum albumin to generate a series of 12 samples covering the clinical range of interest for the measurement of ACR. Albumin and creatinine concentrations in these samples were analyzed three times on each of 3 days by 24 accredited laboratories in Canada and the USA. Reference values (RV) for albumin measurements were assigned by a liquid chromatography-tandem mass spectrometry (LC-MS/MS) comparative method and gravimetrically. Ten random urine samples (check samples) were analyzed as singlets and albumin and ACR values reported according to the routine practices of each laboratory. RESULTS: Augmented urine pools were shown to be commutable. Gravimetrically assigned target values were corrected for the presence of endogenous albumin using the LC-MS/MS comparative method. There was excellent agreement between the RVs as assigned by these two methods. All laboratory medians demonstrated a negative bias for the measurement of albumin in urine over the concentration range examined. The magnitude of this bias tended to decrease with increasing albumin concentrations. At baseline, only 10% of the patient ACR values met a performance limit of RV ± 15%. This increased to 84% and 86% following post-analytical correction for albumin and creatinine calibration bias, respectively. CONCLUSIONS: International organizations should take a leading role in the standardization of albumin measurements in urine. In the interim, accuracy based urine quality control samples may be used by clinical laboratories for monitoring the accuracy of their urinary albumin measurements.

In utero copper treatment for Menkes disease associated with a severe ATP7A mutation.

Menkes disease is a lethal X-linked recessive neurodegenerative disorder of copper transport caused by mutations in ATP7A, which encodes a copper-transporting ATPase. Early postnatal treatment with copper injections often improves clinical outcomes in affected infants. While Menkes disease newborns appear normal neurologically, analyses of fetal tissues including placenta indicate abnormal copper distribution and suggest a prenatal onset of the metal transport defect. In an affected fetus whose parents found termination unacceptable and who understood the associated risks, we began in utero copper histidine treatment at 31.5 weeks gestational age. Copper histidine (900 µg per dose) was administered directly to the fetus by intramuscular injection (fetal quadriceps or gluteus) under ultrasound guidance. Percutaneous umbilical blood sampling enabled serial measurement of fetal copper and ceruloplasmin levels that were used to guide therapy over a four-week period. Fetal copper levels rose from 17 µg/dL prior to treatment to 45 µg/dL, and ceruloplasmin levels from 39 mg/L to 122 mg/L. After pulmonary maturity was confirmed biochemically, the baby was delivered at 35.5 weeks and daily copper histidine therapy (250 µg sc b.i.d.) was begun. Despite this very early intervention with copper, the infant showed hypotonia, developmental delay, and electroencephalographic abnormalities and died of respiratory failure at 5.5 months of age. The patient's ATP7A mutation (Q724H), which severely disrupted mRNA splicing, resulted in complete absence of ATP7A protein on Western blots. These investigations suggest that prenatally initiated copper replacement is inadequate to correct Menkes disease caused by severe loss-of-function mutations, and that postnatal ATP7A gene addition represents a rational approach in such circumstances.