Accuracy of measurement of dual-isotope Schilling test urine samples: a multicenter study [corrected]

As a component of our quality assurance program, this multicenter study was performed to characterize the magnitude and types of error present in measurement of typical dual-isotope Schilling test (DIST) urine samples.

A panel of three simulated DIST urine samples was formulated corresponding to diagnoses of normal excretion, malabsorption and pernicious anemia and was distributed to eight hospitals in our regional area (three novice and five experienced users). Count-rate data and urine volume measurements from each site were analyzed for accuracy against the predicted values and a carefully measured gold standard and were correlated with the methodology and equipment used.

Three of 24 results were uninterpretable due to an overly low ratio of intrinsic factor bound to free vitamin B12 excretion (B/F ratio), inconsistent with possible diagnoses. In 20 of 21 interpretable samples, results corresponded to the appropriate diagnoses, with typical values noted in 18 of the cases and slightly atypical yet diagnostic values seen in the remaining two cases. In only one sample did values correspond to an erroneous diagnosis (low normal or partial malabsorption rather than pernicious anemia). The four major discrepancies (test failure or misdiagnosis) were largely attributable to blunders and were limited to two of the three novice sites and to a single experienced site which had grossly inaccurate raw data (background greater than sample counts).

Quantitation of vitamin B12 excretion in DIST urine samples is a reliable method of evaluation when performed by reasonably experienced and competent clinical laboratories. Improved accuracy may be obtained by increasing the stochastic certainty of the count data and by more careful measurement of the sample and urine volumes.