Ultrafiltration discrepancies in recovery of myoglobin from urine.

We investigated the efficiency, accuracy, and reliability of the ultrafiltration/dipstick methodology commonly used to diagnose myoglobinuria. Twenty-five myoglobin-containing urine specimens were filtered by centrifugation for 15 min at 1500g through a Centricon-30 membrane filter. Both the original specimen and filtrate were assayed for myoglobin. The amount of myoglobin recovered subsequent to filtration varied from <1-38%. This poor and variable recovery was independent of sample matrix or precentrifugation of the specimens. This was most critical for urine specimens with myoglobin concentrations <60 000 microg/L. Fourteen of 18 such filtrates had concentrations <350 microg/L, a concentration below which a negative result would be obtained by using conventional dipstick methods. Thus, the use of this procedure has the potential to misdiagnose patients with myoglobin concentrations associated with increased risk of subsequent renal dysfunction, in particular when urine myoglobin concentrations are <60 000 microg/L.