Use of serum myoglobin assays for urine myoglobin measurements can cause false-negative results.

Several conditions, including rhabdomyolysis, trauma, and surgery, are associated with the release of large amounts of myoglobin into the circulation. Saturation of the salvage system of the kidney will produce myoglobinuria, a condition associated with acute renal failure (ARF). The exact mechanism is not known, but precipitation of myoglobin in the tubules and myoglobin-mediated formation of free radicals have been postulated (1 ). Early identification of severe myoglobinuria permits acute treatment, thereby avoiding ARF. Quantitative myoglobin assays have several pitfalls, however, including discrepancies in myoglobin recovery (2 ). The lack of quantitative information precludes differentiating mild myoglobinuria, which probably does not cause ARF, from severe myoglobinuria. Therefore, myoglobin immunoassays developed for serum applications have been adapted for use with urine samples (3–5 ). Prior reports have recommended alkalinizing urine specimens before storage. Because quantification of urine myoglobin might be valuable in the early evaluation of patients who might have rhabdomyolysis, a rapid in-house myoglobin assay is desirable. To this end, we evaluated 4 different commercial serum myoglobin assays for their potential to measure myoglobin in urine. These 4 assays can be run on the analyzers present in our laboratory [Integra 700 (Roche), Elecsys 1010 (Roche), Aeroset (Abbott Diagnostics), and BN ProSpec (Dade Behring)]. All 4 assays are based on immunogenic detection of myoglobin, but by different techniques, namely turbidimetry, nephelometry, and a heterogeneous immunoassay. The assays were evaluated for imprecision with control materials provided by the manufacturers. For all assays, interassay CVs were 5% for all tested concentrations (50 –1100 g/L). Although spiking of purified myoglobin into urine samples could be used to study myoglobin stability in urine, we preferred to use patient-derived material, thereby circumventing the intrinsic instability of purified myoglobin. We collected urine samples from patients with rhabdomyolysis, who were selected because of their dramatically increased serum activities of creatine kinase. The samples were alkalinized with sodium hydroxide solution to a pH 8. The measured myoglobin concentrations in these patients exceeded the linear ranges of the 4 assays. We therefore diluted the urine samples in myoglobin-free urine, which was adjusted to pH 4.5, pH 7, or pH 8.5. Fig. 1 shows the results for patient A. The original myoglobin concentration in the urine of patient A was approximately 30 000 g/L as measured with the Elecsys 1010 kit. This sample was diluted to a concentration of approximately 2000 g/L myoglobin. Acidifying the samples dramatically decreased the measured myoglobin concentrations on all 4 platforms (Fig. 1). The lower measured concentration for the Elecsys platform probably reflects variation in the calibration procedure, which was not investigated further. Similar results were obtained with 2 other patient samples, which showed a decrease in the measured myoglobin concentration of at least 5-fold after the samples were acidified. The decrease in measured myoglobin was also observed with urine samples acidified up to pH 6, although to a less pronounced extent (data not shown). These findings, which were obtained with assays that are currently widely available, are in line with older studies of myoglobin stability that used spiked myoglobin as well as urine from patients with rhabdomyolysis (4, 5 ). To our knowledge, only the BN ProSpec method has previously been investigated (5 ). We found that the measured myoglobin concentration decreased up to 50% within 1 week of storage of alkalinized urine at 4 °C, 20 °C, or 70 °C, with different kinetics