Near-patient measurements of methemoglobin, oxygen saturation, and total hemoglobin: Evaluation of a new instrument for adult and neonatal intensive care

Objectives: a) To evaluate the performance of a compact, new instrument that uses disposable cuvettes to measure total hemoglobin concentration, oxygen content, and the relative concentrations of oxy- and methemoglobin in 50-μL blood samples; b) to determine whether the instrument can be used for near-patient assessment of methemoglobinemia; and c) to ascertain whether problems commonly encountered in neonatal blood samples affect the instrument's performance. Design: Prospective study, in which the test instrument was compared with a standard method. Samples of whole blood with and without bilirubin, fetal hemoglobin, and hemolysis were analyzed on the new (test) instrument and on a widely used cooximeter (OSM3 hemoximeter, Radiometer; reference instrument). Setting: In vitro analyses of blood samples in clinical and university laboratories. Measurements and Main Results: There was a close linear correlation between the methemoglobin measurements of the test instrument and those measurements of the reference instrument (slope = 0.989; r² = .989). The average difference in mean assay values between the reference instrument and the test instrument was -0.59%, i.e., <1% methemoglobin. Repeated measurements indicated the precision was 0.5% methemoglobin. Complete hemolysis of the sample reduced the methemoglobin reading by only 0.40%. Adding bilirubin (10 to 11 mg/dL [171 to 188.1 μmol/L]), increased the methemoglobin reading by 0.23%, increased the oxyhemoglobin reading by 0.45%, and increased total hemoglobin by 0.21 g/dL. Fetal hemoglobin also had minimal effects on the readings. Conclusions: The test instrument is fast and easy to operate. No sample preparation or pipetting is required. To operate the instrument, the user simply connects a syringe containing the blood sample to one of the disposable cuvettes, injects 50 μL of blood into the cuvette, and inserts the cuvette into the instrument. The test instrument automatically detects the presence of the cuvette, analyzes the sample, and displays the results in <10 secs. The findings in this study indicate that the test instrument has sufficient accuracy for near-patient testing in intensive care units. The errors introduced by hemolysis, fetal hemoglobin, and bilirubin were too small to be of clinical importance. Thus, the test instrument is essentially unaffected by complications commonly encountered in neonatal blood. The capacity of the test instrument to measure methemoglobin makes it particularly useful if inhaled nitric oxide therapy becomes a standard clinical practice.