Evaluation of the AVOXimeter: precision, long-term stability, linearity, and use without heparin

Because the AVOXimeter uses disposable cuvettes and makes its measurements directly in whole blood without first hemolyzing the sample, it does not need the care and maintenance that conventional co-oximeters require, it operates faster than conventional co-oximeters, and it is less expensive. Therefore, the objectives of the study were (1) to evaluate the precision and linearity of the AVOXimeter's measurements of total hemoglobin concentration and oxyhemoglobin saturation; (2) to assess its long-term stability and thus the required interval for recalibration; (3) to determine whether measurements can be made without anticoagulants; and (4) to assess the feasibility of storing blood samples in the disposable cuvettes.

Measurements made by the test instrument were compared with those of conventional co-oximeters or with standardized hemoglobin solutions. Blood samples were also collected with and without heparin to determine whether anticoagulation is necessary.

Our tests confirmed the specified precision of 0.3 g/dl for total hemoglobin and 0.5% for oxyhemoglobin. The results also showed that these measurements were linear when compared with a conventional co-oximeter, and they were consistent with the specified accuracy of 0.45 g/dl for total hemoglobin and 1% for oxyhemoglobin. Weekly checks with control solutions showed that the instrument holds its calibration for a year or more. Although treating syringes with heparin caused dilution errors, heparin did not affect the measurements when dilution was avoided. When blood samples were placed in disposable cuvettes and read repeatedly at 1-min intervals for 20 min, the readings drifted appreciably away from the original value. This drift occurred so slowly that readings taken at the first and second minute after the cuvette was filled were within 1 or 2% of the original reading.

In our experience the test instrument was simple and easy to operate. It met the specification for precision and accuracy, its measurements were highly linear, and it maintained a stable calibration for one year. If the cuvettes are filled as soon as blood is drawn, anticoagulation is unnecessary. However, the cuvettes should be read with 1 min of filling the cuvette.