Hydroxocobalamin and cyanocobalamin interference on co-oximetry based hemoglobin measurements

Abstract

Background

Hydroxocobalamin (OHCbl) and cyanocobalamin (CNCbl) are dark-red colored analogs of vitamin B-12. OHCbl is used as an alternate antidote for cyanide poisoning. Due to the strong red color, if uncorrected, these cobalamins interfere with hemoglobin measurements and can introduce errors in spectrophotometric assay on co-oximeters. The impact of cobalamins on commonly used co-oximetry systems was compared to evaluate the accuracy of hemoglobin measurements and to further assess the ability of the instruments to detect and flag cobalamin interference.

Methods

Blood samples spiked with varying concentrations of hydroxocobalamin or cyanocobalamin (0–2 g/l), were measured on 3 GEM Premier 4000 s, 3 IL 682 CO-Oximeter, 1 Radiometer ABL 735 and 1 Siemens Rapidpoint 405 analyzer. The effect of OHCbl or CNCbl interference on total hemoglobin and hemoglobin derivatives is evaluated using the measured difference between the control sample and the spiked sample.

Results

Spectral measurements, blood measurements and delta spectral data confirmed that hemoglobin measurement accuracy is affected by the presence of OHCbl or CNCbl and that the source of interference is from errors in measurement algorithms, and not due to cobalamin induced hemoglobin changes.

Conclusions

Among the 4 co-oximeters tested in this evaluation, GEM Premier 4000 and Rapidpoint 405 analyzer showed minimal impact for hydroxocobalamin concentrations around 0.5 g/l. Cyanocobalamin displayed similar interference effect on co-oximetry measurements as OHCbl. The error detection system in the GEM Premier 4000 appropriately detected and flagged interferences on sample measurements.

Introduction

Hydroxocobalamin has been gaining attention in Europe and US as the choice antidote for cyanide toxicity [1], [2], [3], [4], [5], [6]. It has been approved by the US Food and Drug Administration (FDA) in December 2006 for treatment of cyanide poisoning. Traditional cyanide antidote kit (CAK) contains nitrites and the presence of nitrites produce high concentrations of methemoglobin (20–30%). Cyanide toxicity together with carbon monoxide poisoning is common for burn-victims from fire accidents and treatment with CAK can significantly reduce the oxygen carrying capacity of blood in case of burn-victims due to high concentrations of carboxyhemoglobin and methemoglobin. For this reason, hydroxocobalamin is very attractive for treating cyanide poisoning cases without impacting the oxygen binding properties of blood [6]. Hydroxocobalamin (vitamin B12a) is a natural analog and a member of the vitamin B12 or cobalamin family. Cyanide binding to hydroxocobalamin in equi-molar quantities to form a non-toxic cyanocobalamin is the basis for the de-toxification mechanism of cyanide [6]. High concentrations of OHCbl and CNCbl are possible in blood samples from patients treated with hydroxocobalamin for cyanide toxicity [7].

Monitoring the oxygenation status of blood of the burn victims or other critically ill patients is very important for appropriate diagnosis and treatment. Co-oximetry is a routine and widely accepted laboratory technique for measuring total hemoglobin and hemoglobin fractions. Both hydroxocobalamin and cyanocobalamin absorb strongly in the visible spectral region between 475 and 650 nm and due to this strong absorbance within the wavelength range used for hemoglobin measurements by most co-oximeters, if uncorrected, these cobalamins can interfere with hemoglobin measurements.

Interference of hydroxocobalamin on co-oximetry based hemoglobin measurements was reported earlier by Lee et al. [8] on blood samples collected from New Zealand rabbits infused with different concentrations of hydroxocobalamin. Measurements on the control and (in-vivo and in-vitro) spiked samples tested on the Avoximeter 4000 co-oximeter (Avox Systems, San Antonio, TX) showed significant interference on hemoglobin measurements at OHCbl concentrations higher than 0.4 mmol/l (0.56 g/l). Interference on methemoglobin (metHb) was reported at OHCbl concentrations higher than 0.7 g/l during the evaluation of new critical care analyzer GEM Premier 4000 (Instrumentation laboratory, Lexington, MA) [9]. In both of these studies, the impact of OHCbl on hemoglobin measurements appear to be related to the presence of unaccounted absorbance in the mathematical modeling algorithms used in co-oximeters, rather than a real change in the concentrations of the hemoglobin fractions. However, in a recent study on the spectrophotometry of hydroxocobalamin and swine hemoglobin [10] the authors claim to identify a new methemoglobin variant from the hemoglobin and hydroxocobalamin mixture. This claim appears to contradict the main advantage of hydroxocobalamin over traditional cyanide antidote kit.

The impact of interference from hydroxocobalamin can vary significantly depending on the concentration, wavelength range or number of discrete wavelengths and mathematical models used in any given co-oximeter. At present, what is lacking is a comprehensive study of the interference produced by hydroxocobalamin on co-oximetry measurements in human blood, using current measurement technologies found in clinical laboratories. Also, for patients undergoing OHCbl treatment for cyanide poisoning, significant concentrations of cyanocobalamin can be present in their blood and the impact of cyanocobalamin on co-oximetry based hemoglobin measurements is also unknown in the literature.

In this study, 4 different commercial co-oximeters are compared to evaluate the magnitude of interference from hydroxocobalamin or cyanocobalamin on the measurement accuracy of total hemoglobin and hemoglobin fractions, and the ability of the analyzer to detect the interference and flag the sample results when needed. In addition to the hemoglobin measurement accuracy, hydroxocobalamin induced changes in hemoglobin fractions were investigated by comparing the spectral data collected on blood samples spiked with and without OHCbl.