Evaluation of 24-second cyanide-containing and cyanide-free methods for whole blood hemoglobin on the Technicon H*1TM analyzer with normal and abnormal blood samples.

Automated methods, with and without cyanide (+CN and -CN), for whole blood hemoglobin (Hb) determination were evaluated on the Technicon H*1TM System. Both automated Hb methods were linear over the range 0-250 g/L (0-25 g/dL) and correlated well with the International Committee for Standardization of Hematology (ICSH) reference method and with the Coulter S+II. Both methods quantitatively converted whole blood containing up to 100% carboxyhemoglobin in less than 24 seconds to their respective end products. With respect to abnormal samples (sickle cell anemia, multiple myeloma, and hyperlipemia), both H*1 methods gave Hb results that were equivalent to the (postfiltration) ICSH method. For samples with white blood cell (WBC) counts less than 36 X 10(9)/L, the +CN method was equivalent to the (postfiltration) ICSH method, whereas for WBC counts greater than 20 X 10(9)/L, the -CN method showed acceptable recovery of the mean but unacceptable imprecision. For WBC counts of 36-164 X 10(9)/L, the +CN method yielded acceptable Hb recovery with unacceptable imprecision. Hyperlipemia, resulting from addition of Intralipid directly to the blood samples, caused large errors in both H*1 methods.

Flow cytometric determination of basophils in whole blood with n-propyl Astra Blue iodide.

Within the past year, it has become apparent, in connection with its use on automatic flow cytometers, that the quality of commercially available Alcian Blue has significantly declined. A homologous series of alkylated (C1-C7) Astra Blue quaternary ammonium halides was prepared, characterized, and evaluated for the detection of basophils in whole blood. On the Technicon H6000 flow cytometer, the resolution of the basophil cluster from the main population of unstained white blood cells was found to depend on the chain length of the quaternizing alkyl group. Optimal basophil resolution was observed for the n-propyl derivative. Correlation of the new method vs Alcian Blue as the reference on the H6000 was expressed as follows: %Baso (Astra Blue) = 0.89% Baso (Alcian Blue) + 0.12% for 180 fresh whole blood samples. Within-run precision at a basophil differential count of 0.73% was characterized by SD = 0.11, identical to that obtained for Alcian Blue. Aqueous solutions of n-propyl Astra Blue iodide, in contrast to Alcian Blue, are thermally stable. Heating the reagent for 1 h at 100 degrees C did not alter solubility or cytochemical behavior. In contrast, parallel treatment of Alcian Blue yielded insoluble material by hydrolysis of the isothiouronium groups. The reagent for basophil detection comprises n-propyl Astra Blue iodide, lanthanum chloride, sodium chloride, Tween 20, and cetylpyridinium chloride. The Astra Blue derivatives were characterized by uv-vis, ir, percentage halide, paper chromatography, and 13C NMR.

Determination of hematocrit based on diffusion of an inert molecular probe from agarose gels into whole blood.

Whole blood hematocrit was determined by an approach which depends on the diffusion of an inert probe, to which red blood cells are impermeable, from a small agarose gel into a stirred, much larger blood sample. Blood cells influence the diffusion rate of the probe by, on the average, physically blocking a fraction of the gel surface. The blocking effect increases with the hematocrit. Cyanocobalamin (B-12) was found to be a suitable probe because it did not penetrate, bind to, or lyse blood cells and was not bound by plasma solutes. The loss of B-12 from gels in contact with blood was monitored by determination of the absorbance change at 540 nm of gels which had been quickly rinsed. The visible spectrum of B-12 in agarose gels was identical to the spectrum in water. Beer's Law was obeyed in 1-mm thick agarose gels over a concentration range of 0.1-0.8 mM. Based on the results from 48 blood samples covering the hematocrit range 25-69, a least-squares line was generated with a slope, -3.46 X 10(-3) delta A/hematocrit unit, a Y intercept of 0.295, and a correlation coefficient of 0.971. The precision of the technique was +/- 9.7%. The assay was insensitive to mean corpuscular volume and sample volume as long as the latter was 50-fold larger than the gel volume. The diffusion coefficient for B-12 in 1% agarose gels was found to be 1.4 +/- 0.2 X 10(-6) cm2 sec-1.