Automated hematology

Background

The roots of automated hematology began with manual microscopy and laborious cell counting techniques. Using the Coulter principle and advances in laboratory methodology, automated hematology is now rapid and inexpensive with the complete blood cell count (CBC) being one of the most commonly ordered tests in medicine.

Content

This chapter describes the principles of automated hematology, including how we measure red blood cells (RBCs), white blood cells (WBCs), and platelets from whole blood. Next, the laboratory parameters that are derived from the measurement of RBCs, WBCs, and platelets are discussed. These include standard laboratory measurements of the CBC which help describe anemia, such as the mean corpuscular volume (MCV), to newer measurements which quantitate immature granulocytes (IG), immature platelets (IPF), and immature reticulocytes (IRF). The uses of recently introduced laboratory parameters in automated hematology are also described.

Manual methods

Hematology has undergone a revolution in technical methods, moving from manual counting to automated techniques. Red blood cells appeared as small red globules under the microscope first invented by van Leeuwenhoek over 300 years ago. The first blood count was developed by Karl Vierordt at the University of Tubingen in 1852, by which a fixed amount of blood was drawn into a capillary tube, spread on a slide, stained, and counted. Based on that principle the hemocytometer was developed, which uses a ruled counting chamber that contains a fixed amount of blood. The chamber’s etched perpendicular grids allow for easier enumeration of the cells. Despite its many drawbacks, the hemocytometer is easy to use and gives better consistency than other manual methods. In the early 1900s, the red blood cell (RBC) count, white blood cell (WBC) count, and the platelet (PLT) count were performed using such counting chambers. The hemoglobin concentration (Hb) was analyzed by the cyanmethemoglobin method, still considered the reference method today. The packed cell volume or hematocrit (HCT) was measured using high-speed centrifugation of a column of blood. This was performed either in a sealed microcapillary tube, also known as the spun hematocrit, or in a specifically designed Wintrobe tube. Finally, the WBC differential was enumerated by the type of WBC (e.g., granulocyte, lymphocyte, monocyte), examining 100 to 200 individual leukocytes on a stained peripheral blood smear. These early methods were laborious, and in particular for manual cell counts, imprecise.

Wintrobe calculations

Wintrobe developed a calculated set of RBC indices in 1932 that estimated the RBC size and hemoglobin content based on the RBC count, Hb, and HCT. The RBC indices included the mean corpuscular volume (MCV), which is the volume of the average erythrocyte measure in femtoliters (fL), the mean corpuscular hemoglobin (MCH), which is the hemoglobin content of the average erythrocyte measured in picograms (pg), and the mean corpuscular hemoglobin concentration (MCHC), which is the hemoglobin concentration within circulating erythrocytes as measured in grams of hemoglobin per deciliters of RBCs.

Coulter principle

Wallace and Joseph Coulter patented a device in the 1950s using electrical impedance to count red and white blood cells. In this method, blood is diluted in a solution that conducts current, such as isotonic saline. A vacuum then draws this solution containing cells through a small aperture located between two electrodes connected across a direct current potential. The resistance between the electrodes increases as each cell passes through this orifice ( Fig. 74.1 ). Cells are poor electrical conductors and generate a transient increase in impedance that results in an electrical pulse. Thus the number of pulses results in the cell count. The amplitude of the pulse is proportional to the volume of that cell. Thresholds could be set such that only particles within a specified range would be counted. For example, a cell count on a diluted whole blood sample could generate an RBC count given that there are approximately 1000 RBCs for every WBC, and platelets can be eliminated by selecting an appropriate lower pulse height threshold to exclude smaller cells. To yield a WBC count, lysing agents would be added to remove RBCs, and then WBCs could be counted using appropriate size thresholds.

The first semi automated systems used separate RBC and WBC manual dilutions. Added to the WBC dilution was a combination lysing agent and potassium ferricyanide solution; this allowed for the enumeration of WBCs and the measurement of hemoglobin, using the spectrophotometric measurement of cyanmethemoglobin. The RBC indices were calculated manually as shown above.

Multichannel instruments

The first multichannel automated hematology analyzers were in use in the 1960s. In these instruments, anticoagulated whole blood was aspirated into the analyzer and automatically aliquoted and diluted into RBC and WBC chambers for cell counting and sizing using impedance apertures, and into a cuvette for hemoglobin determination using the cyanmethemoglobin reaction (see below). The RBC indices were automatically calculated. To correct for coincidence, that is when two or more cells are so close together that they are counted as one cell, a coincidence correction was automatically performed on the raw impedance count. These instruments directly measured the RBC and the MCV, and thus the HCT can be calculated as follows:

By combining a flow device with a laser, Dittrich and Gohde developed a light based high-throughput cytometry instrument, making it possible to count large numbers of cells within a short time period. These multichannel instruments revolutionized the CBC, enabling fast turnaround time, high throughput and lower cost. The platelet count was not included in the first automated CBC, because the aperture resolution on these early instruments could not adequately separate RBCs from platelets.

By the 1970s, improvement of the cell counting apertures led to the automated enumeration of platelets in whole blood samples. Technical problems that needed to be overcome included the nonaxial passage of RBCs through the aperture and the recounting of RBCs that nicked the edge of the aperture; these problems resulted in a small proportion of RBCs resembling platelets in terms of cell size. This could lead to overestimates of the platelet count. These problems were corrected by better apertures that were smaller in diameter and longer in length, thus centralizing the flow of particles. Hydrodynamic focusing was developed in which the cells are forced to enter the aperture in a single file through the center of the aperture, ensuring more accurate cell sizing. At the exit of the aperture, “sweep flow” of the diluent quickly removes counted particles so that the RBCs cannot recirculate and nick the aperture and be erroneously recounted as platelet-size particles. Some analyzers have a von Behrens transducer which minimizes recirculation of cells by forcing the exit stream through a second aperture, and trapping cells behind a plate. Finally, pulse editing eliminates pulses from cells that travel a nonaxial path through the aperture for both platelets and RBCs. An important point to note is that poikilocytes, such as schistocytes or other non-normal erythrocyte shapes, often have pulses with aberrant shapes that are edited out of the MCV computation. In spite of these improvements, impedance methods have a limited linear range for the sizing of particles. This failure to accurately enumerate the smallest and largest erythrocytes can lead to an overestimation of microcytosis and an underestimation of macrocytosis, resulting in an MCHC with limited range.

George and Groner also implemented light scattering in flow cytometers in the 1970s, allowing for the discrimination of different types of leukocytes based on size and light scatter properties. Using light scatter measurements at two different angles allowed for the measurement of red cell size (low angle light scatter) and hemoglobin content (high angle scatter) after isovolumetric sphering of cells.

Manual counting methods

The hemocytometer is still used today in low-resource laboratories around the world because of its simplicity and low cost.

Automated counting methods

The principles of automated instruments include electrical current (impedance), light absorption, and fluorescence. Automated hematology analyzers primarily use impedance counting to enumerate RBCs ( Table 74.1 ). For example, in Beckman Coulter analyzers, RBCs are enumerated as particles within the RBC/PLT aperture with a cell volume ranging from approximately 36 to 360 fL. A plot of frequency versus size generates an RBC volume histogram that typically has a Gaussian distribution ( Fig. 74.2 ). Aberrant pulses are excluded using pulse editing software. The MCV and RBC distribution width or red cell distribution width (RDW) (coefficient of variation [CV] of the RBC volume distribution) are either derived from the RBC histogram or as direct measurements (see Table 74.1 ). Analysis of the RBC histogram can give clues to a patient’s medical condition, including: the presence of spherocytes or RBC fragments (left shoulder indicating RBCs with small size), a dimorphic curve indicating the presence of two populations of RBCs, macrocytes or reticulocytes (right shoulder indicating RBCs of large size), or RBC agglutinins (RBC population to extreme right indicating clumping of RBCs).

Light scatter also can be used to count RBCs with a predetermined size range (e.g., 30 to 180 fL) used to yield an RBC count. To accomplish this, RBCs must first be isovolumetrically sphered and fixed, and thus low and high forward angle light scatter yield RBC size and refractive index (directly proportional to Hb content), respectively. Thus an RBC cytogram can be constructed (RBC volume and Hb concentration) for each individual erythrocyte. Subsequent RBC volume and Hb histograms can be generated and the mean of the directly measured Hb content per RBC, mean of the directly measured Hb concentration, the cellular hemoglobin content distribution width, and the cellular hemoglobin concentration distribution width can be derived. Light scatter also allows better discrimination of WBCs from unlysed RBCs and nucleated red blood cells (nRBCs).

Fluorescence is utilized in modern hematology analyzers to stain nucleic acid RNA and DNA for the measurement of reticulocytes, reticulocyte hemoglobin, immature granulocytes, immature platelets fraction, nucleated RBCs, and WBC differential counts including blasts and abnormal lymphocytes. A few studies have demonstrated that fluorescence is more accurate and reliable than impedance or light scatter for platelet measurements.

Red blood cell parameters

The normal range of RBC parameters may vary by age, sex, and geographic location. For example, newborns have higher RBC counts, Hb, and HCT levels because of the relative hypoxic environment in the uterus. Males usually have higher counts than females. People living at higher elevations typically have higher counts than people at sea level because of the relatively lower oxygen concentration found at higher elevations.