PART III
CHAPTER 15
Sharral Longanbach, Martha K. Miers
OUTLINE
General Principles of Automated Blood Cell Analysis
Electronic Impedance
Radiofrequency
Optical Scatter
Principal Instruments
Overview
Beckman Coulter Instrumentation
Sysmex Instrumentation
Abbott Instrumentation
Siemens Healthcare Diagnostics Instrumentation
Automated Reticulocyte Counting
Limitations and Interferences
Calibration
Instrument Limitations
Specimen Limitations
Clinical Utility of Automated Blood Cell Analysis
Objectives
After completion of this chapter, the reader will be able to:
Since the 1980s, automated blood cell analysis has virtually replaced manual hemoglobin, hematocrit, and cell counting, due to its greater accuracy and precision, with the possible exception of phase platelet counting in certain circumstances. Hematology analyzers are marketed by multiple instrument manufacturers. These analyzers typically provide the eight standard hematology parameters (complete blood count [CBC]), plus a three-part, five-part, or six-part differential leukocyte count in less than 1 minute on 200 μL or less of whole blood. Automation allows more efficient workload management and more timely diagnosis and treatment of disease.
General principles of automated blood cell analysis
Despite the number of hematology analyzers available from different manufacturers and their varying levels of sophistication and complexity, most rely on only two basic principles of operation: electronic impedance (resistance) and optical scatter. Electronic impedance, or low-voltage direct current (DC) resistance, was developed by Coulter in the 1950s1, 2 and is the most common methodology used. Radiofrequency (RF), or alternating current resistance, is a modification sometimes used in conjunction with DC electronic impedance. Technicon Instruments Corporation (Tarrytown, NY) introduced darkfield optical scanning in the 1960s, and Ortho Clinical Diagnostics, Inc. (Raritan, NJ), followed with a laser-based optical instrument in the 1970s.3 Optical scatter, using both laser and nonlaser light, is frequently employed in today’s hematology instrumentation.
Electronic impedance
The impedance principle of cell counting is based on the detection and measurement of changes in electrical resistance produced by cells as they traverse a small aperture. Cells suspended in an electrically conductive diluent such as saline are pulled through an aperture (orifice) in a glass tube. In the counting chamber, or transducer assembly, low-frequency electrical current is applied between an external electrode (suspended in the cell dilution) and an internal electrode (housed inside the aperture tube). Electrical resistance between the two electrodes, or impedance in the current, occurs as the cells pass through the sensing aperture, causing voltage pulses that are measurable (Figure 15-1).4, 5 Oscilloscope screens on some instruments display the pulses that are generated by the cells as they interrupt the current. The number of pulses is proportional to the number of cells counted. The height of the voltage pulse is directly proportional to the volume of the cell, which allows discrimination and counting of cells of specific volumes through the use of threshold circuits. Pulses are collected and sorted (channelized) according to their amplitude by pulse height analyzers. The data are plotted on a frequency distribution graph, or volume distribution histogram, with relative number on the y-axis and volume (channel number equivalent to a specific volume) on the x-axis. The histogram produced depicts the volume distribution of the cells counted. Figure 15-2illustrates the construction of a frequency distribution graph. Volume thresholds separate the cell populations on the histogram, and the count is the cells enumerated between the lower and upper set thresholds for each population. Volume distribution histograms may be used for the evaluation of one cell population or subgroups within a population.5 The use of proprietary lytic reagents to control shrinkage and lysis of specific cell types, as in the older Coulter S-Plus IV, STKR, and Sysmex E-5000 models, allows separation and quantitation of white blood cells (WBCs) into three populations (lymphocytes, mononuclear cells, and granulocytes) for the three-part differential on one volume distribution histogram.6-8
FIGURE 15-1 Coulter principle of cell counting. Source: (From Coulter Electronics: Coulter STKR product reference manual, PN 4235547E, Hialeah, FL, 1988, Coulter Electronics.)
FIGURE 15-2 Oscilloscope display and histogram showing the construction of a frequency distribution graph. Source: (Modified from Coulter Electronics: Significant advances in hematology: hematology education series, PN 4206115A, Hialeah, FL, 1983, Coulter Electronics.)
Several factors may affect volume measurements in impedance or volume displacement instruments. Aperture diameter is crucial, and the red blood cell (RBC)/platelet aperture is smaller than the WBC aperture to increase platelet counting sensitivity. On earlier systems, protein buildup occurred, decreasing the diameter of the orifice, slowing the flow of cells, and increasing their relative electrical resistance. Protein buildup results in lower cell counts, which result in falsely elevated cell volumes. Impedance instruments once required frequent manual aperture cleaning, but current instruments incorporate burn circuits or other internal cleaning systems to prevent or slow protein buildup. Carryover of cells from one sample to the next also is minimized by these internal cleaning systems. Coincident passage of more than one cell at a time through the orifice causes artificially large pulses, which results in falsely increased cell volumes and falsely decreased cell counts. This count reduction, or coincident passage loss, is statistically predictable (and mathematically correctable) because of its direct relationship to cell concentration and the effective volume of the aperture.7-9 Coincidence correction typically is completed by the analyzer computer before final printout of cell counts from the instrument. Other factors affecting pulse height include orientation of the cell in the center of the aperture and deformability of the RBC, which may be altered by decreased hemoglobin content.10, 11 Recirculation of cells back into the sensing zone creates erroneous pulses and falsely elevates cell counts. A backwash or sweep-flow mechanism prevents recirculation of cells back into the sensing zone, and anomalously shaped pulses are edited out electronically.6, 7, 9
The use of hydrodynamic focusing avoids many of the potential problems inherent in a rigid aperture system. The sample stream is surrounded by a sheath fluid as it passes through the central axis of the aperture. Laminar flow allows the central sample stream to narrow sufficiently to separate and align the cells into single file for passage through the sensing zone.12-14The outer sheath fluid minimizes protein buildup and plugs, eliminates recirculation of cells back into the sensing zone with generation of spurious pulses, and reduces pulse height irregularity because off-center cell passage is prevented and better resolution of the blood cells is obtained. Coincident passage loss also is reduced because blood cells line up one after another in the direction of the flow.15 Laminar flow and hydrodynamic focusing are discussed further in Chapter 32.
Radiofrequency
Low-voltage DC impedance, as described previously, may be used in conjunction with RF resistance, or resistance to a high-voltage electromagnetic current flowing between both electrodes simultaneously. Although the total volume of the cell is proportional to the change in DC, the cell interior density is proportional to pulse height or change in the RF signal.Conductivity, as measured by this high-frequency electromagnetic probe, is attenuated by nucleus-to-cytoplasm ratio, nuclear density, and cytoplasmic granulation. DC and RF voltage changes may be detected simultaneously and separated by two different pulse processing circuits.15, 16 Figure 15-3 illustrates the simultaneous use of DC and RF current.
FIGURE 15-3 Radiofrequency/direct current (RF/DC) detection method, showing simultaneous use of DC and RF in one measurement system on the Sysmex SE-9500. Source: (From TOA Medical Electronics Company: Sysmex SE-9500 operator’s manual [CN 461-2464-2], Kobe, Japan, 1997, TOA Medical Electronics Co.)
Two different cell properties, such as low-voltage DC impedance and RF resistance, can be plotted against each other to create a two-dimensional distribution cytogram or scatterplot(Figure 15-4). Such plots display the cell populations as clusters, with the number of dots in each cluster representing the concentration of that cell type. Computer cluster analysis can determine absolute counts for specific cell populations. The use of multiple methods by a given instrument for the determination of at least two cell properties allows the separation of WBCs into a five-part differential (neutrophils, lymphocytes, monocytes, eosinophils, and basophils). DC and RF detection are two methods used by the Sysmex analyzers to perform WBC differentials.15, 16
FIGURE 15-4 Illustration of cell volume measurement with direct current (DC) voltage change versus measurement of cell nuclear volume/complexity with change in the radiofrequency (RF) signal. The two measurements can be plotted against each other to form a two-dimensional distribution scatterplot. Source: (From TOA Medical Electronics Company: Sysmex SE-9500 operator’s manual [CN 461-2464-2], Kobe, Japan, 1997, TOA Medical Electronics Co.)
Optical scatter
Optical scatter may be used as the primary methodology or in combination with other methods. In optical scatter systems (flow cytometers), a hydrodynamically focused sample stream is directed through a quartz flow cell past a focused light source (Figure 32-3). The light source is generally a tungsten-halogen lamp or a helium-neon laser (light amplification by stimulatedemission of radiation). Laser light, termed monochromatic light because it is emitted at a single wavelength, differs from brightfield light in its intensity, its coherence (i.e., it travels in phase), and its low divergence or spread. These characteristics allow for the detection of interference in the laser beam and enable enumeration and differentiation of cell types.12, 17Optical scatter may be used to study RBCs, WBCs, and platelets.
As the cells pass through the sensing zone and interrupt the beam, light is scattered in all directions. Light scatter results from the interaction between the processes of absorption, diffraction (bending around corners or the surface of a cell), refraction (bending because of a change in speed), and reflection (backward scatter of rays caused by an obstruction).18 The detection of scattered rays and their conversion into electrical signals is accomplished by photodetectors (photodiodes and photomultiplier tubes) at specific angles. Lenses fitted withblocker bars to prevent nonscattered light from entering the detector are used to collect the scattered light. A series of filters and mirrors separate the varying wavelengths and present them to the photodetectors. Photodiodes convert light photons to electronic signals proportional in magnitude to the amount of light collected. Photomultiplier tubes are used to collect the weaker signals produced at a 90-degree angle and multiply the photoelectrons into stronger, useful signals. Analogue-to-digital converters change the electronic pulses to digital signals for computer analysis.12, 17
Forward-angle light scatter (0 degrees) correlates with cell volume, primarily because of diffraction of light. Orthogonal light scatter (90 degrees), or side scatter, results from refraction and reflection of light from larger structures inside the cell and correlates with degree of internal complexity. Forward low-angle scatter (2 to 3 degrees) and forward high-angle scatter (5 to 15 degrees) also correlate with cell volume and refractive index or with internal complexity.17, 19 Differential scatter is the combination of this low-angle and high-angle forward light scatter and is primarily used on Siemens systems for cellular analysis. The angles of light scatter measured by the different flow cytometers are manufacturer and method specific.
Scatter properties at different angles may be plotted against each other to generate two-dimensional cytograms or scatterplots, as on the Abbott CELL-DYN instruments.20, 21 Optical scatter may also be plotted against absorption, as on the Siemens systems,22, 23 or against volume, as on the larger Beckman Coulter systems.9 Computer cluster analysis of the cytograms may yield quantitative and qualitative information.
Principal instruments
Overview
Hematology blood cell analyzers are produced by multiple manufacturers, including, but not limited to, Abbott Laboratories (Abbott Park, IL);24 HORIBA Medical (Irvine, CA);25 Siemens Healthcare Diagnostics, Inc. (Deerfield, IL);26 Beckman Coulter, Inc. (Brea, CA);27 and Sysmex Corporation (Kobe, Japan).28 The following discussion is limited to instrumentation produced by four of these suppliers. Emphasis is not placed on sample size or handling, speed, level of automation, or comparison of instruments or manufacturers. Likewise, technology continues to improve, and the newest (or most recent) models produced by a manufacturer may not be mentioned. Instead, a detailed description of primary methods used by these manufacturers is given to show the application of, and clarify further, the principles presented earlier and to enable the medical laboratory scientist or technician to interpret patient data, including instrument-generated histograms and cytograms. Table 15-1 summarizes methods used for the hemogram, reticulocyte, nucleated red blood cell, and WBC differential count determination on four major hematology instruments.
TABLE 15-1
Methods for Hemogram, Reticulocyte, Nucleated RBC, and WBC Differential Counts on Four Major Hematology Instruments
* Instruments auto-correct the WBC count for the presence of nucleated RBCs.
CV, Coefficient of variation; DC, direct current; HCT, hematocrit; HGB, hemoglobin; MCH, mean cell hemoglobin; MCHC, mean cell hemoglobin concentration; MCV, mean cell volume; NRBC, nucleated red blood cell count; RBC, red blood cell (or count); RDW, RBC distribution width; SD, standard deviation; VCS, volume, conductivity, scatter; WBC, white blood cell (or count).
Hematology analyzers have some common basic components, including hydraulics, pneumatics, and electrical systems. The hydraulics system includes an aspirating unit, dispensers, diluters, mixing chambers, aperture baths or flow cells or both, and a hemoglobinometer. The pneumatics system generates the vacuums and pressures required for operating the valves and moving the sample through the hydraulics system. The electrical system controls operational sequences of the total system and includes electronic analyzers and computing circuitry for processing the data generated. Some older-model instruments have oscilloscope screens that display the electrical pulses in real time as the cells are counted. A data display unit receives information from the analyzer and prints results, histograms, or cytograms.
Specimen handling varies from instrument to instrument based on degree of automation, and systems range from discrete analyzers to walkaway systems with front-end load capability. Computer functions also vary, with the larger instruments having extensive microprocessor and data management capabilities. Computer software capabilities include automatic start-up and shutdown, with internal diagnostic self-checks and some maintenance; quality control, with automatic review of quality control data, calculations, graphs, moving averages, and storage of quality control files; patient data storage and retrieval, with δ checks (Chapter 5), critical value flagging, and automatic verification of patient results based on user-defined algorithms; host query with the laboratory or hospital information system to allow random access discrete testing capability; analysis of animal specimens; and even analysis of body fluids.
Beckman coulter instrumentation
Beckman Coulter, Inc., manufactures an extensive line of hematology analyzers, including the smaller Ac-T series that provide complete RBC, platelet, and WBC analysis with a five-part differential. The LH 780 system, part of the LH 700 series, provides a fully automated online reticulocyte analysis.27 The LH series also has the capability to perform CD4 and CD8 counts.29 Coulter instruments typically have two measurement channels in the hydraulics system for determining the hemogram data. The RBC and WBC counts and hemoglobin are considered to be measured directly. The aspirated whole-blood sample is divided into two aliquots, and each is mixed with an isotonic diluent. The first dilution is delivered to the RBC aperture chamber, and the second is delivered to the WBC aperture chamber. In the RBC chamber, RBCs and platelets are counted and discriminated by electrical impedance as the cells are pulled through each of three sensing apertures (50 μm in diameter, 60 μm in length). Particles 2 to 20 fL are counted as platelets, and particles greater than 36 fL are counted as RBCs. In the WBC chamber, a reagent to lyse RBCs and release hemoglobin is added before WBCs are counted simultaneously by impedance in each of three sensing apertures (100 μm in diameter, 75 μm in length). Alternatively, some models employ consecutive counts in the same RBC or WBC aperture. After counting cycles are completed, the WBC dilution is passed to the hemoglobinometer for determination of hemoglobin concentration (light transmittance read at a wavelength of 525 nm). Electrical pulses generated in the counting cycles are sent to the analyzer for editing, coincidence correction, and digital conversion. Two of the three counts obtained in the RBC and the WBC baths must match within specified limits for the counts to be accepted by the instrument.5, 9 This multiple counting procedure prevents data errors resulting from aperture obstructions or statistical outliers and allows for excellent reproducibility on the Beckman Coulter instruments.
Pulse height is measured and categorized by pulse height analyzers; 256 channels are used for WBC and RBC analysis, and 64 channels are used for platelet analysis. Volume-distribution histograms of WBC, RBC, and platelet populations are generated. The RBC mean cell volume (MCV) is the average volume of the RBCs taken from the volume distribution data. The hematocrit (HCT), mean cell hemoglobin (MCH), and mean cell hemoglobin concentration (MCHC) are calculated from measured and derived values. The RBC distribution width (RDW) is calculated directly from the histogram as the coefficient of variation (CV) of the RBC volume distribution, with a reference interval of 11.5% to 14.5%.5 The RDW is an index of anisocytosis, but it may be falsely skewed because it reflects the ratio of the standard deviation (SD) to MCV. That is, an RBC distribution histogram with normal divergence but a decreased MCV may imply a high RDW, falsely indicating increased anisocytosis. MCV and RDW are used by the instrument to flag possible anisocytosis, microcytosis, and macrocytosis.9
Platelets are counted within the range of 2 to 20 fL, and a volume-distribution histogram is constructed. If the platelet volume distribution meets specified criteria, a statistical least-squares method is applied to the raw data to fit the data to a log-normal curve. The curve is extrapolated from 0 to 70 fL, and the final count is derived from this extended curve. This fitting procedure eliminates interference from particles in the noise region, such as debris, and in the larger region, such as small RBCs. The mean platelet volume (MPV), analogous to the RBC MCV, also is derived from the platelet histogram. The reference interval for the MPV is about 6.8 to 10.2 fL. The MPV increases slightly with storage of the specimen in ethylenediaminetetraacetic acid (EDTA).5
Many older-model Beckman Coulter instruments, such as the STKR, and the newer, smaller models, such as the Ac-T series, provide three-part leukocyte subpopulation analysis, which differentiates WBCs into lymphocytes, mononuclear cells, and granulocytes. In the WBC channel, a special lysing reagent causes differential shrinkage of the leukocytes, which allows the different cells to be counted and volumetrically sized based on their impedance. A WBC histogram is constructed from the channelized data. Particles between approximately 35 and 90 fL are considered lymphocytes; particles between 90 and 160 fL are considered mononuclears (monocytes, blasts, immature granulocytes, and reactive lymphocytes); and particles between 160 and 450 fL are considered granulocytes. This allows the calculation of relative and absolute numbers for these three populations (Figure 15-5).6 Proprietary computerized algorithms further allow flagging for increased eosinophils or basophils or both and interpretation of the histogram differential, including flagging for abnormal cells, such as reactive lymphocytes and blasts.7 When cell populations overlap or a distinct separation of populations does not exist, a region alarm (R flag) may be triggered that indicates the area of interference on the volume-distribution histogram. An R1 flag represents excess signals at the lower threshold region of the WBC histogram and a questionable WBC count. This interference is visualized as a high takeoff of the curve and may indicate the presence of nucleated RBCs, clumped platelets, unlysed RBCs, or electronic noise.6, 7
FIGURE 15-5 Printouts from the Coulter STKR showing the interpretive differential. A, Note the three distinct white blood cell (WBC) populations, Gaussian or normal distribution of red blood cells (RBCs), and right-skewed or log-normal distribution of platelets. B, Note the left shift in the WBC histogram with possible interference at the lower threshold region. R2 flag indicates interference and loss of valley owing to overlap or insufficient separation between the lymphocyte and mononuclear populations at the 90-fL region. RM flag indicates interference at more than one region. Eosinophil data have been suppressed. Also note the abnormal platelet volume distribution with a low platelet count. Manual 200-cell differential counts on the same samples: A, 52.5% neutrophils (47% segmented neutrophils, 5.5% bands), 41.5% lymphocytes, 4.0% monocytes, 1% basophils, 0.5% metamyelocytes, 0.5% reactive lymphocytes; B, 51% neutrophils (23% segmented neutrophils, 28% bands), 12% lymphocytes, 9.5% monocytes, 1% metamyelocytes, 1.5% myelocytes, 25% reactive lymphocytes, and 17 nucleated RBCs/100 WBCs.
More recent Beckman Coulter instruments, the LH 700 Series and UniCel DxH series, generate hemogram data (including the WBC count) as before but use Coulter’s proprietary VCS (volume, conductivity, scatter) technology in a separate channel to evaluate WBCs for the determination of a five-part differential. The VCS technology includes the volumetric sizing of cells by impedance, conductivity measurements of cells, and laser light scatter, all performed simultaneously for each cell. After RBCs are lysed and WBCs are treated with a stabilizing reagent to maintain them in a near-native state, a hydrodynamically focused sample stream is directed through the flow cell past the sensing zone. Low-frequency DC measures cell volume, whereas a high-frequency electromagnetic probe measures conductivity, an indicator of cellular internal content. The conductivity signal is corrected for cellular volume, which yields a unique measurement called opacity. Each cell also is scanned with monochromatic laser light that reveals information about the cell surface, such as structure, shape, and reflectivity. Beckman Coulter’s unique rotated light scatter detection method, which covers a 10-degree to 70-degree range, allows for separation of cells with similar volume but different scatter characteristics.27 Beckman Coulter’s newest analyzer, the UniCel DxH 800, uses volume and conductivity as well as five additional parameters: axial light loss (AL2), low-angle light scatter (LALS), median-angle light scatter (MALS), lower median–angle light scatter (LMALS), and upper median–angle light scatter (UMALS).30, 31 Using the data collected by the parameters listed above, the instrument applies data transformation, the process by which populations of cells are separated, allowing the determination of major populations as well as the enhancement of subpopulations of cells. Once those populations are established, a technique called the watershed concept searches for those populations and aids in determining counts as well as flagging based on all the populations found for that sample.30, 31
This combination of technologies provides a three-dimensional plot or cytograph of the WBC populations, which are separated by computer cluster analysis. Two-dimensional scatterplots of the measurements represent different views of the cytograph. The scatterplot of volume (y-axis) versus light scatter (x-axis) shows clear separation of lymphocytes, monocytes, neutrophils, and eosinophils. Basophils are hidden behind the lymphocytes but are separated by conductivity owing to their cytoplasmic granulation. Single-parameter histograms of volume, conductivity, and light scatter also are available.9
Two types of WBC flags (alarms or indicators of abnormality) are generated on all hematology analyzers that provide a WBC differential count: (1) user defined, primarily set for distributional abnormalities, such as eosinophilia or lymphocytopenia (based on absolute eosinophil or lymphocyte counts); and (2) instrument specific, primarily suspect flags for morphologic abnormalities. For distributional flags, the user establishes reference intervals and programs the instrument to flag each parameter as high or low. Suspect flags indicating the possible presence of abnormal cells are triggered when cell populations fall outside expected regions or when specific statistical limitations are exceeded. Instrument-specific suspect flags on the Coulter UniCel DxH 800 system and LH 700 series include immature granulocytes/bands, blasts, variant lymphocytes, nucleated RBCs, and platelet clumps. The UniCel DxH 800 also utilizes the International Society for Laboratory Hematology (ISLH) consensus rules in addition to the user defined and system defined flags for complete data analysis.32, 33 In addition to the flags listed above, inadequate separation of cell populations may disallow reporting of differential results by the instrument and may elicit a review slide message.9, 33, 34
The UniCel DxH 800 system utilizes VCS as well as digital signal processing from five light scatter angles for clear cellular resolution. On the LH 700 series, Coulter utilizes anIntelliKinetics application. This application is used to ensure consistency with the kinetic reactions. It provides the instrument the best signals for analysis independent of laboratory environment variations. Compared with earlier models Coulter IntelliKinetics provides better separation of cell populations for WBCs and reticulocytes, which enables better analysis by the system algorithms.33, 34
The UniCel DxH 800 also includes the number of nucleated RBCs as part of the standard CBC report. They are identified, counted, and subtracted from the white blood cell count using volume, conductivity, and the same five light scatter measurement described above. The AL2 measurement (which reflects the amount of light absorbed as it passes through the flow cell) initially separates the nucleated RBCs from the WBCs. Algorithms are applied using the scatter from the other angles to electronically separate and count the nucleated RBCs. Two scatterplots display the nucleated RBC data by plotting axial light loss (AL2) on the x-axis against low-angle light scatter (RLALS) and upper-median angle light scatter (RUMALS) on y-axis.
Figure 15-6 represents a standard patient printout from the Beckman Coulter UniCel DxH 800.
FIGURE 15-6 Coulter UniCel DxH 800. The DxH 800 printout displays the CBC, DIFF and reticulocyte data for the same patient in Figures 15-7, 15-9, and 15-12. A, CBC data; B, Differential with the nucleated red blood cells (NRBCs); C, Reticulocyte data, including the IRF (immature reticulocyte fraction); D, Impedance histograms for the WBC, RBC, and PLT; E, Advanced two-dimensional optical scatterplots for WBCs, NRBCs, and reticulocytes; F, Suspect area in which any sample or system flags will display.
Sysmex instrumentation
Sysmex Corporation, formerly TOA Medical Electronics Company, Ltd., manufactures a full line of hematology analyzers that provide complete RBC, platelet, and WBC analysis with three-part differential; the larger XT-1800i (SF-3000 and SE-9000) that performs a CBC with five-part differential; and the XE series and the newest XN series that also provide a fully automated reticulocyte count.28, 33 The newest XN series is modular. The series is scalable, and multiple modules can be combined onto one platform. Each module contains the XN-CBC and XN-DIFF with other options available, including XN-BF, the body fluid application. Included standard on the CBC and DIFF modules are NRBCs, RET-He (reticulocyte hemogloblin), and IRF (immature reticulocyte fraction). The platelet analysis on the XN also utilizes a fluorescent count, in addition to the impedance count and optical count, called the PLT-F, performed by optical measurement.35-38 The PLT-F can be performed on each sample or set up as a reflex based on the laboratory’s PLT criteria. The method uses a fluorocell fluorescent dye (oxazine) combined with an extended PLT counting volume and time. The PLTs can be differentiated from other cells based on differences in intensity of the fluorescence combined with forward scattered light.35 The WBC, RBC, platelet counts, hemoglobin, and hematocrit are considered to be measured directly. Three hydraulic subsystems are used for determining the hemogram: the WBC channel, the RBC/platelet channel, and a separate hemoglobin channel. In the WBC and RBC transducer chambers, diluted WBC and RBC samples are aspirated through the different apertures and counted using the impedance (DC detection) method for counting and volumetrically sizing cells. Two unique features enhance the impedance technology: in the RBC/platelet channel, a sheathed stream with hydrodynamic focusing is used to direct cells through the aperture, which reduces coincident passage, particle volume distortion, and recirculation of blood cells around the aperture; and in the WBC and RBC/platelet channels, floating thresholds are used to discriminate each cell population.8, 15, 16
As cells pass through the apertures, signals are transmitted in sequence to the analogue circuit and particle volume distribution analysis circuits for conversion to cumulative cell volume distribution data. Particle volume distribution curves are constructed, and optimal position of the autodiscrimination level (i.e., threshold) is set by the microprocessor for each cell population. The lower platelet threshold is automatically adjusted in the 2- to 6-fL volume range, and the upper threshold is adjusted in the 12- to 30-fL range, based on particle volume distribution. Likewise, the RBC lower and upper thresholds may be set in the 25- to 75-fL and 200- to 250-fL volume ranges. This floating threshold circuitry allows for discrimination of cell populations on a sample-by-sample basis. Cell counts are based on pulses between the lower and upper autodiscriminator levels, with dilution ratio, volume counted, and coincident passage error accounted for in the final computer-generated numbers. In the RBC channel, the floating discriminator is particularly useful in separating platelets from small RBCs. The hematocrit also is determined from the RBC/platelet channel, based on the principle that the pulse height generated by the RBC is proportional to cell volume. The hematocrit is the RBC cumulative pulse height and is considered a true relative percentage volume of erythrocytes.8, 15 In the hemoglobin flow cell, hemoglobin is oxidized and binds to sodium lauryl sulfate (SLS) forming a stable SLS–hemoglobin complex, which is measured photometrically at 555 nm.16
The following indices are calculated in the microprocessor using directly measured or derived parameters: MCV, MCH, MCHC, RDW-SD, RDW-CV, MPV, and plateletcrit. RDW-SD is the RBC arithmetic distribution width measured at 20% of the height of the RBC curve, reported in femtoliters, with a reference interval of 37 to 54 fL. RDW-CV is the RDW reported as a CV. Plateletcrit is the platelet volume ratio, analogous to the hematocrit. MPV is calculated from the plateletcrit and platelet count just as erythrocyte MCV is calculated from the hematocrit and RBC count. The proportion of platelets greater than 12 fL in the total platelet count may be an indicator of possible platelet clumping, giant platelets, or cell fragments.8, 15, 16 The XE series has the capability to run the platelet counting in the optical mode, which eliminates common interferences found with impedance counting. In the optical mode, the, immature platelet fraction or IPF, can be measured to provide additional information concerning platelet kinetics in cases of thromobocytopenia.39
The SE-9000/9500 uses four detection chambers to analyze WBCs and obtain a five-part differential: the DIFF, IMI (immature myeloid information), EO, and BASO chambers. The high-end instrumentation such as the XE-series and the XN series has a six-part differential: neutrophils, lymphocytes, monocytes, eosinophils, basophils, and immature granulocytes. Every differential performed generates a percentage and absolute number for immature granulocytes, thus providing valuable information about the complete differential.28 In the DIFF detection chamber, RBCs are hemolyzed and WBCs are analyzed simultaneously by low-frequency DC and high-frequency current (DC/RF detection method). A scattergram of RF detection signals (y-axis) versus DC detection signals (x-axis) allows separation of the WBCs into lymphocytes, monocytes, and granulocytes. Floating discriminators determine the optimal separation between these populations. Granulocytes are analyzed further in the IMI detection chamber to determine immature myeloid information. RBCs are lysed, and WBCs other than immature granulocytes are selectively shrunk by temperature and chemically controlled reactions. Analysis of the treated sample using the DC/RF detection method allows separation of immature cells on the IMI scattergram. A similar differential shrinkage and lysis method is also used in the EO and BASO chambers. That is, eosinophils and basophils are counted by impedance (DC detection) in separate chambers in which the RBCs are lysed, and WBCs other than eosinophils or basophils are selectively shrunk by temperature and chemically controlled reactions. Eosinophils and basophils are subtracted from the granulocyte count derived from the DIFF scattergram analysis to determine the neutrophil count. User-defined distributional flags may be set, and instrument-specific suspect flags, similar to those described for the Beckman Coulter LH 700 series, are triggered for the possible presence of morphologic abnormalities.15, 16 A POSITIVE or NEGATIVE interpretive message is displayed.
In the XN-1000, fluorescent flow cytometry is used for the WBC count, WBC differential, and enumeration of nucleated RBCs. In the WDF channel, RBCs are lysed, WBC membranes are perforated, and the DNA and RNA in the WBCs are stained with a fluorescent dye. Plotting side scatter on the x-axis and side fluorescent light on the y-axis enables separation and enumeration of neutrophils, eosinophils, lymphocytes, monocytes, and immature granulocytes. In the WNR channel, the RBCs are lysed including nucleated RBCs, and WBC membranes are perforated. A fluorescent polymethine dye stains the nucleus and organelles of the WBCs with high fluorescence intensity and stains the released nuclei of the nucleated RBCs with low intensity. Plotting side fluorescent light on the x-axis and forward scatter on the y-axis enables separation and enumeration of the total WBC count, basophils, and nucleated RBCs. The WBC count is automatically corrected when nucleated RBCs are present in the sample. A WPC channel detects blasts and abnormal lymphocytes in a similar manner using a lysing agent and fluorescent dye and plotting side scatter on the x-axis and side fluorescent light on the y-axis. Figure 15-7 shows a patient report from the Sysmex XN-1000 analyzing the same patient specimen for which data are given in Figure 15-6.
FIGURE 15-7 Sysmex XN-1000. The XN-1000 printout displays the CBC, DIFF, and reticulocyte data for the same patient in Figures 15-6, 15-9, and 15-12. A, CBC data; note the “& F” to indicate the fluorescent platelet count result; B, Nucleated red blood cells (NRBCs), six-part differential, including the IG (immature granulocyte), reticulocyte count (RET), and immature reticulocyte fraction (IRF); C, Reticulocyte hemoglobin (RET-He), immature platelet fraction (IPF), and body fluid counts, if done; D, Two scatterplots, WDF (lymphocytes, monocytes, neutrophils, eosinophils, and immature granulocytes) and WNR (WBC count, basophils, and nucleated RBCs); E, Reticulocyte (RET) and platelet (PLT-F) scatterplots, and RBC and PLT impedance histograms; F, Sample-related flags are listed at the bottom of the printout, if any are generated.
Abbott instrumentation
Instruments offered by Abbott Laboratories include the smaller CELL-DYN Emerald, which provides complete RBC, platelet, and WBC analysis with three-part differential, and the larger CELL-DYN Sapphire and the midrange CELL-DYN Ruby, both of which provide a CBC with five-part differential and random fully automated reticulocyte analysis.33 The CELL-DYN 4000 system has three independent measurement channels for determining the hemogram and differential: an optical channel for WBC count and differential data, an impedance channel for RBC and platelet data, and a hemoglobin channel for hemoglobin determination.20, 21 The WBC, RBC, hemoglobin, and platelet parameters are considered to be measured directly. A 60- to 70-μm aperture is used in the RBC/platelet transducer assembly for counting and volumetrically sizing of RBCs and platelets by the electronic impedance method.
A unique von Behrens plate is located in the RBC/platelet counting chamber to minimize the effect of recirculating cells. Pulses are collected and sorted in 256 channels according to their amplitudes: particles between 1 and 35 fL are included in the initial platelet data, and particles greater than 35 fL are counted as RBCs. Floating thresholds are used to determine the best separation of the platelet population and to eliminate interference, such as noise, debris, or small RBCs, from the count. Coincident passage loss is corrected for in the final RBC and platelet counts. RBC pulse editing is applied before MCV derivation to compensate for aberrant pulses produced by nonaxial passage of RBCs through the aperture. The MCV is the average volume of the RBCs derived from RBC volume distribution data. Hemoglobin is measured directly using a modified hemiglobincyanide method that measures absorbance at 540 nm. Hematocrit, MCH, and MCHC are calculated from the directly measured or derived parameters. The RDW, equivalent to CV, is a relative value, derived from the RBC histogram by using the 20th and 80th percentiles. The platelet analysis is based on a two-dimensional optical platelet count using fluorescent technology, the same technology used for direct nucleated RBC counting by adding a red fluorescence to the sample to stain nucleated red cells.40 Further analysis of platelets and platelet aggregates can be performed by using an automated CD61 monoclonal antibody to generate an immunoplatelet count.41-43 Other indices available include MPV and plateletcrit.20, 21
The WBC count and differential are derived from the optical channel using CELL-DYN’s patented multiangle polarized scatter separation (MAPSS) technology with three-color fluorescent technology. A hydrodynamically focused sample stream is directed through a quartz flow cell past a focused light source, an argon ion laser. Scattered light is measured at multiple angles: 0-degree forward light scatter measurement is used for determination of cell volume, 90-degree orthogonal light scatter measurement is used for determination of cellular lobularity, 7-degree narrow-angle scatter measurement is used to correlate with cellular complexity, and 90-degree depolarized light scatter measurement is used for evaluation of cellular granularity. Orthogonal light scatter is split, with one portion directed to a 90-degree photomultiplier tube and the other portion directed through a polarizer to the 90-degree depolarized photomultiplier tube. Light that has changed polarization (depolarized) is the only light that can be detected by the 90-degree depolarized photomultiplier tube. Various combinations of these four measurements are used to differentiate and quantify the five major WBC subpopulations: neutrophils, lymphocytes, monocytes, eosinophils, and basophils.20, 40, 44 Figure 15-8illustrates CELL-DYN’s MAPSS technology.
FIGURE 15-8 A, Multiangle polarized scatter separation (MAPSS) technology. Cells are measured and characterized by plotting light scatter from four different angles. B, Mononuclear and polymorphonuclear scatter with MAPPS technology. It plots 10 degree scatter (complexity) on the x-axis and 90 degree scatter (lobularity) on the y-axis. The system uses algorithms to further separate the two populations, displaying mononuclear on the lower left and polymorphonuclear on the upper right. C, Separation and plotting of the polymorphonuclear cells into neutrophils and eosinophils based on MAPPS technology. It plots 90 degree scatter (lobularity) on the x-axis and 90 degree depolarized (90 D) scatter on the y-axis. The system uses algorithms to further separate the two populations of cells. D, Scatter of all WBC populations by MAPPS technology plotting 10 degree scatter (complexity) on the x-axis and 0 degree scatter (size or volume) on the y-axis. On the newer instruments, a 7-degree angle for complexity is now used instead of the 10-degree angle. The change reflects use of the midrange of the angle instead of the end range; however, it still provides the same information. Source: (From Abbott Laboratories: CELL-DYN 3700 system operator’s manual [914032C], Abbott Park, IL, 2000.)
The light scatter signals are converted into electrical signals, sorted into 256 channels on the basis of amplitude for each angle of light measured, and graphically presented as scatterplots. Scatter information from the different angles is plotted in various combinations: 90 degrees/7 degrees, or lobularity versus complexity; 0 degrees/7 degrees, volume versus complexity; and 90 degrees depolarized/90 degrees, granularity versus lobularity. Lobularity or 90-degree scatter (y-axis) plotted against complexity or 7-degree scatter (x-axis) yields separation of mononuclear and segmented (polymorphonuclear neutrophil) subpopulations. Basophils cluster with the mononuclears in this analysis, because the basophil granules dissolve in the sheath reagent, and the degranulated basophil is a less complex cell. Each cell in the two clusters is identified as a mononuclear or segmented neutrophil for further evaluation.
The mononuclear subpopulation is plotted on a 0-degree/7-degree scatterplot, with volume on the y-axis and complexity on the x-axis. Three populations (lymphocytes, monocytes, and basophils) are seen clearly on this display. Nucleated RBCs, unlysed RBCs, giant platelets, and platelet clumps fall below the lymphocyte cluster on this scatterplot and are excluded from the WBC count and differential. Information from the WBC impedance channel also is used in discriminating these particles.21
The segmented neutrophil subpopulation is plotted on a 90-degree depolarized/90-degree scatterplot, with granularity or 90-degree depolarized scatter on the y-axis and lobularity or 90-degree scatter on the x-axis. Because of the unique nature of eosinophil granules, eosinophils scatter more 90-degree depolarized light, which allows clear separation of eosinophils and neutrophils on this display. Dynamic thresholds are used for best separation of the different populations in the various scatterplots. Each cell type is identified with a distinct color, so that after all classifications are made and volume (0-degree scatter) is plotted on the y-axis against complexity (7-degree scatter) on the x-axis, each cell population can be visualized easily by the operator on the data terminal screen. Other scatterplots (90 degrees/0 degrees, 90 degrees depolarized/0 degrees, 90 degrees depolarized/7 degrees) are available and may be displayed at operator request. On earlier instruments, the 7-degree angle measurement for complexity was referred to as the 10-degree angle. The change reflects use of the midrange of the angle instead of the end range; however, it still provides the same information.45-46 As on the previously described instruments, user-defined distributional flags may be set, and instrument-specific suspect flags may alert the operator to the presence of abnormal cells.20 , 45 Figure 15-9 represents a patient printout from the CELL-DYN Sapphire analyzing the same patient specimen for which data are given in Figures 15-6 and 15-7.
FIGURE 15-9 CELL-DYN Sapphire. The Sapphire printout displays the CBC, DIFF, and reticulocyte data for the same patient in Figures 15-6, 15-7, and 15-12. A, CBC data; B, Differential count data. Note that the WBC is listed with the differential instead of the CBC data; C, Reticulocyte and nucleated RBC data displayed under the CBC data but before the PLT data; D, Two scattergrams for the differential; both 7-degree scatter (complexity) vs 0-degree scatter (size or volume) and 90-degree scatter (lobularity) vs 90-degree depolarized (90D) scatter (granularity) are plotted for the WBCs; two histograms are also plotted for the nucleated RBC and reticulocyte data; E, Impedance histogram and optical platelet scatterplot side by side; F, At the bottom where the interpretative report flags display if there are any for the sample.
Siemens healthcare diagnostics instrumentation
Siemens Healthcare Diagnostics Inc. manufactures the ADVIA 2120 and 2120i, the next generation of the ADVIA 120.26, 33, 47 Siemens has simplified the hydraulics and operations of the analyzer by replacing multiple complex hydraulic systems with a unified fluids circuit assembly, or Unifluidics technology. The ADVIA 2120, 2120i, and 120 provide a complete hemogram and WBC differential, while also providing a fully automated reticulocyte count.22, 23
Four independent measurement channels are used in determining the hemogram and differential: RBC/platelet channel, hemoglobin channel, and peroxidase (PEROX) and basophil-lobularity (BASO) channels for WBC and differential data. WBC, RBC, hemoglobin, and platelets are measured directly. Hemoglobin is determined using a modified cyanmethemoglobin method that measures absorbance in a colorimeter flow cuvette at approximately 546 nm. The RBC/platelet method uses flow cytometric light scattering measurements determined as cells, in a sheath-stream, pass through a flow cell by a laser optical assembly (laser diode light source). RBCs and platelets are isovolumetrically sphered before entering the flow cell to eliminate optical orientation noise. Laser light scattered at two different angular intervals—low angle (2 to 3 degrees), correlating with cell volume, and high angle (5 to 15 degrees), correlating with internal complexity (i.e., refractive index or hemoglobin concentration)—is measured simultaneously (Figure 15-10). This differential scatter technique, in combination with isovolumetric sphering, eliminates the adverse effect of variation in cellular hemoglobin concentration on the determination of RBC volume (as seen by differences in cellular deformability affecting the pulse height generated on impedance instruments).10, 48 The Mie theory of light scatter of dielectric spheres18 is applied to plot scatter-intensity signals from the two angles against each other for a cell-by-cell RBC volume (y-axis) versus hemoglobin concentration (x-axis) cytogram or RBC map (Figure 15-11).19
FIGURE 15-10 Differential scatter detection as used in the ADVIA 120. Forward high-angle scatter (5 to 15 degrees) and forward low-angle scatter (2 to 3 degrees) are detected for analysis of red and white blood cells. Source: (From Miles: Technicon H systems training guide, Tarrytown, NY, 1993, Miles.)
FIGURE 15-11 Cytograms or red blood cell (RBC) maps derived using the Mie theory of light scatter of dielectric spheres. A, Transformation between scatter angles (2 to 3 degrees and 5 to 15 degrees) and RBC volume (V) and hemoglobin concentration (HC). B, RBC map for a patient sample. Source: (From Groner W: New developments in flow cytochemistry technology. In Simson E, editor: Proceedings of Technicon H-1 hematology symposium, October 11, 1985, Tarrytown, NY, 1986, Technicon Instruments Corp, p. 5.)
Independent histograms of RBC volume and hemoglobin concentration also are plotted. On the ADVIA 2120 and 120 platelets are counted and volumetrically sized using a two-dimensional (low-angle and high-angle) platelet analysis, which allows better discrimination of platelets from interfering particles, such as RBC fragments and small RBCs.22 Larger platelets can be included in the platelet count.23, 47
Several parameters and indices are derived from the measurements described in the previous paragraph. MCV and MPV are the mean of the RBC volume histogram and the platelet volume histogram. Hematocrit, MCH, and MCHC are mathematically computed using RBC, hemoglobin, and MCV values. RDW is calculated as the CV of the RBC volume histogram, whereas hemoglobin distribution width (HDW), an analogous index, is calculated as the SD of the RBC hemoglobin concentration histogram. The reference interval for HDW is 2.2 to 3.2 g/dL. Cell hemoglobin concentration mean (CHCM), analogous to MCHC, is derived from cell-by-cell direct measures of hemoglobin concentration. Interferences with the hemoglobin colorimetric method, such as lipemia or icterus, affect the calculated MCHC but do not alter measured CHCM. CHCM generally is not reported as a patient result but is used by the instrument as an internal check for the MCHC and is available to the operator for calculating the cellular hemoglobin if interferences are present. Unique RBC flags derived from CHCM include hemoglobin concentration variance (HC VAR), hypochromia (HYPO), and hyperchromia (HYPER).22, 23
Siemens hematology analyzers determine WBC count and a six-part WBC differential (neutrophils, lymphocytes, monocytes, eosinophils, basophils, and large unstained cells [LUCs]) by cytochemistry and optical flow cytometry, using the PEROX and BASO channels. LUCs include reactive or variant lymphocytes and blasts.
Peroxidase (perox) channel
In the PEROX channel, RBCs are lysed, and WBCs are stained for their peroxidase activity. The following reaction is catalyzed by cellular peroxidase, which converts the substrate to a dark precipitate in peroxidase-containing cells (neutrophils, monocytes, and eosinophils):
A portion of the cell suspension is fed to a sheath-stream flow cell where a tungsten-halogen darkfield optics system is used to measure absorbance (proportional to the amount of peroxidase in each cell) and forward scatter (proportional to the volume of each cell). Absorbance is plotted on the x-axis of the cytogram, and scatter is plotted on the y-axis.22, 23 A total WBC count (WBC-PEROX) is obtained from the optical signals in this channel and is used as an internal check of the primary WBC count obtained in the basophil-lobularity channel (WBC-BASO). If significant interference occurs in the WBC-BASO count, the instrument substitutes the WBC-PEROX value.23
Computerized cluster analysis allows classification of the different cell populations, including abnormal clusters such as nucleated RBCs and platelet clumps. Nucleated RBCs are analyzed for every sample using four counting algorithms, which permits the system to choose the most accurate count based on internal rules and conditions. Neutrophils and eosinophils contain the most peroxidase and cluster to the right on the cytogram. Monocytes stain weakly and cluster in the midregion of the cytogram. Lymphocytes, basophils, and LUCs (including variant or reactive lymphs and blasts) contain no peroxidase and appear on the left of the cytogram, with LUCs appearing above the lymphocyte area. Basophils cluster with the small lymphocytes and require further analysis for classification.22, 23, 49
Basophil-lobularity (baso) channel
In the BASO channel, cells are treated with a reagent containing a nonionic surfactant in an acidic solution. Basophils are particularly resistant to lysis in this temperature-controlled reaction, whereas RBCs and platelets lyse and other leukocytes (nonbasophils) are stripped of their cytoplasm. Laser optics, using the same two-angle (2 to 3 degrees and 5 to 15 degrees) forward scattering system of the RBC/platelet channel, is used to analyze the treated cells. High-angle scatter (proportional to nuclear complexity) is plotted on the x-axis, and low-angle scatter (proportional to cell volume) is plotted on the y-axis. Cluster analysis allows for identification and quantification of the individual cellular populations. The intact basophils are identifiable by their large low-angle scatter. The remaining nuclei are classified as mononuclear, segmented, and blast cell nuclei based on their nuclear complexity (shape and cell density) and high-angle scatter.22, 23
Basophils fall above a horizontal threshold on the cytogram. The stripped nuclei fall below the basophils, with segmented cells to the right and mononuclear cells to the left along the x-axis. Blast cells uniquely cluster below the mononuclear cells. Lack of distinct separation between the segmented and mononuclear clusters indicates WBC immaturity or suspected left shift. As indicated earlier, this channel provides the primary WBC count, the WBC-BASO. Relative differential results (in percent) are computed by dividing absolute numbers of the different cell classifications by the total WBC count.22, 23
The nucleated RBC method is based on the physical characteristics of volume and density of the nucleated RBC nuclei. These characteristics allow counting in both WBC channels on the ADVIA 2120, and algorithms are applied to determine the absolute number and percentage of nucleated RBCs. Information from the PEROX and BASO channels is used to generate differential morphology flags indicating the possible presence of reactive lymphocytes, blasts, left shift, immature granulocytes, nucleated RBCs, or large platelets or platelet clumps.22, 23, 49 Figure 15-12 shows a patient printout from the ADVIA 2120i analyzing the same patient specimen for which data are given in Figures 15-6, 15-7, and 15-9.
FIGURE 15-12 ADVIA 2120i. The ADVIA 2120i printout displays the CBC, DIFF, and reticulocyte data for the same patient in Figures 15-6, 15-7, and 15-9. A, CBC data; B, Six-part differential, including large unstained cells (LUCs); nucleated RBCs; C, Reticulocyte information includes the CHr (cellular hemoglobin reticulocyte). D, Cytograms for the differential, both the perox and baso channels, on the right; E, Scattergram for the reticulocyte and RBC counts; F, Morphology flags and Sample/System Flags where flags are displayed.
Automated reticulocyte counting
Reticulocyte counting is the last of the manual cell-counting procedures to be automated and has been a primary focus of hematology analyzer advancement in recent years. The imprecision and inaccuracy in manual reticulocyte counting are due to multiple factors, including stain variability, slide distribution error, statistical sampling error, and interobserver error.50 All of these potential errors, with the possible exception of stain variability, are correctable with automated reticulocyte counting. Increasing the number of RBCs counted produces increased precision.51 This was evidenced in the 1993 College of American Pathologists pilot reticulocyte proficiency survey (Set RT-A, Sample RT-01) on which the CV for the reported manual results was 35% compared with 8.3% for results obtained using flow cytometry.52 Precision of automated methods has continued to improve. The manual reticulocyte results for one specimen in the 2000 Reticulocyte Survey Set RT/RT2-A showed a CV of 28.7%, whereas the CV was 2.8% for results obtained using one of the automated methods.53Automated reticulocyte analyzers may count 32,000 RBCs compared with 1000 cells in the routine manual procedure.54
Available automated reticulocyte analyzers include flow cytometry systems such as the FACS system from Becton, Dickinson and Company (Franklin Lakes, NJ) or the Coulter EPICS system; the Sysmex R-3500, R-500, XE-2100, XE-5000, and XN-series systems; the CELL-DYN 3500R, 3700, and 4000 systems; the Coulter LH 750 systems and the UniCel DxH800; and the Siemens ADVIA 2120, 2120i, and 120. All of these analyzers evaluate reticulocytes based on optical scatter or fluorescence after the RBCs are treated with fluorescent dyes or nucleic acid stains to stain residual RNA in the reticulocytes. Because neither the FACS nor EPICS system is generally available in the routine hematology laboratory, the discussion here is limited to the other analyzers.
The Sysmex R-3000/3500 is a stand-alone reticulocyte analyzer that uses auramine O, a supravital fluorescent dye, and measures forward scatter and side fluorescence as the cells, in a sheath-stream, pass through a flow cell by an argon laser. The signals are plotted on a scattergram with forward scatter intensity, which correlates with volume, plotted against fluorescence intensity, which is proportional to RNA content. Automatic discrimination separates the populations into mature RBCs and reticulocytes. The reticulocytes fall into low-fluorescence, middle-fluorescence, or high-fluorescence regions, with the less mature reticulocytes showing higher fluorescence. The immature reticulocyte fraction (IRF) is the sum of the middle-fluorescence and high-fluorescence ratios and indicates the ratio of immature reticulocytes to total reticulocytes in a sample. The XE-5000, the XT-2000i and the XN series also determines the reticulocyte count and IRF by measuring forward scatter and side fluorescence. They also have a parameter called RET-He (reticulocyte hemoglobin equivalent) that measures the hemoglobin content of the reticulocytes.55 It uses a proprietary polymethine dye to fluorescently stain the reticulocyte nucleic acids. This is similar to the reticulocyte hemoglobin content (CHr) parameter on the ADVIA 2120i (discussed below). Platelets, which also are counted, fall below a lower discriminator line.56 The Sysmex SE-9500/9000+RAM-1 module uses the same flow cytometry methodology for reticulocyte counting as the R-3500.16 Off-line sample preparation is not required. The smaller Sysmex R-500 uses flow cytometry with a semiconductor laser as the light source and polymethine supravital fluorescent dye to provide automated reticulocyte counts.28
The CELL-DYN 3500R performs reticulocyte analysis by measuring 10-degree and 90-degree scatter in the optical channel (MAPSS technology) after the cells have been isovolumetrically sphered to eliminate optical orientation noise. The RBCs are stained with the thiazine dye new methylene blue N in an off-line sample preparation before the sample is introduced to the instrument. The operator simply must change computer functions on the instrument before aspiration of the reticulocyte preparation.45 The CELL-DYN Sapphire also uses MAPSS technology but adds fluorescence detection to allow fully automated, random access reticulocyte testing.24, 46 The RBCs are stained with a proprietary membrane-permeable fluorescent dye (CD4K530) that binds stoichiometrically to nucleic acid and emits green light as the cells, in a sheath-stream, pass through a flow cell by an argon ion laser. Platelets and reticulocytes are separated based on intensity of green fluorescence (scatter measured at 7 degrees and 90 degrees), and the reticulocyte count along with the IRF is determined.46
Beckman Coulter also has incorporated reticulocyte methods into its primary cell-counting instruments: LH 700 series systems and the UniCel DxH800. The Coulter method uses a new methylene blue stain and the VCS technology described earlier. Volume is plotted against light scatter (DF 5 scatterplot) and against conductivity (DF 6 scatterplot), which correlates with opacity of the RBC. Stained reticulocytes show greater optical scatter and greater opacity than mature RBCs. Relative and absolute reticulocyte counts are reported, along with mean reticulocyte volume and maturation index or IRF.54
The Siemens ADVIA 2120, 2120i, and 120 systems enumerate reticulocytes in the same laser optics flow cell used in the RBC/platelet and BASO channels described earlier. The reticulocyte reagent isovolumetrically spheres the RBCs and stains the reticulocytes with oxazine 750, a nucleic acid–binding dye. Three detectors measure low-angle scatter (2 to 3 degrees), high-angle scatter (5 to 15 degrees), and absorbance simultaneously as the cells pass through the flow cell. Three cytograms are generated: high-angle scatter versus absorption, low-angle scatter versus high-angle scatter (Mie cytogram or RBC map), and volume versus hemoglobin concentration. The absorption cytogram allows separation and quantitation of reticulocytes, with additional subdivision into low-absorbing, medium-absorbing, and high-absorbing cells based on amount of staining. The sum of the medium-absorbing and high-absorbing cells reflects the IRF. Volume and hemoglobin concentration for each cell are derived from the RBC map by applying Mie scattering theory.26, 57 Unique reticulocyte indices (MCVr, CHCMr, RDWr, HDWr, CHr, and CHDWr) are provided. The CHr or reticulocyte hemoglobin content of each cell is calculated as the product of the cell volume and the cell hemoglobin concentration. A single-parameter histogram of CHr is constructed, with a corresponding distribution width (CHDWr) calculated.22, 23 These reticulocyte indices are not reported on the routine patient printout but are available to the operator. Figure 15-13 is a reticulocyte printout from an ADVIA 120, showing the cytograms and reticulocyte indices.
FIGURE 15-13 Composite cytograms obtained from the ADVIA 120. I, Normal reticulocyte count. II, High reticulocyte count with a large immature reticulocyte fraction (IRF). I-A and II-C, Reticulocyte scatter/absorption cytograms show high-angle (5- to 15-degree) scatter on the y-axis versus absorption on the x-axis, which allows separation of low-absorbing, medium-absorbing, and high-absorbing cells based on the amount of staining with nucleic acid–binding dye (oxazine 750). I-B and II-D, Reticulocyte scatter cytograms show low-angle (2- to 3-degree) scatter on the y-axis versus high-angle (5- to 15-degree) scatter on the x-axis (red blood cell [RBC] map). Because the RBCs are evaluated by the instrument on a cell-by-cell basis, unique reticulocyte indices can be derived.
Automation of reticulocyte counting has allowed increased precision and accuracy and has greatly expanded the analysis of immature RBCs, providing new parameters and indices that may be useful in the diagnosis and treatment of anemias. The IRF, first introduced to indicate immature reticulocytes, shows an early indication of erythropoiesis. The IRF and the absolute reticulocyte count can be used to distinguish types of anemias. Anemias with increased marrow erythropoiesis, such as hemolytic anemia, have a high total reticulocyte count and increased IRF, while chronic renal disease has decreased absolute count and an IRF indicating decreased marrow erythropoiesis.57, 58 An increased IRF and normal to decreased absolute reticulocyte count indicates an early response to therapy in nutritional anemias.58 Utilization of both tests is a reliable indicator of changes in erythropoietic activity and may prove to be a valuable therapeutic monitoring tool in patients.58 The reticulocyte maturity measurements also may be useful in evaluating bone marrow suppression during chemotherapy, monitoring hematopoietic regeneration after bone marrow or stem cell transplantation, monitoring renal transplant engraftment, and assessing efficacy of anemia therapy.57-61 The reticulocyte hemoglobin content, CHr (Advia) and Ret-He (Sysmex), provides an assessment of the availability of iron for erythropoiesis (Chapters 11 and 20). The additional reticulocyte indices derived on the ADVIA 2120 and 120 are valuable in following the response to erythropoietin therapy, and the CHr in particular has proved useful in the early detection and diagnosis of iron deficient erythropoiesis in children.61, 62 The National Kidney Foundation KDOQI (Kidney Disease Outcomes Quality Initiative) recommends the addition of the reticulocyte hemoglobin content to the CBC, in addition to the reticulocyte count and ferritin level to assess the iron status in patients with chronic kidney disease.63Widespread use of the new parameters may be limited by the availability of instrumentation.
Limitations and interferences
Implementing automation in the hematology laboratory requires critical evaluation of the instrument’s methods and limitations, and the performance goals for the individual laboratory. The Clinical and Laboratory Standards Institute (CLSI) has approved a standard for validation, verification, and quality assurance of automated hematology analyzers.64 This standard provides guidelines for instrument calibration and assessment of performance criteria, including accuracy, precision, linearity, sensitivity, and specificity. The clinical accuracy (sensitivity and specificity) of the methods should be such that the instrument appropriately identifies patients who have disease and patients who do not have disease.65 Quality control systems should reflect the laboratory’s established performance goals and provide a high level of assurance that the instrument is working within its specified limits.
Calibration
Calibration is crucial in defining the accuracy of the data produced (Chapter 5). Calibration, or the process of electronically correcting an instrument for analytical bias (numerical difference from the “true” value), may be accomplished by appropriate use of reference methods, reference materials, or commercially prepared calibrators.64 Because few instruments are precalibrated by the manufacturer, calibration must be performed at initial installation and verified at least every 6 months under the requirements of the Clinical Laboratory Improvement Act of 1988.66 Periodic recalibration may be required after major instrument repair requiring optical alignment or part replacement.
Whole-blood calibration using fresh whole-blood specimens requires the use of reference methods, materials, and procedures to determine “true” values.1, 67, 68 The International Committee for Standardization in Haematology has established guidelines for selecting a reference blood cell counter for this purpose,1 but the cyanmethemoglobin method remains the only standard available in hematology for calibration and quality control.69 Whole-blood calibration, which historically has been considered the preferred method for calibration of multichannel hematology analyzers, has been almost completely replaced by the use of commercial calibrators assayed using reference methods. Calibration bias is possible with the use of these calibrators because of inherent differences in stabilized and preserved cell suspensions.70 It is essential that calibrations be carried out properly and verified by comparison with reference methods or review of quality control data after calibration and by external comparison studies such as proficiency testing.1
Instrument limitations
The continual improvement of automated technologies has resulted in greater sensitivity and specificity of instrument flagging with detection of possible interferences in the data. The parallel improvement in instrument walk-away capabilities has increased the importance of the operator’s awareness and understanding of instrument limitations, however, and of his or her ability to recognize factors that may interfere and cause erroneous laboratory results. Limitations and interferences may be related to methodology or to inherent problems in the blood specimen.
Each instrument has limitations related to methodology that are defined in instrument operation manuals and in the literature. A common limitation of impedance methods is an instrument’s inability to distinguish cells reliably from other particles or cell fragments of the same volume. Cell fragments may be counted as platelets in specimens from chemotherapy-treated patients with increased WBC fragility.1 Likewise, schistocytes or small RBCs may interfere with the platelet count. Larger platelet clumps may be counted as WBCs, which results in a falsely decreased platelet count and potentially increases the WBC count. Micromegakaryocytes may be counted as nucleated RBCs or WBCs. RBCs containing variant hemoglobins such as Hb S or Hb C are often resistant to lysis, and the unlysed cells can be falsely counted as nucleated RBCs or WBCs and interfere in the hemoglobin reaction.71 This phenomenon has become more apparent with the use of milder diluent and lysing reagents in the analyzers with automated WBC differential technology. Non-lysis also may be seen in specimens from patients with severe liver disease, those undergoing chemotherapy treatment, and neonates (due to increased levels of Hb F) on the older Sysmex instruments.15 The ADVIA 2120 and 120 reports the WBC-BASO as the primary WBC count.23 An extended lyse cycle may be used on the CELL-DYN 3500, and the newer instruments are able to provide a correct WBC impedance count when lyse-resistant RBCs are present.21 The Sysmex SE-9000 and Sysmex SE-9500 also have an additional WBC impedance channel.28
Suppression of automated data, particularly WBC differential data, may occur when internal instrument checks fail or cast doubt on the validity of the data. Instruments from some manufacturers release results with specific error codes or flagging for further review. The suppression of automated differential data ensures that a manual differential count is performed, whereas the release of data with appropriate flagging mandates the need for careful review of the data and possibly a blood film examination. This suggests a difference in philosophy among the manufacturers and affects the work flow in different ways.72 More importantly, each laboratory must establish its own criteria for directed blood film review based on established performance goals, instrument flagging, and inherent instrument limitations.
Specimen limitations
Limitations resulting from inherent specimen problems include those related to the presence of cold agglutinins, icterus, and lipemia. Cold agglutinins manifest as a classic pattern of increased MCV (frequently greater than 130 fL), markedly decreased RBC count, and increased MCHC (frequently greater than 40 g/dL). Careful examination of the histograms or cytograms from the instruments may yield clues to this abnormality.73 Icterus and lipemia directly affect hemoglobin measurements and related indices.71 Table 15-2 summarizes conditions that cause interference on some hematology analyzers and offers suggestions for manually obtaining correct patient results. As instrumentation advances, instrumentation software can adjust or correct for some of the conditions listed. Historically, a nucleated RBC flag required examination of a blood film to enumerate the nucleated RBCs and correct the WBC. All four major vendors offer online nucleated RBC enumeration and WBC correction, although the laboratory must validate the results. Lipemia interferes with the hemoglobin reading by falsely elevating the hemoglobin and associated indices. The Siemens technology uses direct measurement of the CHCM parameter, which allows back-calculation of the hemoglobin unaffected by lipemia and thus eliminates the need for the manual method of saline replacement in lipemic specimens. These two examples involving nucleated RBCs and lipemia illustrate instrument advances, and continued future improvements in technology will eliminate or decrease the need for manual intervention to obtain accurate results.
TABLE 15-2
Conditions That Cause Interference on Most Hematology Analyzers
* Manufacturer’s labeling.
† Lipemia shows signature pattern on Siemens ADVIA 120 H cytograms.
‡ HGB can be back-calculated from directly measured MCHC on Siemens ADVIA 120 cytograms.
↑, Increased; ↓, decreased; EDTA, ethylenediaminetetraacetic acid; HGB, hemoglobin; HCT, hematocrit; MCH, mean cell hemoglobin; MCHC, mean cell hemoglobin concentration; MCV, mean cell volume; MPV, mean platelet volume; PLT,platelet count; RBC, red blood cell (or count); WBC, white blood cell (or count).
Specimen age and improper specimen handling can have profound effects on the reliability of hematology test results. These factors have even greater significance as hospitals move toward greater use of off-site testing by large reference laboratories. Specific problems with older specimens include increased WBC fragility, swelling and possible lysis of RBCs, and the deterioration of platelets.15 Stability studies should be performed before an instrument is used, and specific guidelines should be established for specimen handling and rejection.
Clinical utility of automated blood cell analysis
The use of automated hematology analyzers has directly affected the availability, accuracy, and clinical usefulness of the CBC and WBC differential count. Some parameters that are available on hematology analyzers, but cannot be derived manually, have provided further insight into various clinical conditions. The RDW, a quantitative estimate of erythrocyte anisocytosis, can be used with the MCV for initial classification of an anemia.74, 75 Although the classification scheme is not absolute, a low MCV with a high RDW suggests iron deficiency, while a high MCV and high RDW suggests a folate/vitamin B12 deficiency or myelodysplasia (Chapter 19). The immature reticulocyte fraction and the immature platelet fraction provide an early indication of engraftment success after hematopoietic stem cell transplant.76 The reticulocyte hemoglobin content (Chr and Ret-He) provides an assessment of the iron available for hemoglobin synthesis. It is useful in the early diagnosis of iron deficiency, functional iron deficiency, as well as an early indicator of recovery after iron therapy.61, 62, 77The MPV may be useful in distinguishing thrombocytopenia due to idiopathic thrombocytopenia purpura (high MPV), inherited macrothrombocytopenia (higher MPV), or bone marrow suppression (low MPV).78, 79 High MPV values are also associated with higher-risk cardiovascular disease and may have use in assessing a patient’s risk of thrombosis.79, 80 However the use of the MPV in these conditions has been hampered by the varying ability of instruments to accurately measure MPV in patients with macroplatelets (they are underestimated in impedance methods), the lack of standardization of MPV cut-off values in various conditions, and the lack of well-controlled prospective studies to prove clinical utility.79 In addition to method variations, anticoagulation and storage time also influence the MPV, which further impacts the relia bility and clinical utility of MPV results.81
Automation of the WBC differential has had a significant impact on the laboratory work flow because of the labor-intensive nature of the manual differential count. The three-part differential available on earlier instruments generally proved suitable as a screening leukocyte differential count to identify specimens that required further workup or a manual differential count. Partial differential counts, however, do not substitute for a complete differential count in populations with abnormalities. The five-part or six-part automated differentials available on the larger instruments have been evaluated extensively and have acceptable clinical sensitivity and specificity for detection of distributional and morphologic abnormalities.42, 43, 88-94 Abnormal cells such as blasts and nucleated RBCs in low concentrations may not be detected by the instruments but likewise may be missed by the routine 100-cell manual/visual differential count.93-96 The CELL-DYN Sapphire, with its added fluorescent detection technology, has been shown to have high sensitivity and specificity for flagging nucleated RBCs and platelet clumps.24, 48, 97 As technology continues to improve, blood film review to confirm the presence of platelet clumps or nucleated RBCs and to correct leukocyte counts for interference from platelet clumps or nucleated RBCs is becoming unnecessary, especially for nucleated RBCs, because the four major vendors now count and correct the WBC for nucleated RBCs on their high-end analyzers.43, 97, 98
Instrument evaluations based on the Clinical and Laboratory Standards Institute H20-A2 standard on reference leukocyte differential counting99 using an 800-cell or 400-cell manual leukocyte differential count as the reference method have shown acceptable correlation coefficients for all WBC types, with the possible exception of monocytes.43, 72, 93, 94, 100-102 However, further studies using monoclonal antibodies as the reference method for counting monocytes suggest that automated analyzers yield a more accurate assessment of monocytosis than do manual methods.103, 104
Histograms and cytograms, along with instrument flagging, provide valuable information in the diagnosis and treatment of RBC and WBC disorders. Multiple reports indicate the usefulness of histograms and cytograms in the characterization of various abnormal conditions, including RBC disorders such as cold agglutination and WBC diseases such as leukemias and myelodysplastic disorders.33, 73, 93, 94
Manufacturers are developing integrated hematology workstations for the greatest automation and laboratory efficiency. The Beckman Coulter LH 1500 Automation Series is Beckman’s solution to integrated hematology. The line can be customized to have two to four LH analyzers as well as SlideMakers and SlideStainers based on the laboratory’s needs for efficiency and automation.33 The Sysmex Total Hematology Automation System (HST series) robotically links the SE-9000, R-3500 (automated reticulocyte analyzer), and SP-100 (automatic slide maker/stainer). The HST line links two XE-2100 units and one SP-100 instrument for complete automation or systemization of hematology testing. The SE-Alpha is a smaller version that links the SE-9000 and SP-100.28 The Siemens ADVIA LabCell links the ADVIA 2120i to the track, and the Autoslide (automatic slide maker/stainer) links to the ADVIA 2120i. Finally, as a result of increasing customer needs, manufacturers have added body fluid counting to their high-end instrumentation. The Beckman Coulter UniCel DxH 800 system, the LH 780, and the Sysmex XN series and XE-5000 count WBCs and RBCs in body fluids; the ADVIA 2120i/120 counts WBCs and RBCs in body fluids and, in addition to cerebrospinal fluid WBC and RBC cell counts, performs a differential count on cerebrospinal fluid.33, 105, 106
Selection of a hematology analyzer for an individual laboratory requires careful evaluation of the laboratory’s needs and close scrutiny of several important instrument issues, including instrument specifications and system requirements, methods used, training requirements, maintenance needs, reagent usage, data management capabilities, staff response, and short-term and long-term expenditures.107 All instruments claim to improve laboratory efficiency through increased automation that results in improved work flow and faster turnaround time or through the addition of new parameters that may have clinical efficacy. All four major vendors offer a slide maker/stainer that can be connected directly to their high-end analyzers. The slide makers/stainers can be programmed to make blood films for every specimen or to make films based on the laboratory’s internal criteria for a film review. This reduces, but does not completely eliminate, the use of manual peripheral blood film review. Automated slide makers/stainers connect only to high-end analyzers and as such are not suitable for some laboratories. Each laboratory must assess its own efficiency needs to determine if a slidemaker and stainer is a value-added instrument to the laboratory.108 The instrument selected should suit the workload and patient population and should have a positive effect on patient outcomes.109 The instrument selected for a cancer center may be different from that chosen for a community hospital.110 Ultimately, however, the instrument decision may be swayed by individual preferences.
Summary
Review questions
Answers can be found in the Appendix.
Examine the histograms/scatterplots obtained from four major instruments for the same patient specimen (Figure 15-14, A-D). Compare the results, and respond to questions 1 to 4 based on the results.
|
___ |
Impedance |
a. Uses diffraction, reflection, and refraction of light waves b. Uses high-voltage electrical waves to measure the internal complexity of cells c. Involves detection and measurement of changes in electrical current between two electrodes |
|
___ |
Abbott CELL-DYN Sapphire |
a. Volume, conductivity, and five angles of light scatter |
|
___ |
Siemens ADVIA 2120i |
b. MAPSS technology and three-color fluorescence |
|
___ |
Sysmex XN-1000 |
c. Peroxidase-staining absorbance and light scatter |
|
___ |
Beckman Coulter UniCel DxH 800 |
d. Detection of forward and side scattered light and fluorescence |
FIGURE 15-14 Composite scatterplots/histograms obtained from four major instruments. A, Coulter UniCel DxH 800; B, ADVIA 2120i; C, Sysmex XN-1000; D, CELL-DYN Sapphire.
References