Response of Red Blood Cell Control Materials to Altered Testing Conditions

Transcription

1 ANNALS OF CLINICAL AND LABORATORY SCIENCE, Vol. 18, No. 1 Copyright 1988, Institute for Clinical Science, Inc. Response of Red Blood Cell Control Materials to Altered Testing Conditions MARK E. SHERMAN, M.D.t, CHARLES F. ARKIN, M.D.$ and ARLENE CHAPLIN, MT(ASCP)SH:j: f Department of Pathology, Montefiore Medical Center Bronx, NY and fdepartment of Clinical Pathology, New England Deaconess Hospital, Boston, MA ABSTRACT W hile many features are regularly considered in selecting control reagents, the responsiveness of these products to méthodologie errors is frequently overlooked. To address this issue with regard to commercially prepared whole blood cell controls, mean cell volume (MCV) determinations were performed on several preserved red blood cell controls and fresh blood in the presence of various artifacts. The effects contaminating the isotonic d ilu ent w ith w ater, saline, or bleach w ere com pared. Although our results indicate that preserved and fresh RBCs generally behave similarly, large differences were observed in the responses to hypotonic stress among controls and between controls and fresh specimens. As this may have clinical relevance, our data highlight the need for testing the responsiveness of controls to commonly encountered m éthodologie variables as part of the routine evaluation of these products. Introduction Use of commercially prepared whole blood cell controls to m onitor performance of cell counters in the clinical hem atology laboratory is an accepted practice. Ideally, these reagents would behave identically to fresh human blood when tested, yet possess indefinite stability, thus allowing repeat testing and * Send rep rin t requests to C harles F. Arkin, M.D., C hief of Clinical Pathology, New England Deaconess Hospital, Meissner Building, Room 323, Boston, MA long-term instrum ent monitoring. Use of controls th a t are less responsive to altered testin g conditions than fresh patient specim ens might allow significant sources of error to go undetected. Alternatively, processed control cells which show greater responsiveness to changes in test variables than fresh blood cells could draw attention to inconsequential méthodologie variation. Accordingly, experiments were carried out on six commercial cell control preparations /88/ $00.90 Institute for Clinical Science, Inc.

2 20 SHERMAN, ARKIN, AND CHAPLIN by com paring their responses to cor tro lled artifacts w ith those of fresl human blood. M aterials F r e s h W h o l e B l o o d S p e c im e n s Patient samples collected in ethylenediamine tetraacetic acid (EDTA) were selected daily from among the specimens submitted to the clinical hematology laboratory in a large hospital for routine complete blood cell counts. These samples w ere chosen to represent a spectrum of p atient diagnoses and a wide range of m ean cell volum es (MCVs). F iv e to 11 sam p les from d iffe re n t patients were evaluated each day of the study. A fresh specimen from a normal volunteer was also included each day. C o m m e r c ia l C e l l C o n t r o l s Preparations of the following p reserved com m ercial cell controls were tested : 4C,* D iff T ro l,t C o unter- C heck^ CBC 3D, Para 8 and Interscience M ulti-param eter Hematology Control with Platelets. H Control preparations w ere stored and used in accordance with the m anufacturers recommendations in the package inserts. Methods S a m p l e P r e p a r a t io n The effects of hypotonicity, hypertonicity, and bleach on the mean cell volume were examined. These conditions were * Coulter Diagnostics, Hialeah, FL. t Dade, Miami, FL. t Diagnostic Technology, Hauppauge, NY. R & D Systems, Minneapolis, MN. Streck, Omaha, NE. 11 Interscience Diagnostics, Fairview, OR. t t American Scientific Products, McGaw Park, IL. achieved by adding contaminating solutions to conventionally prediluted specimens. Fresh whole blood and comercial preparations w ere treated identically. Twenty ml aliquots of each specimen were prepared in duplicate with isotonic buffered salineçi to have a final dilution of 1:50,000. All such duplicates were randomly processed in each run. Exactly one minute prior to testing, the specific solution used to induce the artifact was introduced and the mixture gently swirled. Pilot studies were used to determine the concentration of solution required to produce clinically im portan t artifactu al changes in the MCV of fresh RBC s. Osmolality was determined by freezing point depression. Each experiment was performed on a separate day with a separate set of fresh specimens. I n s t r u m e n t a t io n The C oulter ZBI equipped w ith an M CV/Hct accessory:^ was used for all MCV determinations and RBC counts. It was o p erated in accordance w ith the manufacturer s instructions1. The MCV was calibrated daily using fresh whole blood samples. After testing each sample, the orifice was flushed until background counts fell below 50 per fxl. MCV E x p e r im e n t s The effect of hypotonic media on the MCV was assessed by pipeting 1.0, 2.5, or 4.0 ml of deionized water into each 20 ml sample prediluted in isotonic diluent. The final osmolalities of the isotonic diluent similarly diluted were 325, 300, and 280 mosm as compared with 337 mosm for the unaltered diluent. The effects of hypertonicity were evaluated by pipeting 2. 0 ml of hypertonic saline (20 gm N ac l in 0.5 L deionized water) solution into each sample. The final osmolality was 398 mosm. t t Coulter Electronics, Hialeah, FL.

3 RED BLOOD CELL CONTROL MATERIALS 21 The effects of two concentrations of bleach on the MCV of fresh blood and control samples were studied: (1) 100 xl of pure bleach in each 2 0 ml specimen and (2 ) 2.5 ml of a 1 : dilution of pure bleach in the 20 ml specimens. The final osmolalities of the samples w ere 358 mosm and 306 mosm, respectively. CHANGE IN NCI) (Ü) O s m o t ic F r a g il it y S t u d ie s O sm otic fragility testing was p e r formed according to standard methods2. Samples of five preserved controls were incubated with seven solutions of sodium chloride ranging in concentration from 0 to 70 percent. A fresh sam ple from a normal volunteer was included as a control. Results The change in MCV observed following the introduction of each artifact is expressed as the mean percent change above or below baseline measurements. R esults of testing duplicate sam ples agreed within two xm3(fl). H y p o t o n ic S t r e s s The changes in MCV of samples in response to a hypotonic environment are shown in figure 1. The label, fresh, indicates the pooled results for the six fresh whole blood specimens in diluent having concentrations of 325, 300, and 280 mosm. T he average percentage increase in MCV above baseline for fresh specimens was 1.9 percent, 5.8 percent, and 10.4 p e rc e n t, respectiv ely. The Dade reagent cells responded most like fresh hum an blood, show ing MCV increases of 1.2 percent, 3.5 percent, and 8.2 percent, respectively. Nearly all fresh samples expanded more than those of the preserved controls at each dilution of isotonic saline. The cause of shrinkage 2 F i g u r e 1. The comparative effect on the mean cell volume of suspending fresh blood and commercial controls in hypotonic diluent. Solid bar = 325 mosm, striped bar = 300 mosm, and open bar = 280 mosm. of the 4C cells at 325 mosm is not known but may be related to the osmolality of its preservative diluent. H y p e r t o n ic S t r e s s In table I are depicted the changes in MCV following incubation of controls and patients in hypertonic saline. The decrease in MCV of five of the six controls tested appeared com parable to fresh blood. Interscience cells shrunk, however, over 1.5 times more than any other sample tested. B l e a c h E f f e c t s The effects of concentrated and dilute bleach on erythrocyte MCV are summarized in table II. All samples contracted m arkedly in p u re bleach. D ade cells changed the most in volume and Interscience the least. O s m o t ic F r a g il it y S t u d ie s Results of osmotic fragility testing of three control preparations resem bled those obtained w ith the fresh control (table III). Slight differences were noted betw een the preserved cells and fresh blood.

4 2 2 SHERMAN, ARKIN, AND CHAPLIN Specimen TABLE I Incubation with Hypertonic Saline Percent Change MCV Para 8* Xnterscience Diff-Trol* C* Countercheck* CBC-3Df Fresh blood (-8.1 to -11.4) *High, normal range controls tested fnormal range control tested Eight samples tested Comment Preparation of cells for use as quality control products requires modifications w hich perm it long-term storage and rep eat analysis. U nfortunately, these modifications may alter normal m em brane properties3,4 and thus render control cells unlike native hum an blood cells. The effects of osmotic matrix on MCV determ inations on both patient H em atology Survey sam ples is w ell known7. Our observations show that control materials vary in their sensitivity to altered testing conditions and in their similarity to fresh blood. O ur data confirm that control cells are less distensible than fresh human erythrocytes and vary in their response to hypotonic stress. Decreasing the osmolality from 337 mosm to 300 mosm caused the MCV of patients to increase 5.8 percent on average, whereas the five control preparations showed an average increase of only 2.6 percent. F u rth er reduction to 280 mosm caused the MCV of fresh RBCs to increase over 10 percent, yet the increase in the MCV of commercial controls varied from 2.4 percent to 8.2 percent. Therefore, it is possible that under certain circumstances some commercial controls might falsely signal that the test system is in control in the face of a large systematic error. Dade control cells appeared most like human blood in this experiment; Coulter 4C cells, th e least. T he D ade cells appeared to attain a progressively higher percent increase in MCV as the osmolality decreased, whereas the other produ c ts d e m o n s tr a te d a p ro g r e s s iv e d e c re a se. T he re le v a n c e of th is is unclear, but it suggests some products more closely resemble the human cells than others (figure 1 ). C ontrols and p a tie n t cells resp o n d ed sim ilarly in hypertonic media. Interscience reagent, which had the highest baseline MCV, shrunk 17.8 percent, which was significantly m ore than the o th er products tested. Bleach is commonly used to remove debris from the orifice of automated cell counters. Our data suggest that concentrated bleach causes both patient and control RBCs to crenate markedly without producing cell lysis. D ilute bleach causes fresh and control RBCs to swell to a variable d egree (table II). The effects of dilute bleach on the MCV are out of proportion to the hypotonicity. Osmotic fragility testing demonstrated slight differences betw een some pre- Sample TABLE I I Effect of Bleach on Mean Cell Volume Percent Change in MCV After Exposure to Concentrated Bleach Percent Change in MCV After Exposure to Dilute Bleach Para * +27.2f Interscience Diff-Trol -35.4* +12.Of 4C -33.2* +21.4f Coun te r-check -33.0» + 8.7f 3D -21.6* Fresh BloodS (-34.0 to -38.2) (+8.7 to +19.0) *High, normal, low range controls tested fhigh, normal range controls tested Twelve samples tested in concentrated bleach; eight samples tested in dilute bleach.

5 Percent Saline TABLE I I I Osmotic Fragility Testing of Commercial Cell Controls - Percent Lysis at Various Saline Concentrations Control 4C Para 8 RED BLOOD CELL CONTROL MATERIALS 23 Diff- Trol CBC-3D countercheck served and fresh erythrocytes. Two preparations, C ountercheck and CBC-3D, appeared to lyse slightly more readily than fresh blood, possibly reflecting relatively more cellular rigidity. O ur results demonstrate that fresh and preserved RBCs frequently respond differently to artifacts such as alteration of osmolality or exposure to bleach. Most importantly, some of these differences in responses may be of sufficient magnitude to influence medical decisions and could adversely affect patient care. The large artifacts required to reproduce our findings seem unlikely to occur during routine testing of patient specimens. Therefore, it is anticipated th at all of the commercial control materials evaluated would perform acceptably in most situations. However, it is not possible to predict all the possible artifacts which may be encountered during laboratory testing. Thus, it would seem pru dent to se le c t c o n tro l p re p a ra tio n s w hich respond most similarly to native human blood. Presum ably, by stu dyin g the responses of control preparations and fresh blood to experimentally induced artifacts, it is possible to obtain data which are required to make this comparison. G e n e ra lly, a co n tro l m a te ria l is selected on the basis of cost considerations, safety and convenient packaging, vial-to-vial variability, long term stability, and in d e p e n d e n t assayability 5 6. However, this product, although judged favorably, may, in fact, be undesirable in that it does not fulfill its cardinal purpose, detection of méthodologie errors w hich yield spurious patient results. Therefore, the evaluation of quality control m aterial should include testing its responsiveness to a series of variables or interferences likely to occur in the laboratory environment. Acknowledgments Thanks are extended to the participating manufacturers for providing the RBC controls employed in this study and to Allan Paduano and Shelley Northup for technical assistance. References 1. Coulter Counter Model ZBI, Product Reference M anual, E, F ebruary 1979, C oulter Electronics, Hialeah, FL. 2. D a c ie, J. V.: Osmotic fragility of erythrocytes. Laboratory Medicine: Hematology. Miale, J. B. St. Louis, The CV Mosby Company, 1972, pp H am, T. H., D u n n, R. F., S a y re, R. W., e t al: P h y sic al p ro p e rtie s o f re d c ells as r e la te d to effects in vivo. I. In creased rigidity o f ery th ro c y tes as m e a su re d by viscosity o f cells a lte re d by c h e m ic a l fix a tio n, sic k lin g a n d h y p o to n ic ity. B lood 3 2 : , K o e p k e, J. A.: The calibration of autom ated instruments for accuracy in hemoglobinometry. Am. J. Clin.Pathol. 68: , L e w is, S. M.: Developing a blood-cell standard. M odern Concepts in Hematology; Symposia of the International Committee for Standardization in Hematology. Izak, G. and Lewis, S. M., eds. New York, Academic Press, 1972, pp L e w is, S. M.: Standards and reference preparations. Quality Control in Haematology: Symposium of the International Committee for Standardization in Haematology. Lewis, S. M. and Coster, J. F. eds. New York, Academic Press, 1975, pp Sa v a g e, R. A.: E vidence for hyperglycem ia osmotic matrix effects on the com prehensive hematology survey Am. J. Clin. Pathol. SO(Suppl): , 1983.