ANNALS O F CLINICAL AND LABORATORY SC IEN CE, Vol. 20, No. 1 Copyright 1990, Institute for Clinical Science, Inc. A Comparison of two Macroscopic Platelet Agglutination Assays for von Willebrand Factor RHONDA D. WRIGHT, M.D. and JONATHAN S. KRAUSS, M.D. Department o f Pathology, Medical College o f Georgia Hospital, Augusta, GA 30912-3620 ABSTRACT Plasma drawn from 22 patients with a broad range of von W illebrand factor (vwf) levels, including one patient with rare, severe type III von W illebrand s disease, was used to investigate the adaptability of the Bio/ Data Corp. vw Factor Assay, a ristocetin cofactor-based kit developed for use w ith a platelet aggregom eter, to a macroscopic platelet agglutination technique. The results were compared to those obtained using the G eneral Diagnostics von W illebrand Factor Assay, another ristocetin cofactor-based kit developed for use as a macroscopic platelet agglutination procedure. No significant difference (p > 0.5) existed betw een results obtained by these two kits. The mean and standard deviation of the vw F levels (as percent of normal) were 132 ± 66.7 by the General Diagnostics kit and 133 ± 59.0 by the Bio/Data Corp. kit. The correlation coefficient (r) was 0.81. The Bio/Data Corp. kit is more costly per test than the G eneral Diagnostics kit. However, it was shown to be compatible with the macroscopic agglutination technique and required no modification to be used in that manner. Introduction The level of von W illebrand factor (vw F; factor V H I-related antigen) in norm al and deficient patients can be m easured by several methods, some of which have becom e available com m ercially as kits. One of these kits, the Bio/ Data Corp. vw Factor Assay * is based on quantitation of the ristocetin cofactor activity of vw F1,3 and was developed for u se w ith a p la te le t a g g re g o m e te r. A nother kit, the G eneral Diagnostics von W illebrand Factor Assayf is based on the same principle, but was made to be used either in a macroscopic platelet agglutination procedure or with a platelet aggregom eter. The macroscopic procedure using this kit is currently utilized in this hospital to determ ine vw F levels for clinical purposes. The adaptability of the Bio/Data Corp. kit to a macroscopic technique has been investigated and the results compared to those obtained with the G eneral Diagnostics kit. * Bio/Data Corp., Hatboro, PA 19040. t Organon Teknika Corp., Durham, NC 27704. 73 0091-7370/90/0100-0073 $00.90 Institute for Clinical Science, Inc.
74 WRIGHT AND KRAUSS M aterials and M ethods St u d y S a m p l e s Plasma samples from 22 patients with and without known disorders of hem o stasis, including one patient with severe ty p e I I I von W il le b r a n d s d is e a s e (vwd), one with mild type I vw D, and one with hem ophilia A, w ere selected for study. T h ese sam ples w e re p re p a re d from whole blood which had been col lected by routine venipuncture methods in vacuum tubes containing 38 g per liter sodium citrate. T he blood was ce n tri fuged (15, 2000 X g, 10 min) and the p la te le t-p o o r p lasm a was draw n off. Prothrom bin tim e and activated partial throm boplastin tim e w ere m easured, and th e rem a in d e r of th e sam ple was held at 4 for varying periods of tim e (up to eight hours) before being stored at 20 - P a t ie n t s Among the patients are four who can be singled out: (1) Patient 1 has severe type III vw D9 which was discovered at the age of six months, and suffers from easy bruisability. His factor VIII level was two percent of normal. R istocetin-induced platelet aggregation (using th e p a tie n t s own FIGURE 1. C om parison of ristocetin-induced platelet-rich plasma) was absent (figure platelet aggregation in platelet-rich plasma from 1). He had no detectable antigenic vw F patient with severe type III von W illebrand s disease (vwf:ag; factor V U I-related antigen) by (arrow) vs. normal control. Note lack of aggregation standard q u a n tita tiv e im m unoelectro- in the von W illebrand s disease patient s plasma. phoresis (Laurell rocket technique)4 (fig ure 2) nor by a special electroim m uno which to perform m ultim eric analysis, d iffu s io n m e th o d w h ic h h a s b e e n w hich revealed a norm al pattern. His established using peroxidase-labelled m aternal grandm other has normal levels polyclonal anti-vw F $ to increase the of factor VIII and vwf. sensitivity of the test for detection for (2) Patient 2 has mild type I vw D. very low levels (as low as 1/128 plasma His vw F levels have been 30 to 50 p er d ilution) of v W F :A g.2 H ow ever, the cent of normal consistently in the past. Scripps Clinic in Lajolla, California, was (3) Patient 3 has hemophilia A, with a able to detect trace amounts of vw F on factor V III level of 1.7 p ercen t. This study includes the first determ inations of i Dako Corp., Santa Barbara, CA 93103. his vw F levels.
75 TWO ASSAYS FOR VON W ILLE BRAND FACTOR Q Q C 5-? 6 1 F ig u r e 2. Quantitative immunoelectrophoresis (Laurell rocket technique) for vwf:ag in plasma from patient with severe type III von W illebrand s disease. Note nonspecific diffusion of stained material into surrounding gel without formation of rocket pattern (arrows). Remaining wells contain two sets of serial dilutions of normal reference plasma (from right to left, undiluted, 1:2, 1:4, and 1:8). (4) Patient 4 is the half-sister of pa tient 2. H er own vw F level has not been determ ined previously. T h e r e m a in in g 18 p a tie n ts w e re selected from the hospital inpatient pop ulation on th e basis of having norm al prothrom bin tim e and activated partial throm boplastin tim e. T heir hem ostatic status was otherwise unknown. G e n e r a l D ia g n o s t ic s K it L yophilized p la te le t and risto ce tin reagents from the kit w ere reconstituted in distilled w ater, as specified in the instructions. Normal reference plasma, supplied in the kit, was reconstituted and dilutions (undiluted, 1:2, 1:4, and 1:8) for th e activity calibration curve w ere prepared in TES buffer, also sup plied. Dilutions of 1:2 of all the patient samples, except for Nos. 1 and 2, were p rep ared using the same buffer. (Sam ples 1 and 2, from p a tie n ts known to have low levels of vw F, w ere run undi luted.) All calibration and patient sam ples w ere run in d uplicate. For each sample, two polystyrene tubes w ere pre pared, each of which contained 0.2 ml of reconstituted platelets and 0.05 ml of ristocetin reagent. To each tube 0.1 ml of appropriately dilu ted plasm a was then added, and the m ixture was gently agi tated until 4 + clum ping, defined as the aggregation of all particles into large clumps with clearance of visible turbid ity of the fluid, was observed. The time n e e d e d for c lu m p in g to d ev elo p was reco rd ed in seconds and th e average tim e of the duplicates was taken as the reported tim e for each sample. The cali bration curve was prepared by plotting on semi-log graph paper the reported clum ping tim es (representing vw F ris tocetin cofactor activity) for the u n d i luted, 1:2, 1:4, and 1:8 dilutions of the reference plasma as 100 percent, 50 p er cent, 25 percent, and 12.5 percent activ ity, respectively. The vw F level (as p er cent of normal) of the patient samples was th e n d e te rm in e d from th e ir r e ported clum ping times by interpolation from the calibration curve and m ultipli cation of the resulting percentage value by the applicable dilution factor. B i o / D ata C o r p o r a t io n K it Ristocetin reagent and norm al refer ence plasma from the kit w ere reconsti tuted in distilled water, as specified in th e instructions. L yophilized p late le t
7 6 WRIGHT AND KRAUSS reagent, supplied in the kit, was reconstituted in tris buffered saline, also supplied. Dilutions (undiluted, 1:2, 1:4, and 1:8) of the normal reference plasma for the percent activity calibration curve and 1:2 dilutions of all the patient samples, except for Nos. 1 and 2, were prepared using the same buffer. (Samples 1 and 2 were run undiluted.) All calibration and patient samples were run in duplicate. Amounts of reagents and samples used were, according to the kit instructions, as follows: for each sample, two polystyrene tubes were prepared, each of which contained 0.4 ml of reconstituted platelets and 0.05 ml of ristocetin. To each tube 0.05 ml of appropriately diluted plasma was then added, b u t instead of testing this mixture in a platelet aggregometer, the mixture was gently agitated until 4 + clum ping, defined as before, was observ ed. T he tim e n e e d ed for clum ping to develop was recorded in seconds and the average tim e of the duplicates was taken as the reported tim e for each sam ple. The calibration curve was prepared and vw F ristocetin cofactor percent activity of the patient samples was determ ined in the same m anner as for the General Diagnostics kit. Results Plasm a vw F levels of th e patien ts studied are shown in table I. The normal range for th e G eneral Diagnostics kit when used in the m anner described is 52 to 144 percent, according to the package insert supplied with the product, which also states that results greater than 45 percent may be considered normal. The insert supplied with the Bio/Data Corp. kit states that a result less than 40 p ercen t (when obtained by th e p latelet aggregom eter m ethod) is considered abnormal; no upper lim it of normal is indicated. In this study, no significant difference (p > 0.5) existed betw een results obtained by these two kits when T A B L E I Comparison of Two Kit Procedures for Determining Plasma Levels of von Willebrand Factor by Ristocetin Cofactor Activity vwf (as Percent of Normal Ristocetin Cofactor Activity) Patient Number General Diagnostics Bio/Data 1 < 12.5 < 12.5 2 45 33 3 52 124 4 112 88 5 > 200 180 6 > 200 > 200 7 158 124 8 > 200 200 9 < 25 134 10 140 128 11 170 150 12 > 200 > 200 13 82 154 14 > 200 > 200 15 96 72 16 104 100 17 140 150 18 > 200 > 200 19 52 48 20 > 200 202 21 126 112 22 > 200 118 Mean 132 133 Std. dev. 66.7 59.0 used for the macroscopic platelet agglutination method. Additionally, the results obtained on individual patients showed a high degree of correlation (r = 0.81; p < 0.001), and the relationship was essentially linear (figure 3). Patients 1 and 2, the two patients in the study with known vw F d e ficien cy (see M aterials and M ethods), w ere correctly identified as having low vw F levels by both of the kits. P atient 1 show ed no d etectab le vw F activity by either of the two kits. The vw F activity for patient 2 was 45 percent of normal by the General Diagnostics kit and 33 percent of normal by the Bio/Data Corp. kit, which is within the range previously established in this patient. Both kits show the vw F activity to be within the normal range for the patient with hemophilia A (patient 3) and for the half-sister (patient 4) of patient 2. In reviewing the results, patient 9 is noted to have an abnormally low level of vw F by one kit but not the other. It is
TWO ASSAYS FOR VON WILLEBRAND FACTOR 7 7 F ig u r e 3. Von Willebrand Factor as per cent of normal ristocetin cofactor activity by platelet agglutination in plasma of 22 patients: Results obtained using the Bio/Data Corp. vw Factor Assay kit vs. results obtained using the General Diagnostics von Willebrand Factor Assay kit. likely that the (low) General Diagnostics result represents a random erro r and that the Bio/Data Corp. result is correct, as this patient s health history does not suggest any disorder of hemostasis. Discussion Von W illebrand Factor is a glycoprotein composed of multimeric subunits of varying sizes.7 Higher levels of vw F as vw F:Ag than as ristocetin cofactor activity may be found in normal individuals because quantitative immunoelectrophoresis m easures any immunologically recognizable vw F present in the sample, regardless of the size of the m ultim ers,10 whereas the ristocetin cofactor m ethod measures the ability of a patient's plasma to induce a synthetic hemostatic-system analog to perform a specific vw F-dependent task. Ristocetin cofactor activity is a function of both large and interm ediatesized m u ltim ers of vw F, b u t is not affected by the presence or absence of small m ultim ers.5,10 N either m ethod is a p erfect reflec tio n of th e tru e level of functional vw F, as only the large m ultim ers are particularly efficacious in bringing about hem ostasis.7,810 Some subtypes of vw D (notably those of type II vwd) show different apparent levels of vw F depending on the m ethod used for quantitation because of abnormalities in the proportion or structure of the m ultim ers, the im portance of this fact being that the subtypes also differ in their response to therapy involving 1- desam ino-8-d-arginine vasopressin.8 Initial investigation of suspected vw D should therefore include at least a platelet count, p rothrom bin tim e and activated partial throm boplastin tim e, determ ination of b leeding tim e, and q u a n tita tio n of vw F le v e l b o th as vw F:Ag and as ristocetin cofactor activity. P re lim in a ry su b ty p in g m ay be accomplished with the aid of these tests while the definitive diagnosis is pending. This information may be of value to the clinician in m anaging an acu tely ill patient, as the definitive diagnosis necessarily includes either crossed immunoelectrophoresis or m ultim eric analysis, studies which cannot be perform ed in most institutions and must be referred to an outside laboratory. Furtherm ore, vw D is one of the more common disorders of hemostasis, with a reported incidence as high as 8.2 per 1000 in one study.6 Accordingly, a reasonably rapid and inexpensive screening test for the von W illebrand group of diseases has great utility in the initial investigation of th e non-throm bocytopenic bleeding patient. In this regard, macroscopic platelet agglutination using either of the ristocetin cofactor-based kits can be perform ed im m ediately after the dilutions have been made, whereas the e le ctro p h o resis step of q u a n tita tiv e immunoelectrophoresis requires several hours to com plete, in addition to the time necessary to prepare the antibodyim pregnated gels. The B io/d ata C orp. kit, w hich at $65.90 contains enough reagents to run
7 8 WRIGHT AND KRAUSS one four-point standardization curve and six samples ($65.90 -h 6 = $10.98 per sample), is more costly per sample than the General Diagnostics kit ($116.31 per kit containing reagents sufficient to run one four-point standardization curve and 16 samples, $116.31 16 = $7.27 per sample). However, our data indicate that the Bio/D ata C orp. kit appears to be quite compatible with the macroscopic agglutination technique and requires no modification to be used in that manner. A definitive study is suggested involving larger num bers of vw D patients than are at the authors disposal and comparing the results with those obtained on the same samples by platelet aggregometry, to investigate this issue further. References 1. Br in k h o u s, K. M., G raham, J. E., C o o p e r, H. A., Alla in, J. P., and Wagner, R. H.: Assay of von Willebrand factor in von Willebrand s disease and hemophilia: Use of a macroscopic platelet aggregation test. Thromb. Res. 6:267-272, 1975. 2. B r o e, M. K. and I n g il d, A.: Amplification of immunoprecipitates in agarose gels by horseradish peroxidase-labelled antibody. Scand. J. Immunol. 17, Suppl. 10:255-258, 1983. 3. H o w a r d, M. A. and F ir k in, B. G.: Ristocetina new tool in the investigation of platelet aggregation. Thromb. Diath. Haemorrh. 26:362-369, 1971. 4. L a u r e ll, C. B.: Quantitative estimation of proteins by electrophoresis in agarose gel containing antibodies. Anal Biochem 75:45-52, 1966. 5. M a r tin, S. E., M a r d er, V. J., F ra n c is, C. W. and Ba r lo w, G. H.: Structural studies on the functional heterogeneity of von Willebrand protein polymers. Blood 57:313 323, 1981. 6. Ro d e g h ie r o, F., C astam an, G., and D in i, E.: Epidemiological investigation of the prevalence of von Willebrand s disease. Blood 69:454-459, 1987. 7. R u g g e r i, Z. M. and Z im m e r m a n, T. S.: The complex multimeric composition of factor VIII/ von W illebrand factor. Blood 57:1140-1143, 1981. 8. R u g g e r i, Z. M. and Z im m e r m a n, T. S.: Von Willebrand factor and von Willebrand disease (review). Blood 70:895-904, 1987. 9. Sh e l t o n -I n l o e s, B. B., C h e h a b, F. F., M ann u c c i, P. M., F e d e r ic i, A. B. and Sa d l e r, J. E.: Gene deletions correlate with the development of alloantibodies in von Willebrand disease. J. Clin. Invest. 79:1459-1465, 1987. 10. Z im m e r m a n, T. S. and R u g g e r i, Z. M.: Von Willebrand s disease (review). Clin. Haematol. 12-, 175-200, 1983.