Laboratory diagnosis of heparin-induced thrombocytopenia type II after clearance of platelet factor 4/heparin complex

Transcription

1 Laboratory diagnosis of heparin-induced thrombocytopenia type II after clearance of platelet factor 4/heparin complex J. HARENBERG, L. C. WANG, U. HOFFMANN, G. HUHLE, and M. FEURING HEIDELBERG, GERMANY Laboratory confirmation of heparin-induced thrombocytopenia (HIT) is limited by assay sensitivity. We investigated whether laboratory confirmation can be improved after antigen clearance by determining free antibody and combining the results of antigenic and biologic assays. Blood samples were collected over 5 to 6 weeks in 14 HIT patients. As an antigenic assay, the fluorescence-linked immunofiltration assay (FLIFA) was performed, and as a biologic assay, the carbon 14 labeled serotonin release assay was performed. At day 1 when heparin was stopped, 11 of 14 patients showed positive results in both assays; thus each assay had a sensitivity of 80%. The 3 patients with negative results seroconverted in one or both assays during the subsequent 7 days. Combining the positive results of the assays increased the sensitivity to 100% at day 7, regardless of whether the antigenic or the biologic assay was performed first. Both assays became negative in all patients within 5 to 6 weeks. The sensitivity of antigen and biologic assays in HIT patients increased to 100% after the time course of the heparin-induced antibody. We assume that in some HIT patients the free antibody can be detected after withdrawal of heparin and after clearance of the platelet-factor 4/heparin complex. (J Lab Clin Med 2001;137:408-13) Abbreviations: ELISA = enzyme-linked immunosorbent assay; FLIFA = fluorescence-linked immunofiltration assay; HIT = heparin-induced thrombocytopenia; IgG = immunoglobulin G; PF4 = platelet factor 4; SRA = serotonin release assay Thrombocytopenia is one of the major concerns and side effects during therapy with heparins. The moderate form appears within a few days of the onset of treatment and spontaneously reverses under continued treatment. If thrombocytopenia occurs 5 to 15 days after the onset of heparin therapy, HIT is almost severe (<100, platelets/l), can result in thromboembolism, and normalizes only after with- From the Department of Medicine IV, University Hospital, and the Department of Clinical Pharmacology, Faculty of Clinical Medicine Mannheim of the Ruprecht-Karls University of Heidelberg. Submitted for publication July 10, 2000; revision submitted January 9, 2001; accepted February 6, Reprint requests: J. Harenberg, MD, Professor of Medicine, Department of Medicine IV, University Hospital, Theodor-Kutzer-Ufer, Mannheim, Germany. Copyright 2001 by Mosby, Inc /2001 $ /1/ doi: /mlc drawal of heparin. 1,2 The diagnosis of HIT may be difficult on the basis of clinical symptoms alone, especially in patients with other diseases that may induce thrombocytopenia. Laboratory confirmation of HIT with biologic or antigen assays is therefore required. Biologic assays include the carbon 14 labeled serotonin release assay, 3 heparin-induced platelet activation assay, 4 and flow cytometric assay. 5 Antigen assays determine antibodies toward the PF4/heparin complex by ELISA, 6 heparin-induced IgG by biotin-labeled PF4, 7 and fluorescence-labeled heparin. 8,9 The sensitivity of all assays ranges from 80% to 98%. Antigenic and biologic assays have been compared in several studies to define the sensitivity and specificity. 2,4,10-15 A combination of the results of two assays to enhance the laboratory confirmation of HIT has been suggested 14 but has not yet been studied. Recently an improvement of the sensitivity from 56% to 85% by a combination of assays has been reported 408

2 J Lab Clin Med Volume 137, Number 6 Harenberg et al 409 Fig 1. The time courses (pattern 1) of the FLIFA, the 14 C-labeled SRA, and the platelet count are shown in a in HIT without thrombosis 16 but has not yet been studied in HIT with thromboembolism. There is evidence from the literature that free antibody can be detected after clearance of the immune complex. 17 Based on these observations, we assumed that in HIT patients with negative test results at day 1 of blood sampling, the antibody may appear in the first few days after disappearance of the antigen. The time course of PF4 and heparin antibodies over 50 days has been reported in 1 HIT patient, 18 and another patient was monitored for 2 weeks with the 14 C-labeled serotonin assay. 3 METHODS Patients. The diagnosis of HIT was made in 14 patients during heparin treatment if all of the following criteria were fulfilled: fall in platelet count to below 100, /L or by >50% of the pretreatment value; occurrence of thromboembolism during heparinization; return of platelet count to pretreatment value after end of heparin therapy; exclusion of other causes of thrombocytopenia. Blood samples were taken in all patients for 5 to 6 weeks, starting immediately after the end of heparin treatment, defined as day 1. Blood was collected from all patients into plastic tubes containing kaolin and was centrifuged within 30 minutes at 1800g for 20 minutes at room temperature. Serum was aliquotted, shock frozen in liquid nitrogen, and stored at 80 C. The study was approved by the ethics committee of the University of Heidelberg, and informed consent was obtained from each person. Fluorescence-linked immunofiltration assay. The heparininduced IgG antibody was determined by fluorescence-labeled heparin in the presence of PF4 and the patient s serum. The assay was performed under previously reported experimental conditions C-labeled SRA. The biologic effect of the patient s heparin-induced IgG antibody was determined by using 14 C- labeled SRA. The assay was performed as previously reported. 3 The same positive and negative controls were run in all experiments. RESULTS Eight patients were men and 6 were women, with 69 years of age (range 55 to 79) and 55 to 79 kg body weight as indications for heparinization. Thirteen received unfractionated heparin, and 1 patient received low-molecular-weight heparin for prophylaxis of thromboembolism, caused by cardiovascular (n = 1), neurologic (n = 2), or gynecologic disease (n = 3); by carcinoma (n = 2); or as a postoperative complication (n = 6). Platelets decreased from 234 (179 to 325) 10 6 /ml to 43 (2 to 123) 10 6 /ml between days 5 and 8 during heparin treatment and normalized to 313 (100 to 643) 10 6 /ml in all patients within 9 days of stopping heparin. Eight patients developed deep venous thrombosis, 3 developed pulmonary embolism, and 1 each developed arterial thrombosis, myocardial infarction, and cutaneous necrosis during heparin therapy. After diagnosis, anticoagulation was switched to danaparoid (n = 2) and r-hirudin (n = 12). Figs 1 thorough 4 show the different time courses of the platelet count, the FLIFA, and the 14 C-labeled SRA after heparin was stopped. Fig 1 shows the results in 1 patient who had the highest HIT-IgG antibody antigen and activity levels at the time of the platelet nadir. The platelets normalized within 6 days, and antigen and biologic assays normalized at the same rate in 10 days. Fig 2 shows the results in another patient in whom both assays were positive at the time of the platelet nadir. The IgG antibody concentration in this patient increased over 12 days, while the platelet count and biologic assay normalized. Fig 3 shows the results in a patient in whom heparin-induced IgG antibody appeared at day 7 while the patient still had thrombo-

3 410 Harenberg et al J Lab Clin Med June 2001 Fig 2. The time courses (pattern 2) of the FLIFA, the 14 C-labeled SRA, and the platelet count are shown in a Fig 3. The time courses (pattern 3) of the FLIFA, the 14 C-labeled SRA, and the platelet count are shown in a cytopenia and with negative 14 C-labeled SRA during the observation period. Fig 4 shows the results in a patient in the reverse situation: with negative IgG antibody but positive 14 C-labeled SRA at day 6. The time courses of the assays in Figs 1 through 4 indicate that laboratory confirmation of HIT was still able to be obtained in all 4 patients, even several days after heparin was stopped and after normalization of the platelet count. The time courses of the platelets and the two assays in the other 10 patients followed one of the four patterns presented. The sensitivity for laboratory confirmation of HIT was 80% for both the 14 C-labeled SRA and FLIFA assay at day 1 and increased to 85% at day 1, if a positive result in at least one of the assays was defined as laboratory confirmation of HIT (Fig 5). However, originally both assays were always performed on the same day. The result was not dependent on which assay was performed first on day 1. Between days 2 and 7, 1 patient became positive at day 7 in the 14 C-labeled SRA and 2 patients became positive at days 4 and 7 in the FLIFA. Combining the positive results in the two assays increased the sensitivity of the laboratory confirmation of HIT to 100% of patients over 7 days. All patients had negative HIT-IgG antigen and biologic assays 5 to 6 weeks after stopping heparin (Fig 6). DISCUSSION Investigations to combine the PF4/heparin ELISA with the heparin-induced platelet activation assay or with the platelet aggregation assay have been con-

4 J Lab Clin Med Volume 137, Number 6 Harenberg et al 411 Fig 4. The time courses (pattern 4) of the FLIFA, the 14 C-labeled SRA, and the platelet count are shown in a Fig 5. The positive 14 C-labeled SRA and FLIFA (%) are depicted over time in patients with HIT after cessation of heparin treatment. ducted only in HIT without thrombosis. 16 The present study confirmed reports that antigenic and biologic assays for laboratory confirmation of HIT with thrombosis are positive in about 80% of patients. After withdrawal of heparin, however, and by repeated testing for heparin-induced antibodies up to day 7, combining the results of antigenic and biologic assays increased sensitivity to 100%. The time courses of the assays in our HIT patients demonstrate that a positive result in one assay indeed confirms the laboratory diagnosis of HIT. Positive results in one of two assays can be combined if the assays used have high specificity. The two assays used in the present study both have a specificity of above 95%. 3,9 To reduce the probability of a falsepositive result, positive findings for heparin-induced antibodies had to be determined twice on 2 consecutive days. When conventional assays are used, clinical HIT with thrombocytopenia and thromboembolic complications will not be confirmed by laboratory testing in 10% to 20% of patients. 2,3 An inappropriate decision to continue the heparin in these patients might be made. It remains difficult to diagnose HIT type II based on clinical findings, because patients at greatest risk for HIT are severely ill with many other possible causes for thrombocytopenia or micro- or macrovascular thrombotic events. Because an ideal assay for HIT has not yet been established, a different strategy for diagnosis is necessary. Previously 17 an increase in free antibodies was described in immune complex diseases. On the first

5 412 Harenberg et al J Lab Clin Med June 2001 Fig 6. Cumulative data on positive 14 C-labeled SRA and FLIFA results (%) are depicted over the time in patients with HIT after cessation of heparin treatment. exposure to the antigen, the host develops the antibody response after between a week and 10 days. The synthesized antibodies result in formation of antigenantibody complexes, which facilitate removal of the antigen in general by cells in the reticuloendothelial system. The fate of the disappearance of preformed complexes injected into animals has been described in detail. 19 The antigen disappears from the circulation in three phases: the first represents equilibration of the antigen between the intra- and extravascular compartments; the second is produced by catabolism of the antigen; the third involves the immune clearance of the antigen by newly formed specific antibody. 19 As more antibodies are synthesized, the lattice structure of the immune complexes increases until the complexes reach a critical size and then are rapidly removed from the circulation by the reticuloendothelial system. The complexes are cleared within 7 days, and free antibody can be detected. 16,19 For the results of the present study, this mechanism is believed to be of importance in HIT patients, because all negative results of the antigenic tests in HIT patients became positive in the following 7 days, and most were also positive in the biologic 14 C- labeled SRA after clearance of the antigen. The cumulative frequency of positive antigen and biologic assays increased over 7 days while platelets normalized during this period, indicating termination of the immunologic process of the disease. This complies with the mechanisms described above for immune complex diseases, where the pathogenic process ceases after clearance of the antigen-antibody complexes and free antigen can be detected. The reason for the more rapid decrease in the normalization of the heparin-induced IgG antibody in these patients remains unclear at present as compared with the decrease in IgG antibodies in other immunologic diseases. The present findings also agree with the observation that patients who were tested only on the day when heparin was stopped should be considered still to have a substantial risk of developing thrombotic complication. 6 So far, in contrast to the platelet count, the antigen and biologic assays for laboratory confirmation of HIT have not been followed up in a larger number of patients. Thus both follow-up of 1 week after onset of the disease and the combination of two sensitive assays may improve the laboratory confirmation of HIT. Follow-up in our HIT patients revealed four patterns for heparin-induced antibodies. Pattern 1 is characterized by positive antigenic tests at the platelets nadir and similar time courses for the normalization of the antigen and biologic assays, while pattern 2 shows dissimilar time curves for the assays. Patterns 3 and 4 are positive in only one of the two assays. There are many possible reasons for the differences between the antigenic and biologic assay results, including the polyclonal structure of the IgG antibody and the possibility of interleukin-8 or neutrophil activating peptide as antigens. In conclusion, repeated determination of sensitive heparin-induced IgG antibody and 14 C-labeled SRAs confirmed the diagnosis in all HIT patients. This may be caused by an interruption of the antigen-antibody complex after elimination of the antigen, leading to an increase in free antibody that can then be detected in antigenic and biologic assays for laboratory confirma-

6 J Lab Clin Med Volume 137, Number 6 Harenberg et al 413 tion of HIT. Although the FLIFA, rather than a more standard ELISA, was chosen for immunologic detection of the antibody, it is reasonable to assume that the findings are applicable to the more widely used ELISA. We thank Dr St. de Looze for manuscript preparation and Mrs Ch. Giese and A. Hagedorn for technical assistance. REFERENCES 1. Chong BH, Burgess J, Ismail F. The clinical usefulness of the platelet aggregation tests for the diagnosis of heparininduced thrombocytopenia. Thromb Haemost 1993;69: Kelton JG, Levine MN. Heparin-induced thrombocytopenia. Semin Thromb Hemost 1986;12: Sheridan D, Carter C, Kelton JG. A diagnostic test for heparin-induced thrombocytopenia. Blood 1986;67: Greinacher A, Michels I, Kiefel V, Mueller-Eckhardt CA. Rapid and sensitive test for diagnosing heparin-associated thrombocytopenia. Thromb Haemost 1991;66: Tomer A. A sensitive and specific functional flow cytometric assay for the diagnosis of heparin-induced thrombocytopenia. Br J Haematol 1997;98: Amiral J, Bridley F, Dreyfus M. Platelet factor 4 complexed to heparin is the target for antibodies generated in heparininduced thrombocytopenia. Thromb Haemost 1992;68: Newman PM, Swanson RL, Chong BH. Heparin-induced thrombocytopenia: IgG binding to PF4-heparin complexes in the fluid phase and cross-reactivity with low-molecularweight heparin and heparinoid. Thromb Haemost 1998; 80: Wang LC, Huhle G, Malsch R, Hoffmann U, Song XH, Harenberg J. Determination of heparin-induced IgG antibody in heparin-induced thrombocytopenia type II. Eur J Clin Invest 1999;29: Wang LC, Huhle G, Malsch R, Hoffmann U, Song XH, Harenberg J. Determination of heparin-induced IgG antibody by fluorescence-linked immunofiltration assay (FLIFA). J Immunol Methods 1999;222: Eichler P, Budde U, Haas S, Kroll H, Loreth RM, Meyer O, et al. First workshop for detection of heparin-induced antibodies: validation of the heparin-induced platelet-activation test (HIPA) in comparison with a PF4/heparin ELISA. Thromb Haemost 1999;81: Pouplard C, Amiral J, Borg JY, Laporte-Simitsidis S, Delahousse B, Gruel Y. Decision analysis for use of platelet aggregation test, carbon 14-serotonin release assay, and heparin-factor 4 enzyme-linked immunosorbent assay for diagnosis of heparin-induced thrombocytopenia. Am J Clin Pathol 1999;11: Look KA, Sahud M, Flaherty S, Zehnder JL. Heparininduced platelet aggregation vs platelet factor 4 enzymelinked immunosorbent assay in the diagnosis of heparininduced thrombocytopenia-thrombosis. Am J Clin Pathol 1997;108: Greinacher A, Amiral J, Dummel V, Vissac A, Kiefel V, Mueller-Eckhardt C. Laboratory diagnosis of heparinassociated thrombocytopenia and comparison of platelet aggregation test, heparin-induced platelet activation test, and platelet factor 4/heparin enzyme-linked immunosorbent assay. Transfusion 1994;34: Arepally G, Reynolds C, Tomaski A, Amiral J, Jawad A, Poncz M, et al. Comparison of PF4/heparin ELISA assay with the 14 C-serotonin release assay in the diagnosis of heparin-induced thrombocytopenia. Am J Clin Pathol 1995; 104: Kelton JG, Sheridan D, Brain H, Powers PJ, Turpie AG, Carter CJ. Clinical usefulness of testing for a heparindependent platelet-aggregating factor in patients with suspected heparin-associated thrombocytopenia. J Lab Clin Med 1984;103: Walenga JM, Jeske WP, Wallis DE, Bakhos M, Lewis BE, Leya F, et al. Clinical experience with combined treatment of thrombin inhibitors and GPIIb/IIIa inhibitors in patients with HIT. Semin Thromb Hemost 1999;25(suppl I): Frank MM, Lawley TJ, Hamburger MI, Browh EJ. Immunoglobulin G-Fc receptor mediated clearance in autoimmune disease. Ann Intern Med 1983;98: Cadroy Y, Amiral J, Raynaud P, Mazaleyrat A, Sauer M, Sie P. Evolution of antibodies anti-pf4/heparin in a patient with a history of heparin-induced thrombocytopenia reexposed to heparin. Thromb Haemost 1994;72: Mannik M, Arend WP. Fate of performed immune complexes in rabbits and rhesus monkeys. J Exp Med 1971;134:19-28.