Abnormalities of Platelet Function in Patients with Polycythemia Vera1

[CANCER RESEARCH 33, 2683 2687, November 1973] Abnormalities of Platelet Function in Patients with Polycythemia Vera1

Sam Berger, Louis M. Aledort, Harriet S. Gilbert, John P. Hanson, and Louis R. Wasserman Division oj Hentalology, Department oj Medicine, Mount Sinai School of Medicine of the City University of New York, New York, New York 10029

SUMMARY function abnormalities in polycythemia vera and other myeloproliferative disorders. Little, however, has been The exact mechanism(s) of thrombosis and hemorrhage published with respect to platelet aggregation in these in polycythemia vera has not yet been delineated. Platelet disorders. Spaet et al. (26) recently reported 3 patients with function studies were performed by examining platelet essential thrombocythemia who had marked and character aggregation in 47 polycythemia vera patients (62 studies) istic abnormalities in platelet aggregation. These consisted with low ("spent"), normal, or elevated hematocrits and of absent epinephrine-induced aggregation, reduced ADP- with varying platelet counts. Adenosine diphosphate, epi- induced aggregation but normal collagen-induced aggrega nephrine, collagen, and thrombin were used as aggregating tion. In 1 of these patients, the aggregation abnormalities agents. Eighty-one % of the 62 studies were abnormal. The improved following therapy. TangÃ¼n(27)has studied plate frequency of abnormal tests increased from those with let aggregation in 33 patients with myeloproliferative disor normal hematocrits and platelet counts (75%) through to ders, 6 of whom had polycythemia vera. Abnormal response those with elevated hematocrits and platelet counts (100%). to aggregating agents occurs in 75% of all cases and in 4 of 6 One hundred % of the studies in spent polycythemia vera with polycythemia vera. Of the latter, 3 studies were were abnormal. Abnormal bleeding times and throm- performed prior to therapy (busulfan) and 1 while on no bohemorrhagic complications did not correlate with each therapy but not in remission. Two patients in hematological other or with abnormal aggregation, hematocrits, or plate remission had normal aggregation studies. Repeat studies, let counts. Repeat studies following therapy showed im performed on 2 patients after (busulfan) therapy showed provement in 10 of 13 patients. The incidence of abnormal improvement but not complete normalization. In the entire aggregation in polycythemia vera is high; however, the group of patients, the effect of myelosuppressive therapy on significance of this finding in the pathogenesis of thrombosis platelet function was variable, and the reduction of a high and hemorrhage remains obscure. platelet count to normal following therapy did not eradicate the platelet function abnormalities. The results of platelet aggregation studies carried out on INTRODUCTION 40 patients with polycythemia vera in various stages of the disease are the subject of the present report. Of these, 18 Thrombosis and hemorrhage are the most frequent causes patients were studied both before and after therapy. of morbidity and mortality in polycythemia vera. Throm- botic and hemorrhagic complications occur in 26 to 63% and 16to 35% of patients, respectively, and are the causes of MATERIALS AND METHODS death in 20 to 40% and 6 to 30%, respectively (6, 29, 30). In Subjects. Sixty-two platelet aggregation studies were addition, surgical procedures are accompanied by excessive performed on 47 patients with polycythemia vera. Twenty- bleeding and thrombosis (2, 9, 24, 32). The incidence of three of the patients were male and 24 were female with an complications in 1 series was almost 3 times as great in age range of 8 to 79 years (2 patients under 30 years of age). uncontrolled as controlled patients, and the longer the Care was taken to ensure that the patients had not recently period of effective control, the fewer were the complications ingested any of the drugs known to affect platelet function. (31). A history of clinically apparent hemorrhage or throm The tendency to hemorrhage and thrombosis in polycy bosis was obtained in 15 and 30% of patients, respectively, themia vera has been ascribed to abnormalities of the and both complications occurred together in only 2 patients vasculature, coagulation factors, and platelets. The exact (4.3%). mechanism(s) is, however, unknown and the relative impor Bleeding Times. Bleeding times were performed by a tance of these 3 factors is not agreed upon. modified Ivy method (13) (normal = 3 to 9 min). Venous Recently, interest has turned to the search for platelet blood was collected by gravity into plastic centrifuge tubes 'Supported in part by USPHS Grants HE-10905, HE-12443, and containing 4% sodium citrate with 1 part citrate to 9 parts HE-05488 from the National Heart and Lung Institute and CA-11816 and blood. Platelets were counted by phase microscopy (4). To CA-10728 from the National Cancer Institute; also by the Albert A. List, obtain PRP2 the anticoagulated blood was centrifuged at 4Â° Frederick Machlin, Anna Ruth Lowenberg, and Helen B. Kuyler Funds. Received February I, 1973; accepted July 23, 1973. 2The abbreviation used is: PRP, platelet-rich plasma.

NOVEMBER 1973 2683

Downloaded from cancerres.aacrjournals.org on September 23, 2021. © 1973 American Association for Cancer Research. Berger, Aledon, Gilbert, Hanson, and Wasserman for 20 min at 900 rpm on Model PR-2 International tant elevation of platelet count (Chart 1) and the greatest Centrifuge. Platelet-poor plasma was obtained by centrifu- number of abnormal results (80%) were obtained in the gation for 15 min at 3500 rpm. Siliconized and plastic group (10 patients) with both high hematocrits and platelet equipment was used throughout. count-s. Platelet Aggregation Studies. Platelet aggregation studies Epinephrine-induced Aggregation. Epinephrine-induced were performed by a turbidimetric method (3) using a aggregation was abnormal in 52% of the studies, but in Chrono-Log Corporation Aggregometer with a Bausch and contrast with ADP-induced aggregation, there was essen Lomb recorder. Studies were done at 37Â°on stirred PRP tially no difference in the percentage of abnormal studies in adjusted to a final platelet count of 300,000 per cu mm with the different groups with normal, elevated, or low hemato autologous platelet-poor plasma within 1 hr of blood crits. In contrast to ADP-induced aggregation, an abnormal collection. ADP disodium salt (Sigma Chemical Company, primary wave occurred almost twice as frequently as an St. Louis, Mo.) was stored as an 0.01 M stock solution at absent secondary wave. As seen in Chart 1, within the -20Â° and used, after dilution, at a final concentration of elevated hematocrit group, there was a higher percentage of 2 x 10~6 M. Topical bovine thrombin (Parke, Davis and abnormal studies in those with concomitant elevation of Company, Detroit, Mich.) was stored as a stock solution of platelet count. In the normal hematocrit group, there was 100 units/ml at -20Â°. It was diluted immediately before use no significant difference. Similar to ADP-induced aggrega to a final concentration of 0.2 units of PRP per ml. tion, the highest percentage of abnormal studies occurred in Epinephrine HCL (4.6 x 10-3 M; Parke, Davis and the group with both elevated hematocrits and platelet Company) was kept at 4Â°and was used after dilution at a counts. final concentration of 5 x 10~5 M. Human collagen Collagen-induced Aggregation. Collagen-induced aggre prepared from dried collagen (Sigma Chemical Company) gation was abnormal in 37% of the studies. The patients was used at a final suspension containing 0.72 mg protein with elevated hematocrits had more than twice the number per ml PRP. By using 0.4% sodium citrate it is not necessary of abnormal studies than those with normal hematocrits. to correct for the hematocrit. Multiple studies were carried One hundred % of the studies in the spent group were out varying the 0.4% citrate concentration without any abnormal. In the remainder of the patients, those with high significant difference in results. platelet counts and high hematocrits had the highest percentage of abnormal studies (Chart 1). Thrombin-induced Aggregation. Thrombin-induced ag RESULTS gregation was abnormal in 37% of the 54 times it was performed. Those with spent polycythemia and those with Hematocrit. Of the 62 studies, 38 were done at normal elevated hematocrits had abnormal tests in approximately (<50%) hematocrits, 19 at elevated hematocrits, and 5 at low hematocrits (<37%) in patients with "spent" polycy- 50% of cases. Elevation of the platelet count did not increase the risk of abnormal studies within each hematocrit group themia. (Chart 1). Platelet Count. The platelet count was normal (150 to Absence of Aggregation in Response to Aggregating 350,000 per cu mm) during 41 studies and elevated in the Agents. Absence of aggregation occurred most commonly remaining 21 studies. The degree of elevation was generally with epinephrine followed in decreasing order by collagen, modest, being over 1,000,000 per cu mm (1,600,000) in only ADP, and thrombin in 8, 5, 4, and 1 studies, respectively. 1 patient. The mean platelet count for the 62 studies was Four studies showed an absent response to 2 aggregating 374,000 perÃ§umm. Platelet Aggregation Studies. Results expressed are in units on the Bausch and Lomb recording paper and are a AGGREGATION-SU M MARY measure of light transmittance. Normal values for this laboratory are as follows (23); ADP, 5.2 Â± 1.0 units 100 ADP M Thrombin M Collagen (primary aggregation, mean Â±1 S.D.; epinephrine, 2.9 Â± 80- Epinephrine 0.6 units (primary aggregation; collagen, 6.4 Â±0.5 units; QC O and thrombin, 4.6 Â±0.7 units. Results were considered abnormal if they were less than the mean - 2 S.D.'s. 1 60- Secondary waves of aggregation were recorded as normal if present and abnormal if absent. ADP and epinephrine 40 studies were considered abnormal if either the primary or the secondary wave of aggregation was abnormal. ADP-induced Aggregation. Forty-seven % of the 62 20 studies were abnormal. The group with elevated hematocrits 0 had a greater percentage of abnormalities (74%) than those N.HCT N.HCT. t HCT. I HCT SPENT with normal (32%) or low (60%) hematocrits. Primary wave N.PLT. t PLT. N.PLT I PLT. abnormalities occurred approximately as frequently as (27) (II) (9) (10) (5) secondary wave abnormalities. More abnormalities were Chart I. Numbers in parentheses, number of studies performed. HCT, present within a hematocrit group if there was a concomi hematocrit; PLT, platelet count; N, normal; ], increased.

2684 CANCER RESEARCH VOL. 33

Downloaded from cancerres.aacrjournals.org on September 23, 2021. © 1973 American Association for Cancer Research. Platelet Function Abnormalities in Polycythemia Vera agents. There was no relationship, however, between ab compared to the group as a whole, there was no significant sence of response and platelet count, elevation of bleeding difference between the number with elevated platelet counts, time, or history of thrombohemorrhagic phenomena. In elevated hematocrits, and abnormal aggregation studies. contrast to this, some aggregation responses were increased, Response to Therapy (Table 1). Thirteen patients had i.e., they showed maximal deflections of the absorbance repeat aggregation studies. Five had tests done first at an curves greater than 2 S.D.'s above the mean. Again, how elevated hematocrit and again, following treatment, at a ever, there was no relationship to platelet count, bleeding normal hematocrit. Two, who originally had abnormal time, or incidence of thrombohemorrhagic phenomena. studies, completely normalized after treatment that con Platelet Aggregation: Summary of Findings. The most sisted of phlebotomies only in one and chemotherapy in the frequent abnormalities occurred with epinephrine (52%) and other. The remaining 3 patients, treated with various ADP (47%), followed by collagen and thrombin (37% each). combinations of phlebotomy and chemotherapy, had their In general, patients with elevated hemactocrits had a greater aggregation studies return toward normal. Two had absent percentage of abnormal studies than did those with spent epinephrine-induced aggregation prior to therapy which polycythemia, who in turn had more abnormalities than improved after therapy. One had absent collagen-induced those with normal hematocrits. The commonest abnormal aggregation which became normal. ity was a decrease in aggregation, followed by a complete Two patients were studied first at normal hematocrits and absence of the secondary wave of aggregation. then following p.o. iron therapy at elevated hematocrits. In Considering all the aggregating agents collectively, 81% one, the studies normalized, but in the other, they became of the studies were abnormal. No patient with either spent abnormal. polycythemia or an increased hematocrit plus platelet count A total of 6 patients had repeat studies performed before had a completely normal study. Eighty-nine % of those with and after therapy but with hematocrits that remained in the an increased hematocrit but normal platelet count had same range. One of these 6, in the spent phase of the disease abnormal studies. Approximately 75% of the patients with and on no treatment, showed no change in platelet aggrega normal hematocrits had abnormal studies whether or not tion. Four of the remaining 5 showed improvement, includ the platelet count was abnormal. ing 1 treated with phlebotomies only for 7 months. The 5th Bleeding Time. The Ivy bleeding time was prolonged patient's studies were unchanged 4 months after therapy (greater than 9 min) in 10of the 49 studies. In only 1was the with phlebotomies and chemotherapy. platelet count elevated (496,000 per cu mm). None of the patients had a platelet count below 100,000 per eu m. All 4 patients with spent polycythemia had prolonged bleeding DISCUSSION AND CONCLUSIONS times when studied. Their hematocrits and platelet counts ranged from 21 to 35% and 110,000 to 298,000 per cu mm, The platelet has been incriminated in the throm respectively. The hematocrit was not elevated during any bohemorrhagic complications of polycythemia vera with study with a prolonged bleeding time. Two of the patients respect to both increased numbers and abnormal function. had hemorrhagic complications and 1 had a history of An elevated platelet count is a common occurrence in thrombosis. Four of the 10 studies showed completely polycythemia vera and is a prerequisite of essential throm- normal platelet aggregation. Thus abnormal bleeding times bocythemia, another myeloproliferative disorder. Some in our patients with polycythemia occurred much less have found no influence (31), while others have reported frequently than platelet aggregation abnormalities and did significantly increased complications rates in patients with not correlate with an elevation of hematocrit or platelet elevated platelet counts in polycythemia vera (8, 17), count or a history of thrombohemorrhagic complications. secondary thrombocytosis (12, 17), and thrombocythemia Thrombohemorrhagic Complications. Thrombotic and (10, 22). Our data showed no relationship between elevated hemorrhagic complications occurred in 30 and 15% of the platelet counts and complication rate. However, the platelet patients, respectively. When these groups of patients were counts in our patients, as is usual in polycythemia vera (32),

Table 1 Response lo therapy

Change in studiesTherapyP, hematocrit ofpatients52411Aggregation aftertherapyI

toN-N CIron3ZP, to1N andN1 P,CP, and1SpentNo. CNoneNormalized21Improved34Same11Worse1

Total 13 3 1I, increased; N, normal; P, phlebotomy: C, chemotherapy.

NOVEMBER 1973 2685

Downloaded from cancerres.aacrjournals.org on September 23, 2021. © 1973 American Association for Cancer Research. Berger, Aledort, Gilbert, Hanson, and Wasserman were only mildly to moderately elevated ( < 1,000,000 per cu human plasma has been noted to have an inhibitor to mm) except in 1 patient (1,600,000 per cu mm). The latter connective tissue-induced aggregation which may play a had no thrombohemorrhagic complications. This discrep role in maintaining normal hemostasis (20). Polycythemia ancy may be due to the fact that over 90% of our patients vera patients may have more of this or other types of were studied as outpatients at a time when they were not inhibitors than the normal population, perhaps lacking an suffering from complications. It is possible that our find unknown required plasmatic factor. Peripheral platelets ings relating platelet count (and function) with complica represent a heterogeneous population, the young platelets tion rate would have been significant were we able to study being more active metnbolically and the more responsive to these patients immediately prior to and/or during a throm aggregating agents (14). Recently, emphasis has been bohemorrhagic event. placed on the study of platelet kinetics in the myeloprolifer Qualitative platelet abnormalities of several types have ative disorders. In polycythemia vera, platelet life span been reported in polycythemia vera and other myeloprolif- appears to be slightly decreased and production increased erative disorders. Variable results have been reported for (5, 15). These data point up the dynamics of maintaining a several platelet function tests in patients with polycythemia peripheral platelet count in this disorder. In myeloprolifera vera. In vitro platelet adhesiveness studies have shown both tive disorders, platelet function abnormalities may thus increased (25) and decreased (17) values but the test is represent a lack of heterogeneity of platelets in the periph difficult to standardize and involves several variables includ eral blood. Although their morphology in polycythemia ing the content of the red blood cells in the sample (11), the vera is usually normal, the circulating platelets may in fact hematocrit (25), the volume of anticoagulant (25), but be functionally less active. Further study of the fate of the apparently not the platelet count (7, 17, 25). more rapidly turning over platelet may be helpful in Recently, much attention has turned to the in vitro clarifying this. measurement of another platelet function, platelet aggrega The fact that therapy, irrespective of type, improved tion (1, 3, 18). Abnormalities have been described in the platelet aggregation in 10 of 13 patients suggests that myeloproliferative disorders (7, 16, 19, 26, 27) but there restoration toward a "more normal state" brings forth a have been no reports of platelet aggregation in a large group significant change such that they respond in the tests in a of patients with polycythemia vera and only minimal data more normal fashion. on the effect of therapy. Our patients exhibited a large number of abnormalities regardless of whether they were "controlled" or "uncontrolled." Multiple aggregation ab REFERENCES normalities were more common than single in any given study. In general, the number of abnormalities increased 1. Aledort, L. M. Platelet Aggregation. In: S. A. Johnson (ed.). The from the group with normal hematocrit and platelet count Circulating Platelet, Chap. 9, pp. 218-240. New York: Academic through to the group with elevated hematocrit and platelet Press, 1971. count. The latter, as well as the spent group, showed 2. Birge, H. L. Complications in Surgery, due to Abnormalities in Blood Volume, with Special Reference to the Syndrome of Polycythemia. abnormal platelet aggregation studies 100% of the time. Virginia Med. Monthly,Â«/: 161 165, 1954. Prolonged Ivy bleeding times and thrombohemorrhagic 3. Born, G. V. R. Aggregation of Blood Platelets by Adenosine complications were not related to each other or to the other Diphosphate and Its Reversal. Nature, Â¡94:927-929, 1962. parameters measured. 4. Brecher, G., Schneiderman, M., and Cronkite, E. O. The Reproduci- Repeat aggregation studies in patients whose hematocrits bility and Constancy of the Platelet Count. Am. J. Clin. Pathol., 23: normalized following therapy showed improvement. Simi 15 26, 1953. larly, improvement also occurred in other patients in whom 5. Brodsky, I., Kahn, S. B., Ross, E. M., and Petkov, G. Platelet and therapy did not alter the hematocrit. The findings were not Fibrinogen Kinetics in the Chronic Myeloproliferative Disorders. influenced by the mode of therapy (myelosuppressive ther Cancer, 30: 1444-1450, 1972. 6. Chievitz, E., and Thiede, T. Complications and Causes of Death in apy versus phlebotomy). Polycythemia Vera. Acta M^ed.Scand., 772: 513-523, 1962. These studies further add to the information gathered 7. Cronberg, S., Nilsson, I. M., and Gydell, K. Hemorrhagic Throm- regarding hemostasis in polycythemia vera. There are bocythemia due to Defect in Platelet Adhesiveness. Scand. J. Hae- several possible explanations both for what appears to be a matol., 2: 208-219, 1965. discrepancy between the findings in vitro and the clinical 8. Dawson, A. A., and Ogston, D. Influence of Platelet Count on the histories, and also for the possible mechanisms producing Incidence ol Thrombotic and Hemorrhagic Complications in Polycy abnormal platelet function. It is possible, first, that these themia Vera. Postgraduate Med. J., 46: 76-78, 1970. measurements in vitro have in fact little relation to the in 9. Fitts, W. T., Jr., Erde, A., Peskin, G. W., and Frost, J. W. Surgical vivo situation. Similarly, it is now well recognized that in Complications of Polycythemia Vera. Ann. Surg., 152: 548-558, I960. other conditions, for example, acetylsalicylic acid-induced 10. Gunz, F. W. Hemorrhagic Thrombocythemia: A Critical Review. platelet function abnormality, there exists a discrepancy Blood, /5: 706 723, 1960. 11. Hellem, A. J. The Adhesiveness of Human Blood Platelets in Vitro. between the high percentage of abnormalities in vitro and Scand. J. Clin. Lab. Invest., 12 (Suppl. 51).- 1 117, 1960. the very few clinical complications in otherwise hemostati- 12. Hirsh, J., McBride, J. A., and Dacie, J. V. Thromboembolism and cally normal persons (28, 33). Increased Platelet Adhesiveness in Post-splenectomy Thrombocytosis. The aggregation abnormalities may be due to platelets Australian Ann. Med. 15: 122 128, 1966. that are refractory to aggregating agents (21). Recently, 13. Ivy, A. C., Shapiro, P. F., and Melnick, P. The Bleeding Tendency in

2686 CANCER RESEARCH VOL. 33

Jaundice. Surg. Gynecol. Obstet., 60: 781-784, 1935. 23. Puszkin, E., Gutfreund, D., and Aledort, L. M. Hemostatic Abnor 14. Karpatkin, S. Heterogeneity of Human Platelets, Metabolic and malities in Heroin Addicts. Transfusion, 12: 9-11, 1972. Kinetic Evidence Suggestive of Young and Old Platelets. Ser. 24. Rigby, P. G., and Leavell, B. S. Polycythemia Vera: A Review of Fifty Haematol., 4: 75-97, 1971. Cases with Emphasis on the Risk of Surgery. Arch. Intern. Med., 706: 15. Kutti, J., and Weinfeld, A. Platelet Survival in Active Polycythaemia 622-627, I960. Vera with Reference to the Haematocrit Level. Scand. J. Haematol., 25. Shield, L. K., and Pearn, J. H. Platelet Adhesiveness in Polycythemia 8: 405 414, 1971. Rubra Vera. Med. J. Australia, /. 711-715, 1969. 16. Lumley, S. E., and Powell, B. W. Essential Thrombocythemia in 26. Spaet, T. H., Lejnieks, I., Gaynor, E., and Goldstein, M. L. Defec Childhood. Proc. Roy. Soc. Med., 64: 22-23, 1971. tive Platelets in Essential Thrombocythemia. Arch. Intern. Med.. 17. McClure, P. D., Ingram, G. I. C., Stacey, R. S., Glass, V. H., and 124: 135 141, 1969. Matchett, M. O. Platelet Function Tests in Thrombocythemia and 27. TangÃ¼n,Y. Platelet Aggregation and Platelet Factor 3 Activity in Thrombocytosis. Brit. J. Haematol., 12: 478 498, 1966. Myeloproliferative Disorders. Throm. Diath. Haemorrhg., 25: 18. Mustard, J. F., and Packham, M. A. Factors Influencing Platelet 241-251, 1971. Function: Adhesion, Release, and Aggregation. Pharmacol. Rev., 22: 28. Van-Casteels, D. M., and De Gaetano, G. Purpura and Acetylsali- 97 187, 1971. cylic Acid Therapy. Acta Paediat. Scand., 60: 203 208, 1971. 19. Neemeh, J. A., Bowie, E. J. W., Thompson, J. H., Didisheim, P., and 29. Videbaek, A. Polycythemia Vera. Course and Prognosis. Acta Med. Owen. C. A. Quantitation of Platelet Aggregation in Myeloprolifera- Scand., 138: 179-187, 1950. tive Disorders. Am. J. Clin. Pathol., 57: 336 346, 1972. 30. Wasserman. L. R., and BÃ¤ssen,F.Polycythemia. J. Ml. Sinai Hosp., 20. Nossel, H. L., Wilner, G. D., and Drillings, M. Inhibition of N. Y., 26: 1-49, 1959. Collagen-induced Platelet Aggregation by Normal Plasma. J. Clin. 31. Wasserman, L. R., and Gilbert, H. S. Surgical Bleeding in Polycy Invest., 50: 2168-2175, 1971. themia Vera. Ann. N. Y. Acad. Sci., 115: 122-138, 1964. 21. O'Brien, J. R. A Comparison of Platelet Aggregation Produced by 32. Wasserman, L. R., and Gilbert, H. S. Complications of Polycythemia Seven Compounds and a Comparison of Their Inhibitors. J. Clin. Vera. Seminars Hematol., 3: 199-208, 1966. Pathol., 17: 275-281, 1964. 33. Weiss, H. J.. Aledort, L. M., and Kochwa, S. The Effect of Salicylates 22. Ozer, F. L., Traux, W. E., Miesch, D. C., and Levin, W. C. Primary on the Hemostatic Properties of Platelets in Man. J. Clin. Invest., 47: Hemorrhagic Thrombocythemia. Am. J. Med., 28: 807 823, 1960. 2169-2180, 1968.

NOVEMBER 1973 2687

Sam Berger, Louis M. Aledort, Harriet S. Gilbert, et al.

Cancer Res 1973;33:2683-2687.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/33/11/2683

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/33/11/2683. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.