The purpose of this study was to evaluate retrospectively the incidence and severity of heparin-induced thrombocytopenia (HIT)-related complications in patients undergoing cardiopulmonary bypass. We reviewed the records of 1,500 consecutive patients who underwent cardiopulmonary bypass between August 1987 and December 1991 at Thomas Jefferson University Hospital. During this period of time, there were 1,155 coronary artery bypass graft operations (77 percent); 225 valve replacements and repairs, or both (15 percent); 60 combination coronary artery bypass graft or valve operations, or both (4 percent); and 60 miscellaneous procedures (4 percent). Although not all patients with postoperative complications were tested for the HIT antibody, 11 patients (0.75 percent) were diagnosed with HIT. There were 17 complications in these 5 men and 6 women including 6 cases of ischemic limbs which required amputation, 4 strokes, 2 instances of saphenous vein graft occlusion with resulting myocardial infarction, 2 cases of pulmonary emboli, 1 case of phlegmasia cerulea dolens, and 2 deaths. The complications occurred an average of 3.6 days postoperatively, with a range of occurrence of 1 to 11 days postoperatively. The mean nadir platelet count at the time of recognition was 123,000/[mm.sup.3] (range 32,000 to 193,000/[mm.sup,3]) with 9 of 11 patients (81.8 percent) having counts greater than 100,000/[mm.sup.3]. There was, however, a mean percent decrease in the platelet count of 50 percent (range, 31 to 75 percent) from the time of first exposure to heparin to the time of recognition of HIT. In our patients, HIT was not related to the type, duration of treatment with or amount of heparin, or to pretreatment with aspirin.
(Chest 1993; 104:1436-40)
Since its discovery in 1916 by McLean, heparin has had extensive clinical applications, including the prevention and treatment of venous thrombosis and pulmonary embolism, stroke, myocardial infarction, preventing thrombosis after coronary artery thrombolysis or angioplasty, atrial fibrillation with embolization, peripheral arterial embolization, disseminated intravascular coagulation, fetal growth retardation, and so on. Heparin also has been invaluable in the development and use of extracorporeal devices such as the heart-lung machine, cell-saver devices, hemodialysis, as well as for an expansive list of clinical and experimental laboratory uses.
Heparin-induced thrombocytopenia (HIT) occurs when specific heparin-dependent antibodies attach to platelet membranes, causing platelet activation. The aggregation and degranulation of platelets during this process result in a variety of clinical scenarios ranging from asymptomatic thrombocytopenia to devastating intravascular coagulation.
The purpose of our study was to evaluate the incidence and severity of HIT-related complications in patients undergoing cardiopulmonary bypass. A further review of the literature will be discussed with particular emphasis on early recognition and treatment of patients suspected of having HIT.
Materials and Methods
We conducted a retrospective study of 1,500 consecutive patients undergoing cardiopulmonary bypass between August 1987 and December 1991 at Thomas Jefferson University Hospital to evaluate the morbidity and mortality of such patients when the diagnosis of HIT is made postoperatively. During this time, there was no systematic investigation of platelet counts in postoperative patients. Early in the reported period, a diagnosis of HIT was considered when a thromboembolic complication occurred in association with a decrease in the platelet count. Later, a diagnosis of HIT was considered in any patient with a thromboembolic complication following cardiopulmonary bypass. The diagnostic test used to detect the presence of the anti-platelet antibody was the two-point [C.sup.14]-serotonin platelet release test described by Sheridan et al.
In order to assess normal platelet counts in patients following cardiopulmonary bypass, the platelet counts of 100 additional consecutive patients with no evidence of HIT were studied. Both the nadir and the percent drop in the platelet count were determined.
During the study period, there were 1,155 coronary artery bypass graft operations (77 percent), 225 valve replacements or repairs, or both (15 percent); 60 combination coronary artery bypass graft and valve operations (4 percent); and 60 miscellaneous procedures (4 percent). Cardiopulmonary bypass was performed with clear prime and moderate systemic hypothermia. All patients received bovine lung heparin during cardiopulmonary bypass; however, all prior exposure to heparin, including during cardiac catheterization, was with porcine mucosal heparin. Postoperative heparin exposure, such as in heparin flushes, was also with porcine mucosal heparin.
Eleven of 1,500 patients (0.75 percent) were diagnosed as having HIT, including 5 men (45 percent) and 6 (55 percent) women. All 11 patients had exposore to heparin while on another service prior to cardiopulmonary bypass from a variety of sources, including heparin flushes, subcutaneous heparin, heparin-coated pulmonary artery catheters, and cardiac catheterization. The mean time from initial exposure to heparin to time of cardiopulmonary bypass was 4.5 days, with a range of 3 to 10 days. No patients were suspected of having HIT preoperatively, since there was no significant change in their platelet counts during this time nor were there any thromboembolic events prior to surgery. Six of 11 patients (55 percent) were receiving aspirin immediately prior to surgery and 5 of 11 patients (45 percent) had no recent exposure to aspirin. All 6 patients taking aspirin received a daily dose of 325 mg of aspirin for a minimum of 1 week prior to surgery.
There were 17 complications in these 11 patients, including 6 cases of limb amputation (patients 3, 5, and 6), 4 instances of stroke, 2 cases of saphenous vein graft occlusion with resulting myocardial infarction (patients 3 and 10), 2 instances of pulmonary emboli (patients 2 and 4), 1 case of phlegmasia cerulea dolens (patient 7), and 2 deaths (patients 3 and 6).
In those patients who underwent amputation of an extremity, all had prior intravascular devices present at those sites. Specifically, patients 3 and 6 had an intra-aortic balloon pump in place postoperatively in the affected legs, leading to amputation. Patient 5 had a history of claudication and therefore the intra-aortic balloon pump was placed in a transthoracic position; however, a left femoral arterial line and right femoral venous line were in place postoperatively. Both patients 5 and 6 had radial arterial lines in place in the affected arms, leading to amputation.
The complications occurred at a mean of 3.6 days postoperatively, with a range of 1 to 11 days. Conbined with their preoperative exposure to heparin, complications occurred an average of 7.4 days from time of initial exposure to heparin, with a range of 4 to 14 days. The mean nadir platelet coutit at the time of recognition was 123,000/[mm.sup.3] (range, 32,000 to 193,000/[mm.sup.3]) with 9 of 11 patients (81.8 percent) with platelet counts greater than 100,000/[mm.sup.3]. There was, however, a mean percent decrease in the platelet count of 50 percent (range, 31 to 75 percent) from the time of first exposure to heparin preoperatively to the time of the first postoperative complications and suspicion of type 2 HIT.
The mean nadir platelet count in 100 consecutive normal patients with no evidence of HIT was 207,000/[mm.sup.3] (range, 66,000 to 332,000/[mm.sup.3]), with a mean percent decrease in the platelet count of 26 percent (range, 0 to 70 percent) from the time of first exposure to heparin preoperatively to the time of the nadir postoperative platelet count (Table 1).
[TABULAR DATA OMITTED]
As with virtually all drugs, heparin also is associated with adverse reactions, the most common being hemorrhage. Other reported complications include local irritation from subcutaneous use, skin necrosis after systemic administration, osteoporosis following long-term administration of high doses of heparin, delayed transient alopecia, priapism, rebound hyperlipidemia on discontinuation of heparin therapy, elevations of aminotransferase (serum glutamic oxaloacetic transaminase and serum glutamic pyruvic transaminase) levels, hypoaldosteronism, and hypersensitivity reactions.
One particularly well-studied complication of heparin therapy is thrombocytopenia, with or without thrombosis or thromboembolism. Historically, it was known as early as 1942 that heparin caused a decrease in platelet count in vitro. In 1958, Weismann and Tobin described arterial embolism occurring during systemic heparin therapy. Roberts et al in 1964, reviewed 11 patients who suffered unexplained arterial embolization while being treated with heparin. They wrote: "We have no explanation for the underlying mechanism but feel that the hypothesis of antigen-antibody reaction, with production of antiheparin factors or platelet agglutinates, warrants attention." In 1973, Rhodes et al confirmed that a heparin-dependent antiplatelet antibody was responsible for the heparin-associated thrombocytopenia and thromboembolism.
The clinical features of HIT generally have been described to occur in two forms, type 1 and 2. In type 1 HIT, platelet counts fall between 1 and 5 days after the initiation of therapy and often return to normal values in spite of the continued use of heparin therapy. This form of HIT is not associated with thrombosis or thromboembolic sequelae. The pathogenesis is believed to be due to a direct aggregating effect of heparin and not to any immune-mediated reaction. Unfortunately, often it is difficult to determine exactly when heparin exposure was initiated, since HIT is known to occur with even minute amounts of heparin such as in heparin flushes and with heparin-coated intravenous catheters. Further, heparin is available from many other sources, such as subcutaneous heparin, intravenous heparin therapy, cardiac catheterization, and cardiopulmonary bypass.
In type 2 HIT, platelet counts fall between 4 and 14 days after initiation of therapy, and the fall is not uncommonly associated with thrombosis and thromboembolic complications. The pathogenesis is believed to be immune mediated, whereby specific heparin-dependent antibodies attach to platelet membranes causing platelet activation. I The aggregation and degranulation of platelets during this process result in a variety of clinical scenarios ranging from asymptomatic thrombocytopenia to devastating intravascular coagulation. The thrombotic phase of this syndrome can lead to saphenous vein graft occlusion, myocardial infarction, pulmonary embolus, stroke, limb loss, and even death. The incidence of type 2 HIT has been reported to be from 0.4 to 31 percent. Since mass screening is impractical, clinical suspicion of HIT is needed so that diagnostic confirmation with the antibody test can be performed while prompt withdrawal of all heparin therapy and institution of appropriate therapy are carried out. If further heparin therapy cannot be avoided, as in cases in which the patient requires cardiopulmonary bypass, alternative therapy is available and will be discussed later.
The incidence of detection of type 2 HIT in this series was 0.75 percent of patients undergoing cardiopulmonary bypass. Although this is consistent with reports in the literature ranging from 0.4 to 31 percent, we believe that the prevalence of the syndrome may be much greater than is routinely suspected.
One of the problems with recognizing HIT lies in the very definition of the syndrome; ie, "heparin-induced thrombocytopenia." In this series, 9 of 11 patients (81.8 percent) had platelet counts greater than 100,000 cu/mm at the time of recognition of the syndrome, although on closer inspection there was a mean percent decrease in the platelet count of 50 percent, with a range of 31 to 75 percent. Kappa et al described 3 of 16 patients (19 percent) with thrombotic complications but no thrombocytopenia. As a result of their findings, they suggest the syndrome be referred to as "heparin-induced platelet activation," since thrombocytopenia need not be present.
Another problem in securing a diagnosis of HIT lies in the availability of the appropriate assay for the heparin-dependent platelet antibody. A widely available and often used test is platelet aggregometry. While this test may be highly specific (>92 percent), the sensitivity of the test may vary from about 40 to 80 percent.[20,21] In 1986, Sheridan et al described a test based on platelet [C.sup.14]-serotonin release. This test is both highly sensitive and specific (>99 percent). Unfortunately, the test may not be readily available at most institutions which may further delay the confirmation and allow for more chance of laboratory error.
An additional consideration is the timing of cardiac surgery in relation to the initial exposure to heparin. Many patients are admitted to the cardiology service where they are exposed to heparin from multiple sources, eg, heparin flushes, subcutaneous heparin, intravenous heparin, heparin-coated pulmonary artery catheters, and cardiac catheterization. Depending on the urgency to perform surgery as well as outside pressures to limit hospital stays, the time between the initial exposure to heparin and cardiopulmonary bypass may be less than 24 h, leaving little opportunity to observe changes in platelet counts or to allow for adequate clinical and laboratory evaluation.
It is important for the clinician to have a high index of suspicion for HIT in any patient receiving heparin therapy. In our study of 100 control patients, the mean decrease in platelet counts following cardiopulmonary bypass was 26 percent. We recommend routine monitoring of the platelet counts in patients being exposed to heparin and prompt investigation for the heparin-dependent antibody whenever there is a thromboembolic complication. A 30 percent decrease in the platelet count may be significant.
The variety of thromboembolic events is striking and includes strokes, saphenous vein graft occlusions, peripheral arterial ischemia, emboli, deep venous thrombosis, and pulmonary embolus. Indeed, Stead et al described five patients with pulmonary embolism after coronary artery bypass surgery, all documented to have the platelet-dependent antibody. We are currently studying all patients with perioperative thromboembolic complications, regardless of the platelet count.
Controversy remains concerning the type of heparin used and its relative risk for causing HIT. The literature is conflicting, some series showing a higher incidence of HIT with bovine lung versus porcine mucosal heparin, while others suggest the incidence is equally low in both types of preparations. In this series, patients were exposed to porcine mucosal heparin preoperatively and bovine lung heparin during cardiopulmonary bypass. This switching of heparin preparations did not appear to lower the incidence of HIT or lessen the severity of complications, thus suggesting crossover of antigenicity between different preparations of heparin. Furthermore, the literature is varied on the suggestion that low molecular weight heparin or heparinoids decrease the risk of HIT.[25,26]
Aspirin has been described both for the prevention and treatment of patients with HIT. In this series, 6 of 11 patients (55 percent) were taking aspirin immediately prior to surgery and 5 of 11 patients (45 percent) had no recent exposure to aspirin. No difference was noted in the incidence of HIT or severity of complications in those patients receiving aspirin. This correlates with Kappa et al who found, in 40 percent of patients tested, that aspirin failed to prevent heparin-induced platelet aggregation and release in vitro.
Unfortunately, the cure for complications from HIT is as elusive as the diagnosis. The mainstay of therapy in the care of these patients is to suspect the syndrome as early as possible and to eliminate all sources of possible exposure to heparin. As mentioned above, the success of antiplatelet drugs, such as aspirin, dipyridamole, or dextran, has been mixed at best. Other therapy has included thrombolytic therapy, plasmapheresis, and administration of immunoglobulin. If continued anticoagulation is required, warfarin should be administered.
What if the diagnosis of HIT is made preoperatively? One approach is to wait 4 to 8 weeks, or more, until the antiplatelet reaction has vanished and then perform the surgery. This approach is limited by the variable length of time for the antibody reaction to dissipate and is not practical in most patients requiring urgent cardiac surgery. Furthermore, the heparin antibody may be present for years; therefore, it is imperative that a negative antibody test be noted prior to performing the procedure.
Ancrod, a defibrinogenating agent derived from Malayan pit viper venom, has been described as an alternative to heparin for cardiopulmonary bypass in cases of known HIT. Although successfully used, it appears to be associated with increased bleeding.
Iloprost, a prostacyclin analogue, has been used in conjunction with heparin in patients with HIT requiring cardiopulmonary bypass.[19,28,33,34] Iloprost is a strong inhibitor of platelet function, and yet, because of its short half-life of 15 to 30 min, platelet reactivity returns quickly and is associated with less bleeding. Because of its vasodilating activity, Iloprost can be associated with severe hypotension resistent to large doses of phenylephrine.
Recombinant hirudin is a homogeneous preparation which directly inhibits thrombin without the need of a cofactor and appears to have minimal interaction with platelets and no immunogenicity problems in early studies. In the future, this may provide an alternative to heparin in patients undergoing cardiopulmonary bypass.
 McLean J. The thromboplastic action of cephalin. Am J Physiol 1916; 41:250-57  Hirsh J. Heparin. N Engl J Med 1991; 324: 1565-74  Sheridan D, Carter C, Kelton JG. A diagnostic test for heparin-induced thrombocytopenia. Blood 1986; 67:22-30  Levine M, Hirsh J. Hemorrhagic complications of anticoagulant therapy. Semin Thromb Hemost 1986; 12:39-57  White RW, Sadd JR, Nensel RE. Thrombotic complications of heparin therapy: including six cases of heparin-induced skin necrosis. Ann Surg 1979; 190:595-608  Ginsberg JS, Kowalahuk G, Hirsh J, Brill-Edwards P, Burrows R, Coates G, et al. Heparin effect on bone density. Thromb Haemost 1990; 64:286-89  Jacques LB. Heparins - anionic polyelectrolyte drugs. Pharmacol Rev 1980; 31:99-166  Physicians desk reference. Montuak, NJ: Medical Economics Company, 1992; 1264-66; 2342-44; 2457-59  O'Kelly R, Magee F, McKenna J. Routine heparin therapy inhibits adrenal aldosterone production. J Clin Endocrinol Metab 1983; 56:108-12  Curry N, Bandana EJ, Pirofsky B. Heparin sensitivity: report of a case. Arch Intern Med 1973; 132:744-45  Copley AL, Robb TP. Studies on platelets: II. The effect of heparin on the platelet count in vitro. Am J Clin Pathol 1942; 12:416-23  Weismann RE, Tobin RW. Arterial embolism occurring during systemic heparin therapy. Arch Surg 1958; 76:219-27  Roberts B, Rosato FE, Rosato EF. Heparin - a cause of arterial emboli? Surgery 1964; 55:803-08  Rhodes GR, Dixon RH, Silver D. Heperin-induced thrombocytopenia with thrombotic and hemorrhagic manifestations. Surg Gynecol Obstet 1973; 136:409-16  Laster JL, Nichols WK, Silver D. Thrombocytopenia associated with heparin-coated catheters in patients with heparin-associated antiplatelet antibodies. Arch Intern Med 1989; 149:2285-87  Becker PS, Miller VT. Heparin-induced thrombocytopenia. Stroke 1989; 20:1449-59  Walls JT, Curtis JJ, Silver D, Boley TM, Schmaltz A, Nawarowong W, et al. Heparin-induced thrombocytopenia in open heart surgical patients: sequelae of late recognition. Ann Thorac Surg 1992; 53:787-91  Ansell JE, Price JM, Shah S, Bechner RR. Heparin-induced thrombocytopenia - what is its real frequency? Chest 1985; 88:878-82  Kappa JR, Fisher CA, Berkowitz HD, Cottrell ED, Addonizio VP. Heparin-induced platelet activation in sixteen surgical patients: diagnosis and management. J Vasc Surg 1987; 5:101-09  Kelton JG, Sheridan D, Brain H, Powers PJ, Turpie AG, Carter CJ. Clinical usefulness of testing for a heparin-dependent platelet-aggregating factor in patients with suspected heparin-associated thrombocytopenia. J Lab Clin Med 1984; 103:606-12  Chong BH, Berndt MC. Heparin-induced thrombocytopenia. Blut 1989; 58:53-7  Stead RB, Schafer AI, Rosenberg RD, Handin RI, Josa M, Khuri SF. Heterogeneity of heparin lots associated with thrombocytopenia and thromboembolism. Am J Med 1984; 77:185-88  Bell WR, Royall RM. Heparin-associated thrombocytopenia: a comparison of three heparin preparations. N Engl J Med 1980; 303:788-95  Bailey RT, Ursick JA, Hein KL, Hilleman DE, Reich J. Heparin-associated thrombocytopenia: a prospective comparison of bovine lung heparin manufactured by a new process, and porcine intestinal heparin. Drug Intell Clin Pharm 1986; 20:374-78  Roussi JH, Houbouyan LL, Goquel AF. Use of low molecular weight heparin in heparin-induced thrombocytopenia with thrombotic complications. Lancet 1984; 1:1183  Leroy J, Leclerc MH, Delahousse B, Guerosis C, Foloppe P, Gruel Y, et al. Treatment of heparin-associated thrombocytopenia and thrombosis with low molecular weight heparin (CY 216). Semin Thromb Hemost 1988; 11:326-29  Makhoul RG, McCann RL, Austin EH, Greenberg CS, Lowe JE. Management of patients with heparin-associated thrombocytopenia and thrombosis requiring cardiac surgery. Ann Thorac Surg 1986; 43:617-21  Kappa JR, Fisher CA, Todd B, Stenoch N, Bell P Campbell F, et al. Intraoperative management of patients with heparin-induced thrombocytopenia. Ann Thorac Surg 1990; 49:714-23  Nand S, Robinson JA. Plasmapheresis in the management of heparin-associated thrombocytopenia with thrombosis. Am J Hematol 1988; 28:204-06  Frame JN, Mulvey KP, Phares JC, Anderson MJ. Correction of severe heparin-associated thrombocytopenia with intravenous immunoglobulin. Ann Intern Med 1989; 111:946-47  Olinger GN, Hussey CV, Olive JA, Malik MI. Cardiopulmonary bypass for patients with previously documented heparin-induced platelet aggregation. J Thorac Cardiovasc Surg 1984; 87:673-77  Teasdale SJ, Zulys VJ, Mycyk T, Baird RJ, Glynn MFX. Ancrod anticoagulation for cardiopulmonary bypass in heparin-induced thrombocytopenia and thrombosis. Ann Thorac Surg 1989; 48:712-13  Ellison N, Kappa J, Fisher CA, Addonizio VP. Extracorporeal circulation in a patient with heparin-induced thrombocytopenia. Anesthesiology 1985; 63:336-37  Kraenzler EJ, Starr NJ. Heparin-associated thrombocytopenia: management of patients for open heart surgery-case reports describing the use of Iloprost. Anesthesiology 1988; 69:964-67  Walenga JM, Bakhos M, Messmore HL, Fareed J, Pifarre R. Potential use of recombinant hirudin as an anticoagulant in a cardiopulmonary bypass model. Ann Thorac Surg 1991; 51:271-77
COPYRIGHT 1993 American College of Chest Physicians
COPYRIGHT 2004 Gale Group