Delayed-onset HIT caused by low-molecular-weight heparin manifesting during fondaparinux prophylaxis Khalid A. Alsaleh, 1 Sami M. A. Al-Nasser, 1 Shannon M. Bates, 1 Ameen Patel, 1 Theodore E. Warkentin, 1,2 and Donald M. Arnold 1 * Heparin-induced thrombocytopenia (HIT) is a prothrombotic condition caused by platelet-activating antibod- ies that react with platelet factor 4 (PF4)/heparin complexes. Delayed-onset HIT occurs after heparin is stopped. Fondaparinux, a synthetic pentasaccharide, is thought to be a safe alternative anticoagulant in HIT. We describe a patient with delayed-onset HIT triggered by low-molecular-weight heparin (LMWH) which occurred during fondaparinux prophylaxis and which was complicated by microangiopathic hemolytic ane- mia. Patient serum contained high-titer anti-PF4/heparin antibodies demonstrating heparin-dependent plate- let activation with serial dilutions. Confirmed delayed-onset HIT with LMWH has not been previously reported. Low dose fondaparinux does not necessarily prevent thrombotic complications of HIT. Am. J. Hematol. 83:876–878, 2008. V V C 2008 Wiley-Liss, Inc. Introduction Heparin-induced thrombocytopenia (HIT) is an immune- mediated adverse reaction caused by platelet-activating IgG antibodies that recognize platelet factor 4 (PF4) bound to heparin. In some patients, high titer PF4/heparin antibod- ies can activate platelets in vitro even in the absence of heparin and cause thrombocytopenia and thrombosis sev- eral days after heparin has been stopped—a syndrome called ‘delayed-onset HIT’ [1]. Fondaparinux, a synthetic antithrombin-binding sulfated pentasaccharide, has been shown to be effective for the prevention of deep-vein thrombosis (DVT) in patients under- going orthopedic surgery [2] and has been used as an alternate anticoagulant in HIT [3–5]. However, rare cases of HIT in patients treated exclusively with fondaparinux [6,7] are cause for concern. We describe a patient with delayed-onset HIT following low-molecular-weight heparin (LMWH) exposure that became manifest during fondaparinux thromboprophylaxis. Atypical clinical features were present including microangio- pathic hemolytic anemia (MAHA), which caused a delay in the diagnosis. Serial dilutions of patient serum allowed for the demonstration of heparin-dependent platelet activation in the serotonin release assay (SRA). Case Report A 64-year-old woman with osteoarthritis underwent right total knee replacement at a peripheral hospital. No heparin was given pre- or intraoperatively, and there was no history of previous heparin exposure. Her preoperative platelet count was 266 3 10 9 /L. Following surgery, she received prophylactic LMWH (dalteparin 5000U subcutaneously) for 3 days. On day 3, the platelet count was 211 3 10 9 /L, and dalteparin was substituted by prophylactic fondaparinux (2.5 mg subcutaneously) on hospital discharge (see Fig. 1). On the seventh postoperative day, she presented to the emergency department complaining of headache. The platelet count was 197 3 10 9 /L with no bleeding or throm- bosis. Computed tomography (CT) scan of the head was normal, and the patient was sent home to continue fonda- parinux prophylaxis. On the eleventh postoperative day, she re-presented with fever, disorientation, amnesia, and slurred speech. The platelet count was 68 3 10 9 /L, interna- tional normalized ratio was 1.2, activated partial thrombo- plastin time (aPTT) was 33 sec, and lactate dehydrogenase (LD) was 1.5-times upper limit normal (361 U/L). A chest radiograph was normal. The patient was admitted to hospi- tal with a presumptive diagnosis of postoperative sepsis, started on empiric antibiotics and continued on prophylactic fondaparinux. Cultures remained sterile. On postoperative day 13, a proximal right leg DVT was diagnosed by com- pression ultrasonography, platelet count was 105 3 10 9 /L and HIT was suspected. Red blood cell (RBC) fragments became apparent on the blood film. Initially, fondaparinux was increased to therapeutic doses (7.5 mg subcutane- ously per day) but within 24 hr was switched to danaparoid, given as an intravenous bolus of 2,250 U followed by infu- sion (400 U/hr for 4 hr, then 300 U/hr for 3 hr, then 200 U/ hr) for fear that HIT was triggered by fondaparinux. Despite increasing the dose to 315 U/hr, anti-Xa danaparoid levels remained below 0.41 U/mL (target range, 0.50–0.80) even by day 17. At that time, the platelet count had dropped to 84 3 10 9 /L, acute delirium and renal failure developed, and imaging studies revealed cerebral ischemia and bilateral adrenal hemorrhage. Danaparoid was discontinued and in- travenous argatroban was started at 200 mcg/min (2 mcg/ kg per min) resulting in an aPTT between 80 and 90 sec. By day 20, the platelet count had improved to 131 3 10 9 /L; however because of the persistence of RBC fragments and increased LD to 415 U/L, the diagnosis of thrombotic thrombocytopenic purpura was entertained (ADAMTS13 level was normal, and anti-ADAMTS13 antibodies were not detected as measured subsequently). One treatment with therapeutic plasma exchange was administered with no effect. By day 23, the platelet count continued to rise, neurological symptoms resolved, RBC fragments were not seen and renal function normalized. By day 26 (10 days 1 Department of Medicine, Michael G. DeGroote School of Medicine McMas- ter University, Hamilton, Ontario, Canada; 2 Department of Pathology and Molecular Medicine, McMaster University, Hamilton, Ontario, Canada *Correspondence to: Donald M. Arnold, HSC 3V-48, 1200 Main Street West, Hamilton, Ontario, Canada. E-mail: arnold@mcmaster.ca Received for publication 16 June 2008; Revised 8 August 2008; Accepted 12 August 2008 Am. J. Hematol. 83:876–878, 2008. Published online 19 August 2008 in Wiley InterScience (www.interscience. wiley.com). DOI: 10.1002/ajh.21273 V V C 2008 Wiley-Liss, Inc. American Journal of Hematology 876 http://www3.interscience.wiley.com/cgi-bin/jhome/35105 Brief Research Report