By Lawrence Rice, M.D.
Dr. Rice is currently a Professor of Medicine in Hematology and Thrombosis Research at the Baylor College of Medicine.

Case Presentation: A 51-year-old man had coronary artery bypass surgery, complicated by hemopericardium requiring a pericardial window. By the fifth post-operative day, he was ambulating and eating, and prophylactic low molecular weight heparin was started. He progressed until the ninth post-operative day when temperature spiked to 104.8 degrees, blood pressure fell to 90/70, and the abdomen became distended and tender, with decreased bowel sounds. Platelets had been 182,000/cu mm on the third post-operative day, but now were 40,000/cu mm. Consultants in infectious disease and hematology both judged sepsis with sepsis-thrombocytopenia likely. Antibiotics were started and the patient was transferred back to intensive care. A DIC profile was normal. Although sepsis appeared to explain the thrombocytopenia, the hematologist was also concerned that the time course fit heparin-induced thrombocytopenia (HIT). Heparin exposures were eliminated and a test for heparin-PF4 antibodies was sent.

The question at this point was: Should anything else be done?

What Was Done and What Transpired:
HIT seemed moderately likely to me, perhaps the main problem. Argatroban was initiated immediately. Suspected bowel ischemia (a possible consequence of HIT) led to an emergent CT scan, revealing (to everyone's surprise) bilateral adrenal hemorrhage. A cortisol level, drawn before steroid administration, later came back less than 1µg/dL. The heparin antibody ELISA came back very strongly positive. Fever and hypotension resolved after steroid bolus, and abdominal symptoms abated. The recommended 2µg/kg/min intravenous infusion dose of argatroban produced a supratherapeutic aPTT at three hours, but steady therapeutic level was soon achieved on 0.5µg/kg/min. Argatroban significantly affects the prothrombin time INR, which was 2.6. Warfarin was cautiously introduced four days later when platelets already were 161,000/cu mm. Argatroban was stopped days later when the INR reached 6.0. Off argatroban, the INR was then 3.6. The patient went home fully recovered, continuing steroid replacement. Perspectives: My colleagues John McCarthy and Kelty Baker and I see more than 100 cases of HIT yearly, 75-80 percent in cardiovascular surgery patients, 1 percent of such patients in our two large hospitals. So, HIT is common, approximately ten times more common with unfractionated heparin than low molecular weight heparin. In this case, the high dose heparin given on pump was clearly the sensitizing agent, but low molecular weight heparins cross-react and are contraindicated once sensitization occurs. The enoxaparin very likely perpetuated the problem. HIT creates extreme hypercoagulability. Bilateral adrenal necrosis is a microthrombotic process with secondary hemorrhage that has been seen with HIT. Our patient's fever and hypotension were due to acute adrenal crisis, not sepsis. Despite significant bleeding risks (hemopericardium, adrenal hemorrhages), the decision to start a direct thrombin inhibitor was clear in the face of possible HIT with impending bowel necrosis. Argatroban and lepirudin are FDA-approved for HIT. The former is preferred with renal compromise; diminished urine output influenced the decision in this case. Lepirudin is preferred with liver dysfunction, and is also infused intravenously with a PTT monitoring. This case well illustrates that the key to preventing catastrophes with HIT is a high level of awareness and suspicion.

A Take-Home Management Point: B Generally, clinical suspicion of HIT should trigger initiation of an alternative anticoagulant, even with isolated HIT (no new thrombus). Prior to the availability of safe and effective alternative anticoagulants, it was acceptable to just stop heparin, initiate warfarin or wait on test results, but the inadequacy of these strategies was demonstrated particularly by Ted Warkentin's and Diane Wallis' studies. In Andreas Greinacher's trials, patients waited for positive antibody tests before they could receive a direct thrombin inhibitor; the period between heparin cessation and initiation of alternative anticoagulation was found to be of highest risk for new complications. I do not believe our patient's outcome would have been favorable had anticoagulation been delayed. I recommend a minimum of six weeks of anticoagulation even with isolated HIT because thrombotic complications are known to occur that late, well after platelet recovery.

Further Reading:

• Rice L, Nguyen P, Vann A. Preventing complications in heparin-induced thrombocytopenia. Alternative anticoagulants are improving patient outcomes. Postgraduate Medicine, 112:85-89, 2002.
• Rice L. Heparin-induced thrombocytopenia: myths and misconceptions. Arch Intern Med, 2004, in press.

Return to Table of Contents