Correspondence

Heparin-Induced Thrombocytopenia

N Engl J Med 1995; 333:1006-1007October 12, 1995

Article

To the Editor:

We disagree that the study by Warkentin et al. (May 18 issue)1 identified a difference in outcome favoring the use of low-molecular-weight heparin. Their initial study of the identical patient population found no statistically significant difference in thrombotic events, incidence of venous thrombosis, episodes of major bleeding, or red-cell–transfusion requirements between the group given standard heparin and the group given low-molecular-weight heparin.2

From both studies (really the same study) one can conclude only that heparin-induced thrombocytopenia is a marker of thrombotic disease, because the authors do not demonstrate that thrombosis as a consequence of heparin-induced thrombocytopenia has a less favorable outcome than thrombosis that occurs during treatment with low-molecular-weight heparin. Their data suggest that standard heparin should be administered until the rare complication of late thrombocytopenia, fairly predictive of heparin-induced thrombocytopenia, is observed, at which point alternative anticoagulation therapy should be considered.

We are puzzled by discrepancies in the results of these two studies of the same patient population. The initial study reported the development of thrombocytopenia in “two patients in the standard group” and two pulmonary emboli in this group. The current study describes nine patients with thrombocytopenia in the standard-heparin group, and four pulmonary emboli overall. These differences deserve clarification.

Noah Berkowitz, M.D., Ph.D.
Joshua Beckman, M.D.
Columbia–Presbyterian Medical Center, New York, NY 10032

2 References

1
Warkentin TE, Levine MN, Hirsh J, et al. Heparin-induced thrombocytopenia in patients treated with low-molecular-weight heparin or unfractionated heparin. N Engl J Med 1995;332:1330-1335
Full Text | Web of Science | Medline

2
Levine MN, Hirsh J, Gent M, et al. Prevention of deep vein thrombosis after elective hip surgery: a randomized trial comparing low molecular weight heparin with standard unfractionated heparin. Ann Intern Med 1991;114:545-551
Web of Science | Medline

To the Editor:

One question raised by the study by Warkentin et al. is whether exposure to standard heparin was controlled during the hospitalization of the patients randomly assigned to receive low-molecular-weight heparin. The ubiquitous use of heparin to maintain the patency of intravenous catheters raises the possibility of heparin exposure during a hospitalization of one to two weeks for elective hip surgery.

If the patients randomly assigned to receive low-molecular-weight heparin were also receiving unfractionated heparin, the 2.2 percent incidence of heparin-dependent antibodies in these patients may reflect clinical realities but may also overestimate the antigenicity of low-molecular-weight heparin. A previous study of serum samples from patients exposed to unfractionated heparin found that 60.8 percent and 25.5 percent, respectively, reacted to two different preparations of low-molecular-weight heparin.1 It would be interesting to know whether the heparin-dependent antibodies detected in the current study were all active in the presence of low-molecular-weight heparin. If they were not, then it would seem likely that the antigenicity of the low-molecular-weight heparin was overestimated.

Mark J. Shumate, M.D., M.P.H.
Emory University, Decatur, GA 30033

1 References

1
Kikta MJ, Keller MP, Humphrey PW, Silver D. Can low molecular weight heparins and heparinoids be safely given to patients with heparin-induced thrombocytopenia syndrome? Surgery 1993;114:705-710
Web of Science | Medline

To the Editor:

In the article by Warkentin et al., two points are unclear. The report gives few details about the characteristics of the two groups of patients: Were they similar with respect to previous treatment with heparin and, hence, to possible sensitization to heparin? Were the numbers of patients with other causes of thrombocytopenia or thrombosis similar in the two groups? Was the duration of the hematologic follow-up similar in the two groups?

Our second and main concern is about the data in Table 1 of their article concerning the clinical evolution of the nine patients with heparin-induced thrombocytopenia. Thrombosis developed in eight of those patients. In four patients, thrombosis occurred two to four days after the diagnosis of heparin-induced thrombocytopenia could have been strongly suspected on the basis of a platelet count below 150,000 per cubic millimeter or the presence of heparin-dependent IgG antibodies (Patients 1, 4, 7, and 9). The authors do not specify the circumstances of the thrombotic events in these patients, and it seems to us that heparin was continued in some patients (Patients 4 and 9) until the occurrence of thrombosis. Warkentin et al. do not mention the limits they set in their protocol for the pursuit or arrest of the trial when heparin-induced thrombocytopenia was diagnosed. Heparin-induced thrombocytopenia with thrombosis has a 30 percent death rate1,2 — a statistic that should prompt the discontinuation of heparin whenever a diagnosis of heparin-induced thrombocytopenia is seriously suspected.

Nicolas Hougardy, M.D.
Jean-Pascal Machiels, M.D.
Christophe Ravoet, M.D.
Hospital of Jolimont, 7100 La Louvière, Belgium

2 References

1
Chong BH. Heparin-induced thrombocytopenia. Br J Haematol 1995;89:431-439
CrossRef | Web of Science | Medline

2
King DJ, Kelton JG. Heparin-associated thrombocytopenia. Ann Intern Med 1984;100:535-540
Web of Science | Medline

Author/Editor Response

The authors reply:

To the Editor: Hougardy and colleagues ask whether the thrombi complicating heparin-induced thrombocytopenia could have been prevented. Three thrombotic events occurred during the initial phase of the decrease in the platelet count (before thrombocytopenia), two thrombotic events occurred the first day the platelet count fell below 150,000 per cubic millimeter, and three events occurred despite the discontinuation of the study drug because of thrombocytopenia. This finding of a strong tendency for thrombosis to occur in heparin-induced thrombocytopenia is consistent with the results of our recent retrospective cohort study1 and suggests that many of these thrombotic complications would occur despite routine monitoring of platelet counts. The second question is whether prior exposure to heparin could have influenced the outcomes in our study. This is unlikely, since no patient had detectable heparin-dependent IgG antibodies before postoperative day 5, suggesting that anamnestic responses did not occur.

Berkowitz and Beckman suggest a thrombus caused by heparin-induced thrombocytopenia and a thrombus that occurs despite heparin prophylaxis have a similar clinical impact. We disagree. Thrombi associated with heparin-induced thrombocytopenia are disproportionately severe and are often the presenting manifestation of heparin-induced thrombocytopenia. Most important, thrombosis complicating heparin-induced thrombocytopenia is not as easy to treat as Berkowitz and Beckman imply: rapidly effective, alternative anticoagulants for the treatment of heparin-induced thrombocytopenia (e.g., danaparoid, hirudin, and ancrod)2 are not approved for use in the United States, and warfarin is ineffective as initial therapy.1,3

The data in our two reports are in complete agreement. The first report4 referred only to the two patients with severe thrombocytopenia (platelet count, less than 50,000 per cubic millimeter) that occurred on the days the study drug was administered. Our subsequent study included all patients whose platelet counts fell below 150,000 per cubic millimeter at any point during postoperative days 6 to 14.5 Pulmonary embolism developed in two patients during the first two postoperative weeks4; subsequently, we described two additional patients in whom pulmonary embolism developed during the third postoperative week, before hospital discharge.5 The odds ratio for the association of pulmonary embolism with heparin-induced thrombocytopenia is similarly high for either follow-up period (81.9 and 93.4, respectively; both P<0.05).

Dr. Shumate asks whether incidental exposure to heparin by means of arterial and central venous catheters could have affected our results. Arterial and central venous catheters were used in 77.6 percent of study patients (76.8 percent of those randomly assigned to unfractionated heparin and 78.4 percent of those assigned to low-molecular-weight heparin, P = 0.58). We did not observe a significantly higher frequency of catheter use in patients in whom heparin-induced thrombocytopenia developed (6 of 9 [67 percent], P = 0.43) or who had heparin-dependent IgG antibodies (16 of 20 [80 percent], P = 1.0). However, if incidental exposure to heparin explained the formation of heparin-dependent IgG antibodies in some of the patients given low-molecular-weight heparin, it would indicate that the advantage of reduced immunogenicity of low-molecular-weight heparin may be even greater than our study suggests.

Theodore E. Warkentin, M.D.
Jack Hirsh, M.D.
John G. Kelton, M.D.
McMaster University, Hamilton, ON L8N 3Z5, Canada

5 References

1
Warkentin TE, Kelton JG. Heparin-induced thrombocytopenia: predominance of venous thrombotic complications, and a high risk for subsequent thrombosis in patients who are initially recognized with isolated thrombocytopenia. Blood 1994;84:Suppl 1:188a-188a abstract.
Web of Science

2
Warkentin TE, Kelton JG. Interaction of heparin with platelets, including heparin-induced thrombocytopenia. In: Bounameaux H, ed. Low-molecular-weight heparins in prophylaxis and therapy of thromboembolic diseases. Vol. 19 of Fundamental and clinical cardiology. New York: Marcel Dekker, 1994:75-127.

3
Warkentin TE, Russett JI, Johnston M, Kelton JG. Warfarin treatment of deep vein thrombosis complicating heparin-induced thrombocytopenia (HIT) is a risk factor for initiation of venous limb gangrene: report of nine patients implicating the interacting procoagulant effects of two anticoagulant agents. Thromb Haemost 1995;73:1110-1110 abstract.
Web of Science

4
Levine MN, Hirsh J, Gent M, et al. Prevention of deep vein thrombosis after elective hip surgery: a randomized trial comparing low molecular weight heparin with standard unfractionated heparin. Ann Intern Med 1991;114:545-551
Web of Science | Medline

5
Warkentin TE, Levine MN, Hirsh J, et al. Heparin-induced thrombocytopenia in patients treated with low-molecular-weight heparin or unfractionated heparin. N Engl J Med 1995;332:1330-1335
Full Text | Web of Science | Medline

Citing Articles (1)