This site is intended for health professionals only
A Tu Duy Khiem-El Aatmani*
**Cardiovascular Surgery ICU
Heparin is widely used as an anticoagulant in the prevention and treatment of venous and arterial thromboembolic events. This is all the more true in France, where the prescription rate of low-molecular-weight heparin (LMWH) reached 33% of the world market in 2002, which put France at the head of consumer countries.(1) Heparin-induced thrombocytopenia (HIT) is the most important complication of heparin treatment. It develops in up to 3% of patients,(2) as a consequence of an antibody-mediated reaction against the complex of heparin and platelet factor 4 (PF4), and is associated with high rates of thromboembolic events.(3) Heparin must then be stopped and another anticoagulant therapy substituted. Alternative anticoagulant treatments approved for this indication include direct thrombin inhibitors, such as argatroban (marketed in the USA), lepirudin and agents that have anti-Xa activity, such as danaparoid (marketed in Europe).(4)
Danaparoid is relatively expensive, and in 2003 the drug committee of the University Hospital of Strasbourg drew up recommendations for its use. HIT usually presents as a fall in the platelet count with or without thrombosis. According to the recommendations, diagnosis of acute HIT should be confirmed by positive anti-PF4/heparin antibody finding, or a high probability of such a finding, and thrombocytopenia with a decrease in platelet count of more than 40%, or to a level of less than 100×10(9)/l. The onset of HIT usually occurs more than five days after the start of heparin treatment, although it can be shorter in patients who have received heparin treatment within the past three months. HIT should also be considered in patients who develop thromboembolic complications in spite of heparin therapy. A history of HIT should be documented in the patient’s medical record. HIT should be confirmed by anti-PF4/heparin antibody finding or clinical evidence, and be recorded.
In recent years, we have noticed a threefold increase in the use of danaparoid. The objective of this study was to assess the adherence to institutional recommendations for the use of danaparoid.
Material and methods
A retrospective study was undertaken of all the patients treated with danaparoid between 1 August 2004 and 31 March 2005. The patients’ medical records were retrieved, and data concerning patient characteristics, the context of danaparoid initiation, anti-PF4/heparin antibody testing, duration of treatment, transfer to oral anticoagulants and clinical course were collected. For the last three criteria, we used the number of hospitalisations instead of the number of patients, because some patients were hospitalised twice. The extent of conformity to institutional recommendations was then analysed.
Over the eight-month period, 50 patients were treated using danaparoid. Forty-five patients were evaluable and comprised 32 men (71%) and 13 women (29%). The median age was 69 years for the men and 73 years for the women, with an overall median of 70 years (range 32 to 95 years). Almost half of the patients (21 [46%]) were treated in intensive care units (ICUs), and eight patients (18%) were treated in a cardiology unit (see Figure 1).
Context for danaparoid initiation
Of the 45 patients included, 32 (71%) patients were treated for suspected acute HIT and 11 (25%) for history of HIT. For one patient, treatment with danaparoid was initiated in another hospital, and for the last patient danaparoid was justified by previous thromboembolic events during anticoagulant therapy. Among the 32 patients with suspected acute HIT, LMWH was implicated in 17 (53%) patients and unfractionated heparin in 15 (47%) patients.
Conformity to institutional recommendations
For patients with suspected acute HIT, the changes in platelet counts are shown in Table 1. The mean platelet count decrease was 62%, and the mean time to onset of thrombocytopenia was nine days. For nine patients, treatment did not conform to the institutional recommendations. Eight of these experienced a fall in the platelet count less than five days after starting heparin treatment.
Among the 11 patients with history of HIT, the diagnosis was confirmed in six by anti-PF4/heparin antibody measurements. In two patients, the diagnosis was confirmed by the clinical course. For the remaining three patients, the details were not documented.
Anti-PF4/heparin antibody testing
Among the 32 patients with suspected acute HIT, anti-PF4/heparin antibody tests were not done for four patients. For the 28 remaining patients, tests were carried out on the first day of danaparoid treatment and were positive in five patients. Overall, 16% of cases of suspected acute HIT were confirmed by positive antiheparin antibody tests.
Duration of treatment and transfer to oral anticoagulant
There were 46 courses of treatment as one patient was hospitalised twice. The median duration of treatment was nine days (range 1–78). Figure 2 shows that 30% of danaparoid treatments exceed two weeks. Overall, 33% of patients were transferred to oral anticoagulants.
The clinical course under danaparoid treatment was favourable in 32 (74%) of the patients. Unfavourable outcomes included seven deaths and two persistent thrombocytopenias for patients with suspected acute HIT, and two deaths in patients with histories of HIT.
Cases of suspected acute HIT were three times greater than histories of actual HIT, which suggests that the medical teams are very concerned about the possibility of HIT. A previous study undertaken in our institution (results not published) examined the use of danaparoid over 29 months (October 1998–March 2001). Twenty-four patients were included (12 men and 12 women). The proportion of cases of suspected HIT was then 46% against 54% of history of HIT. Compared with our previous work, it appears that the total number of patients treated using danaparoid has increased sevenfold, and that the proportion of cases of suspected acute HIT increased from 46% to 71%. The rate of anti-PF4/heparin antibody testing increased from 64% to 88%, whereas the frequency of positive test findings decreased from 29% to 18%.
Among the long durations of treatment, six were over 20 days and concerned primarily patients in ICUs where transfer to an oral anticoagulant was not possible.
Moreover, after a confirmed diagnosis of acute HIT, apart from a positive anti-PF4/heparin result, no document is given to the patient certifying a history of HIT. Because people with a history of HIT are more likely to develop another episode of HIT, heparin must not be given again. Clinicians have to be informed that anticoagulation must be achieved with an alternative agent.
Overall, the drug committee guidelines of our
institution were followed. The number of patients treated increased sevenfold in six years and is related to a higher proportion of cases of suspected HIT. This, together with the higher frequency of anti-PF4/heparin antibody testing, shows improved management of HIT by medical teams.
As a result of this study, clinicians, haematologists and the Regional Centre for Pharmacovigilance have collaborated closely to prepare new recommendations to continue improving the management of HIT. A record card that confirms the positive anti-PF4/heparin antibody status for an individual was created. This document will be given by haematologists to the patients concerned, and the Centre for Pharmacovigilance will be informed. The impact of these actions will be evaluated in future studies.