Heparin Induced Thrombocytopenia: Descriptive Analysis of Diagnosis and Management

Heparin induced thrombocytopenia (HIT) is a well-recognized serious complication of heparin therapy. A key step in diagnosing HIT is the timing of ordering of the diagnostic tests, ELISA HIT antibody (HIT Ab) and the functional Serotonin release assay (SRA). A pretest clinical score (T score) has been developed and prospectively validated in two centers and includes thrombocytopenia, timing of platelet count fall, presence of thrombosis and other possible causes of thrombocytopenia. Initial positive or negative HIT Ab testing in the absence of accurate clinical suspicion might result in significant morbidity and mortality due either to unnecessary anticoagulation or holding necessary anticoagulation.

To shed more light on this issue and to increase the awareness of established protocols, we reviewed cases in which HIT Ab tests were sent between January 2008 and June 2010 in a tertiary hospital center and did a descriptive analysis according to clinical suspicion based on the T score.

A total of 1173 HIT Ab tests were sent between January 2008 and June 2010. 826(70.4%) patients had low clinical score. From these patients, 764(92.4%) had negative HIT Ab test of which 4 patients had SRA checked and all (100%) were negative; 62(7.6%) patients had positive HIT Ab test, of which 11 patients had SRA checked, with 9(81.8%) being negative and 2(18.2%) being positive. 347(29.6%) patients had intermediate/high clinical score. From these patients, 260(74.9%) had negative HIT Ab, of which 6 patients had SRA checked and all (100%) were negative, and 87(24.1%) had positive HIT Ab, of which 18 patients had SRA checked with 4(22.2%) being positive and 14(77.8%) being negative. A total of 97 HIT Ab tests were repeated with 97 %(94 cases) concordance between the results and 3%(3 cases) discordance. Two out of these 3 cases had negative HIT Ab which on repeating turned positive; however, HIT was excluded by a negative SRA in one, and clinically in the other one. A total of 322(27.4%) patients had discordance between clinical score and HIT Ab test, with 96 (29.8%) cases generating a hematology consult, while in the 851(73.6%) concordant cases, 150(17.6%) generated a hematology consult. DTI was started in 20(2%) patients with low clinical score and 78(22%) patients with intermediate/high clinical score.

This study confirms the importance of following the established protocols of ordering the HIT score only when there is at least an intermediate or high clinical suspicion for diagnosis as only 7.6% of patients with low clinical score had positive HIT Ab test, and even in these patients 81.8% of SRA tests sent were negative, compared to 24.1% positive HIT Ab in patients with intermediate/high clinical score. However, it also shows that the current pretest clinical scores should be improved since in patients with intermediate/high clinical score and positive HIT Ab, 77.8% of SRA tests checked were negative, ruling out HIT, and in patients with low clinical score and positive HIT Ab test, 18.2% of SRA tests checked were positive, confirming HIT. Awaiting improvement in the pretest clinical scores, we feel that SRA should be always sent in positive HIT Ab cases and low/intermediate pretest probability. The study also confirms the high negative predictive value of a negative HIT Ab test as all the SRA sent (100%) were negative even in patients with intermediate/high clinical score. This study also shows for the first time that there is a very high concordance rate between repeated HIT Ab tests (97%), and there is no need to repeat HIT Ab test when positive or negative. The study also revealed a low frequency of hematology consults even with discordant results between clinical score and HIT Ab tests (29.8%) and we feel that a hematology consult can be more helpful in such cases. We also encourage the use of DTI, when possible, in all patients with intermediate/high clinical score while awaiting the HIT diagnostic tests as the study revealed only 22% of patients receiving DTI in this subset of patients.

This study confirms the importance of following the established protocols and having an accurate pretest clinical probability for diagnosing HIT, but it also shows that there is a need to improve the current clinical scoring system to avoid unnecessary anticoagulation or holding necessary anticoagulation, and also avoiding repeating HIT Ab tests with its ensuing costs and low benefit.

No relevant conflicts of interest to declare.