Are We Doing Too Many HIT Tests? II

Back Ground: Heparin Induced Thrombocytopenia is a complex immune disorder related to exposure to heparin that results in arterial and venous thrombosis and moderate to severe thrombocytopenia. Incidence is 0.8-1% in hospitalized patients exposed to heparin. However there is a problem with over testing for HIT syndrome that it is not only financially burdensome but that treating false positive patients with anticoagulation can increase their risk for bleeding.

Objective: To determine the actual incidence of Type II-HIT compared to the frequency and necessity of testing in a single urban, teaching, community hospital in Chicago.

Methods: A retrospective review of lab records for patients who were tested for HIT syndrome by H-PF4 -ELISA through 2 years Jan 2013-Dec 2014. Then each patient's chart was reviewed. Patients were categorized according to requesting department (i.e. ICU, ED, Medical and surgical floors) then screened for date of admission, date of onset of platelet drop, degree of platelet drop, incidence of DVT, PE, arterial thrombosis and bleeding and finally for possible reasons of platelet count drop other than HIT for calculation of Pretest probability by the 4T score. According to the 4T criteria patients were categorized as low (score 0-3), intermediate (score 4, 5) and high probability (score 6-8). All charts with positive ELISA were reviewed for SRA (Serotonin Release Assay) results. Also contacted lab and pharmacy for cost of HIT testing, confirmatory test with SRA and the cost of Argatroban to be able to determine the economic burden of over testing for HIT syndrome.

Results: an average of about 20,000 admissions per year. Over the year 2013, 110 patients were screened for HIT. Over the year 2014, 87 patients were tested for HIT. Of total of 197 tests ordered over the 2 years period 19 Patients did not have enough data in the charts for determination of 4T score and were so excluded. Of the remaining 178, 85(45%) were sent from ICU, 61(32.6%) from Medical floors, 28(15%) from CCU, 9(4.8%) from Surgical floors, 1(0.5%) from Rehab unit, 1(0.5%) from ED and 1(0.5%) from Observation Unit. 17(9.5%) Patients had proven DVT, PE or arterial thrombosis. 2(1%) Patients had evidence of bleeding. According to 4T score139 (74.3%) of 178 were low, 34(18%) intermediate and 5(2.6%) were high probability. ELISA test was positive for only 34(19%) patients of the 178 of which only 3(1.6%) were proven true positive by SRA, 12(6%) had no SRA result in their chart and 19(10.6%) had negative SRA results. ELISA was positive in 22 cases with Low Probability score, 9 Intermediate probability and 3 high probability. Of the 22 low probability ELISA positive cases only 1 (4%) was SRA positive, 15(68%) were SRA negative and 6(27%) did not have SRA in the chart. ELISA test costs 233$, SRA test costs 50$ and one day of Argatroban costs average of 663.44$.Thus the calculated cost of testing for low probability patients reached 32,387 $ for ELISA and 800$ for SRA. Cost of treating false positive patients with Argatroban for an average of 3 days until SRA results are available is 43,758$.

Conclusion: From this study we concluded that in our facility we continue to do too many HIT studies without appropriate prescreening with added cost of testing and treatment for low probability patients of almost 77,000$ over the period of two years. In planning to avoid unnecessary testing and treatment for false positive patients we plan to build a 4T score calculator into our Electronic Medical Record System that is started once a HIT test is ordered to improve the screening process. We will also continue work with the Internal Medicine residency program and medical staff to improve teaching on HIT syndrome and other conditions with similar presentations.