A 2 Year of Review of HIT Evaluations at an Urban Medical Center with an Aim Towards Quality Improvement and the Development of Practical Testing Strategies

Introduction:

Thrombocytopenia is a common occurrence among inpatients while heparin-induced thrombocytopenia (HIT) is relatively rare and diagnostic strategies can vary. Our goal was to evaluate testing habits over a two year period at an urban tertiary care center.

Methods:

All cases with HIT panels over a two year period were reviewed to elucidate ordering patterns, test turnaround times, and areas for potential improvement. Sensitivities and specificities were determined for the 4T score, which was retrospectively calculated for all patients, the heparin/PF4 ab immunoassay (H/PF4), the SRA functional assay (SRA) and a combined low risk 4T and H/PF4 OD ratio less than 0.4. Average and median OD ratios were calculated for different levels of risk as determined by 4 T score. Given the lower risk of HIT in patients exposed to low molecular weight heparin (LWMH) (Martel et al. 2005) an exploratory analysis was made by which one point was deducted from the 4T for patients who were only exposed to LMWH during the HIT work-up admission.

Results:

245 patients received work-ups for HIT that included either a H/PF4 and/or an SRA, with 7 patients ultimately being determined to have HIT (yield of 2.86%). Workup issues included 28 incorrect tests being sent and a lack of 4T score documentation in the charts reviewed (15% with documentation, 38/245). Given that hematology consults were obtained in 60% (147/245) of the cases this was not a problem specific to primary inpatient teams.

The sensitivities (SN) and specificities (SP) were as follows: for a low risk 4T score (SN 100%, SP 61.76%), for a low risk 4T after subtracting one point for sole exposure to LMWH (SN 100%, SP 63.03%), for the H/PF4 using Weak Positive as negative (SN 100%, SP 74.84%) for an OD<0.4 as negative (SN 100%, SP 79.05%), for the combined low risk 4T and OD<0.4 (SN 100%, SP 84.62%), for the SRA using Indeterminate as negative (SN 83.33%, SP 100%). Mean and median OD ratios by 4T risk category can be seen in the accompanying table and by score in the accompanying graph.

Our calculated turnaround times for these tests were similar; 156 H/PF4 tests ordered for 146 patients took an average of 3.01 days to result, 229 SRA tests ordered for 212 patients took an average of 3.21 days to result. For the 55 HIT negative patients that ultimately received treatment the average length of therapy for those treated with argatroban was 5.11 days and for fondaparinux 7.11 days.

Discussion:

In an effort to improve the 4T score subtraction of 1 point for exposure to only LMWH was investigated with minimal improvement in specificity. However, 19 of the 23 patients in this category had a 4T score of four or less and this adjustment could have helped to avoid a sizable amount of subsequent testing (18 SRA's), imaging (5 studies) and alternate therapies (9 days total) if it moved these patients from intermediate to low risk.

As is true in many hospitals, H/PF4's and SRA's are both send out tests in our institution and thus often ordered simultaneously. High turnaround times are also present in institutions where these tests are "batched" and only performed on certain days. With the small difference in test turnaround times it is worth questioning whether both tests are necessary to send concurrently in every case.

One setting to evaluate simultaneous testing is in patients who are low risk by 4T and thus not switched from heparin to alternative agents. In these cases it is practical to not order any testing given the sensitivity of a low risk 4 T score. But when testing is opted for ordering only the H/PF4 and reserving the SRA for cases with OD ratios greater than 0.4 may be prudent. This is supported by the combined low risk 4T H/PF4 testing characteristics (Kim et al 2011). This strategy is already recommended but in institutions with long turnaround times simultaneous testing is much more common.

In our cohort, 70 patients with low risk 4T scores had SRA's sent. Of these 70 patients, none of whom had HIT, 57 (81.4%) had H/PF4 ab OD ratios less than 0.40 and would not have needed additional testing. 57 SRA's could have been avoided and given that heparin can almost always be safely continued in these patients the potential costs of continuing alternate therapy are virtually non-existent.

Conclusion:

The low-cost, highly sensitive 4T remains a vital screening tool for HIT and while a low risk score should obviate the need for further testing our data supports that in cases where more testing is pursued a H/PF4 alone is the correct option.