Background/Purpose:

Heparin induced thrombocytopenia (HIT) is the destruction of platelets in patients exposed to heparin products. While there are two types of HIT, the term is often used without modifier to refer to HIT Type II, which is immune-mediated and severe. Up to 5% of patients on unfractionated heparin and 1-2% of patients on low molecular weight heparin will develop HIT Type II, with 30-80% of them developing thrombotic sequelae. The gold standard for treatment is to stop heparin products and start a different anticoagulant. Although not explicitly indicated, steroids are often incidentally given to suppress the immune-mediated destruction of platelets. However, the impact of steroids on patient outcomes, such as length of stay, mortality, and adverse events like bleeding or thrombosis, is unclear. A literature review of steroid use patients with HIT in the ICU setting did not show any definitive studies looking at this association. We believe that investigating incidental steroid use in this patient population could clarify its role and impact and help us develop better guidelines for treatment.

Methods:

We conducted a retrospective analysis using the MIMIC-III Critical Care Database. For our primary endpoint, we selected patients diagnosed with HIT using ICD-9 codes, filtered for earliest admission date and divided them into two groups for comparison: one group who received steroids, and the other group who did not receive steroids. Our sample is composed of 2,205 patients without incidental steroid use and 753 with incidental steroid use, with a total of 2,958. The chart was also abstracted for variables such as length of stay, admit type, mortality, and bleeding, as well as age (stratified by: neonate, 17-29, 30-49, 50-63, 64-75, 76+) and sex (male vs. female). Analysis was performed using chi-square test for categorical variables, Fisher’s exact test for the admit type variable, and unpaired 2-sided t-test with unequal variances for the duration of stay variable

Results:

Holding all other variables constant, we found that patients who were prescribed steroids had longer hospital stays by 6.5 days on average, when compared to patients who were not prescribed steroids (p<0.001, 95% CI = 5.30 to 7.67). Moreover, patients who were prescribed steroids were almost twice as likely to experience mortality compared to those who were not prescribed steroids (p<0.001, 95% CI = 1.63 to 2.45), and have increased odds of bleeding by 33% compared to those who were not given steroids (p=0.004, 95% CI = 1.10 to 1.62).

Conclusion:

This study represents the first attempt to discern the impact of steroid treatment on outcomes in patients with HIT Type II in the ICU setting. Given the results, it seems that steroid use is associated with worsened outcomes, such as longer length of stay, increased mortality, and increased bleeding. Although conventional wisdom has convinced us to give steroids in a large proportion of HIT cases, this study suggests that overall, steroid use in HIT patients may be more harmful than beneficial. We recommend that there be further investigation into the use of steroids in HIT, with demographic stratification, so that we can develop appropriate best practice guidelines for treatment.

« Back to 2018 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/incidental-steroid-use-may-worsen-outcomes-in-patients-with-heparin-induced-thrombocytopenia-in-the-icu-setting/