Appropriateness of Heparin-Induced Thrombocytopenia (HIT) Testing and the Reliability of 4-Ts Scoring in Critically Ill Patients

Introduction:

Thrombocytopenia is a common occurrence in critically ill patients and results in an array of diagnostic workup. Heparin induced thrombocytopenia (HIT) is one of the most frequently ordered tests in this scenario, although it is a less common cause (0.02-0.5%) The over testing and treatment of HIT in Intensive Care Unit (ICU) can lead to a change in medications, and result in side effects, increased ICU length of stay and increased costs. Current guidelines recommend 4-Ts criteria to evaluate the clinical HIT probability in general population, but it may also be used in Intensive Care Unit (ICU) patients. Here, we want to analyze the the usage of 4Ts score in a community hospital and to understand its correlation with HIT testing to see if HIT testing could have been avoided.

Methods:

We performed a retrospective chart review of adult patients who underwent testing for Heparin Induced Thrombocytopenia anti-platelet factor 4 antibody (anti- PF4 Ab) in ICU at our institution from 03/01/2012 to 02/01/2018. The primary outcomes were set to identify (1) the extent of inappropriate of HIT testing and its consequences in patients with low 4Ts scores, and (2) if 4Ts score as a potential predictor of length of stay and mortality in ICU. As a secondary outcome, we assessed if Body Mass Index (BMI) and Recent Surgery are related to false positive anti-PF4 Ab test. Kruskal-Wallis test and Fisher Exact test were used.

Results:

A total of 66 patients were identified. The mean age was 64 yr and the male-female ratio was similar. 40 patients had recent surgery in the past 3 months. The cause of thrombocytopenia other than HIT was possible or definite in 62 patients with sepsis being the most common cause (56.45%). 30 patients had low probability 4-Ts scores, 28 had intermediate and 7 had high probability. HIT was confirmed (Serotonin Release Assay-SRA positive) in 0%, 7.14% and 12.5% of patients with low, intermediate and high probability 4Ts scores, respectively. In the patients with low probability 4Ts and anti-PF4 Ab testing, heparin was held in 24 (80%) and the anti-coagulation was switched (to either fondaparinux or argatroban) in 10 (33.33%). In those patients with anticoagulation switch, 3 had minor bleeding.

There was no difference in the ICU length of stay (p=0.1712) and mortality (p=0.149) between patients with low, intermediate and high probability 4Ts scores. Patients with BMI ≥40 kg/m2 had twice the percentage of false positive anti-PF4 Ab testing compared with the rest, but the results were not significant (p=0.285). Recent surgery was not related to the false positive anti-PF4 Ab testing (p=1).

Conclusion:

As no patient with low-probability 4Ts score had confirmed HIT, low 4Ts may be a good estimator of the unlikeliness of HIT even in critically ill patients. We found discrepancy in the recommendations and usage of Anti-PF4 Ab testing in our institution. This has resulted in switching of heparin to other, more expensive anti-coagulations. This may very well be the situation in many other similar institutions. This could partly be due to lack of awareness of HIT testing guidelines in the critical care setting. We recommend the use of 4Ts scoring and more sensible testing and treatment of HIT in ICU patients. Although we could not demonstrate the statistical association between morbid obesity and false positive anti-PF4 Ab, our study might have been underpowered by low subgroup population of such patients. We propose further studies about HIT testing and treatment to include morbidly obese patients as a separate sub-group.