Current Diagnostics May Misclassify Suspected Heparin-Induced Thrombocytopenia

Published on: 

When the current recommended diagnostic algorithm was used in clinical practice, antibody testing was required in half of patients with suspected HIT.

A recent study examining the accuracy of diagnosing suspected heparin-induced thrombocytopenia (HIT) found only half of clinically suspected cases required antibody testing using the currently recommended diagnostic algorithm.1

However, this analysis, conducted across 11 study centers in Switzerland, Germany, and the United States, showed a notable percentage of patients with suspected HIT were not properly classified, which may lead to delayed diagnosis or change of the anticoagulant.

“This exposes those often-vulnerable patients to the risk of serious thromboembolic complications or bleeding complications due to overtreatment with risky anticoagulants,” wrote the investigative team, led by Michael Nagler, MD, PhD, department of clinical chemistry, Inselspital, Bern University Hospital, and the University of Bern. “The utilization of a clinical decision support (CDS) can potentially improve the diagnostic algorithms for HIT.”

A life-threatening complication after heparin administration, HIT is uncommon and often unfamiliar to practicing clinicians across diverse settings. Evidence has pointed to the vital importance of knowing when to suspect HIT and how to treat it properly to avoid severe thromboembolic complications.

However, the utility of recommended diagnostic tests can often differ in daily practice from published study findings—Nagler and colleagues noted there is limited knowledge on the utility of diagnostic tests used to recognize HIT in clinical practice.

For this study, the team sought to evaluate the diagnostic accuracy of currently recommended tests for suspicion of HIT in clinical practice, including the 4Ts score, the AcuStar HIT-IgC chemiluminescent immunoassay (CLIA), and the recommended diagnostic algorithm serially combining the results of both tests.

The analysis occurred as part of Towards Precise and Rapid Diagnosis of HIT: A Prospective, Multi-Center Cohort Study (TORADI-HIT). Between January 2018 and May 2021, participants with suspected HIT, aged ≥18 years, and provided informed consent were recruited and enrolled in the study. Specially trained study nurses collected a set of prespecified clinical characteristics and laboratory test results at diagnosis into an electronic case report form.

Within 1 week after attaining a residual serum sample, investigators assessed platelet factor 4 (PF4)/heparin antibodies using a CLIA. Moreover, as a reference standard, the presence of HIT was determined using a washed-platelet heparin-induced platelet activation (HIPA) test.

Between the study period, 1448 patients were included from 11 study centers—after exclusions due to insufficient serum samples or lack of clinical information, 1318 patients were left eligible for the current analysis. The median age was 67 and 849 [64.6%] patients were male.

Overall, HIPA was positive in 111 patients, for a prevalence of 8.4% in the study population. Data showed the most common settings for HIPA were the intensive care unit (487 [37.0%]) or cardiovascular surgery (434 [33.0%]). The 4Ts score was determined to be low risk in 625 patients (46.8%), intermediate risk in 611 patients (46.9%), and high risk in 82 patients (6.2%).

Upon analysis, the 4Ts score correctly classified 101 patients as HIT positive and 615 as HIT negative. The number of patients with false negatives and false positives was 10 (9.0%) and 592 (49.0%), respectively. Meanwhile, the CLIA correctly categorized 106 patients as HIT positive and 1134 as HIT negative, with numbers of false negatives and false positives being 5 (4.5%) and 73 (6.0%), respectively.

The currently recommended diagnostic algorithm, consisting of the 4Ts score followed by CLIA, correctly assessed 96 patients as HIT positive and 1157 as HIT negative. The number of false negative and false positives were 15 (13.5%) and 50 (4.1%), respectively.

To address the limited sensitivity of the current algorithm, Nagler and colleagues validated a machine-learning algorithm for patients with suspected HIT using the study cohort. Their findings showed the TORADI-HIT algorithm was significantly more accurate than the currently recommended diagnostic algorithm.

“The algorithm reduces the number of patients with false-positive and false-negative results so that functional assays are necessary in approximately 10% of HIT patients only,” they wrote. “Prospective cohort studies are currently running to validate the algorithm in other settings and situations.”


Larsen EL, Nilius H, Studt J, et al. Accuracy of Diagnosing Heparin-Induced Thrombocytopenia. JAMA Netw Open. 2024;7(3):e243786. doi:10.1001/jamanetworkopen.2024.3786