Diagnosing Heparin-Induced Thrombocytopenia: How Accurate Is the Recommended Algorithm?
Sensitivity and specificity were high, but a small proportion of patients assessed as low risk might deserve further testing.
Heparin-induced thrombocytopenia (HIT) is a high-risk condition that requires immediate therapy with high-risk medications; thus, accurate diagnosis is critical. Because HIT-antibody testing has a relatively high false-positive rate, guidelines direct clinicians to use an algorithm that starts with the 4Ts score and then recommends antibody testing in patients whose 4Ts scores indicate intermediate or high risk for HIT. In this prospective study of 1300 patients with possible HIT, investigators assessed the accuracy of the 4Ts score and the guideline-recommended diagnostic algorithm. The referencestandard test for comparison was a heparin-induced platelet activation assay; prevalence of HIT was 8.4% by reference-standard testing.
Positive predictive values of 4Ts score and the recommended diagnostic algorithm were expectedly low (15% and 66%, respectively). However, nearly half of patients were low risk using 4Ts score and would not have warranted antibody testing by the recommended diagnostic algorithm. That algorithm had high sensitivity (87%) and specificity (96%) and a very high negative predictive value (99%).
COMMENT
Because clinicians order reference-standard testing (i.e., serotonin release assay or heparin-induced
platelet activation assay) to confirm positive HIT-antibody testing, false positives by the recommended
algorithm eventually get resolved. However, 9% of patients with confirmed HIT (10 of 111) had falsenegative
(i.e., “low-risk”) 4Ts scores in this study. Additional testing still should be considered if suspicion
for HIT remains despite a low-risk 4Ts score; for example, a score of 3 points is technically
“low risk” but might warrant further testing based on clinical gestalt.
— Daniel D. Dressler, MD, MSc, MHM, FACP
Dr. Dressler is Professor of Medicine at Emory University School of Medicine, Atlanta.
Larsen EL et al. Accuracy of diagnosing heparin-induced thrombocytopenia. JAMA Netw Open 2024 Mar; 7:e243786. (https://doi.org/10.1001/jamanetworkopen.2024.3786)
GKB-NON-2024-00279