Background:
Heparin-induced thrombocytopenia (HIT) is a prothrombotic and potentially fatal adverse drug reaction mediated by platelet-activating antibodies against multimolecular complexes of platelet factor 4 (PF4) and heparin. The 4Ts is a pretest scoring system for HIT that was developed to improve and standardize clinical diagnosis. The strength of the 4Ts score lies in the negative predictive value of a low pretest probability score. Those with a 4T score ≥4 is considered to have moderate to high probability of HIT and should undergo confirmatory laboratory test. Delays in treatment are associated with risk of thrombosis. Immediate cessation of heparin and initiation of a non-heparin anticoagulant is therefore standard of care.
Method:
This is a single institutional retrospective study. We reviewed the charts of 219 patients with thrombocytopenia with suspected HIT to evaluate the appropriateness of ordering the HIT testing based on the pretest probability of the 4Ts scoring system. All patients had enzyme linked immunosorbent assay (ELISA) test for HIT antibodies (the screening test). We studied tests that were sent from 03\16\2021 through 07\03\2023. The serotonin release assay (SRA) which is the confirmatory test for HIT was only sent for patients who had a positive ELISA test. 4T scores were calculated retrospectively using chart's information available to clinicians at the time of the HIT tests were sent.
Results:
Out of the 219 patients, 80(36.5%), 120(54.7%) and 19(8.6%) patients had a low, intermediate and high pretest probability respectively. The HIT ELISA test was positive in 33 (15.7%) patients and 196 (84.9%) patients had negative results. HIT ELISA had 100% and 93.5% sensitivity and specificity respectively while the positive and negative predicative values were 60% and 100% respectively. A positive SRA test was reported in 23 patients (10.5% of the total study population). Seventy-nine patients (36% of our total sample) scored 4 points which represents intermediate pretest probability. Out of these 79 patients only 6 (11.3%) patients had a positive ELISA test, and 5 (6.3%) patients had a positive SRA test. Aiming to minimize the HIT over testing we examined the increase in the cutoff point of the intermediate pretest probability from 4 points to 5 points which resulted in significant increase in the sensitivity and positive predictive value of the intermediate pretest probability (75.4% and 21% versus 39% and 12.9% respectively). However, the sensitivity dropped from 90% to 65%. The 4th part of 4Ts score gives points based on the probability of HIT versus other pathology as a cause of the thrombocytopenia. Most of the patients with suspected HIT will score one point out of the 4th part of 4Ts score which makes this part of the score is less helpful in distinguishing between patients with suspected HIT and will result in higher overall pretest probability and more testing. We studied the 3Ts score (the 4Ts scoring system after elimination of the 4th part) effect on the intermediate pretest probability. The cutoff points for the 3Ts score are 1-3, 4-5 and 6 for low, intermediate and high pretest probability respectively. The 3Ts score model resulted in significant increase in the specificity and the positive predictive value (76% and 21.7% versus 39.2% and 12.9% respectively) however the sensitivity dropped from 90% to 65%. The average turnaround time for HIT ELISA test and the SRA test was 2.3 days and 4.4 days respectively. Fifteen patients (21% out the patient had low pretest probability on the 4Ts score), 41 patients (59% of the patient who had an intermediate pretest probability on the 4Ts score) and 13 patients (18.8% of the patients who had high pretest probability on the 4Ts score) started on non-heparinoid anticoagulation meanwhile awaiting on the HIT testing result.
Conclusion:
Over testing of HIT is common secondary to underuse of 4Ts scoring system. Multicomponent educational interventions to increase the awareness and utility of the 4Ts score in addition to implantation of clinical decision support based on HIT computerized score will help to reduce unnecessary diagnostic testing of HIT in patients with low pretest probability. Increasing the cutoff point of the intermediate pretest probability or applying the 3Ts score model resulted in significant increase in the specificity and the positive predictive value of the 4Ts scoring system on expense of the scoring system sensitivity.
Disclosures
No relevant conflicts of interest to declare.
