Comparison of Interobserver Agreement between the 4T and the HIT Expert Probability (HEP) Scores: We Agreed Not to Agree

Abstract

Background: Diagnosis of heparin induced thrombocytopenia (HIT) is challenging. The 4T score is the principal scoring system for HIT. The HIT Expert Probability (HEP) score, based on expert opinion, was proposed in 2010. Both scores have a very high sensitivity, being capable of ruling out HIT in low-risk individuals. However inter-observer agreement (IOA) of HEP score in comparison to the 4T score remains uncertain. Aim: This study aimed to compare IOA between the 4T and the HEP score among hospitalized patients. Specific objectives: Comparison of IOA between the 4T and the HEP score in the 1-whole cohort 2-non-ICU 3- ICU Identify which item of the score have the lowest IOA (magnitude of thrombocytopenia, timing of thrombocytopenia, existence of other causes of thrombocytopenia, presence of thrombosis) Methods: Patients: This is a retrospective study at a single tertiary hospital. All inpatients 16 years or older who have been tested for HIT between June 2019 to end of June 2021 were included. Patients were excluded if the reason for sending HIT was a repeat test. Data were collected retrospectively by manual chart review. Assessment of HEP score and 4Ts score: Two hematologists rated the HEP score and 4Ts score independently for every patient. The score's calculation was blinded between observers and, they were blinded to the HIT Eliza result. Low 4Ts score is defined as score less than 4 and for HEP score if less than 3. HIT test was done using Stago Asserachrom HPIA-IgGAM ELISA with level of < 0.4 considered negative. Statistical analysis: Continuous variables are reported as mean ± standard deviation (SD) or median (interquartile range; IQR), as appropriate. Categorical variables are reported as frequency and percentage. The Kappa coefficient (к) with 95 % confidence interval (CI) was used to assess the IOA of 4T score and HEP score, it was calculated for the individual risk categories - low vs high score in: 1-the whole patients 2-ICU 3- non-ICU We also calculated the IOA for each item of the score, magnitude of thrombocytopenia, timing of thrombocytopenia, existence of other causes of thrombocytopenia, presence of thrombosis (as total score). A P-value less than 0.05 was considered significant. Statistical analyses were performed using SAS version 9.4 (SAS institute, CARY, NC, USA). The study was approved by the local ethical board. Informed consent was waived. Results: After exclusion of the repeat test (13), 302 patients were included in this study. Median age 66 years old. Most common reason for request was thrombocytopenia 288 (95%). HIT ELISA ≥ 0.4 in 34 (11%). Inter-observer agreement: All patients: We found similar IOA between the 4T score and the HEP score к (95 % CI) of 4T score was .14(-0.13-0.41, P < 0.01), compared 0.17(-0.16-0.5, P < 0.01) in the HEP score. Non-ICU patients: The IOA was similar between the 4T score and the HEP score, к of 4T score was 0.24(0.11-0.38, P < 0.01) compared 0.17(0.03-0.31 P < 0.01) in the HEP score. ICU patients: The IOA of 4Tscore was lower than HEP score к of 4T score was -020(-0.20.16, P = 0.83) compared 0.16(-0.037-0.37, P = 0.86) in the HEP score. Among the four individual items of 4T score, “existence of other causes of thrombocytopenia” achieved the lowest к with 0.03 and 0.1 in 4Tscore and HEP score respectively. Conclusion: In this retrospective study, we found a lower interobserver agreement of the 4T score compared to the HEP score in ICU patients. The poor IOA of the 4T score in ICU patients is attributed to worse agreement in (existence of other causes of thrombocytopenia item). Ascertaining the cause of thrombocytopenia in ICU patients is challenging and the 4T score seems to depend on physician gestalt rather than objective points. Our data suggests that the HEP score is the preferred score in in patients with critical illness.