Pitfalls of Clinical Diagnosis Strategies for Heparin-Induced Thrombocytopenia (HIT) After Cardiac Surgery (CS).

Abstract

Abstract 3516Poster Board III-453 BackgroundHIT is an infrequent but potentially serious complication of heparin therapy. Its diagnosis is complex and depends on a combination of clinical suspicion and laboratory confirmation through ELISA and functional tests such as the serotonin release assay (SRA). The 4Ts score comprises 4 clinical parameters (severity and timing of onset of thrombocytopenia, development of thrombosis, and clinician's appraisal of the likelihood of alternate causes for thrombocytopenia) and has been proposed to predict the probability of HIT in patients deemed to be at risk. However, the validity of the 4Ts score in patients undergoing cardiac surgery (CS) is questionable considering the numerous other factors that predispose such patients to thrombocytopenia and thrombosis. In addition, in CS patients the HIT ELISA assay has been reported to have 25 - 50% false positive results making it less useful. ObjectivesTo determine the usefulness of the 4T score in the post cardiac surgical population and the value of the HIT ELISA optical density for predicting HIT. MethodsRetrospective case-control study of patients admitted for cardiac surgery to the London Health Sciences Centre between January 2006 and December 2008 and for whom a HIT ELISA assay was requested. Patients with an equivocal or positive ELISA test were tested by SRA which considered the gold standard. Information collected included clinical variables related to the surgery and post-operative period, calculated 4T scores, ELISA optical density (OD) and SRA results. Categorical variables were compared using chi2 or Fisher's exact tests as appropriate. Continuous variables were compared using a Mann-Whitney U test. Covariates achieving a p value ≤0.1 in univariate analysis and the components of the 4Ts score were incorporated in logistic regression models using stepwise forward selection. Finally, we constructed a Receiver Operating Characteristic (ROC) curve for the ELISA OD. Results73 patients were included in the analysis. Results of the univariate analysis are shown in the table. On regression analysis only the ELISA optical density (per each OD arbitrary unit increase) was correlated with a diagnosis of HIT (OR 37.266; 95% CI 2.342-593.013; p=0.010). For the ELISA OD the area under the ROC curve was 0.990 (SE 0.013) (Figure). A cutoff value for the OD of 0.475 had a Sensitivity of 1, a specificity of 0.9, a positive likelihood ratio (LR) of 10 and a negative LR of 0.00. Assuming a prevalence proportion of 0.082 the posterior probability of HIT if the ELISA has an OD <0.475 is 0 (95% CI 0 – 9). On the other hand, an OD >0.92 resulted in a LR+ of 20 with a posterior probability of 64% (95% CI 35 – 80). ConclusionsIn this study, we found that the 4T score does not accurately predict HIT in post CS patients. Limitations of this study include a reduced sample size and its retrospective nature. Our findings suggest that in post CS patients developing thrombocytopenia between 10 and 100 × 109 or a platelet drop of 50% or more (100% of our population) a HIT ELISA with an OD < 0.475 could be used to rule out HIT. Our findings need to be confirmed in prospective studies.ControlsCasesPAge (years) [Median]74.00066.0000.175Gender1.000Male [%]58.20066.700Cardiac surgery type [%]0.556CABG22.433.3Valve replacement28.40CABG + Valve replacement43.366.7Other30Transplant30Use of intra-aortic balloon pump [%]26.933.30.663Clamp time (minutes) [Median]121.000112.0000.651Pump time (minutes) [Median]174.000167.0000.507Heparin used in surgery (units) [Median]55000.00085500.0000.007Baseline platelet count (x 10∧9) [Median]211.000250.0000.315Nadir platelet count [Median]39.00038.5000.912Percent platelet fall [Median]81.41080.7970.659Time from cardiac Sx to HIT ELISA request OR Plt Nadir (days) [Median]4.0007.0000.032Time to HIT ELISA /Plt nadir >=5 days [%]29.983.30.016Heparin exposure within 100 days [%]49.333.30.676Confirmed new thrombosis [%]033.30.006Suspected new thrombosis [%]1.501Skin necrosis [%]016.70.082Alternate cause for thrombocytopenia[%]0.221No34.30Possible20.933.3Definite44.866.74T Score Calculated [Median]4.0005.0000.0594Ts Score pretest probability category [%]0.125Low37.30Intermediate43.383.3High19.416.7HIT ELISA Assay<0.001Negative80.60Equivocal11.90Positive7.5100ELISA optical density (arbitrary units) [Median]0.1702.306<0.001 Disclosures:No relevant conflicts of interest to declare.