Abstract Number: PB1055
Meeting: ISTH 2022 Congress
Background: Heparin-induced thrombocytopenia is a common, potentially lethal, clinical issue in intensive care setting.
However, while HIT three-step diagnostic work-up is well defined, there is no way to easily quantify how likely a patient has HIT.
Aims: We developed an iOS and Android based app (NomoHIT) for quantifying HIT diagnosis probability at patient’s bedside, by fusing 4Ts score, immunoassay and functional test results.
Methods: The method is a digital equivalent of Fagan nomogram which is a graphical tool for quantifying the probability that a patient has the disease, on the basis of test performance and pre-test probability of the condition.
The Fagan nomogram belongs to the statistical field of Bayesian inference. It is however difficult to apply in practice, especially to a three-step diagnostic workup.
The app we developed consists in successive screens (Figures 1 and 2) with query fields to be sequentially completed with 4Ts score output, and the results of immunoassay and of functional assay, after having specified the sensitivity and specificity of the used assays.
The app then applies the well-known Bayes’ formula to compute the probability of positive diagnosis, by using each post-test probability from a prior step as pre-test probability for the following step.
Results: Results will be exemplified through an interactive interface.
Interestingly enough, the app displays the intermediate probabilities, enabling the physician to assess the information gain she may expect from moving forward into the diagnostic workup.
Conclusion(s): Bayesian approach provides clinicians with a systematic framework for estimating the probability of HIT in their patients, to help in informing their medical decision making.
The app we developed offers the lab specialists and physicians a user-friendly tool to benefit from the Bayesian approach in practice.
Figure 1
Example of how quieries and readouts are displayed in the app
Figure 2
Example of how pre-test and post-test probabilities are displayed in the app
To cite this abstract in AMA style:
Allemand F, Stoll M. “Does my patient have HIT? There should be an app for that.” [abstract]. https://abstracts.isth.org/abstract/does-my-patient-have-hit-there-should-be-an-app-for-that/. Accessed August 9, 2022.« Back to ISTH 2022 Congress
ISTH Congress Abstracts - https://abstracts.isth.org/abstract/does-my-patient-have-hit-there-should-be-an-app-for-that/