A machine-learning model for reducing misdiagnosis in heparin-induced thrombocytopenia: A prospective, multicenter, observational study

Henning Nilius; Adam Cuker; Sigve Haug; Christos Nakas; Jan-Dirk Studt; Dimitrios A Tsakiris; Andreas Greinacher; Adriana Mendez; Adrian Schmidt; Walter A Wuillemin; Bernhard Gerber; Johanna A Kremer Hovinga; Prakash Vishnu; Lukas Graf; Alexander Kashev; Raphael Sznitman; Tamam Bakchoul; Michael Nagler

doi:10.1016/j.eclinm.2022.101745

A machine-learning model for reducing misdiagnosis in heparin-induced thrombocytopenia: A prospective, multicenter, observational study

EClinicalMedicine. 2022 Nov 24:55:101745. doi: 10.1016/j.eclinm.2022.101745. eCollection 2023 Jan.

Authors

Henning Nilius¹, Adam Cuker², Sigve Haug^{3

4}, Christos Nakas^{1

5}, Jan-Dirk Studt⁶, Dimitrios A Tsakiris⁷, Andreas Greinacher⁸, Adriana Mendez⁹, Adrian Schmidt¹⁰, Walter A Wuillemin¹¹, Bernhard Gerber¹², Johanna A Kremer Hovinga¹³, Prakash Vishnu¹⁴, Lukas Graf¹⁵, Alexander Kashev³, Raphael Sznitman¹⁶, Tamam Bakchoul¹⁷, Michael Nagler¹

Affiliations

¹ Department of Clinical Chemistry, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland.
² Department of Medicine and Department of Pathology and Laboratory Medicine, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, USA.
³ Mathematical Institute, University of Bern, Bern, Switzerland.
⁴ Albert Einstein Center for Fundamental Physics and Laboratory for High Energy Physics, University of Bern, Bern, Switzerland.
⁵ Laboratory of Biometry, School of Agriculture, University of Thessaly, Volos, Greece.
⁶ Division of Medical Oncology and Hematology, University and University Hospital Zurich, Zurich, Switzerland.
⁷ Diagnostic Haematology, Basel University Hospital, Basel, Switzerland.
⁸ Institut für Immunologie und Transfusionsmedizin, Universitätsmedizin Greifswald, Greifswald, Germany.
⁹ Department of Laboratory Medicine, Kantonsspital Aarau, Aarau, Switzerland.
¹⁰ Clinic of Medical Oncology and Hematology, Municipal Hospital Zurich Triemli, Zurich, Switzerland.
¹¹ Division of Hematology and Central Hematology Laboratory, Cantonal Hospital of Lucerne and University of Bern, Switzerland.
¹² Clinic of Hematology, Oncology Institute of Southern Switzerland, Bellinzona, Switzerland.
¹³ Department of Hematology and Central Hematology Laboratory, Inselspital, Bern University Hospital, University of Bern, Bern, Switzerland.
¹⁴ Division of Hematology, CHI Franciscan Medical Group, Seattle, United States.
¹⁵ Cantonal Hospital of St Gallen, Switzerland.
¹⁶ ARTORG Center for Biomedical Engineering Research, University of Bern, Bern, Switzerland.
¹⁷ Centre for Clinical Transfusion Medicine, University Hospital of Tübingen, Tübingen, Germany.

Abstract

Background: Diagnosing heparin-induced thrombocytopenia (HIT) at the bedside remains challenging, exposing a significant number of patients at risk of delayed diagnosis or overtreatment. We hypothesized that machine-learning algorithms could be utilized to develop a more accurate and user-friendly diagnostic tool that integrates diverse clinical and laboratory information and accounts for complex interactions.

Methods: We conducted a prospective cohort study including 1393 patients with suspected HIT between 2018 and 2021 from 10 study centers. Detailed clinical information and laboratory data were collected, and various immunoassays were conducted. The washed platelet heparin-induced platelet activation assay (HIPA) served as the reference standard.

Findings: HIPA diagnosed HIT in 119 patients (prevalence 8.5%). The feature selection process in the training dataset (75% of patients) yielded the following predictor variables: (1) immunoassay test result, (2) platelet nadir, (3) unfractionated heparin use, (4) CRP, (5) timing of thrombocytopenia, and (6) other causes of thrombocytopenia. The best performing models were a support vector machine in case of the chemiluminescent immunoassay (CLIA) and the ELISA, as well as a gradient boosting machine in particle-gel immunoassay (PaGIA). In the validation dataset (25% of patients), the AUROC of all models was 0.99 (95% CI: 0.97, 1.00). Compared to the currently recommended diagnostic algorithm (4Ts score, immunoassay), the numbers of false-negative patients were reduced from 12 to 6 (-50.0%; ELISA), 9 to 3 (-66.7%, PaGIA) and 14 to 5 (-64.3%; CLIA). The numbers of false-positive individuals were reduced from 87 to 61 (-29.8%; ELISA), 200 to 63 (-68.5%; PaGIA) and increased from 50 to 63 (+29.0%) for the CLIA.

Interpretation: Our user-friendly machine-learning algorithm for the diagnosis of HIT (https://toradi-hit.org) was substantially more accurate than the currently recommended diagnostic algorithm. It has the potential to reduce delayed diagnosis and overtreatment in clinical practice. Future studies shall validate this model in wider settings.

Funding: Swiss National Science Foundation (SNSF), and International Society on Thrombosis and Haemostasis (ISTH).

Keywords: Anticoagulants; Diagnosis; Heparin; Heparin-induced thrombocytopenia; Low-molecular-weight; Platelet count; Thrombocytopenia.