APTTO: a clot waveform analysis-based algorithm for diagnostic triage of prolonged activated partial thromboplastin...

Area of Science:

• Hematology
• Clinical Diagnostics
• Coagulation Science

Background:

• Prolonged activated partial thromboplastin time (APTT) is a frequent lab finding leading to extensive, often unnecessary, diagnostic workups.
• Many patients with prolonged APTT do not have a clinically significant bleeding disorder, causing diagnostic delays.
• Clot waveform analysis (CWA), a dynamic coagulation assessment during APTT, is underutilized for prolonged APTT evaluation.

Purpose of the Study:

• To develop and validate a CWA-based algorithm (APTTO) for evaluating prolonged APTT.
• To create a two-step model to identify patients unlikely to have etiological findings and differentiate specific coagulation disorders.

Main Methods:

• A multicenter cohort with prolonged APTT and normal prothrombin time was used to extract CWA features.
• Two logistic regression models (APTTO1 and APTTO2) were developed to stratify patients based on risk and specific diagnoses.
• Model performance was assessed using discrimination, calibration, and decision-curve analysis (DCA) in a temporally independent cohort.

Main Results:

• APTTO1 showed excellent discrimination (AUC 0.88) and high sensitivity (99.5%) and NPV (95.2%) for identifying patients without etiological findings.
• APTTO2 effectively differentiated lupus anticoagulant (LA) from factor deficiencies or von Willebrand disease (VWD) (AUC 0.86).
• The APTTO algorithm demonstrated clinical utility by reducing the need for etiologic testing while maintaining a high NPV.

Conclusions:

• The APTTO algorithm is a pragmatic, CWA-based tool for evaluating prolonged APTT, offering efficient diagnostic triage.
• The algorithm showed robust performance in reducing unnecessary testing and improving diagnostic efficiency.
• Further prospective external validation is recommended before widespread clinical implementation.