Biological variation outweighs pre-analytical factors in coagulation reference intervals: a multicenter big data...

Area of Science:

• Clinical Chemistry
• Hematology
• Medical Diagnostics

Background:

• Establishing reference intervals (RIs) for coagulation tests like PT and APTT is difficult due to pre-analytical variability and donor recruitment challenges.
• This study addresses the need for accurate RIs by employing a "Big Data" indirect approach across a harmonized hospital network.

Purpose of the Study:

• To define RIs for Prothrombin Time (PT) and Activated Partial Thromboplastin Time (APTT) using an indirect "Big Data" method.
• To evaluate the impact of pre-analytical and biological variables on these coagulation test RIs.

Main Methods:

• A multicenter retrospective study analyzed data from over 300,000 patients across nine hospitals using the refineR algorithm.
• Local estimates were aggregated via random-effects meta-analysis and validated against direct methods on healthy donors.

Main Results:

• Indirect estimates strongly correlated with direct methods.
• Harmonized upper reference limits (URLs) were established for APTT ratio (1.19) and PT ratio (1.25, or 1.17 excluding inpatients).
• Age and sex significantly influenced PT, with a J-shaped increase after 50 and higher values in males.

Conclusions:

• Harmonizing RIs across hospital networks is achievable, with minimal impact from technical variations.
• Age and sex stratification for PT RIs in patients over 50 may enhance diagnostic specificity.