Improving diagnostic accuracy in pregnancy with individualised, gestational age-specific reference intervals.

Area of Science:

• Clinical Chemistry
• Maternal-Fetal Medicine
• Hematology

Background:

• Interpreting laboratory results during pregnancy requires pregnancy-specific reference intervals (RIs).
• Fixed RIs may lack accuracy due to the dynamic physiological changes throughout gestation.
• Gestational age-specific RIs are hypothesized to improve diagnostic precision.

Purpose of the Study:

• To evaluate the diagnostic accuracy of gestational age-specific platelet RIs.
• To compare the predictive value of early pregnancy low platelet counts using centile-based RIs versus a fixed threshold for term thrombocytopaenia.

Main Methods:

• Analysis of platelet count trajectories in 32,778 pregnant women using 116,798 results.
• Mapping platelet changes from 8 to 40 weeks of gestation.
• Assessing the accuracy of predicting term thrombocytopaenia (<100 × 10⁹/L) based on early pregnancy platelet counts (<3rd centile) compared to a fixed limit (<150 × 10⁹/L).

Main Results:

• Platelet counts decreased by 14.8% between 8 and 40 weeks of gestation.
• Platelet counts below the 3rd centile before 20 weeks accurately predicted thrombocytopaenia at term (AUC 0.86).
• This prediction was significantly more accurate than using a fixed limit (AUC 0.76, p=0.004).

Conclusions:

• Routinely collected hospital data can establish pregnancy-specific RIs, enabling detailed analysis of gestational changes.
• Individualized RIs derived from such data offer superior accuracy compared to fixed or general pregnancy-specific limits.
• Clinicians should consider implementing individualized RIs to enhance diagnostic accuracy for various biomarkers in clinical practice.