反復無刺激唾液採取：診断精度を向上させる信頼性の高いアプローチ

Unstimulated whole saliva (UWS) sialometry is fundamental for diagnosing salivary gland hypofunction, but single-timepoint measurements are limited by physiological variability.

To evaluate intraindividual variability and reproducibility of a repeated-measures UWS collection protocol in healthy young adults, and to propose a standardised approach to improve diagnostic accuracy.

Sixty-two healthy subjects (19-33 years old) collected UWS by passive drooling three times daily (morning, afternoon and evening) over three consecutive days (nine timepoints). Data were analysed using linear mixed-effects models, intraclass correlation coefficients (ICC), Bland-Altman plots and coefficients of variation (CV%).

Mean UWS flow rate collected over 5 min remained stable across days (0.38-0.39 mL/min), ranging from 0.14 to 0.93 mL/min. Aggregated daily means showed the highest reproducibility (ICC₃ = 0.890 for Days 1-2), while single timepoints were less consistent (afternoon ICC₃ = 0.524). Evening collections demonstrated the best reproducibility among individual timepoints (ICC₃ = 0.751). Bland-Altman analysis indicated low bias and acceptable agreement across days; within-subject variability (CVw%) was 32.24%.

Single UWS measurements are influenced by substantial physiological variability, which may lead to diagnostic misclassification. Averaging multiple standardised collections, particularly in the evening, enhances reliability and may optimise diagnostic accuracy. This protocol supports refinement of clinical guidelines and the implementation of home-based saliva testing for salivary gland dysfunction.