唾液是一种对患者友好的治疗药物监测 (TDM) 基质，但在常规护理中很少使用。这是由于基于唾液的 TDM 结果无法确定剂量。本研究旨在检索唾液血浆浓度数据，并随后确定影响药物排泄到唾液中的理化特性，以增加基于唾液 TDM 的基础知识。检索了 Medline、Web of Science 和 Embase (1974-2023)用于人体临床研究，确定唾液和血浆中的药物药代动力学。包括至少十名受试者和每个受试者五对唾液血浆浓度的研究。对于每项研究，确定唾液和血浆之间浓度-时间曲线下面积的比率以评估唾液中的排泄。每种药物的理化特性（例如 pKa、亲脂性、分子量、极性表面积、可旋转键和未与血浆蛋白结合的药物分数）均从 PubChem 和 Drugbank 获得。药物根据其电离性进行分类，然后通过 Henderson-Hasselbalch 方程调整蛋白质结合和生理 pH 值来预测唾液与血浆的比率。对每种药物类别进行 Spearman 相关分析，以确定预测唾液排泄的因素 (α = 5%)。研究质量通过干预工具非随机研究的偏倚风险进行评估。总体而言，纳入了 42 项研究，包括 40 种药物（抗精神病药物、抗微生物药物、免疫抑制剂、抗血栓药物、抗癌药物和心脏药物）。两性组（0.59）、碱性组（0.43）和酸性组（0.41）药物的唾液与血浆比率中值相似，中性组药物最低（0.21）。唾液中酸性药物的排泄量 (n = 5) 较高，与较低的电离和蛋白质结合相关（预测与观察到的唾液与血浆比率之间的相关性：R2 = 0.85，p = 0.02）。对于碱性药物 (n = 21)，pKa 预测唾液排泄（Spearman 相关系数：R = 0.53，p = 0.02）。对于两性药物 (n = 10)，氢键供体 (R = - 0.76，p = 0.01) 和极性表面积 (R = - 0.69，p = 0.02) 是预测因子。对于中性药物 (n = 10)，蛋白质结合 (R = 0.84，p = 0.004)、亲脂性 (R = - 0.65，p = 0.04) 和氢键供体计数 (R = - 0.68，p = 0.03) 是预测因子。被认为可能适合基于唾液的 TDM 的药物有苯妥英、他克莫司、伏立康唑和拉莫三嗪。这些研究具有低至中度的偏倚风险。许多常用药物会排泄到唾液中，这可以通过药物的电离状态、蛋白质结合、亲脂性、氢键供体计数和极性表面积来部分预测。需要评估药物转运蛋白和生理因素对排泄的贡献。对可能适合基于唾液的 TDM 的药物的持续研究将有助于采用这种以人为本的 TDM 方法来改善患者的治疗结果。© 2024。作者。

Saliva is a patient-friendly matrix for therapeutic drug monitoring (TDM) but is infrequently used in routine care. This is due to the uncertainty of saliva-based TDM results to inform dosing. This study aimed to retrieve data on saliva-plasma concentration and subsequently determine the physicochemical properties that influence the excretion of drugs into saliva to increase the foundational knowledge underpinning saliva-based TDM.Medline, Web of Science and Embase (1974-2023) were searched for human clinical studies, which determined drug pharmacokinetics in both saliva and plasma. Studies with at least ten subjects and five paired saliva-plasma concentrations per subject were included. For each study, the ratio of the area under the concentration-time curve between saliva and plasma was determined to assess excretion into saliva. Physicochemical properties of each drug (e.g. pKa, lipophilicity, molecular weight, polar surface area, rotatable bonds and fraction of drug unbound to plasma proteins) were obtained from PubChem and Drugbank. Drugs were categorised by their ionisability, after which saliva-to-plasma ratios were predicted with adjustment for protein binding and physiological pH via the Henderson-Hasselbalch equation. Spearman correlation analyses were performed for each drug category to identify factors predicting saliva excretion (α = 5%). Study quality was assessed by the risk of bias in non-randomised studies of interventions tool.Overall, 42 studies including 40 drugs (anti-psychotics, anti-microbials, immunosuppressants, anti-thrombotic, anti-cancer and cardiac drugs) were included. The median saliva-to-plasma ratios were similar for drugs in the amphoteric (0.59), basic (0.43) and acidic (0.41) groups and lowest for drugs in the neutral group (0.21). Higher excretion of acidic drugs (n = 5) into saliva was associated with lower ionisation and protein binding (correlation between predicted versus observed saliva-to-plasma ratios: R2 = 0.85, p = 0.02). For basic drugs (n = 21), pKa predicted saliva excretion (Spearman correlation coefficient: R = 0.53, p = 0.02). For amphoteric drugs (n = 10), hydrogen bond donor (R = - 0.76, p = 0.01) and polar surface area (R = - 0.69, p = 0.02) were predictors. For neutral drugs (n = 10), protein binding (R = 0.84, p = 0.004), lipophilicity (R = - 0.65, p = 0.04) and hydrogen bond donor count (R = - 0.68, p = 0.03) were predictors. Drugs considered potentially suitable for saliva-based TDM are phenytoin, tacrolimus, voriconazole and lamotrigine. The studies had a low-to-moderate risk of bias.Many commonly used drugs are excreted into saliva, which can be partly predicted by a drug's ionisation state, protein binding, lipophilicity, hydrogen bond donor count and polar surface area. The contribution of drug transporters and physiological factors to the excretion needs to be evaluated. Continued research on drugs potentially suitable for saliva-based TDM will aid in adopting this person-centred TDM approach to improve patient outcomes.© 2024. The Author(s).