Clinical Pharmacology, Royal Brisbane & Women's Hospital, Brisbane, Australia.
Faculty of Medicine, University of Queensland, Brisbane, Australia.
Br J Clin Pharmacol. 2021 Nov;87(11):4341-4353. doi: 10.1111/bcp.14853. Epub 2021 May 4.
Equations to calculate albumin-adjusted total concentrations have been validated to correlate with measured free concentrations for both phenytoin and valproate, but there is a lack of data to assess correlation with clinical outcomes. We aimed to assess the association of hypoalbuminaemia and albumin-adjusted total concentrations with concentration-dependent toxicity for phenytoin and valproate and review the impact on management decisions following concentration monitoring in hypoalbuminaemia.
Patients undergoing concentration monitoring for phenytoin or valproate between January and December 2018 were included. Patients were identified using a centralised laboratory database with data extracted from medical records.
Total phenytoin concentrations were measured for 144 patients, with hypoalbuminaemia (≤30 g L ) recorded in 59 (41%) patients. Albumin-adjusted phenytoin concentration >20 mg L was associated with increased neurological adverse effects (77% vs. 43%, P < .001). On logistic regression, higher albumin-adjusted phenytoin concentration was an independent risk factor for neurotoxicity (OR 1.06, 95% CI 1.01-1.12, P = .011). Total valproate concentrations were measured for 383 patients, with hypoalbuminaemia (≤30 g L ) noted in 53 (14%) patients. For the valproate cohort, hypoalbuminaemia (42% vs. 28%, P = .039) and albumin-adjusted valproate concentration >100 mg L (49% vs. 23%, P < .001) were both associated with increased neurotoxicity. On multiple logistic-regression, valproate daily dose (aOR = 1.01, 95% CI 1.00-1.02, P = .006) and albumin-adjusted valproate concentration (aOR 1.01, 95% CI 1.00-1.02, P = .033) were independent risk factors for neurotoxicity after accounting for confounders.
While measuring free drug concentrations in hypoalbuminaemia would be ideal, the adjustment equations can help identify vulnerable patients needing further assessment of potential concentration-dependent toxicity.
已经验证了用于计算白蛋白校正后总浓度的方程,以与苯妥英和丙戊酸的实测游离浓度相关,但缺乏数据来评估其与临床结局的相关性。我们旨在评估低白蛋白血症和白蛋白校正后的总浓度与苯妥英和丙戊酸浓度依赖性毒性的关系,并回顾在低白蛋白血症时进行浓度监测对管理决策的影响。
纳入 2018 年 1 月至 12 月期间进行苯妥英或丙戊酸浓度监测的患者。使用中央实验室数据库识别患者,数据从病历中提取。
共检测了 144 例患者的总苯妥英浓度,其中 59 例(41%)患者记录有低白蛋白血症(≤30g/L)。白蛋白校正后的苯妥英浓度>20mg/L 与神经不良反应增加相关(77%比 43%,P<0.001)。在逻辑回归中,较高的白蛋白校正后的苯妥英浓度是神经毒性的独立危险因素(OR 1.06,95%CI 1.01-1.12,P=0.011)。共检测了 383 例患者的总丙戊酸浓度,其中 53 例(14%)患者记录有低白蛋白血症(≤30g/L)。对于丙戊酸队列,低白蛋白血症(42%比 28%,P=0.039)和白蛋白校正后的丙戊酸浓度>100mg/L(49%比 23%,P<0.001)均与神经毒性增加相关。在多元逻辑回归中,丙戊酸钠日剂量(aOR=1.01,95%CI 1.00-1.02,P=0.006)和白蛋白校正后的丙戊酸浓度(aOR 1.01,95%CI 1.00-1.02,P=0.033)是在控制混杂因素后神经毒性的独立危险因素。
虽然在低白蛋白血症时测量游离药物浓度是理想的,但调整方程可以帮助识别需要进一步评估潜在浓度依赖性毒性的易受影响的患者。