From the Department of Anaesthesiology, Radboudumc, Nijmegen, The Netherlands (MCTT, TAJvL, PvdP, CK), the Department of Anaesthesiology, Imeldaziekenhuis, Bonheiden, Belgium (PvdP), the Department of Anesthesia and Pain Management, Toronto Western Hospital (AP), the Department of Anesthesiology and Pain Medicine, University of Toronto, Toronto, Ontario, Canada (AP).
Eur J Anaesthesiol. 2021 Dec 1;38(12):1223-1229. doi: 10.1097/EJA.0000000000001465.
Enteral nutrition is essential in the treatment of critically ill patients. Current methods to monitor enteral nutrition such as aspiration of residual volume may be inaccurate. Gastric ultrasonography estimates total gastric fluid volume using the Perlas model, but this model is validated for clear fluids only, and its accuracy for measuring thick fluids is unknown.
The primary aim of this study was to evaluate the Perlas model for gastric volume estimation of enteral nutrition, a thick fluid product.
A single-centre, single blinded, randomised controlled study.
Single university hospital, from May to July 2019.
Seventy-two healthy fasted volunteers were randomly allocated to different fluid volume groups.
Participants randomly ingested predetermined volumes between 50 and 400 ml of a feeding-drink (Nutricia Nutridrink). Following a standardised gastric ultrasound scanning protocol, a blinded sonographer measured the antral cross-sectional area in the supine and right-lateral decubitus positions.
Measurements were performed at baseline, 5 min postingestion and 20 min postingestion. Gastric volumes were predicted using the previously established Perlas model and compared with total gastric fluid volumes after ingestion of the study drink.
The Perlas model underestimated the volume of thick gastric fluid and yielded a suboptimal fit for our data. However, antral cross-sectional area and total gastric thick fluid volumes were significantly correlated (Pearson's correlation coefficient 0.73, P < 0.01). A new model was fitted to predict gastric volumes of thick fluids, using the antral cross-sectional area (cm2) in the right-lateral decubitus position: Volume (ml) = 79.38 + 13.32 x right-lateral cross-sectional area.
The Perlas model for clear gastric fluid volume estimation is suboptimal for thick fluid volume assessment and an alternative model is presented.
Netherlands Trial Register Trial NL7677, Registration date: 16 April 2019; https://www.trialregister.nl/trial/7677.
肠内营养是危重症患者治疗的重要组成部分。目前监测肠内营养的方法（如残留量抽吸）可能不准确。胃超声检查使用 Perlas 模型估计总胃液体量，但该模型仅针对透明液体进行验证，其测量浓稠液体的准确性尚不清楚。
本研究的主要目的是评估 Perlas 模型在测量浓稠肠内营养液体（一种浓稠液体产品）方面的胃容量估计值。
一项单中心、单盲、随机对照研究。
单所大学医院，2019 年 5 月至 7 月。
72 名健康空腹志愿者被随机分配到不同的液体量组。
参与者随机摄入 50 至 400ml 之间的特定容量喂养饮料（Nutricia Nutridrink）。在标准化的胃超声扫描方案下，一名盲法超声技师在仰卧位和右侧卧位时测量胃窦横截面积。
在基线、摄入后 5 分钟和摄入后 20 分钟进行测量。使用先前建立的 Perlas 模型预测胃容量，并将其与摄入研究饮料后的总胃液体量进行比较。
Perlas 模型低估了浓稠胃液体的体积，并且对我们的数据拟合效果不佳。然而，胃窦横截面积和总浓稠胃液体量之间呈显著相关（Pearson 相关系数 0.73，P<0.01）。使用右侧卧位时的胃窦横截面积拟合了一个新的模型来预测浓稠液体的胃容量：体积（ml）=79.38+13.32×右侧卧位横截面积。
Perlas 模型用于估计透明胃液体量的模型对于浓稠液体体积评估效果不佳，因此提出了一种替代模型。
荷兰临床试验注册 NL7677，注册日期：2019 年 4 月 16 日；https://www.trialregister.nl/trial/7677。
