Journal of clinical monitoring and computing
-
J Clin Monit Comput · Aug 2019
Machine learning based framework to predict cardiac arrests in a paediatric intensive care unit : Prediction of cardiac arrests.
A cardiac arrest is a life-threatening event, often fatal. Whilst clinicians classify some of the cardiac arrests as potentially predictable, the majority are difficult to identify even in a post-incident analysis. Changes in some patients' physiology when analysed in detail can however be predictive of acute deterioration leading to cardiac or respiratory arrests. ⋯ A positive predictive value of 11% and negative predictive value of 98% was obtained with a prevalence of 5% by our method of prediction. While clinicians predicted 4 out of the 69 cardiac arrests (6%), the prediction system predicted 63 (91%) cardiac arrests. Prospective validation of the automated system remains.
-
J Clin Monit Comput · Aug 2019
Selection of cuffed endotracheal tube for children with congenital heart disease based on an ultrasound-based linear regression formula.
It remains to be discovered whether a formula predicting the subglottic transverse diameter measured by ultrasound (SGDformula) for the selection of an appropriate endotracheal tube (ETT) for children without congenital heart disease (CHD) is useful for children with CHD. A formula for predicting SGD was established after assessing 60 children ≤ 8 years without CHD and validated on 60 children with CHD. We selected the cuffed ETT size based on the SGD by ultrasound (SGDultra). ⋯ And the mean bias (SGDformula-ETT size and SGDultra-ETT size) was 0.21 mm (95% confidence interval, - 0.59 to 1.01 mm) and 0.00 mm (- 0.79 to 0.84 mm). For the CHD group, the ultrasound-based method yielded a 78% success rate of ETT size choice, while the formula-based method permitted an appropriate ETT size in only 32% of subjects (P < 0.001). Our analysis showed that measuring the SGDultra was more accurate in predicting the correct OD of the ETT in children with CHD undergoing cardiovascular surgery, based on the correlation and agreement with ETT OD.
-
J Clin Monit Comput · Aug 2019
Comparative Study Observational StudyA comparison of propofol-to-BIS post-operative intensive care sedation by means of target controlled infusion, Bayesian-based and predictive control methods: an observational, open-label pilot study.
We evaluated the feasibility and robustness of three methods for propofol-to-bispectral index (BIS) post-operative intensive care sedation, a manually-adapted target controlled infusion protocol (HUMAN), a computer-controlled predictive control strategy (EPSAC) and a computer-controlled Bayesian rule-based optimized control strategy (BAYES). ⋯ Both computer-based control systems are feasible to be used during ICU sedation with overall tighter control than HUMAN and even with lower required CePROP. EPSAC control required higher CeREMI than BAYES or HUMAN to maintain stable control. Clinical trial number: NCT00735631.
-
J Clin Monit Comput · Aug 2019
Mathematical arterialisation of peripheral venous blood gas for obtainment of arterial blood gas values: a methodological validation study in the clinical setting.
Arterial blood gas (ABG) analysis is an essential tool in the clinical assessment of acutely ill patients. Venous to arterial conversion (v-TAC), a mathematical method, has been developed recently to convert peripheral venous blood gas (VBG) values to arterialized VBG (aVBG) values. The aim of this study was to test the validity of aVBG compared to ABG in an emergency department (ED) setting. ⋯ Bland-Altman plot revealed clinically acceptable mean difference and limits-of-agreement intervals between ABG and aVBG pH and pCO2, but not between ABG and aVBG pO2. Arterialization of VBG using v-TAC is a valid method for measuring pH and pCO2, but not for pO2. Larger clinical studies are required to evaluate the applicability of v-TAC in different patient subpopulations.
-
J Clin Monit Comput · Aug 2019
Can variable practice habits and injection port dead-volume put patients at risk?
Injection ports used to administer medications and draw blood samples have inherent dead-volume. This volume can potentially lead to inadvertent drug administration, contribute to erroneous laboratory values by dilution of blood samples, and increase the risk of vascular air embolism. We sought to characterize provider practice in management of intravenous (IV) and arterial lines and measure dead-volumes of various injection ports. ⋯ Mean (SD) dead-volume in microliters ranged from 0.1 (0.0) to 5.6 (1.0) in 1-way injection ports and from 54.1 (2.8) to 126.5 (8.3) in 4-way injection ports. The practices of our providers when giving medications and drawing blood samples are variable. The dead-volume associated with injection ports used at our institution may be clinically significant, increasing errors in medication delivery and laboratory analysis.