David Murley

Using physiological models and decision theory for selecting appropriate ventilator settings.

ObjectiveTo present a decision support system for optimising mechanical ventilation in patients residing in the intensive care unit.MethodsMathematical models of oxygen transport, carbon dioxide transport and lung mechanics are combined with penalty functions describing clinical preference toward the goals and side-effects of mechanical ventilation in a decision theoretic approach. Penalties are quantified for risk of lung barotrauma, acidosis or alkalosis, oxygen toxicity or absorption atelectasis, and hypoxaemia.ResultsThe system is presented with an example of its use in a post-surgical patient. The mathematical models describe the patient’s data, and the system suggests an optimal ventilator strategy in line with clinical practice.ConclusionsThe system illustrates how mathematical models combined with decision theory can aid in the difficult compromises necessary when deciding on ventilator settings.

Oxygenation within the first 120 h following coronary artery bypass grafting. Influence of systemic hypothermia (32 ⁰C) or normothermia (36 ⁰C) during the cardiopulmonary bypass: a randomized clinical trial

Background:  Lung function is often impaired after cardiac surgery performed under cardiopulmonary bypass (CPB). Normothermic CPB has become more common, but it remains unknown whether it reduces post‐operative lung function compared with hypothermic CPB. The aim of this study was to investigate oxygenation within the first 120 h after systemic hypothermia and normothermia under CPB.

Reproducibility and comparability of insulin sensitivity indices measured by stable‐label intravenous glucose tolerance test

We have investigated the reproducibility of (1) insulin sensitivity (S*I) and glucose effectiveness (S*G) as measured by the stable‐label (one compartment) minimal model, and (2) insulin sensitivity (S*Ib), plasma clearance rate (PCR), basal hepatic output (HGOb), and total hepatic glucose output (HGO0–240) as measured by the novel stable‐label two compartment model of glucose disappearance during labelled intravenous glucose tolerance test (IVGTT) using 6,6‐2H‐glucose. Ten normal male subjects were studied on two occasions one week apart. Both models provided estimates of all indices with acceptable precision (CV of parameter estimates ≤50 %). The within subject CVs of S*I and S*Ib were comparable (17 % vs 19 %) as were the within subject CVs of S*G and PCR (13 % vs 16 %). A highly significant linear relationship was observed between S*Ib and S*I (0.303 ± 0.046 ml kg−1 min−1 per mU l−1 vs 13.04 ± 1.89 10−4 min−1 per mU l−1, y = 0.0037 x + 0.0002, r = 0.90, p <0.001; mean ± SE), but not between PCR and S*G (1.98 ± 0.15 ml kg−1 min−1 vs 0.0089 ± 0.0005 min−1, rs = 0.34, NS). The two compartment model provided a plausible time‐profile of hepatic glucose output during IVGTT, reproducible estimates of HGOb (1.96 ± 0.18 mg kg−1 min−1, 15 %; mean ± SE, within subject CV), and a highly reproducible HGO0–240 (7 %; within subject CV). We conclude that the stable‐label (one compartment) minimal model and the stable‐label two compartment model provide reproducible estimates of parameters of glucose kinetics in normal subjects. Insulin sensitivity indices estimated by the two models are strongly linearly related.

Decision support of inspired oxygen selection based on Bayesian learning of pulmonary gas exchange parameters

OBJECTIVE To investigate if the real-time Bayesian learning of physiological model parameters can be used to support and improve the selection of inspired oxygen fraction. METHODS AND MATERIAL Supporting the selection of inspired oxygen fraction relies on predictions of arterial oxygen saturation. The efficacy of using these predictions to select inspired oxygen was tested retrospectively in a system for estimating gas exchange parameters of the lung (Automatic Lung Parameter Estimator, ALPE). For the predictions to offer effective decision support they need to be accurate and above all safe. These qualities were tested with data from 16 post-operative cardiac patients, using two different tests. The aim of the first test was to assess retrospectively if the predictions could have supported clinical decisions. The second test sought to establish if the predictions could support improving the efficiency of inspired oxygen selection during an ALPE oxygen titration. RESULTS The predictions were found to be reasonably accurate, and most importantly safe in both of the tests. CONCLUSION The method described can be used to support the selection of inspired oxygen fraction, and it has the potential to improve the efficiency of inspired oxygen selection during an oxygen titration.

Can new pulmonary gas exchange parameters contribute to evaluation of pulmonary congestion in left-sided heart failure?

BACKGROUND Assessment of pulmonary congestion in left-sided heart failure is necessary for guiding anticongestive therapy. Clinical examination and chest x-ray are semiquantitative methods with poor diagnostic accuracy and reproducibility. OBJECTIVES To establish reference values, describe reproducibility, and investigate the diagnostic and monitoring properties in relation to pulmonary congestion of new pulmonary gas exchange parameters describing ventilation/perfusion mismatch (variable fraction of ventilation [fA2] or the drop in oxygen pressure from the mixed alveolar air of the two ventilated compartments to the nonshunted end-capillary blood [DeltaPO(2)]) and pulmonary shunt. METHODS Sixty healthy volunteers and 69 patients requiring an acute chest x-ray in a cardiac care unit were included. The gas exchange parameters were estimated by analyzing standard bedside respiratory and circulatory measurements obtained during short-term exposure to different levels of inspired oxygen. Nine patients were classified as having pulmonary congestion using a reference diagnosis and were followed during 30 days of anticongestive therapy. Diagnostic and monitoring properties were compared with chest x-ray, N-terminal probrain natriuretic peptide (NT-proBNP), spirometry values, arterial oxygen tension, alveolar-arterial oxygen difference and venous admixture. RESULTS The 95% reference intervals for healthy subjects were narrow (ie, fA2 [0.75 to 0.90], DeltaPO(2) [0.0 kPa to 0.5 kPa] and pulmonary shunt [0.0% to 8.2%]). Reproducibility was relatively good with small within subject coefficients of variation (ie, fA2 [0.05], DeltaPO(2) [0.4 kPa] and pulmonary shunt [2.0%]). fA2, DeltaPO(2) and NT-proBNP had significantly better diagnostic properties, with high sensitivities (100%) but low specificities (30% to 40%). During successful anticongestive therapy, fA2, DeltaPO(2), NT-proBNP and spirometry values showed significant improvements. CONCLUSIONS The gas exchange parameter for ventilation/perfusion mismatch but not pulmonary shunt can have a possible role in rejecting the diagnosis of pulmonary congestion and in monitoring anticongestive therapy.

Use of physiological models in selecting ventilator settings

JOINT ESCTAIC – SFIMAR MEETING Pierre Baudis Congress Center, Toulouse, France,

Bayesian Learning of the Gas Exchange Properties of the Lung for Prediction of Arterial Oxygen Saturation

This paper describes how real-time Bayesian learning of physiological model parameters is used to predict arterial oxygen saturation at the bedside. The efficacy of using these predictions as a decision support tool in a system for estimating gas exchange parameters of the lung (ALPE) was tested retrospectively. For the predictions to offer effective decision support they need to be accurate and safe. These qualities were tested for two patient groups, using two different test strategies for each group. The prediction accuracy when used in combination with the predictions’ safety margin was found to be adequate in all the test cases. Thus the method described can be used as the basis for effective model-based decision support in ALPE.