Charlotte Allerød

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.

A decision support system for suggesting ventilator settings: Retrospective evaluation in cardiac surgery patients ventilated in the ICU

Selecting appropriate ventilator settings decreases the risk of ventilator-induced lung injury. A decision support system (DSS) has been developed based on physiological models, which can advise on setting of tidal volume (Vt), respiratory frequency (f) and fraction of inspired oxygen (FiO2). The aim of this study is to assess the feasibility of the DSS by comparing its advice with the values used in clinical practice. Data from 20 patients following uncomplicated coronary artery bypass grafting (CABG) with cardiopulmonary bypass was used to test the DSS. Ventilator settings suggested by the DSS were compared to the settings selected by the clinician. When compared to the clinician the DSS suggested: lowering FiO2 (by median 7%, range 2-17%) at high SpO2 and increasing FiO2 (by median 2%, range 1-5%) at low SpO2; lowering ventilation volume (by median 0.57 l min(-1), range 0.2-1.1 l min(-1)) at high pHa and increasing ventilation volume (by median 0.4 l min(-1), range 0.1-0.9 l min(-1)) at low pHa. Suggested changes in ventilation volume were such that simulated values of PIP were < or = 22.9 cmH2O and respiratory frequency < or = 18 breaths min(-1). In all cases, computer suggested values of FiO2, Vt or f were consistent with maintaining sufficient oxygenation, normalising pH and obtaining low values of PIP.

Prospective evaluation of a decision support system for setting inspired oxygen in intensive care patients

PURPOSE The aim of the study was to prospectively evaluate a decision support system for its ability to provide appropriate suggestions of inspired oxygen fraction in intensive care patients comparing with levels used by clinicians in attendance. MATERIALS AND METHODS Thirteen mechanically ventilated patients were studied in an intensive care unit where up to 4 experiments were performed during 2 consecutive days. Inspired oxygen fraction was selected in each experiment by both the decision support system and attending clinicians, and each selection was evaluated by measuring arterial oxygen saturation. RESULTS Median (interquartile range [range]) changes in inspired oxygen fraction from baseline level by attending clinicians and the decision support system were 0.00 (-0.05 to 0.00 [-0.10 to 0.05]) and -0.03 (-0.07 to 0.01 [-0.16 to 0.12]), respectively. Clinician ranges of inspired oxygen fraction and arterial oxygen saturation were 0.25 to 0.70 and 0.92 to 0.99, respectively. Decision support system ranges of inspired oxygen fraction and arterial oxygen saturation were 0.26 to 0.54 and 0.94 to 0.99, respectively. CONCLUSIONS The decision support system selects appropriate levels of inspired oxygen fraction in intensive care patients and could be used for automatic frequent assessment of patients, freeing the focus of clinicians to concentrate on more challenging therapy.

Retrospective evaluation of a decision support system for controlled mechanical ventilation

Management of mechanical ventilation in intensive care patients is complicated by conflicting clinical goals. Decision support systems (DSS) may support clinicians in finding the correct balance. The objective of this study was to evaluate a computerized model-based DSS for its advice on inspired oxygen fraction, tidal volume and respiratory frequency. The DSS was retrospectively evaluated in 16 intensive care patient cases, with physiological models fitted to the retrospective data and then used to simulate patient response to changes in therapy. Sensitivity of the DSS’s advice to variations in cardiac output (CO) was evaluated. Compared to the baseline ventilator settings set as part of routine clinical care, the system suggested lower tidal volumes and inspired oxygen fraction, but higher frequency, with all suggestions and the model simulated outcome comparing well with the respiratory goals of the Acute Respiratory Distress Syndrome Network from 2000. Changes in advice with CO variation of about 20% were negligible except in cases of high oxygen consumption. Results suggest that the DSS provides clinically relevant and rational advice on therapy in agreement with current ‘best practice’, and that the advice is robust to variation in CO.

Decision support of inspired oxygen fraction using a model of oxygen transport

Abstract Setting inspired oxygen fraction (FiO 2 ) is a complicated balance between ensuring adequate oxygenation and minimizing the risk of lung damage. This paper presents a retrospective test of a model-based decision support system (INVENT) for advising on FiO 2 levels in intensive care patients. Clinically determined FiO 2 levels and the resulting blood oxygenation are compared with INVENT determined FiO levels and model simulated blood oxygenation. The results indicate that INVENT can maintain an acceptable level of oxygenation using similar or more appropriate levels of FiO compared to clinical practice.

The intelligent ventilator project: application of physiological models in decision support

This paper describes progress in a model-based approach to building a decision support system for mechanical ventilation. It highlights that the process of building models promotes generation of ideas and describes three systems resulting from this process, i.e. for assessing pulmonary gas exchange, calculating arterial acid-base status; and optimizing mechanical ventilation. Each system is presented and its current status and impact reviewed.

Retrospective evaluation of a decision support system for advising on ventilator settings in patients with ARDS/ALI

INTRODUCTION. Host infection by pathogens triggers an innate immune response leading to a systemic inflammatory response, often followed by an immune dysfunction which can favour the emergence of secondary infections. Dendritic cells (DCs) have a unique ability to link innate and adaptive immunityand may be centrally involved in the regulationof sepsis-induced immune suppression. We previously reported that polymicrobial sepsis durably affects the functions of DCs and confers long-term susceptibility to P. aeruginosa pneumonia. In this study, we assessed the contribution of DCs to lung defence towards secondary P. aeruginosa pneumonia.

The Intelligent Ventilator project: application of physiological models in decision support

This paper describes progress in a model-based approach to building a decision support system for mechanical ventilation. It highlights that the process of building models promotes generation of ideas and describes three systems resulting from this process, i.e. for assessing pulmonary gas exchange, calculating arterial acid-base status; and optimizing mechanical ventilation. Each system is presented and its current status and impact reviewed.

Use of the invent system for standardized quantification of clinical preferences towards mechanical ventilator settings

for ESCTAIC 2010 “Glucosafe A model-based medical decision support system for tight glycemic control in critical care” Ulrike Pielmeier a a Center for Model-based Medical Decision Support, Aalborg University, Fredrik-Bajers-Vej 7, 9220 Aalborg, Denmark Introduction Hyperglycemia during critical illness is common and is associated with increased mortality, morbidity and prolonged stay in intensive care [1][2]. The past decade has seen many attempts to improve survival by regulating blood glucose using intensive insulin therapy (IIT) protocols [3][4]. However, consistent control has proven elusive, not least because typical IIT protocols ignore the carbohydrate intake of patients [5]. An effective method that achieves and maintains “tight” blood glucose levels (i.e. in the range from 4 to 6 mmol/l) without high glucose variances and without increasing insulin-induced severe hypoglycemia (<2.2 mmol/l) has yet to emerge [6]. This work assesses the effectiveness of the computerized decision support system “Glucosafe” for tight glycemic control in critical care. This system advises insulin therapy and infusion rates of enteral and parenteral nutrition, based on blood glucose predictions with a physiological insulin-glucose model and patient-specific data [7]. Pilot testing shows significant improvements of glycemic control in a prospectively controlled cohort of intensive care patients [8]. In a retrospective analysis of the pilot study data the model is assessed with regard to how accurately blood glucose was predicted, and whether the predictive accuracy can be improved by two physiological model extensions, regarding the decreased delivery rate of nutrients that is often observed in critical care patients with delayed gastric emptying [9], and the dependency of pancreatic insulin secretion on the blood glucose level [10]. Methods The blood glucose concentrations of 10 hyperglycemic patients admitted to a neuroand trauma intensive care unit were retrospectively predicted using a) the original Glucosafe model [7] b) the Glucosafe model including a feedback loop between blood glucose and pancreatic insulin secretion rate c) the Glucosafe model and a reduced rate of appearance of enterally administered nutrition in the intestinal reservoir d) both extensions as described in b) and c). Prediction errors were expressed as absolute percent error (APE) from measured concentrations; the comparison was based on median APEs for different prediction time lengths, reflecting intervals between measurements of up to 5 hours. Results The model predictive accuracy improved modestly for each one of the two model extensions. The greatest reduction in prediction error was achieved when both model extensions were included in the Glucosafe model. For predictions time lengths (in hours) of 0.5-1.5h, 1.5-2.5h, 2.5-3.5h, 3.5-4.5h and 4.5-5.5h, the median APE was 9.7%, 11.2%, 14.8%, 15.1% and 17.7% with the Glucosafe model, compared to 9.2%, 10.1%, 12.3%, 13.2% and 16.6% with both of the model extensions included, for the same prediction time lengths. Discussion Predicted blood glucose concentrations with the Glucosafe model in its original form [7] are sufficiently accurate for typical time intervals between two measurements. The pilot trial results [8] showed that glycemic control was significantly improved, while no hypoglycemic event was observed. Thus, model-based predictive control based on the Glucosafe model may be a step towards a consistent reduction of elevated blood glucose levels. This retrospective analysis also explored two physiological model extensions, which modestly improved the model’s predictive accuracy. However, as the data used in this study were from a small cohort of patients with similar admission diagnosis, groups of other patients with a different disease background should be used to verify these preliminary results. References [1] Falciglia M, Freyberg RW, Almenoff PL, et al. Hyperglycemia-related mortality in critically ill patients varies with admission diagnosis. Crit Care Med 37 (12): 3001-3009, 2009. [2] Krinsley JS: Association between hyperglycemia and increased hospital mortality in a heterogeneous population of critically ill patients. Mayo Clinic Proceedings 78 (12): 1471-1478, 2003. [3] Meijering S, Corstjens AM, Tulleken JE, et al.Towards a feasible algorithm for tight glycaemic control in critically ill patients: a systematic review of the literature. Critical Care 10 (1): R19, 2006. [4] Chase JG, Hann CE, Shaw GM, et al. An overview of glycemic control in critical care relating performance and clinical results. Journal of Diabetes Science and Technology 1 (1): 82-91, 2007. [5] Kalfon P, Preiser JC. Tight glucose control: should we move from intensive insulin therapy alone to modulation of insulin and nutritional inputs? Crit Care 12:156, 2008. [6] Dossett LA, Collier B, Donahue R, et al. Intensive insulin therapy in practice: Can we do it? J Parenter Enteral Nutr 33 (1): 14-20, 2009. [7] Pielmeier U, Andreassen S, Nielsen BS, et al. A simulation model of insulin saturation and glucose balance for glycemic control in ICU patients. Computer Methods and Programs in Biomedicine 97 (3): 211-222, 2010. [8] Pielmeier U, Andreassen S, Juliussen B, et al. The Glucosafe system for tight glycemic control in critical care: A pilot evaluation study. J Crit Care 25 (1): 97-104, 2010. [9] Chapman M, Fraser R, Matthews G, et al. Glucose absorption and gastric emptying in critical illness. Crit Care 13 (4): R140, 2009. [10] Polonsky KS, Given BD, Van Cauter E. Twenty-four-hour profiles and pulsatile patterns of insulin secretion in normal and obese subjects. J Clin Invest 81: 442-448, 1988.SELECTED ABSTRACTS PRESENTED AT THE 21ST MEETING OF THE EUROPEAN SOCIETY FOR COMPUTING AND TECHNOLOGY IN ANAESTHESIA AND INTENSIVE CARE (ESCTAIC) Amsterdam, The Netherlands, 6th–9th October, 2010 Edited by: A. A. van Dusseldorp, C. Boer, D. S. Karbing, L. Krummreich, S. E. Rees, S. A. Loer LIST OF ABSTRACTS Soraya Abbasi: The Role of Physiological Models in Critiquing Mechanical Ventilation Treatments Gracee Agrawal: Real-time Detection of Suppression in EEG Christa Boer: Clinical Experience with Perioperative Non-invasive Beat-to-beat Arterial Blood Pressure Monitoring Nadja Bressan: Infusion Rate Control Algorithm for Target Control Infusion using Optimal Control Chih-Yen Chiang: Rule-based Evaluation for the Patientcontrolled Analgesia Clinical Effectiveness Wolfgang Friesdorf: Professional Design of Clinical Working Systems According to Human Factors Fred de Geus: CAROLA: An Open Source PDMS, After 25 Years Still Experimental? Yori Gidron: The Effects of Stress and Hemispheric Lateralization on Managerial Decisions Johan Groeneveld: Value of Central or Mixed Venous O2 Saturation in Guiding Treatment in the Intensive Care Unit Gabriel M. Gurman: Professional Stress and the Anesthesiologist-how Evident is it? Eliahu Heldman: Salivary Cortisol as a Measure of Professional Stress; An Overview and a Description of a Study with Paramedics Martin Hurrell: Implementation of a Standards-based, CDA-Compliant Anesthesia Record Mathieu Jeanne: Analgesia Nociception Index Online Computation and Preliminary Clinical Test During Cholecystectomy Under Remifentanil-Propofol Anaesthesia Christian Jeleazcov: Pharmacodynamic Modeling of Changes in Pulse Waveform During Induction of Propofol Anaesthesia in Volunteers: Comparison between Invasive and Continuous Non-invasive Measurements of Pulse Pressure Pierre Kalfon: Assessing Performances of Glucose Control Algorithms on a set of Virtual ICU Patients Cor Kalkman: Automation and Automation Surprises: Lessons from Aviation. Should Health Care Brace Itself. Dan S Karbing: Use of the INVENT System for Standardized Quantification of Clinical Preferences Towards Mechanical Ventilator Settings Talma Kushnir: Moods and Burnout Among Physicians: Associations with Prescribing Medications Communicating with Patients, and Referrals for Specialists and Diagnostic Tests Johannes J van Lieshout: Non-invasive Pulse Contour Cardiac Output by Nexfin Technol Egbert Mik: Monitoring Mitochondrial Oxygenation Journal of Clinical Monitoring and Computing (2011) 25:3–43 DOI: 10.1007/s10877-011-9276-2 Springer 2011

INVENT - a decision support system for managing inspired oxygen

SELECTED ABSTRACTS PRESENTED AT THE 20TH MEETING OF THE EUROPEAN SOCIETY FOR COMPUTING AND TECHNOLOGY IN ANAESTHESIA AND INTENSIVE CARE (ESCTAIC) Berlin, Germany, Technical University, 23–26 September, 2009 Edited by: I. Marsolek, A. Dellermalm, L. Krummreich, S. E. Rees, W. Friesdorf LIST OF ABSTRACTS Bibian, S & Zikov, T: Predictability of BIS, Entropy and NeuroSENSE Bornemann, M: Data – Information – Knolwedge: The Way Up in the ICU? Using Information Making Knowledge Available? Bunker, N & Handy, J: Current Clinical Problems in Interpreting Electrolytes, Acid–base & Metabolites Dai, CY; Chen, CY; Chang, YK; Lin, WT; Lin, CP & Sun, WZ: A Multifunctional Endoscopic Platform with Detachable Probe DeVience, S; Moretti, E & Shang, AB: A Retrofittable Anesthesia Agent Alarm Dollman, M; Thien, C; Ingenlath, M; Pappert, D & Friesdorf, W: Sustainable Information of ICU Patient Family Members Eden, A; Barak, Y & Pizov, R: The Impact of an Electronic Reminder on the Administration of Preoperative Prophylactic Antibiotics Fuchs, D; Marsolek I & Friesdorf, W: Analyzing the Requirements for a Computer Based Optimization of the Medication Process Gerber, D.; Eberle, B & Trachsel, S: A Web-based Knowledge Database (Wiki Platform) for Standard Operational Procedures (SOPs) in Cardiac Anesthesia Grottke, O; Ullrich, S; Fried, E; Deserno, T; Kuhlen, T & Rossaint R: Regional Anaesthesia in Virtual Environments Gurman, G: Prevention of Ventilator Associated Pneumonia: Where Are We Now? Heinlein, M: Structured Medical Documentation in the OR: MediColor Web and MediAnes Web: Web Based Solutions for Surgery and Anaestehsia Huh, J; Ahn, W; Ro, Y; Min, S & Kim, C: Comparison of Perfusion Index, T-wave Amplitude and Heart Rate as an Indicator for Intravascular Injection of Epinephrine-Containing Test Dose in Anaesthetized Adults Karbing, DS; Allerød, C; Thorgaard, P; Andreassen, S; Kjærgaard, S & Rees, SE: INVENT – A Decision Support System for Managing Inspired Oxygen – Prospective Evaluation in an Intensive Care Unit Kennedy, R; Minto, C; French, R & McKellow, M: Effect Site Sevoflurane Levels for Airway Manipulations During Rapid Wash-In Kofránek, J; Mateják, M; Tribula, M & Matoušek, S: Educational Application of Acid–base, Volume and Electrolyte Modelling Koller, W: Data – Information – Knowledge: The Way Up in the ICU? Everyday Clinical Challenges Journal of Clinical Monitoring and Computing (2010) 24:1–33 DOI: 10.1007/s10877-009-9211-y Springer 2010