A knowledge- and model-based system for automated weaning from mechanical ventilation: technical description and first clinical application.

To describe the principles and the first clinical application of a novel prototype automated weaning system called Evita Weaning System (EWS). EWS allows an automated control of all ventilator settings in pressure controlled and pressure support mode with the aim of decreasing the respiratory load of mechanical ventilation. Respiratory load takes inspired fraction of oxygen, positive end-expiratory pressure, pressure amplitude and spontaneous breathing activity into account. Spontaneous breathing activity is assessed by the number of controlled breaths needed to maintain a predefined respiratory rate. EWS was implemented as a knowledge- and model-based system that autonomously and remotely controlled a mechanical ventilator (Evita 4, Dräger Medical, Lübeck, Germany). In a selected case study (n = 19 patients), ventilator settings chosen by the responsible physician were compared with the settings 10 min after the start of EWS and at the end of the study session. Neither unsafe ventilator settings nor failure of the system occurred. All patients were successfully transferred from controlled ventilation to assisted spontaneous breathing in a mean time of 37 ± 17 min (± SD). Early settings applied by the EWS did not significantly differ from the initial settings, except for the fraction of oxygen in inspired gas. During the later course, EWS significantly modified most of the ventilator settings and reduced the imposed respiratory load. A novel prototype automated weaning system was successfully developed. The first clinical application of EWS revealed that its operation was stable, safe ventilator settings were defined and the respiratory load of mechanical ventilation was decreased.

Standardizing intensive care device data to enable secondary usages.

To represent medical device observations in a format that is consumable by clinical software, standards like HL7v3 and ISO/IEEE 11073 should be used jointly. This is demonstrated in a project with Dräger Medical GmbH focusing on their Patient Data Management System (PDMS) in intensive care, called Integrated Care Manager (ICM). Patient and device data of interest should be mapped to suitable formats to enable data exchange and decision support. Instead of mapping device data to target formats bilaterally we use a generic HL7v3 Refined Message Information Model (RMIM) with device specific parts adapted to ISO/IEEE 11073 DIM. The generality of the underlying model (based on Yuksel et al. [1]) allows the flexible inclusion of IEEE 11073 conformant device models of interest on the one hand and the generation of needed artifacts for secondary usages on the other hand, e.g. HL7 V2 messages, HL7 CDA documents like the Personal Health Monitoring Report (PHMR) or web services. Hence, once the medical device data are obtained in the RMIM format, it can quite easily be transformed into HL7-based standard interfaces through XSL transformations because these interfaces all have their building blocks from the same RIM. From there data can be accessed uniformly, e.g. as needed by Dräger´s decision support system SmartCare [2] for automated control and optimization of weaning from mechanical ventilation.

[SmartCare: automatizing clinical guidelines]. / SmartCare: Automatisierung klinischer Leitlinien.

In critical care environments, important medical and economic challenges are presented by the enhancement of therapeutic quality and the reduction of therapeutic costs. For this purpose, several clinical studies have demonstrated a positive impact of the adoption of so-called clinical guidelines. Clinical guidelines represent well documented best practices in healthcare and are fundamental aspects of evidence-based medicine. However, at the bedside, such clinical guidelines remain difficult to use by clinical staff. The knowledge-based technology SmartCare allows incorporation of arbitrary computerized clinical guidelines into various medical target systems. SmartCare constitutes a clinical guideline engine because it executes one or more clinical guidelines on a specific medical device. SmartCare was initially applied for the automated control of a mechanical ventilator to assist the process of weaning from a medical device. The methodology allows further applications to be implemented effectively with other medical devices and/or with other appropriate guidelines. In this paper, we report on the methodology and the resulting versatility of such a system, as well as the clinical evaluation of SmartCare/PS and its perspectives.

A multicenter randomized trial of computer-driven protocolized weaning from mechanical ventilation.

RATIONALE AND OBJECTIVES: Duration of weaning from mechanical ventilation may be reduced by the use of a systematic approach. We assessed whether a closed-loop knowledge-based algorithm introduced in a ventilator to act as a computer-driven weaning protocol can improve patient outcomes as compared with usual care. METHODS AND MEASUREMENTS: We conducted a multicenter randomized controlled study with concealed allocation to compare usual care for weaning with computer-driven weaning. The computerized protocol included an automatic gradual reduction in pressure support, automatic performance of spontaneous breathing trials (SBT), and generation of an incentive message when an SBT was successfully passed. One hundred forty-four patients were enrolled before weaning initiation. They were randomly allocated to computer-driven weaning or to physician-controlled weaning according to local guidelines. Weaning duration until successful extubation and total duration of ventilation were the primary endpoints. MAIN RESULTS: Weaning duration was reduced in the computer-driven group from a median of 5 to 3 d (p=0.01) and total duration of mechanical ventilation from 12 to 7.5 d (p=0.003). Reintubation rate did not differ (23 vs. 16%, p=0.40). Computer-driven weaning also decreased median intensive care unit (ICU) stay duration from 15.5 to 12 d (p=0.02) and caused no adverse events. The amount of sedation did not differ between groups. In the usual care group, compliance to recommended modes and to SBT was estimated, respectively, at 96 and 51%. CONCLUSIONS: The specific computer-driven system used in this study can reduce mechanical ventilation duration and ICU length of stay, as compared with a physician-controlled weaning process.