SmartCare™/PS (Pressure Support): knowledge-based
system for the management of mechanical ventilation in
Intensive Care Units
|knowledge-based systems, decision
support systems, clinical guidelines, workflow, knowledge
engineering, knowledge representation, temporal reasoning,
||Intensive care, emergency care,
perinatal care, critical care; ventilation management,
weaning patients from mechanical ventilation, respiratory
|| Dräger Medical GmbH & Co KGaA, Lübeck, Germany in association with
Hôpital Henri Mondor and INSERM, Créteil, France.
|location of use
||Commercial product, in clinical
FDA-approved (USA), TPD-approved (Canada).
|| SOLVATIO® knowledge-based
system shell developed by IISY (Germany).
Rule-based Forward Chaining;
Categorical Problem Solving Method;
Non-monotonic Reasoning using ITMS (Immediate-Check Truth Maintainance System);
Past-oriented Temporal Reasoning;
Multiple knowledge bases allowing multiple guidelines.
|integration with patient data
At present, all patient data are keyed in by the user before using SmartCare/PS (either for a single or multiple sessions).
Extraction of patient data from SmartCare/PS is a semi-automated, two step process. Data
are firstly exported as XHTML (using SmartCare's
web-server) before being imported into an EPR or other application.
includes an Ethernet interface (plus a web-server) which would support future direct connectivity
to CIS, HIS, PDMS - as well as data mining
and machine learning.
|source for knowledge base
SmartCare™ is a generic
framework built by Dräger Medical for constructing
intelligent applications. The technology is designed
to allow clinical guidelines and protocols to be executed
by automatically operated medical devices.
The current focus of use of SmartCare is in critical
care. SmartCare/PS (Pressure Support) is a knowledge
based weaning system. It is the latest (and most advanced)
in a series of systems for the management of mechanical
ventilation in Intensive Care Units that started with
NéoGanesh (1992-2001) and continued through Automedon,
Automedon/PS and EvitaXL. SmartCare/PS is essentially
an enhanced version of the EvitaXL ventilator system
to which it adds protocol-based care for diagnostics
and treatment with the goal of weaning patients off
SmartCare/PS is based on a clinical protocol for
weaning. The system "divides the control process
into three steps:"
- "Step 1: Stabilizing the patient within a
respiratory comfort zone by regulating the level
of pressure support based on three parameters: breathing
rate, tidal volume and end tidal CO2."
- "Step 2: Reducing invasiveness by testing
whether the patient can tolerate a lower pressure
support level without leaving the comfort zone."
- "Step 3: Testing readiness for extubation
by maintaining the patient at the lowest limit of
support" (Dräger Medical).
Lellouche F, Mancebo J, Jolliet P et al.
A multicenter randomized trial of computer-driven protocolized weaning from mechanical ventilation.
Am J Respir Crit Care Med. 2006 Oct 15;174(8):894-900.
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.
Mersmann S and Dojat M. SmartCare
- Automated clinical guidelines in critical care.
In: R. Lopez de Mantara and L. Saitta, eds., 16th
European Conference on Artificial Intelligence
(ECAI'04). IOS press, Valencia (ES) 22-27 Aug,
" In critical care environments
important medical and economical 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 health care
and are fundamental aspects of evidence-based
medicine. However, at the bedside, such clinical
guidelines remain difficult to use by the clinical
staff. Recently, we have designed and implemented
the knowledge-based SmartCare™ system that allows
automated control of medical devices in critical
care. SmartCare™ constitutes a clinical guideline
engine since it executes one or more clinical
guidelines on a specific medical device. The underlying
methodology comprises two sequential phases and
seamlessly combines knowledge engineering with
expert system techniques, e.g. rule-based forward
chaining and temporal reasoning, for clinical
guidelines modelling and software engineering
techniques for source code generation and for
integration to the target platform. SmartCare™
was initially applied for the automated control
of a mechanical ventilator and is currently being
evaluated in a European multicentre clinical study
started two years ago. Intermediate reports have
been extremely positive and suggest a statistically
significant reduction in the duration of mechanical
ventilation using SmartCare™. The methodology
allows SmartCare™ to be implemented effectively
with other medical devices and/or with other appropriate
guidelines. In this paper we report on the methodology,
architecture and the resulting versatility of
SmartCare™ for the automated execution of clinical
guidelines. Benefits and lessons learned during
its development are discussed "
Mersmann S, Kück K.
SmartCareTM – Optimizing
Workflow Processes in Critical Care through Automation.
Accepted for presentation at ESCTAIC
Improving the quality and efficiency of health
care delivery are important objectives in critical
care. Process engineering approaches to identify,
organize and standardize health care workflows
have been employed to meet these goals. Evidence-based
clinical guidelines (CGs) for critical care are
among these approaches. Their impact on outcome
measures have been investigated and quantified
in several clinical studies, e.g. . Outcome
measures that were studied include the reduction
of hospital stay, mortality, human errors, medical
device induced complications and workload of clinical
staff. A logical next step is now the implementation
of standardized health care processes into medical
technology by allowing CGs to be executed by medical
devices. This could provide automated standardized
workflow process support. Dräger Medical's SmartCareTM
technology is a platform that allows the implementation
and automatic execution of various CGs within
a wide range of medical devices. The SmartCareTM
expert system comprises a universal engine and
a set of executable knowledge bases that each
reflects a certain critical care process, as described
by a CG. An expert system construction suite (Solvatio,
iisy AG, Rimpar, Germany) is used to facilitate
efficient, visual-oriented knowledge modeling
as well as the transition to the runtime environment.
It seamlessly combines process-, knowledge- and
software-engineering tasks. The core paradigm
is that if a medical device allows for reading
access to its measurements, settings, and contextual
information as well as for writing access to its
settings, then every clinical guideline for that
medical device is potentially automatable .
Currently the automation of a specific process
for weaning patients from mechanical ventilation
has been implemented in a commercial product.
SmartCare™/PS as an add-on for EvitaXL (Dräger
Medical, Germany) provides automated control in
pressure support ventilation. It implements a
weaning CG clinically developed by Dojat and Brochard
. Methods. A multi-center, randomized controlled
study was carried out in five university hospitals.
144 medico-surgical ICU patients were enrolled
in this study. Approximately half of the patients
(n=70) were randomized to be weaned following
the conventional weaning protocol used in the
respective hospital, the other half were weaned
using the automated SmartCare™ approach (n=74).
Results. In comparison with manual implementation
of conventional weaning CGs used in these intensive
care units, SmartCare™/PS reduced weaning duration
by 50%, total duration of mechanical ventilation
by more than 30% and the ICU length of stay by
almost 30 % . Conclusion. The automated execution
of CGs by medical devices is a logical and beneficial
progression of workflow support in health care.
The implementation of additional CGs is expected
to demonstrate the efficiency of SmartCare™ technology
throughout the complex development process from
knowledge acquisition to knowledge execution."
 SM Burns et al., Crit Care Med
2003, Vol. 31, No. 12:2752-2763
 S Mersmann,
M Dojat, 16th European Conference on Artificial
 M Dojat et al.,
Art Intell Med 1997, 11:97-117
 F Lellouche
et al., Intensive Care Medicine 2004, Vol. 30,
Supplement 1, 254:P69
Bouadma L, Lellouche F, Cabello B et al.
Computer-driven management of prolonged mechanical ventilation and weaning: a pilot study.
Intensive Care Med. 2005 Aug 23;
OBJECTIVE: To evaluate the ability of a computer-driven system (CDS) to manage pressure-support ventilation over prolonged periods and to predict weaning readiness compared to intensivists. The system continuously adapts pressure support, gradually decreases ventilatory assistance when possible, and indicates weaning readiness.DESIGN AND SETTING: A two-center, prospective, open, clinical, pilot study in medical ICUs of two university hospitals.PATIENTS AND PARTICIPANTS: 42 consecutive mechanically ventilated patients (60+/-14 years, SAPS II 39+/-15), 9 of whom were excluded.INTERVENTIONS: As soon as patients could tolerate pressure support, they were ventilated with the CDS. The times of weaning readiness determined by the intensivists and CDS were compared.MEASUREMENTS AND RESULTS: Weaning was successful in 25 patients and failed in 7; unplanned extubation occurred in 1 patient. Time on CDS ventilation was 3+/-3 days (maximum, 12 days). The CDS detected weaning readiness earlier than the intensivists in 17 patients, and intensivists earlier than the CDS in 4; in 11 patients detection times coincided.CONCLUSIONS: A CDS was successful in fully managing pressure-support ventilation over prolonged periods and often proposed weaning readiness earlier than the intensivists did. Use of this CDS may reduce the duration of mechanical ventilation.
Dojat M, Pachet F, Guessoum Z et al.
NéoGanesh: a working system for the automated control of assisted ventilation in ICUs.
Artif Intell Med. 1997 Oct;11(2):97-117.
Automating the control of therapy administered to a patient requires systems which integrate the knowledge of experienced physicians. This paper describes NeoGanesh, a knowledge-based system which controls, in closed-loop, the mechanical assistance provided to patients hospitalized in intensive care units. We report on how new advances in knowledge representation techniques have been used to model medical expertise. The clinical evaluation shows that such a system relieves the medical staff of routine tasks, improves patient care, and efficiently supports medical decisions regarding weaning. To be able to work in closed-loop and to be tested in real medical situations, NeoGanesh deals with a voluntarily limited problem. However, embedded in a powerful distributed environment, it is intended to support future extensions and refinements and to support reuse of knowledge bases.
Dräger Medical AG & Co. KG
Moislinger Allee 53-55
Fon +49 451 882 4062
Fax +49 451 882 2856
Research and Development Critical Care, Dräger Medical;
Michel Dojat, INSERM (France).
| Entry on OpenClinical: 23 August 2005
Last main updates: 01 September 2005, (25 October 2006)
Design template v0.3: 24 July 2005