Towards Implementation of OMOP in a German University Hospital Consortium

Background  In 2015, the German Federal Ministry of Education and Research initiated a large data integration and data sharing research initiative to improve the reuse of data from patient care and translational research. The Observational Medical Outcomes Partnership (OMOP) common data model and the Observational Health Data Sciences and Informatics (OHDSI) tools could be used as a core element in this initiative for harmonizing the terminologies used as well as facilitating the federation of research analyses across institutions.

Objective  To realize an OMOP/OHDSI-based pilot implementation within a consortium of eight German university hospitals, evaluate the applicability to support data harmonization and sharing among them, and identify potential enhancement requirements.

Methods  The vocabularies and terminological mapping required for importing the fact data were prepared, and the process for importing the data from the source files was designed. For eight German university hospitals, a virtual machine preconfigured with the OMOP database and the OHDSI tools as well as the jobs to import the data and conduct the analysis was provided. Last, a federated/distributed query to test the approach was executed.

Results  While the mapping of ICD-10 German Modification succeeded with a rate of 98.8% of all terms for diagnoses, the procedures could not be mapped and hence an extension to the OMOP standard terminologies had to be made.

Overall, the data of 3 million inpatients with approximately 26 million conditions, 21 million procedures, and 23 million observations have been imported.

A federated query to identify a cohort of colorectal cancer patients was successfully executed and yielded 16,701 patient cases visualized in a Sunburst plot.

Conclusion  OMOP/OHDSI is a viable open source solution for data integration in a German research consortium. Once the terminology problems can be solved, researchers can build on an active community for further development.

[Does annual simulation training influence the safety climate of a university hospital? : Prospective 5‑year investigation using dimensions of the safety attitude questionnaire]

BACKGROUND

Simulation-based training with a focus on non-technical skills can have a positive influence on safety relevant attitudes of participants. If an organization succeeds in training sufficient staff, it may experience a positive change in the safety climate. As the effects of a single training are of a transient nature, annual training sessions may lead to an incremental improvement of safety relevant attitudes of employees over time. In spring 2012 the Department of Anesthesia at the University Hospital of Erlangen established an annual simulation-based training for staff members (e.g. consultants, trainee anesthetists and nurse anesthetists).

OBJECTIVE

The study aimed to test whether an annual simulation-based training would result in an incremental longitudinal improvement in attitudes towards teamwork, safety and stress recognition.

METHODS

A survey comprising three domains (teamwork climate, safety climate and stress recognition) of the safety attitudes questionnaire (SAQ) and items addressing briefing and speaking up was distributed to all participants in an annual in-house simulation training. Participants filled out the questionnaire in the morning of each training day. The attitudes were measured before the first training series in 2012, 6 months after the first training and then every year (2013-2016). Participants generated a personalized identification code which allowed individuals to be anonymously tracked over time. Results of the 5‑point Likert scale were transformed to a 100-point scale. Results were calculated at the group level and at the individual level. Univariable linear regression was used to calculate mean changes per year.

RESULTS

Over a period of 5 years (2012-2016) a total of 255 individuals completed the questionnaire. Each year, 14-20% of all nurse anesthetists and 81-90% of all anesthetists participated in the simulation-based training. As a result of annual staff turnover 16-24% of participants were new staff members. A personalized code allowed the before and after comparison of 99 staff members who had participated twice or more. Physicians had a higher mean score for teamwork climate before the first training (+8.7 p < 0.001). Mean teamwork climate and safety climate scores before the first training increased over a period of 5 years (3.11 for teamwork climate, p < 0.001 and 2.73 for safety climate, p < 0.001). Repeat participation led to a bigger mean change of individual attitudes in nurse anesthetists: teamwork climate 5.2 (nurses) vs. 1.4 (physicians) and safety climate 5.3 (nurses) vs. 2.8 (physicians) without reaching significance. Participants acknowledged the importance of briefings but confirmed their existence in less than half of the cases. The frequency of briefings increased over the 5‑year period. There were no changes in attitude towards speaking up.

CONCLUSION

Over a 5-year period, small positive changes in attitudes towards teamwork and safety occurred. Low participation of nurse anesthetists as well as personnel turnover may have weakened the impact of simulation-based training on the safety climate.

Septic cardiomyopathy in rat LPS-induced endotoxemia: relative contribution of cellular diastolic Ca(2+) removal pathways, myofibrillar biomechanics properties and action of the cardiotonic drug levosimendan

Cardiac dysfunction is a common complication in sepsis and is characterized by forward pump failure. Hallmarks of septic cardiomyopathy are decreased myofibrillar contractility and reduced Ca(2+) sensitivity but it is still not clear whether reduced pump efficiency is predominantly a diastolic impairment. Moreover, a comprehensive picture of upstream Ca(2+) handling mechanisms and downstream myosin biomechanical parameters is still missing. Ca(2+)-sensitizing agents in sepsis may be promising but mechanistic insights for drugs like levosimendan are scarce. Here, we used an endotoxemic LPS rat model to study mechanisms of sepsis on in vivo hemodynamics, multicellular myofibrillar Ca(2+) sensitivity, in vitro cellular Ca(2+) homeostasis and subcellular actomyosin interaction with intracardiac catheters, force transducers, confocal Fluo-4 Ca(2+) recordings in paced cardiomyocytes, and in vitro motility assay, respectively. Left ventricular ejection fraction and myofibrillar Ca(2+) sensitivity were depressed in LPS animals but restored by levosimendan. Diastolic Ca(2+) transient kinetics was slowed down by LPS but ameliorated by levosimendan. Selectively blocking intracellular and sarcolemmal Ca(2+) extrusion pathways revealed minor contribution of sarcoplasmic reticulum Ca(2+) ATPase (SERCA) to Ca(2+) transient diastole in LPS-evoked sepsis but rather depressed Na(+)/Ca(2+) exchanger and plasmalemmal Ca(2+) ATPase. This was mostly compensated by levosimendan. Actin sliding velocities were depressed in myosin heart extracts from LPS rats. We conclude that endotoxemia specifically impairs sarcolemmal diastolic Ca(2+) extrusion pathways resulting in intracellular diastolic Ca(2+) overload. Levosimendan, apart from stabilizing Ca(2+)-troponin C complexes, potently improves cellular Ca(2+) extrusion in the septic heart.

Influence of intensive care treatment on the protein binding of sufentanil and hydromorphone during pain therapy in postoperative cardiac surgery patients

BACKGROUND

Our objective was to evaluate the effect of intensive care treatment on the protein binding of sufentanil and hydromorphone in cardiac surgery patients during postoperative analgesia using a target-controlled infusion (TCI) and patient-controlled analgesia (PCA).

METHODS

Fifty adult patients were enrolled in this prospective randomized study; of which, 49 completed the study (age range 40-81 yr). Sufentanil was administered as an analgesic intraoperatively, and hydromorphone was dosed after operation with TCI and PCA until 8 a.m. on the first postoperative day. Arterial plasma samples were collected for drug and protein concentration measurements up to 24 h after cardiac surgery. Corresponding patient data were collected from the electronic patient data system. After explorative data analysis with principal component analysis, multivariate regression analysis and non-linear mixed effects modelling was used to study the effect of treatment on protein binding.

RESULTS

Data of 35 patients were analysed. The median protein binding of sufentanil and hydromorphone was 88.4% (IQ range 85.7-90.5%) and 11.6% (IQ range 9.5-14.3%), respectively. Free fraction of sufentanil increased towards the end of the study period, whereas hydromorphone free fraction remained nearly constant. The total sufentanil concentration and volume balance were identified as significant covariates for the protein binding of sufentanil. For the protein binding of hydromorphone, no significant covariate effects were found.

CONCLUSIONS

Sufentanil protein binding was significantly dependent on changes in the total drug concentration and volume balance addressing the importance of adequate dosing and fluid-guided therapy. Hydromorphone protein binding was nearly constant throughout the study period.

CLINICAL TRIAL REGISTRATION

EudraCT 2011-003648-31 and ClinicalTrials.gov: NCT01490268.

[Implementation of a patient data management system. Effects on intensive care documentation]

BACKGROUND

Patient data management systems (PDMS) enable digital documentation on intensive care units (ICU). A commercial PDMS was implemented in a 25-bed ICU replacing paper-based patient charting. The ICU electronic patient record is completely managed inside the PDMS. It compiles data from vital signs monitors, ventilators and further medical devices and facilitates some drug dose and fluid balance calculations as well as data reuse for administrative purposes. Ventilation time and patient severity scoring as well as coding of diagnoses and procedures is supported. Billing data transferred via interface to the central billing system of the hospital. Such benefits should show in measurable parameters, such as documented ventilator time, number of coded diagnoses and procedures and others. These parameters influence reimbursement in the German DRG system. Therefore, measurable changes in cost and reimbursement data of the ICU were expected.

MATERIAL AND METHODS

A retrospective analysis of documentation quality parameters, cost data and mortality rate of a 25-bed surgical ICU within a German university hospital 3 years before (2004-2006) and 5 years after (2007-2011) PDMS implementation. Selected parameters were documented electronically, consistently and reproducibly for the complete time span of 8 years including those years where no electronic patient recording was available. The following parameters were included: number of cleared DRG, cleared ventilator time, case mix (CM), case mix index (CMI), length of stay, number of coded diagnoses and procedures, detailed overview of a specific procedure code based on daily Apache II and TISS Core 10 scores, mortality, total ICU costs and revenues and partial profits for specific ICU procedures, such as renal replacement therapy and blood products.

RESULTS

Systematic shifts were detected over the study period, such as increasing case numbers and decreasing length of stay as well as annual fluctuations in severity of disease seen in the CM and CMI. After PDMS introduction, the total number of coded diagnoses increased but the proportion of DRG relevant diagnoses dropped significantly. The number of procedures increased (not significantly) and the number of procedures per case did not rise significantly. The procedure 8-980 showed a significant increase after PDMS introduction whereas the DRG-relevant proportion of those procedures dropped insignificantly. The number of ventilator-associated DRG cases as well as the total ventilator time increased but not significantly. Costs and revenues increased slightly but profit varied considerably from year to year in the 5 years after system implementation. A small increase was observed per case, per nursing day and per case mix point. Additional revenues for specific ICU procedures increased in the years before and dropped after PDMS implementation. There was an insignificant increase in ICU mortality rate from 7.4 % in the year 2006 (before) to 8.5 % in 2007 (after PDMS implementation). In the following years mortality dropped below the base level.

CONCLUSION

The implementation of the PDMS showed only small effects on documentation of reimbursement-relevant parameters which were too small to set off against the total investment. The method itself, a long-term follow-up of different parameters proved successful and can be adapted by other organizations. The quality of results depends on the availability of long-term parameters in good quality. No significant influence of PDMS on mortality was found.

Equipment for pre-hospital airway management on Helicopter Emergency Medical System helicopters in central Europe

BACKGROUND

For advanced out-of-hospital airway management, skilled personnel and adequate equipment are key prerequisites. There are little data on the current availability of airway management equipment and standards of medical staff on Helicopter Emergency Medical System (HEMS) helicopters in central Europe.

METHODS

An internet search identified all HEMS helicopters in Austria, Switzerland and Luxembourg. We identified 15 HEMS helicopter bases in Switzerland, 28 in Austria and three in Luxembourg. A questionnaire was sent to all bases, asking both for the details of the clinical background and experience of participating staff, and details of airway management equipment carried routinely on board.

RESULTS

Replies were received from 14 helicopter bases in Switzerland (93%), 25 bases in Austria (89%) and all three bases in Luxembourg. Anaesthesiologists were by far the most frequent attending physicians (68-85%). All except one bases reported to have at least one alternative supraglottic airway device. All bases had capnometry and succinylcholine. All bases in the study except two in Austria had commercial pre-packed sets for a surgical airway. All helicopters were equipped with automatic ventilators, although not all were suitable for non-invasive ventilation (NIV; Switzerland: 43%, Austria: 12%, Luxembourg: 100%). Masks for NIV were rarely available in Switzerland (two bases; 14%) and in Austria (three bases; 12%), whereas all three bases in Luxembourg carried those masks.

CONCLUSION

Most HEMS helicopters carry appropriate equipment to meet the demands of modern advanced airway management in the pre-hospital setting. Further work is needed to ensure that appropriate airway equipment is carried on all HEMS helicopters.

[Airway management in the German air rescue service]

BACKGROUND

The equipment carried in rescue helicopters concerning airway management is not standardized in Germany. The current DIN/EN equipment regulations are not detailed for every single item resulting in differences between the various rescue bases. The aim of this survey was to show existing differences in airway management equipment.

METHODS

All 72 German air rescue helicopters listed in the ADAC air rescue stations atlas 2006/2007 received a questionnaire.

RESULTS

A total of 64 bases returned the questionnaire (88.9%). The most frequent types of supraglottic airway equipment were laryngeal masks (67.2%), Combitubes (32.8%) and laryngeal tubes (29.7%). Cricothyrotomy sets were more frequent (68.8%) than tracheostomy sets (35.9%). Of the bases 18.8% had masks suitable for non-invasive ventilation (NIV). All helicopters were equipped to carry out capnometry.

CONCLUSIONS

The rescue helicopters were found to have differences with respect to the equipment carried for airway management. Medical treatment according to current guidelines is possible on almost all bases.

Prehospital airway management on rescue helicopters in the United Kingdom

Adequate equipment is one prerequisite for advanced, out of hospital, airway management. There are no data on current availability of airway equipment on UK rescue helicopters. An internet search revealed all UK rescue helicopters, and a questionnaire was sent to the bases asking for available airway management items. We identified 27 helicopter bases and 26 (96%) sent the questionnaire back. Twenty-four bases (92%) had at least one supraglottic airway device; 16 (62%) helicopters had material for establishing a surgical airway (e.g. a cricothyroidotomy set); 88% of the helicopters had CO(2) detection; 25 (96%) helicopters carried automatic ventilators; among these, four (15%) had sophisticated ventilators and seven (27%) helicopters carried special face masks suitable for non-invasive ventilation. We found a wide variation in the advanced airway management equipment that was carried routinely on air ambulances. Current guidelines for airway management are not met by all UK air ambulances.

[Patient-controlled analgesia (PCA) in outpatients with cancer pain. Analysis of 1,692 treatment days]

INTRODUCTION

In the home-care setting, cancer pain patients in need of parenteral analgesia have to be switched to patient-controlled analgesia using portable pumps. But there is a paucity on data on the logistic requirements or the success rate of such a cost-intensive therapy performed by specialized home-care services.

METHODS

In a retrospective study we analyzed data on care intensity, logistics and outcome of 46 consecutive palliative cancer patients with patient-controlled analgesia (PCA) in a home-care setting.

RESULTS

On days 1, 2, and 3 of PCA the switch to parenteral analgesia resulted in a significant increase of the median daily opioid dose in comparison to the dose just prior to PCA. Concurrently, pain scores were significantly reduced. The median duration of PCA was 25 days (range 2-189 days). On average, each patient was seen by the home-care team every 7.4 days. The median duration of the home visits was 60 min (range, 10-190 min). Of the visits 20% were unscheduled, most of these visits being due to problems regarding analgesia. Most patients died at home. Insufficient analgesia required prefinal hospitalization in only a single case.

CONCLUSION

If the indications are correct, intravenous PCA for palliative cancer pain patients results in higher opioid consumption and better pain control. Home-care PCA requires a lot of human and financial resources, but pain-related hospitalization can be prevented.

Target controlled anaesthetic drug dosing

It belongs to the particularities of anaesthesia that the conscious response of the patient to drug therapy is not available for the adjustment of drug therapy and that the side-effects of anaesthetic drug therapy would be in general lethal if no special measures were taken such as artificial ventilation. Both conditions do not allow for a slow, time-consuming titration of drug effect towards the therapeutically effective window, but measures have to be taken to reach a therapeutic target fast (within seconds to a few minutes), reliably, and with precision. Integrated pharmacokinetic-pharmacodynamic models have proved to be a useful mathematical framework to institute such drug delivery to patients. The theory of model-based interactive drug dosing on the basis of common pharmacokinetic-pharmacodynamic (pk-pd) models is outlined and the target-controlled infusion system (TCI) is presented as a new anaesthetic dosing technique that has developed during the last decade. Whereas TCI presents an open-loop dosing strategy (the past output does not influence the future input), current research deals with the model-based adaptive closed-loop administration of anaesthetics. In these systems the past output is used to adapt and individualize the initial pk-pd model to the patients and thus has an influence on future drug dosing which is based on the adapted model.

Pharmacokinetics and pharmacodynamics of GPI 15715 or fospropofol (Aquavan injection) - a water-soluble propofol prodrug

Propofol (2,6-diisopropylphenol) is inadequably soluble in water and is therefore formulated as a lipid emulsion. This may have disadvantages when propofol is used to provide total intravenous anaesthesia or especially during long-term sedation. There has been considerable interest in the development of new propofol formulations or propofol prodrugs. GPI 15715 or fospropofol (Aquavan injection; Guilford Pharmaceutical, Baltimore, MD) is the first water-soluble prodrug that has been thoroughly studied in human volunteers and patients. GPI 15751 or fospropofol is cleaved by alkaline phosphatase to phosphate, formaldehyde and propofol. Formaldehyde is rapidly metabolised to formate. Although a formate accumulation is the principal pathomechanism responsible for the toxicity of methanol ingestion, so far there has been no report of toxicity due to the administration of fospropofol or other phosphate ester prodrugs, such as fosphenytoin. Fosphenytoin has been successfully introduced into the market for the treatment of status epilepticus in 1996. The main side-effects were a feeling of paraesthesia after rapid i.v. administration of GPI 15715 or fospropofol, which has also been described for fosphenytoin. The pharmacokinetics of GPI 15715 or fospropofol could be described by a combined pharmacokinetic model with a submodel of two compartments for GPI 15715 and of three compartments for propofol(G). The liberated propofol(G) compared to lipid-formulated propofol showed unexpected pharmacokinetic and pharmacodynamic differences. We found a significantly greater V(c), V(dss), significantly shorter alpha- and beta-half-life and a longer MRT (mean residence time) for propofol(G). The pharmacodynamic potency of propofol(G) appears to be higher than propofol when measured by EEG and clinical signs of hypnosis. In summary, GPI 15715 or fospropofol was well suited to provide anaesthesia or conscious sedation.

[Equipment of a land-based emergency medical service in Bavaria: a questionnaire]

BACKGROUND

Currently valid DIN regulations concerning the emergency equipment in physician-staffed rescue vehicles are not well defined for every single item. This leads to variations in the equipment in different ambulances. The aim of this study was to demonstrate the differences with some exemplary equipment purchases.

METHODS

All 218 physician-staffed rescue vehicles in Bavaria received a questionnaire asking for current emergency medical equipment and planned items.

RESULTS

A reply was received from 177 of the 218 bases (81%). The main results were that 88% of all bases had a 12-lead ECG, a portable emergency ventilator was available at 93% of all bases, 77% had alternative airway management devices (set for emergency coniotomy 71%, laryngeal mask 26%), expiratory CO(2 )measuring was available at 32% and 31 bases (18%) had fibrinolytic drugs in the emergency vehicles.

CONCLUSIONS

There are vast differences among the emergency physician-staffed ambulances concerning the equipment which means that medical treatment according to current recommendations is not always possible at all bases.

Teletherapeutic drug administration by long distance closed-loop control of propofol †

BACKGROUND

The objective of this pilot study was to investigate the feasibility of an EEG-controlled closed-loop administration of propofol over a long distance of about 200 km.

METHODS

We performed a teletherapeutic propofol infusion during total intravenous anaesthesia with propofol in 11 patients undergoing general surgery. The teletherapeutic system consisted of a computer at the patient site in Munich and a computer at the control site in Erlangen, which were connected via the internet through a virtual private network. The patient's EEG signal was sent to the control site computer, where the median frequency (MEF) of the EEG power spectrum was calculated. The propofol infusion, determined by a model-based adaptive feedback algorithm to maintain a MEF of 1.5 to 2 Hz, was sent to the patient site computer connected to the infusion pump. The quality of the control was assessed by the performance error defined as the percentage deviation of the measured MEF from the set point and the necessity of interventions by the anaesthetist at the patient site.

RESULTS

During closed-loop administration of propofol [83 (52) min] the median performance error of the system was - 4.6 (4.4)% and the median absolute performance error was 18.8 (5.7)%. From a total number of 10 905 transmitted EEG epochs, there were five epochs with transmission errors, without further consequences for drug control. In one patient, teletherapy was stopped because the internet connection was interrupted.

CONCLUSIONS

Teletherapeutic drug administration could be realized over a longer distance. Further studies have to investigate the practicability and safety of teletherapeutic drug control in anaesthesia.

[Pharmacodynamics of two different propofol formulations]

BACKGROUND

Propofol is nowadays available in various lipid formulations. We compared two different propofol formulations with respect to pharmacodynamics, using the EEG and clinical signs.

MATERIALS AND METHODS

Ten volunteers received Diprivan 1% and Propofol 1% MCT Fresenius as a computer controlled infusion with increasing propofol target concentrations. A sigmoid E(max) model with effect compartment was estimated for the median frequency of the EEG power spectrum, based on measured arterial propofol plasma concentrations. Clinical pharmacodynamics were assessed by reaction on acoustic stimuli, eyelid reflex and corneal reflex.

RESULTS

The drugs did not differ in pharmacodynamics with respect to EEG (EC(50) 2.1+/-0.6 for Diprivan and 2.1+/-0.5 microg/ml for Propofol Fresenius) and clinical signs. The pharmacodynamic model was characterized by a steep concentration effect relationship and a distinct hysteresis between propofol plasma concentration and effect (k(e0) 0.12+/-0.04 and 0.12+/-0.5 min(-1)).

CONCLUSIONS

The investigated lipid formulations have no influence on the pharmacodynamics of propofol.

Concurrent recording of AEP, SSEP and EEG parameters during anaesthesia: a factor analysis

BACKGROUND

Spontaneous EEG, mid-latency auditory evoked potentials (AEP) and somatosensory evoked potentials (SSEP) have been used to monitor anaesthesia. This poses the question as to whether or not EEG, AEP and SSEP vary in parallel with varying conditions during surgical anaesthesia.

METHODS

A total of 81 variables (31 EEG, 22 SSEP, 28 AEP) were simultaneously recorded in 48 surgical patients during anaesthesia. A total of 307 cases of the 81 variables in stable anaesthetic states were recorded. A factor analysis was performed for this data set.

RESULTS

Sixteen variables were excluded because of multicollinearity. We extracted 13 factors with eigenvalues >1, representing 78.3% of the total variance, from the remaining 65 x 307 matrix. The first three factors represented 12%, 11% and 10% of the total variance. Factor 1 had only significant loadings from EEG variables, factor 2 only significant loadings from AEP variables and factor 3 only significant loadings from SSEP variables.

CONCLUSION

EEG, AEP and SSEP measure different aspects of neural processing during anaesthesia. This gives rise to the hypothesis that simultaneous monitoring of these quantities may give additional information compared with the monitoring of each quantity alone.

["The intensive care simulator": a new teaching-concept to train severe sepsis management]

BACKGROUND AND OBJECTIVE

In addition to basic research and development of new therapeutic strategies, the education of health care professionals who manage sepsis patients is an important step to decrease the high mortality of severe sepsis. Patient simulators are increasingly used for teaching in anaesthesia. A training program in sepsis management was developed, using a full-scale anaesthesia simulator including the setting of a modern intensive care unit, and its results were evaluated by means of a questionnaire.

METHODS

The simulator is controlled from a separate room using a controlling computer provided with physiological models and pharmacokinetic as well as pharmacodynamic patterns of substances commonly used in anaesthesia and intensive care. An important element of the training program is the subsequent debriefing with different modules, according to the individual deficits and needs of the participants detected during simulation.

RESULTS

From September 2002 to July 2004 82 physicians participated in the training program. 4 weeks after the training 52 % of the participants stated that they had changed their treatment behaviour due to the training content. They assessed the interactive simulator workshop semiquantitatively on a scale from 1 ("absolutely correct") to 7 ("not correct at all") as follows: Sepsis simulation training (SST) improves identification (mean+/-SD) (2.3 +/- 1.3) and treatment (2.5 +/- 1.2) of patients with severe sepsis, and SST including true-life scenarios is more appropriate than traditional lectures (1.5 +/- 0.7).

CONCLUSION

The presented SST could be an effective way to train intensive care specialists in severe sepsis management.

Pr�zision von ?target-controlled infusion? (TCI) mit zwei unterschiedlichen Propofolformulierungen

BACKGROUND

Target-controlled infusion (TCI) of propofol was initially realized as a device for prefilled syringes (Diprifusor). New TCI systems can be used with any propofol formulation. We compared two different propofol formulations with respect to accuracy of TCI and pharmacokinetics.

MATERIALS AND METHODS

A total of 10 volunteers received Diprivan 1% and Propofol 1% MCT Fresenius as TCI using the pharmacokinetic model of the Diprifusor. The prediction error was determined from measured arterial concentrations. A three-compartment model was fitted to the concentration data.

RESULTS

The median prediction error and the median absolute prediction error were -1.4% and 23.3% for Diprivan, and -5.9% and 17.8% for Propofol Fresenius. The drugs did not differ in pharmacokinetics but showed a smaller central volume of distribution than used for infusion control.

CONCLUSIONS

The pharmacokinetic model of Diprifusor can also be used for TCI of Propofol Fresenius. The large volume of distribution in this model may cause an overshoot in concentration.

Simulatorgestütztes, modulares Human Factors Training in der Anästhesie

BACKGROUND

Human factors (HF) play a major role in crisis development and management and simulator training can help to train HF aspects. We developed a modular training concept with psychological intensive briefing. The aim of the study was to see whether learning and transfer in the treatment group (TG) with the module "communication and team-cooperation" differed from that in the control group (CG) without psychological briefing ("anaesthesia crisis resource management type course").

METHODS

A total of 34 residents (TG: n=20, CG: n=14) managed 1 out of 3 scenarios and communication patterns and management were evaluated using video recordings. A questionnaire was answered at the end of the course and 2 months later participants were asked for lessons learnt and behavioral changes.

RESULTS

Good communication and medical management showed a significant correlation (r=0.57, p=0.001). The TG showed greater initiative ( p=0.001) and came more often in conflict with the surgeon ( p=0.06). The TG also reported more behavioral changes than the CG 2 months later. The reported benefit of the simulation was training for rare events in the CG, whereas in the TG it was issues of communication and cooperation ( p=0.001).

CONCLUSIONS

A training concept with psychological intensive briefing may enhance the transfer of HF aspects more than classical ACRM.

Modelling the pharmacodynamic interaction between remifentanil and propofol by EEG-controlled dosing

BACKGROUND AND OBJECTIVE

Knowledge of the pharmacodynamic interaction between remifentanil and propofol is important to permit optimal dosage strategies. We studied this pharmacodynamic interaction using the median power frequency of the processed electroencephalogram as a control parameter for feedback-controlled dosing of propofol.

METHODS

Twenty-one patients were given total intravenous anaesthesia with remifentanil and propofol. During three target-controlled infusion regimens, the target concentrations of remifentanil (5, 10, 15 ng mL(-1)) and propofol dosing were automatically adjusted to keep the median power frequency in the range 2 +/- 0.5 Hz. In each patient and during each remifentanil target concentration, four arterial propofol/remifentanil concentration pairs were measured. The type of interaction was tested using the relative distance from the line of additivity and the isobole was modelled using Bernstein splines.

RESULTS

The results from 13 patients were used for data analysis. The measured remifentanil concentrations during the three targets were (mean +/- SD): 3.6 +/- 0.9, 8.1 +/- 2.5 and 12.4 +/- 2.8 ng mL(-1). The corresponding propofol concentrations were 2.64 +/- 0.86, 2.13 +/- 0.58 and 2.09 +/- 0.58 microg mL(-1). The data were best described with an additive type of interaction and the isobole was estimated using: ((c)Remifentanil/64.2 ng mL(-1)) + ((c)Propofol/2.61 microg mL(-1)) = 1.

CONCLUSIONS

Within the studied concentration range, remifentanil and propofol showed an additive type of pharmacodynamic interaction on the electroencephalogram.

Testing and modelling the interaction of alfentanil and propofol on the EEG

BACKGROUND AND OBJECTIVE

For total intravenous anaesthesia an opioid is often combined with a hypnotic. A supra-additive interaction has been reported for clinical signs such as loss of consciousness or loss of the eyelash reflex. This study investigated the type of interaction of alfentanil and propofol on the electroencephalogram.

METHODS

Twenty patients scheduled for abdominal surgery were enrolled in the study. Anaesthesia was induced and maintained with alfentanil and propofol. Each patient received a target-controlled infusion of alfentanil. Three target concentrations of 150, 225 and 300 ng mL(-1) were applied to each patient in random order. Propofol was added to the alfentanil infusion by a feedback system. The set point was the range of 1.5-2.5 Hz median frequency of the electroencephalogram. Four arterial blood samples were taken within the last 20 min of each period. The mean drug concentrations were used to determine the type of interaction and an isobole was estimated by fitting Bernstein spline functions to the data.

RESULTS

In 17 patients, all three alfentanil target concentrations could be administered. The test for supra-additivity as well as the isobole construction resulted in an additive type of interaction. The line of additivity cA/cA0 + cP/cP0 = 1 was best fitted for the values (standard deviation) cA0 = 1240 (51)ng mL(-1) and cP0 = 5.21 (0.36) microg mL(-1).

CONCLUSIONS

The type of interaction between alfentanil and propofol on the electroencephalogram in the investigated dose range is additive. This gives the freedom and need to select the appropriate dosing ratio of alfentanil and propofol by other considerations.

Stereoselective pharmacokinetics of ketamine: R(-)-ketamine inhibits the elimination of S(+)-ketamine

OBJECTIVE

We investigated the pharmacokinetics of ketamine with special regard to enantiomer-specific differences.

METHODS

Ten healthy young male volunteers (mean age, 28 +/- 4 years; mean weight, 79 +/- 11 kg) received racemic ketamine and S(+)-ketamine in a randomized double-blind crossover study. Drugs were administered by a computer-controlled device. Two infusion cycles with linearly increasing targets [slope, 0.1 microg x ml(-1) x min(-1) for S(+)-ketamine and 0.2 microg x ml(-1) x min(-1) for racemic ketamine] were administered. Concentrations of the ketamine enantiomers were determined from arterial blood, and pharmacokinetic parameters were estimated with a 2- and 3-compartment model.

RESULTS

The total doses needed to reach defined end points were 271 +/- 80 mg and 409 +/- 75 mg for S(+)-ketamine and racemic ketamine, respectively (P <.05). S(+)-ketamine showed a significantly higher clearance (26.3 +/- 3.5 ml x kg(-1) x min(-1)) compared with racemic ketamine (14.8 +/- 1.7 ml x kg(-1) x min(-1); P <.05) and R(-)-ketamine (13.8 +/- 1.3 ml x kg(-1) x min(-1); P <.05). Furthermore, the clearance of the S (+)-ketamine was smaller in the racemate (18.5 +/- 0.7 ml x kg(-1) x min(-1); P <.05) than for the pure isomer.

CONCLUSIONS

These results demonstrate that R(-)-ketamine inhibits the elimination of S(+)-ketamine.

200 years of nitrous oxide (laughing gas)--and the end of an era?

The history of nitrous oxide is more than 200 years old and its clinical use as anaesthetic is more than 150 years old. The symposium discussed the question whether nitrous oxide should maintain its traditional role as a component of the anaesthetic breathing gas in everdays anaesthetic procedure or whether there are indications not to continue the regular use of nitrous oxide. As a matter of fact the continued use of nitrous oxide will not change its clinical pharmacology and one may argue that every year of additional experience with this drug is likely to increase the safety of its application. However, one has steadily to judge the risks of this drug against its alternatives and these have changed dramatically over the past decades. The new anaesthesia machines allow the combination of oxygen and air as breathing gas, there are new inhalational agents (e. g. desflurane, xenon) as controllable as nitrous oxide and new i. v. agents. As a conclusion of these developments the Department of Anaesthesiology at the Universität Erlangen-Nürnberg has decided to cease the traditional use of nitrous oxide.

[First results in an animal model on stents sheathed with a polytetrafluoroethylene membrane]

OBJECTIVES

Mechanisms of restenosis after coronary stent implantation include marked intimal proliferation as well as vascular tissue protrusion through the meshes. Thus, stent sheathed with membranes may be an alternative to improve the long-term outcome.

METHODS

Seventeen cylindric serpentine shaped 316L stainless steel stents (nominal diameter 3.0 mm, length 15 mm) lined and covered by a polytetrafluoroethylene (PTFE) membrane and 10 unsheathed devices were implanted into the iliac arteries of 14 New Zealand White Rabbits for an observation period of 1 - 10, 11 - 20, and 21 - 30 weeks. After sacrificing the animals, specimens were harvested, fixed in formalin, processed in paraffin, serially sliced into 5 microm thick preparations, and stained (hematoxylin & eosin, elastica von Gieson).

RESULTS

The polytetrafluoroethylene membrane stents increased the vascular lumen significantly (p < 0.04) to 1185.3 - 1620.4 microm compared with the native segments (655.6 +/- 268.8 microm). In the stainless steel stents the lumen decreased from 1873.1 microm to 719.1 microm. None of the devices penetrated the internal elastic membrane. There was no inflammatory vascular reaction. Compared to the native segments, the amount of elastic fibres was slightly less (native: 8.9 %, PTFE: 2.3 - 3.5 %, without PTFE: 1.9 - 5.5 %) whereas the collagen fibres increased marginally (native: 5.1 %, PTFE: 6 - 6.9 %, without PTFE: 6.4 - 8.4 %).

CONCLUSIONS

In the rabbit iliac artery, stents sheathed with a microporous polytetrafluoroethylene membrane showed good tissue compatibility with no restenosis. These results warrant clinical trials.

A new model of electrically evoked pain and hyperalgesia in human skin: the effects of intravenous alfentanil, S(+)-ketamine, and lidocaine

BACKGROUND

The authors used the analgesics alfentanil, S(+)-ketamine, and systemic lidocaine to examine a new human model of experimental pain and hyperalgesia.

METHODS

Transcutaneous electrical stimulation at a high current density (5 Hz, 67.5+/-6.6 mA) was used to provoke acute pain (numeric rating scale, 5 of 10), stable areas of secondary mechanical hyperalgesia to pin prick (43.6+/-32.1 cm2), and light touch (27.5+/-16.2 cm2) for 2 h. Alfentanil, S(+)-ketamine, and lidocaine were applied for 20 min in a double-blind, placebo-controlled, crossover design in 12 subjects using target controlled infusions.

RESULTS

In the placebo session, pain ratings and areas of hyperalgesia were stable during the stimulation period, which facilitated the assessment of analgesic effects. Alfentanil effectively inhibited electrically evoked pain and reduced pin prick hyperalgesia and allodynia during its infusion. S(+)-ketamine-induced inhibition of secondary hyperalgesia was more pronounced and lasted for the whole experimental protocol. Therapeutic levels of systemic lidocaine showed only marginal analgesic effects, but lasting antihyperalgesic effects.

CONCLUSIONS

A new model of electrically induced pain and hyperalgesia was established, which enabled assessment of the time course of analgesic and antihyperalgesic effects with high temporal resolution and minimum tissue damage and which was further validated by use of common intravenous anesthetics.

Postoperative pain control following remifentanil-based anaesthesia for major abdominal surgery

Eighty patients undergoing major abdominal surgery using remifentanil-based anaesthesia were randomly allocated in a double-blind manner to receive an intravenous bolus of fentanyl, buprenorphine, morphine or piritramide 20 min before the end of surgery. A reduced dose was administered postoperatively when patients reported moderate pain. Subsequent analgesia was provided by patient-controlled analgesia (PCA). The mean time from the end of anaesthesia to spontaneous respiration was 9 +/- 5 min. At first pain assessment, 63% of patients reported no or mild pain; 80% of patients required the second opioid bolus, those receiving piritramide needed the bolus significantly later than patients receiving buprenorphine or fentanyl. First PCA requirement also occurred significantly later in the piritramide group. This technique provided effective postoperative pain relief and transition to routine PCA and did not compromise recovery.

Population pharmacokinetics of propofol: a multicenter study

BACKGROUND

Target-controlled infusion is an increasingly common type of administration for propofol. This method requires accurate knowledge of pharmacokinetics, including the effects of age and weight. The authors performed a multicenter population analysis to quantitate the effects of covariates.

METHODS

The authors analyzed 4,112 samples of 270 individuals (150 men, 120 women, aged 2-88 yr, weighing 12-100 kg). Population pharmacokinetic modeling was performed using NONMEM (NONMEM Project Group, University of California, San Francisco, CA). Inter- and intraindividual variability was estimated for clearances and volumes. The effects of age, weight, type of administration and sampling site were investigated.

RESULTS

The pharmacokinetics of propofol were best described by a three-compartment model. Weight was found to be a significant covariate for elimination clearance, the two intercompartmental clearances, and the volumes of the central compartment, the shallow peripheral compartment, and the deep peripheral compartment; power functions with exponents smaller than 1 yielded the best results. The estimates of these parameters for a 70-kg adult were 1.44 l/min, 2.25 l/min, 0.92 l/min, 9.3 l, 44.2 l, and 266 l, respectively. For patients older than 60 yr the elimination clearance decreased linearly. The volume of the central compartment decreased with age. For children, all parameters were increased when normalized to body weight. Venous data showed a decreased elimination clearance; bolus data were characterized by increases in the volumes of the central and shallow peripheral compartments and in the rapid distribution clearance (Cl2) and a decrease in the slow distribution clearance (Cl3).

CONCLUSIONS

Pharmacokinetics of propofol can be well described by a three-compartment model. Inclusion of age and weight as covariates significantly improved the model. Adjusting pharmacokinetics to the individual patient should improve the precision of target-controlled infusion and may help to broaden the field of application for target-controlled infusion systems.

Present state of closed-loop drug delivery in anesthesia and intensive care

Although closed-loop systems have a long tradition in engineering, their continued use for patient care is limited to the last 20 years, with the exception of BICKFORD's pioneering work in 1950. During the past 2 years it has been shown that automated closed-loop systems for drug delivery can provide unique study designs for clinical research allowing experimental setups not realizable by traditional means. It is, however, evident that a lot of research and development has to be done on therapeutic closed-loop systems to solve the many questions related to their reliability, safety and use in the routine clinical setting.

[Analgo-sedation in intensive care: a quantitative, EEG-based trial with propofol 1% and 2%]

OBJECTIVE

The primary aim of this study was to find out whether adequate long-term sedation (> or = 72 h) can be achieved in critically ill patients with an EEG median frequency controlled closed-loop system for the application of propofol 1% and 2%. Moreover, we investigated the pharmacokinetics and pharmacodynamics of propofol with respect to possible tolerance and compared the quality of sedation of both propofol formulations and their lipid load.

PATIENTS AND METHODS

After institutional approval and written consent, 16 ASA II-IV patients were included in this study. Main inclusion criterion was the necessity for prolonged sedation/analgesia for at least 72 h. Sedation was induced and maintained using continuous infusion of propofol 1% (n = 7) or 2% (n = 9). Analgesia was maintained with continuous infusion of alfentanil. The EEG was recorded from four leads (Fp1,2 and C3,4) and the EEG median frequency was obtained from the power spectrum (0.5-32 Hz). Propofol was administered computer-controlled with a median frequency setpoint depending on the depth of sedation which was assessed clinically using a modified Ramsay score. Alfentanil was applied as TCI. Arterial plasma concentrations were measured by HPLC (propofol) and RIA (alfentanil). Pharmacokinetics of propofol and alfentanil were derived using a three compartment model.

RESULTS

All patients were successfully sedated for 77 +/- 9 h. The median EEG frequency during sedation was stable at 1.5 +/- 0.2 Hz. The sedation score increased from 1 in the first 12 h to values between 2 and 3 for the remaining sedation period. At the same time, propofol plasma concentrations increased from 0.7 +/- 0.3 microgram/ml to 1.8 +/- 1.3 micrograms/ml. The patients required an average of 2.5 mg/kg/h propofol and 0.030 mg/kg/h alfentanil. Pharmacokinetics of propofol 2% showed an increased volume of distribution when compared to propofol 1%. Alfentanil clearance was found to be reduced with four patients having extremely small clearance values (33 +/- 3 ml/min). Triglyceride values increased up to 4.5 +/- 1.2 mmol/l for patients receiving propofol 1% and remained within normal range for propofol 2%.

CONCLUSIONS

The EEG median frequency can be used for closed-loop control of propofol even for long-term sedation in critically ill patients. EEG median frequencies were similarly low as in deeply anaesthetised patients. No differences in quality of sedation were seen between the two propofol formulations, but propofol 2% seems to be advantageous due to lower lipid load and triglyceride values. Increasing concentrations of propofol at unchanged sedation scores and EEG median frequencies may indicate development of tolerance.

[Pharmacokinetic-pharmacodynamic modeling in phase II of drug development. A comparative study with young and old volunteers with benzdiazepine as an example]

Especially for medical disciplines like anesthesiology, which represent only a small economic market, drug development is a cost intensive and with respect to financial aspects a high risk task. This situation requires methods, which in the early stages of the drug development process of an interesting compound allow the establishment of a reliable and valid data set, in order to make decisions on the continuation or discontinuation of a project. Given a compound out of the group of benzodiazepines as an example this paper represents todays methods of integrated pharmacokinetic-pharmacodynamic modelling as well as the special computer aided strategies of drug dosing as powerful tools in the anaesthetic drug development process. It is shown that one can generate data concerning differences in drug requirement and drug duration in the therapeutic dose range between different groups of possible patients, e.g. young and elderly, as early as the early phase II. It is concluded that these clinical pharmacological tools and the high resolution data generated by them facilitate the Go/No-Go decision to proceed to phase III and enhance the likelihood of passing phase III successfully.

A rational approach to the control of sedation in intensive care unit patients based on closed-loop control

Optimal control of long-term sedation during therapy in the intensive care unit is difficult to achieve in a number of patients when based on commonly used clinical sedation scores alone. We therefore used the median frequency of the EEG power spectrum as a quantitative measure for closed-loop administration of propofol in 21 artificially ventilated patients (nine trauma, 12 non-trauma). The EEG setpoint was correlated with a clinical sedation score and defined such, that mechanical ventilation was tolerated. The sedative therapy was given for 31 +/- 30 h. Non-trauma patients required sedation with an EEG median frequency between 2 and 3 Hz (propofol consumption: 1.4 +/- 0.8 mg kg-1 h-1) and sedation seemed to follow some circadian patterns, whereas trauma patients needed significantly deeper sedation (EEG median frequency: 1-2 Hz; propofol consumption: 2.6 +/- 0.8 mg kg-1 h-1). We conclude that the EEG closed-loop system could safely and reliably administer propofol to maintain a predetermined level of sedation for patients in intensive care unit over a protracted time.