Patient simulation for training basic and advanced clinical skills.

INTRODUCTION: Patient simulators are increasingly used in the education and training of healthcare professionals. This paper describes the history of human patient simulator development, the features of contemporary simulators, the acquisition of basic and advanced clinical skills using patient simulators, and the benefits, cost, limitations and effectiveness of this innovative learning modality. SIMULATOR DEVELOPMENT: The development of human patient simulators began in the late 1960s, and accelerated in the late 1980s and early 1990s. Several simulator systems are now professionally manufactured, commercially available, and used at hundreds of medical centres, universities and colleges in the USA and throughout the world. Contemporary patient simulators have many clinical features, and look and respond to interventions with ever-increasing degrees of realism because sophisticated physiological and pharmacological models automatically control many features. SIMULATOR USE IN MEDICAL EDUCATION: Simulators are used to teach basic skills, such as respiratory physiology and cardiovascular haemodynamics, and advanced clinical skills, e.g. management of difficult airways, tension pneumothorax, pulmonary embolism and shock. BENEFITS, COSTS AND LIMITATIONS: The simulation laboratory offers distinct educational advantages, especially for learning how to recognise and to treat rare, complex, clinical problems. Costs of simulator-based educational programmes include facility, equipment and personnel. Current limitations include clinical realism of the patient manikin and faculty development.

Minocycline: stain devil?

Minocycline is the treatment of choice for acne vulgaris, the most common form of inflammatory acne, despite the increase in awareness of rare but significant side-effects. This paper discusses the undesirable side-effect of minocycline staining in permanent teeth.

Educational simulation of the electroencephalogram (EEG).

We describe a model for simulating a spontaneous electroencephalogram (EEG) and for simulating the effects of anesthesia on the EEG, to allow anesthesiologists and EEG technicians to learn and practice intraoperative EEG monitoring. For this purpose, we developed a linear model to manipulate the amplitude of the activity in each of the traditional EEG frequency bands. Burst suppression patterns are simulated by manipulating an overall gain. To demonstrate the model feasibility, model parameters for thiopental and isoflurane were estimated guided by published data on the EEG effects of these anesthetic drugs. Using these estimates, EEG time signals were simulated for isoflurane at various partial pressures, and for bolus intravenous doses of thiopental. Comparison with actual recorded EEG signals showed that the changes produced by isoflurane and thiopental in the simulated signals are very similar to the changes in the actual signals, which was confirmed by two clinicians with experience and routine practice in intraoperative EEG monitoring.

TWITCHER: a device to stimulate thumb twitch response to ulnar nerve stimulation.

OBJECTIVE: To design and fabricate a device to simulate evoked thumb adduction in response to ulnar nerve stimulation. METHODS: We implemented a computer-controlled, motorized thumb (TWITCHER) that responds to ulnar nerve stimulation by an unmodified peripheral nerve stimulator. Clinically realistic response patterns are generated for both depolarizing and non-depolarizing muscle relaxants and three modes of stimulation (single twitch, train-of-four, tetanus). RESULTS: The device has been used in a full-scale patient simulator for the last six years. DISCUSSION: TWITCHER has been well received by participants in simulation exercises including the use of neuromuscular blocking drugs.

Pharmacokinetic-pharmacodynamic model for educational simulations.

Pharmacokinetic-pharmacodynamic (PK-PD) models play an important role in educational simulations. The parameters of PK-PD models described in the scientific literature are obtained from studies in which the drug concentrations and the drug-effect data are measured simultaneously. Simultaneous PK-PD studies cannot be expected to incorporate all possible combinations of drugs and patient physiology that are desired for educational simulations. To solve this problem, we elaborate on the traditional simultaneous PK-PD model, creating a new model that accepts parameter data from different, more readily available, nonsimultaneous pharmacologic studies. These data are incorporated in the model using a novel estimation procedure for the parameters kc0 and EC50. A sensitivity analysis of the parameter estimation procedure confirms that the time of peak effect following a bolus and the dose-response curve are accurately reflected by the new model. It also demonstrates how inconsistencies among the different parameter sets affect simulation of the recovery phase. The model is extended to incorporate any monotonic parametric or nonparametric dose-response curve. For the neuromuscular relaxant vecuronium, we demonstrate that data from different pharmacologic studies are available, and that the described estimation procedure leads to parameter estimates that are within the standard deviations of the parameters determined in a simultaneous PK-PD study.

Influence of pulse oximetry and capnography on time to diagnosis of critical incidents in anesthesia: a pilot study using a full-scale patient simulator.

OBJECTIVE: Many studies (outcome, epidemiological) have tested the hypothesis that pulse oximetry and capnography affect the outcome of anesthetic care. Uncontrollable variables in clinical studies make it difficult to generate statistically conclusive data. In the present study, we eliminated the variability among patients and operative procedures by using a full-scale patient simulator. We tested the hypothesis that pulse oximetry and capnography shorten the time to diagnosis of critical incidents. METHODS: A simulator was programmed to represent a patient undergoing medullary nailing of a fractured femur under general anesthesia and suffering either malignant hyperthermia, a pneumothorax, a pulmonary embolism or an anoxic oxygen supply. One hundred thirteen anesthesiologists were randomly assigned to one of two groups of equal size, one with access to pulse oximetry and capnography data and the other without. Each anesthesiologist was further randomized to one of the four critical incidents. Each anesthetic procedure was videotaped. The time to correct diagnosis was measured and analyzed. RESULTS: Based on analysis of 91 of the subjects, time to diagnosis was significantly shorter (median of 432 s vs. >480 s) for the anoxic oxygen supply scenario (p = 0.019) with pulse oximetry and capnography than without. No statistical difference in time to diagnosis was obtained between groups for the other three critical incidents. CONCLUSIONS: Simulation may offer new approaches to the study of monitoring technology. However, the limitations of current simulators and the resources required to perform simulator-based research are impediments to wide-spread use of this tool.

Modeling obstetric cardiovascular physiology on a full-scale patient simulator.

To our knowledge, this is the first attempt at adapting an existing cardiovascular model to simulate the hemodynamics of a particular patient population. Despite attempts to define the physiologic alterations in advance, we discovered there were critical parameters not completely defined in the literature. These were discovered through the iterative process of testing, comparing resulting vital signs with targets, and literature review. A list of the parameters that should be sought for future modeling efforts is provided (Table 3), but this list is by no means exhaustive. As further work is performed in this area, additional independent and essential parameters will be identified (pressure characteristics of valvular anomalies, for example). To define a physiology that is less well described in the literature, empirical alterations and best-guess estimates of parameter changes will be required with significantly more iterations. Finally, we have described only modeling of cardiovascular physiology, modeling the respiratory system will require a similar process.

Comparison of tests for detecting leaks in the low-pressure system of anesthesia gas machines.

Small leaks in the low-pressure system (LPS) of the anesthesia gas machine can cause hypoxia or patient awareness. We sought to determine the relative sensitivities of the various tests recommended for detecting LPS leaks before anesthesia. Special adapters were fashioned to create leaks of six different sizes in the LPS that were equivalent to the following: a single 25-, 22-, 20-, or 15-gauge needle, two 15-gauge needles, or a 2.5-mm endotracheal tube connector. With each leak condition, five different leak tests were performed on three each of the following machines: Ohmeda Modulus I, Ohmeda Modulus II-Plus, and North American Dräger Narkomed (2A, 3 and 4), for a total of 54 leaks to be detected for each leak test (3 x 3 x 6). The number of leaks detected with each test was compared by Fisher's exact test, P < 0.05 being considered significant. Only the negative pressure leak test detected all 54 leaks, a significant difference from the positive pressure test, which detected the least number of leaks, 28 (P < 0.05). Some leak tests are more suitable for specific anesthesia machines. Adoption of the negative pressure test as a universal LPS leak test may prevent the risks associated with using the wrong test for the particular anesthesia machine: hypoxic gas or patient awareness.

Logistics of conducting a large number of individual sessions with a full-scale patient simulator at a scientific meeting.

OBJECTIVE: To design and implement the logistics of accommodating a large number of participants in individual, hands-on sessions on a full-scale patient simulator during a major scientific meeting or continuing medical education course. METHODS: We used our method during the 11th World Congress of Anaesthesiologists in Sydney, Australia to facilitate studying the impact of pulse oximetry and capnography on the time taken by anesthesiologists to correctly identify critical incidents on a full-scale patient simulator. Each study participant spent 15 minutes in 4 sections of the study area: the anesthesia and monitoring equipment briefing room, the simulator briefing room, the simulation room and the debriefing room. RESULTS: There were 113 participants during five days (15 during instructor training and 25, 23, 24 and 26 on subsequent exhibit days). We were oversubscribed daily. However, there were 9 no-shows during the 4 days of the study, which generated a participant absence rate of 9.2%. The average number of participants over the 4 days of the study was 24.5 per day compared to our capacity of 27 per day. The feedback we obtained from the participants about the simulation experience and the format of the exercise was positive and enthusiastic. CONCLUSIONS: We have developed a practical and viable method that can be adapted for use at scientific meetings and courses, which improves accessibility of individual, hands-on sessions on full-scale patient simulators to a larger audience than previously attainable. Our method is applicable for continuing medical education courses as well as research purposes in the form of prospective studies during scientific meetings and courses.

Duration of carbon dioxide absorption by soda lime at low rates of fresh gas flow.

STUDY OBJECTIVE: To determine the impact of a low fresh gas flow rate on the duration of carbon dioxide (CO2) absorption by soda lime. DESIGN: Nonclinical, experimental. SETTING: Experimental laboratory. METHODS: In vitro test with Sodasorb and a semiclosed breathing circle ventilating a test lung with a CO2 inflow of 250 ml per minute. Fresh gas flow rates of 0.25, 0.5, 1, 2, and 4 L/min were studied. MEASUREMENTS AND MAIN RESULTS: CO2 was measured at the breathing circuit test lung interface with a mainstream capnometer. Duration of CO2 absorption was determined as the time for the inspired CO2 tension (PICO2) to increase from 0 mm to 7 mm of mercury. The times of this interval were recorded four times for each fresh gas flow rate and compared by analysis of variance; p less than 0.05 was considered significant. Time to soda lime failure was significantly longer at 2 L/min than at 1 L/min fresh gas flow and at 1 L/min than at 0.25 L/min fresh gas flow. CONCLUSION: Because soda lime color indicators are unreliable, when a semiclosed breathing circle is used at a low rate of fresh gas flow without CO2 monitoring, the CO2 absorbent must be replaced more frequently.