Optimizing the tactile display of physiological information: vibro-tactile vs. electro-tactile stimulation, and forearm or wrist location

Anesthesiologists use physiological data monitoring systems with visual and auditory displays of information to monitor patients in the operating room (OR). The efficacy of visual-audio systems may impose an increase in patient risk when the demand for constant switching of attention between the patient and the visual monitoring system is high. This is evidenced by auditory alarms frequently being neglected in a noisy OR environment. Hence, the use of a complementary patient data monitoring system, which utilizes other sensory modalities, could be of great value. In this paper, we describe a series of experiments designed to determine the performances of a tactile display that could be used to convey patient's physiological information to the attending anesthesiologist. We tested both vibro-tactile and electro-tactile display prototypes in their ability to convey information using an alert scheme of four distinct tactile stimuli. Using pseudo-clinical data, the display was designed, for example, to provide an alert when a change in the monitored heart rate occurred. Based on previous research in human physiology and psychophysics, we selected the forearm and wrist of the user's non-dominant hand as the stimulation site. In our study of 30 subjects, we evaluated the response time and accuracy of tactile pattern recognition to compare (1) the performance of a vibro-tactile display on the forearm (VF) and an electro-tactile display on the forearm (EF), and (2) the localization of stimulation between the forearm (VF) and a vibro-tactile display on the wrist (VW). A post-study questionnaire was completed by each subject to assess the comfort and usability of the three prototypes. We found that both VF and VW were superior to the EF in both accuracy and comfort and, that there were no differences between the wrist and the forearm. In conclusion, the tactile-display prototypes designed to alert the clinician of adverse changes in a patient's physiological state efficaciously and unobtrusively delivered these data and warranted further investigation and development.

A non invasive wearable sensor for the measurement of brain temperature

As the thermoregulation centres are deep in the brain, the cerebral temperature is one of the most important markers of fever, circadian rhythms physical and mental activities. However due to a lack of accessibility, the brain temperature is not easily measured. The axillary, buccal, tympanic and rectal temperatures do not reflect exactly the cerebral temperature. Nevertheless the rectal temperature is used as probably the most reliable indicator of the core body temperature. The brain temperature can be measured using NMR spectroscopy, microwave radiometry, near infrared spectroscopy, ultra-sound thermometry. However none of those methods are amenable to long term ambulatory use outside of the laboratory or of the hospital during normal daily activities, sport, etc. The brain core temperature "BCT" sensor, developed by the Biomedical Microsensors dpt of LPM at INSA-Lyon is a flexible active sensor using "zero-heat-flow" principle. The sensor has been used for experimental measurement: brain temperature during mental activity, and in hospital for the study of circadian rhythms. The results are in agreement with the measurement by the rectal probe. There are 2 versions of this sensor: a non ambulatory for the use in hospitals, and an ambulatory version using teletransmission. We are working for improving the autonomy of the ambulatory version up to several days. This wearable biomedical sensor (WBS) can be used for circadian assessment for chronobiology studies and in medical therapies.

Evaluations of tactile displays of physiological monitoring

Physiological monitoring devices are used in the operating room (OR) to identify abnormal changes. They are currently detected by the anesthesiologist from direct observation of visual displays or by auditory cues. In an effort to improve OR safety, we explored the use of an alternative sensory modality, touch, to effectively and accurately convey patients' physiological information. To be accepted [1] such devices need to fulfill five criteria: learnability, errors, efficiency, memorability and satisfaction. We designed and compared various tactile displays in respect to some of these criteria.

Autonomic nervous system response to vibrating and electrical stimuli on the forearm and wrist

In today's operating rooms, anesthesiologists use physiological data monitoring systems with visual and auditory cues to receive patient information. The efficacy of these visual-audio systems is limited by the human limitations of these modalities. Previous studies have shown the potential use of a complementary, or alternate, patient data monitoring technology utilizing another psychophysically relevant modality: the sense of touch via vibro-tactile or electro-tactile stimulation. In this paper, we describe an experiment designed to determine whether the specific type and/or location of such a tactile stimulation device on the arm affects the autonomic nervous system response. In our study, each of 10 participants tested a vibro-tactile display on the forearm (VF), a vibro-tactile display on the wrist (VW), and an electro-tactile display on the forearm (EF) in random order. Using the LifeShirt, system, electrocardiogram (ECG), respiratory rate (Br), tidal volume (Vt) data were collected. Results showed a higher value of the heart rate and heart rate variability (HRV) when using the VF compared to the VW and EF. We also found that the HRV response for the three tactile prototypes was correlated with the accuracy of tactile pattern identification.

Target controlled infusion for kids: trials and simulations

Target controlled infusion (TCI) for Kids is a computer controlled system designed to administer propofol for general anesthesia. A controller establishes infusion rates required to achieve a specified concentration at the drug's effect site (C(e)) by implementing a continuously updated pharmacokinetic-pharmacodymanic model. This manuscript provides an overview of the system's design, preclinical tests, and a clinical pilot study. In pre-clinical tests, predicted infusion rates for 20 simulated procedures displayed complete convergent validity between two software implementations, Labview and Matlab, at computational intervals of 5, 10, and 15s, but diverged with 20s intervals due to system rounding errors. The volume of drug delivered by the TCI system also displayed convergent validity with Tivatrainer, a widely used TCI simulation software. Further tests, were conducted for 50 random procedures to evaluate discrepancies between volumes reported and those actually delivered by the system. Accuracies were within clinically acceptable ranges and normally distributed with a mean of 0.08 +/- 0.01 ml. In the clinical study, propofol pharmacokinetics were simulated for 30 surgical procedures involving children aged 3 months to 9 years. Predicted C(e) values during standard clinical practice, the accuracy of wake-up times predicted by the system, and potential correlations between patient wake-up times, C(e), and state entropy (SE) were assessed. Neither Ce nor SE was a reliable predictor of wake-up time in children, but the small sample size of this study does not fully accommodate the noted variation in children's response to propofol. A C(e) value of 1.9 mug/ml was found to best predict emergence from anesthesia in children.

Seizure detection by a novel wavelet packet method

We describe a novel wavelet-based method for the detection of seizure in patients with temporal lobe epilepsy. This method uses local discriminant bases and cross- data entropy algorithms to identify nodes of a wavelet packet dictionary that best discriminate preictal from ictal EEG signals. The algorithms are based on relative entropy criterion as a measure of discrepancy between different classes of signals. The frequency bands associated with these nodes were in the range of interest for seizure events. After selecting two bands, we determined the ratio of energies at the level of wavelet packet chosen by the cross-data entropy algorithm. Preliminary results demonstrate that the wavelet packet energy ratio could serve as a robust criterion in seizure detection.

Continuous noninvasive blood pressure measurement by pulse transit time

Blood pressure measurement is performed either invasively by an intra arterial catheter or noninvasively by cuff sphygmomanometry. The invasive method is continuous and accurate but has increased risk; the cuff is safe but less reliable and infrequent. A reliable continuous noninvasive blood pressure measurement is highly desirable. While the possibility of using pulse transit time to monitor blood pressure has previously been investigated, most studies were limited to calculating the correlation of the pulse transit time and blood pressure under rather static conditions. The relationship between the pulse transit time and blood pressure is yet to be clearly identified. This paper focuses on the modeling between the two values and presents results on cases where dramatic variation in blood pressure of the patient was induced by drug administration or surgical stimulation.

A Wavelet Approach to Detecting Electrocautery Noise in the ECG

A software approach has been developed for detecting electrocautery noise in the electrocardiogram (ECG) using a wavelet decomposition of the signal. With this approach, a clinical monitoring expert system can be forewarned of potential artefacts in trend values derived from the ECG, allowing it to proceed with caution when making decisions based on these trends. In 15 operations spanning 38.5 hours of ECG data, we achieved a false positive rate of 0.71% and a false negative rate of 0.33%. While existing hardware approaches detect the source of the noise without any ability to assess its impact on the measured ECG, our software approach detects only the presence of noise in the signal itself. Furthermore, the software approach is cheaper and easier to implement in a clinical environment than existing hardware approaches.

A Wavelet Approach to Detecting Electrocautery Noise in the ECG

A software approach has been developed for detecting electrocautery noise in the electrocardiogram (ECG) using a wavelet decomposition of the signal. With this approach, a clinical monitoring expert system can be forewarned of potential artefacts in trend values derived from the ECG, allowing it to proceed with caution when making decisions based on these trends. In 15 operations spanning 38.5 hours of ECG data, we achieved a false positive rate of 0.71% and a false negative rate of 0.33%. While existing hardware approaches detect the source of the noise without any ability to assess its impact on the measured ECG, our software approach detects only the presence of noise in the signal itself. Furthermore, the software approach is cheaper and easier to implement in a clinical environment than existing hardware approaches.

Adaptive change point detection for respiratory variables

Current alarm strategies for physiological monitoring depend on predetermined thresholds without consideration for the heterogeneity between patients or intraoperative variations. To improve upon this situation, we developed an adaptive change point detection scheme to automatically notify the clinician when a change of clinical significance has occurred in the respiratory variables. We modeled End-Tidal Carbon Dioxide, Expiratory Minute Volume, and Respiratory Rate using a dynamic linear growth model, whose noise covariances are estimated by an adaptive Kalman filter based on a recursive Expectation-Maximization method. Change points are detected by the CUSUM testing. The comparison of the results with post-hoc expert annotations demonstrates that the algorithm can accurately detect relevant changes in the respiratory signals.

Adaptive change detection in heart rate trend monitoring in anesthetized children

The proposed algorithm is designed to detect changes in the heart rate trend signal which fits the dynamic linear model description. Based on this model, the interpatient and intraoperative variations are handled by estimating the noise covariances via an adaptive Kalman filter. An exponentially weighted moving average predictor switches between two different forgetting coefficients to allow the historical data to have a varying influence in prediction. The cumulative sum testing of the residuals identifies the change points online. The algorithm was tested on a substantial volume of real clinical data. Comparison of the proposed algorithm with Trigg's approach revealed that the algorithm performs more favorably with a shorter delay. The receiver operating characteristic curve analysis indicates that the algorithm outperformed the change detection by clinicians in real time.

Variation in Blood Pressure as a Guide to Volume Loading in Children Following Cardiopulmonary Bypass

OBJECTIVE

Intravascular volume loading is used to optimize cardiac output in children following weaning from cardiopulmonary bypass. Central venous pressure (CVP) is frequently used to titrate fluid administration but it is often misleading in predicting fluid responsiveness. Variation in the arterial pressure waveform is exaggerated in patients with deficient intravascular volume and has been shown to be a good predictor of fluid responsiveness in adults following cardiac surgery. The aim of this study was to compare the measures of variation in blood pressure as a guide to volume loading in children following cardiopulmonary bypass.

METHODS

After ethical approval, we collected continuous real-time measurements from 25 children during volume loading after cardiopulmonary bypass. Subjects with moderate or severe tricuspid incompetence or who did not require volume loading during weaning from cardiopulmonary bypass were excluded from the study. Unstable readings were excluded from analysis. Systolic Pressure Variation (SPV), Pulse Pressure Variation (PPV) and Systolic Volume Variation (SVV) were retrospectively calculated before and after each bolus of fluid. Fluid responsiveness was classified as a change in blood pressure of > or =80 mmHg/L/m(2).

RESULTS

Forty-four boluses were analyzed from the 25 children. Respiratory variables were similar. CVP was a poor predictor of fluid responsiveness and a negative relationship between change in blood pressure and Delta Down was observed. Performance in predicting fluid responsiveness as measured by the areas under the ROC curves were CVP (0.58), PPV (0.67), SPV (0.74) and SVV (0.74).

CONCLUSIONS

Variation in blood pressure was a better guide to volume loading in children than CVP. Delta down was not useful in predicting fluid responsiveness in children with open chests following bypass surgery. SPV and SVV require further testing in prospective clinical trials.

[Intestinal obstruction in pregnancy]

Intestinal obstruction is a rare but dreadful complication of pregnancy. Both the mother and the foetus may be severely affected and even die. The authors here report their recent experience and review the literature. They emphasize that diagnostic pitfalls are common during pregnancy and there appropriate management most often delayed. A multidisciplinary approach is advocated and the specific aspects of this high-risk situation are discussed.

An adaptive Cusum test based on a hidden semi-Markov model for change detection in non-invasive mean blood pressure trend

A hidden semi-Markov model is proposed to describe a trend signal for non-invasive mean blood pressure. Based on the model, a Viterbi beam search algorithm working with an adaptive cumulative sum test is proposed to detect change points and recognize change patterns online. Testing results on the simulated signals and clinical signals demonstrate that the algorithm has improved performance over the standard cumulative sum test for change detection, and has the potential to provide a clinically relevant description of trend changes.

Estimating arterial stiffness using transmission line model

An arterial stiffness index based on the transmission line model is defined. The parameters of the transmission line model are initially estimated using measured pressure, flow and aortic root diameter. Pressure is measured at the carotid using applanation tonometry. Flow is measured using Doppler at the ascending aorta. Aortic root diameter is measured using 2-D echocardiography. The initial estimates are then refined using grey-box identification. The resulting estimate of the distributive compliance of the transmission line model is proposed to be an arterial stiffness index. Similar to the Windkessel compliance, this index describes the global stiffness. However, it is based on a more realistic 1-D model that can simulate wave propagation and wave reflection.

An Evaluation of a Vibro-Tactile Display Prototype for Physiological Monitoring

Visual displays and auditory alarms are used to convey information on physiological variables in an operating room. However, the exponential growth in the number of physiological variables and the high probability of false alarms has amplified demands on the clinician's attention. We have extended existing tactile technology to improve situational awareness and produce a practical clinical advisory device. A vibro-tactile display, using two vibrating motors applied to the volar surface of the forearm, was compared to an auditory alarm in a simulated clinical environment. Compared with auditory alarms, the vibro-tactile alarm was as easy to learn and had a better identification rate when used alone or combined with the auditory alarm. Most users preferred the vibro-tactile alarm although the prototype caused some discomfort. Furthermore, a combined vibro-tactile and auditory alarm had reduced accuracy when compared with the vibro-tactile alarm alone. The vibro-tactile modality shows considerable promise for clinical practice but will require further clinical testing and refinement, especially with regard to user comfort.

Evaluation of GHb assays in France by national quality control surveys

OBJECTIVE

To evaluate the state of the art concerning GHb assays through analysis of a large-scale quality control survey and to compare the results with those of previous surveys.

RESEARCH DESIGN AND METHODS

A lyophilized hemolyzate prepared from human erythrocytes containing a physiological HbA1c level (5.5%) was sent to 3,500 French laboratories in February of 1995 and assayed as a patient's sample under routine conditions. Distribution of values was analyzed from the reported results for each method. The results were compared with the assigned value (acceptable range: +/- 20%) and with the upper value of the reference range currently used.

RESULTS

Results were obtained from 2,674 laboratories, among which 39% used cation-exchange chromatography methods, 37.5% affinity chromatography, 16% immunological methods, and 7.5% electrophoresis. The number of laboratories using immunological methods increased from 100 to 400 between 1993 and 1995. The overall interlaboratory coefficient of variation (CV) was 20.2%, with within-method CVs ranging between 3.2 and 29.5%. Method-to-method accuracy varied dramatically, with mean HbA1c values ranging from 4.4 to 8.2%. Results from 75% of the laboratories were comprised in the acceptable range; 88% of them reported a value within the normal range of the method used.

CONCLUSIONS

The interlaboratory variability of results illustrates the difficulties encountered by diabetologists in the follow-up of diabetic patients using results obtained from different laboratories. It demonstrates the usefulness of the internationally developed standardization process of GHb measurements and points out the need for laboratories to fulfill good practices.

Cast nephropathy in mu heavy chain disease

The occurrence of kidney diseases was very rarely reported in heavy chain diseases (HCD). At variance with gamma and alpha HCD in which there is no free light chain secretion, about two-thirds of mu HCD patients have urinary Bence Jones (BJ) proteins. We report on a 66 year-old man affected with typical mu HCD who developed renal failure after a 3-year follow-up. He had presented with chronic lymphocytic leukemia with bone marrow vacuolated plasma cells, serum mu HCD protein and serum and urine BJ protein. After an apparent hematological remission following fludarabine therapy, anemia and blood hyperlymphocytosis recurred together with microscopic hematuria, proteinuria and increased creatininemia. Kidney biopsy showed numerous tubular eosinophilic casts which stained for kappa chain determinants by immunofluorescence and an interstitial infiltration by lymphocytes and plasma cells. The hematological and renal condition improved after reinitiation of chemotherapy. This appears to be the first documented report of a light chain-dependent visceral complication in HCD.

High prevalence and usual persistence of serum monoclonal immunoglobulins evidenced by sensitive methods in renal transplant recipients

BACKGROUND

The occurrence of serum monoclonal immunoglobulins in kidney transplant recipients is well known but their significance and predictive value for the occurrence of lymphoma are a matter of debate. We therefore conducted a study of monoclonal immunoglobulins by a sensitive method during the long-term follow up of grafted patients.

METHODS

Monoclonal immunoglobulins were characterized by high-resolution electrophoresis, conventional immunoelectrophoretic analysis, and a sensitive Western blotting procedure in the serum from 84 renal transplant recipients prior to grafting and subsequently, with a 1-8-year follow-up and excluding the patients who developed posttransplant lymphoma.

RESULTS

Low abundance monoclonal immunoglobulins were detectable prior to transplantation in 56 cases (66.6%) and after graft in 72 cases (85.5%) (and in 1 case (1.2%) and 18 cases (21.4%) of cases respectively, by immunoelectrophoresis). These abnormalities were often multiple in individual sera. Monoclonal components detected by immunoblotting were transient in 23.8% of patients only (whereas those evidenced by immunoelectrophoresis usually became undetectable by this method) and their pattern was remarkably stable in the majority of cases. The frequency of post-transplant monoclonal immunoglobulins was higher in patients of more than 50 years of age than in younger patients. The appearance of monoclonal components after grafting and their transient character correlated with CMV infections. No correlation was found with various other parameters. The isotypic distribution of monoclonal immunoglobulins with an IgM, IgG3, and IgG1 predominance and an abnormally low kappa/lambda ratio was the same as that observed in various immunodeficiency states. The monoclonal immunoglobulin pattern in three further patients who developed post-transplant lymphoma was unremarkable.

CONCLUSION

Monoclonal immunoglobulins hence are not discriminant for lymphoma and their characterization does not appear to be necessary in the evaluation of followed up grafted patients, at least for a prediction of post-transplant lymphoma.