Automated real-time data acquisition and analysis of cardiorespiratory function

Adequate interval for the monitoring of vital signs during endotracheal intubation

Background

In the perioperative period, it may be inappropriate to monitor vital signs during endotracheal intubation using the same interval as during a hemodynamically stable period. The aim of the present study was to determine whether it is appropriate to use the same intervals used during the endotracheal intubation and stable periods to monitor vital signs of patients under general anesthesia.

Methods

The mean arterial pressure (MAP) and heart rate (HR) were continuously measured during endotracheal intubation (15 min after intubation) and hemodynamically stable (15 min before skin incision) periods in 24 general anesthesia patients. Data was considered “unrecognized” when continuously measured values were 30% more or less than the monitored value measured at 5- or 2.5-min intervals. The incidence of unrecognized data during endotracheal intubation was compared to that during the hemodynamically stable period.

Result

There were significantly more unrecognized MAP data measured at 5-min intervals during endotracheal intubation than during the hemodynamically stable period (p value <0.05). However, there was no difference in the incidence of unrecognized MAP data at 2.5 min intervals or HR data at 5 or 2.5 min intervals between during the endotracheal intubation and hemodynamically stable periods.

Conclusion

A 5-min interval throughout the operation period was not appropriate for monitoring vital signs. Therefore, , a 2.5-min interval is recommended for monitoring the MAP during endotracheal intubation.

Comparison of four methods of automated recording of physiologic data at one minute intervals

INTRODUCTION

This study aims to evaluate possible differences between the maximum, minimum, and mean value over each epoch, or the instantaneous value at the end of the interval, when recording physiological data at 1-minute intervals.

METHODS

The mean arterial pressure (MAP), heart rate derived from the radial pulse (PULSE), end-tidal CO2 (etCO2), and the arterial O2 saturation by pulse oximeter (SpO2) were measured continuously in 10 patients during general anesthesia. Anesthesia was induced and maintained using the same technique for the first 10 minutes. At 1-minute intervals for each parameter, the maximum, minimum, and mean value (over the preceding minute), as well as the instantaneous value at the end of the recording epoch, were automatically downloaded to an electronic spreadsheet. The differences between the maximum and minimum, the maximum and mean, the mean and minimum, and the instantaneous and mean values were calculated for each patient.

RESULTS

We found differences between the maximum and minimum values recorded over each interval, between the maximum and mean values, between the mean and minimum values, and between the mean and instantaneous values recorded at the end of each minute interval for all physiological parameters examined.

CONCLUSIONS

From the data presented, we conclude that during the first 10 minutes of a general anesthetic, including anesthesia induction and endotracheal intubation, the values of physiologic parameters derived from different summary indices, namely the maximum, minimum, and mean values over the 1-minute interval or instantaneous at the end of the interval differ from each other. This source of error must be considered when designing systems for computerized recordkeeping of anesthesia charts and when interpreting the data stored in electronic databases.

Techniques to identify clinical contexts during automated data analysis

The interpretation of automatically collected data to produce intelligent alarms and identify particular conditions is nearly impossible without identifying the specific context in which the data are obtained. Shifts in clinical context occur because of changes in the patient's physiologic state, or due to the passage of time, or due to changes imposed by therapeutic intervention such as surgery. Techniques to identify such changes in clinical context are discussed with particular attention to the application of cluster analysis, discriminant analysis, and statistical predictors. An example of these analyses applied to EEG data is presented, showing an unexpected hysteresis of EEG behavior in response to an hypoxic challenge.