JP2006503617A - System and method for trend analysis in sedation and analgesia systems - Google Patents

Abstract

The present invention includes a sedation and analgesia system that can collect data from a single monitor associated with a single patient parameter to reduce the probability of a false positive alarm response due to data artifacts. The present invention also includes a monitoring system that can detect an impending bad patient condition and can detect such a condition before the bad patient condition actually develops. The present invention further includes a method of incorporating trend analysis into a sedation and analgesia system.

Description

[Background of the invention]
The present invention relates generally to trend analysis, and more particularly to trend analysis built into the monitoring, processing and output functions of sedation and analgesia systems.

[Cross-reference of related applications]
This application is based on 35 U.S.C. 119 (e) (35 USC §119 (e)) and is filed on Oct. 3, 2002, which is incorporated herein by reference. Claims priority to US Provisional Patent Application No. 60 / 415,524 entitled “Trend Analysis in a Sedation and Analgesia System”.

[Description of research and development funded by the federal government]
Not applicable

[Reference to “Microfiche Appendix”]
Not applicable

[Description of related technology]
Excessive use of sedatives, analgesics and / or amnesia for patients undergoing painful, uncomfortable, or terrible (rising anxiety) medical or surgical procedures, with or without a qualified anesthetist Sedation and analgesia systems have been developed that provide a safe means of accepting in a way that reduces the risk of medication. Significant advances in technology can make sedation and analgesia systems safe for use in hospitals and outpatient settings, for example C.I. R. N. A. A person other than a trained anesthesiologist, such as a trained doctor or other trained technician, can operate. The sedation / analgesic system addresses the needs of practitioners, where safe and effective sedation and analgesia cannot arrange anesthesia personnel for all procedures that can substantially reduce the effects of fear and pain. Has helped to meet. With the advent of sedation and analgesia systems aimed at these purposes, drug delivery systems integrated into patient monitoring systems that reduce the cognitive and manual workload required for anesthesia machine operation for these people Although provided, clinicians still remain at the center of patient management. In accordance with the principle “the clinician knows best”, the clinician is still given the responsibility to make a final decision. This advanced technology makes the sedation / analgesic system lower than general anesthesia even without anesthesia personnel, providing the patient with a cost-effective and easily accessible means of sedation, memory loss and / or analgesia Can be manipulated by drug concentration effect.

  An example of a sedation and analgesia system is described in US patent application Ser. No. 09 / 324,759, filed Jun. 3, 1999 and incorporated herein by reference in its entirety. This sedation / analgesic system can be used to electronically monitor the physiological state of one or more patients while delivering, for example, one or more sedatives, analgesics and / or memory loss agents, and positive airway pressure. Electronically integrates delivery of drugs, decrease or increase in drug delivery, delivery of oxygen, drug changes to eg opioid antagonists, requests for further information from patient monitors, and alarm triggers . In one form, the system of US patent application Ser. No. 09 / 324,759 uses one or more sets of stored data definition parameters reflecting patient and system status, which parameters are Accessed to conservatively manage drug delivery and correlate with safe cost-effective optimized values related to conscious patient vital signs and other physiological conditions.

  Due to spurious monitor data or other factors, the sedation / analgesic system may take a potentially dangerous action, take no action in critical situations, or give an unnecessary alarm. For example, a sedation / analgesic system may monitor a patient's heart rate with an unstable electrocardiograph (ECG). Based on a single monitor, the sedation / analgesic system may signal an alarm indicating, for example, a dangerously low heart rate if the unstable ECG data is actually false. The high frequency of false alarms can be frustrating to the clinician, and can be less focused on true life-threatening conditions.

  In general, monitoring systems built into medical devices monitor a given patient parameter with a dedicated monitor. Then, a safety data set is established for the monitored parameter, and when the monitored data is out of the safe range, an alarm response is started. Such a system may provide high sensitivity when most true patient conditions are detected, but false positive alerts resulting from data artifacts that fall outside the safety data set are also There is a possibility that it is likely to occur. In addition, many patient parameters such as heart rate, in the case of an impending bad patient condition, decrease linearly or monotonically towards the threshold of the safety data set, indicating a bad patient condition. In existing monitoring systems, such a drop is generally not detected until the data is out of the safety data set, but by looking at the patient's heart rate over a given period, a few seconds before the patient parameters actually exit the safety data set. In some cases, it is clear that bad patient events are imminent. By waiting for the data to exceed the established safety data set threshold, the clinician may struggle to catch up in the situation where the patient is already in poor condition.

[Summary of Invention]
The present invention includes a sedation and analgesia system that can collect data from a single monitor associated with a single patient parameter to reduce the probability of a false positive alarm response due to data artifacts. The present invention also includes a monitoring system that can detect an impending bad patient condition and detect such a condition before the bad patient condition actually occurs.

  The present invention also includes a method of incorporating trend analysis into a sedation and analgesia system. In one embodiment, such a method includes providing a patient monitor that monitors a single patient parameter, and monitoring the patient with the monitor, thereby collecting data collected from the monitor of the sedation and analgesia system. Send to controller. The method further includes inputting a trend into the sedation and analgesia system algorithm or analyzing the trend. Finally, the method further includes initiating appropriate treatment based on the trend analysis.

Detailed Description of the Invention
FIG. 1 illustrates one implementation of a sedation and analgesia system 22 according to the present invention having a user interface 12, a software controller 14, a peripheral device 15, a power supply 16, an external communication 10, a pressure delivery unit 11, a patient interface 17 and a drug delivery unit 19. A block diagram of the form is shown, wherein the sedation and analgesia system 22 is operated by the user 13 to provide sedation and / or analgesia to the patient 18. An example of a sedation and analgesia system 22 is disclosed and enabled by US patent application Ser. No. 09 / 324,759 filed Jun. 3, 1999 and incorporated herein by reference in its entirety. Embodiments of the user interface 12 are disclosed and enabled by US patent application Ser. No. 10 / 285,689 filed Nov. 1, 2002, which is incorporated herein by reference in its entirety.

  The patient interface 17 includes, but is not limited to, a noninvasive blood pressure monitor, pulse oximeter, capnometer, ECG, patient awareness evaluation system, ventilation flow monitor, ventilation pressure monitor, impedance plethysmograph (IPG), gas analyzer, ventilation temperature. Includes one or more patient health monitors such as vital signs monitors and consciousness monitors, including monitors, ventilation and humidity monitors and auditory monitors. A patient monitor at the patient interface 17 may be electronically coupled to the controller 14 to provide signals representative of the actual physiological state of the patient. In one embodiment of the present invention, the at least one monitor monitors the first patient parameter over a predetermined period of time, where the trend of the patient parameter is analyzed to determine whether an unfavorable patient condition is imminent. A determination is made and / or it is determined whether the data may be due to artifacts or may represent a true patient condition. Monitored parameters may include, for example, heart rate, carbon dioxide level, oxygen saturation, and blood pressure.

  Before a patient's monitored parameter falls outside a predetermined safety data set (eg, the heart rate drops to a value deemed to be too low), the parameter generally exceeds the safety data set threshold eventually. There is a slow change, i.e. a period of trend (e.g. heart rate slowly decreases). Without trend analysis, the monitoring system will generally alert the clinician only when the parameter is outside its safety data set, so that the attendant can correct a potentially dangerous situation. In many cases, you will be slain. The trend analysis provided by the system 22 can provide early alerts to the clinician for potentially dangerous situations that may be present. In addition, when patient parameters fall outside the safety data set due to data artifacts, the trend analysis of the present invention allows the sedation and analgesia system 22 to recognize the artifacts because there is no prior information indicating an impending bad patient condition. There is a possibility that you can. The controller 14 may compare the electronic feedback from the patient interface 17 with data stored in the memory device over a predetermined period of time, in which case the data may be evaluated as a trend of information rather than point by point. Good.

  The controller 14 may be programmed to control an effector (not shown) in response to the results of trend analysis and / or stored data comparison. The actuator may be any suitable control function that can ensure patient safety and clinician awareness. Actuators include, but are not limited to, drug reduction, drug increase, positive airway pressure change, alarm, advance alarm, oxygen delivery, trigger for further data sampling from the monitor, eg change of drug to carbon dioxide and opioid antagonist And patient response queries. The actuator may begin silently without alerting the attending clinician, may be signaled by the user interface 12, and / or may require confirmation from the user before it is initiated. Good.

  FIG. 2 illustrates one embodiment of a display 30 that shows a heart rate trend 33 from the patient being monitored. In the illustrated display, a safe heart rate range for a given patient and clinical situation may be considered 90-110 bpm. A trend 33 may be established based on the heart rate y-axis 31 and the time x-axis 32. The display 30 further shows a period 34 during which the patient's heart rate is in the safety data set (95 bpm), alerting the clinician and / or taking no action to make the patient safe. Following the period 34, the display 30 shows a period 35 of the trend 33. Time period 35 falls outside a predetermined safe range of acceptable heart rate given an established safety data set. However, if there is no evidence to support the existence of a bad patient condition, only a single data point is below the 90 bmp threshold, and a slope towards a threshold that matches most truly serious bad patient conditions If there is no decrease, period 35 may be the result of a data artifact. Alerting the clinician during the illustrated example can result in false alarms, which can reduce the specificity of the monitoring system and frustrate the attending clinician. With false detection alerts consistently initiated, clinicians may end up paying less attention to alerts that may ultimately indicate a really bad patient condition.

  FIG. 3 shows a further example of a display 40 showing a heart rate trend 43 from the patient being monitored. In the illustrated display, a safe heart rate range for a given patient and clinical situation may be considered, for example, 90-110 bpm. A trend 43 may be established based on the heart rate y-axis 41 and the time x-axis 42. The display 40 further shows a period 44 during which the patient's heart rate is in the safety data set (100 bpm) and no action is taken to alert the clinician and / or make the patient safe. Following the period 44, the display 40 shows a period 45 of the trend 43. Period 45 is the downward slope of trend 43, and period 45 eventually exceeds the established 90 bpm safety threshold. In the illustrated example, an existing monitoring system may only have alerted when period 45 has finally exceeded the 90 bpm threshold, and such impending events can be detected earlier by trend morphology studies. There may have been. As described further herein, the present invention may be able to detect such imminent patient conditions before they are completely aggravated by detecting tilt variation in trend analysis, Thereby, patient safety can be ensured in the valuable time of the attending clinician. Furthermore, based on the slope of period 45, such data indicative of a debilitating patient condition is unlikely to come from spurious data. The present invention incorporates trend analysis into the sedation and analgesia system 22 to reduce the impact of data artifacts while capturing truly critical patient events early on by assessing trends in monitored patient parameters. This may improve the specificity of patient monitoring.

  FIG. 4 illustrates one embodiment of a method 100 for incorporating trend analysis into the sedation and analgesia system 22. The method 100 includes a step 101 that includes providing a patient monitor that monitors a given patient parameter. The monitor of step 101 may be, for example, a pulse oximeter that measures heart rate, but any suitable monitor of any suitable patient parameter is in accordance with the present invention. Step 102 includes monitoring the patient with the monitor of step 101, and data collected from the monitor may be transmitted to the controller 14 of the sedation and analgesia system 22 (FIG. 1).

  Step 103 includes creating a trend based on the data received from the monitor of step 101. For example, FIGS. 2 and 3 show trends created by combining data points over a predetermined period of time. Such trends may be established by any suitable means, and such trends may be further displayed for visual analysis by attending clinicians. Following step 103, method 100 may proceed to step 104 and / or step 105.

  Step 104 includes analyzing the trend established by step 103. For example, such a tendency may be analyzed by the following method. (1) If the trend has a linear, quasi-linear, or monotonic nature, calculating the slope of the trend allows the trend to progress inexorably towards the outer limit of the safety data set (2) If the trend does not tend to follow a single (eg, linear) path, it is possible to determine where the trend is by calculating multiple slopes for the trend variation. (3) If the trend is inherently a polynomial, it may determine the direction of the trend by calculating the coefficients of the polynomial, and (4) the least mean square error technique and others Using such an algorithm, trend curve fitting is performed to predict when and when the trend will exceed the safety data set. From such an analysis, the present invention involves monitoring the trend of any suitable patient parameter in a manner that provides the most accurate depiction of the true patient condition.

  According to step 104, based on the above analysis, the method 100 may then evaluate the monitored trend against the established safety data set. For example, referring to FIG. 3, if the slope of trend 45 exceeds a certain rate, such slope indicates an impending bad patient condition, and sedation and analgesia system 22 is described in further detail herein. A pre-alarm or other suitable treatment may be initiated. Step 104 further includes calculating a probability value for whether the data presented to the controller 14 actually indicates a change in slope by means generally known in the art. Based on a comparative analysis, if such data reflects a change in tilt, the sedation / analgesic system 22 then evaluates such data to determine whether the tilt indicates an impending good patient condition. Also good. Further, referring to FIG. 2, such an analysis allows the period 35 as a data artifact to be truncated, thereby reducing the probability that the sedation / analgesic system 22 will initiate a false alarm. Even incorporating trend analysis into a single monitored patient parameter can help improve the specificity of the sedation and analgesia system 22 (by reducing the impact of data artifacts) and poor patient status Helps capture the facts before they become obvious. Such trend analysis may also be applied to multiple parameters for the sedation and analgesia system 22, thereby providing more advantageous results.

  Step 105 includes inputting the trend created in step 103 into any suitable algorithm of the sedation and analgesia system 22. By combining such trends with multiple other trends from relevant patient parameters, the effects of data artifacts can be further reduced and non-deterministic data can be revealed by incorporating such sensor fusion. Further, such a tendency may be incorporated as a function of redundancy having orthogonality. Here, orthogonal redundancy refers to simultaneously monitoring a single patient parameter with multiple monitors. For sensor fusion, a co-pending US patent assigned to the same assignee as the present application entitled “Systems and Methods for Providing Sensor Fusion” filed on Oct. 3, 2003, which is incorporated herein by reference. It is further described in the application. Orthogonal and redundant monitoring is identical to the present application entitled “Methods and Systems for Providing Orthogonally Redundant Monitoring in a Sedation and Analgesia System” filed Oct. 3, 2003, which is incorporated herein by reference. Is further described in a co-pending US patent application assigned to the assignee of the present application. By incorporating trend analysis into such redundancy, the specificity and sensitivity of the sedation / analgesic system 22 can be further improved by reducing the probability of initiating both a missed alarm condition and a false alarm condition. Moreover, you may integrate this tendency into a neural network. Here, a neural network is a computerized intelligence system that can extract complex patterns and reach a correct judgment even when an incomplete or ambiguous picture is presented.

  The present invention includes incorporating trend analysis into the sedation and analgesia system 22, and such integration may allow the controller 14 to analyze data more accurately with respect to patient status. Monitoring a given parameter may be able to reduce the presence of artifacts and anticipate imminent bad patient events, but single parameter trend monitoring, trend monitoring of multiple related parameters and neural By using trends in conjunction with the network, the ability of the sedation and analgesia system 22 to perform treatment based on the true state of the patient's condition is further improved. Treatments performed based on the algorithm associated with the sedation and analgesia system 22 at step 105 include, in addition to the treatment shown at step 106, any other suitable treatment that helps to ensure patient safety.

  Step 106 includes taking appropriate action based on the trend analysis of step 104. For example, if trend analysis determines that data outside the safety data set is due to artifacts (as in FIG. 2), the method 100 may proceed to step 102 and the sedation and analgesia system 22 will not take any other action. May be. Maintaining normal functionality in the presence of data artifacts can reduce the probability of false alarms and allow the sedation and analgesia system 22 to monitor the actual patient condition more directly. Can be. If, based on the trend analysis, it is determined that bad patient symptoms are imminent (as in FIG. 3), step 106 includes initiating a pre-alarm. The pre-alarm may be any suitable action taken to alert the attending clinician that there is a high probability of an impending bad patient condition. Such pre-alarms may be notified visually and / or audibly and, for example, by reducing the drug level, changing the drug to, for example, carbon dioxide and / or opioid antagonists, from the patient monitor Triggering a request to collect a lot of information, delivering oxygen, testing patient responsiveness, and delivering positive airway pressure may be included. Such treatment may be performed if the patient exceeds an established safety threshold, and the alert for such an event may be greater than for a pre-alarm, for example. By taking action earlier in the trend analysis, the sedation and analgesia system 22 can alert the clinician at an earlier stage and can avoid many bad patient conditions in advance. There is sex. For example, if the trend representing heart rate indicates that the heart rate is dropping sharply and not due to artifacts, the sedation and analgesia system 22 falls outside the established safety data set of the patient's heart rate. Drug delivery may be paused before. Such prior treatment can avoid or reduce the severity of a bad patient condition.

  While exemplary embodiments of the present invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided merely as examples. Numerous insubstantial variations, modifications and substitutions will be apparent to those skilled in the art without departing from the scope of the invention disclosed herein by the applicant. Accordingly, it is intended that the invention be limited only by the spirit and scope of the appended claims as permitted.

1 is a block diagram showing an embodiment of a sedation / analgesic system according to the present invention. It is a figure which shows an example of the heart rate tendency display by this invention. It is a figure which shows the further example of the heart rate tendency display by this invention. FIG. 4 illustrates one embodiment of a method for incorporating trend analysis into a sedation and analgesia system according to the present invention.

Claims (14)

At least one patient health monitoring device adapted to be coupled to the patient and to measure patient parameters reflecting the physiological state of the patient;
A user interface;
A drug delivery controller for supplying one or more drugs to the patient;
A memory device storing a safety data set reflecting at least one safe parameter and an undesirable parameter of the monitored patient's physiological condition;
Interconnected with the patient health monitor, the user interface, the drug delivery controller, and the memory device to determine whether an unhealthy patient condition is imminent by receiving the patient parameters and analyzing the patient parameter trends An electronic controller to determine whether
A sedation / analgesic system.   One or more actuators for ensuring patient safety and clinician awareness, wherein the electronic controller is interconnected with the actuators and the actuators according to the trends of the patient parameters The sedation / analgesic system according to claim 1, wherein the sedation and analgesia system is controlled.   The sedation and analgesia system according to claim 1, wherein the electronic controller provides an early warning in response to a bad trend of the patient parameters within the safety data set.   The sedation / analgesic system of claim 1, wherein the electronic controller evaluates the patient parameter measurements against the trend.   The sedation / analgesic system according to claim 1, wherein the patient parameter is derived from capnometry, pulse oximetry and blood pressure.   The sedation / analgesic system according to claim 1, wherein the patient parameter is a heart rate.   The sedation / analgesic system according to claim 1, wherein the at least one patient health monitor is one of an electrocardiograph and a pulse oximeter. A method of incorporating trend analysis into a sedation and analgesia system,
Providing a patient monitor for a patient monitoring single patient parameters;
Monitoring the patient with the monitor, wherein the data collected from the monitor may be sent to a controller of the sedation and analgesia system;
Creating a trend based on data received from the monitor;
At least one of a step of inputting the trend to the algorithm of the sedation / analgesic system and a step of analyzing the trend;
Starting appropriate treatment based on the trend analysis;
To incorporate trend analysis into sedation and analgesia systems.   The analyzing step includes calculating a slope of the trend, calculating a plurality of slopes for trend variation, calculating a coefficient of a polynomial, and curve fitting of the slope using a minimum mean square error technique. 9. A method of incorporating trend analysis into a sedation and analgesia system according to claim 8, comprising at least one of:   9. The method of incorporating trend analysis into a sedation and analgesia system according to claim 8, further comprising calculating a probability value for whether the data presented to the controller actually indicates a change in slope.   9. The method of incorporating trend analysis into a sedation and analgesia system according to claim 8, wherein the trend is analyzed in combination with a plurality of other trends from associated patient parameters.   10. The method of incorporating trend analysis into a sedation and analgesia system according to claim 9, wherein the trend is incorporated as a function of orthogonal redundancy.   9. The method of incorporating trend analysis into a sedation and analgesia system according to claim 8, wherein the trend is integrated into at least one neural network. 9. The sedation according to claim 8, wherein the appropriate treatment includes at least one of continuing monitoring, initiating a pre-alarm, pausing drug delivery, and initiating a full alarm. How to incorporate trend analysis into an analgesic system.

Images