SE1200155A1 - Apparatus for monitoring mechanical ventilation - Google Patents

Abstract

The invention relates to a monitoring device for physiological parameters during mechanical ventilation. The purpose is to identify beneficial and harmful effects of fan changeover within a few breaths while the operator remains at the bedside, monitored parameters include e.g. tidal volume, airway pressure, end-tidal C0, hemodynamics, and the volume of C0 eliminated per minute, VC0. Data sets for each signal before and after conversion are statistically analyzed to reduce noise and allow accurate presentation of changes associated with conversion. The ratio between VC0 after and before adjustment is used to demonstrate the effect on alveolar ventilation and thus effects on arterial partial pressure of C0. The course of VC0 after conversion is analyzed to obtain an accurate value of VC0 after conversion. The invention can be applied to different ventilation methods due to noise reduction strategies, which are particularly important in irregular ventilation.

Description

The object of the present invention is to enable the evaluation of physiological effects due to a fan changeover within a limited number of breaths after the changeover, preferably while the operator performing the changeover is present at the bedside. The physiological effects affected include effects on airway pressure such as mean pressure, peak pressure, plateau pressure after inspiration, external PEEP, internal PEEP and total PEEP. Other physiological effects are end-time OCZ concentration, the volume of CO2 eliminated per minute, VMmCOz and change in the PaCOg of the blood and pH, which in a new steady state is caused by the fan change. Oxygen saturation in blood studied in the periphery, SPOZ, is an important parameter.

Physiological effects are measured with sensors that provide signals, which are analyzed using a computer. The sensors and the computer can be integrated with the fan in one and the same device. An alternative is that the sensors are the same as those within the fan and that a separate computer receives the signals from these sensors and performs the analysis. Signals representing circulation such as arterial pressure are usually recorded with special monitoring equipment. information therefrom can be supplied to the system according to a preferred embodiment of the invention.

According to an alternative embodiment of the invention, both the sensor and the computer are separate from the fan. This embodiment can be used with different types of fans.

Measurement and analysis of the signals begins before the fan changeover and continues through the changeover and for a period after this. within a number of breaths or within a few minutes after the changeover, the computer reports how selected physiological parameters change and allows the operator to assess whether the changes are in line with expectations and match the goals behind the changeover. According to a preferred embodiment of the invention, the signals are analyzed with respect to noise and trends, which are caused by physiological phenomena which reflect that steady state is not immediately established. This demonstrates, for example, effects on VMNCOZ and PaCOz caused by the changeover.

DESCRIPTION OF THE DRAWINGS Figure 1 Figure 1 illustrates an apparatus according to a preferred embodiment of the invention. The apparatus 1 is only shown schematically, as alternative embodiments are almost unlimited with modern technology.

A pneumatic inspiratory system within the fan comprises inlets for gases such as air and oxygen 2, a mixer for the gases 3 and a fate regulator in the inspiration line 4. According to an alternative embodiment of the invention, the mixer 3 and the fate regulator 4 are integrated within one and the same unit. the inspiration line is equipped with a fl fate meter 5. In addition to the fan or integrated inside the fan, the inspiration line is equipped with a humidifier 6 and continues in the form of an fl visible inspiration hose 7 which leads to a Y-piece 8.

The ventilator is connected to the patient 10 by a tracheal tube 9 but can be connected by other means. Expiration takes place through the ventilator's expiratory pneumatic system which begins at the Y-piece 8 and leads on through an exhalable expiratory hose 11, an expiratory valve 12 and an expiratory flow meter 13. The order of 12 and 13 may be the opposite. A CO 2 analyzer 14 measures the fraction of CO 2 at the Y-piece. A pressure sensor 15 measures the airway pressure. Alternatively, it can be connected to the expiration line 11 or doubled within both the inspiration line and the expiration line.

The functions of the fan atom are controlled by an electronic control unit 17, which may be an analogue or digital apparatus. In a preferred embodiment of the invention, the control unit contains at least one computer, which registers and analyzes the signals from the flow pressure and CO 2 sensors 5, 13, 15 and 14. The computer can also receive signals from monitoring systems for circulation such as arterial pressure. The control unit can also communicate with the user through a keyboard, through touch controls or other means.

Remote communication is also possible, e.g. from a central system within an intensive care unit.

All stipulated parts can be integrated within a single device or be functionally distributed between different devices. The latter option may mean that e.g. the function 10 15 20 25 30 35 40 45 50 which serves to control the pneumatic systems is located within the fan, while e.g. calculation and monitoring are physically located in another device such as an external computer.

The control unit receives analogue or digital signals representing speed of discharge, pressure and CO2 and sends signals to the inspiratory and expiratory valves 4 and 12 via electronic communication means 16. In addition to components within the fan itself, the control unit may include components and systems outside the fan. Today's technology offers seemingly limitless opportunities to realize the invention in terms of technical solutions for electronic components and their communication with each other by wired or wireless means. Monitoring of the ventilation process can be achieved with a system built into the fan or with a system outside the fan but which communicates with the control unit inside the fan. According to a preferred embodiment of the invention, the control unit 17 is equipped with a monitor for monitoring flow and pressure signals and other information.

Figure 2 Figure 2 illustrates an alternative preferred embodiment of the invention 1 which 1-17 indicates the same structures as in Figure 1. The system used for monitoring according to the present invention is realized by an apparatus separate from the fan 1. The monitoring apparatus comprises a computer 20 and transducers for CO 2 14 fl fate rate 18 and airway pressure 19, which by wired or wireless means of communication 21 send signals to the computer 20. According to the embodiment illustrated in Figure 2, the invention can be applied together with all kinds of fans.

According to a further embodiment of the invention, the computer 20 can receive signals for one or fl era of the parameters fl fate rate, airway pressure and CO 2 from sensors integrated in the ventilator thereby avoiding duplication of sensor equipment. The computer 20 may also have access to other information from the fan 1 such as the setting of the fan, respiratory rate as well as information on time distribution for parts of the respiratory cycle by digital or analogue, wired or wireless communication path. Likewise, the computer may receive information from other sources as such used to monitor circulation such as arterial pressure.

Figure 3 Figure 3 illustrates a monitor for monitoring in connection with a ventilator adjustment in a patient with ARDS performed at the time of 2 minutes. Top panel A shows respiratory rate, tidal volume, VT, and Positive End Expiratory Pressure, PEEP. VT is obtained by integrating the flow signal over each breath. PEEP is measured by the sensor 15 or 19.

Panel B shows post-inspirational plateau pressure, Ppw, and the difference between PPW and PEEP called delta-P. Ideally, delta-P would represent the true difference between elastic resilience pressure of the respiratory system before and after inspiration. To measure these values accurately, the patient needs to be passive and the flow rate to be zero at these times.

Zero flow can be achieved with modem fans by closing valves 4 and 12.

In some circumstances, this is not possible or desirable. Then PPW can be determined with minimal mot fate towards the end of the inspiration. PEEP can be determined with minimal expiratory mot fate towards the end of the expiration. Such PEEP values can be corrected for flow and resistance dependent airway pressure gradients according to principles described by Jonson et al.

(Bull. Physiopath. Resp. 1975, v. 11, pp. 729-743) Panel C illustrates the volume of CO 2, which is eliminated per minute by ventilation, VM.NCO 2.

VM.NC02 is calculated from the volume of C02, eliminated during each breath divided by the duration of the breath. VM.NC02 is obtained by integrating the product between the flow rate and the fraction of CO2 measured at the airway opening. Panel D is a record of peripheral saturation of oxygen, S202, measured by conventional means. Panel E shows end-tidal C02, ETC02. In the example: At initial fan setting, monitoring showed that VT, PPW and delta-P were higher than usually recommended. PaCO 2 was also high, 62 mmHg. To reach compliance with standard recommendations, RR was increased at 2 minutes from 25 to 33 breaths per minute, VT was reduced from 6.9 to 5.8 ml per kg body weight and PEEP was reduced from 12 to 8 cmHzO. Immediately after the changeover, the pressure dropped to what was the recommended value. The immediate increase in VM.NCO2 was followed by a slow decline towards the initial level.

SpOz and ETCO; fell somewhat with some delay.

Figure 4 Figure 4 Panel A illustrates how different types of disturbances can affect the signals, e.g. the one representing VM | NCO2. Such disturbances will adversely affect the accuracy of the determination of the change of VMWCOZ which occurs after adjustment of the fan, in this example at 2 minutes. Slow operation of VMWCOZ is common. This operation may reflect changes in metabolism due to variations in body temperature or varying inflammatory activity in the body as well as other causes of varying homeostasis. In the example in Figure 4, there is a trend towards increasing vMmCOz. Another type of disturbance is caused by variation from breath to breath of the volume of eliminated CO2. This variation has a number of causes such as measurement errors, variation of the tidal volume from breath to breath and in cardiac arrhythmia varying lung perfusion.

The slow operation in the example in Figure 4 was characterized for 2 minutes before the changeover. The thick straight line in Panel A shows this trend extrapolated to 11 minutes.

After setting the trend line value at 2 minutes to zero, the trend was subtracted from the signal shown in Panel B. After adjusting the fan, determining the immediate change of VMWCOZ, AVWNCOZ, is an important aspect of the gain. The impact of respiratory disturbances on AVWNCOZ is therefore important to minimize. This must be done without applying conventional filtering of the signal over the conversion period.

Before a changeover, noise is reduced breath by breath in the VMWCO signal; by statistical analysis of values before the changeover as shown by the solid line during the first 2 minutes. The value for VMWCO, before conversion, is called VMNCOZbaSeIine. Upon conversion, VM | NCO2 will immediately increase or decrease to a new value vMmCOzreset.

The change will be inversely proportional to the change in effective alveolar ventilation. During the following period of 15-30 minutes, VM | NCO2 will slowly, according to an exponential course, return to the value representing the current metabolic production of C02. To reduce noise from breath to breath in VMWCOZ after adjustment, values on VWNCO are used, after adjustment to analyze this process as described below and in Figure 4 is shown by a thick line representing the case against the baseline. AVM | NCO2 is determined from the difference between vMmCOzreset and vMmCOzbaseline as shown by the broken vertical line in Figure 4.

Figure 5 Figure 5 shows a numerical display of the same parameters depicted in Figure 3 based on values for 2 minutes before conversion and 15 s to 2 minutes after conversion. According to a preferred embodiment of the invention, a change is accompanied by information about its statistical significance. In this example, significant changes (p <0.05) are marked in bold. SE is the standard error of the estimate. 10 15 20 25 30 35 40 45 50 DESCRIPTION OF PREFERRED EMBODIMENTS The system is based on sensors for airway flow rate, airway pressure and CO 2 as illustrated in Figures 1 and 2. Flow rate and airway pressure can be measured within the ventilator Figures 1, 13 and 15 at the patient's airway opening 18, 19 in Figure 2. According to preferred embodiments, CO 2 is measured at the airway opening 14. According to an alternative embodiment of the invention, CO2 is measured somewhere along the expiratory pneumatic system. In such cases, the small amount of CO 2 from the Y-piece 8 and the inspiration and expiratory lines 7 and 11 will not be measured. This limitation can be weighed against the technical advantages of this alternative embodiment. For each embodiment of the invention, the signals for fate rate, airway pressure and CO 2 shall have adequate frequency response and be adequately synchronized with each other so that events during respiration relating to each signal or combination of signals can be properly recorded and monitored. Selected sensors for SPO 2, arterial pressure or other signals are contemplated to be incorporated into alternative embodiments of the invention.

A computer that can be integrated in the fan 17 or consist of a separate computer 20 samples signals for CO 2, airway pressure and med fate with adequate frequency. These signals together with data calculated from the signals and other information can be displayed and stored by the computer in accordance with conventional monitoring systems. Thus, volumes are calculated by integrating flow rate over time. Respiratory rate per breath per minute can be derived from signals controlling the ventilator valves 4, 12 or through the computer 17, 20 analysis of pressure and fate signals. The purpose of this invention is to monitor parameters that provide the most important information regarding goal fulfillment of mechanical ventilation as well as risks associated with it and in particular how the adjustment of the fan affects these parameters. Figure 3 shows an example of combinations of such parameters. The combination of parameters can vary e.g. against the background and the patient's disease nature as well as the availability of selected donor signals. The registered signals are accompanied by numerical information. An example is given in Figure 5. Figure 3 shows respiratory rate, PEEP and VT, all of which in controlled ventilation are parameters that directly relate to the ventilator setting. Panel B shows the end inspiratory plateau pressure, Ppm, which is the pressure at zero fl fate or very low flow rate at the end of the inspiration. Ppm is the commonly used parameter indicating the degree of lung distension, which causes lung damage related to over-stretching when it is too high. delta-P, ie the difference between PPLAT and PEEP is a parameter that is recommended to estimate the risk of VILI. Panel C shows C02 eliminated per minute, VMWCO2. According to a preferred embodiment of the invention, this value is calculated breath by breath by integrating the product of flow rate and CO 2 measured at the airway opening with sensors 5, 13, 14 and 18. Panels D and E show records of peripheral oxygen saturation and end-tidal CO 2 concentration.

The primary goal of ventilation, mechanical or spontaneous, is the exchange of O 2 and CO 2 so that the arterial blood acquires suitable properties with respect to these gases. With regard to oxygen, the focus is often on oxygen saturation of hemoglobin, which can easily be measured peripherally as Sp02. For CO 2, the arterial partial pressure of CO 2, PaCO 2, is the parameter of primary interest. PaC02 cannot be estimated correctly without analysis of arterial blood tests. Apart from blood sampling and costs, there is the problem that a change in PaCO 2 occurs slowly after a change in the alveolar ventilation caused by a change of the ventilator. This is due to large layers of C02 in body fluids. It may take more than 30 minutes for a new steady state to be reached after altered alveolar ventilation. An alternative to PaCO 2 is to measure the final partial pressure of CO2, ETC02. ETC02 can easily be obtained from the donor 14. In lung disease, however, it is often the case that ETC02 is very different from PaC02. Furthermore, a change of ETC02 occurs after a delay similar to that of PaCO2 and is affected by many physiological effects, which change over time and by adjustment of the ventilator. There are no methods for directly measuring or accurately estimating PaCO 2 bedside in connection with fan switching. However, a change in PaC02 in connection with a fan changeover, APaC02, can be estimated soon after the changeover as explained below.

The mean value of VM | NC02 over such a long period of time that the body's stocks of CO2 can be considered as constant represents the rate of production of CO2, which is proportional to aerobic metabolism. A change in VMmC02 observed after Ventilator conversion over such a short time that the metabolism and body stores of C02 can be considered constant, reflects a change in the effective alveolar ventilation, which is equal to total ventilation minus ventilation of alveolar deadspace. Therefore, determining the change in VM | NC02 that occurs immediately after conversion, AVM | NC02, is of particular interest. Under stable conditions such as controlled ventilation of a sedated patient, AVM.NC02 can often be roughly estimated by observing the recorded signal Figure 3, Panel C.

The signal representing VM | NC02 is very often affected by noise from different sources. Then AVWNC02 cannot be correctly estimated by visual analysis of the VM.NC02 signal.

Figure 4 with accompanying text shows how the influence of noise within measured AVM.NCO 2 according to a preferred embodiment of the invention can be reduced to a level which in most circumstances is sufficiently low. VM | NC02baseline is calculated by statistical analysis of values during a period preceding the fan conversion. This analysis not only minimizes variation in breath to breath but also allows determination of slow operation. Values after the fan changeover are corrected for slow operation determined before the changeover. According to preferred embodiments of the invention, VM.NCO 2 reset is obtained by statistical analysis of values after the conversion. This analysis serves to characterize the slow return to a steady state, which corresponds to the metabolic production of C02. According to a preferred embodiment of the invention, an exponential return is expected, but similar results can be obtained with alternative mathematical expressions.

Fan change does not always occur at a specific moment, e.g. when more than one parameter changes as in the example behind Figure 3. Furthermore, some fans are such that fan changeover is not immediately realized at the moment of the changeover but is carried out during a few breaths. Thus, a few breaths must pass before the process that they fi nm VM | NC02 after a conversion is characterized. 5 breaths or 15 s after the last adjustment is according to a preferred embodiment of the invention default for postponing the analysis. The delay can be increased by the operator, but it is proposed to a maximum of 30 s after the last changeover. With greater delay, the precision may decrease due to altered layers of CO 2 in the body fluid coma and when one loses such data, which can reduce the input of noise. When the course of VM | NC02 after conversion has been characterized, the equation is extrapolated back to the time of the conversion to obtain VM.NC02reset representing the time of the conversion and to calculate AVWNC02 as the difference between VMmC02reset and VM.NC02base | ine. AVM.NC02 is the change in the C02 elimination in connection with the fan conversion.

The accuracy of AVMmC02 is reduced by noise in the VM.NC02 signal. According to a preferred embodiment of the invention, the noise level is analyzed statistically. Thereby, the influence of noise on the determination of AVM | NC02 can be estimated by applying standard statistical methods.

The ratio AVM.NC02reseWM, NC02baseline shows the extent to which the alveolar ventilation changed in connection with the fan conversion and can be expressed as a percentage, AVM | NC02%. In the example in Figure 4, data sets for VM, NC02 were analyzed 2 minutes before and 15 s to 2 minutes after the changeover. Two minutes may be an appropriate default value for these periods.

In the event of a pronounced noise in the registration of VM | NC02, longer periods can be applied automatically or manually to increase the precision in the calculation of AVM | NC02. With controlled ventilation, the respiratory rate is usually constant. Then values for eliminated volume C02 per breath can be used to calculate the change in CO2 elimination in connection with fan change instead of values for VM | NC02.

PaC02 is proportional to metabolic C02 production and indirectly proportional to efficient alveolar ventilation. Consequently, the change of PaCO 2, which after equilibration of the body's stores of CO 2 follows a conversion, according to a preferred embodiment of the invention will be estimated from AVM | NCO2%. In the example in Figure 4, AVM.NCO2% was estimated at + 7% with an estimated range between +6 and + 8%. The value for this range was measured using standard statistical methods such as 95% confidence intervals. This interval serves by default according to a preferred embodiment of the invention. Thus, the expected relative change of PaCOz after equilibration was -7%, estimate range -8 to -6%. An alternative to the estimation area is to report the standard error of the expected change of PaCOz as shown in Figure 5. If information of the PaCOz value before the conversion is available within the system, the estimated value of PaCOg after the conversion according to a preferred embodiment of the invention will be presented to the operator. . If the computer 17 or 20 also has access to the current pH value and acid base status e.g. expressed in terms of Base excess, the computer can also calculate the expected change in pH by applying traditional well-known equations.

The following serves as an example: Current PaCOz 8.3 kPa. Estimated PaCOg after conversion 7.7 kPa (confidence interval 7.6-7.8).

Current pH 7.28. Estimated pH after conversion 7.32.

It is worth noting that few studies have been published on the relationships between observed change in VM | NCO2 and change in PaOz in different patient groups. Reactions in the body in the form of secondary feedback can in certain circumstances lead to complementary mechanisms, which can to a marginal extent disrupt the calculation of PaCOz after fan conversion. It can therefore be expected that different algorithms for predicting a change in PaCO 2 and pH based on AVM.NCO2% will be developed for different patient groups. Figures 3 and 4 represent each data point as a single breath. With controlled ventilation, this is a preferred embodiment of the invention. In case of very irregular breathing, VWNCOZ and other parameters such as tidal volume can vary greatly between breaths. An alternative embodiment of the invention is then to measure parameters over longer periods of time rather than per breath.

Determination of volumes such as of CO2 based on gas fl flow rate and fraction of CO 2 is affected by the circumstances at which flow rate and CO 2 are measured. Circumstances may change before and after fan change, especially with respect to pressure. According to a preferred embodiment of the invention, correction is made to standardized circumstances, e.g. BTPS (body temperature, atmospheric pressure and saturated with water vapor) or STPD (standard temperature and pressure, saturated). The standard chosen does not matter with respect to the present invention. Correction to a certain standard takes place according to well-known physics by using information from pressure sensors 15,19. The signal from the C02 sensor can be affected somewhat by the oxygen content in the breathing air. When the oxygen content is changed in connection with a fan changeover, the accuracy of the AVWNCOZ determination according to a preferred embodiment of the invention is increased by correcting the CO 2 signal for variation of the oxygen content. A signal corresponding to the oxygen content of the breathing air is available in existing advanced fans.

Analysis of physiological parameters other than VMWCO2 before and after conversion is done by statistical analysis of data sets for each signal sampled before and after conversion. For most parameters, steady state is expected within a few breaths after adjustment. Thus, the level after adjustment is determined for a certain parameter based on data from a period beginning 5 breaths or 15 seconds after the last adjustment and ending 2 minutes thereafter, as for the analysis of vMmCO 3. For this period, the mean, range or standard error of the parameter in question is reported. For parameters that reach a steady state after a few breaths, analysis of trends before and after adjustment is not necessary, in contrast to AvMjNCOz. When the fan change does not take place at an exact specified moment, a change caused by the fan change is called “a change in connection with the fan change”.

End-tidal C02, ETCOZ, behaves in its own way after fan conversion. As a first approximation, it will fall at a rate and extent similar to that of PaCOz, which follows an exponential path. Nevertheless, ETCOZ is affected by more physiological phenomena than PaCOz. For example, as the respiratory rate increases and the tidal volume decreases, in most circumstances the difference between ETCOZ and PaCO 2 will increase. ETCOZ is also affected by cardiac output and intrapulmonary shunt fraction, which are often affected by a fan switch. Since a change in ETCOZ is complicatedly affected by both slow and fast phenomena, it is usually not useful to predict its future steady state value after a fan change, but rather to track its variation over time as in Figure 3. Despite the complexity of changes of ETCOZ, it is of value to monitor this parameter. ETCOZ may fall suddenly upon rapid suppression of lung perfusion, which may occur after an inappropriate ventilator adjustment. A sudden fall of both EfCOz and VM | NCO2 warns of suppressed circulation.

The examples given in Figure 3 and Figure 4 refer to controlled ventilation of a sedated patient.

In such circumstances, the signal noise is usually low. The signal for respiratory rate and tidal volume is apparently free of noise. The invention can be applied to other ventilation methods, e.g. different types of assisted ventilation. In such cases, the variability of the parameters, leading to noise in the observations, is often of much greater importance. The observation periods over which calculations are performed should then be adapted to the noise level. When the noise becomes so high that values of AVM.NCO2% cannot be determined statistically, according to a preferred embodiment of the invention, the prediction of a change in PaCOZ is suppressed.

In addition to the Greek presentation as in Figure 3, data are presented before and after fan conversion in numerical format. Figure 5 shows an example based on the data in Figure 3.

According to a preferred embodiment of the invention, the parameters shown graphically or numerically can be selected in a setup procedure of systems such as in Figures 1 and 2. This option includes parameters not shown in Figure 3. Examples are airway mean pressure and total PEEP.

Total PEEP is the pressure in the alveoli at the end of an expiration, which can be measured during an end-expiratory pause or estimated according to the principles described by Jonson et al. (Bull. Physiopath.

Resp. 1975, v 11, pp 729-743). According to a preferred embodiment of the invention, signals from hemodynamic monitoring systems and sensors for SpO 2 are transmitted to the computer 17, 20.

Such parameters are monitored and analyzed in accordance with parameters from sensors within the system.

According to a preferred embodiment of the invention, both Greek and numerically presented information can be saved and recalled for treatment documentation and for research.

Claims (1)

1. 0 15 20 25 30 35 40 Requirement 10. A monitoring device for mechanical ventilation, comprising sensors for measuring airway fl fate (5, 13, 18) and CO 2 (14) and a computer (17, 20), which monitors physiological parameters produced from the sensors characterized in that the change of CO 2 elimination per unit time in connection with fan change is calculated on the basis of measured parameters and monitored. A mechanical ventilation monitoring apparatus comprising sensors for measuring airway fl fate (5, 13, 18) pressure (15, 19) and CO 2 (14) and a computer (17, 20), which monitors physiological parameters produced by the sensors characterized by changes in a number of parameters in connection with fan adjustment are calculated on the basis of measured parameters and monitored. A monitoring device according to claims 1 and 2, characterized in that the change of CO 2 elimination per unit time is one of the plurality of monitored parameters in connection with fan changeover. A monitoring device according to claims 1 and 3, characterized in that the change of CO2 elimination per unit time in connection with fan conversion is calculated by statistical analysis of different data amounts of CO2 elimination observed before and after conversion. A monitoring device according to claim 4, characterized in that the statistical analysis of the change of CO 2 elimination per unit time in connection with fan conversion is based on calculation of a slow change trend of CO 2 elimination per unit time, which is characterized by data before fan change. A monitoring device according to claims 1 and 2, characterized in that the statistical analysis of the change of CO 2 elimination per unit time in connection with fan change is based on an equation that describes return to values, which represent steady state. A monitoring apparatus according to any one of claims 1 to 6 characterized in that the relative change of CO 2 elimination per unit time in connection with fan conversion is used to calculate and monitor the effects of fan conversion on the arterial partial pressure of CO 2. A monitoring device according to either of claims 1 to 7, characterized in that the relative change of CO 2 elimination per unit time in connection with fan switching is used to calculate and present the effects of fan switching on pH in blood. A monitoring device according to claim 2, characterized in that changes in parameters from supplementary monitoring systems are monitored in analogy with parameters from built-in sensors in connection with fan changeover. A monitoring device according to claims 2 and 9, characterized in that changes in a plurality of parameters in connection with fan changeover are monitored on the basis of statistical analysis of data sets before and after fan changeover.