Hello, and welcome to the session entitled, Can't I Trust My Monitors? Problems with Pulse Oximetry, Arterial Line, and Pulmonary Artery Catheter Monitoring. My name is Talia Ben-Jacob. I'm the Division Chief of Critical Care Medicine in the Department of Anesthesiology at Cooper University Hospital, and I'm also an Associate Professor of Anesthesiology at Cooper Medical School of Rowan University. Today I will be discussing, is my arterial line accurate? This slide demonstrates the outline of what we will be discussing today. So why does a patient need an arterial line? Arterial lines have actually been determined to be the gold standard of arterial pressure management in high-risk, critically ill patients. This is because while they are semi-invasive, as they do require catheter insertion into an artery, they still are low risk, and they have been determined to be more accurate than non-invasive blood pressure management with an oscillometric cuff. Arterial lines provide continuous blood pressure management. They detect nearly twice as many episodes of hypotension compared to the blood pressure cuff, and oftentimes the arterial waveform allows for pulse contour analysis that can be used to estimate cardiac output, as well as other cardiac variables. Arterial lines are better at detecting labile blood pressure. They can often help a clinician anticipate hemodynamic instability and prevent it from happening due to the fact that they monitor blood pressure continuously. Because they do monitor the blood pressure continuously, they allow for titration of vasoactive drugs, and they are usually found to be more accurate than non-invasive blood pressure cuffs, as non-invasive blood pressure cuffs have been often found to have more discrepancies and usually overestimate a low blood pressure. There have been valid—however, the validation studies that have compared continuous non-invasive blood pressure measurements to arterial line blood pressure management have not really revealed any of the contradictory results that have been observed. Another interesting reason to use an arterial line is because they have been shown to have a mortality benefit in patients who have hypertension at baseline. A total of 11,732 critically ill patients who were enrolled in the EISU collaborative research database were included in this study. Patients were then divided into two groups according to whether they received an arterial line versus whether they never received one. The primary outcome in this study that was discussed was in hospital mortality. Prior to conducting the study between the two groups, demographic variables were assessed as well as sensitivity analysis were performed, and they basically showed similar results in the subgroups with regards to obesity and sepsis and other demographic variables. Propensity, score matching, and inverse probability of treatment weighing models were used to balance any of the confounding covariates. Once this was done, it was found that invasive blood pressure measurement was associated with a lower in-hospital mortality rate. The confidence intervals did not cross one, and they were found within 95% of the sensitivity. What exactly is an arterial line system? Arterial line systems include one of the arterial line catheter. They include non-dispensable rigid-walled fluid-filled tubing. They include a pressure transducer, automated slow infusion of pressurized normal saline, and electronic transducer amplifier with display. The sequence of events is as follows. Basically the artery is cannulated by a provider who is trained in cannulating, and a stiff short catheter and pressure tubing is used and connected to pressure tubing to the transducer. The reason why it has to be a stiff short catheter with regards to pressure tubing is because we are essentially replacing a small part of the wall of an artery with a stiff membrane in order to gain the knowledge of what the actual blood pressure is. In order to measure said pressure, a hydrostatic reference level needs to be defined, and it's usually the level of the right atrium, and the transducer needs to be kept at this level at all times to ensure accurate blood pressure readings. Once the readings are transduced, they are amplified onto a monitor where a waveform is created and blood pressure numbers are displayed. So what exactly does the waveform look like? So what's interesting about an arterial line as compared to a non-invasive blood pressure cuff is that the arterial line will directly measure the systolic pressure, the diastolic pressure, and the mean arterial pressure. As you can see in the waveform above, the nadir corresponds to the diastolic blood pressure. The upstroke immediately after the diastolic blood pressure represents ventricular contraction and culminates at the systolic peak, which represents the systolic blood pressure. As the ventricle relaxes, its pressure falls below that of the aorta, resulting in the closure of the aortic valve, which creates a reflected pressure wave represented by the dichrotic notch. As one moves distally in the body, as in comparing A-lines from the radial artery or comparing to the dorsalis pedis artery, the dichrotic notch appears more delayed. Also, for example, in severely vasodilated patients, such as those with septic shock or neurogenic shock, the dichrotic notch will occur at a much lower pressure. When we talked earlier about arterial lines being able to obtain estimates for cardiac output, you can see that the calculation is basically down with the area under the curve, meaning that the arterial line waveform is measured, the area under the curve is measured, and then multiplying it by the heart rate can often give you an estimate of cardiac output. And therefore, there have been new creations from different things, such as the Vigileo A-line or the PICO, that are used to calculate other data points to assume hemodynamic variables in critically ill patients. So what you see here with regards to the A-line is that this is a normal peripheral arterial waveform. And so it does correlate with the EKG, so oftentimes you can use the arterial line comparing to the EKG to look for perfusing beats. And so what's noted is that the systolic blood pressure is always after the QRS complex. This phenomenon reflects the time that it takes the cardiac systolic pressure wave to reach the peripheral catheter and sensor. The dichrotic notch reflects the closure of the aortic valve. Of course, the same time delay applies to the dichrotic notch. The aortic valve has closed prior to the display of the notch. So now that we understand how the arterial line system works, as well as the creation of the waves, obviously, just like any monitoring that we do, there's always benefits and disadvantages. And so now we can discuss about whether or not what the issues are with the accuracy of the arterial line. So it's important to note that, as we can see here, there are multiple components to the arterial line. And unfortunately, each component of the measuring system can introduce inaccuracies in the measurement errors. And some of these inaccuracies can be to the location of the transducer, the leveling, the cannulas, the tubing. So across the board, it's always important that when your waveform looks to be inaccurate, you have to assess multiple areas to figure out the exact cause and how to fix it. So one of the issues is the location of the cannulation. So cannulation of the radial or dorsalis pedis arteries are the preferred sites of measuring invasive blood pressure through an arterial line. That's because the palmar implant or arches allow for collateral blood flow to the hand and foot. There are some times where brachial artery cannulation has been performed. This area is higher at risk for ischemic injury. Oftentimes, you can cannulate the femoral artery. However, that site has a higher risk of infection. What's interesting also is that it's important to note that systolic blood pressure tends to increase when measured at an increasing distance from the heart. So therefore, the site of arterial cannulation along the vascular tree is also an important determinant of systolic blood pressure. A dorsalis pedis arterial line will typically show a higher systolic blood pressure than a radial A-line, which in turn measures a higher systolic blood pressure than a femoral arterial line. This phenomenon occurs because of the complex summation and reflection of the pressure waves traveling over the arterial tree. What's interesting to note that despite the changes in the systolic blood pressure, the map, the mean arterial pressure, is again much less affected by this phenomenon. Therefore, when determining patient's blood pressure targets, the site of measurement of the invasive blood pressure monitoring must be taken into consideration. Therefore, also, this is a reason why treating MAP is often more foolproof than making treatment based on systolic blood pressure management. The measurement of the MAP is not only less affected by dampening and resonance, but also by arterial catheter location and the systolic blood pressure. Using the MAP to guide hemodynamic therapy can help to avoid mistreating patients based on erroneous values that are prominently displayed on a monitor for everyone to see. What's interesting to note is that the mean arterial pressure between the locations, dorsalis pedis versus the radial, should be similar. And that's because the more distal foot dorsalis pedis arterial line will have a higher systolic blood pressure and a lower diastolic blood pressure compared to the radial A-line. This is a consequence of the reflected pressure waves at all branch points blood encounters as it exits the heart and barrels down the arterial tree. So we discussed with regards to from the location, so after the artery is cannulated in the location that is specified for whatever reason is most conducive to the patient, the arterial line then needs to be connected to the transducer through pressure tubing. So at this point in time, once the arterial line is connected and the cable is connected to the monitor to be amplified, the transducer needs to be zeroed. What's important to note is that sometimes transducers can drift and they are prone to baseline drift. And so therefore, the transducer needs to be zeroed at a regular interval to ensure that there are no quantitative errors. However, oftentimes this drift is not that significant and it leads to a bias of less than three millimeters of mercury. However, it's very important that when the transducer, as we mentioned before, is leveled with the level of the right atrium. As we noticed that a leveling error of 10 centimeters will cause a measurement error of 7.4 millimeters of mercury in the blood pressure. It's also important to note that even if the transducer does not move, however, patient positioning can change the blood pressure management. For example, if the transducer gets accidentally dropped on the floor, the blood pressure will be higher. If the patient ends up going into, if the patient ends up lying flat or going into trendellumbary, the transducer will be higher and the pressure will be lower. So leveling and zeroing of the pressure transducer. As we mentioned before, a vertical difference of 10 centimeters between the pressure transducer and the artery of interest results in a pressure difference of 7.4 millimeters of mercury due to the hydrostatic pressure. It's very important to monitor the location of this transducer, especially in patients that are supine, whether they're in beach chair or prone, because wrong leveling and zeroing can result in wrong therapeutic actions with consequences of low perfusion in the circle of Willis. As we see here, depending on where the transducer is located, the map reading can change. So as we see, so as this patient was supine, the map was reading 64 millimeters of mercury. However, now that the patient is in beach chair and the transducer remains in the same location, we see that even though the map is reading 65, because that's where the blood pressure is reading from the arterial line, we can see that the level of difference between the patient and the circle of Willis up here in the brain, the map is actually 20. And so this patient, who despite having a normal map reading, is actually not perfusing his brain. Because as you change the location, whereas in a patient that's sitting up elevated or greater than 30 degrees of the head of bed, we see here that despite the fact that this transducer is at the level of the right atrium, he once again is still having marginal pressures with regards to his circle of Willis. So oftentimes, even if we accurately transduce and level the patient at zero, we have to worry about perfusing all our end organs. So we mentioned before that in addition to measuring blood pressure, oftentimes using the calculation of the area under the curve, you can measure other cardiac variables such as stroke volume variation, cardiac output. And so with regards to measuring accuracy, it's also important to determine are the numbers that we are calculating based on the arterial line, are they correct? And so when we talked about leveling, it was determining if the level of the transducer changes but the blood pressure still reads normal, do these pulse contour waveform derived measurements that we mentioned, are they still accurate? And so we know that a 10 centimeter departure, as we mentioned from the reference level of a Prussians transducer is equal to about 7.5 millimeters of mercury change of invasive hemodynamic pressure monitoring in a fluid system. However, this study decided to study the relationship between the site level of a variable arterial pressure transducer and the pulse contour derived parameters because these have yet to been established in a critically ill patient. So this is the first study of the kind. They looked at the elevation of the pressure transducer. So when they noticed and they studied this in their patients in their single site study, they found that the elevation of the pressure transducer caused significantly positive changes in the continuous cardiac index, stroke volume index, stroke volume variation and had negative changes in the rate of left ventricular pressure rise during systole as well as the systemic vascular resistance. What they noted was on average for every centimeter change of the transducer location, there was a corresponding change in the continuous cardiac index, there was a continuous change in the stroke volume variation, there was a continuous change in the stroke volume index. So they found that the variation of the arterially transducer position can result in inaccurate measurements of pulse contour waveform derived parameters, especially the further you moved from the phlebostatic axis and the phlebostatic axis is where they choose to zero it, which once again we said was at the right atrium. So there are many ways to tell with regards to the accuracy of your arterial line. And so the waveforms can give you a heads up as to whether or not your A-line is accurate. So over here with B and D, we actually see what is a normal waveform. And so physicians need to be aware that the systolic blood pressures may be inaccurate in a significant number of patients and pay attention to the shape of arterial blood pressure waveform due to damping and resonance phenomenon. Wrong and potentially harmful therapeutic interventions may be undertaken by health care providers who have not been trained to recognize these resonances in damping artifacts because they will misinterpret the systolic blood pressure value displayed on the monitor as real systolic blood pressure. The blood pressure waveform is a complex amalgamation of both anterograde and retrograde pressure waves and is affected by vascular compliance, distance from the left ventricle, and the 3D structure of the vascular tree. The map is much easier to measure accurately because it is less affected by dampening and resonance than systolic and diastolic blood pressure. An underdamped hyper resonant trace, for example, overestimates while a damped trace underestimates systolic blood pressure. The map is not significantly affected by these phenomenon and is essentially the same for both traces. So when in doubt with regards to the utility of what your curve looks like, of what your arterial waveform looks like, it's always safe to use the map. An underdamped or hyper resonant arterial pressure waveform results in overestimation of systolic arterial pressure and underestimation of diastolic arterial pressure. Curve A here is what we call an underdamped arterial pressure waveform, as you can see by the several dichrotic notches. B here is the optimized waveform with a single dichrotic notch. Curve C here is an overdamped arterial pressure waveform without a dichrotic notch, and an overdamped arterial pressure waveform results in an underestimation of systolic arterial pressure and an overestimation of diastolic arterial pressure. And once again, comparatively, you can see the difference between that with waveform D here, which is an optimized waveform with a single dichrotic notch. If the trace looks hyper resonant or overdamped, the treatment decision should be based on the map. If clinicians insist on making treatment decisions based on systolic blood pressure, then the damping within the measurement system must first be optimized before it is safe to use systolic blood pressure to guide therapy. And so once again, when we look at our waveforms, we can see here in cases of underdamping, the representative waveform differs from the true waveform as follows. You have a systolic blood pressure overestimation, a systolic pressure overshoot with a narrow peak can be observed, and a diastolic blood pressure underestimation. There will also be a blood pressure overshoot with a narrow peak can be observed, and a diastolic blood pressure underestimation. There will also be a pulse pressure overestimation and a deep dichrotic notch. Non-physiological oscillations during the diastolic phase will also be seen. The main reasons for why you can have an underdamped blood pressure signal in these cases are excessively stiff tubing and a defective transducer. In cases of overdamping, the representative waveform will differ from the true waveform as follows. A systolic blood pressure underestimation, diastolic blood pressure overestimation, pulse pressure underestimation, a slurred upstroke, an absent dichrotic notch, general loss of all the detail of the waveform. The main reasons for an overdamped signal are low infusion bag pressure, air bubbles in the circuit, blood clots, looser open connections, kinking, or obstruction of the catheter. As we mentioned before, because of that this is a system that is man-made that requires to be put together by a provider, there's always a chance for error at any course. So once again, as we mentioned, that any artifact along the lines of the system that is set up can lead to error. So this is the combined system of cannula tubing transducer as a second-order transmission line that guides the intra-arterial pulse wave to the transducer membrane. This second-order system can be characterized by its natural resonance frequency and damping factor. The natural frequency of the measurement system must exceed the frequency range of the arterial pulse because the goal is to accurately determine the maximum rate of pressure during oscillovolumetric contraction. Higher natural frequencies can be obtained by making the cannula and connecting tubing shorter, wider, and stiffer. The systems also exhibit damping caused by friction and the viscosity of the filling fluid. Critical damping is the amount of damping required to prevent overshoot. The damping coefficient of a critically damped system is one. However, this results in a relatively slow responding system. Therefore, the damping coefficient is actually less than one, and it's at 0.64, which is called optimal damping because it provides a good compromise between the responsiveness of the system as well as minimal distortion. So therefore, when your blood pressure is measured, your amplitude is actually only two-thirds with regards to the natural frequency. And we'll have some distortion as well as some issues with frequency. So it is noted oftentimes that there may sometimes be an artificial increase in invasive blood pressure monitoring due to the fact that the damping coefficient is not perfect at one. In addition to this intrinsic error, there is also oftentimes where arterial lines may develop blood clotting, and this is a very common case. In addition to this intrinsic error, there is also oftentimes where arterial lines may develop blood clots, kinking in the cannula, clamping of the arterial line tubing, air bubbles within the tubing. Narrow, long, or compliant tubing can actually cause the system to be over-damped with damping coefficients larger than critical damping. So it's also very important to make sure that you use the correct tubing with regards to your A-line setup. Damping will result in under-reading of the systolic blood pressure and over-reading of the diastolic blood pressure. It's important to note that the longer the catheter that's used for insertion, the higher the damping coefficient, the larger the radius of the catheter, the lower the damping coefficient, and the higher probability of recording an under-damped signal. It also depends on the size of the vessel that you have chosen to cannulate with regards to the catheter that you've selected to cannulate said vessel. So we discussed bubbles within the arterial line, and so the question becomes, why are having bubbles within the pressure tubing bad? Oftentimes, you'll see that people have used air in order to improve accuracy with regards to arterial line reading, but that's not suggested because obviously air bubbles entering into the bloodstream through the radial or the dorsalis pedosa artery are not healthy. So air is compressible and can actually lead to incorrect readings, and air may be flushed into the arteries leading to ischemia and embolus like we just mentioned. A damped arterial trace is often a blunted trace with a low systolic and high diastolic reading and mean arterial pressure. However, mean arterial pressure will often remain the same despite the inclusion of air bubbles. Causes of over-damping are the kinked catheter, a blocked line, or air bubbles in the line. In addition to over-damping and under-dampening, there's also another phenomenon known as whip, and so you'll often hear the discussion on rounds where you'll see, oh, there's a whip in the arterial line. What does that mean? So exaggerated waveforms with elevated systolic pressure and additional peaks in the waveform. As we mentioned before, usually there's only two peaks within the waveform, the actual peak of the systolic blood pressure and the dichrotic notch. But when you have additional peaks or a jagged reading, as you can see below, it's called catheter whip, and that's the result of excessive movement of the catheter within the artery. Typically, this problem is actually self-limited, but care must be taken not to interpret typically normal systolic blood pressure values with evidence of catheter whipping. Usually, catheter whipping will overestimate the pressure, meaning that the pressure will read higher than normal, and so oftentimes treatment will not be performed when necessary due to the inaccurate reading. However, once again, if you see your catheter with a lot of whip, then the principal guideline of the assessment is to use the MAP because, once again, the MAP is prone to less error. So, when you have discovered that your A-line is malfunctioning or is not reading accurately, what is the best way to troubleshoot said A-line? So, it's really important to keep an eye and to visually check the blood pressure waveform to ensure impeccable signal quality and identify artifacts at all times. Ideally, also, you should zero your transducer repeatedly and throughout the day at timed intervals to ensure the accuracy and functionality of the transducer. One way to test the damping properties of the measurement system is a fast flush test, and these should be repeatedly performed throughout the day. The fast flush test, or also known as the square wave test, is performed by flushing the crystallized fluid that fills the tubing transducer system with 300 millimeters of mercury pressure via the flush system. This maneuver generates high amplitude oscillating waves that will fade exponentially after the flushing maneuver depending on the damping coefficient. The natural frequency of the system is calculated by dividing the monitor speed. The time between oscillations will be short, and this is the natural frequency of the system, and it should be less than 20 to 30 milliseconds in order to resolve the details in the arterial pulse waveform. Ideally, post the fast flush test, you should see one bounce oscillation. If the system does not oscillate, then there's too much damping. There should be no more than two oscillations. A system which oscillates too many times is underdamped. There should also be a distinct dichrotic notch. The dichrotic notch is resolved from high frequency waveforms, which are usually of low amplitude and therefore more susceptible to damping. If the arterial line is progressively becoming more and more damped, the dichrotic notch will actually be the first feature to disappear. This figure illustrates the characteristic changes of the arterial blood pressure waveform in case of under and over damping and the corresponding fast flush test. The red arterial blood pressure waveform represents a normal non-distorted waveform with a normal fast flush test, whereas the blue arterial blood pressure waveforms represent an underdamped in the upper part of the figure and an overdamped in the lower part of the figure arterial blood pressure waveforms. Damping will change depending on the pressure of the system, and it should be Damping will change the amplitude of the oscillations as it influences the energy in an oscillating system. Thus the amplitude ratio of two consecutive resonant waves can be calculated by dividing the amplitude of the smaller wave by the amplitude of the higher one. Once the amplitude ratio is calculated, it can be plotted against the natural frequency in a specific graph that shows three areas adequate dynamic response over damping and under damping. In clinical practice, the oscillating waves induced by the fast flush test usually are visually inspected. To reduce the likelihood of underdamping, the operator should use short, stiff, non-compliant tubing, reduce movement of the catheter in the artery, and limit the number of stopcocks. A general recommendation is not to modify the transducer package unless it is absolutely necessary for clinical purposes. Since the components of a transducer kit are carefully selected with the aim to find optimal physical properties to avoid artifacts due to over- or underdamping. If the transducer package has to be modified, it is important to use only extra lines and stopcocks that are made for blood pressure measurement symptoms, i.e. pressure tubing. If these precautions are insufficient, adjustable damping devices can be used to modify the damping coefficient of the system when underdamping affects the signal transmission. In cases of overdamping, since the main reasons are air bubbles or blood clots in the circuit or kinking of the catheter, the only procedures potentially effective are modifying the wrist position in cases of kinking, removing air or blood clots from the tubing, or changing the catheter and arterial site. In case of under- or overdamping, the monitor scaling should be checked, as inappropriate scaling can imitate an under- or overdamped blood pressure signal. Beside the above-mentioned artifacts, patient movement during blood pressure measurements or leaning against the patient's arm used for BP monitoring may falsify blood pressure reading. In addition, substantial variations in blood pressure reading between different measurement systems may occur because different monitor devices use different algorithms for data processing, data averaging, and artifact filtering. So while we've discussed the fact that invasive arterial lines are actually still the gold standard for so many reasons as we've mentioned before, there are many inaccuracies as we've discussed. So when we look at different studies that actually compare the reliability of non-invasive blood pressure monitoring, what's interesting is that a lot of these studies have found that using the map is still safe. So this study here that was published in Critical Care Med in 2012 reviewed 150 patients where the cuff was either placed on the arm, thigh, or ankle, compared to also blood pressure management readings from an arterial line. And what they found was that the agreement with regards to mean arterial pressure, the non-invasive blood pressure cuff and the invasive arterial line actually had a lot of agreement for mean arterial pressure. However, systolic and diastolic pressure did not correlate at all. So this study that was conducted in AMICU in Montpelier, France, and there are 83 patients who were in shock, noted that the maps actually correlated. And so they determined that for patients who are critically ill for which arterial line access cannot be obtained, it would be safe to guide our management with regards to map, which is also very interesting because as we noted before, with regards to all the inaccuracies with regards to our arterial lines, that when in doubt, it is always safe to use the map. And so one of the conclusions we can obtain from the inaccurate arterial lines as well as the blood pressure cuff is, it's all about the map. I don't know how many people watched Dory the Explorer, but if anybody knows the song, I'm the map, I'm the map, I'm the map. So when in doubt, if you can't read your arterial line properly, or the system's overdamped or underdamped, or you've got WIP, or all you have is your non-invasive blood pressure cuff, just remember, it's okay to go off the map. So in summary, invasive blood pressure management through an arterial line is currently the gold standard to measure arterial blood pressure, although it definitely comes with sources of error as we've discussed. Accurate systolic blood pressure is often difficult to measure in routine clinical practice because of problems of hyperresonance or damping within the measurement system. The shape of the arterial pressure waveform has to be carefully and continuously assessed prior to making any treatment decisions. The location of the arterial line needs to be taken into consideration when interpreting systolic blood pressure measurements. Once again, making treatment decisions based on the map when monitoring any blood pressure in any form is less prone to error. Antitreating treatment according to the map may avoid mistreating patients based on erroneous values that resulted from measurement artifacts. I'd like to thank Society of Critical Care Medicine as well as the programming committee for allowing me to speak, and please let me know if there are any questions.

invasive arterial line monitoring

blood pressure management

continuous blood pressure measurements

pulse contour analysis

waveform interpretation

troubleshooting techniques

mean arterial pressure