Good morning, ladies and gentlemen. It is indeed a great honor and pleasure for me to be presenting the William C. Shoemaker Oration. I am really grateful to the Society of Critical Care Medicine for giving me this opportunity to have a woman coming from India delivering this honorary oration. It reflects the commitment of SCCM to diversity both in letter and spirit. And I'm really grateful to all of you to come here early in the morning to listen to me. William C. Shoemaker has been an inspiration to several generations of critical care specialists. I started my career in critical care medicine reading the textbook of Shoemaker and also his work on normal oxygen delivery. He's been a doyen in critical care medicine, especially hemodynamics, and has been an inspiration to many of us. And it's indeed a great honor for me to deliver this honorary lecture. I bring you greetings from the largest democracy in the world to the oldest one. I see many Indians in the audience and many of Indian origin. And I'm sure you do know that India is one of the second most populous countries in the world, 1.35 billion people. So if you take six people in the world, one will be an Indian. And what's very interesting, I'm sure many of you don't know, that it's a very young country. The median age of India is only 27. I come from the city of Mumbai. It's called the city of dreams, also maximum city. It's on the west coast of India by the Arabian Sea. The Marine Drive is one of the most beautiful promenades in the city. And you have the ceiling that goes across the Arabian Sea and iconic buildings like the Gateway of India, the Taj Mahal Hotel, and the Victoria Terminus, where I live in the city. I work at the Tata Memorial Hospital, as Ashish has already mentioned. This is one of Asia's largest cancer centers. It drains a lot of population from around India and around the subcontinent. Despite having a large population, the proportion of doctors is not as large as the population. And just to give you a comparison with another cancer center, the MD Anderson Cancer Center that you're all familiar with, if you look at the patient load, and this is, of course, data from the pre-COVID times, and about 41,000 patients annually in unit admissions and 1,700 faculty. And if you look at the hospital where I work, the Tata Memorial Hospital, about 42,000 patients, and this is about 200 attendings looking after this patient. So though the number of the population is large, and the workload for the clinicians is really, really huge in comparison, because we have a lot of patients to look after. I work in a mixed medical surgical intensive care unit. And of course, during COVID, we were also looking after general patients in the intensive care unit. I'm going to be speaking about assessing fluid responsiveness in 2023. Before I start, I'd like to say that I have no conflict of interest for this presentation. I'm going to be describing the concept of fluid responsiveness. I'm going to be talking about the current methods and the tests that are available, the limitations of some of these tests, some of the newer tests, and some of the ways to overcome these limitations. Now when you have a patient who presents an acute circulatory failure, volume expansion is usually considered as the first line of therapy. And I'm sure you agree one of the first few things that we do is we try to optimize the preload by giving IV fluids to the patient. Now it's interesting that, and this is an old meta-analysis that was published in Chess that shows that almost half the patients in ICU to whom we give fluids are actually not fluid responsive. We don't give antibiotics like this. There's nothing that we give so arbitrarily in the intensive care unit like we give fluids. Someone says give 500 mils. Someone says no, don't give any more fluids. Someone says give about 100 mils of fluid to this patient. So there's a wide variation in clinical practice as far as giving fluids to the patients. So does it really matter whether we give less or more fluids? And we know that we have a lot of data now to show that excessive fluid loading is associated with increased complications, mortality, and length of ICU stay. On the other hand, if you have uncorrected hypovolemia, even this may affect the tissue oxygenation and leads to organ dysfunction and death. So it really does matter whether you give more or less fluid to a patient. Just to demonstrate this with an example, the typical kind of patient that we present in our intensive care unit, I'll just give you this case, is a 50-year-old male, a case of CA colon, presents with intestinal perforation. He's had an inferior wall myocardial infarction one year back, has an ejection fraction of 40%. He's febrile. He's got elevated white cell count. The respiratory rate is 36, tense distended abdomen. He's been intubated, ventilated on high FIO2 and PEEP. About three liters of fluid has already been given to this patient in the emergency department. His heart rate is 120. The blood pressure is 80 by 50. A central line has already been inserted, and the CVP is 14. This patient has a lactate of 4.5. This is the kind of typical patient that will come to your unit who'd already have been resuscitated a bit and has these kind of numbers. Now, I just want to ask you, how many of you would want to give more fluid to this patient? Can I have a show of hands? Okay. Very few people. How many of you feel that we should not give fluid to this patient? Can you raise your hands? Okay. A few more. Some are not raising their hands, but we have to do something for this patient. Should we give more fluid? This is really the dilemma. Are there reasons to give this patient fluid? Yes. This patient has, we know, intra-abdominal sepsis, perforation, there's relative hypovolemia. We know there's sequestration of fluid in the bowel, persistent tachycardia, and this patient also has high lactate. So there are some reasons why I would consider giving fluid to this patient. On the other hand, are there reasons not to give fluid to this patient? If you look at the CVP, it's 14. This patient's already received about three liters of fluid, and he also has this history of the previous inferior bowel MI. So determining which patients will be fluid responsive is really a challenge in the intensive care unit, not during initial resuscitation, but definitely following the initial fluid administration. Because on one side, you want to restore the organ perfusion. You want to correct the hypotension, while on the other side, you're worried about pushing the patient into pulmonary edema. So when we ask the question, should I give fluid to the patient, what we really want to know is that, will the cardiac output increase with fluid loading? Because if the cardiac output is not going to increase, it's not going to serve any purpose. It's not going to benefit the patient, and it might actually produce harm to this patient. So when I decide whether to give fluids or not, I ask myself three questions. The first question is, is there the presence of acute circulatory failure? Now the normal human heart is also fluid responsive. That doesn't mean that we need to give fluid to the patient. So there has to be some clinical signs of shock before you decide to give fluid to this patient. And the second question, as I've just mentioned, is, is the patient fluid responsive? And the third question we should ask ourselves is, is there any major risk of giving fluid? So even if the patient is fluid responsive, if there's a risk of giving fluid to this patient, I might not consider giving more fluid, I might consider giving vasopressors. So if my answer is yes to all these three questions, that's when I decide to give fluid to the patient. And the second question is, is my patient fluid responsive? So what do we mean by fluid responsiveness? Fluid responsiveness essentially is a state where administration of fluid will lead to improvement in the stroke volume and hence the cardiac output. Now one of the ways of assessing fluid is by giving fluid to the patient and looking at an increase in the fluid, you know, in looking at increase in the stroke volume. As I mentioned earlier, this is one of the oldest, you know, described fluid challenges by Weill and Henning. And this was a little complicated, different volumes of fluid, looking at different values of CVP. And subsequently, Professor Jean-Louis Vasseau had the modified fluid challenge. And then more recently, you have this fluid challenge that has been described where you look, you give about 4 mL per kg of crystalloids to the patient over a period of 10 minutes. And you look at the response, whether there's an increase in cardiac output or no increase in cardiac output. And then you also look at the filling pressures. If at any point the filling pressures increase, whether or not the cardiac output increase, you would stop giving fluid. If the filling pressures do not increase and the cardiac output doesn't, you should also consider whether the amount of fluid you've given is enough or there are ongoing losses in these patients. And of course, if the cardiac output increases and the filling pressures do not increase, you could consider giving more fluid. So this is just a simplified algorithm as to how you could look at a fluid challenge. But when you give a fluid challenge, you're actually giving fluid to a patient. So this could actually be detrimental in a patient who's not fluid responsive because then you're giving unnecessary fluids to this patient. Now, there's a wide variation in the way people give fluid, both in terms of volume, the duration and timing. And this is a very elegant study conducted by the group from Professor Maurizio Cecconi and Antonio Messina, where they've looked at, done a pharmacodynamic analysis of the fluid challenge. And they've given four mils of fluid over a period of 10 minutes and also 20 minutes. And it was a crossover randomized clinical trial. And the important findings of this is that a fixed volume worked better than a variable volume and when given over 10 minutes. And the important thing is that the effect of the fluid challenge fades away after five minutes. So you can't just give it over a long period of time. And you need at least a volume of four mils per kg. Now, as I said, you have to give fluid. So if a patient is not a fluid responder, it might actually produce harm. So one of the ways of doing this is by giving a mini fluid challenge to this patient. Smaller volumes. The original study described in anesthesiology about 10 years back used colloids, but we use about crystalloids given over one minute. And then you do echocardiography, you look at the VTI and increase more than 10% fluid responsiveness. Now, some people look at this as a way of predicting fluid responsiveness, but you are giving fluid to this patient, 100 ml. And if you're repeating this a couple of times, then it would be a larger volume of fluid. One of the other downsides of this is that you need someone who's very good at echocardiography because you're giving very small volume of fluid. And so you need to have a very accurate determination of increase in VTI. So a lot of people look at predicting fluid responsiveness, and they look at the change in CVP. And I'm sure you all know this meta-analysis from Paul Merrick, which was published several years ago now. And it's does CVP predict fluid responsiveness? And this was an updated meta-analysis, 43 articles, papers, almost 2,000 patients. And look at the area on the curve, it's just 0.56. It's almost like tossing a coin. And the meta-analysis concluded that there's no widespread data to support the practice of using CVP, and this approach of fluid resuscitation should be abandoned. I'm not saying that CVP is useless, but CVP does not help you predict fluid responsiveness. None of the static parameters like CVP, PAOP, GDV are accurate in assessing fluid responsiveness. So preload is not really preload responsive. We're not interested today in knowing whether the patient can take in more fluid. What we're interested in knowing is that by giving fluid, is your cardiac output or your stroke volume going to increase? Now when we give fluid to a patient, we presume that this is a Frank-Starling curve. We presume that they're on the steep part of the Frank-Starling curve, so that by giving a certain amount of preload, you have a proportionate increase in the stroke volume or the cardiac output. But after you've given large amounts of fluid, or if the patient's cardiac function is not good, you might give the same volume of fluid and you might not have as much increase in the stroke volume. And this is a dynamic assessment, and this is one of the reasons why the static parameters like CVP, PAOP, GDV, et cetera, may not help you assess fluid responsiveness. Now if you look at the recommendations from the Surviving Sepsis Campaign Guidelines that were published in 2021, or the Fluid Administration of Acute Circulatory Failure, these were the ESICM guidelines, and I was very fortunate to be a part of this group. And they suggest and they also recommend that dynamic over static parameters should be used to assess fluid responsiveness, and we should not rely on static parameters like CVP anymore for this purpose. So what are these dynamic parameters that we were talking about? So these are respiratory variation, stroke volume, like pulse pressure variation, stroke volume variation, or variations in IVC or SVC, aortic fall velocity, or some non-invasive techniques using pulse oximeter, plethysmographic waveforms. We also have a new test, which was developed in our unit, which I'm going to be speaking to you about. That is the tidal volume challenge, end-expiratory occlusion test, lung recruitment maneuvers, PEEP challenge. And these are some of the newer methods that are being developed. Now all these depend, are based on heart-lung interactions, and there's one test that does not depend on heart-lung interactions, and this is the passive leg raising test. I'm just going to give you a brief overview of these tests. Now I'm sure all of you have seen, you know, when you put in an arterial line, you see these kind of dynamic changes in the arterial waveform during mechanical ventilation, during inspiration and expiration, and this is normal physiology. And you will see these, that when a patient is fluid responsive, you will see big swings in the arterial waveform in response to these airway pressure changes during mechanical ventilation. And this is exactly, that's a principle that is used while looking at Prandtl pressure variation and stroke volume variation. You look at the difference in the pulse pressure, and you divide it by the mean, or the difference in the stroke volume, and you get this kind of number on the monitor. So for example, if it's more than 12%, this patient is likely to be fluid responsive. So how accurate are these parameters like pulse pressure variation and stroke volume variation in predicting fluid responsiveness? Now if you look there, this is one of the most extensively studied parameters, and more than three meta-analyses looking at it, and if you look at the area under the curve, it's pretty good. They're pretty good predictors of fluid responsiveness. However, they aren't without limitations, and there are many reasons why they could be false positive or false negative. They don't work well in spontaneous breathing, when there's raised internal pressure, when the thorax is open. But one of the biggest limitations is the use of low tidal volume, which we're using a lot now in the intensive care unit, and even the operating room. So when you use low tidal volume, we're using less than 8 mL, sometimes 6, or even less than 6 mL, and we're not only using it in acute respiratory distress syndrome, or ALI, we're also using it in anesthesia, trauma patients, septic patients. Now the problem with these tests is when you use low tidal volume, the low tidal volume may be insufficient to produce significant changes in intrathoracic pressure. So the pulse pressure variation and the stroke volume variation may actually indicate a non-responsive status even in a patient who are fluid responders. So you will get false negative readings when you use pulse pressure variation and stroke volume variation. So PPV and SVB do not reliably predict fluid responsiveness at 6 mL per kg. So really good tests, very reliable, but they don't work at 6 mL per kg, and we're using about 6 mL per kg ideal body weight these days. So in our unit, we developed a new test, and this is called the Tidal Volume Challenge, which we call the test the Tidal Volume Challenge, which we published in Critical Care Medicine in 2017. And we hypothesized that, OK, you have these excellent variables, very reliable, PPV and SVB, which work very well at 8 mL per kg, but they don't work at 6 mL per kg. So our hypothesis is that why don't we transiently increase the tidal volume from 6 mL per kg to 8 mL per kg, and then look at the pulse pressure variation and the stroke volume variation. So that's what we did. We increased the tidal volume from 6 mL, took one set of readings to 8 mL. And like a fluid challenge, we were giving tidal volume, so I decided to call it the Tidal Volume Challenge. And then, of course, we don't want to keep the tidal volume at 8 mL for too long, and we brought the tidal volume back to 6 mL per kg. And then we also confirmed how well this test worked by actually giving fluid to the patient and looking at how much was the increase, whether they were responders or not. And this was what we did. And what's very interesting is if you look at the area under the curve, as we expected at 6 mL per kg for pulse pressure variation, the area under the curve was not very good when we increased the tidal volume to 8 mL per kg. As we hypothesized, the pulse pressure variation was very good at predicting fluid responsiveness after giving a tidal volume challenge. But what was more interesting is better than the 8 mL per kg was the delta. That is, the difference between the 8 and the 6 mL per kg was a better predictor of fluid responsiveness. So the delta PPV had an area under the curve of almost 0.99 and was a better predictor. And when we looked at stroke volume variation, also we had similar results. Of course, this was a proof of concept study. This was a small number of patients, but it showed that increasing the tidal volume like that could make a difference. And we got cutoff values of about 3.5 and 2.5. That means if your pulse pressure variation increased by more than 3.5 after giving a tidal volume challenge, this was a good cutoff to discriminate between fluid responders and non-responders with a good sensitivity and a specificity. I'd like to just demonstrate this, how easy it is to perform the tidal volume challenge using pulse pressure variation. All you need is an arterial line. And as you can see, when you have an arterial line, you can have your pulse pressure variation as a number on the monitor. And I've just put the ventilator side by side. And this is a patient with ideal body weight of 45, ventilated with 270 mils of tidal volume. And you can see that what we do is very simple. We have to increase the tidal volume from 6 mL per kg to 8 mL per kg for a minute. And here you can see that from 270, the tidal volume is being increased to 360. And we do this just for one minute. And after doing this, you observe the pulse pressure variation. And it's very interesting, because you look at the changes. And when you increase the tidal volume, you will see a rise in the pulse pressure. And you can see the pulse pressure variation has now increased from 8 to 10. It's gone to 13. And you just do this for one minute, because we don't want to keep the tidal volume at 8 mL for too long. And it's increased to 15 mils. And then after a minute, we need to reduce the tidal volume back to 6 mL per kg. And look at the pulse pressure variation. And there we're putting the tidal volume back to the initial 270. And what's interesting is when you get the tidal volume back to 6 mL per kg, you actually see the pulse pressure variation going back to the baseline. And we started with 8. And after a minute, the pulse pressure variation increased to 15. The difference between 15 and 8 is 7. This is more than 3.5. And so this is a patient is a fluid responder with a good sensitivity and specificity. And it's quite reliable. And what's interesting is that to perform this test, you don't need continuous cardiac output monitoring. All you need is an arterial line. And this is very helpful in resource-limited settings. And when I say resource-limited settings, it's not only certain countries. But even within a hospital in the developed world, you have resource-limited settings. The kind of monitors and infrastructure you have probably in intensive care unit may not be the same that you have in ward areas and in the emergency department. So subsequently, this test became very popular, of course, because of not using cardiac output monitoring and simple to perform. And interestingly, this test works very well in the operating room, because you give muscle relaxant to this patient. So there's no spontaneous breathing. You need the patient to be deeply sedated, not necessarily paralyzed. But in the operating room, the accuracy is really good. And this was studied by the group of Maurizio Ciccone and Antonio Messina. And they used it in neurosurgical patients. It's also been used in laparoscopic-assisted surgery, Trendelenburg position, and also in prone position in neurosurgical patients. And they found that the tidal volume challenge reliably predicted fluid responsiveness in these settings. and this was a study that came from Egypt where they use both the tidal volume challenge and the passive leg raising and after giving a tidal volume challenge they looked at the reliability of this in using PPV to discriminate between responders and non-responders and they got similar cut off values as we got in our study. And this is a study was published in BJA from Olfa Hamzawi and Professor Jean-Louis Taboul and they looked at patients who also had some spontaneous breathing because you know in patients in the ICU are not paralyzed and they do have some spontaneous breathing efforts and they try to look at the reliability of tidal volume you know performing a tidal volume challenge both on pulse pressure variation and also by performing a passive leg raising so they did something very similar and interestingly the the values that they got the area under the curve was not that great but definitely it was better than when you use it at 6 ml per kg when you increase you give up you do a tidal volume challenge the area under the curve was 0.73 so it gives you some confidence that by giving a tidal volume challenge you can improve the reliability even in patients who have some spontaneous breathing efforts and they concluded that in patients mechanically ventilated with spontaneous breathing the changes in pulse pressure variation obtained after both the tidal volume challenge or after giving a passive leg raising tests are superior to the baseline values of pulse pressure variation and subsequently this has been tested in many studies and they've looked at the effect of the tidal volume challenge in patients with decreased respiratory system compliance so spontaneous breathing decreased compliance these are some of the limitations of using the pulse pressure variation and other indices that look at heart-lung interactions and the tidal volume challenge helps you overcome some of these limitations and even during COVID times you know where people were hesitant to put in femoral lines and do you know thermo dilution or do continuous cardiac output monitoring the tidal volume challenge was recommended as one of the tests that could be used and this is a very a very nice algorithmic approach that was given by Professor Jean-Louis Vasseau which was published in the Blue Journal and he talked about a practically how pulse pressure variation could be used and also how it could be integrated with the tidal volume challenge now of course pulse pressure variation doesn't work well when you have arrhythmias and spontaneous breathing but in a patient on controlled mechanical ventilation you look at the pulse pressure variation if you have values above 13 of course this is a fluid responder you don't need to do anything else the gray area is between 9 and 13 and in this kind of setting you can do a tidal volume challenge and this may help you unmask some of the fluid responders now if your pulse pressure variation is less than 9 then you look at the tidal volume if the tidal volume in which the patient ventilated is more than 8 ml then you know this patient is likely to be fluid responsive however if the tidal volume that is used is lower less than 8 ml then you could consider doing a tidal volume challenge and this is how you could integrate using pulse pressure variation and stroke volume variation echocardiographic variations are very popular of course we have SV you can look at SVC variability you can of course look at aortic root velocity but the most popular one is usually looking at respiratory variations in the IVC diameter using trans thoracic echocardiography and this has become like the new stethoscope in the intensive care unit all critical care fellows and are trained in doing bedside focus and we look at respiratory variations in the IVC diameter does not work too well in spontaneous breathing but during control mechanical ventilation you can lose either the distensibility index or you could look at the Delta IVC and depending upon what formula you're using the different cut off values but this is the largest study that's compared all the you know echocardiographic indices to predict fluid responsiveness and interestingly the the area under the curve was not very great when you look at IVC and you know a lot of people do not realize that even I'm looking at respiratory variations in IVC diameter depend on heart-lung interactions so it has all the limitations of the tests that use heart-lung interactions including low tidal volume so this is a little disappointing to see you know values like 0.6 and 0.7 area under the curve and the Delta SVC that's the superior veneer cover variability was more reliable than the IVC variability if you look more closely at this paper almost two-thirds of these patients were ventilated using low tidal volume and that might be one of the reasons why you know the the which reduce the accuracy of these tests now more recently this is a meta-analysis that looks looks at the diagnostic accuracy of IVC respiratory variations that we commonly use to predict fluid responsiveness so it's about 0.8 and 0.7 is the sensitivity and specificity so quite good and it's it's really good because it's you could do it at the bedside it's very easy to do you don't need to put in any lines so it's a great way of quickly you know assessing the patient's fluid status the only problem is that it's not continuous in my unit you know the fellows would do it once and then they don't repeat it they don't repeat it many times we have one machine among 12 patients so that could become a challenge but very quick non-invasive way of quickly determining the fluid status of the patients and now you have many other studies that are looking at brachial artery velocity they're looking at carotid artery peak velocity so move people are moving more and more peripheral because this is very easy to do and more easy to interpret and you know instead of looking at IVC variability and this is a study that comes from the group Michael Pinsky are looking at you know jugular vein distensibility but these need to be still validated and studied further before we can use them as surrogates instead of looking at IVC variability and other surrogates for stroke volume. So moving on to another test that looks at heart-lung interactions and this comes from Professor Tabool and Professor Monet from Paris and this is called the end expiratory occlusion test and a very smart test this also depends on heart- lung interactions so what you're doing is you're giving a 15 second expiratory hold to the patient on the of course it requires a patient to be intubated and mechanically ventilated and when you give this 15 second expiratory hold what you're doing is you're preventing the next inspiration from occurring and there will be an increase in cardiac output in sudden increase in preload and this if this is more than 5% then it's very reliable with a good sensitivity and specificity to predict fluid responsiveness. The only problem is that the 5% of 5% is a very small you know sort of margin it's almost like you know the margin for error when you're looking at cardiac output. This test has been extensively studied and this is a meta-analysis of several you know studies using the end-expiratory occlusion test used both in the operating room and in the intensive care unit and they've got a cut-off value of 5% which increases increase in cardiac output can reliably predict fluid responsiveness. Very easy test to do at the bedside all you need is a ventilated patient but as I mentioned the 5% cut-off becomes a little difficult so the unit the same group has done about another very interesting study they've combined the end-expiratory hold along with the end-inspiratory hold so doing both the expiratory hold and the inspiratory hold when you do an end-expiratory hold you'll have an increase in stroke volume of course and when you give an inspiratory hold you'll have a drop in stroke volume this would be larger in fluid responders so what they've done is they've combined the two and they've been able to get a larger you know a combined effect which is larger than the 5% that you see with just doing the end-expiratory occlusion test. Of course quite complicated but once you have automation on ventilators which can make you do these maneuvers this would probably be the future because you have a 13% cut-off that's more reliable than a 5% cut-off. The only downside of this test is that you have you need real-time cardiac output monitoring that's one of the limitations so they've investigating further studies where they're trying to look at other variables like plethysmorphic perfusion index etc where they could use this as reliable surrogate but of course these need to be tested in future before we can look at them. Then the more and more studies like the tidal volume challenge looking at other tests to increase intrathoracic pressure and use this to discriminate between responders and non-responders and this comes from the group of Matthew Beer they've looked at recruitment maneuvers what they've done is they've given continuous positive airway pressure of 30 centimeters for 30 seconds and of course when you increase the intrathoracic pressure you'll have a decrease in stroke volume and a decrease more than 30% discriminated well between you know flu responders and non-responders with a good sensitivity and specificity. Many other groups have looked at stroke volume you know increase you know doing a recruitment maneuver and looking at the drop in the stroke volume but this again the even the the way the recruitment was given and also the results need to be standardized before this can be adopted into clinical practice. Another test that's been tried is the peep the peep challenge where they've you know used increase the peep to about five and try to discriminate again between responders and non-responders and this improves the reliability of both pulse pressure variation and stroke volume variation. Another interesting study is from the group of Antonio Messina and Maurizio Cecconi what they've tried to do is they've tried to look at patients on pressure support ventilation and give a sigh maneuver so they gave a sigh of 15 20 25 at different intervals and then they look at the systolic arterial pressure this the pulse pressure as well as the stroke volume index at these different intervals and then they plot graphs like this and they found a difference in the slope you know between responders and non-responders very complicated to perform but as I mentioned these this is like proof of concept and with automation in in you know with on the ventilator this might really be the flu future. There are certain non-invasive techniques like using plethysmographic indices that could be used and this is used a lot in the operating room and this is a meta-analysis of various studies which combine studies both in the operating room and in the ICU and if you look at the reliability of these non-invasive plethysmographic indices it's about 0.85 which is which is quite good in fact if you've looked at an arterial waveform and you looked at a pulse oximeter waveform and you look at the swings in the arterial waveform you I'm sure you've noticed that the pulse oximeter often follows it follows the same trend as the arterial waveform and this is the principle that is used where they look at the perfusion index which is the difference between the you know ratio of that which is a ratio of the non-pulsatile and the pulsatile you know blood flow through the capillary bed and you the PVI that is the pleth variability index is the difference between the perfusion index divided by the mean and you get this kind of cutoff value and these are you know easy to use non-invasive you don't need to put in the arterial line but they work well in the operating room they don't you know because you need you know if your patient is in shock and the peripheries are cold you need you know you need a good signal capture for these non-invasive devices and this is a very elegant study from the group of Professor Tabool who's shown that when a patient is in shock and on norepinephrine they don't you know these devices don't the pleth variability index don't doesn't work very well so a good monitor you to use in the operating room but does not fare so well in the inpatients who are on vasopressors. So I've just outlined all the tests that look at heart-lung interaction those respiratory variations in stroke volume, tidal volume challenge and expiratory occlusion tests and some newer tests as I mentioned all of them are based on heart-lung interactions and they have the same you know the limitations of low tidal volume, arrhythmias, open thorax etc and you have the passive leg raising test that does not depend on heart-lung interactions so it can overcome a lot of the limitations that are related to heart-lung interactions and I'm sure all of you are aware of the passive leg raising test and this is a very nice cartoon from Professor Jean-Louis Tabool who talks about how to do this test well and of course you have to start from semi recumbent position and you need to move the bed and not the patient because you shouldn't you know though it's called passive leg raising you don't raise the legs there could be tachycardia and this itself could you know affect the cardiac output and you raise the legs for about 30 seconds to 1 minute before making the interpretation what's very important as I mentioned is you need real-time cardiac output monitoring to make an accurate interpretation of the patient so this is the correct way of doing a passive leg raising and they have done a meta-analysis of all the studies this has been extensively studied and you can see the area under the curve is 0.95 and they have a very nice threshold of 10% so if the cardiac output increases more than 10% after performing the passive leg raising this patient is likely to be a fluid responder with a good sensitivity and the specificity and the limitation as I mentioned you need real-time cardiac output monitoring so the group has now further tried to look at other ways to use a surrogates for cardiac output monitoring like using plethyscoma graphic saturation and also using capillary refill times these are very interesting newer concepts that they're using and they found that a drop in the capillary refill time of more than 25% after a volume expansion helps to discriminate between responders and non-responders so you could use them as like surrogates instead of cardiac output real-time cardiac output monitoring which may not be possible so this is a great test because it overcomes most of the limitations with the test using heart-lung interactions the only limitations are as I mentioned the need for real-time cardiac output monitoring doesn't work very well when a patient has intra-abdominal hypertension and also if you're wearing your compression stockings various you know tests have been you know methods for I have been used as surrogates for using real-time cardiac output but the most reliable is when you use real-time cardiac output or you use echocardiography the other methods that have been tested don't really work as well as when you use continuous cardiac output monitoring but a great test that could overcome most of these limitations some newer tests you know you can't do passive leg raising in a patient who's in prone position so the Trendelenburg position has been recently tested and what they find is that is it you know the change in cardiac index during the Trendelenburg maneuver reliably predicts fluid responsiveness and this is in ARDS patients and we had a lot of these kind of patients during COVID and we were proning a lot of patients so this is one of the newer tests that you could use in patients who are in prone position and with severe ARDS and this is a recent meta-analysis of looking at the predictors of fluid responsiveness and this is looked at those tests in only those patients who are ventilated with low tidal volume and that's like most of our patients in intensive care unit and I was very happy to see that you know the tidal volume challenge the stroke volume variation and end-expert occlusion tests have a good area under the curve along with the other tests like of course many fluid challenge the passive leg raising and the PPV so this of course needs to be further studied now people always say okay you have all these tests they really work well in predicting fluid responsiveness despite having limitations but what about outcomes you know it does using all these tests for assessing fluid responsiveness really alter outcomes and we have very few studies we have only about three trials and if you see very small numbers and not much no difference in mortality you know just differences in the volume of fluid or the positive fluid balance in 72 hours or just the duration of mechanical ventilation renal replacement therapy or blood cell transfusion no difference really in mortality very small numbers you need very large trials to demonstrate that using these parameters to assess fluid responsiveness can actually make a difference to you know a mortality difference but what's interesting is you know this is the meta-analysis that looked at incorporating the dynamic assessment of fluid responsiveness into a goal-directed therapy and this was a systematic review and meta-analysis and what is it very interesting that they demonstrated they concluded in this that it favored using you know tests for assessing fluid responsiveness into a goal-directed therapy so a goal-directed therapy guided by assessing fluid responsiveness appears to be associated with reduced mortality ICU length of stay and duration of mechanical ventilation so definitely it's promising we don't have you know a mortality difference when you use the test directly but when it's incorporated into a goal-directed therapy there seems to be definitely a benefit of using the test to assess fluid responsiveness definitely you'll end up giving less volumes to this patient and you may not give unnecessary fluid to a patient who's non-responsive so again coming back to the three questions I asked myself before giving more fluid is this patient in acute circulatory failure yes is this patient fluid responsive yes but again this is only one part of the question is the patient fluid responsive or not is not a reason to give fluid to the patient what's really important is to find out whether there is any risk of giving fluid to this patient as I mentioned that even if the patient is fluid responsive that's only one part of the question the second part is is there a risk of giving fluid because if there's a risk of giving fluid to this patient for example a very weak heart or very severe ARDS even if the patient is fluid responsive I might not give fluid to this patient I like this cartoon very much it says either he's gone or the net is down and this is pretty much the situation nowadays everybody's just looking at the monitor at the numbers nobody's really looking at the patient anymore we're so so dependent on technology and I always say that when technology is master disaster is faster so I believe of course technology is important and we must use technology to our advantage but don't make it your master you have to integrate the numbers that you see and the technology that you have with your clinical assessment and it has to be used in conjunction because just these numbers on the monitor and the technology is really not going to you know give you the right answers and I just end with my take-home message that today when we assess fluid responsiveness we need to look at dynamic over static parameters parameters like CVP POP are not useful in predicting fluid responsiveness and we shouldn't give more fluid after the initial resuscitation we shouldn't be giving more fluid to the patient without assessing fluid responsive because not only will it be not beneficial if the patient is known not a fluid responder but it will actually could produce harm if you're giving fluid to someone who is not fluid responsive it may produce harm the commonly used tests are blood pressure variation stroke volume variation echocardiographic variables variations like IVC diameters newer tests like tidal volume challenge and expert occlusion tests and passive leg raising are the most commonly used at the bedside most important the presence of fluid responsiveness is only one part of the question being fluid responsive does not mean you need to give fluid to the patient you need to see whether they pay there are any risk of giving fluid to this patient and if there are any risk you should probably consider giving vasopressors rather than giving more fluid to the patient and there definitely is more and more data now showing that incorporating dynamic assessment of fluid responsiveness into goal-directed therapy does help reduce mechanical ventilation ICU stay and also mobile mortality of course we need you know more studies and larger numbers before we can say that there can be a mortality benefit but we don't need to wait until that time to use the various tests that are available to assess fluid responsiveness and what's most important is that no test or tool is perfect and integration of various dynamic indices along with your clinical assessment is really really important and I think that's that's important not just using technology but using your clinical assessment along with the test in fact using a conjunction of several different tests when we have patients coming to the intensive care unit we first use echocardiography then I would put in arterial line I would look at the pulse pressure variation I would do a tidal volume challenge when I get the patient back on spontaneous breathing I would you know since the pulse pressure variation is not reliable I would consider doing a passive leg raising so you can use a combination of different tests along with your clinical assessment to evaluate whether this patient is fluid responsive or not and I'll end with what this very bright man I would answer and said that it's not we know but there's still a lot we don't know and what we know may not really be meaningful so not everything that counts can be counted and not everything that can be counted counts and with that I thank you very much for your attention

fluid responsiveness

assessment

critical care medicine

patient outcomes

pulse pressure variation

echocardiographic indices

clinical assessment

dynamic parameters