Hello, and welcome to the multi-professional adult critical care review course. I'm Greg Martin, a professor and pulmonary and critical care faculty member at Emory University and Grady Memorial Hospital in Atlanta, and the president for the Society of Critical Care Medicine. I'm going to talk about evidence-based management of sepsis and septic shock. There are three things we're going to cover together. We're going to define sepsis and septic shock. We're going to talk about sepsis management, and we're going to talk about septic shock management. I don't have any particular conflicts of interest, other than the fact that I do research and serve as a consultant, particularly around things in critical care, such as sepsis, ARDS, and COVID-19. First, let's start with the question, what is sepsis? Number one, is sepsis bacteria circulating in the bloodstream? Or is it number two, the SERS criteria? Number three, a host immune response to infection? Or number four, a complex syndrome related to infection that is difficult to diagnose and may be even more difficult to treat? Sepsis has been defined many ways over many years. It's not necessarily bacteria in the bloodstream, although it is related to an infection. And sepsis is more complex than just the immune response or the SERS criteria. It is a host immune response to infection, but it remains more complex than that. And the way we define sepsis now, I'll show you in a few moments, is the host immune response to an infection as that immune response becomes dysregulated. From a historical perspective, sepsis has been described many times. In this first description, sepsis is described as an infection caused by small creatures that are invisible to the eye that cause dangerous diseases. And then in the second description, actually a very good description of sepsis, hectic fever at its inception is difficult to recognize but easy to treat. But left unattended, it becomes easy to recognize and difficult to treat. That we understand now as sepsis. A second question, which of the following was part of the original 1992 consensus clinical definition of sepsis? Multiple organ dysfunction, procalcitonin, SERS, or shock? The earliest clinical definition of sepsis was defined by the American College of Chest Physicians and the Society of Critical Care Medicine back in 1992, now described as sepsis one. In this criteria, there were important components described, including the systemic inflammatory response syndrome, SERS, and multiple organ dysfunction. The SERS criteria were defined in that original sepsis one definition, identifying these four physiologic derangements that may occur with infection or with other abnormalities as well. And sepsis then was defined as the intersection between infection and SERS. When those two occur together, then sepsis was considered to be present. It also defined severe sepsis as those patients who had infection and SERS, meaning sepsis, but they also had organ dysfunction in one of several different organ systems. This shows the intersection of infection and SERS, where sepsis lies, as well as those patients with organ dysfunction, severe sepsis. But it also reminds us that SERS criteria may exist for other reasons other than infection, such as trauma, burns, and pancreatitis, as examples. And it also reminds us that there are several types of infections which may lead to sepsis. Bacterial infections, fungal infections, parasitic infections, and viral infections, whether they're in the bloodstream or not, may be sources of sepsis. Pancreatic shock in the original sepsis definition was defined as persistent or refractory hypotension despite adequate fluid loading, which leads to the conceptual model of increasing levels of severity, increasing levels of risk of death, and stepwise moving from at the bottom SERS and infection to being together as a cause of sepsis, to then causing organ dysfunction as in severe sepsis, and finally at the top of the pyramid with the highest risk of death in those with septic shock. That original sepsis 1 definition was revisited in 2001, and there were several additional criteria considered in that to try and make the definition potentially more sensitive or more specific. But despite that, all these additional criteria, it turns out that there really was nothing better than the original definition in terms of clinical performance. And that lack of additional performance served to underscore the challenge that was present in changing the definition. There were several concerns with the prior clinical consensus definitions, both sepsis 1 and sepsis 2. The first of those is that they were SERS-based definitions, which resulted in lack of sensitivity and specificity. The term severe sepsis was confusing, as sepsis was always considered to be a severe disease. Clearly from an epidemiologic perspective, there were different criteria that would produce different results depending on how the definition was applied, and also the definition was based on expert consensus and not based on data, as it may have been able to in more recent times. Those challenges led to the creation of the sepsis 3 definition, which was published in JAMA in 2016. So question number 3, which of the following is part of the current sepsis 3 consensus clinical definition of sepsis? Organ dysfunction, procalcitonin, SERS, or shock? So the new current sepsis 3 definition of sepsis is life-threatening organ dysfunction caused by a dysregulated host response to infection. So the element here is organ dysfunction as part of the definition of sepsis. In terms of making the diagnosis of sepsis, organ dysfunction can be identified in this case, maybe the most easy way, is an acute change in total SOFA score of at least 2 points related to the infection that underlies the cause of sepsis. For those patients where a baseline SOFA score is unknown, they can be presumed to have a score of 0 and therefore a score of 2 points would qualify them as having organ dysfunction and therefore potentially sepsis. Septic shock is then defined as hypotension despite adequate fluid resuscitation requiring the use of vasopressors to maintain a mean arterial pressure of at least 65 and having a lactate of at least 2. Patients with septic shock then can be identified as those patients with persistent hypotension requiring vasopressors with the elevated serum lactate. And the reason that definition was chosen, as you can see, is that the mortality was highest, the risk of death was highest for the patients who had all three of those criteria, hypotension, requiring vasopressors, and with a lactate greater than 2. From an epidemiologic perspective, we know that gram-positive bacteria, shown here in blue, are the most common cause of sepsis, with gram-negative bacteria being the next most common, and then fungal infections, in yellow, shown third. The U.S. burden of sepsis as published a few years ago shows that there's probably nearly 2 million people developing sepsis each year and at least 270,000 people dying each year of sepsis. That makes sepsis the third most common cause of death in the United States. And using those numbers, it means that one in every three patients who die in a hospital have sepsis, and there's a substantial amount of additional information from the CDC and from SCCM. The global burden of sepsis is even greater, and it was shown just in the last year that there's approximately 50 million cases of sepsis each year and more than 11 million deaths globally. That makes sepsis the number one cause of death globally. And although the incidence and mortality have declined over time, it still contributes to about 20% of all global deaths each year. And the burden of sepsis, as you might guess, is particularly associated with income and development, with 85% of all sepsis-related deaths in lower and middle-income countries. So there's several ways to think about sepsis management, and you can look at it from the highest level, or you can focus down to the smallest level. So one way to think about that is zooming in. And if you begin to zoom in on the patient with sepsis, you can get right down to the cellular level. This is not just the patient, but now thinking, how do we really treat the patient? What are the factors involved in sepsis for them? And zoom in further and further, not just to the level of the white bud cell, but actually beyond that. So as we think about diagnosing sepsis, one of the challenges in diagnosing sepsis is that it's a very heterogeneous disease. It looks different in every patient. There's different types of infections, respiratory infections, skin infections, brain, GI, and others. There's different organisms. There's different forms of organ dysfunction, and there's a full age spectrum that it occurs in from neonates through the elderly. That heterogeneity makes sepsis incredibly difficult to diagnose. So as we talked about, sepsis may be more difficult to recognize early on at a time when its treatment might actually be easier. But in fact, it becomes more obvious, more evident that sepsis is the cause of illness when it's much more difficult to treat. So sepsis is a syndrome. It's not a specific disease that we can point a finger at or have a diagnostic test for, a single test that tells us that sepsis is present. Which means we rely on the sepsis definition, which includes organ dysfunction as one of the characteristics. But there's two important components of organ dysfunction to remember. One is that sepsis-related organ dysfunction may be occult. So if you have a patient with infection, you should screen and look for organ dysfunction in those patients, because it may not be evident until you look for it. Conversely, if you have a patient with organ dysfunction, there may be unrecognized infection. And therefore, if you have a patient, particularly one who's critically ill with organ dysfunction, you should also look for underlying infection as the cause. And therefore, sepsis management relies on sepsis identification. In order to diagnose sepsis, you have to screen for it, you have to be attentive, and you proactively looking for it. Differentiating sepsis from septic shock determines two important factors, what care is needed, and the urgency of care. There are several key sepsis management strategies, largely which are laid out by the Surviving Sepsis Campaign. The Surviving Sepsis Campaign was launched in 2002 as a global commitment to reducing morbidity and mortality from sepsis and septic shock. It's a collaboration between the Society of Critical Care Medicine and the European Society of Intensive Care Medicine. And at this point, it's progressed through multiple phases of expanding scope with four editions of evidence-based guidelines and the implementation of a truly global performance improvement program. You can see the several editions of the guidelines here, and they continue to evolve as they will again this year. An important sepsis principle is that sepsis and septic shock are medical emergencies, and we recommend that treatment and resuscitation begin immediately. From the first perspective, there's infection diagnosis, and from the Surviving Sepsis Campaign, it specifically says, we recommend that appropriate routine microbiologic cultures, including blood, be obtained before starting antimicrobial therapy in patients with suspected sepsis or septic shock, particularly if doing so results in no substantial delay in start of antimicrobials. That antimicrobial therapy is critically important for treating the underlying infection that causes sepsis. So the recommendation is for the administration of IV antimicrobials as soon as possible and within one hour for both sepsis and septic shock. That should be empiric broad-spectrum therapy with one or more antimicrobials to cover all the likely pathogens. The surviving sepsis campaign recommends that a specific anatomic diagnosis of infection requiring emergent source control be identified or excluded as rapidly as possible. Because while antibiotics are important, antibiotics may not be sufficient to treat sepsis. And source control, meaning surgical or other drainage interventions, may be key. Fluid resuscitation is another cornerstone of sepsis therapy. And the surviving sepsis campaign recommends the resuscitation of patients with sepsis-induced hypotension or hypoperfusion be accomplished with the administration of at least 30 cc's per kilo of intravenous crystalloid given over the first three hours, and ideally initiated and accomplished within the first hour. Fluid resuscitation is specifically chosen to be crystalloid as the fluid of choice because it's most readily available and there's no evidence that other fluids are superior. Specifically, albumin is only suggested to be used in patients where crystalloids are becoming ineffective or require substantial administration. And particularly, the surviving sepsis campaign recommends against the use of hydroxyethyl starch solutions because those have been shown to be harmful in patients with sepsis and septic shock. For patients with persistent hypotension and septic shock, the first choice vasopressor is norepinephrine. And it should be used to target blood pressure. For patients with persistent hypotension and septic shock, the vasopressor of choice is norepinephrine. For those patients who remain hypotensive, the addition of vasopressin or epinephrine may be useful for raising arterial pressure to target. And it should be individualized to the patient depending on cardiac function. Angiotensin II notably was approved in 2018 as an additional adjunct for patients with persistent hypotension. The blood pressure target for septic shock is a mean arterial pressure of 65 derived from predominantly this trial published in 2014, which shows that inpatients who have a mean arterial pressure of 65 do just as well as those patients who had a high target of mean arterial pressure of 85. And furthermore, for patients with septic shock, it's important that further hemodynamic assessments continue to be done so that fluid and vasopressor therapy can be titrated as needed. And particularly as a surviving sepsis campaign suggests the dynamic over static variables we use to predict fluid responsiveness and make decisions for fluid and vasopressor therapy. The goal of hemodynamic resuscitation using these hemodynamic assessments is to use resuscitation to normalize lactate levels. And that's the appropriate target based on this meta-analysis published earlier. One way to think about increased lactate from this algorithm is to assess the imbalance between oxygen demand and oxygen delivery. And think about whether there are things that can be done to decrease oxygen demand, such as reducing workload or providing sedation, or increasing oxygen delivery through improved cardiac output, arterial oxygen content, or hemoglobin. For patients, as you see on the left, if lactate is decreasing, you can consider the resuscitation method effective and continue them as long as lactate remains elevated. For patients without a decrease in lactate, those on the right, it's important to rethink, reassess, and then resolve what the real problem is and ensure that we're not missing something in a critically ill patient as a source of increased lactate. There are several adjunctive therapies for the treatment of sepsis and septic shock. Particularly for septic shock, there is the use of IV hydrocortisone. Usually at a dose of about 200 milligrams per day, which may be added for patients who are persistently hypotensive despite fluid resuscitation and the use of at least a single vasopressor. Also, for patients who are anemic, transfusion is appropriate if their hemoglobin level is less than 7. But otherwise, transfusion of red blood cells is not indicated for otherwise stable patients with sepsis or even septic shock. Another important supportive therapy we think of is for respiratory support, particularly mechanical ventilation, where we follow the same strategies that we do for other critically ill patients, such as those with ARDS. We'd choose a target tidal volume of 6 cc per kilo of predicted body weight and use an upper limit plateau pressure goal of 30 centimeters of water. And we would even suggest using those lower tidal volumes in patients with respiratory failure who do not yet have ARDS, as low tidal volume ventilation can be preventive for the development of ARDS. The other components of supportive therapy for patients requiring mechanical ventilation are to use a conservative fluid strategy, particularly those with sepsis-induced ARDS. Using spontaneous breathing trials to identify those who are ready for weaning and to accelerate the liberation from mechanical ventilation. Use of a ventilator weaning protocol Use of a ventilator weaning protocol and minimizing continuous or intermittent sedation or eliminating it completely beyond specific titration endpoints where possible. All of this falls into the ICU Liberation Bundle. If you're not familiar with the ICU Liberation Bundle, this is a strategy that combines several elements together to accelerate the liberation from mechanical ventilation of critically ill patients. And what you see here is that bundle compliance, as you move to the right on the graph, dramatically increase the likelihood of someone being discharged alive from the ICU. And in fact, experiencing fewer days of delirium, coma and other complications. The other supportive therapies that we think of in critically ill patients with sepsis and septic shock, protocolized blood glucose management to prevent severe hyper and hypoglycemia, venous thromboembolism prophylaxis, stress ulcer prophylaxis when appropriate, the early initiation of internal nutrition, particularly in those patients who can tolerate inner nutrition and avoiding peripheral nutrition in the first seven days of illness. And finally, but maybe most importantly, discussing the goals of care and the prognosis early with patients and families in order to incorporate those into treatment planning. Summarizing what we've talked about today, sepsis and septic shock are medical emergencies where treatment must begin immediately. The goal is to obtain cultures, administer antibiotics and consider the need for diagnostic studies and achieve source control as early as possible, ideally within the first hour or few hours. Administering intravenous fluids and the use of vasopressors early and also according to both patient needs and those hemodynamic assessments to individualize them. And finally, pay close attention to those ancillary supportive therapies and the application of the ICU liberation bundle to accelerate recovery and the liberation from mechanical ventilation. Let's start with the first of two cases. Our first case is a 58-year-old white male with COPD who presents to your emergency department with a two-day history of fever, cough with purulent sputum, dyspnea and right-sided chest pain. On exam, his vital signs show fever and an oxygen saturation of 94%. He's noted to be dyspneic and has ralls in the right lower chest. His laboratory exam shows an elevated white blood cell count, a creatinine of 2.1, lactate of 2.2, a blood gas that you see here and a rapid gram stain of sputum shows gram-positive diplococci. So the first question in the case, this patient has sepsis because, A, he has evidence of an infection with fever and leukocytosis. B, he has infection with a lactate of at least two. C, his primary care doctor called the ED and told the clerk that he was coming and had sepsis. D, he has infection with a creatinine elevation that counts for two SOFA points. Or E, he doesn't have sepsis at all. In this case, the patient has sepsis. Recall that sepsis is an infection causing a dysregulated immune response that causes new organ dysfunction and is most often measured by SOFA score. This patient had an elevated creatinine that qualified for the elevated SOFA score in the presence of infection that therefore met the definition of sepsis. While SIRS criteria may be relevant for sepsis screening and may be early markers of sepsis, they aren't part of the sepsis diagnosis. And elevated lactate is only part of the septic shock diagnosis and doesn't necessarily count for SOFA points. Case number two, a 59-year-old female, postoperative day number eight, after a bowel resection with a primary anastomosis at an outside hospital. Admitted to your ICU from the emergency department with worsening abdominal pain, nausea, and vomiting. Her vitals show tachycardia, hypotension, tachypnea, an oxygen saturation of 94% on four liters, and fever. Her laboratory showed leukocytosis, anemia, an elevated creatinine, elevated AST, as well as a lactate of 3.1. She has a CT scan of the abdomen that shows an anastomotic leak. She has cultures taken, she's given antibiotics, and has now received three liters of Ringer's lactate. In case number two, here's the question. This patient has septic shock because, A, she has infection, hypotension, and a lactate greater than two. B, she has infection with a QSOFA score greater than two. C, the outside hospital called you to tell you that she was coming to you with septic shock. D, she has infection with organ dysfunction that counts for at least two SOFA points. Or E, she doesn't have septic shock. The answer to the question for case number two is that the patient does not have septic shock. Remember, septic shock is sepsis causing hypotension sufficient to require vasopressors to maintain arterial blood pressure of at least 65 despite fluid administration and with a lactate greater than two. So while she was hypotensive and received fluids and she did have an elevated lactate, we did not give you a follow-up blood pressure and we also did not tell you that she was receiving vasopressors. Hypotension alone with an elevated lactate is not septic shock. Remember also that QSOFA is really only a sepsis screening tool, particularly outside the ICU, and it's not used to diagnose sepsis or septic shock. And SOFA is not used to diagnose septic shock. Septic shock is the combination of hypotension and vasopressors in patients who have been adequately fluid resuscitated with an elevated lactate. In conclusion, the sepsis three definition provides the methods to screen and diagnose sepsis focusing on organ dysfunction and illness severity. And early patient identification and rapid treatment improves the likelihood of a patient surviving sepsis. And because this diagnosis of sepsis is not simple, you should screen for organ dysfunction when suspecting sepsis and screen for sepsis when unexplained organ dysfunction is present. Thank you very much for your attention. And I look forward to hearing questions and ideally meeting you all in person. Thank you very much.

sepsis

septic shock

evidence-based management

immune response

organ dysfunction

antibiotics