Medical School · Year 3 · Internal Medicine · includes a discussion video

Seminar 12: Sepsis

Internal Medicine Clerkship

Learning Objectives

By the end of this seminar, students will be able to:

1. Define sepsis and septic shock using the current Sepsis-3 consensus criteria and explain the evolution from prior definitions
2. Describe the pathophysiology of sepsis including the inflammatory cascade, cardiovascular dysfunction, and microvascular injury
3. Apply rapid bedside identification strategies using qSOFA and the full SOFA scoring system to screen for and diagnose sepsis
4. Outline the Hour-1 bundle and principles of initial resuscitation including fluid therapy, antibiotic timing, and hemodynamic targets
5. Select appropriate empiric antibiotic therapy based on suspected infectious source and de-escalate based on culture results
6. Describe vasopressor selection, adjunctive therapies, and the importance of source control in the management of septic shock

Seminar Outline

Section 1: Definitions

The Sepsis-3 definitions, published in 2016 by the Third International Consensus Definitions for Sepsis and Septic Shock, represent the current standard for defining these clinical syndromes. Sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection, emphasizing that sepsis is fundamentally a problem of organ injury rather than simply an exaggerated inflammatory response. Septic shock is a subset of sepsis in which underlying circulatory, cellular, and metabolic abnormalities are profound enough to substantially increase mortality, and it is operationally defined by the requirement for vasopressors to maintain a mean arterial pressure of sixty-five millimeters of mercury or higher combined with a serum lactate level greater than two millimoles per liter despite adequate fluid resuscitation. The key conceptual change from prior definitions is the replacement of the systemic inflammatory response syndrome criteria with the Sequential Organ Failure Assessment score as the primary tool for identifying organ dysfunction in the setting of suspected infection.

The evolution of sepsis definitions reflects the medical community's evolving understanding of the syndrome. The original Sepsis-1 definition, established in 1991, defined sepsis as the presence of two or more systemic inflammatory response syndrome criteria in the setting of suspected or confirmed infection. The Sepsis-2 revision in 2001 expanded the diagnostic criteria but maintained the fundamental framework of SIRS plus infection. The Sepsis-3 definition in 2016 represented a paradigm shift by removing the SIRS requirement entirely and focusing instead on organ dysfunction as the defining feature of sepsis. This change was motivated by the recognition that SIRS criteria are neither sensitive nor specific for sepsis: many hospitalized patients meet SIRS criteria without infection, and some patients with severe sepsis do not meet SIRS criteria.

The systemic inflammatory response syndrome criteria, while no longer required for the diagnosis of sepsis, remain clinically relevant as markers of physiologic stress and are worth reviewing for historical context. The four SIRS criteria are temperature greater than thirty-eight degrees Celsius or less than thirty-six degrees Celsius, heart rate greater than ninety beats per minute, respiratory rate greater than twenty breaths per minute or PaCO2 less than thirty-two millimeters of mercury, and white blood cell count greater than twelve thousand or less than four thousand per microliter or greater than ten percent immature band forms. The presence of two or more criteria constitutes SIRS. The primary limitation of SIRS is its non-specificity, as these physiologic changes can occur in response to a wide variety of non-infectious insults including surgery, trauma, burns, pancreatitis, and autoimmune conditions, making SIRS criteria insufficient as the sole basis for diagnosing sepsis.

Organ dysfunction in sepsis manifests across multiple organ systems, and its recognition is central to the Sepsis-3 definition. Cardiovascular dysfunction presents as hypotension requiring vasopressor support to maintain adequate perfusion pressure, reflecting the distributive shock physiology of sepsis. Respiratory dysfunction manifests as hypoxemia and may progress to acute respiratory distress syndrome, with PaO2/FiO2 ratio serving as the quantitative measure of severity. Renal dysfunction presents as acute kidney injury with rising creatinine and diminished urine output. Hepatic dysfunction is reflected by elevated bilirubin levels. Hematologic dysfunction manifests as thrombocytopenia and may progress to disseminated intravascular coagulation with consumptive coagulopathy. Neurologic dysfunction presents as altered mental status, ranging from confusion to obtundation and coma, and reflects the effects of systemic inflammation, hypoperfusion, and metabolic derangements on cerebral function.

<image>Panel A: Timeline showing the evolution of sepsis definitions from Sepsis-1 (1991, SIRS plus infection) through Sepsis-2 (2001, expanded criteria) to Sepsis-3 (2016, organ dysfunction focus), with key conceptual changes highlighted at each transition. Panel B: Venn diagram comparing the Sepsis-3 definitions of sepsis (organ dysfunction from dysregulated host response to infection) and septic shock (sepsis requiring vasopressors for MAP 65 or higher plus lactate greater than 2 despite fluids), with mortality rates annotated. Panel C: Display of the four SIRS criteria (temperature, heart rate, respiratory rate, WBC) with their thresholds, accompanied by a sensitivity-specificity analysis showing why SIRS was replaced. Panel D: Six-organ system diagram showing manifestations of organ dysfunction in sepsis: cardiovascular (hypotension), respiratory (hypoxemia and ARDS), renal (AKI), hepatic (elevated bilirubin), hematologic (thrombocytopenia and DIC), and neurologic (altered mental status).</image>

Section 2: Pathophysiology

The pathophysiology of sepsis begins with the recognition of invading pathogens by the innate immune system. Pathogen-associated molecular patterns, which are conserved molecular structures on microbial surfaces such as lipopolysaccharide on gram-negative bacteria and lipoteichoic acid on gram-positive bacteria, are recognized by pattern recognition receptors including toll-like receptors on the surface of immune cells. This recognition triggers the activation of intracellular signaling cascades that lead to the production and release of pro-inflammatory cytokines, including interleukin-1, interleukin-6, and tumor necrosis factor-alpha. In sepsis, this inflammatory response becomes dysregulated and excessive, with a systemic rather than localized cytokine release that produces widespread tissue injury. Simultaneously, damage-associated molecular patterns released from injured host cells further amplify the inflammatory cascade, creating a self-perpetuating cycle of inflammation and tissue damage. A subsequent immunosuppressive phase may follow, characterized by lymphocyte apoptosis, monocyte deactivation, and impaired adaptive immunity, which predisposes to secondary nosocomial infections.

The cardiovascular effects of sepsis are central to the hemodynamic derangement that characterizes septic shock. Vasodilation, mediated by nitric oxide and other inflammatory mediators acting on vascular smooth muscle, is the hallmark of septic shock and produces a distributive shock physiology with low systemic vascular resistance. Myocardial depression, mediated by circulating cytokines including tumor necrosis factor and interleukins, results in reduced ventricular contractility and biventricular dilation, although cardiac output is often initially maintained or elevated due to compensatory tachycardia. Capillary leak, caused by endothelial dysfunction and breakdown of endothelial barrier integrity, leads to extravasation of fluid and protein into the interstitial space, contributing to tissue edema and intravascular volume depletion. The net hemodynamic effect is a distributive shock pattern characterized by maldistribution of blood flow, with some tissue beds receiving excessive flow while others are hypoperfused, resulting in regional tissue hypoxia despite apparently adequate global hemodynamic parameters.

Microvascular dysfunction is increasingly recognized as a critical mechanism of organ injury in sepsis, operating independently of macrovascular hemodynamics. Endothelial injury from inflammatory mediators disrupts the glycocalyx layer, increases permeability, and promotes leukocyte adhesion and transmigration, perpetuating local tissue inflammation. Microthrombosis develops as the procoagulant state of sepsis causes fibrin deposition in small vessels, obstructing microvascular blood flow and creating regional tissue hypoxia. Arteriovenous shunting, in which blood bypasses the capillary bed through pathological shunt vessels, further reduces oxygen delivery to metabolically active tissues. Mitochondrial dysfunction, termed cytopathic hypoxia, impairs cellular oxygen utilization even when oxygen delivery is adequate, representing a fundamental defect in aerobic metabolism that contributes to organ failure through energy depletion and accumulation of metabolic waste products including lactate.

The coagulation cascade is profoundly disturbed in sepsis, shifting toward a procoagulant state that promotes microvascular thrombosis and organ injury. Tissue factor, released from activated monocytes and endothelial cells, initiates the extrinsic coagulation pathway and generates thrombin, which in turn activates platelets and converts fibrinogen to fibrin. Simultaneously, the natural anticoagulant pathways are suppressed: protein C and antithrombin levels fall due to consumption, impaired synthesis, and degradation by neutrophil elastase. Fibrinolysis is inhibited by increased production of plasminogen activator inhibitor-1, preventing the breakdown of microvascular thrombi. The culmination of these procoagulant, anti-fibrinolytic, and anticoagulant-depleted states is disseminated intravascular coagulation, a consumptive coagulopathy characterized by simultaneous widespread microthrombosis and bleeding due to consumption of clotting factors and platelets. Disseminated intravascular coagulation is a marker of severe sepsis and is associated with significantly increased mortality.

<image>Panel A: Illustration of the innate immune recognition pathway showing pathogen-associated molecular patterns (PAMPs) on bacterial surfaces binding to toll-like receptors (TLRs) on macrophages and neutrophils, triggering intracellular signaling and release of pro-inflammatory cytokines (IL-1, IL-6, TNF-alpha) into the systemic circulation. Panel B: Cardiovascular effects diagram showing vasodilation from nitric oxide causing decreased systemic vascular resistance, myocardial depression from cytokines reducing contractility, and capillary leak from endothelial dysfunction causing interstitial edema, all converging on distributive shock physiology. Panel C: Microvascular dysfunction cross-section showing injured endothelium with glycocalyx disruption, fibrin microthrombi occluding capillaries, arteriovenous shunting bypassing the capillary bed, and mitochondria with impaired oxidative phosphorylation (cytopathic hypoxia). Panel D: Coagulation cascade diagram in sepsis showing tissue factor activation of the extrinsic pathway, depletion of protein C and antithrombin, increased PAI-1 inhibiting fibrinolysis, and the endpoint of DIC with simultaneous microthrombosis and consumptive bleeding.</image>

Section 3: Identification and Screening

The quick Sequential Organ Failure Assessment score, known as qSOFA, is a bedside screening tool designed for rapid identification of patients with suspected infection who are at high risk for poor outcomes. The three criteria are respiratory rate of twenty-two or more breaths per minute, altered mental status (defined as a Glasgow Coma Scale score of less than fifteen), and systolic blood pressure of one hundred millimeters of mercury or less, each scored as one point. A qSOFA score of two or more should prompt clinicians to suspect sepsis and initiate a comprehensive evaluation including full SOFA scoring, lactate measurement, and blood cultures. The qSOFA is intended as a bedside screening tool for use outside the intensive care unit, where its simplicity allows rapid assessment without laboratory data. However, it has moderate sensitivity, and clinicians should not rely on qSOFA alone to exclude sepsis in patients with clinical features concerning for infection with organ dysfunction.

The full SOFA score provides a comprehensive, quantitative assessment of organ dysfunction across six organ systems and is the foundation of the Sepsis-3 definition. The respiratory component uses the PaO2/FiO2 ratio, scored from zero for a ratio of four hundred or greater to four for a ratio of less than one hundred with mechanical ventilation. The coagulation component uses the platelet count, scored from zero for one hundred fifty thousand or greater to four for less than twenty thousand. The hepatic component uses bilirubin levels, scored from zero for less than 1.2 milligrams per deciliter to four for greater than twelve milligrams per deciliter. The cardiovascular component uses mean arterial pressure and vasopressor requirements, ranging from zero for a MAP of seventy or greater to four for high-dose dopamine, epinephrine, or norepinephrine. The neurologic component uses the Glasgow Coma Scale, and the renal component uses creatinine levels or urine output. Sepsis is defined as a suspected infection with an acute change of two or more points in the SOFA score from the patient's baseline.

Clinical recognition of sepsis requires a high index of suspicion and awareness of the varied presenting features. Fever is the most common presenting sign and is a strong indicator of infection, although hypothermia (temperature below thirty-six degrees Celsius) is also recognized in sepsis and may indicate a more severe immune dysregulation with a worse prognosis. Tachycardia is an early compensatory response to vasodilation and hypovolemia. Tachypnea may reflect direct respiratory involvement, metabolic acidosis with compensatory hyperventilation, or both, and is one of the strongest bedside predictors of clinical deterioration. Hypotension is a late sign that indicates failure of compensatory mechanisms and the development of overt shock. Altered mental status, ranging from subtle confusion to frank delirium or obtundation, reflects the effects of systemic inflammation, hypoperfusion, and metabolic derangements on cerebral function. Lactate elevation, even in the absence of overt hypotension, is a critically important marker of tissue hypoperfusion and occult shock.

Serum lactate has emerged as one of the most important biomarkers in the identification and management of sepsis. A lactate level below two millimoles per liter is considered normal and indicates adequate tissue perfusion in most clinical contexts. Intermediate lactate levels between two and four millimoles per liter indicate an increased risk of adverse outcomes and warrant close monitoring and aggressive evaluation for the source of infection. A lactate level above four millimoles per liter is associated with high mortality and mandates immediate aggressive resuscitation including intravenous fluid bolus and empiric antibiotics per the Hour-1 bundle. Serial lactate measurements are essential for assessing the adequacy of resuscitation, with a target of greater than ten percent clearance within the first six hours serving as a prognostic marker of improving tissue perfusion. Lactate clearance, defined as the percentage decrease in lactate from baseline to a subsequent measurement, has been shown to correlate with improved outcomes and can guide ongoing fluid and vasopressor therapy.

<image>Panel A: qSOFA scoring card showing the three bedside criteria (respiratory rate 22 or more, altered mental status, systolic BP 100 or less) with point assignments, and the decision pathway: score of 2 or more triggers full sepsis evaluation with SOFA scoring, lactate, and cultures. Panel B: Complete SOFA score grid with six organ systems (respiratory, coagulation, hepatic, cardiovascular, neurologic, renal) displayed in rows, with scoring parameters from 0 to 4 in columns, showing threshold values for each level and the definition of sepsis as an acute change of 2 or more points. Panel C: Clinical recognition panel showing six bedside features of sepsis: fever or hypothermia on thermometer, tachycardia on cardiac monitor, tachypnea with respiratory rate counter, hypotension on blood pressure display, altered mental status depicted as confused patient, and lactate elevation on point-of-care test. Panel D: Lactate interpretation ladder showing normal (less than 2), intermediate risk (2-4), and high risk (greater than 4) thresholds with corresponding mortality rates, alongside a graph of serial lactate clearance over 6 hours demonstrating the greater than 10% clearance target.</image>

Section 4: Initial Resuscitation

The Hour-1 bundle represents the current standard for initial sepsis resuscitation and emphasizes the time-critical nature of early intervention. The five elements of the bundle, all to be initiated within one hour of sepsis recognition, are as follows: measure serum lactate and remeasure if the initial level is greater than two millimoles per liter; obtain blood cultures from two separate sites before antibiotic administration; administer broad-spectrum antibiotics directed at the most likely source; begin rapid infusion of thirty milliliters per kilogram of crystalloid fluid if the patient is hypotensive or has a lactate level of four or greater; and apply vasopressors if hypotension persists after initial fluid resuscitation to maintain a mean arterial pressure of sixty-five millimeters of mercury or higher. The bundle framework emphasizes parallel rather than sequential implementation of these elements, meaning that all five components should be initiated simultaneously rather than in a stepwise fashion. The one-hour time frame is aspirational and reflects the evidence that delays in each element, particularly antibiotics and fluids, are independently associated with increased mortality.

Fluid resuscitation is the cornerstone of initial hemodynamic management in sepsis and addresses the relative and absolute intravascular volume depletion that results from vasodilation, capillary leak, and insensible losses. Crystalloid solutions are the preferred first-line resuscitation fluid, with balanced crystalloid solutions such as lactated Ringer solution and Plasmalyte preferred over normal saline because they cause less hyperchloremic metabolic acidosis and may be associated with improved renal outcomes and reduced mortality. The initial fluid bolus of thirty milliliters per kilogram of ideal body weight should be administered within the first three hours, which translates to approximately two to three liters for most adults. Response to fluid resuscitation should be assessed using a combination of clinical parameters including blood pressure, heart rate, urine output, and lactate clearance, as well as dynamic measures of fluid responsiveness such as pulse pressure variation, passive leg raise testing, and point-of-care echocardiography. After the initial bolus, additional fluid administration should be guided by ongoing assessment of fluid responsiveness to avoid fluid overload, which is associated with worse outcomes.

The choice of resuscitation fluid has been the subject of significant study and ongoing debate. Balanced crystalloid solutions, including lactated Ringer solution and Plasmalyte, have a chloride content and strong ion difference that more closely approximate plasma physiology and cause less hyperchloremic metabolic acidosis than normal saline. Normal saline contains a supraphysiologic chloride concentration of one hundred fifty-four millimoles per liter, which can contribute to acidosis, renal vasoconstriction, and potentially worse clinical outcomes when administered in large volumes. Albumin may be considered as an adjunctive colloid solution in patients who have received large volumes of crystalloid without adequate hemodynamic response, although evidence supporting a mortality benefit is limited. Hydroxyethyl starch solutions are contraindicated in sepsis due to evidence of increased acute kidney injury and mortality. The current consensus favors balanced crystalloid solutions as the first-line fluid choice, with the recognition that the total volume of resuscitation is likely more important than the specific type of crystalloid.

The endpoints of resuscitation guide the clinician in determining when adequate tissue perfusion has been restored and when further fluid administration is unlikely to be beneficial. The primary hemodynamic target is a mean arterial pressure of sixty-five millimeters of mercury or higher, which is considered the minimum perfusion pressure necessary to maintain adequate organ blood flow. A decreasing serum lactate level is one of the most reliable indicators of improving tissue perfusion and should be monitored serially every two to four hours during the resuscitation phase. Urine output of at least 0.5 milliliters per kilogram per hour provides a real-time indicator of renal perfusion and is easily monitored with a urinary catheter. Central venous oxygen saturation of seventy percent or higher, when available through a central venous catheter, reflects the balance between systemic oxygen delivery and consumption and indicates adequate global perfusion. Skin perfusion, assessed by capillary refill time and skin mottling, provides a simple bedside measure of peripheral perfusion that correlates with clinical outcomes.

<image>Panel A: Hour-1 bundle infographic displayed as a clock face with five elements positioned around the dial: measure lactate, obtain blood cultures, administer broad-spectrum antibiotics, begin 30 mL/kg crystalloid, and start vasopressors if needed, all within 60 minutes, with evidence-based mortality impact data for each element. Panel B: Fluid resuscitation diagram showing crystalloid bags (lactated Ringer and Plasmalyte preferred over normal saline), 30 mL/kg initial bolus calculation, and response assessment using blood pressure, heart rate, urine output, lactate clearance, and dynamic measures (passive leg raise, pulse pressure variation). Panel C: Comparison of balanced crystalloid versus normal saline showing chloride content, pH, and physiologic effects, with albumin as adjunctive option and hydroxyethyl starch marked as contraindicated, with outcome data from key clinical trials. Panel D: Endpoints of resuscitation dashboard showing target MAP of 65 or higher on arterial line tracing, decreasing lactate trend on serial measurements, urine output of 0.5 mL/kg/hr on hourly chart, central venous oxygen saturation of 70% or higher, and capillary refill time assessment on fingertip.</image>

Section 5: Antibiotic Therapy

The principles of antibiotic therapy in sepsis emphasize early, broad-spectrum, source-directed treatment with subsequent de-escalation. Timing is the single most critical factor: antibiotics should be administered within one hour of sepsis recognition, as each hour of delay in antibiotic administration has been associated with an approximately seven percent increase in mortality. The initial empiric regimen should provide broad-spectrum coverage against the most likely pathogens based on the suspected source of infection, the patient's comorbidities, prior microbiology data, and local resistance patterns. Source-directed therapy means selecting antibiotics that specifically target the pathogens most commonly associated with the suspected infection site, while providing adequate tissue penetration to the affected organ. De-escalation should occur as soon as culture and susceptibility results are available, typically within forty-eight to seventy-two hours, to narrow the antibiotic spectrum, reduce the risk of resistance emergence, and minimize adverse effects. The typical duration of antibiotic therapy in sepsis is seven to ten days, although this may be shorter for some infections and longer for others depending on the source, pathogen, and clinical response.

Empiric antibiotic selection is guided by the suspected source of infection, which should be identified through careful clinical assessment. For sepsis of unknown source, a common empiric regimen is vancomycin combined with piperacillin-tazobactam, providing coverage against gram-positive organisms including methicillin-resistant Staphylococcus aureus, gram-negative organisms including Pseudomonas, and anaerobes. For suspected pulmonary sources, ceftriaxone combined with azithromycin provides coverage against typical and atypical community-acquired pneumonia pathogens, while broader coverage with an anti-pseudomonal beta-lactam is appropriate for hospital-acquired or ventilator-associated pneumonia. For suspected abdominal sources, piperacillin-tazobactam or a carbapenem provides broad coverage against enteric gram-negative organisms and anaerobes. For urinary tract sources, ceftriaxone or a fluoroquinolone provides appropriate coverage, while broader coverage may be needed for complicated or healthcare-associated urinary tract infections. For skin and soft tissue sources, vancomycin combined with piperacillin-tazobactam covers both gram-positive organisms including MRSA and polymicrobial infections including anaerobes. For catheter-related infections, vancomycin combined with gram-negative coverage addresses the most common pathogens.

Special considerations in antibiotic selection require attention to patient-specific and pathogen-specific factors. Neutropenic patients require empiric anti-pseudomonal coverage due to the high risk of Pseudomonas bacteremia in the setting of profound immunosuppression. Patients at risk for methicillin-resistant Staphylococcus aureus, including those with prior MRSA colonization, recent hospitalization, intravenous drug use, or hemodialysis access, should receive vancomycin or an alternative anti-MRSA agent. When Pseudomonas is a concern, initial double gram-negative coverage with two agents from different classes may be considered to maximize the likelihood of providing at least one effective agent until susceptibility results are available. Patients at risk for Clostridioides difficile infection should avoid fluoroquinolones when possible, as these agents are particularly associated with C. difficile colonization and infection. Recent antibiotic exposure within the preceding ninety days should be considered, as it increases the likelihood of infection with resistant organisms, and the prior antibiotic class should ideally be avoided in the empiric regimen.

De-escalation of antibiotic therapy is a critical component of antibiotic stewardship in sepsis and should be pursued systematically once culture data become available. Culture-directed therapy involves narrowing the antibiotic regimen to the most targeted effective agent based on susceptibility testing, which reduces unnecessary broad-spectrum exposure and the associated risks of resistance selection and adverse effects. If cultures are negative for MRSA and the patient is clinically improving, vancomycin should be discontinued. If cultures do not grow Pseudomonas, dual gram-negative coverage should be simplified to a single agent. The duration of therapy may be guided by procalcitonin trends, with decreasing procalcitonin levels supporting discontinuation of antibiotics, thereby avoiding unnecessary prolongation of treatment. Clinical improvement, assessed by resolution of fever, normalization of white blood cell count, improving organ function, and lactate clearance, should inform decisions about when to transition from intravenous to oral antibiotics and when to discontinue therapy.

<image>Panel A: Central diagram showing antibiotic timing impact with a clock and mortality curve demonstrating the approximately 7% increase in mortality per hour of delay, surrounded by the principles of empiric therapy: broad-spectrum, source-directed, culture-guided de-escalation, and 7-10 day typical duration. Panel B: Source-based empiric antibiotic selection matrix showing six suspected sources (unknown, pulmonary, abdominal, urinary, skin/soft tissue, line-related) with corresponding recommended regimens and targeted pathogens for each source. Panel C: Special considerations panel showing neutropenic patient (anti-pseudomonal required), MRSA risk factors checklist (prior colonization, hospitalization, IVDU, hemodialysis), Pseudomonas risk (double coverage), and C. difficile risk (avoid fluoroquinolones). Panel D: De-escalation flowchart showing initial broad empiric coverage transitioning to culture-directed narrow therapy at 48-72 hours, with decision nodes for MRSA negative (stop vancomycin), no Pseudomonas (single agent), and procalcitonin-guided duration assessment.</image>

Section 6: Vasopressor Therapy

Vasopressor therapy is indicated when hypotension persists despite adequate fluid resuscitation, with the goal of maintaining a mean arterial pressure of sixty-five millimeters of mercury or higher to ensure adequate organ perfusion. Vasopressors should not be delayed while completing the full thirty milliliters per kilogram fluid bolus if the patient remains hemodynamically unstable, as early initiation of vasopressors in combination with ongoing fluid resuscitation may improve outcomes by restoring perfusion pressure more rapidly. The decision to initiate vasopressors should be based on clinical assessment of the patient's hemodynamic status, including blood pressure, heart rate, mental status, urine output, and lactate levels. Arterial line placement for continuous blood pressure monitoring is recommended when vasopressors are initiated, although initiation should not be delayed for placement of invasive monitoring.

Norepinephrine is the recommended first-line vasopressor in septic shock based on evidence demonstrating superior efficacy and safety compared to other agents. The primary mechanism of norepinephrine is alpha-1 adrenergic receptor agonism, which produces potent arterial vasoconstriction and raises systemic vascular resistance, directly counteracting the pathologic vasodilation of septic shock. Norepinephrine also has mild beta-1 adrenergic activity, providing modest inotropic support to augment cardiac output. The initial dose is typically two to five micrograms per minute, titrated upward in increments of two to five micrograms per minute to achieve the target mean arterial pressure. Norepinephrine is preferred over dopamine as the first-line agent because it is associated with fewer arrhythmias and lower mortality. Central venous access is preferred for vasopressor administration due to the risk of tissue necrosis with peripheral extravasation, but peripheral administration through a large-bore proximal vein is acceptable as a temporary measure when central access is not immediately available.

Second-line vasopressor agents are added when norepinephrine alone is insufficient to maintain the target mean arterial pressure. Vasopressin at a fixed dose of 0.03 units per minute is the recommended second-line agent, added to norepinephrine rather than used as a replacement. Vasopressin acts on V1 receptors on vascular smooth muscle to produce vasoconstriction through a non-adrenergic mechanism, making it effective in the setting of catecholamine-resistant vasodilation. Adding vasopressin may allow dose reduction of norepinephrine, potentially decreasing the adverse effects of high-dose catecholamine therapy. Epinephrine is considered when additional inotropic support is needed beyond what norepinephrine provides, as it has potent beta-1 activity in addition to alpha-1 and beta-2 effects. Dopamine has largely fallen out of favor as a first- or second-line agent due to evidence of more arrhythmias and higher mortality compared to norepinephrine, although it may be considered in selected patients with significant bradycardia. Phenylephrine is a pure alpha-1 agonist that may be appropriate in patients with tachyarrhythmias where additional beta-adrenergic stimulation is undesirable.

Vasopressor titration and monitoring require careful attention to hemodynamic targets, perfusion markers, and potential complications. Central venous access is preferred for vasopressor administration to minimize the risk of peripheral tissue necrosis from extravasation, although temporary peripheral administration through a large-bore proximal vein is acceptable for brief periods. The target mean arterial pressure is sixty-five millimeters of mercury or higher, although higher targets may be considered in patients with chronic hypertension. A key principle is to add a second vasopressor agent at a moderate dose rather than escalating a single agent to maximum dose, as this multi-agent approach may improve hemodynamics through complementary mechanisms while minimizing dose-dependent adverse effects. Continuous monitoring should include arterial blood pressure, heart rate, lactate trends, urine output, and clinical signs of perfusion including mental status, skin temperature, and capillary refill time. Vasopressor doses should be weaned as clinically tolerated once the patient demonstrates sustained hemodynamic stability and improving markers of tissue perfusion.

<image>Panel A: Vasopressor initiation criteria showing persistent hypotension despite fluid resuscitation (MAP below 65), with notation that vasopressors should not be delayed for fluid completion and that early initiation improves outcomes, displayed alongside clinical assessment parameters. Panel B: Norepinephrine pharmacology diagram showing alpha-1 receptor binding on vascular smooth muscle (vasoconstriction) and mild beta-1 receptor activity on cardiac myocytes (inotropy), with dose range, titration increments, and superiority data over dopamine. Panel C: Second-line vasopressor comparison showing vasopressin (V1 receptor, 0.03 units/min, catecholamine-sparing), epinephrine (beta-1 and alpha-1, added inotropy), dopamine (more arrhythmias, limited role), and phenylephrine (pure alpha-1, for tachyarrhythmia), with mechanism diagrams for each agent. Panel D: Vasopressor titration and monitoring display showing arterial line tracing with MAP target of 65, multi-agent approach concept (add second agent rather than maximize single agent), and monitoring parameters: continuous BP, heart rate, serial lactate, hourly urine output, and peripheral perfusion assessment.</image>

Section 7: Source Control

Source control is a critical component of sepsis management that refers to the identification and elimination of the anatomical focus of infection through physical means. Antibiotics alone are often insufficient to resolve sepsis when the infectious source involves an undrained collection, devitalized tissue, an infected device, or an obstructed system that perpetuates bacterial growth and systemic inflammation. The principle of source control is to remove or drain the infectious focus as rapidly and completely as possible while minimizing the physiologic insult of the procedure itself. Source control should ideally be achieved within six to twelve hours of sepsis identification when feasible, as delays in source control are associated with increased mortality. The specific intervention required depends on the nature and location of the infectious focus and must be individualized to the clinical scenario.

Common infectious sources and their corresponding interventions illustrate the range of source control procedures encountered in clinical practice. Intra-abdominal abscess requires drainage, which may be accomplished percutaneously under imaging guidance by interventional radiology or surgically if percutaneous drainage is not feasible or adequate. Cholangitis from biliary obstruction requires decompression, typically through endoscopic retrograde cholangiopancreatography with sphincterotomy and stone extraction, or percutaneous transhepatic cholangiography when endoscopic access is not possible. Bowel perforation with peritonitis is a surgical emergency requiring operative repair or resection of the perforated segment. Necrotizing fasciitis requires emergent and aggressive surgical debridement, often requiring multiple returns to the operating room for serial debridement until all necrotic tissue is removed. Infected intravascular catheters should be removed, with the catheter tip sent for culture. Empyema requires chest tube drainage, which may be supplemented by intrapleural fibrinolytics or surgical decortication.

Imaging studies play a critical role in identifying the source of infection when the clinical examination alone is insufficient. A chest radiograph can identify pneumonia, pleural effusion, and empyema as potential pulmonary sources. Computed tomography of the abdomen and pelvis is the most important study for evaluating intra-abdominal sources, including abscess, cholecystitis, appendicitis, diverticulitis, and bowel perforation, and can simultaneously provide guidance for percutaneous drainage procedures. Ultrasound is useful for evaluating the gallbladder for acute cholecystitis and for identifying superficial abscesses. Computed tomography of the chest can better characterize empyema, lung abscess, and mediastinal collections. Echocardiography, particularly transesophageal echocardiography, is essential for evaluating suspected endocarditis and should be obtained in patients with bacteremia from organisms commonly associated with endocarditis, such as Staphylococcus aureus, or in patients with persistent bacteremia despite appropriate antibiotics.

Certain clinical situations require delayed or modified source control approaches. Infected pancreatic necrosis is best managed with delayed intervention, typically four or more weeks after the onset of pancreatitis, to allow the collection to mature and wall off, facilitating a less invasive step-up approach with percutaneous drainage followed by minimally invasive necrosectomy if needed. In patients with severe hemodynamic instability, stabilization with fluids, vasopressors, and antibiotics should take priority before undertaking a major surgical procedure, although emergent situations such as bowel perforation with free air may necessitate immediate operative intervention regardless of hemodynamic status. When the source of infection is unclear despite thorough clinical evaluation and initial imaging, additional diagnostic studies or empiric broadening of antibiotic coverage may be necessary while continuing to evaluate for an occult source. Multidisciplinary collaboration between intensivists, surgeons, interventional radiologists, and infectious disease specialists is essential for optimal source control decision-making.

<image>Panel A: Central concept diagram showing the principle of source control: identifying and eliminating the anatomical focus of infection (undrained collection, devitalized tissue, infected device, obstructed system) within 6-12 hours of sepsis identification, with evidence for timing impact on mortality. Panel B: Six-panel illustration of common source control interventions: percutaneous abscess drainage (CT-guided needle in abdominal collection), ERCP for cholangitis (endoscope with sphincterotomy), surgical repair of bowel perforation (operative field), surgical debridement of necrotizing fasciitis (wound after debridement), infected catheter removal (central line extraction), and chest tube placement for empyema. Panel C: Imaging modality guide showing chest radiograph (pneumonia, effusion), CT abdomen/pelvis (abscess, perforation, cholecystitis), ultrasound (gallbladder, superficial abscess), CT chest (empyema, lung abscess), and echocardiography (endocarditis vegetations), with corresponding clinical indications. Panel D: Decision framework for delayed or modified source control showing pancreatic necrosis (delay 4+ weeks), hemodynamic instability (stabilize first except for emergent surgical conditions), unclear source (additional imaging and empiric broadening), and multidisciplinary collaboration.</image>

Section 8: Adjunctive Therapies

Corticosteroid therapy in septic shock is indicated for patients with refractory hemodynamic instability despite adequate fluid resuscitation and vasopressor therapy. The recommended regimen is hydrocortisone at a total daily dose of two hundred milligrams, typically administered as fifty milligrams intravenously every six hours or as a continuous infusion. The primary benefit of corticosteroids in septic shock is faster resolution of shock, with earlier weaning of vasopressors compared to placebo, which reduces the total vasopressor exposure and its associated adverse effects. However, the evidence regarding mortality benefit remains controversial and uncertain, with some studies showing no significant reduction in overall mortality while others suggest a modest benefit in the most severely ill patients. The indication for corticosteroids is typically defined as ongoing vasopressor requirement despite adequate fluid resuscitation, and most clinicians initiate steroids when moderate to high doses of norepinephrine are required for more than six to twelve hours. Corticosteroids should be weaned gradually once vasopressors are discontinued to avoid rebound hemodynamic instability.

Blood product transfusion in sepsis follows a restrictive strategy to minimize transfusion-related complications while maintaining adequate oxygen-carrying capacity. Red blood cell transfusion is recommended when the hemoglobin level falls below seven grams per deciliter, with a target hemoglobin of seven to nine grams per deciliter, unless the patient has active bleeding, acute coronary syndrome, or other conditions that warrant a higher hemoglobin threshold. Platelet transfusion is indicated when the platelet count falls below ten thousand per microliter to reduce the risk of spontaneous bleeding, or below twenty thousand per microliter in patients with significant bleeding risk such as those requiring invasive procedures. Fresh frozen plasma should be administered in the setting of active bleeding with coagulopathy or prior to invasive procedures when the international normalized ratio is significantly prolonged. The restrictive approach to transfusion in sepsis is supported by evidence demonstrating that liberal transfusion strategies do not improve outcomes and may increase complications including transfusion-related acute lung injury and immunomodulation.

Glucose management in sepsis targets a moderate glycemic range to avoid both hyperglycemia-related immune dysfunction and the significant dangers of hypoglycemia. The recommended target blood glucose range is one hundred forty to one hundred eighty milligrams per deciliter, maintained with an insulin infusion when needed. Tight glucose control targeting eighty to one hundred ten milligrams per deciliter is no longer recommended based on evidence from the NICE-SUGAR trial demonstrating increased mortality with tight control, primarily due to the increased incidence of severe hypoglycemia. Hypoglycemia in critically ill patients is independently associated with increased mortality and neurologic injury, making avoidance of hypoglycemia a higher priority than achieving normoglycemia. Stress hyperglycemia is common in sepsis due to cortisol-mediated insulin resistance, catecholamine-induced glycogenolysis, and hepatic gluconeogenesis, and moderate hyperglycemia may represent an adaptive response that should be treated but not aggressively corrected.

Additional supportive care measures in sepsis address the multiple complications and physiologic derangements that occur in critically ill patients. Deep venous thrombosis prophylaxis with low-molecular-weight heparin or unfractionated heparin should be initiated in all patients without active bleeding or severe thrombocytopenia, as immobilized, critically ill patients are at extremely high risk for venous thromboembolism. Stress ulcer prophylaxis with a proton pump inhibitor or histamine-2 receptor antagonist is recommended for patients at risk of gastrointestinal bleeding, including those requiring mechanical ventilation or those with coagulopathy. Early enteral nutrition, initiated within twenty-four to forty-eight hours of intensive care unit admission, is preferred over parenteral nutrition and is associated with improved outcomes including reduced infectious complications, shorter hospital stay, and better preservation of gut mucosal integrity. Sedation in mechanically ventilated patients should be minimized using sedation protocols, daily sedation interruption, and assessment of readiness for extubation to reduce the duration of mechanical ventilation and the risk of ventilator-associated complications. Lung-protective ventilation with low tidal volumes of six milliliters per kilogram of predicted body weight and plateau pressure limitation is mandatory for all patients with acute respiratory distress syndrome.

<image>Panel A: Corticosteroid therapy decision tree showing indication (refractory shock despite fluids and vasopressors), recommended regimen (hydrocortisone 200 mg/day IV), expected benefit (faster shock reversal, earlier vasopressor weaning), controversy (uncertain mortality benefit), and gradual weaning after vasopressor discontinuation. Panel B: Blood product transfusion thresholds displayed as hemoglobin gauge (transfuse below 7 g/dL), platelet count gauge (transfuse below 10,000 or 20,000 with bleeding risk), and FFP indication (active bleeding with coagulopathy), with evidence summary for restrictive strategy superiority. Panel C: Glucose management target range illustrated as a glucometer display showing the 140-180 mg/dL target zone with red danger zones for hypoglycemia (below 70, increased mortality) and severe hyperglycemia (above 200, immune dysfunction). Panel D: Supportive care checklist showing DVT prophylaxis (LMWH or UFH syringe), stress ulcer prophylaxis (PPI or H2RA), early enteral nutrition (feeding tube and formula), minimal sedation (sedation protocol with daily wake-up), and lung-protective ventilation (ventilator with 6 mL/kg tidal volume setting).</image>

Section 9: Monitoring and Reassessment

Clinical monitoring parameters in sepsis require frequent reassessment to detect deterioration, guide ongoing therapy, and assess response to treatment. Vital signs, including heart rate, blood pressure (preferably via arterial line in patients on vasopressors), respiratory rate, oxygen saturation, and temperature, should be monitored continuously in the acute phase of sepsis management. Mental status should be assessed at least hourly, as changes in consciousness are among the earliest and most sensitive indicators of worsening organ perfusion or the development of septic encephalopathy. Urine output should be monitored hourly via an indwelling urinary catheter, as it provides a real-time indicator of renal perfusion and is one of the most accessible markers of end-organ blood flow. Peripheral perfusion should be assessed regularly by evaluating skin temperature, color, mottling, and capillary refill time, as these simple bedside findings provide information about microvascular blood flow that complements macrovascular hemodynamic parameters.

Laboratory monitoring in sepsis requires serial assessment of biomarkers and organ function tests to track disease progression and response to therapy. Serum lactate should be measured every two to four hours during the initial resuscitation phase and until the level normalizes or shows a consistent downward trend, as lactate clearance is one of the most reliable prognostic markers in sepsis. Serum creatinine should be monitored daily to assess for the development or progression of acute kidney injury, which is one of the most common organ dysfunctions in sepsis. Liver function tests should be monitored daily to assess for hepatic dysfunction, which may affect drug metabolism and indicate worsening systemic illness. Complete blood count should be monitored daily to track white blood cell response to therapy, hemoglobin trends, and platelet count, the latter being important for monitoring for disseminated intravascular coagulation. Coagulation studies, including prothrombin time, partial thromboplastin time, fibrinogen, and D-dimer, should be obtained when disseminated intravascular coagulation is suspected, as consumptive coagulopathy can significantly impact management decisions.

Response assessment in sepsis guides decisions about continuing, modifying, or escalating the treatment plan. A patient who is improving will demonstrate resolving fever, decreasing vasopressor requirements, improving urine output, clearing lactate, and stabilizing or improving organ function parameters. For the improving patient, the appropriate action is to continue the current therapy while beginning to de-escalate antibiotics based on culture results and wean vasopressors and supplemental oxygen as tolerated. A patient who is not improving at forty-eight to seventy-two hours should prompt a systematic reassessment of the diagnosis, antibiotic coverage, fluid status, and source control. The key questions include: Is the infectious source adequately controlled? Are the antibiotics covering the correct pathogens based on available culture data? Is the patient adequately resuscitated or fluid overloaded? Is there a missed or second source of infection? For the worsening patient, escalation of care should include broadening antibiotic coverage, re-imaging to identify new or worsening collections, reassessing the need for procedural source control, and considering additional hemodynamic support.

Goals of care discussions are an essential component of sepsis management that are often underemphasized in the acute phase of treatment. Prognostic information should be communicated honestly and clearly to patients and families, including the expected mortality rate based on disease severity, the potential for prolonged intensive care unit stay and recovery, and the possibility of significant long-term morbidity. Patient preferences regarding the aggressiveness of treatment should be elicited early, including discussion of advance directives, do-not-resuscitate status, and acceptable quality of life outcomes. Family communication should be regular, ideally daily, and should involve a multidisciplinary approach including the attending physician, nursing staff, social workers, and chaplains when appropriate. Palliative care involvement should be considered early in the course of sepsis, not only for patients with a poor prognosis but also for symptom management, family support, and assistance with complex medical decision-making, as early palliative care consultation has been associated with improved patient and family satisfaction without reducing the intensity of curative treatment.

<image>Panel A: Clinical monitoring dashboard showing continuous vital sign displays (heart rate, arterial blood pressure, respiratory rate, SpO2, temperature), hourly assessments (mental status check, urine output measurement), and peripheral perfusion evaluation (skin temperature, mottling score, capillary refill time). Panel B: Laboratory monitoring schedule table showing serial lactate (every 2-4 hours initially), daily creatinine, daily liver function tests, daily CBC, and coagulation studies (when DIC suspected), with trend graphs showing expected patterns during recovery. Panel C: Three-pathway response assessment showing improving patient (de-escalate antibiotics, wean vasopressors), not improving at 48-72 hours (reassess source control, antibiotic coverage, fluid status, second source), and worsening patient (broaden antibiotics, re-image, procedural source control, escalate hemodynamic support). Panel D: Goals of care discussion framework showing prognostic communication (mortality data, ICU trajectory, long-term outcomes), advance directive review, family meeting structure (daily updates, multidisciplinary team), and palliative care integration (symptom management, family support, complex decision-making).</image>

Section 10: Outcomes and Complications

The mortality of sepsis and septic shock remains substantial despite advances in early recognition and evidence-based management. In-hospital mortality for sepsis ranges from ten to twenty percent, reflecting the broad spectrum of severity from organ dysfunction that quickly resolves with appropriate treatment to refractory multi-organ failure. Septic shock carries a dramatically higher mortality of forty to fifty percent, underscoring the prognostic significance of refractory hypotension and elevated lactate. Each hour of delay in antibiotic administration is associated with an approximately seven percent increase in mortality, making timely antibiotic delivery one of the most impactful interventions. The number of organ systems affected directly correlates with mortality: single-organ dysfunction may carry a mortality of ten to fifteen percent, while failure of four or more organ systems is associated with mortality exceeding sixty to seventy percent.

Long-term sequelae of sepsis survivorship are increasingly recognized as a significant public health burden. Cognitive impairment, including deficits in memory, attention, executive function, and processing speed, affects thirty to fifty percent of sepsis survivors and may persist for years after hospital discharge. Physical disability, including reduced functional status, decreased mobility, and inability to perform activities of daily living, affects twenty-five to fifty percent of survivors and significantly impacts quality of life. Psychological sequelae including post-traumatic stress disorder, depression, and anxiety affect ten to thirty percent of survivors and are often under-recognized and undertreated. Hospital readmission rates are alarmingly high, with twenty to forty percent of sepsis survivors readmitted within ninety days, often for recurrent infection or exacerbation of chronic conditions. Long-term mortality remains elevated for at least two years following a sepsis episode, even in patients who appear to have made a good recovery.

Intensive care unit complications in sepsis are common and require proactive prevention and management. Acute respiratory distress syndrome is one of the most serious complications and requires lung-protective ventilation with low tidal volumes and appropriate positive end-expiratory pressure to minimize ventilator-induced lung injury. Acute kidney injury occurs in approximately fifty percent of patients with septic shock and may require renal replacement therapy; prevention strategies include avoiding nephrotoxic agents and optimizing hemodynamic parameters. ICU-acquired weakness, resulting from critical illness polyneuropathy and myopathy, affects a significant proportion of patients with prolonged ICU stays and is mitigated by early mobilization and physical therapy. Delirium occurs in up to eighty percent of mechanically ventilated patients and is managed through minimization of sedation, early mobilization, and reorientation strategies. Pressure injuries are prevented through regular repositioning, specialized mattresses, and attention to nutrition and moisture management.

Post-sepsis syndrome is an emerging concept that describes the constellation of long-term physical, cognitive, and psychological sequelae experienced by sepsis survivors. Cognitive deficits may include impaired memory, difficulty with attention and concentration, and reduced executive function, mimicking mild cognitive impairment or early dementia and significantly impacting the ability to return to work and independent living. Physical manifestations include persistent weakness, fatigue, decreased exercise tolerance, and reduced functional capacity, often requiring prolonged rehabilitation. Psychological sequelae encompass anxiety, depression, post-traumatic stress disorder, and sleep disturbances, which may interact with and exacerbate cognitive and physical symptoms. Management of post-sepsis syndrome requires a comprehensive, multidisciplinary approach including physical rehabilitation, cognitive rehabilitation, psychological support, and close outpatient follow-up with primary care to address the multiple chronic conditions that commonly coexist in sepsis survivors. Recognition of post-sepsis syndrome is important for setting realistic expectations during recovery and for directing appropriate rehabilitative resources.

<image>Panel A: Mortality statistics displayed as a stepped bar chart showing in-hospital mortality for sepsis (10-20%), septic shock (40-50%), and the impact of antibiotic delay (7% increase per hour), alongside a graph showing mortality by number of organ failures (one organ 10-15%, four or more organs 60-70%). Panel B: Long-term sequelae infographic showing five domains of post-sepsis morbidity: cognitive impairment (30-50%, brain icon), physical disability (25-50%, wheelchair icon), psychological sequelae (10-30%, brain with storm), hospital readmission (20-40% at 90 days, hospital building), and increased long-term mortality (2+ years, survival curve). Panel C: ICU complications prevention and management panel showing ARDS (lung-protective ventilation settings), AKI (avoid nephrotoxins, optimize hemodynamics), ICU-acquired weakness (early mobilization, physical therapy), delirium (minimize sedation, reorientation), and pressure injuries (repositioning schedule, specialized mattress). Panel D: Post-sepsis syndrome overview showing cognitive deficits (memory and executive function), physical manifestations (weakness, fatigue, reduced exercise tolerance), psychological sequelae (anxiety, depression, PTSD), and multidisciplinary management approach (rehabilitation, psychology, primary care follow-up).</image>

Summary

• Sepsis-3 defines sepsis as life-threatening organ dysfunction caused by a dysregulated host response to infection
• Septic shock is defined as sepsis requiring vasopressors to maintain MAP of 65 or higher combined with lactate greater than 2 despite adequate fluid resuscitation
• qSOFA (respiratory rate 22 or more, altered mentation, systolic BP 100 or less) is a bedside screening tool; SOFA score quantifies organ dysfunction
• The Hour-1 bundle includes: measure lactate, obtain blood cultures, administer broad-spectrum antibiotics, infuse 30 mL/kg crystalloid, and initiate vasopressors if needed
• Fluid resuscitation with balanced crystalloid is preferred; reassess fluid responsiveness after the initial bolus
• Antibiotics should be administered within one hour of sepsis recognition, source-directed, and de-escalated based on culture results
• Norepinephrine is the first-line vasopressor; vasopressin is the recommended second-line agent; target MAP is 65 or higher
• Source control within 6-12 hours is essential when an undrained collection, infected device, or obstructed system is identified
• Adjunctive therapies include corticosteroids for refractory shock, restrictive transfusion, moderate glucose control, and comprehensive supportive care
• Sepsis mortality is 10-20% overall and 40-50% for septic shock; long-term sequelae are common and require multidisciplinary follow-up

Key Terms

This content is subject to the MIT License. © 2024–2026 Hibbert School of Medicine.

Read this lecture as Markdown