Infectious Diseases

Sepsis and Septic Shock

Last Updated 9/1/2023

#Sepsis and Septic Shock


-- ABCs: determine if need ICU for pressors or intubation (often from resp distress 2/2 PNA, ARDS, or hyperventilation from acidosis); do they need central access?

-- Chart Check: calculate qSOFA (specific for sepsis) and SIRS (sensitive, less specific); look for prior culture data, co-morbidities that may increase risk of overload, hypotension, inability to fight off infection

-- HPI Intake: previous infections, recent abx, indwelling lines, sick contacts, immune status, recent travel

-- Can't Miss: shock, poor source control (abscess, empyema, etc)

-- Admission Orders: continuous pulse ox and tele, CBC, BMP, LFTs, coags, VBG, lactate, consider procalcitonin, make sure infectious workup sent (2 sets of peripheral BCx, culture off any line present for >48 hours, UA, UCx BEFORE antibiotics), CXR if not done

-- Initial Treatment to Consider: fluid resuscitation - commonly 30cc/kg LR in first few hours (cautious not to overload), check for response, pressor if no longer fluid responsive, decide on ongoing abx


-- History: previous infections, recent abx, indwelling lines, sick contacts, immune status, recent procedures

-- Clinical: fever, altered mental status (AMS), localizing symptoms, urine output

-- Exam: hypotension, widened pulse pressure, low diastolic pressure, tachycardia, AMS, warm, rigors, diaphoresis, volume assessment, rash, mottling, cool extremities, cap refill, abscess, decrease breath sounds, crackles, RUQ pain, signs of peritonitis, CVA tenderness, joint pain

-- Data: WBC, trend lactate, creatinine, CXR, UA

-- qSOFA:  ***/3 - (RR >22, AMS, systolic BP <100)

-- SIRS: ***/4 - (HR>90, RR>20, WBC <4 or >12, T < 96.8 or > 100.4)


-- Pulmonary - pneumonia

-- Urinary - UTI, pyelonephritis,

-- Gastrointestinal - cholangitis, C Diff, appendicitis, diverticulitis, obstruction, perforation

-- SSTI - toxic shock, necrotizing fasciitis

-- CNS - meningitis, spinal abscess

-- Bacteremia - line infection, gut translocation (neutropenic), endocarditis

-- Other Infectious - tick-borne (anaplasmosis), candida, aspergillus, PJP

-- Mimics - acute mesenteric ischemia, bowel obstruction, pancreatitis, liver failure, decompensated cirrhosis, adrenal crisis, thyroid storm, DKA,



-- Labs: f/u BCx, UA, UCx, procalcitonin; consider 1,3 BDG, galactomannan, cryptococcal Ag if c/f fungemia

-- Imaging: CXR; Consider CT C/A/P to identify a potential source if unknown

-- Monitoring: trend daily CBC, CMP, LFTs, Coags; strict I/O’s for UOP


-- Abx: Empiric treatment with *** (vancomycin PLUS ceftriaxone OR cefepime OR pip/tazo OR carbapenem) to cover for ***; If c/f toxic shock or Fournier’s, add clindamycin; tailor to clinical presentation, resistance patterns

-- Volume: s/p ***; plan for *** (goal to bolus at least 30 mL/kg) with *** to assess for response; cautious not to overload given h/o ***

-- Pressor: titrate to MAP goal >65 (most commonly norepinephrine --> vaso --> epinephrine; phenylephrine for afterload in Afib, AS)

-- Steroid: hydrocortisone ~200mg daily in divided doses (50mg q6 or 100mg q8) in refractory shock

-- O2: continuous pulse ox; titrate as needed for goal *** >94%

-- Other: transfusion, dialysis, bicarb


The patient remains septic from *** (known/suspected) *** (etiology).

They are currently *** (improving/stable/worsening) based on *** (fevers, hypotension, pressor requirement, WBC, UOP, other end organ-dysfunction, etc).

We are covering for *** with *** (antibiotics)

We have resuscitated the patient with ***, and their current volume status is *** with an ScVO2 of ***.

There *** (is/is not) currenty a pressor requirement ***

We’ve sent *** and are waiting for *** to come back.

Today, I propose we ***

If You Remember Nothing Else

Sepsis, a condition with evolving definitions over the years, can manifest in various ways. While the SIRS criteria are sensitive in detecting sepsis, they lack specificity. On the other hand, qSOFA offers greater specificity, helping clinicians discern common patterns among presentations. When suspecting sepsis, it's imperative to obtain cultures promptly and initiate broad-spectrum antibiotics without delay. While initial fluid resuscitation is crucial, it's equally important to monitor for signs of fluid overload, such as deteriorating oxygenation and respiratory status, which might hint at pulmonary edema. Should the patient's response to fluids diminish or if they remain hypotensive despite adequate fluid resuscitation, don't delay in introducing vasopressors. Norepinephrine is typically the first choice in sepsis or undifferentiated shock. Continuous monitoring remains at the heart of sepsis management. Aim to maintain a mean arterial pressure between 60-70 mmHg and ensure a urine output of at least 0.5 mL/kg/hour. Tracking serum lactate levels, along with other relevant lab results, provides valuable insights into the patient's response to treatment. Identifying and addressing the source of infection is paramount. This may involve surgical interventions, abscess drainage, or the removal of infected devices. If there's no improvement in the patient's condition or if they show signs of deterioration, it's time for a thorough re-evaluation. This includes revisiting the initial diagnosis, searching for other potential infection sources, verifying the suitability of the current antibiotic regimen, and being vigilant for complications or nosocomial infections.

Clinical Pearls

History, Definitions, and Terminology

  • In 1991, SIRS criteria were introduced as a way to identify sepsis and patients at high risk of sepsis-related death. Sepsis was defined as the presence of 2/4 SIRS criteria with evidence of infection. Severe sepsis was sepsis with organ dysfunction, and septic shock was if there was hypotension unresponsive to fluids. 
  • In 2001, Sepsis-2 expanded and clarified these definitions without a complete overhaul. However, the broad nature of SIRS commonly led to misclassifications, necessitating further revision.
  • Between 2014-2016, the Sepsis-3 conference elected to eliminate SIRS and severe sepsis, instead introducing SOFA and qSOFA to identify sepsis in the absence of SIRS and to focus more on organ dysfunction which narrows the spectrum of sepsis
  • Now, sepsis is defined as life-threatening organ dysfunction caused by a dysregulated host response to infection
  • Septic Shock is defined as sepsis PLUS pressor need and lactate >2 without hypovolemia (after being resuscitated)

Etiology, Pathophysiology, and Clinical Complications

  • Pneumonia is the most common cause of sepsis; gram-positive bacteria are more common in the U.S. but gram-negative sepsis tends to make patients sicker faster
  • Gram positives have exotoxins and superantigens, GNRs have lipopolysaccharide (LPS)
  • Pathophysiology - immune response to inflammation → vessel dilation → capillary leak → edema and intravascular hypovolemia → extrinsic coagulation cascade and DIC → microvascular thrombosis → tissue ischemia → organ dysfunction and acidosis
  • Good urine output is reassuring, but low urine output is nonspecific
  • Significant clinical complications of sepsis to be on the lookout for include vasodilation, ARDS, cholestasis, AKI, hepatic dysfunction, DIC
  • Risk factors for fungemia include neutropenia, chemo, transplant, and indwelling lines

Laboratory Values in Sepsis

  • Lactate elevation is most commonly attributed to hypoperfusion and is viewed as something that needs to be rapidly corrected. However, lactate may be more appropriately thought of as a reflection of endogenous epinephrine production and persistent or worsening elevation may prompt you to rethink the source of infection, your antibiotic selection, or if there is another diagnosis.
  • The differential for lactatemia includes Type A (Tissue Hypoxia) caused by hypovolemia, shock, localized ischemia (mesenteric, limb), decreased oxygenation, anemia, and increased metabolic demand (seizure, rigors, exercise). Type B (No tissue hypoxia) can be caused by increased adrenergic state (albuterol, cocaine, epinephrine, pheo), decreased Krebs (thiamine deficiency, alcohol use, cyanide), liver dysfunction (decreased gluconeogenesis, reduced clearance)
  • The Neutrophil/Lymphocyte Ratio (NLR) can help identify patients with early physiologic stress (endogenous cortisol and catecholamines mobilize neutrophils and kill lymphocytes) - If the ratio is 6-10+ it may suggest clinically relevant stress warranting closer monitoring
  • Procalcitonin is a measure of inflammation that is sensitive to bacterial pathogens, most notably gram negatives. It has been investigated for the purpose of identifying patients unlikely to have a bacterial infection and de-escalating antibiotics in the ICU setting; however it can reasonably be used in patients already started on abx with normal immune systems who have a presentation equivocal for sepsis - if the procalcitonin is <0.5, it is less likely to be sepsis/septic shock, noting that it can take a day to become elevated

Initial Resuscitation and Fluid Management

  • Some suggest the goals of resuscitation are to clear lactate, return the cap refill to <3 seconds, and improve hemodynamics and UOP. However, be wary of misinterpreting the etiology of lactate elevation and assuming poor UOP is due to prerenal AKI instead of intrinsic ATN from sepsis as it may lead to inappropriately overloading a patient. The focus should be on ensuring tissue perfusion.
  • Hypovolemia is NOT the core problem in sepsis. Rather, it is vasodilation and poor perfusion from poor distribution of blood to vital organs. However, many patients with sepsis will present with hypovolemia and benefit from fluid resuscitation to achieve euvolemia. 
  • A great deal of administered fluid may likely leak out of the vessels in sepsis - unless there is a compelling clinical reason to suspect otherwise, giving more than 2-3 liters of fluid is unlikely to help, and may even cause harm via overload and pulm edema
  • In general, MAP is the most reliable parameter for assessing response to therapy; the primary concern with low MAP is renal injury
  • A fluid challenge (250-500cc bolus) can be utilized to see if further fluid may be beneficial, however common ways we measure fluid responsiveness should be taken with a grain of salt
  • Passive Leg Raise can be done rather than bolusing - raise legs to a 45-degree angle for 1 minute while lying supine and watch CVP - this is the equivalent of giving ~250-350cc
  • Some prior work suggests that an adequate fluid challenge is if CVP increases >2 
  • IVC diameter collapsibility <50% with inspiration suggests CVP is elevated
  • Pulse Pressure Variation - validated in mechanically ventilated patients 

Vasopressor Selection and Use

  • Start with Norepinephrine (Levophed) as the initial pressor; norepinephrine can be administered peripherally, but prolonged use can lead to increased risk of extravasation leading to necrosis 
  • Consider adding Vasopressin when the Levophed dose reaches 5-15mcg/min or in patients who are tachycardic or at risk for arrhythmia.  Vasopressin is usually considered either on or off. It should not be used peripherally. 
  • Epinephrine is typically the third agent introduced, especially when Levophed exceeds 25mcg/min; may be beneficial in patients with reduced EG, inappropriately low/normal heart rate; epinephrine infusions will increase lactate
  • Phenylephrine (Neo) is chosen for increasing afterload, particularly when Levophed causes side effects like arrhythmias or in cases of AFib with RVR and hypotension, provided cardiac output is satisfactory
  • Dopamine is not commonly used but can be considered for patients with bradycardia who are at low risk for arrhythmias.
  • Methylene blue is a rare choice, reserved for suspected vasoplegia.

Acidosis Management

  • In cases of severe acidemia (pH <7.1), which can lead to myocardial depression, decreased pressor efficacy, and increased arrhythmias, consider using bicarbonate. This is especially true if there's significant AKI.
  • Bicarbonate can be administered as a 50mEq/50mL ampule or as an infusion with 150mEq in 1L D5W.
  • Note: Administering bicarbonate increases pCO2 levels, which the patient must ventilate. Ensure the patient has a strong respiratory drive or be prepared for potential intubation.
  • Early dialysis initiation hasn't shown consistent benefits unless there's an urgent need, remembered by the acronym AEIOU.

Steroid Use in Sepsis

  • The use of steroids in sepsis remains a topic of debate. However, for septic shock, in patients refractory to low-medium doses of pressor may benefit from IV hydrocortisone at doses of 50mg every 6 hours or 100mg every 8 hours. This treatment typically lasts 5-7 days, followed by a taper based on the patient's response.
  • Fludrocortisone is unlikely to add any additional benefit
  • The body naturally increases cortisol production during stress, including severe infections like sepsis, as a defense mechanism against inflammation and to maintain cardiovascular stability.
  • Some septic patients might experience "relative adrenal insufficiency" or "critical illness-related corticosteroid insufficiency (CIRCI)", where the adrenal glands don't produce adequate cortisol.
  • Contraindications may include immunocompromise, fungal infection, poorly controlled diabetes

Trials and Literature

Fluid Management in Critically Ill Patients

  • SMART Trial (NEJM, 2018): Compared balanced crystalloids to saline in critically ill adults. Found 14.3% vs. 15.4% adverse kidney events; 10.3% vs. 11.1% for in-hospital mortality at 30 days.
  • Will This Hemodynamically Unstable Patient Respond to a Bolus of IV Fluids (JAMA, 2016): A study focusing on the response of hemodynamically unstable patients to IV fluid boluses.

Vasopressor Selection and Use

  • SOAP II Trial (NEJM, 2010): Compared dopamine to norepinephrine in shock. Found similar 28-day mortality rates, but dopamine had an increased risk of arrhythmia.

Renal Replacement Therapy (RRT) in Septic Shock

  • IDEAL-ICU Trial (NEJM, 2018): Compared early vs. delayed RRT in septic shock. For those without an urgent need for dialysis, there was no difference in 90-day mortality, and the delayed group had overall lower RRT use.

Blood Transfusion Guidelines in Sepsis

  • TRISS Trial (NEJM, 2014): Compared transfusion guidelines in sepsis for patients with Hgb <7 vs. <9 goal. Found similar 90-day mortality rates, but the lower threshold used 50% fewer blood units.

Early Goal-Directed Therapy (EGDT)

  • EGDT is a structured approach to treating sepsis and septic shock in the early stages, particularly during the first six hours after recognition. The concept of EGDT was introduced to optimize the management of sepsis by targeting specific physiological goals including CVP, MAP, ScvO2, UOP, and lactate.
  • The approach was initially popularized by a landmark study (NEJM 2001), which suggested that EGDT could significantly reduce mortality in septic patients. 
  • However, subsequent large-scale trials including ARISE (NEJM, 2014), ProCESS (NEJM, 2014), and ProMISe (NEJM, 2015) did not show a mortality benefit of EGDT over usual care, leading to some changes in sepsis management guidelines.

Adjunctive Therapies in Sepsis

  • CORTICUS Trial (NEJM, 2008): Investigated the use of hydrocortisone in septic shock. Found that hydrocortisone did not significantly reduce mortality but did speed up shock reversal.
  • Meta-Analysis of Corticosteroids in Treatment of Sepsis (NNT): Suggested corticosteroids might reduce the risk of short-term 28-day mortality, but likely not long-term at 90 days.
  • Vitamin C, Thiamine, and Hydrocortisone in Severe Sepsis (Chest, 2017): A small trial that showed positive outcomes but hasn't become routine practice.

Other Resources