Show Summary Details
Page of

Management of septic shock in the critically ill 

Management of septic shock in the critically ill
Management of septic shock in the critically ill

Sandra L. Peake

and Matthew J. Maiden

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2020. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

date: 01 December 2020

Key points

  • The pathophysiology of septic shock involves variable degrees of macrovascular, microvascular, and cellular dysfunction.

  • Early recognition, resuscitation, appropriate antimicrobials, and source control are essential.

  • Intravenous fluids are the first therapeutic strategy, but the optimal type and volume of fluid and resuscitation targets remain uncertain.

  • If hypotension persists despite fluid resuscitation, noradrenaline is the currently recommended first-line vasopressor. If cardiac contractility is impaired despite adequate fluid replacement, consider adding dobutamine or adrenaline.

  • Lactate and/or ScvO2 can be used as a guide to adequacy of resuscitation. However, high ScvO2 is associated with high mortality.


Over the last 20 years, numerous clinical trials of novel therapeutic agents for septic shock have been evaluated. Nonetheless, conclusive evidence of a mortality benefit has not been forthcoming. Accordingly, the fundamental management principles of septic shock, namely early recognition, source control, appropriate and timely antibiotics, and haemodynamic resuscitation, remain the most important and effective therapeutic strategies.

The 2012 Surviving Sepsis Campaign (SSC) guidelines [1]‌ recommend early (within 6 hours), protocolized resuscitation for septic patients with evidence of tissue hypoperfusion (hypotension despite an initial fluid challenge or blood lactate ≥4 mmol/L). Key recommendations and the levels of supportive evidence are outlined in Table 298.1.

Table 298.1 Summary of the key 2012 Surviving Sepsis Campaign guidelines recommendations for early infection control and haemodynamic resuscitation in patients with severe sepsis or septic shock#

Early resuscitation strategies

Level of evidence

Strength of recommendation

Determining infection source and causative organisms

  • Appropriate cultures before starting antimicrobials, provided administration not delayed



  • ≥2 blood culture sets (other sites e.g. vascular access devices as clinically indicated)



Appropriate, timely antimicrobial therapy and source control

  • Broad-spectrum iv antimicrobials within 1hr of recognizing septic shock



  • Broad-spectrum iv antimicrobials within 1hr of recognizing severe sepsis without shock



  • Source identification and control within 12hr of diagnosis if feasible, including removal of potentially infected intravascular devices



Haemodynamic resuscitation

  • Goal-directed resuscitation for first 6hr after hypoperfusion recognized



  • iv fluid to achieve CVP 8–12 mmHg (12–15 mmHg if mechanically ventilated or decreased ventricular compliance)



  • Initial fluid resuscitation with crystalloids



  • Minimum initial fluid challenge 30 mL/kg of crystalloids



  • Suggest adding albumin for initial fluid resuscitation



  • Noradrenaline as first-line vasopressor



  • Adrenaline added or substituted to maintain MAP



  • MAP target ≥65 mmHg



  • Dobutamine if evidence of myocardial dysfunction or hypoperfusion



  • PRBC transfusion (if Hct < 30%) and/or dobutamine to attain ScvO2 ≥ 70% or SvO2 ≥ 65%



# Quality of evidence assessed using the Grading of Recommendations Assessment Development and Evaluation (GRADE) system. Evidence graded as high (A), moderate (B), low (C) and very low (D). Strength of recommendation graded as strong (1) or weak (2).

iv, intravenous; hr, hour; CVP, central venous pressure; MAP, mean arterial pressure; PRBC, packed red blood cells; Hct, haematocrit; ScvO2, central venous oxygen saturation; SvO2, mixed venous oxygen saturation (SvO2) of ≥65%.

Adapted from Dellinger R et al., ‘Surviving Sepsis Campaign: International Guidelines for Management of Severe Sepsis and Septic Shock: 2012’, Critical Care Medicine, 41(2), copyright 2013, with permission from Wolters Kluwer Health and Society of Critical Care Medicine.

Of note, the haemodynamic goals and the time frame for resuscitation were particularly influenced by one small, single-centre, randomized, controlled trial of protocolized early goal-directed therapy (EGDT) in patients presenting to the emergency department (ED) with severe sepsis or septic shock [2]‌. This trial, published in 2001, reported that delivery of EGDT for 6 hours was associated with a 16% reduction in hospital mortality. Implementation of ‘bundled’ care, incorporating the EGDT resuscitation algorithm, has been reported to improve survival in subsequent non-randomized studies. More recently, three randomized trials of protocol-based resuscitation for patients presenting to the ED with early septic shock in the United States (Protocol-based Care for Early Septic Shock, ProCESS) [3], Australasia (Australasian Resuscitation In Sepsis Evaluation, ARISE) [4] and England (Protocolised Management In Sepsis, ProMISe) [5] have failed to demonstrate a survival benefit with EGDT compared with non-protocolized usual care.

The aim of this chapter is to review the current management of septic shock, with particular emphasis on the importance of early infection control and haemodynamic resuscitation.

Infectious source identification and control

Urgent identification of the infectious focus should occur for all septic patients. If the source is not apparent clinically, imaging studies are often required to assist in site determination.

To help identify the infecting organism(s) and direct antimicrobial management, ≥2 sets of blood cultures (and other sites as appropriate) should be taken prior to commencing antimicrobials. Prospective data from >15,000 patients in the SSC performance improvement initiative demonstrated that mortality was lower for patients in whom blood cultures were obtained before starting antimicrobials (odds ratio [OR] 0.86, 95% Confidence Interval [CI] 0.79–0.93) [6]‌. Nevertheless, obtaining cultures must not unduly delay antimicrobial administration.

Although there are no, and unlikely to ever be, any randomized trials studying the effect of source control, once the site of infection is identified attention must be directed towards definitive treatment (e.g. drainage of collections, removal of infected tissue or intra-vascular devices). Highly lethal diseases like necrotising soft tissue infections and intestinal ischaemia require immediate intervention. Timing for other infections depends upon the type of infection, severity of physiological disturbance and the likely risk of deterioration.

Antimicrobial therapy

Timely administration of empirical broad-spectrum agents to cover all likely pathogens and susceptibility patterns (with subsequent narrowing of therapy) is vitally important. The importance of providing appropriate initial antimicrobials has been highlighted in a number of retrospective and observational studies in which delayed or inappropriate antimicrobial administration is associated with increased mortality [7,8]. Survival benefit with timely antimicrobials has also been described for immunosuppressed patients and for critically ill patients with specific infections such as candida, nosocomial infection, and ventilator-associated pneumonia.

Although delayed antimicrobial administration may be a critical determinant of outcome, the observed relationship does not necessarily imply causality. Early and appropriate antimicrobial therapy could simply be a surrogate marker for better overall processes-of-care and hence, better outcomes from septic shock. This provides impetus for processes-of-care that facilitate early recognition and definitive treatment of severe sepsis. A recent meta-analysis of providing ‘bundled care’ for sepsis, reported reduced time to antibiotic initiation (weighted mean difference –0.58 hrs [–0.85 to –0.33]) and increased likelihood of providing appropriate antimicrobials (OR 3.06, 95%CI 1.69–5.53) [9]‌.

Haemodynamic resuscitation

Resuscitation of septic shock is a medical emergency. Early recognition is vital as survival is decreased if treatment is delayed. Resuscitation should occur in an environment with continuous physiological monitoring and close medical and nursing attention.

The fundamental disorder in septic shock is inadequate cellular oxygen delivery (DO2) and/or impaired utilization. The circulatory disturbance involves decreased preload (secondary to increased capillary permeability, venodilatation, hypovolaemia), myocardial depression, arteriolar dilatation, and peripheral shunting. The causes of impaired cellular oxygen utilization remain unclear.

Whilst the resuscitation principles of optimizing preload, afterload, contractility, heart rate, haemoglobin concentration and oxygen saturation seem straightforward, a number of issues merit further consideration, including:

  • The choice of resuscitation fluid.

  • How much fluid to administer.

  • Which vasoactive agent(s) to use.

  • The best way to reliably guide resuscitation.

Resuscitation fluids

The predominant circulatory disturbance in septic shock is a maldistribution of circulating blood volume and inadequate preload. Myocardial dysfunction also occurs, but fluid resuscitation often increases stroke volume (SV), resulting in a hyperdynamic circulation. Hence, fluids are usually the first therapeutic strategy in the management of septic shock and an initial challenge with ≥500–1000 mL of either crystalloid or colloid should be immediately and rapidly infused.

However not all patients respond to fluid administration, particularly if the heart is fully loaded (Frank-Starling law). Of interest, results from the Fluid Expansion as Supportive Therapy (FEAST) trial even suggest that a fluid bolus (5% albumin or 0.9% saline) may be harmful; at least in children with severe infection in Africa [10]. A conservative approach to ongoing fluid management is also supported by a randomized trial of liberal versus restrictive fluid administration in mechanically ventilated patients with acute lung injury.

Unfortunately, there is little evidence to indicate how much fluid to give or what are the most appropriate end-point(s) to guide fluid resuscitation. For example, there are no large-scale randomized studies evaluating the recommended ‘≥30 mL/kg crystalloid as initial resuscitation’. In the absence of such a trial, we are left with clinical assessment of the heart rate and blood pressure (BP) response to fluid administration, indices of end-organ perfusion (e.g. urine output, capillary refill) and evidence of fluid overload (e.g. oedema).

Clinical assessment may be supplemented with invasive and/or non-invasive haemodynamic monitoring. Although the SSC guidelines recommend targeting a central venous pressure (CVP) of 8–12 mmHg (mechanically ventilated 12–15 mmHg), cardiac filling pressures often fail to discriminate when a patient will respond to a fluid challenge.

Repeated assessment of the ability to increase CO with fluid boluses or a passive leg-raise are more reliable than static measures such as CVP or pulmonary artery occlusion pressure (PAOP). Dynamic measures of systolic pressure variation, pulse pressure variation, and SV variation (using pulse contour analysis) are currently the most reliable techniques of assessing fluid responsiveness. However, these techniques have only been validated in patients in sinus rhythm and receiving positive-pressure ventilation. Other dynamic measurement techniques include ultrasound assessment of inferior vena cava diameter, left ventricle end-diastolic area (echocardiography), and global end-diastolic volume index (GEDVI) using transpulmonary thermodilution; albeit these techniques have proven too unreliable to adopt widely.

Which resuscitation fluid?

The choice of fluid has engendered considerable debate for many years and there is no evidence that one fluid is better than another for patients with septic shock. An international survey of resuscitation fluid use in 391 intensive care units (ICU) and 25 countries reported that whilst colloid is more commonly administered than crystalloid, considerable geographical variation is evident.

In principle, the choice of fluid should be determined by the type of fluid lost, adverse effects of the fluid administered, degree of peripheral/interstitial oedema, serum albumin, availability, and costs. Whilst crystalloids are cheap and readily available, the potential for interstitial oedema, electrolyte, and acid-base disturbances (e.g. hypernatraemia and hyperchloraemic acidosis with large volume 0.9% saline resuscitation) are considered problematic. Conversely, colloids restore circulating volume more quickly, but are more expensive than crystalloids, can accumulate in the interstitium contributing to persistent interstitial oedema and some solutions have specific side-effects.

In the Saline versus Albumin Fluid Evaluation (SAFE) study comparing 4% albumin and 0.9% saline for fluid resuscitation, albumin was associated with less fluid administration on days 1–4 (ratio 1:1.4). Over the same time-frame, no difference in mean arterial pressure (MAP) was observed, although CVP was statistically, but not clinically, significantly different (albumin ~1 mmHg higher). Twenty-eight-day mortality was also not different. Analysis of the pre-defined severe sepsis subgroup suggested that albumin improved survival (OR 0.71, 95% CI 0.52–0.97) [11]. A recent meta-analysis of albumin resuscitation in sepsis also supports a beneficial role (OR 0.82, 95% CI 0.67–1.0) [12].

Whilst hydroxyethyl starch (HES) is widely used in many countries, there is considerable concern regarding the safety profile, particularly with respect to renal function. The Intensive Insulin and Pentastarch Resuscitation in Severe Sepsis Randomized Trial reported a significant increase in renal dysfunction and a trend to increased 90-day mortality with 10% HES (200/0.5). The Scandinavian 6S Trial Group has also recently reported that resuscitation with 6% HES (130/0.42) is associated with both an increased mortality risk and requirement for renal-replacement therapy (RRT) compared with Ringer’s acetate in a multi-centre, blinded trial of 798 severe sepsis patients [13]. However, a small, randomized trial of HES 130/0.4 versus 0.9% saline in severe sepsis (CRYSTMAS study) reported that HES was not associated with adverse outcomes. In contrast, the Crystalloid versus Hydroxyethyl Starch Trial (CHEST) trial involving 7000 ICU patients requiring fluid resuscitation reported that 6% HES was associated with increased requirement for RRT, although 90-day mortality was not different compared to patients receiving 0.9% saline for resuscitation [14].

Vasoactive agents

When fluid resuscitation alone fails to optimize end-organ perfusion, vasoactive agents should be commenced. As with the choice of resuscitation fluid, the optimal agent(s) is also controversial and the survival benefit of any particular agent has not been proven in large-scale trials.

The SSC guidelines advocate noradrenaline (predominantly α‎-agonist) as the initial vasopressor of choice to maintain a MAP ≥65 mmHg. Adrenaline has not been recommended as first-line therapy; in part because of concerns about tachycardia, hyperlactataemia and impaired organ perfusion. There is no evidence that adrenaline is associated with increased mortality and concerns are ameliorated by a recent blinded, randomized trial of adrenaline versus noradrenaline in patients with shock (57% septic shock), which reported that adrenaline-associated tachycardia and lactic acidosis were non-sustained and attainment of MAP goals was similar for both catecholamines. Mortality (secondary end-point) was also similar [15]. Haemodynamic end-points and survival were also similar in a randomized trial of septic shock patients comparing adrenaline with noradrenaline plus dobutamine.

Dopamine has often been considered an alternative first-line catecholamine for septic shock. A recent meta-analysis comparing dopamine and noradrenaline in septic shock suggested increased mortality and arrhythmias with dopamine. Accordingly, routine use of dopamine is not supported by the SSC guidelines.

Although the Vasopressin in Septic Shock Trial (VASST) failed to show a survival benefit for adding vasopressin (vs. noradrenaline) in septic shock patients, low-dose vasopressin administration may still be considered for catecholamine-resistant hypotension and tissue hypoperfusion [16]. Early, rather than late, administration may also be more beneficial.

Irrespective of the vasopressor choice, it is recommended that BP is continuously monitored using invasive measurement and that the MAP target is ≥65 mmHg. However, the appropriate BP target must be individualized to account for factors such as pre-morbid BP, vascular disease, and the clinical response to resuscitation. Although an increase in MAP from 65 to 85 mmHg with noradrenaline has been reported to improve the macrocirculation and microvascular blood flow, the ideal BP goal has never been formally evaluated in large-scale trials.

Finally, if MAP is restored, but signs of organ hypoperfusion remain, some authors advocate commencing an inotrope (generally dobutamine) to increase CO and improve DO2. In particular, the Rivers’ EGDT protocol calls for dobutamine commencement (≤20 µg/kg/min) if ScvO2 is <70% and haematocrit is >30% despite intravenous (iv) fluid repletion and restoration of MAP [2]‌. Although the addition of dobutamine is recommended by the SSC guidelines, the level of evidence supporting its use is low.

Glyceryl trinitrate (GTN) has been reported to improve microvascular flow in a small observational study using sublingual orthogonal polarization spectral imaging in septic shock patients. A subsequent randomized, blinded trial of 70 severe sepsis patients demonstrated that whilst fluid resuscitation improved microvascular flow, GTN was not different to placebo and was associated with a trend to increased mortality.

Resuscitation targets

Tissue perfusion and O2 utilization may remain inadequate despite seemingly normal global haemodynamic parameters. The ability to monitor adequacy of cellular resuscitation is becoming increasingly important. Given the complex and varied interplay of macro-vascular, micro-vascular, and cellular changes that occur in septic shock, it is unlikely that any one variable is reliable enough to guide resuscitation. Based on available evidence, a combination of clinical assessment, venous oxygen saturation monitoring, and/or lactate measurement provides the most useful guide to the adequacy of resuscitation.

Venous haemoglobin-O2 saturation

Mixed venous oxygen saturation (SvO2) monitoring provides information regarding the balance between DO2 and global oxygen consumption. However, measurement requires pulmonary artery catheter placement. More recently, continuous central venous oxygen (ScvO2) monitoring, with therapies titrated to achieve a ScvO2 >70%, has been proposed as an alternative resuscitation target. Although ScvO2 measurement does not include coronary venous blood, and in health is usually 5–20% higher than SvO2, the two measures behave similarly in shock and ScvO2 may be a useful surrogate marker.

Low ScvO2 (<70%) suggests inadequate DO2 and is associated with higher mortality rates. However high ScvO2 (>85%) is also commonly encountered in septic shock and is associated with even higher mortality rates [17]. High ScvO2 represents inadequate oxygen utilization or vascular shunting and how best to manage microvascular dysfunction and cellular dysoxia is an area of burgeoning research.

The utility of ScvO2-guided resuscitation as described by Rivers’ et al. has been investigated in three independent, multicentre, randomized trials of EGDT in the United States (ProCESS) [3]‌, Australasia (ARISE) [4], and the United Kingdom (ProMISE) [5]. No evidence of a survival benefit has been demonstrated. Given concerns regarding the generalizability of Rivers’ trial findings, the complexity and potential risks associated with elements of the resuscitation algorithm (e.g. blood, dobutamine), financial and infrastructure implications, and limited adoption despite incorporation into the SSC guidelines, widespread implementation of EGDT for the management of patients with septic shock cannot be recommended at this stage.


Hyperlactataemia occurs in the setting of oxygen supply/demand imbalance and/or microcirculatory or cellular dysfunction. Serum lactate is a useful parameter to guide resuscitation as it is rapidly and easily measured at point-of-care and may be more readily available than continuous ScvO2 monitoring. However, lactate is not specific for cellular hypoxia and can be elevated with β‎-agonists, some iv fluids and reduced hepatic clearance.

The utility of lactate as a marker of severity of sepsis has been reported by several investigators. In a retrospective study of 830 severe sepsis patients, adjusted mortality was increased with higher presenting lactate levels [18]. Normotensive septic patients with a lactate >4 mmol/L also have a similar mortality risk as hypotensive septic patients.

Lactate clearance may also provide a useful resuscitation guide in severe sepsis. Observational studies suggest that failure to achieve ≥10% lactate clearance during resuscitation is a better predictor of mortality than ScvO2 or persistent hypotension. In controlled studies, lactate clearance was non-inferior to ScvO2 as a resuscitation target and in a separate trial of ICU patients with lactate >3 mmol/L (39% sepsis), targeting therapy to achieve a ≥20% lactate clearance every 2 hours increased fluid administration and reduced mortality (hazard ratio 0.61, 95% CI 0.43–0.87) [19].

Other resuscitation tools

Monitoring venous–arterial CO2 difference, tissue oxygenation, and sublingual micro-circulatory flow provide fascinating insights into the pathophysiology of septic shock. Currently, these techniques remain experimental.


1. Dellinger RP, Mitchell MM, Rhodes A, et al. (2013). Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock: 2012. Critical Care Medicine, 41(2), 580–637.Find this resource:

2. Rivers E, Nguyen B, Havstad S, et al. (2001). Early goal-directed therapy in the treatment of severe sepsis and septic shock. New England Journal of Medicine, 345(19), 1368–77.Find this resource:

3. The ProCESS Investigators (2014). A randomised trial of protocol-based care for early septic shock. New England Journal of Medicine, 370(18), 1683–93.Find this resource:

4. The ARISE Investigators and the ANZICS Clinical Trials Group (2014). Goal-directed resuscitation for patients with early septic shock. New England Journal of Medicine, 371(16), 1496–506.Find this resource:

5. Mouncey PR, Osborn TM, Power GS, et al. (2015). Trial of early, goal-directed resuscitation for septic shock. New England Journal of Medicine, 372(14), 1301–11.Find this resource:

6. Levy MM, Dellinger P, Townsend SR, et al. (2010). The Surviving Sepsis Campaign: results of an international guideline-based performance improvement program targeting severe sepsis. Intensive Care Medicine, 36(2), 223–31.Find this resource:

7. Kumar A, Ellis P, Arabi Y, et al. (2009). Initiation of inappropriate antimicrobial therapy results in a fivefold reduction of survival in human septic shock. Chest, 136(5), 1237–48.Find this resource:

8. Kumar A, Roberts D, Wood KE, et al. (2006). Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Critical Care Medicine, 34(6), 1589–96.Find this resource:

9. Barochia AV, Cui X, Vitberg D, et al. (2010). Bundled care for septic shock: an analysis of clinical trials. Critical Care Medicine, 38(2), 668–78.Find this resource:

10. Maitland K, Kiguli S, Opoka R, et al. (2011). Mortality after fluid bolus in African children with severe infection. New England Journal of Medicine, 364(26), 2183–95.Find this resource:

11. SAFE Study Investigators, Finfer S, McEvoy S, et al. (2011). Impact of albumin compared to saline on organ function and mortality of patients with severe sepsis. Intensive Care Medicine, 37(1), 86–96.Find this resource:

12. Delaney AP, Dan A, McCaffrey J, and Finfer S. (2011). The role of albumin as a resuscitation fluid for patients with sepsis. A systematic review and meta-analysis. Critical Care Medicine, 39(2), 386–91.Find this resource:

13. Perner A, Haase N, Guttormsen A, et al. (2012). Hydroxyethyl starch 130/0.42 versus Ringer’s acetate in severe sepsis. New England Journal of Medicine, 367(2), 124–34.Find this resource:

14. Myburgh JA, Finfer S, Bellomo R, et al. (2012). Hydroxyethyl starch or saline for fluid resuscitation in intensive care. New England Journal of Medicine, 367(20), 1901–11.Find this resource:

15. Myburgh JA, Higgins A, Jovanovska A, et al. (2008). A comparison of epinephrine and norepinephrine in critically ill patients. Intensive Care Medicine, 34(12), 2226–34.Find this resource:

16. Russell J, Walley K, Singer J, et al. (2008). Vasopressin versus norepinephrine infusion in patients with septic shock. New England Journal of Medicine, 358(9), 877–87.Find this resource:

17. Pope JV, Jones AE, Gaieski DF, et al. (2010). Multicentre study of central venous oxygen saturation (ScvO2) as a predictor of mortality in patients with sepsis. Annuals of Emergency Medicine, 55(1), 40–6.Find this resource:

18. Mikkelsen ME, Miltiades AN, Gaieski DF, et al. (2009). Serum lactate is a associated with mortality in severe sepsis independent of organ failure and shock. Critical Care Medicine, 37(5), 1670-77.Find this resource:

19. Jansen TC, van Bommel J, Schoonderbeek FJ, et al. (2010). Early lactate-guided therapy in intensive care unit patients. A multicentre, open-label, randomised controlled trial. American Journal of Respiratory Critical Care Medicine, 182(6), 752–61.Find this resource: