In 2004, experts in the diagnosis and management of infection and sepsis from 11 international medical organizations published the first guidelines for improving the prognosis of severe sepsis and septic shock.
These guidelines represent Phase II of the Save Sepsis Campaign (SSC), an international effort to increase awareness of severe sepsis and improve its prognosis. In conjunction with several other organizations, this working group met again in 2006 and 2007 to update the guideline document with a new evidence-based methodological system to assess the quality of evidence and strength of recommendations. These recommendations are intended to be used to guide clinicians in the treatment of patients with severe sepsis and septic shock. It is important to note that the recommendations in these guidelines are not a substitute for clinician decision making when physicians are faced with unique clinical indicators for specific patients.
GRADE System 1 (strong recommendation: do or don’t do) 2 (weak recommendation: may or may not do)
A (high quality randomized controlled study (RCT) or meta-analysis study)
B (moderate quality RCT or high quality observational and cohort study)
C (well-done, controlled observational and cohort studies)
D (case summary or expert opinion, low quality study)
A. Initial resuscitation
Sepsis shock is characterized by inadequate tissue perfusion, persistent hypotension, blood lactate ≥ 4 mmol/L, and transfer to an ICU unit for treatment as soon as possible after the onset of hypotension Objectives for the first 6 hours of resuscitation
a) Central venous pressure (CVP): 8-12 mmHg b) Mean arterial pressure (MAP) ≥ 65 mmHg c) Urine output ≥ 0.5 ml/kg/hd) Central venous (superior vena cava) oxygen saturation ≥ 70% or mixed arteriovenous oxygen saturation ≥ 65% (1C) e) CVP has reached the target but ScvO2 still cannot reach 70% or SvO2 still cannot reach 65%, then infusion of concentrated red blood cell suspension Hct ≥ 30% and/or infusion of dobutamine (maximum amount of 20 μg/kg.min) to achieve this goal (2C)
B. Diagnosis
1. At least two blood cultures should be obtained prior to antibiotic administration! That is, blood specimens from a percutaneous puncture and from an intravascular tube left in place for more than 48 hours, and other culture specimens, including urine, cerebrospinal fluid, wounds, respiratory secretions, or other body fluids that may be the source of infection, should be obtained prior to the administration of antibiotics whenever possible (1C).
2. Perform imaging as soon as possible to confirm the underlying infection (1C) {E}
C. Antibiotic therapy
1. Early intravenous antibiotic treatment within 1 hour is recommended when septic shock (1B) or severe sepsis has not yet developed (1D) is confirmed. Appropriate specimens should be obtained prior to the application of antibiotics, but antibiotics should not be delayed for the purpose of obtaining specimens (1D).
2a. The recommended initial empiric anti-infective therapy includes one or more drugs against all suspected pathogenic microorganisms (bacteria and/or fungi) and penetration into the infected lesion causing sepsis at a sufficiently high concentration (1B)
{D}
2b. Recommend daily evaluation of antibiotic regimens to achieve desired clinical outcomes, prevent the development of bacterial resistance, reduce toxicity, and reduce costs (1C)
2c. Combination therapy is recommended for patients with severe sepsis due to known or suspected Pseudomonas spp. infection (2D)
2d.Empirical combination therapy is recommended for patients with neutropenia (2D).
2e. For patients with severe sepsis when applying empirical therapy, combination therapy is recommended for no more than 3-5 days. Once the pathogen is found, the most appropriate single treatment (2D) should be chosen.
3. The recommended course of treatment is generally 7-10 days, but it should be extended appropriately for patients with slow response to clinical treatment, incomplete clearance of the infected lesion or immunodeficiency (including neutropenia) (1D).
D Infection source control
1a. Specific infections requiring urgent management such as necrotizing fasciitis, diffuse peritonitis, cholangitis, and intestinal infarction should be searched for an etiology and a diagnosis established or excluded as soon as possible (1C) and within 6 hours of symptom onset (1D).
1b. All patients with severe sepsis should be evaluated to determine if a controllable source of infection exists. Controls include drainage of abscesses or localized foci of infection, debridement of necrotic tissue after infection, removal of medical devices that can cause infection, or source control of microbial infections still present (1C).
D. Source control of infection 2. It is recommended that interventions for those identified as having peripancreatic necrosis that may become a potential site of infection should preferably wait until viable tissue and necrotic tissue are clearly delineated (2B).
3. When pathogenic treatment is required, effective interventions with minimal physiologic damage, such as percutaneous drainage of abscesses rather than surgical drainage, are recommended (1D).
4. Endovascular devices that may become foci of severe sepsis or septic shock infection should be removed immediately after other vascular access is established (1C).
E. Fluid therapy
1. Fluid resuscitation with natural/artificial colloid or crystalloid fluids is recommended. There is no evidence to support the superiority of one fluid over the other (1B).
a. The use of albumin has been shown to be safe and equivalent to crystalloids. b. The use of colloidal fluids significantly reduced mortality (P=0.09). c. There was no difference in the effect of crystalloid and colloid resuscitation.
d. The amount of crystalloid was significantly greater than the amount of colloid to achieve the same treatment goal. e. Crystalloids were cheaper.
2. The recommended initial therapeutic goal for fluid resuscitation is to achieve a CVP of at least 8 mmHg (12 mmHg for mechanically ventilated patients), after which further fluid therapy is usually required (1C).
3a. Fluid shock therapy with continuous fluid replacement until hemodynamics (e.g., arterial pressure, heart rate, urine output) improve is recommended (1D).
3b. For fluid shocks in patients with suspected hypovolemia, a minimum of 1000 ml of crystalloid or 300-500 ml of colloid should be administered within the first 30 minutes.
colloidal fluid. In patients with sepsis resulting in inadequate organ perfusion, a faster and higher dose of fluid must be administered (1D). 3c. In cases where there is only an increase in cardiac filling pressure (CVP or pulmonary artery wedge pressure) without
If there is no hemodynamic improvement, the rate of fluid replacement should be reduced (1D). F. Maintaining MAP at ≥65 mmHg (1C) is recommended for vasopressors.
When hypovolemia is not corrected, vasopressors should be used to ensure perfusion during hypotension. The use of norepinephrine should be gradually increased until MAP reaches 65 mmHg in order to maintain tissue perfusion. In addition, prior complications of the patient should be taken into account when setting MAP treatment goals.
2. Norepinephrine or dopamine is recommended as the vasopressor of choice for correction of hypotension in septic shock (administered as soon as possible after central venous access is established) (1C).
3a. Epinephrine, phenylephrine, or antidiuretic hormone is not recommended as the vasopressor of choice in septic shock (2C). 0.03 U/min of antidiuretic hormone combined with norepinephrine is equivalent to norepinephrine alone.
3b. If norepinephrine or dopamine is not effective, epinephrine is recommended as the drug of choice (2B).
F. Vasopressors
4. low-dose dopamine is not recommended as a renal protective agent (1A).
A large randomized clinical trial and meta-analysis showed no significant differences when comparing the effects of low-dose dopamine and placebo. Therefore, there is no evidence to support that low-dose dopamine protects renal function.
5. It is recommended that arterial access (1D) be established as soon as possible for patients requiring vasopressor drugs.
In shock, arterial catheterization of blood pressure is more accurate, the data can be analyzed repeatedly, and continuous monitoring data helps one to develop the next treatment plan based on blood pressure.
G. Positive inotropic drugs should be administered intravenously with dobutamine (1C) in the presence of elevated cardiac filling pressures and reduced cardiac output suggestive of myocardial dysfunction.
2. Oppose the use of methods that increase the cardiac index to supernormal levels.
G. Positive inotropic drugs When the patient’s left ventricular filling pressures and MAP are sufficiently high (or when fluid resuscitation therapy has been clinically assessed as adequate) and low cardiac output is measured or suspected, dobutamine is the preferred myocardial contraction drug.
If cardiac output is not monitored, a combination of a myoconstrictor/vasopressor such as norepinephrine or dobutamine is recommended.
When cardiac output and blood pressure can be monitored, a vasopressor such as norepinephrine alone may be used to achieve target MAP and cardiac output.
d. Two large prospective clinical studies of critically ill ICU patients with sepsis did not show benefit from the use of dobutamine to increase patient oxygen delivery to supra-maximal levels.
H. Glucocorticoids
In adults with septic shock, intravenous hydrocortisone is recommended only for patients whose blood pressure is insensitive to fluid resuscitation and vasopressant therapy (2C).
2. For the identification of subgroups of adult septic patients subject to glucocorticoids, ACTH excitation tests are not recommended (2B).
3. If hydrocortisone is available, dexamethasone is not recommended (2B).
4. If hydrocortisone is not available and the alternative hormone preparation does not have significant salt corticosteroid activity, an increase in oral fludrocortisone (50 μg) daily is recommended. If hydrocortisone is used, fludrocortisone is optional (2C).
5. Discontinuation of glucocorticoid therapy is recommended when the patient no longer requires vasopressors (2D).
6. For the purpose of treating sepsis, a daily glucocorticoid dosage of no more than 300 mg equivalent of hydrocortisone is recommended for patients with severe sepsis or septic shock (1A).
7. For patients with sepsis without shock, the application of hormones is not recommended. However, there are no contraindications to hormone maintenance therapy or the use of stress doses of hormones in cases where the patient requires endocrine or glucocorticoid therapy (1D).
I. Recombinant human activated protein C (rhAPC)
In adult patients with organ insufficiency due to sepsis, clinically assessed as high risk of death (most APACHE II ≥ 25 or with multiorgan failure), rhAPC therapy is recommended if there are no contraindications (2B, 2C for patients operated within 30 days).
{B} 2. In adult patients with severe sepsis and low risk of death (most APACHE II <20 or single organ failure), rhAPC therapy is not recommended (1A).
J. Blood product use
Erythrocyte transfusion is recommended for hemoglobin below 7.0 g/dl (70 g/L) to maintain hemoglobin at 7.0-9.0 g/dl (70-90 g/L) (1B)
2. Erythropoietin is not recommended as specific treatment for severe sepsis anemia, but is available for other acceptable causes such as impaired erythropoiesis induced by renal failure (1B).
3. Fresh frozen plasma is not recommended for correction of laboratory coagulation abnormalities in the absence of clinical bleeding and when no invasive procedures are planned (2D).
4. Antithrombin is not recommended in the treatment of severe sepsis and septic shock (1B).
5. In patients with severe sepsis, platelet transfusion is recommended when the platelet count is <5000/mm3 (5×109/L), regardless of the presence of bleeding. Platelet transfusion may be considered when the platelet count is 5000-30,000/mm3 (5-30×109/L) and there is a clear risk of bleeding. When surgical or invasive operations are required, platelet count should be ≥
50,000/mm3 (50×109/L) (2D).
Part II Supportive therapy for severe sepsis
A Mechanical ventilation
1. For patients with acute lung injury (ALI)/acute respiratory distress syndrome (ARDS) due to sepsis, it is recommended to set the tidal volume of mechanical ventilation at 6 ml/kg (1B) based on the predicted body weight.
It is recommended to monitor end-inspiratory plateau pressure in patients with ALI/ARDS and to set the initial plateau pressure limit to ≤30 cm H2O. The patient’s thoracic compliance should be taken into account when assessing the plateau pressure (1C).
The final recommendation is that high plateau pressure and high tidal volume ventilation should be avoided in patients with ALI/ARDS. The tidal volume should be set at 6 ml/kg for the first 1-2 hours to keep the end-inspiratory plateau pressure below 30 cmH2O. If the tidal volume is still higher than 30cmH2O at 6ml/kg, lower the tidal volume to 4ml/kg.
A Mechanical ventilation
3. To reduce the plateau pressure and tidal volume as much as possible, patients with ALI/ARDS were allowed to have hypercapnia (PaCO2 above normal, called “permissive hypercapnia”) (1C). However, the trial did not consider “permissive hypercapnia” as the primary treatment target. Such hypercarbia should be limited in patients with pre-existing metabolic acidosis and prohibited in patients with high intracranial pressure.
PEEP > 5 cmH2O is the lower limit to prevent alveolar atrophy.
5. In experienced units, prone positioning should be considered for patients with ARDS requiring high inspiratory oxygen levels (FiO2) and plateau pressures that may cause lung injury, if there is no excessive risk of changing position (2C).
Some trials have demonstrated that prone positioning improves oxygen exchange in patients, but a large multicenter study did not show that maintaining prone positioning for approximately 7 hours per day reduced mortality in patients with ALI/ARDS.
A Mechanical ventilation
6A. In the absence of contraindications, it is recommended that mechanically ventilated patients remain in the semiprone position to prevent aspiration and ventilator-associated pneumonia (VAP) (1B). 6B. It is recommended that the head of the bed be elevated 30-45 degrees (2C).
7. NIV is recommended for only a small number of patients with ALI/ARDS who meet the following criteria: mild respiratory failure (relatively low pressure support and PEEP are effective), hemodynamically stable, more comfortable and easily aroused, able to cough sputum and protect the airway on their own, and subjective expectation of early recovery. Maintaining a low threshold for tracheal intubation (2B) is recommended.
Avoiding tracheal intubation has many benefits, such as facilitating communication, reducing the chance of infection and anesthetic use. Two randomized controlled clinical trials have demonstrated that successfully performed NIV improves patient prognosis. Unfortunately, only a small proportion of patients with life-threatening hypoxemia are suitable for this approach.
8. It is recommended that an appropriate deconditioning plan be developed to assess the ability to decondition mechanically ventilated patients by performing a spontaneous breathing test, provided that the patient also meets the following conditions:
(i) arousable, (ii) hemodynamically stable (no boosters), (iii) no new potentially serious medical conditions, (iv) only low ventilation and low PEEP, and (v)
Oxygen can be administered by mask or nasal cannula to meet the oxygen concentration requirement. A low level pressure support, continuous positive airway pressure (CPAP, ≈5 cmH2O) or T-tube for autonomic breathing test (1A) should be chosen.
9. Pulmonary artery catheter application is not recommended as a routine for patients with ALI/ARDS (1A).
For patients with existing ALI and no evidence of tissue hypoperfusion, a conservative rehydration strategy is recommended to reduce the number of days of mechanical ventilation and ICU stay (1C).
B. Sedation, anesthesia, and inotropes When sedation is required in critically ill mechanically ventilated patients, anesthesia documentation should be performed and anesthesia goals set (1B). There is increasing evidence that this may reduce the duration of mechanical ventilation and the number of days in the ICU.
If anesthesia sedation is required for mechanically ventilated patients, intermittent injections or continuous drips are recommended to achieve a predetermined sedation endpoint, with daily interruptions/decreases in sedation to allow the patient to wake up/re-drip the drug (1B). Studies have shown that continuous drip sedation increases the duration of mechanical ventilation and ICU stay in patients.
Given the long duration of inotropes after discontinuation, it is recommended that inotropes (NMBA) be avoided in patients with sepsis. If it must be used, it should be given intermittently or with a 4-hour serial monitored block depth (1B) during a continuous drip.
However, a randomized controlled trial showed that the application of NBMA did not improve oxygen delivery and oxygen consumption in patients with severe sepsis.
Therefore, NBMA is not recommended when intubation or ventilation cannot be safely performed without obvious indications such as proper sedation and analgesia.
C Glycemic control For patients with severe sepsis combined with hyperglycemia who have been initially stabilized after admission to the ICU, intravenous insulin therapy is recommended for glycemic control (1B).
It is recommended to adjust the insulin dose with an effective regimen to control blood glucose below 150 mg/dl (2C).
It is recommended that all patients receiving intravenous insulin therapy use glucose as a source of calories and monitor blood glucose every 1-2 hours, and then every 4 hours once blood glucose and insulin dosage have stabilized (1C).
When monitoring peripheral blood glucose levels by rapid bedside testing, caution should be exercised if blood glucose values are low, as arterial blood or plasma glucose levels may be lower than the test value (1B).
A large randomized single-center study conducted in a cardiac surgery ICU showed that intensive intravenous insulin therapy (Leuven
regimen) to control glucose at 80-110 mg/dl was shown to reduce ICU mortality. It also reduced organ dysfunction and shortened the duration of ICU stay in patients who stayed in ICU for more than 5 days. However, the risk of hypoglycemia in patients was increased approximately threefold.
Two studies suggested a glycemic threshold between 145 and 180 mg/dl for reducing patient mortality.
An observational study with a large sample (7049 patients) found that lowering mean blood glucose levels was as important as reducing blood glucose fluctuations.
D Renal replacement therapy
1. In patients with severe sepsis combined with acute renal failure, continuous renal replacement therapy is equivalent to intermittent hemodialysis (2B).
2. In hemodynamically unstable patients, continuous renal replacement therapy is recommended to assist in maintaining fluid balance (2D).
Two meta-analyses showed no significant difference in the reduction of in-hospital mortality between continuous and intermittent renal replacement therapy.
Two studies showed that continuous therapy was more beneficial in achieving the goal of maintaining fluid balance. In conclusion, the current evidence is insufficient to draw conclusions about the mode of replacement therapy chosen in patients with sepsis complicated by acute renal failure.
E Bicarbonate therapy
In patients with hypoperfusion causing hyperlactatemia and pH ≥7.15, the use of sodium bicarbonate to improve hemodynamics or reduce the use of elevating agents is contraindicated (1B).
There is no evidence to support the use of sodium bicarbonate for the treatment of hyperlactatemia due to sepsis hypoperfusion. Two randomized blinded crossover studies showed no significant difference between equimolar saline and bicarbonate for improving hemodynamic indices, or reducing the need for boosters, in patients with hyperlactatemia, but studies were less likely to include patients with pH <7.15.
Bicarbonate may increase water-sodium load, increase blood lactate and PCO2, and decrease serum ionized calcium, but the relationship of these parameters to patient prognosis is uncertain. The effect of bicarbonate on hemodynamic parameters or the need for elevating agents in patients with low or any pH is unknown.
F Prevention of deep vein thrombosis In patients with severe sepsis, low-dose normal heparin (UFH) 2-3 times daily or daily low molecular weight heparin (LMWH) is recommended
Prevention of deep vein thrombosis (DVT) unless there are contraindications such as thrombocytopenia, severe coagulopathy, active bleeding, recent cerebral hemorrhage, etc. (1A).
2. For those with contraindications to heparin, device prophylaxis such as graduated compression stockings (GCS) or intermittent compression devices (ICD) is recommended unless there is a contraindication (1A).
3. In very high-risk patients such as those with severe sepsis combined with a history of DVT, trauma or orthopedic surgery, combined pharmacologic and mechanical prophylaxis is recommended unless contraindicated or impossible to perform (2C).
4. Given the proven benefits of LMWH in other high-risk patients, LMWH is recommended over UFH in very high-risk patients (2C).
G Stress ulcer prevention
Prevention of upper gastrointestinal bleeding due to stress ulcers with H2 receptor blockers (1A) or proton pump inhibitors (PPI) (1B) is recommended in patients with severe sepsis, but also considering that elevated intragastric pH may increase the risk of VAP.
A trial by Cook et al. that included 1200 patients and a meta-analysis showed that H2 receptor blockers were more effective than aluminum thioglycollate for acid suppression. Two studies support the equivalence of H2 receptor blockers to PPIs.
H Selective bowel decontamination Experts are divided on the issue of selective bowel decontamination
(SDD) issue is widely divided, with almost equal numbers of pro- and anti-users. Therefore, the use of SDD in patients with severe sepsis is not recommended at this time.
Experience has shown that prophylactic use of SDD (enteral non-absorbable antibiotics and short-course intravenous antibiotics) reduces infections (mainly pneumonia) and decreases overall mortality in critically ill and trauma patients without increasing the risk of resistance to Gram-negative bacteria.
Analysis of two prospective blinded studies showed that SDD reduced nosocomial (secondary) infections in patients admitted to the ICU for primary infections and reduced their mortality. The primary purpose of SDD in patients with severe sepsis or septic shock may be to prevent secondary infections.
The primary role of SDD is to prevent VAP, and therefore it is necessary to compare SDD with non-antimicrobial VAP interventions such as ventilator intervention systems. Although studies including enteral vancomycin have demonstrated its safety, there is still a potential for the emergence of drug-resistant Gram-positive infections.
I. Limitations of Support Considerations A discussion with the patient and family about further treatment plans, including possible regression and realistic treatment goals (1D), is recommended. Key recommendations
(1) Early targeted resuscitation of patients with sepsis in the first 6 hours after diagnosis (1C) Various diagnostic measures should be taken rapidly to identify possible sources of infection (1C) Treatment with broad-spectrum antibiotics within 1 hour of sepsis diagnosis (1B) The use of anti-infective drugs should be analyzed in conjunction with clinical and bacterial culture results to justify the use of appropriate narrow-spectrum antibiotics (1C) The duration of antibiotic use is generally 7-10 days, which can be adjusted according to clinical response (1D)
The choice of infection source control methods should be weighed against the pros and cons (1B) Fluid resuscitation should pay attention to circulating perfusion pressure (1C) The rate of fluid resuscitation depends on the increase in circulating perfusion pressure, but there is no improvement in tissue perfusion (1D) Key recommendations
(2) Use vasopressors norepinephrine or dopamine to maintain the initial resuscitation target mean arterial pressure ≥65 mmHg
(1C) For patients with adequate fluid replacement but still low cardiac output, use the positive inotropic agent dobutamine to increase cardiac output or combine the positive inotropic agent with a vasopressor (1C) Corticosteroids are recommended only if fluid replacement is adequate in septic shock but an elevating agent is still needed to maintain normotension (2C) For critically ill patients in septic shock with a high probability of death from clinical treatment Use of recombinant human activated protein C (2B, 2C for post-operative patients) Once the tissue is hypoperfused and there are no other comorbidities such as coronary artery disease, acute bleeding, the therapeutic target for hemoglobin should be 7-9 g/dl; use of low tidal volume (1B) and respiratory plateau pressure limitation strategy (1C) for patients with acute lung injury (ALI) or acute respiratory distress syndrome (ARDS) Key recommendations
(3) A minimum amount of positive end-expiratory pressure ventilation should be given to patients with acute lung injury (1C) This goal should be clearly defined at the outset for mechanically ventilated patients, otherwise it will lead to inappropriate treatment (1B) Avoid routine use of pulmonary artery catheter monitoring in patients with ALI/ARDS (1A) The duration of use of mechanical ventilation and ICU admission should be shortened, and conservative for ALI/ARDS patients without shock Fluid therapy strategies (1C) The use of sedative/analgesic medications should follow guidelines (1B) Either intermittent or continuous injections combined with interruption or reduction of daily continuous injections is the method of sedation management (1B) Avoid the use of inotropes if not completely necessary (1B); hyperglycemia should be controlled (1B), and once stabilized, patients with severe sepsis should maintain blood glucose at 150 mg/dL or less (2C) Continuous veno-venous hemofiltration and intermittent hemodialysis have the same effect (2B) Prevention of deep vein thrombosis (1A) Key recommendations.
(4)The use of H2 receptor blockers or proton pump inhibitors is more effective than thioglycollate for the prevention of stress ulcers.(1B)There are no studies directly comparing proton pump blockers with H2 receptor blockers in the prevention of stress ulcers.(1D)Effective recommendations for the treatment of pediatric sepsis include the important use of physical examination as the ultimate point of treatment.(2C)Dopamine as the drug of choice for maintaining blood pressure.(2C ) Corticosteroids are used only in children suspected or proven to be adrenaline deficient (2C) The recommendation is against the use of recombinant human activator protein-C.