Sepsis

Introduction

• Mortality of patients presenting to ED with temp >38 = 3%
  • 6% require ICU
• Cryptic shock
  • Arterial lactate 2.0-3.9 with SBP >90 and not requiring ventilation
  • 25% deteriorate and need ICU care
  • 12.5% die
• 50% of patients with sepsis have temp <38.3
  • Mortality inversely correlated with temperature (decreases 5% for every degree rise)
• BP normal in 20% of severe sepsis
• Only 50% have SBP <90 despite elevated lactate
• Bacteraemia only detected in 35-50% of all severe sepsis cases

Definitions

• Sepsis: Life-threatening organ dysfunction caused by a dysregulated host response to infection (Sepsis-3)
  • Clinical criteria
    • Organ dysfunction is defined as an increase of 2 or more points in SOFA score
    • Equates to overall mortality rate of 10%
  • SOFA score
    • Sepsis-related organ failure assessment score or Sequential OFA
    • Complex point scoring system for respiratory, CNS, CVS, liver, renal, coags
  • qSOFA
    • 2 or more of below = likely to have prolonged ICU stay or die in hospital
    • Hypotension (SBP <= 100mmHg)
    • Altered mental status (GCS <15)
    • Tachypnoea (RR>=22)

• Septic shock: Subset of sepsis with circulatory and cellular/metabolic dysfunction associated with higher rates of mortality
  • Clinical criteria
    • Sepsis (despite adequate volume resuscitation) with both of:
      • Persistent hypotension requiring vasopressors to achieve MAP of 65
      • Lactate >= 2
  • Hospital mortality >40%
• Severe sepsis no longer used

Surviving Sepsis Campaign

• Initial resuscitation
  • At least 30mL/kg IV crystalloid within first 3 hours
  • Frequent re-assessment
  • Dynamic > static variables
  • Initial target MAP >65
  • Guide resuscitation to normalise lactate

• Cultures
  • Guidelines state 45 minutes as a reasonable maximal delay to achieve collection of cultures from suspected sites of infection
  • Discourages against pan-culture and instead targeted culture is preferred as prevents inappropriate antimicrobial therapy for potential commensals
  • Intravascular catheters should have a set of cultures taken off simultaneously but data is conflicting whether positive culture at >2 hours earlier than peripheral samples constitutes evidence for catheter being source of infection

• Antibiotics
  • Within 1 hour of recognition of sepsis or septic shock
  • 4% increase in mortality for every hour delay
  • Survival reduced 5-fold if inappropriate empiric agent used
  • Antifungals
    • Risk factors for invasive candida infection include immunocompromise, prolonged invasive devices, TPN, necrotising pancreatitis, recent abdominal surgery, prolonged broad-spectrum antibiotic use, prolonged admission, recent fungal infection, multisite colonisation
    • Empiric anidulafungin/caspofungin/ampho B is preferred for most
    • Azoles are acceptable if less unwell

• Antibiotic therapy
  • 7-10 days sufficient duration for most cases
    • Not S. aureus bacteraemia, slow clinical response, undrainable focus, some fungal/viral infections (herpes, CMV) or immunodeficiencies
  • <7 days is as effective for rapid clinical improvement in acute pyelonephritis (with/without bacteraemia), uncomplicated cellulitis, SBP
  • Uncomplicated S. aureus bacteraemia requires at least 14 days
    • Defined as exclusion of endocarditis, no implanted prostheses, negative BC at 2-4 days, defervescence within 72 hours of treatment and no evidence of metastatic infection
  • Complicated S. aureus bacteraemia requires 6 weeks (endovascular duration)

• Galactomannan and beta-D-glucan testing in assessment of invasive Aspergillus (and other fungal pathogens) is accepted
  
• Procalcitonin
  • Low quality evidence that can be used for de-escalation of therapy and reduce duration of therapy without adverse effect on mortality (and maybe benefit)
  • Strongest evidence for confirming bacterial infection is in bacterial pneumonia vs. non-infectious pulmonary pathology

• Source control
  • Target of 6-12 hours at most
  • Prolonged efforts at medical stabilisation in deteriorating patients is likely to be harmful if it impedes source control
  • Intravascular devices suspected of being the source need to be removed promptly once alternative access is achieved
  • As long as patient is not in septic shock or suffering fungaemia, if removal of catheter is not practical (e.g. tunnelled catheter), prolonged antimicrobial courses may be preferred

• Fluids
  • Initial 20mL/kg bolus crystalloid
  • Targets: HR< 100, SBP >100, MAP >65, JVP >3cm, adequate IVC filling on USS, UO >0.5mL/kg and ScvO2 >70% within 6 hours
  • Fluid challenge technique advised  beyond initial resuscitation
  • Crystalloids advised over colloids unless very large volumes of crystalloids required
  • Overall weak recommendation

Vasopressors

• Vasopressors
  • Noradrenaline first-choice 
  • Recommend adding vasopressin up to 0.03U/min or adrenaline to raise MAP to target or adding vasopressin to decrease noradrenaline dosage
  • Dopamine recommended as alternative to noradrenaline only in highly selected patients (low risk of tachyarrhythmias and absolute or relative bradycardia)
  • Dobutamine recommended if persistent hypoperfusion despite adequate fluid loading and vasopressor use

• Noradrenaline vs. dopamine
  • NA increases MAP through vasoconstriction with little change in HR and less increase in SV compared to dopamine
  • Dopamine increases MAP and cardiac output through increase in SV and HR
  • Noradrenaline is more potent than dopamine and may be more effective at reversing hypotension in septic shock
  • Dopamine may be helpful if reduced systolic function but carries risk of arrhythmia and causes more tachycardia than noradrenaline
  • Noradrenaline has mortality benefit and less risk of arrhythmia than dopamine

• Noradrenaline vs. Adrenaline
  • No difference in mortality
  • Adrenaline increases lactate production and may preclude lactate clearance guided resuscitation
• Vasopressin
  • Levels are lower in septic shock than other shock states
  • Low doses may improve MAP refractory to other vasopressors
  • VASST trial (NA vs. NA + Vasopressin)
    • No difference in outcome but subgroup analysis showed potential improved outcomes if NA dosing limited to <15 via vasopressin
  • Advised to use with caution if not euvolaemic or at doses >0.03

• Dobutamine
  • Routine supranormal CO targets with dobutamine does not improve outcomes
  • If reduced CO AND reduced vital organ perfusion with adequtate volume resuscitation and MAP – dobutamine may be indicated
  • Target at perfusion (lactate, ScvO2, clinical) vs. MAP
• Milrinone
  • Phosphodiesterase inhibitor increases cAMP levels to increase CO
• Levosimendan
  • Increased cardiac myocyte calcium responsiveness and opens ATP-dependent K channels to produce inotropic and vasodilatory properties
  • No mortality benefit

• Metaraminol
  • Alpha-1 agonist and releases NA from stored sites
  • 25% converted to NAD
  • 1-2 min onset with effect duration 20 – 60 minutes

Other therapies

• Corticosteroids
  • No benefit to routine hydrocortisone 200mg/day
  • If failure to restore haemodynamic stability – 200mg/day indicated
  • Random cortisol levels or determination of ‘relative adrenal insufficiency’ has no proven benefit
  • Taper steroids once vasopressors no longer required (weak evidence)
• Blood products
  • Transfuse once <70 in absence of MI, severe hypoxaemia or acute haemorrhage
  • TRISS trial 70 vs. 90 showed similar 90-day mortality, ischaemic events and use of life support
  • EPO advised against
  • FFP advised against in the absence of bleeding or planned invasive procedures
    • Usually fails to correct PT in non-bleeding patients with mild coagulation abnormalities
    • No studies suggest benefit if not bleeding or undergoing procedure even with severe coagulation abnormalities
  • Platelets
    • Advise to replace if <10 or <20 if significant risk of bleeding
    • Target >50 if actively bleeding, surgery or invasive procedures

• Immunoglobulins
  • IVIG not recommended as no survival benefit in the only large study
• Blood purification
  • Lack of evidence but may be of benefit
• Glucose control
  • Recommend insulin once two consecutive BSL >10mmol/L and target this as upper limit
  • More intensive therapy (as in NICE-SUGAR) showed increased hypoglycaemia episodes and increased mortality
• RRT
  • Higher doses of RRT show no benefit (RENAL study)
  • ’Early’ RRT not advisable if no clinical indication for it exists

• Thromboprophylaxis
  • Alhazzani et al. showed mortality benefit of 7% with LMWH vs. UFH (not statistically significant though)
  • Reduced HIT with LMWH, easier to administer, cheaper once delivered and less VTE

• Lactate in sepsis
  • 0-2.5 = 4% mortality
  • >4 = 28%
• Procalcitonin
  • In ED patients >65 a level >0.2ng/mL (mcg/L)
    • 80-100% sensitive for bacteraemia
    • 30-50% specific for bacteraemia (LR- 0.2)
  • In adult ICU patients, >0.28 is 92% sensitive and 95% specific for bacteraemia
  • Interpretation
    • <0.5 = Low risk of systemic bacterial/fungal infection
    • 0.5-2 = High risk of systemic bacterial/fungal infection
    • 2-10 = High risk of sepsis/progression to septic shock
    • >10 = High risk of septic shock

• CRP
  • 90% sensitive and 75% specific if at level of 40

Catheter-related blood stream infection (CRBSI)

• Clinical signs
  • Suspect if invasive line and no other source for BSI
  • Fever is most sensitive but specificity is poor
  • Inflammation or purulence at site is specific but poorly sensitive
  • Haemodynamic instability, ALOC, catheter dysfunction and clinical signs of sepsis that start with catheter use
  • Complications i.e. IE, thrombophlebitis, osteomyelitis
  • BC positive for S. aureus, S. epidermidis, Candida
  • Clinical improvement within 24 hours of removal

• When is treatment not required?
  • Positive catheter tip culture in absence of clinical signs of infection
  • Positive BC from catheter with concomitant negative peripheral cultures in absence of clinical features
  • Phlebitis in absence of infection

• Catheter management
  • Removal
    • Severe sepsis
    • Metastatic infection
    • Suppurative thrombophlebitis
    • Persistent bacteraemia for 72 hours despite therapy to which organism is susceptible
• Salvage
  • Only if uncomplicated CRBSI (i.e. not septic or metastatic infection) with long-term catheters and NOT S. aureus, P. aeruginosa, fungi or mycobacteria

• Antibiotics
  • Vancomycin
  • Gram-negative cover (PipTaz or Gent) if neutropaenic, septic
  • Treat known MDR colonization
• Empiric candida therapy if:
  • TPN
  • Prolonged broad-spectrum antibiotic use
  • Haematological malignancy
  • Transplant
  • Femoral catheterization
  • Colonisation with Candida at multiple sites

Sepsis in the older person (Burkitt)

• 2/3 of cases are in the over 65’s
• Incidence rate increasing 20% faster than young people
• When present to ED are:
  • More unwell
  • Higher levels of reversible organ dysfunction
  • Higher mortality
  • Higher morbidity if survive
  • 1/3 not back to baseline at 6 months
• Age increases risk of infection, bacteraemia as a result of infection and sepsis

Sepsis in the older person

• Immunosenescence (cell-mediated and humoral)
• Imbalance of pro- and anti-inflammatory cytokines
• Increased procoagulant state
• Comorbidities
• Immunosuppressant medications
• Reduced cardiopulmonary response to infection
• Increased indwelling devices
• MROs
• Malnutrition and frailty

Sepsis in the older person

• Exhibit fewer clinical signs
• Risk factors:
  • Recent hospitalization
  • Invasive procedures
  • Frequent presentations
  • Immunosuppression
• Fever absent in 50% of frail older persons with serious infections
• Less likely to suffer specific signs of bacteraemia
  • Hypothermia
  • Rigors
  • Sweating
  • Altered mental status
  • Leukopaenia
  • Lymphopaenia

Sepsis in the older person

• Tachycardia response is blunted due to reduced myocardial responsiveness or beta-blockade
• Tachycardia may be incorrectly attributed to AF/CCF
• Relatively high baseline bp means under-recognition of hypotension
• SIRS and qSOFA criteria 
  • Sensitivity of 65% and 28% respectively for mortality at 30 days
  • qSOFA remains highly specific

Sepsis in the older person

• Management
  • Collaborative decision making with patient, family, usual physicians, ICU
  • Must educate re: likelihood of returning to baseline even if survives
  • Timely management produces 16% absolute RR of mortality in the elderly with septic shock
  • Target MAP should be 65mmHg as for others as studies are conflicting when higher targets are used
    (65 study)
    • Although urine output/mentation may be better indicators
    • Some evidence that matching a patients pre-illness basal perfusion pressure reduces the risk of kidney injury in septic patients requiring vasopressors (REACT Shock Study)

Fungaemia

Candidaemia is the most common invasive fungal infection seen in adults, particularly those in ICU.

Risk factors

• CVL’s
• TPN
• Broad-spectrum antibiotics
• Acute renal failure
• Prior surgery, especially upper GI
• GI tract perforations/anastamotic leaks
• Pancreatitis
• Malnourished
• Immunosuppression
• IVDU
• Urinary tract obstruction/instrumentation if Candida exists as a commensal organism

Distinguishing between localised candida infection i.e. candiduria and invasive candidaemia/pyelonephritis is particularly difficult.

Candiduria is common and the vast majority of patients have only local commensal fungal growth. Patients presenting unwell, with clinical pyelonephritis, radiological kidney involvement and/or with risk factors above warrant empirical antifungal therapy.

Clinical clues that suggest invasive haematogenous spread of Candida include characteristic eye lesions (chorioretinitis), skin lesions and muscle abscesses.

Multiorgan involvement is then heralded by involvement of heart (infective endocarditis), kidneys (abscess, pyelonephritis), liver, spleen, lungs, eyes and brain.

Last Updated on September 21, 2022 by Andrew Crofton