See also Toxic Shock Syndrome

I. Pathophysiology

  1. O2 delivery is usually normal or supranormal
  2. Usually associated with Gram(-); can also be Gram(+), fungal
  3. Early ("warm") shock :
  1. Low systolic BP but relatively normal SV and pulse pressure
  2. Normal or high CO
  3. Patients are vasodilated with warm dry skin, "relative hypovolemia"
  4. Decreased SVR, marked tachycardia and tachypnea
  5. O2 consumpt. is despite normal or delivery
  6. ABG may reveal moderate respiratory alkalosis
  1. Late (Cold) Shock :
  1. Increased capillary permeability (bradykinin)hypovolemia
  2. Cell membrane dysfn (O2 resistance)Na+, Ca2+, H2O can enter the cell
  3. Altered serum Ca2+ leading to impaired myocardial function
  4. Rapid rise in lactate, HC03acidosis vasoconstriction ( SVR)skin becomes cold, clammy, mottled, cyanotic
  5. PAP, PVR high CVP despite low PCWP
  6. Intravascular coagulation (small vessels), resulting in consumption of endogenous anticoagulants
  7. Increasing organ dysfunction

II. Diagnosis

  1. Fever, most frequently ass'd with UTI in the elderly
  2. Onset w/shaking chill, rapid rise in temperature
  3. Tachycardia
  4. Labs:
  1. WBC w/left shift, resp. alkalosis & met. acidosis
  2. Hyperglycemia
  3. At least 50%will have repeat neg. blood cx
  4. Decreased platelets and increased coags PT/PTT
  1. Identifying the causative organism
    1. A chromatographic assay of urine for Streptococcus pneumoniae C polysaccharide (checked no later than 24h after abx started) in a study of hospitalized adult pts with bacteremia had sensitivity of 82% and a specificity of 97% (for identification of S. pneumoniae as the etiology of the infection) compared with blood cx. (J. Clin. Microbiol. 41:2810, 2003--JW)

III. Management

  1. Control primary process e.g. with antibiotics
  2. Obtain cultures immediately
  3. Activated Protein C-Drotrecogin Alfa (Xigris)
    1. In a randomized trial in 1690 pts with severe sepsis randomized to 4d infusion of Xigris vs. placebo, 28d mortality rate was sig. lower in active tx group (24.7% vs. 30.8%); serious bleeding was nonsig. more common in active tx patients (3.5% vs. 2.0%) (NEJM 344:699, 2001--JW)
    2. In a post-hoc analysis of data from the above study, the benefit was seen to be limited to those patients with APACHE II scores > 24 (NEJM 347:1027, 2002--JW)
    3. In a study in 2,640 adults with severe sepsis and APACHE II scores < 25 or only single-organ dysfunction randomized to drotrecogin alfa (via single 96h infusion) vs. placebo; 28d incidence of mortality were not sig. diff. in the two groups but incidence of serious bleeding was sig. greater among drotrecogin alfa recipients (2.4% vs. 1.2% during the infusion period). (NEJM 353:1332, 2005--JW)
    4. In a study in 477 children aged 38wks-17y w/sepsis-induced cardiovascular and respiratory failure randomized to drotrecogin alfa vs. placebo x 96h, there was no sig. diff. in time to complete organ failure resolution, or in 28d mortality ("RESOLVE" Trial; Lancet 369:836, 2007--JW).
    5. Can be associated with risk of severe bleeding
  4. Antithrombin-III
    1. In a study in 2,339 pts with severe sepsis, AT-III c/w placebo did not have any sig. mortality benefit in the overall cohort.  However, in a subgroup analysis, among pts with intermediate predicted mortality risk (30-60% on the Simplified Acute Physiology Score II), 90-day mortality was sig. lower in the AT-III group, though AT-III group had sig. higher incidence of bleeding events (Crit. Care Med. 34:285, 2006--JW)
  5. Corticosteroids
    1. 40 pts with septic shock randomized to Hydrocortisone 100mg then 0.18mg/kg/h, gradually tapering after shock reversed, vs. placebo; median time on vasopressors sig. less in hydrocortisone group (2d vs. 7d); hydrocortisone also ass'd with nonsig. trends toward less organ dysfunction; no sig. diff. in in-hospital death rate (Crit. Care Med. 27:723, 1999--JW)
    2. In a randomized trial of 299 adults with septic shock ranodmized to hydrocortisone 50mg IV Q6h + fludrocortisone 50ug PO QD vs. placebo x 7d, 28d mortality was sig. less (53% vs. 63%) in the steroid group in the subset of pts who had adrenal insufficiency at enrollment ("nonresponders" to a corticotropin test); among the subset (n = 70) who did not have adrenal insufficiency, there was no sig. diff. in mortality (JAMA 288:862, 2002--abst)
    3. In a meta-analysis of 16 randomized trials of corticosteroids vs. placebo in 2,063 pts w/severe sepsis, steroids were not ass'd with sig. differences in 28d mortality overall except in the subgroup of pts tx'd with 5d or more of low-dose tx (300mg or less/day of hydrocortisone equivalents) (RR 0.8). Steroid tx was not ass'd with increased risk for GI bleeding, superinfections, or hyperglycemia (BMJ 329:480, 2004--JW)
    4. In a meta-analysis of 20 studies involving 2,384 pts, corticosteroid use was associated with a sig. reduction in 28-day mortality (RR 0.92); corticosteroid use was associated with sig. higher incidence of hyperglycemia and hypernatremia (JAMA 391:2388, 2009-JW)
    5. Treatment of corticosteroid-associated hyperglycemia with insulin in patients with septic shock
      1. In a study in 509 pts with septic shock and multi-organ dysfunction, all of whom had received hydrocortisone, randomized to continuous IV insulin vs. "conventional insulin therapy", there was no sig. diff. in in-hospital mortality but pts on intensive insulin had sig. higher incidence of severe hypoglycemia (JAMA 303:341, 2010-abst)
  6. Granulocyte-macrophage colony stimulating factor ("GM-CSF")
    1. Sepsis is thought to induce immunosuppression which contributes to advancement of the condition
    2. In a study in 38 pts with severe sepsis or septic shock and low levels of HLA-DR on circulating monocytes (as a marker for immunosuppression) randomized to GM-CSF x 8d vs. placebo, the GM-CSF recipients had sig. shorter times on ventilator and sig. lower APACHE-II scores (Am. J. Resp. Crit. Care Med.)
  7. Hemoperfusion with polymyxin B filtration device
    1. Polymyxin B binds endotoxin
    2. In a study in 64 pts with severe sepsis from intra-abdominal infection requiring emergency surgery, randomized to hemoperfusion with polymyxin B + conventional therapy vs. conventional therapy alone, the study was halted when it was found that 28-day mortality was sig. lower in the hemoperfusion group (32% vs. 53%) (JAMA 301:2445, 2009-JW)
  8. Other supportive treatments
  1. Ventilatory support and O2 as needed
  2. IV fluids
  3. Keep Hgb 12-14
  4. HCO3 if needed for acidosis
  5. Inotropes (Dopamine 5-15 microg/kg/min or Dobutamine) e.g. if cardiac index remains < 3-3.5 after volume loading
  6. Vasodilators (Nitroprusside) if SVR causes decreased CO ("cold shock")
  7. Vasopressors may be needed for hypotension