PROWESS-SHOCK: Drotrecogin Alfa (Activated) in Adults with Septic Shock

PROWESS-SHOCK Study Group. N Engl J Med 2012;366:2055-64.

Clinical Question

• In patients with septic shock, does Activated Protein C (also known as drotrecogin alfa or DrotAA) compared with placebo reduce mortality?

Design

• Randomised
• Multi-centre
• Double blind
• Placebo controlled
• 6 month follow up period
• Powered at 80% with ⍺ significance level of 0.05 to detect absolute mortality difference of 7% (relative difference 20%), from expected baseline mortality rate of 35%, if 1500 patients were recruited
  • Recruitment target was increased mid-study to 1696 due to lower-than-expected mortality rate of 27.6%

Setting

• March 2008 to August 2011
• 208 sites worldwide (Europe, North and South America, Australia, New Zealand and India)

Population

• Inclusion
  • Age ≥18 years
  • Infection requiring IV antibiotics
  • ≥2 SIRS criteria
  • Septic shock, defined as:
    • At least 30 mL/kg IV fluids
    • ≥1 vasopressor for ≥4h
    • Continuing vasopressor requirements through time of randomization with goal SBP of 90 mmHg or MAP of 65 mmHg
  • Hypoperfusion in the prior 36 hours, defined as ≥1 of:
    • Metabolic acidosis, defined as base deficit of at least 5 mmol/L, bicarbonate <18 mmol/L, lactate >2.5 mmol/L
    • Urine output <0.5 mL/kg/h for an hour or a 50% increase of creatinine from baseline
  • Acute hepatic dysfunction with AST or ALT >500 IU/dL or bilirubin >2 g/dL

• Exclusion
  • Coexisting illnesses with high risk of death (e.g. metastatic cancer)

• 27816 potential patients, 1697 recruited and randomised
• Mean APACHE II scores were 25.2 (DrotAA group) and 25.5 (placebo group)

Intervention

• DrotAA 24mcg/kg of body weight per hour for 96 hours
  • Mean duration administered was 83.3 hours
  • Interrupted for procedures in 36.7%
  • Study drug stopped early in 25.9%

Control

• Placebo dissolved in 0.9% saline solution at equivalent infusion rate
  • Mean duration administered was 85.1 hours
  • Interrupted for procedures in 34.4%
  • Study drug stopped early in 22.9%

Outcome

• Primary outcome: 28-day mortality
  • 26.4% in intervention vs 24.2% in control
  • Relative risk 1.09 (95% CI 0.92-1.28; p = 0.31)
  • Absolute risk reduction -2.2 (95% CI -2.0-6.3; NNH = 47)
• Secondary outcomes:
  • 90-day mortality
    • 34.1% in intervention vs 32.7% in control
    • Relative risk 1.04 (95% CI 0.90-1.19; p = 0.56)
    • Absolute risk reduction -1.36 (95% CI-3.17-5.89; NNH = 74)
  • SOFA score at day 7
    • No difference:
      • Cardiovascular
      • Respiratory
      • Renal
      • Coagulation
      • Liver
  • Adverse Events
    • Any serious events by day 28
      • 14.3% in intervention vs 11.5% in control
      • Relative risk 1.23 (95% CI 0.96-1.59; p = 0.11)
      • Absolute risk reduction -2.76 (95% CI -0.46-5.98; NNH = 37)
  • Any bleeding event during treatment period
    • Non-serious
      • 8.6% in intervention vs 4.8% in control
      • Relative risk 1.80 (95% CI 1.23-2.61; p = 0.002)
      • Absolute risk reduction -3.84 (95% CI 1.44-6.24; NNH = 27)
    • Serious
      • 1.2% in intervention vs 1.0% in control
      • Relative risk 1.25 (95% CI 0.49-3.15; p = 0.81)
      • Absolute risk reduction -0.24 (95% CI -0.75-1.23; NNH = 417)
    • Any cerebral haematoma, cerebral haemorrhage, subarachnoid haemorrhage, haemorrhagic stroke by day 28
      • 0.4% in intervention vs 0.4% in control
      • Relative risk 1.00 (95% CI 0.20-4.90; p = 1.00)

Authors’ Conclusions

• DrotAA did not significantly reduce mortality at 28 or 90 days, as compared with placebo, in patients with septic shock.

Strengths

• Randomised, blinded, placebo-controlled
• Targeted the group of patients identified by PROWESS as most likely to benefit i.e. the sickest patients
• Adaptive trial design: statistical power maintained by increasing recruitment numbers when mortality < 30%

Weaknesses

• Funded by manufacturer
• Each centre recruited an average of 8.2 patients over the 3 years and 5 months period. This very low recruitment rate for severe sepsis (a not uncommon syndrome) is unexplained.