From Wiki Journal Club
Jump to: navigation, search
Bouadma L, et al. "Use of procalcitonin to reduce patients' exposure to antibiotics in intensive care units". The Lancet. 2010. 375(9713):463-74.
PubMedFull textPDF

Clinical Question

In critically ill ICU patients with suspected bacterial infections, what is the benefit of a procalcitonin-guided strategy on mortality and antibiotic-free days?

Bottom Line

Among critically ill patients with suspected bacterial infections, a procalcitonin-guided antibiotic strategy to treat suspected bacterial infections led to fewer days of antibiotic exposure without a >10% increase in mortality at 28 or 60 days.

Major Points

Prolonged courses of antibiotics increase selective pressure in a population of bacteria to allow resistant bacteria to thrive, and exposure of bacteria to suboptimal antibiotic concentrations increases the frequency of antibiotic resistant organisms. As a result, antimicrobial resistance has emerged as a public health issue of increasing magnitude, and finding effective solutions to address this problem is a critical focus of the CDC.

Procalcitonin, one of the precursors to calcitonin, is a useful biomarker for the diagnosis of bacterial sepsis. The Procalcitonin to Reduce Antibiotic Treatments in Acutely ill patients (PRORATA) trial aimed to establish the effectiveness of an algorithm based on procalcitonin to reduce unnecessary antibiotic exposure in critically ill patients with suspected sepsis. PRORATA randomized 630 critically ill patients in France with suspected bacterial infections to procalcitonin or control. In the procalcitonin arm, the trial used a procalcitonin-guided strategy to help physicians start, continue, or stop antibiotics. However, in the end, the final decision to start or stop antibiotics was at the discretion of treating physicians. In the control arm, patients received antibiotics according to conventional guidelines.

The PRORATA trial demonstrated that the mortality of patients in the procalcitonin arm was non-inferior to those in the control group at day 28 and at day 60, using 10% as the margin of non-inferiority. However, patients in the procalcitonin group had approximately 25% more antibiotic-free days than those in the control group (14.3 days vs 11.6 days, P<0.0001) in the intention-to-treat analysis. This strategy could be beneficial for reducing antibiotic resistance in the ICU.

There were multiple criticisms regarding this trial. A major criticism centered around the use of 10% margin of non-inferiority for mortality. If a 5% margin had been used instead, the procalcitonin arm would have had inferior mortality at 60 days. Also, more than half of the patients randomized to the procalcitonin arm were not given algorithm-guided treatment, either because it was overruled by the treating physician, or because the patients were discharged from the ICU. However, after excluding patients who were not treated according to protocol, the procalcitonin arm still had significantly more days without antibiotics (3.2 days) without a significant difference in mortality.


Surviving Sepsis Campaign severe sepsis and septic shock (2016, adapted)[1]

  • Suggest measuring procalcitonin to support shortening duration of antimicrobial therapy in patients with sepsis (weak recommendation, low quality of evidence)
  • Suggest using procalcitonin to support the decision to discontinue empiric antibiotics in patients who initially seemed to have sepsis but subsequently had limited evidence of an infection (weak recommendation, low quality of evidence)


  • Multicenter, open-label, parallel-group, randomized, controlled trial
  • N=630 critically ill patients with suspected bacterial infections
    • Procalcitonin (n=311)
    • Control (n=319)
  • Setting: 7 (5 medical, 2 surgical) ICUs in 5 university-affiliated hospitals, 1 medicosurgical ICU in a general hospital, in France
  • Enrollment: 2007-2008
  • Analysis: Intention-to-treat
  • Primary outcomes: Mortality at days 28 and 60 (non-inferiority analysis), and number of days without antibiotics by day 28 (superiority analysis)


Inclusion Criteria

  • Age ≥18 years
  • Suspected bacterial infection before study inclusion or sepsis during ICU stay
    • If suspected bacterial infection before study inclusion, must have received no antibiotics or antibiotics for <24h

Exclusion Criteria

  • Known pregnancy
  • Expected stay in ICU <3 days
  • BMT or chemotherapy-induced neutropenia (<500 neutrophils/mL)
  • Infections for which long-term antibiotic treatment is strongly recommended (i.e., infective endocarditis, osteoarticular infections, anterior mediastinitis after cardiac surgery, hepatic or cerebral abscesses, chronic prostatis, infection with Mycobacterium tuberculosis, Pneumocystis jirovecii, or Toxoplasma gondii)
  • Poor chance of survival, defined as SAPS II >65 points at screening
  • Code status: DNR

Baseline Characteristics

At admission to ICU

  • Mean age: 61.6 years
  • Male: 66%
  • Severe comorbidities
    • NYHA III/IV CHF: 4.5%
    • IDDM: 8%
    • Cirrhosis: 5.5%
    • O2 therapy at home 6.5%
    • CKD on dialysis 5%
    • Metastatic cancer: 2.5%
    • Immunocompromised: 15.5%
  • SAPS II: 17.6
  • SOFA score: 4.7
  • Organ or system failure, defined by SOFA score >2
    • Respiratory: 46%
    • Cardiovascular: 45.5%
    • Renal: 18%
    • CNS: 36.5%
    • Hepatic: 5.5%
    • Coagulation: 8%
  • Reason for admission to ICU
    • Septic shock: 17.5%
    • Non-septic shock: 15%
    • Acute respiratory failure: 38.5%
    • Renal failure: 2.5%
    • Neurological failure: 11%
    • Multiorgan failure: 6.5%
    • Other: 9%

At inclusion to study

  • SAPS II: 15.8
  • SOFA score: 4.4
  • Organ or system failure, defined by SOFA score >2
    • Respiratory: 46%
    • Cardiovascular: 42%
    • Renal: 17%
    • CNS: 32.5%
    • Hepatic: 4%
    • Coagulation: 7.5%
  • Mechanical ventilation: 67.5%
  • Type of infection
    • Community-acquired infection: 52.5%
    • Hospital-acquired infection: 47.5%
  • Septic shock: 43%
  • Serum lactate >2 mmol/L: 37.5%
  • Positive blood cultures: 17.5%
  • Infection site
    • Pulmonary: 72.5%
    • Urinary tract: 7.5%
    • Skin and soft tissue: 2%
    • Intra-abdominal: 6%
    • CNS: 2.5%
    • Catheter-related infection: 1.5%
    • Primary blood stream infection: 3.5%
    • Other: 3.5%


  • Randomized to procalcitonin vs. control arms
  • Investigators blinded to assignments before, but not after, randomization, as per open-label design
  • Procalcitonin arm: antibiotics were started or stopped based on predefined cut-off ranges of procalcitonin concentrations
    • Starting of antibiotics discouraged if <0.5µg/L and strongly discouraged if <0.25µg/L; procalcitonin remeasured 6-12h later
    • Starting of antibiotics encouraged if ≥0.5µg/L and strongly encouraged if ≥1µg/L
    • Stopping of antibiotics encouraged if ≥80% decrease from peak concentration or absolute concentration <0.5µg/L and strongly encouraged if absolute concentration <0.25µg/L
    • Continuing of antibiotics encouraged if <80% decrease from peak concentration or absolute concentration ≥0.5µg/L and strongly encouraged if increased concentration and concentration ≥0.5µ/L
  • Control group received antibiotics according to present guidelines
    • Physicians free to decide optimum duration of therapy based on clinical assessment
  • Drug selection and final decision to start or stop antibiotics were ultimately at the discretion of physician in both arms.


Comparisons are procalcitonin vs. control.

Primary Outcomes

28-day mortality
21.2% vs. 20.4% (P value not stated)
60-day mortality
30.0% vs. 26.1% (P value not stated)
Number of days without antibiotics by day 28
14.3 vs. 11.6 days (P<0.0001)

Secondary Outcomes

6.5% vs. 5.1% (P=0.45)
34.5% vs. 30.9% (P=0.29)
Number of days without mechanical ventilation
11.1 vs. 10.9 days (P=0.47)
SOFA score
Day 1: 4.4 vs. 4.4 (P=0.39)
Day 7: 4.2 vs. 4.2 (P=0.73)
Day 14: 3.5 vs. 3.6 (P=0.87)
Day 21: 3.3 vs. 3.1 (P=0.52)
Day 28: 3.0 vs. 2.4 (P=0.037)
Length of stay in ICU from inclusion
16.1 vs. 14.1 days (P=0.23)
Length of stay in hospital from inclusion
19.3 vs. 18.3 days (P=0.87)
Multidrug-resistant bacteria
17.9% vs. 16.6% (P=0.67)
Days of antibiotic exposure per 1000 inpatient days
653 vs. 812 days (P<0.0001)
Duration of first episode of antibiotic treatment
Overall: 6.1 vs. 9.9 days (P<0.0001)
Community-acquired pneumonia: 5.5 vs. 10.5 days (P<0.0001)
Ventilator-associated pneumonia: 7.3 vs. 9.4 days (P=0.021)
Intra-abdominal infection: 8.1 vs. 10.8 days (P=0.29)
Urinary tract infection: 7.4 vs. 14.5 days (P=0.0053)
Infection with positive blood culture: 9.8 vs. 12.8 days (P=0.06)


  • Although multicenter and randomized, design was open-label and only 8 ICUs participated.
  • Surgical ICU patients represented only 10% of study population.
  • Many physicians in the procalcitonin group did not follow the recommended algorithm for antibiotic discontinuation. 53% of patients randomized to procalcitonin arm did not follow the algorithm-guided treatment
  • A slightly higher number of patients in the procalcitonin arm died between days 29 and 60. The analysis used a 10% margin of non-inferiority threshold for mortality. If a 5% margin of non-inferiority for mortality had been used instead, the procalcitonin arm would have had inferior mortality at 60 days.
  • No formal cost-effectiveness analysis done.


Authors with multiple disclosures.
  1. Rhodes A, et al. "Surviving Sepsis Campaign: International guidelines for management of sepsis and septic shock: 2016." Critical Care Medicine. 2017;45(3)1-67.