Whenever you order a lab you must ask yourself the following questions?
Why am I ordering this test?
What am I going to do with the result?
Given that PCT has become a frequently ordered test for many of you I wanted to answer some straightforward questions, and dive into the body of evidence to help support your decision making in the Emergency Department.
What is Procalcitonin?
- Precursor of calcitonin which is involved in calcium homeostasis
- Produced by thyroid, lung and intestine w/ half-life of 25 to 30 hours.
- Below the limit of detection (10 pg/mL) in healthy patients
- Rises in response to proinflammatory stimuli, especially bacterial, mostly via lung and gut production
- Starts to rise 6 hours after the onset of infection
- “It is a helpful biomarker for early diagnosis of sepsis in the critically ill patient,” Wacker et. al. Lancet Infectious Dis, 2013
- Seems to be the new darling of critical care providers
In general, what do the results mean?
- <0.5 normal
- >0.5-2 moderate risk for progression to severe systemic infection
- >2 high risk for progression to severe systemic infection
- >10 high likelihood of severe sepsis/septic shock
Let’s take a look at some of the evidence
Procalcitonin to detect invasive bacterial infection in non-toxic-appearing infants with fever without apparent source in the emergency department
Luaces-Cubello et al
Pediatr Infect Dis J, 2012
- This was a study of 868 non-toxic, febrile infants <36 months of age. 1.7% had an invasive infection.
- The area under the receiver operating characteristic curve for procalcitonin was 0.87 (optimum cutoff 0.9 ng/mL, sensitivity 86.7%, specificity 90.5%)
- CRP = 0.79 (optimum cutoff 91 mg/L, sensitivity 33.3%, specificity 95.9%)
- In fever <8 hours duration the of <8 hours duration, the area under the receiver operating characteristic curve was 0.97 for procalcitonin and 0.76 for C-reactive protein
- Summary: The higher the area under the ROC, the better the ability to help predict disease. Especially with fever <8 hours procalcitonin was superior to CRP in this study.
Comparison of the Test Characteristics of Procalcitonin to C-Reactive Protein and Leukocytosis for the Detection of Serious Bacterial Infections in Children Presenting With Fever Without Source: A Systematic Review and Meta-analysis
Chia-Hung et al.
Annals of Emerg Med, 2012
- The authors reviewed 8 PCT, 6 CRP and 7 elevated WBC studies
- In summary, they found that PCT was better than CRP or leukocytosis at ruling out SBI in FUS
- Procalcitonin OR=10.6 sens 83% spec 69%
- CRP OR=9.83 sens 74% spec 76%
- Leukocytosis OR=4.26 sens 58% spec 73%
Procalcitonin test in the diagnosis of bacteremia: a meta-analysis
Jones et al
Annals of Emerg Med, 2007
- Meta-analysis of adults and kids to evaluate PCTs ability to evaluate for bacteremia in in the ED
- Data were reviewed across 17 studies with a total of 2,008 subjects
- The degree of heterogeneity was high across studies (I2=64%)
- When they analyzed a subgroup using only studies with the test threshold of >0.5 or 0.4 ng/mL the pooled estimate for yielded pooled estimates for sensitivity was 76% (95% CI 0.66 to 0.84) and specificity was 70% (95% CI 0.60 to 0.79) indicating a moderate ability to detect for bacteremia
Procalcitonin and C-reactive protein as markers of bacterial infection in critically ill children at onset of systemic inflammatory response syndrome
Simon et al
Pediatr Crit Care Med, 2008
- A prospective cohort of PICU patients comparing PCT and CRP in differentiating bacterial vs non-bacterial SIRS
- PCT levels were significantly higher in patients with bacterial infection vs those without (p = .01)
- The AOC for PCT was better than CRP – 0.71 vs 0.65 – indicating a small degree of superiority
- PCT >2.5 ng/mL added to bedside clinical judgment, increased the likelihood of bacterial infection from 39% to 92%
- CRP (<4 mg/dL) decreased the probability of bacterial infection from 39% to 2%
Use of serum procalcitonin in evaluation of febrile infants: a meta-analysis of 2317 patients
England et al
J Emerg Med, 2014
- Investigation of the utility of a single PCT measurement to identify SBI in infants <91 days of age
- Meta-analysis of 7 studies involving 2,317 patients
- 0.3 ng/mL was the PCT cutoff value, infants above that had a 42.7% prevalence of SBI, below 0.3 had a 12.5% prevalence (kind of high…)
- Overall relative risk (RR) of PCT >0.3 was 3.97 (95% CI 3.41-4.62)
- The RR from a systematic review of clinical prediction rules was 30.6 (95% CI 7.0-68.13) and 8.75 (95% CI 2.29-15.2) for infants untreated and treated with antibiotics, respectively.
But wait, there’s more!
Given that PCT has become a frequently ordered test for many of you I wanted to answer some straightforward questions, and dive into the body of evidence to help support your decision making in the Emergency Department.
What is Procalcitonin?
- Precursor of calcitonin which is involved in calcium homeostasis
- Produced by thyroid, lung and intestine w/ half-life of 25 to 30 hours.
- Below the limit of detection (10 pg/mL) in healthy patients
- Rises in response to proinflammatory stimuli, especially bacterial, mostly via lung and gut production
- Starts to rise 6 hours after the onset of infection
- “It is a helpful biomarker for early diagnosis of sepsis in the critically ill patient,” Wacker et. al. Lancet Infectious Dis, 2013
- Seems to be the new darling of critical care providers
In general, what do the results mean?
- <0.5 normal
- >0.5-2 moderate risk for progression to severe systemic infection
- >2 high risk for progression to severe systemic infection
- >10 high likelihood of severe sepsis/septic shock
Let’s take a look at some of the evidence
Procalcitonin to detect invasive bacterial infection in non-toxic-appearing infants with fever without apparent source in the emergency department
Luaces-Cubello et al
Pediatr Infect Dis J, 2012
- This was a study of 868 non-toxic, febrile infants <36 months of age. 1.7% had an invasive infection.
- The area under the receiver operating characteristic curve for procalcitonin was 0.87 (optimum cutoff 0.9 ng/mL, sensitivity 86.7%, specificity 90.5%)
- CRP = 0.79 (optimum cutoff 91 mg/L, sensitivity 33.3%, specificity 95.9%)
- In fever <8 hours duration the of <8 hours duration, the area under the receiver operating characteristic curve was 0.97 for procalcitonin and 0.76 for C-reactive protein
- Summary: The higher the area under the ROC, the better the ability to help predict disease. Especially with fever <8 hours procalcitonin was superior to CRP in this study.
Comparison of the Test Characteristics of Procalcitonin to C-Reactive Protein and Leukocytosis for the Detection of Serious Bacterial Infections in Children Presenting With Fever Without Source: A Systematic Review and Meta-analysis
Chia-Hung et al.
Annals of Emerg Med, 2012
- The authors reviewed 8 PCT, 6 CRP and 7 elevated WBC studies
- In summary, they found that PCT was better than CRP or leukocytosis at ruling out SBI in FUS
- Procalcitonin OR=10.6 sens 83% spec 69%
- CRP OR=9.83 sens 74% spec 76%
- Leukocytosis OR=4.26 sens 58% spec 73%
Procalcitonin test in the diagnosis of bacteremia: a meta-analysis
Jones et al
Annals of Emerg Med, 2007
- Meta-analysis of adults and kids to evaluate PCTs ability to evaluate for bacteremia in in the ED
- Data were reviewed across 17 studies with a total of 2,008 subjects
- The degree of heterogeneity was high across studies (I2=64%)
- When they analyzed a subgroup using only studies with the test threshold of >0.5 or 0.4 ng/mL the pooled estimate for yielded pooled estimates for sensitivity was 76% (95% CI 0.66 to 0.84) and specificity was 70% (95% CI 0.60 to 0.79) indicating a moderate ability to detect for bacteremia
Procalcitonin and C-reactive protein as markers of bacterial infection in critically ill children at onset of systemic inflammatory response syndrome
Simon et al
Pediatr Crit Care Med, 2008
- A prospective cohort of PICU patients comparing PCT and CRP in differentiating bacterial vs non-bacterial SIRS
- PCT levels were significantly higher in patients with bacterial infection vs those without (p = .01)
- The AOC for PCT was better than CRP – 0.71 vs 0.65 – indicating a small degree of superiority
- PCT >2.5 ng/mL added to bedside clinical judgment, increased the likelihood of bacterial infection from 39% to 92%
- CRP (<4 mg/dL) decreased the probability of bacterial infection from 39% to 2%
Use of serum procalcitonin in evaluation of febrile infants: a meta-analysis of 2317 patients
England et al
J Emerg Med, 2014
- Investigation of the utility of a single PCT measurement to identify SBI in infants <91 days of age
- Meta-analysis of 7 studies involving 2,317 patients
- 0.3 ng/mL was the PCT cutoff value, infants above that had a 42.7% prevalence of SBI, below 0.3 had a 12.5% prevalence (kind of high…)
- Overall relative risk (RR) of PCT >0.3 was 3.97 (95% CI 3.41-4.62)
- The RR from a systematic review of clinical prediction rules was 30.6 (95% CI 7.0-68.13) and 8.75 (95% CI 2.29-15.2) for infants untreated and treated with antibiotics, respectively.
Obviously this is just scratching the surface on PCT:
- There is no evidence to show that it supplants a thorough H&P and clinical judgement – Just that it supports it! So don’t go ordering a procalcitonin and hoping that it will make the sick vs not-sick decision for you.
- You can use it to help differentiate serious infection in children with FUS that are well appearing – but there’s still something that worries you about them
- At least in the PICU it is better than CRP at helping to identify bacterial vs non-bacterial SIRS
- Though England et al. did show a statistically significant RR (approximately 4), compare this with what we know from Huppler et al’s review of 21 studies on the Rochester criteria in which babies 29-90 days old identified to be low risk had a prevalence of SBI of only 2.7%, with a RR for SBI in high-risk patients of 30.6 (95% CI 7.0-68.13)
- The risk risk Rochester criteria were:
- Previously healthy term infant without perinatal complications and with no previous antibiotic treatment
- Normal physical examination findings (including no otitis media)
- White blood cell count: 5000–15 000 cells per mm3
- Band count: <1500 cells per mm3
- Urinalysis: <10 white blood cells per high-power field in centrifuged catheterized specimen
- Applying the low risk criteria per Huppler allowed 30% of patients to be treated safely without empiric antibiotics
- The risk risk Rochester criteria were:
- And don’t forget, that this really applies to infants older than 30 days in most cases. most experts support doing the whole shebang in babies under that threshold
[…] Check out my recent post on Procalcitonin here […]
[…] This episode of PEM Currents tackles a bread and butter issue in Pediatric Emergency Medicine, the newborn with fever. I discuss management, specifically how it differs for babies under 28 days of age as well as when to get labs and how to interpret them. I also reference procalcitonin, and touch on its emerging role. Read more here. […]