Procalcitonin in the ED: The evidence

Is procalcitonin of any value in the emergency department? Considering that procalcitonin testing is not available in any of my hospitals, I have never been all that interested in this question. But Swami requested a procalcitonin evidence review for EM:RAP, so I waded through the endless morass of PubMed in his honour.

Procalcitonin is a precursor for calcitonin, the production of which is upregulated in response to inflammatory states. Unlike other inflammatory markers, procalcitonin elevation is somewhat specific to certain bacterial antigens. Conversely, viral infections typically result in the production of interferon gamma, which decreases production of procalcitonin. Therefore, theoretically, procalcitonin might be helpful as an inflammatory marker that could distinguish between viral and bacterial illnesses.

When considering the use of procalcitonin, it is important to be clear about your desired outcomes. Some people try to use it to delineate between viral and bacterial infections. Some try to use it to determine the appropriate length of antibiotic therapy. Others use it because it sounds more modern than a white cell count. Its value may depend on how it is used.

Based on my initial search, there appear to be thousands of publications of procalcitonin. I will admit right up front, I have not read every single paper. Not even the majority. Therefore, there is a chance I am missing something, but there is nothing in the studies that I have read that convinces me that this is a topic worthy of multiple years of my life. 

Procalcitonin to exclude bacterial infection (aka initiating antibiotic therapy)

Accuracy (aka observational trials)

There are a number of trials that attempt to estimate the sensitivity and specificity of procalcitonin for bacterial infections. Considering that there is no perfect gold standard for bacterial infections, these trials are severely limited, and the numbers should be taken with a grain of salt. 

In one multicenter prospective study of 1735 hospitalized patients with potential community acquired pneumonia, there was no procalcitonin level that would safely exclude bacterial infection. Even the lowest procalcitonin level (0.1 ng/mL) only had a negative predictive value of 82%. (Self 2017)  (This study, like many pneumonia studies, is limited because a definitive pathogen was only identified in 37% of patients.) If you focus on the subset of patients with a known pathogen, patients with typical bacterial infections had higher average procalcitonin levels, but the overlap of values between groups demonstrates its lack of clinical utility. 23% of patients with bacterial pneumonia still had a low procalcitonin (<0.25). Furthermore, patients with atypical bacterial infections had procalcitonin levels almost identical to patients with viral infection.

The BACH trial was an observational trial of 1641 patients presenting to the ED with dyspnea. (Maisal 2012) 9% of patients were diagnosed with pneumonia, and although procalcitonin was associated with the diagnosis, the area under the curve was poor at 0.72, and essentially no better than just checking whether the patient had a fever (AUC .70). More importantly, procalcitonin performed worse than physician judgment (AUC 0.85). Even taken at face value, these numbers would not support the use of procalcitonin, and this is almost certainly an overestimate of procalcitonin’s value. The procalcitonin values were known to the clinicians and could have biased the final diagnosis. Furthermore, there was not a standardized definition of positive or negative for procalcitonin, so these numbers were retrospectively fit to the data after it was collected. 

Although I think we should probably stop using sensitivity and specificity in medicine, they are still widely reported. In the diagnosis of sepsis, procalcitonin appears to be only moderately accurate, with a pooled sensitivity of 77% and specificity of 79%. (Wacker 2013) Translated into useful numbers, this equates to a positive likelihood ratio of about 3.5 and a negative likelihood ratio of 0.29. Thus, although procalcitonin may budge your probability of sepsis a little bit, it is nowhere close to good enough to either rule in or rule out the diagnosis. Furthermore, these numbers are certainly over-estimates, because there is not a universally agreed upon definition of ‘positive’ or ‘negative’ procalcitonin, so each study was able to retrospectively fit their data to whatever threshold looked best.

For pneumonia, this was summarized in a 2020 meta-analysis which found that procalcitonin had a sensitivity of 55% and specificity of 76%. They conclude that “a procalcitonin level is unlikely to provide reliable evidence either to mandate administration of antibiotics or to enable withholding such treatment in patients with CAP.” (Kamat 2020)

There are many known causes of both false positive and false negative procalcitonin levels, which help explain the moderate to poor accuracy seen here. There are numerous nonbacterial causes of procalcitonin elevation, including candidiasis, dengue, malaria, renal disease, thyroid disease, burns, trauma, major surgery, lung cancer, and hematologic malignancies. (Maves 2020) Furthermore, procalcitonin is not routinely elevated in immunocompromised patients, nor with atypical bacterial infections. (Self 2017)

The one exception to the poor accuracy of procalcitonin may be meningitis. I have not found time to critically appraise all the individual studies, but a meta-analysis on the topic suggests pretty impressive numbers. (Wei 2016) The pooled sensitivity and specificity of serum procalcitonin in the diagnosis of bacterial meningitis was 95% (95% CI 89-99%) and 97% (95% CI 89-99%). This translates to a negative likelihood ratio of 0.06 and a positive likelihood ratio of 31.7. These numbers are far more impressive than the numbers seen in sepsis and respiratory infections. I am not sure why, but it might just be that there are a much smaller number of causative organisms for bacterial meningitis. Importantly, unlike sepsis and respiratory infections, the utility of procalcitonin in meningitis has not been tested in an RCT, and given the rarity of the condition in most settings, it still may not provide clinical value. However, considering that meningitis is our primary concern in febrile neonates, these numbers might explain why procalcitonin has found its way into various pediatric algorithms. On the other hand, there are a number of reasons to be cautious of these results. There were less than 100 patients on average in each of these studies, and selection bias will be an issue, as these were all patients sick enough to require a lumbar puncture, but presumably we want to use procalcitonin on a less sick cohort to potentially avoid performing the LP. 

Bottom line: The studies are imperfect, but it is pretty clear that procalcitonin is not accurate enough to rule out bacterial infections, and therefore does not seem to have a role in decisions about whether to start antibiotics in the emergency department. (Meningitis might be the one exception to this general rule.)

Usefulness (aka RCTs)

Luckily, we don’t have to rely on sensitivities and specificities. Unlike many of our diagnostic tests, procalcitonin has actually been studied in RCTs.

The HiTEMP trial is an emergency department based RCT in which 551 febrile adult patients were randomized to either have a single procalcitonin drawn as part of their workup, or to receive standard care. (van der Does 2018) Procalcitonin had no effect on antibiotic prescription (73% vs 77%). If there is any ED population where procalcitonin could have helped, it was clearly this population, as antibiotic overuse must be rampant if 77% of febrile patients are receiving an antibiotic prescription. There was no difference in their safety outcome, and procalcitonin only looks moderately accurate with an AUC of 0.681, which was essentially identical to CRP.

The ProACT study is a multicenter emergency department RCT that enrolled 1656 adult patients with an initial diagnosis of respiratory tract infection in whom the ED physician was unsure about the need for antibiotics. (Huang 2018) There was no difference in antibiotic use between the two groups. 77% of the patients were in the lowest tier of procalcitonin, where antibiotics were “strongly discouraged” and another 15% were in the low tier where antibiotics were “discouraged”. Despite this, 65% of the total group received antibiotics, and 35% received them in the ED. (The primary outcome is far from perfect, because it includes all antibiotics in the month after the ED visit. A procalcitonin in the ED really can’t be expected to change the management of a walk-in clinic next week.) Either way, procalcitonin had no impact on patient outcomes. 

The BPCTrea study looked at an antibiotic algorithm to guide both initiation and continuation of antibiotics in COPD patients admitted to the ICU. (Daubin 2018) Total antibiotic use was not really different between the two groups, but the procalcitonin group had a higher mortality (20% vs 14%, with a 95% confidence interval that just crossed 1). In the subgroup of patients who were on antibiotics at baseline, the increase in mortality in the procalcitonin group was statistically significant. This would be an important finding, but much like the ICU studies claiming a mortality benefit discussed below, I just don’t believe it. How exactly is procalcitonin changing mortality? (Especially with no decrease in antibiotic use, and the fact that the increased mortality was in patients without pneumonia). 

There was an outpatient study that concluded that procalcitonin decreased antibiotic prescribing. (Briel 2008) They recruited 455 patients visiting their primary care physician’s office for a respiratory tract infection, in whom the physician thought antibiotics were needed, and randomized them to a procalcitonin algorithm or standard care. One major caveat for emergency departments: if the physician decided not to prescribe antibiotics, a second procalcitonin measurement was required in 6-24 hours for safety reasons. (I don’t think the manuscript tells us how often this changed management.) The procalcitonin strategy significantly decreased antibiotic usage (97% vs 25%) without any change in patient symptoms. However, there are a couple major issues with this trial. First, the baseline antibiotic usage is just way too high. The vast majority of these patients were diagnosed with bronchitis, pharyngitis, or influenza. None of them needed antibiotics, so instead of using procalcitonin, you could have just told the doctors to practice good medicine. More importantly, despite the conclusion that antibiotic prescriptions decreased, every single patient was given an antibiotic prescription. The protocol was actually to give patients ‘delayed antibiotic prescriptions’ which are psychologically very different for doctors and physicians. They don’t tell us exactly how many filled the delayed prescription, but the mean days with antibiotics was not different between the groups (6.2 versus 7.1), so it seems like almost everyone in both groups took a week of antibiotics anyway.

Another multicenter non-inferiority RCT – the proHOSP study – found that a procalcitonin strategy is non-inferior to standard care in emergency department patients with lower respiratory tract infections. (Schuetz 2009) In general, if we are going to start using a new test, we want to know that it is better than standard care, but the non-inferiority approach is supported by the idea that procalcitonin can decrease antibiotic usage, which could be a benefit. (I have a long blog post coming about non-inferiority trials. They are massively overused, often as advertising by industry, when we would be better off just proving the claim of superiority. This might be a reasonable time to mention the lead author of this paper is a paid consultant of the company that makes procalcitonin assays, and his name is on many of the positive procalcitonin papers, including the Cochrane review). This study mixes strategies, as procalcitonin was used to both guide initiation of antibiotics, as well as their discontinuation. Although the overall use of antibiotics was decreased, procalcitonin had no effect on initial prescribing in the ED (93.7 and 91.4% of patients received antibiotics initially). It is also worth noting that, like all the studies we will discuss, this trial was unblinded, and so the Hawthorne effect is very likely to bias these results. Digging into the numbers, it is even less clear that procalcitonin is really helping. The rate of antibiotic prescriptions in patients with confirmed pneumonia dropped from 99.1% with standard care to 90.7% with procalcitonin. Does that sound like an improvement? Most of the change was in patients with bronchitis, but once again the baseline was absolutely atrocious, with 50% of patients receiving antibiotics. Such rates of antibiotic overuse don’t need procalcitonin, but a medical school refresh. Either way, for our purposes, there was no difference in the initiation of antibiotics in the emergency department.

Bottom line: As one would expect based on the poor sensitivity and specificity, RCTs of procalcitonin used to determine whether or not to initiate antibiotics therapy fail to show any benefit from the test.

Procalcitonin to guide length of antibiotic therapy

There are a number of studies that demonstrate shorter courses of antibiotic therapy when guided by procalcitonin. However, the devil may be in the details, as so much depends on what is happening in the control groups.

The PRORATA study randomized 621 ICU patients to either procalcitonin guided antibiotic therapy or usual care, and found that the procalcitonin group received antibiotics an average of 3 days less (14.3 versus 11.6), with no change in mortality or hospital length of stay. (Bouadma 2010) However, even if this difference in an unblinded trial is real, it is unlikely that any of these patients truly needed 2 weeks of antibiotic therapy. For almost all conditions studied, shorter courses of antibiotics have been found to be equivalent to longer courses. With the exception of abscesses or other difficult to clear locations, most bacterial infections were probably cleared well before either of these groups stopped antibiotics. You could imagine that a simple rule to stop all antibiotics after 7 days might outperform either of these strategies. This is especially true because this trial excluded all patients with infections in which longer antibiotics are recommended, such as osteomyelitis and endocarditis. Furthermore, a couple day decrease in antibiotics is a pretty marginal benefit if it doesn’t change any other patient important outcomes, and requires daily procalcitonin measurement. 

The SAPS trial is another open-label multicenter RCT in 1575 adult ICU patients that randomized patients to procalcitonin guided antibiotic therapy or standard care. (de Jong 2016) The procalcitonin arm received approximately 2 fewer days of antibiotics, with no signs or harm. In fact, mortality was statistically lower in the procalcitonin group, but that doesn’t make any sense to me, and I think suggests a problem with the study. How could 2 days fewer antibiotics translate into a 5% absolute decrease in mortality. It just doesn’t make sense.

The ProGUARD study is a multicenter, partially blinded RCT that included 394 adult ICU patients and randomized them to a procalcitonin guidance group or standard care. (Shehabi 2014) There was not a statistical difference in the time to antibiotic cessation, although it was 2 days better in the procalcitonin group (9 vs 11 days, p=0.58). There were no changes in other clinical outcomes. 

Another open label RCT compared procalcitonin to CRP in the management of sepsis and found no difference. The mean length of antibiotic therapy was 7 days with procalcitonin and 6 days with CRP. (Oliveira 2013)

The small decrease in antibiotic use seen in RCTs also seems to occur in the real world. There are numerous large observational studies that show that procalcitonin use is associated with a shorter duration of antibiotic therapy. (Balk 2017, Meir 2019) These studies don’t really add a lot, but tell us that physicians trust procalcitonin and almost certainly overuse antibiotics when left to their own devices. However, even in the procalcitonin groups of these studies, the duration of antibiotics seems unnecessarily long, at almost 2 weeks. 

There is a relatively strong indication that the unblinded nature of these trials matters a lot, and that we only get what we are looking for: there are trials in which elevated procalcitonin actually leads to greater use of antibiotics. The PASS study was an RCT of 1200 ICU patients comparing a procalcitonin driven strategy to standard of care, with the procalcitonin focused on escalating care. The procalcitonin group used significantly more broad spectrum antibiotics, received antibiotics for longer, and had significantly more days in which 3 different antibiotics were given. However, despite all these extra antibiotics, there was no difference in mortality and most clinical outcomes look a bit worse in the procalcitonin group. (Jensen 2011) In other words, procalcitonin doesn’t seem to change clinical outcomes in any of these studies, and the amount of antibiotics used just seems to go in the direction the researchers want it to.

There are a number of systematic reviews on this topic. One focused specifically on mortality in ICU patients. (Pepper 2019) They include 16 RCTs, and I find the results a little bizarre. There was almost no difference in antibiotic use between the groups (1.3 days), but they still conclude that there was a mortality benefit from using a procalcitonin algorithm (RR 0.89, 95% 0.83-0.97). I imagine this is a fluke. There was only 1 paper with a statistically significant improvement in mortality, and if this trial was removed from the analysis the results are no longer statistically significant. It doesn’t make a lot of sense that an antibiotic stewardship strategy that barely changes antibiotics use would have a large impact on mortality. (Although, I suppose if we are making a lot of misdiagnoses in the ICU, and the procalcitonin is the thing that reminds physicians that not all shock is sepsis, so they go looking for other causes, there could be a mortality benefit? However, I am not sure we need a blood test to remind us to do a thorough assessment of critically ill patients.)

Another systematic review concludes that there is no effect of procalcitonin algorithms on mortality overall, but if one just looks at the the algorithms used for antibiotic cessation, they find a statistical mortality benefit (RR 0.87 95% CI 0.77-0.98). (Lam 2018) However, like the above review, this conclusion seems to be based entirely on the SAPS trial (discussed above), and a simple inspection of the Forest plot gives one the impression that mortality is unaffected. (I don’t think most procalcitonin proponents are really convinced that it is a life saving lab test.)

Bottom line: None of these trials are blinded, which I think probably completely explains the marginal benefit described. Researchers tend to get whatever they want from clinicians who know their antibiotic prescriptions are being monitored. Even if the difference is real, it seems very modest (perhaps 2 fewer days of antibiotics), and has not been compared to other antibiotic stewardship programs that may be more effective and less costly.

Procalcitonin in pediatrics

One of the reasons for the recent interest in procalcitonin is its inclusion in new pediatric fever algorithms and guidelines. In particular, procalcitonin has played a central role in 2 major publications over the past 2 years, both of which were covered for the blog by Dr. Dennis Ren, as well as in the ‘step by step’ approach to febrile infants:

• The PECARN febrile infants study
• The 2021 AAP guidelines on the Evaluation and management of well-appearing febrile infants 8 to 60 days old

Before getting to these decision tools, it is worth noting that procalcitonin doesn’t not appear to be any more accurate in children than it is in adults. In one large multicenter observational study focused on infants from 7 to 91 days with fever, procalcitonin was found to have a moderate accuracy for “serious bacterial illness” that was not different from CRP (an AUC of 0.8). (Milicent 2016) However, seeing as “serious bacterial illness” includes UTIs which probably aren’t all that serious, this number may not be all that important. When focusing specifically on bacteremia and meningitis, procalcitonin was better than CRP, and had a reasonable AUC at 0.91. The sensitivity and specificity vary by cut-off, and are almost certainly inflated in this data set because a single cut-off wasn’t pre-specified. At the lowest cut-off of 0.3 ng/Ml, procalcitonin had a sensitivity of 90% (95% CI 68-99%), specificity of 78% (95%CI 75-80%), positive likelihood ratio of 4 and negative likelihood ratio of 0.1. These numbers are OK, but probably not good enough on their own, especially when considering the very wide confidence intervals on the sensitivity. 

So despite having questionable accuracy for ‘serious bacterial illness’, procalcitonin has found its way into two major pediatric decision tools. Let’s start with the PECARN tool. In their derivation and internal validation study, they included 1821 febrile infants less than 60 days old, and after playing with some numbers, come up with a decision rule with a sensitivity of 98.8% (95% CI 92.5-99.9%) in the derivation and 97.7% (95% CI 91.3-99.6%) in the validation. (Kuppermann 2019) The specificity is about 60%, negative likelihood ratio of 0.03, and positive likelihood ratio of 2.5. There are a number of problems with this study, including the fact that UTI might not be so serious, but in the under 60 day group I am certainly OK overtreating some UTIs. However, it is just not clear that procalcitonin is adding any value to this rule. They don’t provide sensitivity and specificity information for procalcitonin alone, but the average procalcitonin in the ‘serious bacterial illness group’ was 0.7, which is well below the cutoff they use in the decision tool of 1.71. Speaking of which, they use a retrospectively defined procalcitonin cutoff of 1.71 ng/mL, which is entirely inconsistent with the rest of this literature, where cutoffs between 0.1-0.3 ng/mL are generally used to define low risk. This number simply doesn’t make any sense, and is an indication that their rule is over-fit to their population, and will almost certainly fail when externally validated. It is also worth noting that even before procalcitonin was added to the algorithm, the risk of ‘serious bacterial illness’ was already less than 1%. If we plug the numbers from this final procalcitonin step of the algorithm into a simple 2 by 2 table, it turns out that procalcitonin only has a sensitivity of 75%, specificity of 98.5%, negative likelihood ratio of 0.25, and positive likelihood ratio of 50. To catch 1 extra ‘serious bacterial illness’, you have to test more than 175 infants. There may be a small marginal gain here, but it is incredibly small, still misses some infants, and is almost certainly a massive over-estimate considering the bizarre cut-off chosen.

The other pediatric decision tool that incorporates procalcitonin is ‘step by step’. You can review the tool itself here on MDCalc. The initial retrospective look at ‘step by step’ included 1123 febrile infants less than 3 months of age. It would have classified 488 as low risk and only missed 1 (0.2%) invasive bacterial illness, although it would have missed 46 (9.4%) possible serious bacterial illnesses. (Mintegi 2014) The procalcitonin picked up 3 invasive bacterial illnesses and missed 3, 2 of which were caught by the next step of CRP and neutrophil count. There is no way to know from this data whether procalcitonin added any value above the other blood tests. ‘Step by step’ was validated in a cohort of 2185 febrile infants. (Gomez 2016) 991 were classified as low risk and 7 (0.7%) had invasive bacterial illnesses. Procalcitonin caught 6 invasive bacterial illnesses and missed 12 (5 of which were caught at the CRP/ANC step). Again, it is impossible to know from this data whether procalcitonin is adding anything here, but it looks pretty bad, as it missed more sick kids than it catches.

From a procalcitonin standpoint, these decision tool studies are pretty useless. They throw the test in without really demonstrating that it is a helpful test. Its like cough and cold medicines that have dozens of ingredients. When they are all mixed together like that, it is impossible to know which (if any) is actually providing benefit.

Bottom line: Despite being a major component of pediatric guidelines, I am not convinced that procalcitonin adds significantly to the care of febrile infants

A few guidelines

The IDSA recommends against using procalcitonin when deciding on antibiotics in pneumonia. “We recommend that empiric antibiotic therapy should be initiated in adults with clinically suspected and radiographically confirmed CAP regardless of initial serum procalcitonin level.” (Metlay 2019) They also recommend against using procalcitonin in determine length of therapy for pneumonia, because the studies that demonstrated a decrease in length of therapy were treating for much longer than is considered standard in North America, and procalcitonin may miss important pathogens such as Legionella and Mycoplasma species. 

Conservely, the IDSA guidelines on antibiotic stewardship provide a weak recommendation that serial procalcitonin measurements should be used in ICU patients with suspected infections. (Barlam 2016)

The Choosing Wisely group says “don’t perform procalcitonin testing without an established, evidence-based protocol.” (Choosing Wisely 2021) Specifically, they note that procalcitonin is really only indicated when measured serially over time in select patients to make antibiotic treatments, which by definition means this is not an emergency department test in their eyes.

Although I am weary to mention the surviving sepsis campaign guidelines, considering their many problems, they suggest against using procalcitonin to decide when to start antibiotics is sepsis, but they do suggest using procalcitonin combined with clinical judgment to decide when to discontinue antibiotics in sepsis and septic shock. (Evans 2021)

When it comes to pediatrics, the new AAP guidelines do include and strongly recommend procalcitonin, but their recommendation is based on the data I discussed above, and I think unsupported by the evidence. They do say that if procalcitonin is unavailable, then CRP and ANC are ok as an alternative.

Discussion

There are many problems with this data, but it is actually a lot stronger than the data we have for most of the tests we use. We have RCTs, which is somewhat of a rarity, and the RCTs pretty clearly show that there is no role for procalcitonin in the emergency department or outpatient settings. There may still be a role in the inpatient or ICU setting for limiting antibiotic use, but I think that data is pretty weak.

In terms of ultimate goals, I agree entirely with supporters of procalcitonin. We massively overuse and abuse antibiotics in modern medicine, and we need to stop that practice. However, I am not convinced that procalcitonin is the best mechanism to achieve that goal. Even in the ICU setting, where there are some RCTs to support its use, I think the comparison group was incorrect. We need to see procalcitonin compared to other antibiotic stewardship programs. As compared to poor prescribing, procalcitonin may provide a marginal benefit, but when compared to appropriate antibiotic use, it is possible that procalcitonin actually makes things worse. (We introduced CT head rules because we wanted to decrease CT usage, but it in real life practice, the Canadian CT head rule actually increased CT usage.) (Stiell 2010)

In fact, it is worth noting that in at least one study, procalcitonin did increase antibiotic usage. (Jensen 2011) With an imperfect specificity, if broadly implemented, it is quite possible that procalcitonin algorithms increase antibiotic usage in some hospitals rather than decreasing it. 

If doctors were perfect rational decision makers, procalcitonin would probably be completely unnecessary. Unfortunately, we are imperfect, and it is a mistake to fail to account for those imperfections. There is great data to suggest that making retirement savings opt-out rather that opt-in results in more people saving for retirement. Personally, I have managed to save just fine in an opt-in system, but an individual’s success doesn’t negate the value of the opt-out systems. Held to a high standard, I think any individual doctor could outperform procalcitonin. However, in aggregate, it is pretty clear that we currently fail to meet this standard. 

I am all for psychological nudges. We need as much help as we can get in medicine. However, it is unclear whether procalcitonin represents the best possible nudge. If the goal is simply to reduce total antibiotic use, an automated message from the EMR stating “your patient has been on antibiotics for 7 days, so it is time to stop” is likely to be at least as effective, if not more effective than procalcitonin. If a patient is not better after 7 days of a meropenem, it is pretty unlikely that 14 days is the answer, no matter what the procalcitonin says. If procalcitonin was compared to other antibiotic stewardship options, like a mandated antibiotic review at 7 days, my guess is that procalcitonin would be worse.

That being said, even the decrease in antibiotics usage seen in these RCTs is questionable. These studies could not be blinded, and the Hawthrone effect means that clinicians who knew that their antibiotic prescribing was being assessed were probably more likely to stick to the provided algorithms than they would be in real life. 

Procalcitonin just isn’t that great. The diagnostic accuracy (sensitivity/specificity) of the test is only moderate at best, and the numbers in these studies can’t be taken at face value. Because the threshold for calling procalcitonin positive or negative was often not prespecified, the authors could choose the threshold that provided the best numbers. Furthermore, most bacterial infections (such as pneumonia) don’t have a perfect gold standard, so there is even more room for bias when assessing the accuracy of procalcitonin. 

Even if procalcitonin was very accurate, it shouldn’t be measured in a vacuum. We don’t need to know whether tests work, but rather whether they add anything to what we already have. Not many studies compared procalcitonin directly to physician judgment, but when the comparison is made, judgment appears to outperform procalcitonin. (Maisel 2012) 

Either way, the only data supporting procalcitonin appears to be in the inpatient setting. From an emergency department standpoint, there just doesn’t seem to be any role for procalcitonin at this point in time. 

Bottom line

I can’t order a procalcitonin in any of the emergency departments I work in, and this data suggests that is probably a good thing.

UPDATES

The pediatric rules were not my main focus when starting this literature search, but they are certainly the topic that has garnered the most debate on social media. I may tackle them independently in the future. I will note that it is possible to use a well structured tool to transform a mediocre test into something valuable. For example, DDimer is horrible as a stand alone test, but when combined with a structured risk assessment, it is very valuable. Procalcitonin could play that role in the step by step or PECARN tools. To that point, Dr. Nathan Kuppermann (lead author of PECARN) engaged on this topic on twitter:

And Justin, the reason that original cut off’s are different than others is that our rule used statistics to generate the best cutoffs – not pre-existing thresholds! @PECARNteam @UCDavisEM https://t.co/Go7JoJbOTh

— Nathan Kuppermann (@nkuppermann) February 16, 2022

Other FOAMed

EM News: Special Report: Procalcitonin: The Über WBC Count?

PulmCast: Deep Dive: Procalcitonin

Washington University Dept of Emergency Medicine journal club: Procalcitonin and Antiobiotics for Lower Respiratory Tract Infections

Washington University Dept of Emergency Medicine journal club: Biomarkers (e.g. Procalcitonin) to Diagnose Sepsis in the ED

Procalcitonin Testing in Suspected Infection: Review

Not free, but for EM:RAP subscribers: Is Procalcitonin Useful in the ED?

References

Balk RA, Kadri SS, Cao Z, Robinson SB, Lipkin C, Bozzette SA. Effect of Procalcitonin Testing on Health-care Utilization and Costs in Critically Ill Patients in the United States. Chest. 2017 Jan;151(1):23-33. doi: 10.1016/j.chest.2016.06.046. Epub 2016 Aug 25. PMID: 27568580

Barlam TF, Cosgrove SE, Abbo LM, MacDougall C, Schuetz AN, Septimus EJ, Srinivasan A, Dellit TH, Falck-Ytter YT, Fishman NO, Hamilton CW, Jenkins TC, Lipsett PA, Malani PN, May LS, Moran GJ, Neuhauser MM, Newland JG, Ohl CA, Samore MH, Seo SK, Trivedi KK. Implementing an Antibiotic Stewardship Program: Guidelines by the Infectious Diseases Society of America and the Society for Healthcare Epidemiology of America. Clin Infect Dis. 2016 May 15;62(10):e51-77. doi: 10.1093/cid/ciw118. Epub 2016 Apr 13. PMID: 27080992

Bouadma L, Luyt CE, Tubach F, Cracco C, Alvarez A, Schwebel C, Schortgen F, Lasocki S, Veber B, Dehoux M, Bernard M, Pasquet B, Régnier B, Brun-Buisson C, Chastre J, Wolff M; PRORATA trial group. Use of procalcitonin to reduce patients’ exposure to antibiotics in intensive care units (PRORATA trial): a multicentre randomised controlled trial. Lancet. 2010 Feb 6;375(9713):463-74. doi: 10.1016/S0140-6736(09)61879-1. Epub 2010 Jan 25. PMID: 20097417

Bloos F, Marshall JC, Dellinger RP, Vincent JL, Gutierrez G, Rivers E, Balk RA, Laterre PF, Angus DC, Reinhart K, Brunkhorst FM. Multinational, observational study of procalcitonin in ICU patients with pneumonia requiring mechanical ventilation: a multicenter observational study. Crit Care. 2011;15(2):R88. doi: 10.1186/cc10087. Epub 2011 Mar 7. PMID: 21385367

Briel M, Schuetz P, Mueller B, Young J, Schild U, Nusbaumer C, Périat P, Bucher HC, Christ-Crain M. Procalcitonin-guided antibiotic use vs a standard approach for acute respiratory tract infections in primary care. Arch Intern Med. 2008 Oct 13;168(18):2000-7; discussion 2007-8. doi: 10.1001/archinte.168.18.2000. PMID: 18852401

Choosing Wisely. American Society for Clinical Pathology. Thirty Five Things Physicians and Patients Should Question. 2021. Available at: https://www.choosingwisely.org/societies/american-society-for-clinical-pathology/

de Jong E, van Oers JA, Beishuizen A, et al. Efficacy and safety of procalcitonin guidance in reducing the duration of antibiotic treatment in critically ill patients: a randomised, controlled, open-label trial. Lancet Infect Dis. 2016 Jul;16(7):819-827. doi: 10.1016/S1473-3099(16)00053-0. Epub 2016 Mar 2. PMID: 26947523

Evans L, Rhodes A, Alhazzani W, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock 2021. Intensive Care Med. 2021 Nov;47(11):1181-1247. doi: 10.1007/s00134-021-06506-y. Epub 2021 Oct 2. PMID: 34599691

Gomez B, Mintegi S, Bressan S, Da Dalt L, Gervaix A, Lacroix L; European Group for Validation of the Step-by-Step Approach. Validation of the “Step-by-Step” Approach in the Management of Young Febrile Infants. Pediatrics. 2016 Aug;138(2):e20154381. doi: 10.1542/peds.2015-4381. Epub 2016 Jul 5. PMID: 27382134

Hankey B, Riley B. Towards evidence based emergency medicine: best BETs from the Manchester Royal Infirmary. BET 1: use of a procalcitonin algorithm to guide antimicrobial therapy in COPD exacerbations can reduce antibiotic consumption with no increase in rates of treatment failure or mortality. Emerg Med J. 2015 Jun;32(6):493-5. doi: 10.1136/emermed-2015-204865.1. PMID: 25991774

Herd D. Towards evidence-based emergency medicine: best BETs from the Manchester Royal Infirmary. BET 3: In well-appearing children suspected of meningococcal disease can procalcitonin reduce the need for empiric antibiotic treatment? Emerg Med J. 2013 Jun;30(6):515. doi: 10.1136/emermed-2013-202697.3. PMID: 23673786

Huang DT, Yealy DM, Filbin MR, Brown AM, Chang CH, Doi Y, Donnino MW, Fine J, Fine MJ, Fischer MA, Holst JM, Hou PC, Kellum JA, Khan F, Kurz MC, Lotfipour S, LoVecchio F, Peck-Palmer OM, Pike F, Prunty H, Sherwin RL, Southerland L, Terndrup T, Weissfeld LA, Yabes J, Angus DC; ProACT Investigators. Procalcitonin-Guided Use of Antibiotics for Lower Respiratory Tract Infection. N Engl J Med. 2018 Jul 19;379(3):236-249. doi: 10.1056/NEJMoa1802670. Epub 2018 May 20. PMID: 29781385

Jensen JU, Hein L, Lundgren B, Bestle MH, Mohr TT, Andersen MH, Thornberg KJ, Løken J, Steensen M, Fox Z, Tousi H, Søe-Jensen P, Lauritsen AØ, Strange D, Petersen PL, Reiter N, Hestad S, Thormar K, Fjeldborg P, Larsen KM, Drenck NE, Ostergaard C, Kjær J, Grarup J, Lundgren JD; Procalcitonin And Survival Study (PASS) Group. Procalcitonin-guided interventions against infections to increase early appropriate antibiotics and improve survival in the intensive care unit: a randomized trial. Crit Care Med. 2011 Sep;39(9):2048-58. doi: 10.1097/CCM.0b013e31821e8791. PMID: 21572328

Kamat IS, Ramachandran V, Eswaran H, Guffey D, Musher DM. Procalcitonin to Distinguish Viral From Bacterial Pneumonia: A Systematic Review and Meta-analysis. Clin Infect Dis. 2020 Jan 16;70(3):538-542. doi: 10.1093/cid/ciz545. PMID: 31241140

Kuppermann N, Dayan PS, Levine DA, Vitale M, Tzimenatos L, Tunik MG, Saunders M, Ruddy RM, Roosevelt G, Rogers AJ, Powell EC, Nigrovic LE, Muenzer J, Linakis JG, Grisanti K, Jaffe DM, Hoyle JD Jr, Greenberg R, Gattu R, Cruz AT, Crain EF, Cohen DM, Brayer A, Borgialli D, Bonsu B, Browne L, Blumberg S, Bennett JE, Atabaki SM, Anders J, Alpern ER, Miller B, Casper TC, Dean JM, Ramilo O, Mahajan P; Febrile Infant Working Group of the Pediatric Emergency Care Applied Research Network (PECARN). A Clinical Prediction Rule to Identify Febrile Infants 60 Days and Younger at Low Risk for Serious Bacterial Infections. JAMA Pediatr. 2019 Apr 1;173(4):342-351. doi: 10.1001/jamapediatrics.2018.5501. PMID: 30776077

Lam SW, Bauer SR, Fowler R, Duggal A. Systematic Review and Meta-Analysis of Procalcitonin-Guidance Versus Usual Care for Antimicrobial Management in Critically Ill Patients: Focus on Subgroups Based on Antibiotic Initiation, Cessation, or Mixed Strategies. Crit Care Med. 2018 May;46(5):684-690. doi: 10.1097/CCM.0000000000002953. PMID: 29293146

Maisel A, Neath SX, Landsberg J, Mueller C, Nowak RM, Peacock WF, Ponikowski P, Möckel M, Hogan C, Wu AH, Richards M, Clopton P, Filippatos GS, Di Somma S, Anand I, Ng LL, Daniels LB, Christenson RH, Potocki M, McCord J, Terracciano G, Hartmann O, Bergmann A, Morgenthaler NG, Anker SD. Use of procalcitonin for the diagnosis of pneumonia in patients presenting with a chief complaint of dyspnoea: results from the BACH (Biomarkers in Acute Heart Failure) trial. Eur J Heart Fail. 2012 Mar;14(3):278-86. doi: 10.1093/eurjhf/hfr177. Epub 2012 Feb 2. PMID: 22302662

Maves RC. Procalcitonin Is Not an Adequate Tool for Antimicrobial De-Escalation in Sepsis. Crit Care Med. 2020 Dec;48(12):1848-1850. doi: 10.1097/CCM.0000000000004547. PMID: 32885936

Meier MA, Branche A, Neeser OL, Wirz Y, Haubitz S, Bouadma L, Wolff M, Luyt CE, Chastre J, Tubach F, Christ-Crain M, Corti C, Jensen JS, Deliberato RO, Kristoffersen KB, Damas P, Nobre V, Oliveira CF, Shehabi Y, Stolz D, Tamm M, Mueller B, Schuetz P. Procalcitonin-guided Antibiotic Treatment in Patients With Positive Blood Cultures: A Patient-level Meta-analysis of Randomized Trials. Clin Infect Dis. 2019 Jul 18;69(3):388-396. doi: 10.1093/cid/ciy917. PMID: 30358811

Metlay JP, Waterer GW, Long AC, Anzueto A, Brozek J, Crothers K, Cooley LA, Dean NC, Fine MJ, Flanders SA, Griffin MR, Metersky ML, Musher DM, Restrepo MI, Whitney CG. Diagnosis and Treatment of Adults with Community-acquired Pneumonia. An Official Clinical Practice Guideline of the American Thoracic Society and Infectious Diseases Society of America. Am J Respir Crit Care Med. 2019 Oct 1;200(7):e45-e67. doi: 10.1164/rccm.201908-1581ST. PMID: 31573350

Milcent K, Faesch S, Gras-Le Guen C, Dubos F, Poulalhon C, Badier I, Marc E, Laguille C, de Pontual L, Mosca A, Nissack G, Biscardi S, Le Hors H, Louillet F, Dumitrescu AM, Babe P, Vauloup-Fellous C, Bouyer J, Gajdos V. Use of Procalcitonin Assays to Predict Serious Bacterial Infection in Young Febrile Infants. JAMA Pediatr. 2016 Jan;170(1):62-9. doi: 10.1001/jamapediatrics.2015.3210. Erratum in: JAMA Pediatr. 2016 Jun 1;170(6):624. PMID: 26595253

Mintegi S, Bressan S, Gomez B, Da Dalt L, Blázquez D, Olaciregui I, de la Torre M, Palacios M, Berlese P, Benito J. Accuracy of a sequential approach to identify young febrile infants at low risk for invasive bacterial infection. Emerg Med J. 2014 Oct;31(e1):e19-24. doi: 10.1136/emermed-2013-202449. Epub 2013 Jul 14. PMID: 23851127

Oliveira CF, Botoni FA, Oliveira CR, Silva CB, Pereira HA, Serufo JC, Nobre V. Procalcitonin versus C-reactive protein for guiding antibiotic therapy in sepsis: a randomized trial. Crit Care Med. 2013 Oct;41(10):2336-43. doi: 10.1097/CCM.0b013e31828e969f. PMID: 23921272

Pepper DJ, Sun J, Rhee C, Welsh J, Powers JH 3rd, Danner RL, Kadri SS. Procalcitonin-Guided Antibiotic Discontinuation and Mortality in Critically Ill Adults: A Systematic Review and Meta-analysis. Chest. 2019 Jun;155(6):1109-1118. doi: 10.1016/j.chest.2018.12.029. Epub 2019 Feb 14. PMID: 30772386

Plata-Menchaca EP, Ferrer R. Procalcitonin Is Useful for Antibiotic Deescalation in Sepsis. Crit Care Med. 2021 Apr 1;49(4):693-696. doi: 10.1097/CCM.0000000000004776. PMID: 33315698

Schuetz P, Christ-Crain M, Thomann R, Falconnier C, Wolbers M, Widmer I, Neidert S, Fricker T, Blum C, Schild U, Regez K, Schoenenberger R, Henzen C, Bregenzer T, Hoess C, Krause M, Bucher HC, Zimmerli W, Mueller B; ProHOSP Study Group. Effect of procalcitonin-based guidelines vs standard guidelines on antibiotic use in lower respiratory tract infections: the ProHOSP randomized controlled trial. JAMA. 2009 Sep 9;302(10):1059-66. doi: 10.1001/jama.2009.1297. PMID: 19738090

Schuetz P, Wirz Y, Sager R, Christ-Crain M, Stolz D, Tamm M, Bouadma L, Luyt CE, Wolff M, Chastre J, Tubach F, Kristoffersen KB, Burkhardt O, Welte T, Schroeder S, Nobre V, Wei L, Bucher HC, Bhatnagar N, Annane D, Reinhart K, Branche A, Damas P, Nijsten M, de Lange DW, Deliberato RO, Lima SS, Maravić-Stojković V, Verduri A, Cao B, Shehabi Y, Beishuizen A, Jensen JS, Corti C, Van Oers JA, Falsey AR, de Jong E, Oliveira CF, Beghe B, Briel M, Mueller B. Procalcitonin to initiate or discontinue antibiotics in acute respiratory tract infections. Cochrane Database Syst Rev. 2017 Oct 12;10(10):CD007498. doi: 10.1002/14651858.CD007498.pub3. PMID: 29025194

Self WH, Balk RA, Grijalva CG, Williams DJ, Zhu Y, Anderson EJ, Waterer GW, Courtney DM, Bramley AM, Trabue C, Fakhran S, Blaschke AJ, Jain S, Edwards KM, Wunderink RG. Procalcitonin as a Marker of Etiology in Adults Hospitalized With Community-Acquired Pneumonia. Clin Infect Dis. 2017 Jul 15;65(2):183-190. doi: 10.1093/cid/cix317. PMID: 28407054

Shehabi Y, Sterba M, Garrett PM, Rachakonda KS, Stephens D, Harrigan P, Walker A, Bailey MJ, Johnson B, Millis D, Ding G, Peake S, Wong H, Thomas J, Smith K, Forbes L, Hardie M, Micallef S, Fraser JF; ProGUARD Study Investigators; ANZICS Clinical Trials Group. Procalcitonin algorithm in critically ill adults with undifferentiated infection or suspected sepsis. A randomized controlled trial. Am J Respir Crit Care Med. 2014 Nov 15;190(10):1102-10. doi: 10.1164/rccm.201408-1483OC. PMID: 25295709

Stiell IG, Clement CM, Grimshaw JM, Brison RJ, Rowe BH, Lee JS, Shah A, Brehaut J, Holroyd BR, Schull MJ, McKnight RD, Eisenhauer MA, Dreyer J, Letovsky E, Rutledge T, Macphail I, Ross S, Perry JJ, Ip U, Lesiuk H, Bennett C, Wells GA. A prospective cluster-randomized trial to implement the Canadian CT Head Rule in emergency departments. CMAJ. 2010 Oct 5;182(14):1527-32. doi: 10.1503/cmaj.091974. Epub 2010 Aug 23. PMID: 20732978

Wacker C, Prkno A, Brunkhorst FM, Schlattmann P. Procalcitonin as a diagnostic marker for sepsis: a systematic review and meta-analysis. Lancet Infect Dis. 2013 May;13(5):426-35. doi: 10.1016/S1473-3099(12)70323-7. Epub 2013 Feb 1. PMID: 23375419

Wei TT, Hu ZD, Qin BD, Ma N, Tang QQ, Wang LL, Zhou L, Zhong RQ. Diagnostic Accuracy of Procalcitonin in Bacterial Meningitis Versus Nonbacterial Meningitis: A Systematic Review and Meta-Analysis. Medicine (Baltimore). 2016 Mar;95(11):e3079. doi: 10.1097/MD.0000000000003079. PMID: 26986140

Wirz Y, Meier MA, Bouadma L, Luyt CE, Wolff M, Chastre J, Tubach F, Schroeder S, Nobre V, Annane D, Reinhart K, Damas P, Nijsten M, Shajiei A, deLange DW, Deliberato RO, Oliveira CF, Shehabi Y, van Oers JAH, Beishuizen A, Girbes ARJ, de Jong E, Mueller B, Schuetz P. Effect of procalcitonin-guided antibiotic treatment on clinical outcomes in intensive care unit patients with infection and sepsis patients: a patient-level meta-analysis of randomized trials. Crit Care. 2018 Aug 15;22(1):191. doi: 10.1186/s13054-018-2125-7. PMID: 30111341

van der Does Y, Limper M, Jie KE, Schuit SCE, Jansen H, Pernot N, van Rosmalen J, Poley MJ, Ramakers C, Patka P, van Gorp ECM, Rood PPM. Procalcitonin-guided antibiotic therapy in patients with fever in a general emergency department population: a multicentre non-inferiority randomized clinical trial (HiTEMP study). Clin Microbiol Infect. 2018 Dec;24(12):1282-1289. doi: 10.1016/j.cmi.2018.05.011. Epub 2018 Jun 2. PMID: 29870855

And a few studies with too many flaws to be included in the main write up, but you can review them if you would like:Verduri A, Luppi F, D’Amico R, Balduzzi S, Vicini R, Liverani A, Ruggieri V, Plebani M, Barbaro MP, Spanevello A, Canonica GW, Papi A, Fabbri LM, Beghè B; FARM58J2XH Study Group. Antibiotic treatment of severe exacerbations of chronic obstructive pulmonary disease with procalcitonin: a randomized noninferiority trial. PLoS One. 2015 Mar 11;10(3):e0118241. doi: 10.1371/journal.pone.0118241. PMID: 25760346