Procalcitonin: physiology

Procalcitonin (PCT) is the precursor of the hormone calcitonin, initially described more than 20 years ago by Moya and colleagues (1975) [1] in chicken ultimobranchial glands incubated in vitro. During the last decade, there has been a growing body of evidence underscoring its value as a novel marker of infection in the intensive care unit (ICU) [2].

PCT is a protein consisting of 116 amino acids with a molecular weight of 13 kDa, and constitutes a pro-peptide encoded by the Calc-1 gene on the short arm of human chromosome 11 inside neuroendocrine C-cells of thyroid gland under normal conditions [3]. The original product of Calc-1 gene is the 141 amino acid chain of pre-procalcitonin (pre-PCT), which binds to the endoplasmic reticulum of the C-cells of thyroid gland with its N terminus, where it is cleaved by an endopeptidase to give rise to PCT. Subsequently, procalcitonin itself is cleaved by a convertase enzyme in calcitonin, katacalcin, and a proteic residue [3].

Figure 1 depicts the PCT sites of production in physiologic and pathologic conditions. In healthy subjects, all the PCTs formed in C-cells are converted into calcitonin, so that serum circulating procalcitonin levels (sPCT) are below the detection level, which is less than 0.5 ng/mL [4]. Furthermore, no enzymes in the plasma can break down the PCT circulating molecule, so that if somehow PCT escapes intracellular proteolysis and is secreted into the blood stream, it remains unchanged with a half-life of about 30 min in physiological conditions, with respect to 4–5 min for calcitonin.

Fig. 1
figure 1

Procalcitonin physiopathology and sites of production

Full size image

Pre-procalcitonin mRNA is over expressed in human medullar thyroid carcinoma, in small cell lung tumor, and sometimes in other rare neuroendocrine tumors such as phaeochromocytoma [3]. Calcitonin-related peptides have also been found in brain and pituitary tissue [3]. Little is known about the physiological increase of PCT in humans, but it seems that circulating levels of calcitonin precursors show increase in newborns, with a spontaneous reversal in the first week [5].

In case of severe bacterial infections or sepsis, procalcitonin is produced not only by extra-thyroidal parenchymal tissues, mainly hepatocytes [6], but also by granulocytes and other immune system components [3–5].

PCT production in the C-cells of the thyroid gland in physiologic conditions and during inflammation states are deeply different. In the first case, C-cells react to hypercalcemia as well as to glucocorticoids, glucagon, gastrin or β-adrenergic stimulation, while somatostatin and vitamin D suppress calcitonin production; none of these stimuli contributes to a rise in PCT levels.

In the presence of inflammatory states, it has been proven that PCT production is extra-thyroidal, and that it is linked to bacterial endotoxin, the most potent stimulator of PCT release into the circulation [6], and inflammatory cytokines: tumor necrosis factor (TNF), and interleukin 1 (IL-1), IL-2 and IL-6 [7]. Cells of neuroendocrine origin express calcitonin-related peptides (Calc-1 gene), and many speculate that PCT is coded by the same gene as PCT in the C-cells of thyroid, but it is not cleaved by proteolytic enzymes intracellularly. Confirmed sites of production of “inflammatory” PCT are neuroendocrine lung and intestine cells, while it is still debated whether the liver and splanchnic area can produce PCT in infective states, since it has been demonstrated only in patients after cardiac surgery with extracorporeal circulation, that is a very peculiar condition (intestine mucosal barrier damage). Furthermore, it has been proved that sPCT rises in non-infective clinical conditions, such as acute respiratory distress syndrome [3–5] and tissue damage of various origins (traumas, surgical interventions, burns) [8]. More recently, elevated sPCT levels in the normal range have been found associated with measures of obesity, insulin resistance, and metabolic syndrome in the general population [9].

After administration of endotoxin, TNF rises first with a peak at 90 min, followed by IL-6 with its peak at 180 min. PCT reacts after only 3–6 h, peaking at 6–8 h and its levels culminate after 12–48 h after endotoxin administration. After PCT, C-reactive protein (PCR) levels also rise; this cytokine pattern of production is typical of acute bacterial infection. As for as IL-1, IL-2, and IL-6 concerned, these cytokines may block PCT proteolysis to calcitonin, but it has not been yet proven. In contrast to TNF and IL-6, whose rise are not specific for certain types of inflammation, PCT selectively rises in bacterial inflammatory processes.

Procalcitonin in sepsis

PCT is known as a sensitive and specific biomarker for bacterial sepsis, and the increase in its plasma levels correlates with the severity of infection [2–5]. Moreover, its serum levels can be helpful in predicting the presence of serious bacterial infection in children with fever but without localizing signs [10]. It has been recently documented [11] that, in the Emergency Department (ED), PCT may be a valuable addition to currently used markers for diagnosis of infection (including C-reactive protein––CRP), and for prognosis in patients with fever. PCT levels are significantly associated with admission to a special care unit, duration of intravenous antibiotic use, total duration of antibiotic treatment and length of hospital stay, whereas CRP is related only to the latter two variables. Recent data support the role of PCT measurement in discriminating between severe lower respiratory tract infections of bacterial and 2009 H1N1 origin [12], and its role to identify bacterial co-infection in patients with H1N1 influenza [13].

The latest International Guidelines for management of severe sepsis and septic shock published in 2008 include in the diagnostic criteria for sepsis a value of serum PCT of more than 2 standard deviations above the upper limits [14]. In our opinion, serial PCT measurements are useful in monitoring the host’s response to antibiotic therapy.

Assicot et al. [2] initially described in children high PCT concentration in the presence of a bacterial infection, with a rapid decrease after antibiotic therapy, with normal values measured of calcitonin. Subsequently, investigations were made to discover the sites and mechanism of PCT production in infective states [6, 7]. Dandona et al. [6] also show an increase in PCT concentrations in normal volunteers after Escherichia coli endotoxin injection, creating the same clinical scenario as in septic shock.

In fact, in septic shock, the presence of circulating bacteria or bacterial products, such as endotoxin, leads to the systemic release of cytokines, mainly by cells of the monocytic lineage (hallmark of this clinic condition). Pro-inflammatory cytokines, i.e., TNF, IL-1, IL-6, IL-8, and interferon-gamma, are postulated to play a pivotal role in the pathogenesis of the syndrome, but regulatory mechanisms exist to counterbalance this overproduction of cytokines, as the shedding of neutralizing soluble TNF receptors, the production of IL-1 and IL-10 receptor antagonist [2, 5, 7]. PCT concentrations are markedly increased during the acute phase of septic shock, until levels equal or exceed 10 ng/mL [6], and during the recovery phase its level decreases to normal. Some data show that PCT may be a secondary mediator that might augment and amplify, but is not able to initiate the septic response. Immunoneutralization of PCT may prove to be an important clinical strategy, in view of its sustained elevation and the difficulty in initiating therapy for sepsis during the early phases of illness [2, 5, 7].

PCT elevation is closely dependent upon the host cytokine response to microbial challenge, which might be mitigated by the antibacterial effect of the antibiotics. Furthermore, the magnitude and the time course of this response may be related to the patient’s outcome [15] and to bacterial load.

Consequently, in the last decade, PCT has been used to guide antibiotic therapy in individual patients as a surrogate biomarker: using a highly sensitive PCT immunoassay with functional assay sensitivity of 0.06 μg/L, antibiotic selection based on PCT cut-off ranges has successfully been implemented in patients with lower respiratory tract infections (LRTI) in different clinical settings [16]. More recently, a multicenter randomized controlled trial (proHOSP) demonstrates, on the contrary, that in patients with LRTI, a strategy of PCT guided antibiotic therapy compared with standard guidelines results in similar rates of adverse outcomes, as well as lower rates of antibiotic exposure and antibiotic-associated adverse effect [17]. The results of the PRORATA trial (a multicentre, prospective, parallel-group, open-label trial) have been recently published [18]. In the procalcitonin group, antibiotics were started or stopped based on predefined cut-off ranges of procalcitonin levels; the control group received antibiotics according to the present guidelines. The mortality of the patients in the procalcitonin group seems to be superior to those in the control group at day 28 (21.2% [65/307] vs. 20.4% [64/314]; absolute difference 0.8%, 90% CI −4.6 to 6.2) and day 60 (30.0% [92/307] vs. 26.1% [82/314]; 3.8%, −2.1 to 9.7). The patients in the procalcitonin group had significantly more antibiotic free days than those in the control group (14.3 days [SD 9.1] vs. 11.6 days [SD 8.2]; absolute difference 2.7 days, 95% CI 1.4 to 4.1, p < 0.0001). The authors conclude that a procalcitonin-guided strategy to treat suspected bacterial infections in non-surgical patients in intensive care units might be able to reduce antibiotic exposure and selective pressure (i.e., induction of resistance to antibiotics) with no apparent adverse outcomes.

Some authors debate the PCT dosage in critically ill patients in the ICU, in fact in this setting, it is very easy to observe a systemic inflammatory response syndrome (SIRS) as the consequence of trauma, surgery, and hypoxic injuries [2–5]. In postoperative patients, if the PCT level increases for more than 24 h, bacterial sepsis should be suspected [19]. Unfortunately, the PCT diagnostic accuracy in the critical care setting is not certain, because many studies also included patients who did not have SIRS or who were not critically ill. A meta-analysis by Uzzan et al. [20] conducted on 49 studies in medical, surgical or polyvalent ICUs or postoperative wards, shows that PCT represents a good biological diagnostic marker for sepsis, severe sepsis or septic shock, so that PCT should be included in diagnostic guidelines for sepsis and in the clinical practice of ICUs. On the other side, a more recent meta-analysis by Tang et al. [21] evaluates all published studies that assessed the diagnostic use of procalcitonin in critical care settings, and concludes that PCT has a low diagnostic performance in differentiating sepsis from SIRS in critically ill adult patients, so that a widespread diffusion of its dosage in sepsis diagnosis may not be useful in ICUs. Furthermore, a mortally endpoint, large scale randomized controlled trial with a biomarker-guided strategy compared with the best standard of care, is still in progress to answer the question: can the survival of the critically ill patient be improved by actively using PCT in the treatment of infections? [22]

Another issue is whether PCT monitoring (that is assessing the dynamic PCT variations) can add information about infections both the diagnostic and the prognostic approach to critically ill patients. One recently published study shows that daily monitoring of PCT allows the medical staff to identify patients with the highest risk of mortality [23]. Besides, several investigations have shown that both PCT elevation and time course are helpful in differentiating between patients who acquire infection in the ICU and those who do not [24]. Most of these studies included postoperative patients.

Luyt et al. [25] document that the positive predictive value of an increase in PCT within the previous 5 days reaches 79% in 73 patients with clinically suspected ventilator-associated pneumonia (VAP). In contrast, the diagnostic accuracy of PCT elevation on the day VAP was initially clinically suspected was poor. Thus, positive and negative predictive values were 43 and 53%, respectively, if a threshold of 0.5 ng/mL was applied.

In a recent paper by Charles et al. [26] performed in an observational cohort of medical ICU patients, it is reported that PCT is helpful in the early detection of septic complications in critically ill medical patients. Moreover, the level of PCT obtained on the day the infection is suspected proves to be a better predictor than such clinical parameters as body temperature and other elements of the SIRS. The authors report that a PCT elevation of at least 0.26 ng/mL over the previous 24 h is strongly associated with the diagnosis of infection since the positive predictive value reaches 100%. It is, however, worth noting that the negative predictive value of PCT using such threshold values is quite low, reflecting the low sensitivity of PCT in their study population, and thus underlining the risk of false negative results.

Available data on this topic are so far controversial and discrepancies may be mainly related to differences in population selection criteria (surgical vs. medical ICU patients) as well as number consistency.

Recently, two different immunoassays for PCT measurement have been compared in 305 patients in the ED [27]. A highly significant correlation was observed between the two automated assays so that the authors conclude that procalcitonin concentrations obtained from both methods lead to the same clinical interpretation. Similar results have been obtained by Schuetz and colleagues [28].

Procalcitonin in acute coronary syndromes

So far, data on serum PCT levels in patients with cardiogenic shock and in those with acute coronary syndromes (ACS) are scarce and controversial. While some studies [29, 30] report that PCT levels are increased in ACS patients on admission, other investigations [31, 32] document that plasma PCT concentrations are in the normal range in patients with uncomplicated acute myocardial infarction.

Buratti et al. [31] assessed PCT values, together with IL-6, another index of inflammation, in a homogeneous population of patients with uncomplicated acute myocardial infarction (AMI). They observe that, on admission, PCT levels are in the normal range, whereas IL-6 levels are tenfold elevated. The extent of PCT release is less pronounced than that of other acute phase reactants (such as CRP and IL-6), so the authors conclude that PCT may have a predominant role as a marker of severe infections, but not of the acute phase response to acute myocardial tissue injury in patients with uncomplicated acute myocardial infarction. On the other hand, more recently, Kafkas et al. [29], who evaluated PCT levels in AMI patients on admission, report that PCT concentrations are elevated on admission, and are detectable in serum earlier than CK-MB or troponin I in most patients. The authors conclude that PCT can be considered as a novel sensitive myocardial index since its release in AMI is probably due to the inflammatory process that occurs during AMI. Similarly, Senturk et al. [33] document elevated PCT levels in patients with acute myocardial infarction with ST-elevation (STEMI), acute myocardial infarction without ST-elevation (NSTEMI), and unstable angina (UA), but fail to find any correlation between PCT and CRP concentrations, and between PCT and CRP levels and severity of underlying coronary artery disease and early prognosis.

Recently our group evaluated [34], on admission in our intensive cardiac care unit (ICCU), the plasma levels of PCT in the following subgroups of acute cardiac care patients submitted to percutaneous coronary intervention (PCI): (a) patients with cardiogenic shock (CS) following STEMI (STEMI); (b) patients with uncomplicated STEMI (Killip class I); and (c) patients with UA and NSTEMI. Furthermore, we investigated whether PCT levels were correlated or not to CRP concentrations.

The main findings of our investigation were as follows (Fig. 2): (1) PCT values are positive in all patients with cardiogenic shock, but in only a small percentage (33%) of uncomplicated STEMI and in very few (8.3%) NSTEMI/UA patients; (2) in our population the behavior of CRP and PCT values is not similar: while the percentage of positive CRP gradually increases from UA/NSTEMI setting to cardiogenic shock patients, PCT is positive only in some STEMI patients and in all CS.

Fig. 2
figure 2

Procalcitonin in cardiogenic shock and acute coronary syndromes

Full size image

According to our results, the degree of myocardial ischemia (clinically indicated by the whole spectrum of ACS, from unstable angina to cardiogenic shock following ST-elevation myocardial infarction) and the related inflammatory-induced response are better reflected by CRP (which was positive in most acute cardiac care patients of all our subgroups) than by PCT, which seems more sensitive to a higher degree of inflammatory activation, being positive only in all CS patients. This phenomenon is strongly supported by the finding of the lack of correlation between CRP and PCT levels in cardiogenic shock patients. Our results agree with those by Remskar et al. [32] who document normal PCT values in uncomplicated STEMI patients, whereas conventional inflammatory parameters are all significantly increased. Recently, Biasucci et al. [35], reviewed current available data on PCT and other biomarkers in acute coronary syndromes, and concluded that future investigations are needed to better understand the utility of PCT as a diagnostic and prognostic tool in this clinical setting besides the markers currently used in clinical practice, and that standardization of methods of assessment and cutoff are needed as well as larger studies of comparison of PCT with CRP.

Procalcitonin and cardiogenic shock

Elevated concentrations of PCT have been reported in patients with cardiogenic shock [29, 36]. In patients with cardiogenic shock, increased PCT values have been related to several factors, besides the inflammatory response, and especially to the release of cytokines, TNF-alpha, IL-6, and soluble TNF receptors [7]. Another proposed mechanism is the exposure to bacterial endotoxin due to bowel congestion or ischemia and altered gut permeability [36].

In a more recent retrospective study [30], it was observed that CS patients show a high PCT concentration, especially in the presence of multiorgan failure (MOF) and in the absence of signs of infections (cultures and clinical findings).

Few studies investigate the dynamics of PCT in cardiac acute patients. Sponholz et al. [37] describe the evolution of serum procalcitonin levels after uncomplicated cardiac surgery, and observe a progressive return to normal levels within the first week. Peak PCT levels are reached within 24 h postoperatively, and this increase seems to be dependent on the surgical procedure, more invasive procedures being associated with higher PCT levels, and on intraoperative events, including aortic cross-clamping time, duration of cardiopulmonary bypass, and duration of surgery.

In infected patients, PCT levels are elevated throughout the first postoperative week, with a more pronounced increase in bacterial and fungal infections than in viral infections of SIRS [38]. The authors conclude that the dynamics of PCT levels, rather than absolute values, may be more important for identifying patients with infectious complications after cardiac surgery. More recently Prat et al. [39] confirm a slight increase in PCT values in the first postoperative day after cardiac surgery, in agreement with previous results [19] and with Adamik et al. [40], who showed that the development of postoperative complications after cardiac surgery with cardiopulmonary bypass is associated with increased postoperative neopterin and PCT levels.

Similarly, after heart transplantation, serum PCT levels display a rise in response to surgery, with a peak on day 2, whereas high peak levels with delayed return to normal values should lead to a search for inflammatory processes, as they are often associated with increased morbidity and mortality [41]. Likewise, in patients with cardiogenic shock and no sign of infections, we document a reduction of PCT levels only in survivors CS patients [42].

This time course of procalcitonin levels can probably be explained, both in postsurgical and in CS patients, by normal PCT kinetics. In healthy subjects, the injection of endotoxin is followed by a rise in PCT, reaching a maximum in 24 h. [6]. The return of PCT levels to normal within a few days in surgical patients (after an uncomplicated postoperative course) and in CS survivors can be explained by the half-life of PCT (18–24 h) [3], in the absence of a further insult that might induce PCT production. Our findings, together with those observed in cardiac surgery [37, 39, 41] strongly support the contention that the “dynamic” approach [43] may be more reliable that the static one (that is an absolute single PCT value) especially in the challenging conditions characterized by a systemic inflammatory response, such as cardiac surgery and cardiogenic shock.

Indeed, in a cohort of unselected critically ill patients, Jensen et al. [44] observe that a PCT increase is an independent predictor of 90-day survival, and that the PCT day-by-day changes are able to identify critically ill patients at a higher risk of ICU mortality. On the other hand, the initial PCT level does not predict mortality, even though many patients were admitted with a PCT 1.0 ng/mL. This suggests that several PCT measurements should be made consecutively to assess the critically ill patient’s infection-related mortality risk (to monitor day-by-day the treatment of infection).

Procalcitonin after cardiac surgery

Cardiac surgery with cardiopulmonary bypass (CPB) is a highly sterile type of surgery. Nevertheless, it can lead to a SIRS since the exposure of blood to non-physiological surfaces, as well as myocardial and pulmonary ischemia/reperfusion due to aortic clamping and extracorporeal circulation, can be responsible for the development of a SIRS. Cardiac surgery with CPB triggers an inflammatory response involving pro-inflammatory cytokines such as TNF-α, IL-6, and IL-8 [45] as well as activation of the complement system because of exposure of the blood to artificial surfaces [46]. Endotoxin release from the ischemic gut is considered the main inducer of this inflammatory response [47], and alterations in hemostasis may also initiate alterations in inflammation at molecular level.

Therefore, in patients submitted to cardiac surgery, it is often impossible to distinguish a SIRS from a systemic inflammation induced by microorganisms.

Prat et al. [39] document in 151 patients undergoing cardiac surgery, a slight increase in PCT values on the first postoperative day (PCT 1), compared to PCT levels at admission and to the control group. These results are consistent with previous studies that report a moderate and transient peak on the first postoperative day in adults with and without CPB followed by a rapid return to normal levels [38, 40]. On the other hand, Boeken et al. [38] do not find an increase in PCT levels, neither during operation nor postoperatively when patients recover uneventfully. Unfortunately in their series, no samples were collected on the first postoperative day. Some authors find higher levels in patients after on-pump coronary artery bypass grafting (CABG) than in those after off-pump CABG [48] or higher after valvular surgery than after CABG [37]. In a recent review on this topic, it is stated that uncomplicated cardiac surgery induces a postoperative increase in serum PCT levels, and that PCT values reported in infected patients are generally higher than in non-infected, though no cut-off levels are recommended.

Prat et al. [39] observe that in patients submitted to cardiac surgery, highly increased PCT levels (above 3 ng/mL) are not found in the absence of postoperative complications, and are not specific to the surgery itself, duration of CPB, or time of aortic clamping. The authors conclude that PCT yields an elevated negative predictive value of complications after cardiac surgery.

Lastly, Jereb et al. [49] evaluated the accuracy of procalcitonin (PCT) in predicting infective endocarditis (IE) in 23 adult patients with IE, 30 patients with sepsis and 30 with tick-borne encephalitis. sPCT level, C-reactive protein (CRP), total leukocytes, and immature polymorphonuclear (PMN) cell counts were determined on admission, prior to the institution of antibiotic therapy, and compared according to the diagnosis: the study fails to demonstrate the superiority of PCT as a diagnostic laboratory parameter in predicting IE compared to CRP.

Conclusions

Data concerning PCT levels in patients with cardiogenic shock and in those with acute coronary syndromes (ACS) are scarce and controversial. Across the whole spectrum of ACS, from unstable angina to cardiogenic shock following STEMI, PCT seems to be more sensitive to a higher degree of inflammatory activation, being positive in patients with cardiogenic shock.

In our opinion, PCT measurements may help the clinician in the early detection of infectious complications in acute cardiac patients (both in patients with acute coronary syndromes and in postoperative cardiac surgical patients). Despite the scarce evidence on this topic, we strongly suggest obtaining serial measurements of PCT (and of CRP) in all patients with a clinical suggestion of infection before starting antibiotics. This could allow the confirmation of the clinical suspicion, and might represent a useful and effective way to monitor antibiotic efficacy. We believe that further research should develop in two directions: (a) assessing the prognostic role (if any) of PCT in a larger series of patients with acute coronary syndromes, both short and long term; and (b) evaluating whether, in the presence of an infectious complication, PCT is able to follow the host’s response to antibiotics even in acute cardiac patients, where a systemic inflammatory activation is detectable.