Procalcitonin and other biomarkers to improve assessment and antibiotic stewardship in infections – hope for hype ?

This review aims to provide physicians with an overview of the potential of procalcitonin to guide antibiotic therapy in respiratory tract infections and in sepsis. Knowledge of the strengths and weaknesses of procalcitonin are prerequisites for a rational and safe use in clinical routine. In most infections a true gold standard for diagnosis does not exist, therefore physicians must remain sceptical towards observational studies evaluating procalcitonin. Interpretation of procalcitonin levels must always include the clinical setting and knowledge of assay characteristics, particularly the setting of specific cut-off ranges and functional assay sensitivities. Highly sensitive procalcitonin measurements, embedded in a clearly defined setting and prospectively validated with clinical algorithms were repeatedly effective in markedly reducing the (over)-utilisation of antimicrobial therapy. Today, this concept has been proven for lower respiratory tract infections and in pilot studies for meningitis and critically ill patients with sepsis. The higher the absolute risk for adverse outcome of a patient, the more cautious physicians must remain and empirical antibiotic therapies must be considered despite initial low procalcitonin levels at the initial presentation. In these patients a procalcitonin-guided shortening of antibiotic courses seems appropriate. The prognostic utility of initial procalcitonin measurement in respiratory tract infections is suboptimal. Other biomarkers including cortisol, human growth hormone and prohormones from adrenomedullin and vasopressin (“copeptin”) have a superior predictive potential to estimate the risk for short and long term mortality and other adverse outcomes in different diseases. An accurate prognostic assessment has the potential to optimise the management of patients and the allocation of our limited health care resources by lowering unnecessary hospitalisations and associated cost. Future intervention studies must prove if these biomarkers indeed improve clinical decision making and thus the overall medical management of patients.


Summary
In bacterial sepsis, pathogens and their antigens stimulate a plethora of pro-and anti-inflammatory mediators which characterise the host defence and orchestrate leukocyte recruitment to the acute site of infection.Precursors, mature forms and degradation products of these different mediators penetrate from the initial site of action into the circulation, where, in theory, they can all be measured.As surrogate biomarkers these substances mirror the extent and severity of an infection.Enormous attempts have been undertaken to correlate the levels of different mediators with the presence of sepsis [1][2][3].Procalcitonin (PCT) has been demonstrated to be clinically most use-ful and superior to commonly used clinical variables and other laboratory tests in the diagnosis of sepsis and to correlate with the extent and severity of microbial invasion [2,[4][5][6][7][8][9][10]. The dual function of PCT as a precursor peptide from the hormone calcitonin and a cytokine mediator, which is elevated upon systemic bacterial infections in line with other cytokines, has led to the term "hormokine" mediator [11].Other surrogate markers have been put forward to diagnose the presence of bacterial infections and guide antibiotic therapy.These include soluble triggering receptor expressed on myeloid cells-1 (s-TREM-1), [12,13] C-reactive protein (CRP) [14] and white

Biomarkers in sepsis
PS, MCC and BM received support from BRAHMS to attend meetings and fullfilled speaking engagements.BM has served as a consultant and received research support.
blood cell count [15].However, the usefulness of these markers for guiding antimicrobial therapy is not supported by controlled intervention trials and hampered by a delayed response with late peak levels, the need for invasive sampling of bronchoalveolar fluid (s-TREM-1) and their modulation by anti-inflammatory drugs such as corticosteroids, respectively [5,16,17].Therefore, these tests have suboptimal practicability and diagnostic accuracy, especially in patients with systemic inflammation [13,[18][19][20].
Importantly, unlike other diagnostic biomarkers such as C-reactive protein (CRP), there is strong evidence from animal studies demonstrating that PCT plays a pathophysiological role in the development of severe sepsis and associated mortality.Administration of PCT to septic hamsters with peritonitis doubled their death rate [21].Conversely, treatment with PCT-reactive antiserum increased survival of septic hamsters and pigs, with mono-and polymicrobial sepsis, respectively [21][22][23][24].A one hour intravenous immuno-neutralisation of porcine PCT, improved all vital parameters in septic pigs.It increased short-term survival from 0% to 80% and was effective even when administered after the animals were moribund [22].In humans, similar studies need to be funded.
As a diagnostic marker, PCT has several advantages over other inflammatory markers.The findings of many clinical studies have established the superior diagnostic accuracy of PCT in severe infections [19,25].PCT shows an earlier increase upon infection and a more rapid decrease when the infection is controlled by the immune system supported by antibiotic therapy.PCT correlates with the extent and severity of infection and has prognostic implications, as the course of PCT predicts the risk for mortality in critically ill patients with infections [26] and in patients with ventilator-associated pneumonia (VAP) [16,17,26].Furthermore, the production of PCT, in contrast to other biomarkers including CRP, seems not to be significantly attenuated by non-steroidal and steroidal anti-inflammatory drugs.[16,17,27] In this context it is interesting to note that the favourable effects of corticosteroids were reported in patients with pneumococcal pneumonia as early as 1955 [28].More recently, a small prospective trial [29], and a large retrospective single centre trial suggested that the use of corticosteroids was associated with a lower mortality [30].Other studies [31,32] were not able to show a mortality benefit, however, a shorter duration of intravenous antibiotics and more rapid normalisation of inflammatory markers was found [32].PCT concentrations were not assessed within these studies.Whether PCT is a useful biomarker for risk stratification and selection of those patients likely to profit from corticosteroids remains an open question

The "gold standard" dilemma in sepsis
The presence of a diagnostic gold standard or reference standard remains the best available method for establishing the presence or absence of a disease [33,34].Optimally, a morphological verification such as histopathology or, in the case of sepsis, growth of typical pathogens in blood cultures can be obtained to establish the "correct" diagnosis.In this context, repeated collection of blood cultures to identify causative microorganisms and to study resistance patterns is propagated to be a cornerstone in the diagnostic work up of patients with suspicion of a sepsis syndrome.Regrettably, blood culture as the alleged "gold standard" lacks both sensitivity and/or specificity and is only available after a time delay.The causative microorganisms cannot be detected in up to 80% of patients with suspected BSI [8,35].Of note, in more than 70% of patients with radiologically confirmed community acquired pneumonia (CAP) the causative microbe is never identified.Conversely in patients with acute exacerbations of chronic obstructive pulmonary disease (COPD), typical respiratory bacterial strains are isolated from sputum in up to 50%, representing colonisation in many cases.
In this diagnostic uncertainty of bacterial infections, diagnostic surrogate biomarkers are of interest.In the past years, countless observational studies have analysed a plethora of novel biomarkers and particularly PCT.In conventional diagnostic accuracy studies, the usefulness of a novel test is determined by comparing the results with the definite diagnosis ascertained by the gold standard.Sepsis is merely a clinical syndrome, not a final diagnosis, encompassing highly heterogeneous groups of disorders varying with respect to the site, bacteriology and even presence of infection, where a true gold standard is lacking [36].In such circumstances, two fundamentally different concepts are used [34].One concept tends to ignore potential dilemmas in the accuracy of the alleged "gold standard" but assumes a well-defined illness, which is represented by the assumption of a diagnostic test or a clinical diagnosis [34].Many observational studies in different clinical settings using PCT assays have been published investigating the diagnostic accuracy of PCT for the diagnosis of sepsis (fig.1).The majority of these studies used the clinical evaluation of the patient and the presence of the "septic syndrome" as the reference standard ("gold standard").Accordingly, these observational studies and resulting metaanalyses [19,25,[37][38][39] reported in part conflicting and confusing results since they are biased by the interpretation of the investigator, and thus, should be interpreted with caution.The other concept discards alleged gold standards and focuses primarily on the outcomes of patients.In the case of sepsis, the clinical benefit of PCT can be measured by clinical outcomes of randomised intervention studies, assuming that if the patient recovers without antibiotics there was no relevant bacterial illness.In our opinion, only randomised controlled trials (RCTs), in which antimicrobial therapy is guided by specific cut off ranges of PCT or other biomarkers and in which the primary measure of efficacy is medical outcome, have the potential to evaluate the ultimate clinical usefulness of a diagnostic biomarker.

Procalcitonin for guidance of antibiotic therapy in respiratory tract infections
As PCT level increases upon bacterial infection and decreases upon recovery, it can be used to guide antibiotic therapy in individual patients as a surrogate biomarker.Highly sensitive PCT assays are needed to reliably diagnose CAP and non-CAP lower respiratory tract infection (LRTI) [40][41][42][43].Using a highly sensitive PCT immunoassay with functional assay sensitivity of 0.06 μg/l, antibiotic stewardship based on PCT cut off ranges has successfully been implemented in patients with lower respiratory tract infections (LRTI) in different clinical settings [44][45][46][47][48][49][50].For this purpose, specific PCT cut-off ranges, reflecting the setting specific likelihood of relevant bacterial infections, have been proposed and translated into a clinical algorithm (fig.2).Based on these specific cut-off ranges, initiation or continuation of antibiotics was more or less discouraged (<0.1 μg/l or <0.25 μg/l) or encouraged (>0.5 μg/l or >0.25 μg/l), respectively.In case antibiotics were withheld, clinical re-evaluation and a repeated measurement of PCT was recommended after 6-24 hours.If PCT values were increased and antibiotic therapy was initiated, repeated PCT measurements were recommended and an-tibiotics were discontinued using the same cut-off ranges.In patients with very high PCT values on admission (e.g.>10 μg /l), discontinuation of antibiotic therapy was encouraged if levels decreased to below 80-90% of the initial value.To assure the safety of patients, specific "overruling" criteria were predefined, where this algorithm could be bypassed (e.g.life-threatening disease or immediate need for ICU admission).Physicians were advised that persistently elevated PCT levels may indicate a complicated course, while PCT levels may remain relatively low in localised infections (e.g.empyema or abscess).This clinical algorithm was prospectively tested in different intervention trials in the emergency room (ER) [45][46][47][48], in a pilot study in an intensive care unit (ICU) [51] and in a primary care setting with over 50 primary care physicians [49,50].Currently, this concept is been externally validated in the nationwide multicentre ProHOSP study including over 1300 patients [46].The validity of this algorithm was measured by good clinical outcomes.Thereby, PCT-guided antibiotic stewardship reduced the antibiotic prescription rate by 40-50% in patients with LRTI presenting to the emergency room, in- Potential drawbacks of observational studies in procalcitonin research.In recent years, a plethora of observational studies have investigated the diagnostic accuracy of PCT in different clinical settings.The majority of these observational studies used the clinical evaluation of the patient as the alleged "gold standard" to differentiate true bacterial infection or no infection.In addition, selection of patients, heterogeneity of clinical settings underlying diagnoses of patients and PCT assays with different functional assay sensitivities may lead to confounding study results.Unsurprisingly, these studies reported, at least in part, confusing and contradicting results.As a consequence, different metaanalyses drew different and even opposite conclusions depending on the inclusion of underlying studies.Only randomised controlled trials (RCTs), in which antimicrobial therapy is guided by PCT and in which the primary measure of ef f icacy is the medical outcome of patients, have the potential to resolve this dilemma and evaluate the clinical usefulness of PCTguided antibiotic stewardship.RCTs denote randomised controlled trials, observational studies [103][104][105][106][107], metaanalyses [19,25,[37][38][39] and RCTs [45][46][47][48][49][50][51]  cluding patients with acute exacerbation of chronic bronchitis (COPD), with no increase in relapse rate over a six month follow up period [48].In patients with community-acquired pneumonia (CAP), PCT-guidance reduced the initial prescription rate by about 10%, but importantly shortened the duration of antibiotic therapy by 65% with a similar outcome in patients with all degrees of severity of CAP [47].In the PARTIstudy the safety and feasibility of the PCT guided algorithm was proven in primary care, with enrolment of more than 450 patients with acute upper and lower respiratory tract (RTI) [49,50].In a low antibiotic prescription setting in Switzerland, PCT-guidance safely reduced antibiotic exposure by more than 75% and thereby significantly re-duced antibiotic-related side effects, especially diarrhoea.Using a similar PCT algorithm, a recently published trial including critically ill septic patients from a medical ICU reported a four day shorter duration of antibiotic therapy in PCTguided patients (6 as compared to 10 days in the control group) with a similar mortality and recurrence rate [51].Importantly, a two day shorter intensive care unit stay was observed in patients assigned to the PCT guided group.Large multicentre trials (ProHOSP [46], PRORATA, PASS) are currently being finalised as this review is in print.The cumulated evidence from RCT's will include almost 5000 patients, a truly remarkable number.

Vascular catheter-related bloodstream infection
Commensals from the human skin flora, mainly coagulase-negative staphylococci, play a major role in catheter colonization.The distinction of blood contamination from blood stream infection (BSI) is important to promptly initiate an adequate therapy in true infection and, in case of contamination, to avoid unnecessary antimicrobial usage.Conventional methods of diagnosing catheter-related BSI generally require catheter removal and culture with quantitative or semi quantitative methods.However, the majority of catheters are withdrawn unnecessarily and the removal of a central venous catheter may be undesirable because of limited vascular access and the potential complications associated with reinsertion.Thus, diagnostically reliable and more costeffective diagnostic approaches that do not re-quire catheter removal would be desirable.In this context, the diagnostic value of serum PCT to distinguish blood contamination from BSI due to coagulase-negative staphylococci has been evaluated in a pilot study [10].Therein, PCT demonstrated a better discriminatory ability as compared to white blood count and serum CRP.In this study [10] increased PCT concentrations were found even prior to the clinical manifestation of BSI (i.e.fever) illustrating the high sensitivity of PCT to detect colonization of the catheter with only subclinical infection ultimately leading to BSI [52].Importantly, the optimal cut off to achieve a 100% sensitivity in this study was at a low PCT cut-off of 0.1 μg/l illustrating, once again, the need for a highly sensitive PCT assay in many clinical settings outside the ICU [53].Early differentiation between septic and non-septic arthritis and the possible decision for early therapy remain a difficult task for the physician.In a pilot study including 42 patients with bacterial and non-bacterial arthritis, PCT was shown to be the most accurate diagnostic biomarker as compared to CRP and white blood cell count [108].Importantly, using a highly sensitive PCT assay with at a cut-off in the range of healthy persons (0.1 μg/l), sensitivity for septic arthritis was 100% and specificity 46%.The area under the receiver operating characteristic (ROC) curve (AUC) of PCT was 0.92 as compared to 0.72 of CRP.Non-septic arthritis with coinfection refers to patients with systemic bacterial infections and concomitant diagnosis of gout arthritis.

Differentiation of septic from non-septic arthritis
Early differentiation between septic and nonseptic arthritis is of utmost importance to prompt adequate initial therapy and surgical measures.As clinical signs and traditional markers of infection are of only limited value in ruling out a bacterial aetiology, arthrocentesis with synovial gram stain and culture is mandatory.Even so the sensitivity of gram staining is only 50-75% and culture is not available immediately [54].A more specific diagnostic biomarker would thus be most useful.Previous studies evaluating PCT in this setting have been inconclusive, mainly because of small and heterogenic patient populations and because they used only semi quantitative PCT assays [55][56][57][58].A recent study including 42 patients with septic and non septic arthritis found PCT to be most useful and superior to other markers of infection.Importantly, PCT had a sensitivity of 100% (93%) to rule out bacterial infection, when a low PCT cut off in the normal range of 0.1 μg/dl (0.25 μg/dl) was used (fig.3).

Biomarkers for prognostic assessment
LRTI and particularly CAP is the leading cause of sepsis and death from infectious diseases in western countries and health expenditures in particular for inpatient management are substantial [59,60].Accurate assessment of disease severity, risk stratification and prediction of outcome are prerequisites for safe decision making on the need for hospitalisation and identifying patients at low risk of complications and thus suitable for outpatient management.Despite their widespread use in clinical routine, traditional markers such as severity of disease estimation by the patient, fever, white blood cells and CRP do not reliably assess disease severity and mortality risk [2].Several organizations have developed prediction rules and disseminated guidelines to stratify management of patients based on predicted mortalities in order to optimise hospital referral and lower hospital admission rates [61,62].The pneumonia severity index (PSI) is a extensively validated and widely propagated American scoring system that assesses the risk of death in a two step algorithm [63].However, its complexity jeopardizes its dissemination and implementation in everyday practice.Therefore, the CURB-65 and the CRB-65 scores, modified versions of the British Thoracic Society (BTS) assessment tool, which are based on only five and four predictors, respectively, have been proposed as a simpler but somewhat less reliable alternative [64,65].All these scores, however, have only been evaluated in CAP and not in all LRTI, and for mortality and not other serious adverse events requiring hospitalisation.They are heavily age-dependent and have a considerable risk of miscalibration and thus misclassification of patients depending on the clinical setting [66].In this context, new measurable biomarkers mirroring distinct pathogenetic mechanisms to predict severity and outcome may improve prognoses of patients.Importantly, the utility of a biomarker in this context is defined by the degree it improves clinical decision making and adds timely information beyond that of readily available information from clinical examination [67].
While a prognostic value of PCT in critically ill patients from the intensive care unit has repeatedly been demonstrated [68,69], the prognostic value of a single level on admission is only suboptimal.A recent study suggested that very low PCT values on admission (optimal PCT cut off at 0.23 μg/l) in patients with community acquired pneumonia (CAP) have a moderate negative prognostic value and thereby improve the CRB65 score [70].Similarly, in patients with CAP due to Legionella, high initial PCT concentrations predicted adverse outcome in patients including mortality and need for ICU admission (fig.4) [71].This being said, it is even more helpful to consider the dynamics of PCT levels in LRTI, as persistently elevated levels are associated with adverse outcome and decreasing PCT levels suggest a favourable outcome, usually showing a log-linear drop off and a half life of 20 to 24 hours [4,70,72].Similarly in the ICU setting, elevated PCT levels, and especially a PCT increase for one day were independent predictors for 90-day all cause mortality in septic patients [26], while increasing levels of CRP and white blood cells did not pre-dict mortality.Similarly in our studies, PCT showed a better prognostic accuracy compared to CRP and white blood cells, but there was a wide overlap in PCT levels between different severities of CAP and only a small difference in PCT levels between survivors and non-survivors [2,5,43,73].Based on these data, PCT seems to be a reliable diagnostic marker able to guide decisions on antibiotic therapy rather than an ideal prognostic tool.This is a disadvantage especially on admission where the costly decision for in-or outpatient treatment has to be made [2,5,73,74].
A more promising prognostic biomarker on admission is adrenomedullin (ADM).It is also a member of the CALC gene family, and is very potent vasodilating agent with additional immune modulating and metabolic properties [75][76][77][78].Measurement of ADM is challenging, since it is rapidly cleared from the circulation [75,76].A new sandwich immuno-assay, which measures the more stable mid regional fragment of proadrenomedullin (proADM) directly reflecting levels of the rapidly degraded active peptide, has been developed [79].In patients with CAP, proADM levels measured on admission showed a comparable correlation with disease severity and outcome of CAP as compared to the PSI.Importantly, proADM improved the prognostic accuracy of the PSI alone, acting as an additional margin of safety [73].
During severe illness a series of other important hormonal changes happen mirroring the patients individual "stress level" [80][81][82].For this reason, stress hormones per se are promising candidate "biomarkers" for the prognostic assessment of patients with systemic infections.The classical stress response of the body includes a stimulation of the hypothalamic-pituitary-adrenal axis with an increased production of cortisol and vasopressin levels and a functional deficit of anabolic hormones [80,81].As plasma cortisol levels mirror the individual stress level associated with the underlying illness, we evaluated the prognostic value of cortisol in patients with CAP [83].Cortisol concentrations measured on presentation accurately predicted the severity and outcome of CAP at a level similar to the PSI and higher as compared to commonly measured laboratory parameters, namely CRP, white blood cells and PCT.The higher level of pro-inflammatory cytokines in patients who subsequently died may additionally explain the pronounced increase of cortisol as the strongest known natural inhibitor of inflammation in non surviving patients.
Circulating levels of growth hormone are increased during critical illness and correlate with outcome in children with meningococcal sepsis.Growth hormone plasma concentrations on admission are independent predictors for mortality in critically ill adult patients and may complement existing risk prediction scores, namely the APACHE II and the SAPS II score [84] .
Another, less known but nevertheless pivotal stress hormone is the anti-diuretic hormone (ADH), also known as vasopressin, which has haemodynamic and osmoregulatory effects and reflects the individual stress response.This hormone co-regulates the secretion of ACTH together with corticotrophin releasing hormone.
Copeptin which is stoichiometrically co-secreted with vasopressin, and directly mirrors vasopressin production, has a longer half life in the circulation and is thus easier to measure [85].We evaluated the prognostic value of copeptin and found significantly higher concentrations in patients with LRTI as compared to controls with highest levels in patients with CAP [86].Copeptin levels increased with increasing severity of CAP, and in patients who died, copeptin levels on admission were significantly higher as compared to levels in survivors [86].Recently, in patients with acute exacerbations of chronic obstructive pulmonary disease (COPD) copeptin was shown to be predictive for long-term clinical failure, independent of age, comorbidity, hypoxaemia and lung functional impairment in multivariate analysis [87].The combination of copeptin and previous hospitalisation for COPD increased the risk of poor outcome.
Another interesting endothelium derived biomarker is endothelin-1 (ET-1) and its precursor peptide proET-1 [88].In humans, elevated plasma levels of mature ET-1 are found during systemic infections and correlate with mortality risk [89][90][91].Preliminary studies in animals have demonstrated beneficial effects of ET-1 antagonism using a selective ET-1 receptor antagonist [9] during septic shock [92][93][94][95].Using a new sandwich immunoassay, we recently evaluated proET-1 levels A recent retrospective study including patients with CAP due to Legionella found that initial PCT concentrations and the course of PCT have prognostic implications: patients with initially high (optimal cut off >1.5 μg/dl) or increasing PCT concentration have a higher risk for death and/or need for ICU admission as compared to patients with low or decreasing PCT concentrations [71].
in patients with CAP and found proET-1 on admission to be independent predictors of short term mortality and need for ICU admission.In this setting, proET-1 levels improved the prognostic accuracy of the commonly used CURB65 score to predict adverse outcome [96].Interestingly in critically ill and septic patients from a medical ICU, the ratio of the endothelium-derived and counteracting substances proADM and proET-1 showed the highest predictive accuracy [97].
It is advisable to base the difficult task of prognostic assessment and treatment decisions on several rather than a single parameter.We have thus studied other hormonal and metabolic mediators and biomarkers in CAP and in LRTI, in-cluding brain-natriuretic-peptide (BNP) [98], Nterminal pro-atrial natriuretic peptide (MR-proANP) levels [99,100] and plasma lipid levels [101,109], each mirroring different pathophysiological aspects, showing promising results in predicting disease severity and outcome [2,102].
There is rational and emerging evidence that different biomarkers have the potential to complement existing clinical severity scores and thus improve the risk stratification of patients with LRTI.However, specific cut off ranges need to be proposed and prospective intervention studies conducted to order to decide if these biomarkers really do improve the daily clinical management of patients with sepsis.

Limitations of biomarkers
Obviously, the septic syndrome is far too heterogeneous and complex to be reduced to a single cut-off of any one surrogate marker.Different microbes might induce distinct responses resulting in a variable upregulation of circulating biomarkers and mediators.Still, the likelihood of a bacterial infection increases gradually with increasing serum levels of PCT.Knowledge of assay characteristics, particularly the functional assay sensitivity and the strengths, pitfalls and optimal cut off ranges in a predefined clinical settings are prerequisites for its optimal use in clinical routine.
Importantly, biomarker levels must always be evaluated in the context of a careful clinical and microbiological assessment.As the kinetics of biomarkers are of particular diagnostic and prognostic interest, repeated measurements should always be performed, especially if antibiotics are withheld and in persistently sick patients.In addition, immuno-modulating drugs may suppress the upregulation of different biomarkers.While the kinetics of PCT are less influenced by corticosteroids, CRP, proADM, proANP and Copeptin are inhibited in a dose-dependent way upon prednisone treatment [27].Limitations of every bio-marker include false-positive and false-negative results [16].Unspecific elevations of PCT levels in the absence of a bacterial infection can typically be seen in situations of massive stress, e.g. after severe trauma or surgery [16,25].In these situations, PCT values are usually only moderately elevated and show a rapid decline in the follow up measurements.Conversely, falsely low PCT levels, typically seen during the early course or localised state of an infection, often show an increase in the follow-up measurements.Importantly, in these situations highly sensitive PCT assays are required, as subtle changes of PCT at very low concentration can be monitored increasing its sensitivity and thus the safety of patients.

Figure 3
Figure 3Differentiation of septic and non-septic arthritis.Early differentiation between septic and non-septic arthritis and the possible decision for early therapy remain a difficult task for the physician.In a pilot study including 42 patients with bacterial and non-bacterial arthritis, PCT was shown to be the most accurate diagnostic biomarker as compared to CRP and white blood cell count[108].Importantly, using a highly sensitive PCT assay with at a cut-off in the range of healthy persons (0.1 μg/l), sensitivity for septic arthritis was 100% and specificity 46%.The area under the receiver operating characteristic (ROC) curve (AUC) of PCT was 0.92 as compared to 0.72 of CRP.Non-septic arthritis with coinfection refers to patients with systemic bacterial infections and concomitant diagnosis of gout arthritis.

Figure 4
Figure 4PCT concentrations in CAP due to Legionella.A recent retrospective study including patients with CAP due to Legionella found that initial PCT concentrations and the course of PCT have prognostic implications: patients with initially high (optimal cut off >1.5 μg/dl) or increasing PCT concentration have a higher risk for death and/or need for ICU admission as compared to patients with low or decreasing PCT concentrations[71].
are referenced.AJRCCM denotes American Journal of Respiratory and Critical Care Medicine; Archives denotes Archives of Internal Medicine; CID denotes Clinical Infectious Diseases; CCM denotes Critical Care Medicine; LID denotes Lancet infectious Diseases, AEM denotes American journal of Emergency Medicine, JCEM denotes Journal of Clinical Endocrinology and Metabolism, JLCM denotes Journal of Laboratory and Clinical Medicine.