The value of serum procalcitonin level for differentiation of infectious from noninfectious causes of fever after orthopaedic surgery

COPYRIGHT  2010 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED The Value of Serum Procalcitonin Level for Differentiation of Infectious from Noninfectious Causes of Fever After Orthopaedic Surgery By Sabina Hunziker, MD, Thomas H¨ugle, MD, Katrin Schuchardt, MD, Isabelle Groeschl, MD, Philipp Schuetz, MD, Beat Mueller, MD, Walter Dick, MD, Urs Eriksson, MD, and Andrej Trampuz, MD Investigation performed at the Departments of Orthopaedic Surgery and Traumatology, University Hospital Basel, Basel, Switzerland Background: Early diagnosis of postoperative orthopaedic infections is important in order to rapidly initiate adequateantimicrobial therapy. There are currently no reliable diagnostic markers to differentiate infectious from noninfectiouscauses of postoperative fever. We investigated the value of the serum procalcitonin level in febrile patients afterorthopaedic surgery.
Methods: We prospectively evaluated 103 consecutive patients with new onset of fever within ten days after ortho-paedic surgery. Fever episodes were classified by two independent investigators who were blinded to procalcitoninresults as infectious or noninfectious origin. White blood-cell count, C-reactive protein level, and procalcitonin level wereassessed on days 0, 1, and 3 of the postoperative fever.
Results: Infection was diagnosed in forty-five (44%) of 103 patients and involved the respiratory tract (eighteen pa-tients), urinary tract (eighteen), joints (four), surgical site (two), bloodstream (two), and soft tissues (one). UnlikeC-reactive protein levels and white blood-cell counts, procalcitonin values were significantly higher in patients with infectioncompared with patients without infection on the day of fever onset (p = 0.04), day 1 (p = 0.07), and day 3 (p = 0.003).
Receiver-operating characteristics demonstrated that procalcitonin had the highest diagnostic accuracy, with a value of0.62, 0.62, and 0.71 on days 0, 1, and 3, respectively. In a multivariate logistic regression analysis, procalcitonin was asignificant predictor for postoperative infection on days 0, 1, and 3 of fever with an odds ratio of 2.3 (95% confidenceinterval, 1.1 to 4.4), 2.3 (95% confidence interval, 1.1 to 5.2), and 3.3 (95% confidence interval, 1.2 to 9.0), respectively.
Conclusions: Serum procalcitonin is a helpful diagnostic marker supporting clinical and microbiological findings formore reliable differentiation of infectious from noninfectious causes of fever after orthopaedic surgery.
Level of Evidence: Diagnostic Level II. See Instructions to Authors for a complete description of levels of evidence.
trauma surgery and can be caused by infections or not helpful in differentiating infectious from noninfectious noninfectious conditions1-3. Damaged tissue due to causes of postoperative fever.
trauma and surgical intervention and the postoperative healing For the diagnosis of bacterial infections, elevated serum process can lead to the production of proinflammatory cyto- procalcitonin has been demonstrated to have higher diagnostic kines and can induce a nonspecific systemic inflammatory accuracy than clinical findings or standard laboratory pa- response syndrome4 without true infection. In addition, other rameters, such as the white blood-cell count and serum C- factors such as hematoma in the surgical site, transfusion of reactive protein levels, in various clinical settings5-17. The value blood or blood products, lung atelectasis, deep venous of elevated serum procalcitonin in the diagnosis of infections thrombosis, and adverse drug reactions also may provoke has been demonstrated for specific surgical settings, such as postoperative fever. Conventional laboratory parameters are cardiac surgery after cardiopulmonary bypass, lung decorti- Disclosure: In support of their research for or preparation of this work, one or more of the authors received, in any one year, outside funding or grants inexcess of $10,000 from the Swiss National Science Foundation (3200B0-112547/1), Stanley Thomas Johnson Foundation, and Gebert R¨ uf Stiftung.
Neither they nor a member of their immediate families received payments or other benefits or a commitment or agreement to provide such benefits froma commercial entity.
J Bone Joint Surg Am. 2010;92:138-48 d doi:10.2106/JBJS.H.01600 cation, major neurosurgery, and abdominal surgery5,18-24. In culture), and collection of two pairs of aerobic and anaerobic addition, elevated procalcitonin values correlate with the ad- blood cultures (each pair of blood samples was harvested thirty verse prognosis of patients after thoracic surgery and ventilator- to sixty minutes apart). Blood cultures were collected as early associated pneumonia25,26. However, the diagnostic accuracy of as possible after fever onset as part of the routine procedure. To procalcitonin levels to distinguish infectious from noninfec- study the kinetics of laboratory parameters, blood was col- tious causes of fever in patients after orthopaedic surgery is not lected for laboratory investigations at the onset of fever (day 0) and at one day (day 1) and three days (day 3) thereafter. In Given the high prevalence of postoperative infection and patients with a suspected respiratory tract infection, the results its impact on mortality and morbidity in patients undergoing of sputum or tracheal secretion cultures or antigen testing for orthopaedic and nonorthopaedic procedures26, a reliable Legionella pneumophila and Streptococcus pneumoniae in the marker for the diagnosis of infection would be of great im- urine were collected. In patients with a suspected wound in- portance. It would allow the initiation of empirical antimi- fection, a wound swab sample was obtained for microbiolog- crobial therapy rapidly in patients with an infection and avoid ical testing. Patient records were prospectively abstracted with unnecessary antimicrobial usage in patients without an infec- use of a standardized data-collection case report form to re- tion, thereby saving health-care costs and preventing the trieve demographic, clinical, microbiological, radiographic, development of antimicrobial resistance. For serum pro- and laboratory data. All patients were followed until hospital calcitonin, this approach has been successfully validated for discharge by one of the study physicians and were monitored respiratory tract infections6-10,27.
for the occurrence of any complications including a recurrent We therefore prospectively evaluated consecutive patients or new infection.
with fever after orthopaedic procedures and trauma, using astandardized diagnostic procedure to diagnose or exclude in- fection. We then assessed the diagnostic accuracy of standard Laboratory analyses included the determination of the white (white blood-cell count and C-reactive protein level) and blood-cell count, C-reactive protein level, and procalcitonin investigational (procalcitonin) laboratory parameters to distin- level from the routinely collected blood samples. C-reactive guish fever episodes of infectious causes from those with non- protein concentrations were determined by an enzyme immu- infectious causes in a blinded manner. Repeated determinations noassay (EMIT; Merck Diagnostica, Zurich, Switzerland) having of blood values were performed during the postoperative period a detection limit of <5 mg/dL. Procalcitonin was determined to determine the kinetics of the inflammatory parameters.
with use of a rapid ultrasensitive immunoluminometric assaywith an assay turnaround time of less than twenty minutes Materials and Methods and a functional detection limit of 0.06 ng/mL (KRYPTOR; B.R.A.H.M.S., Hennigsdorf, Germany). Blood cultures were pro- The study was conducted in the Departments of Ortho- cessed with use of an automated colorimetric detection system paedic Surgery and Traumatology at the University Hos- (BacT/ALERT; bioM´erieux, Durham, North Carolina)28.
pital Basel in Switzerland, an 800-bed tertiary health-carecenter. Between May 2006 and October 2007, we prospectively included consecutive hospitalized patients who were eighteen On the basis of the diagnostic procedures, patients were pro- years of age or older with a new onset of fever within ten days spectively assessed and febrile episodes were independently after surgery, including fracture fixation (upper limb, lower classified by the physician in charge on the ward and by a limb, and spine) and orthopaedic procedures (joint arthro- senior infectious diseases consultant. Classification of feb- plasty, fracture stabilization, or fracture repair). Patients were rile episodes was done after collection of all diagnostic in- screened daily for study eligibility by one of the study coor- formation and was based on clinical judgment, routine dinators (S.H., T.H., K.S., and I.G.), who were internal med- laboratory results including the C-reactive protein level and icine physicians in charge on the surgical ward. For this study, white blood-cell count measurements, and definitions of fever was defined as a core body (tympanic membrane) tem- nosocomial infections according to the Centers for Disease perature of ‡38.5C (‡101.3F) at a single measurement or Control and Prevention29. In brief, pneumonia was defined ‡38.0C (‡100.4F) determined at two consecutive measure- as the presence of (1) at least one respiratory symptom ments within one hour. Patients who had, or were incubating, (cough, sputum production, dyspnea, tachypnea, or pleu- a preexisting infection before surgery or who had a fever on the ritic pain) and at least one finding during auscultation (rales day of surgery were excluded. The local ethical committee or crepitation), or (2) one sign of infection (a core body classified this study as a quality control study and waived the temperature of >38.0C, shivering, or a white blood-cell need for obtaining patient informed consent.
count of >10 · 109/L or <4 · 109/L) and a new infiltrate on achest radiograph; urinary tract infection was defined as (1) Standardized Diagnostic Procedure substantial leukocyturia (>10 white blood cells per visual Patients included in the study were assessed with a standard- field on microscopy of sediment per high-power field) or (2) ized diagnostic procedure, including full clinical examination, substantial bacteriuria (>105 bacteria per milliliter of chest radiograph, urine sediment investigation (analysis and urine). Surgical site infection was defined as microbiologi- cally proven superficial incisional, deep incisional, or organ analysis and report odds ratios (i.e., the ratio of the probability and/or space infection; prosthetic joint infection was defined that an event will occur compared with the probability that the as the presence of (1) visible purulence, (2) acute inflam- event will not occur). Further, we calculated a receiver-operating mation on histopathological analysis, (3) a sinus track, or characteristic analysis and report sensitivity (the proportion (4) microbial growth in synovial fluid or periprosthetic of actual positives that are correctly identified as such, i.e., tissue; and bloodstream infection was defined as growth of the percentage of patients with infectious fever who are relevant bacteria in blood cultures. Both investigators were identified as having an infection) and specificity (the proportion blinded to each other and blinded to the procalcitonin of negatives that are correctly identified, i.e., the percentage of values. Patients were classified as having (1) a fever of in- patients with a noninfectious fever who are identified as not fectious origin or (2) a fever of noninfectious origin. In case having an infection) of procalcitonin at different cutoff points30.
of disagreement between the two specialists, a consensus by The area under the receiver-operating characteristic curve was a third independent physician was reached, and the most the overall performance measure of the accuracy of the labo- likely (‘‘best guess'') diagnosis was presumed. On the basis ratory parameter to distinguish patients with infectious fevers of this classification, empirical or targeted (if the pathogen from those with noninfectious fevers. A p value of <0.05 (for a was isolated) antimicrobial therapy was prescribed for pa- two-sided test) was considered significant.
tients with a fever of infectious origin.
Source of Funding Statistical Methods This study was supported by the Swiss National Science Founda- To evaluate differences between groups, the unpaired Student tion (#3200B0-112547/1), Stanley Thomas Johnson Foundation, t test or the Mann-Whitney U test for continuous variables and and Gebert R¨uf Stiftung.
the chi-square or Fisher exact test for categorical variables wasused, as appropriate. If the C-reactive protein or procalcitonin level was not detectable, a value equal to the detection limit forthe respective assay was assigned. To compare the diagnostic Atotal of 103 patients who were febrile, with a median age of seventy-seven years (range, nineteen to ninety-seven value of individual laboratory markers to diagnose infectious years; interquartile range, sixty-one to eighty-five years), were and noninfectious fever, we performed a logistic regression included in this study; forty-five (44%) were men. A local or TABLE I Characteristics of Study Patients Noninfectious Fever (N = 58) Infectious Fever (N = 45) Median age (interquartile range) (yr) Median length of hospital stay (interquartile range) (days) Insertion of dynamic hip screw Knee or foot arthroplasty Upper extremity surgery Median tympanic temperatureafter onset of fever(interquartile range) (C) Median time after surgery until onset of fever(interquartile range) (days) *The values are given as the number of patients, with the percentage in parentheses.
TABLE II Laboratory Parameters in Patients with Postoperative Fever of Infectious and Noninfectious Origin Characteristic Curve† White blood-cellcount (·109/L) C-reactive proteinlevel (mg/L) Procalcitonin (ng/mL) *The values are given as the median with the interquartile range in parentheses. †The values are given as the mean with the 95% confidenceinterval in parentheses.
systemic infection was diagnosed in forty-five patients (44%), In the forty-five patients with an infectious cause of the and these febrile episodes were classified as infectious, whereas fever, forty-four (98%) received antibiotic therapy including no infection could be found in fifty-eight patients (56%) and amoxicillin-clavulanic acid (eighteen patients), a cephalosporin these febrile episodes were classified as noninfectious (Table I).
(nine), a quinolone (eleven), sulfamethoxazole-trimethoprim The most common surgical procedures were insertion of a (three), or broad-spectrum antibiotics (three). In the fifty- dynamic hip screw (thirty-four patients) and hip arthroplasty eight patients with a noninfectious fever, seven (12%) received (twenty-eight), followed by knee or foot surgery (sixteen) and antibiotics including sulfamethoxazole-trimethoprim (three), spine surgery (twelve).
a cephalosporin (two), a quinolone (one), or amoxicillin- The median tympanic membrane temperature at the clavulanic acid (one).
onset of fever was 38.4C in both groups, and the course of Table II shows the laboratory parameters in patients with fever during the following three days did not differ between postoperative fever of infectious origin and those with a fever of groups. Fever tended to occur earlier with regard to the sur- noninfectious origin. Overall, at the onset of fever (day 0), the gical procedure in patients without infection compared with patients showed normal median white blood-cell count values patients with infection (median postoperative day 2 compared (8.9 · 109/L) but increased median concentrations of C-reactive with day 3, p = 0.06).
protein (147 mg/L) and procalcitonin (0.28 ng/mL). The areas Patients with infection had involvement of the lung under the receiver-operating characteristic curve for the diag- (eighteen), urinary tract (eighteen), prosthetic joint (four), nosis of an underlying infection on days 0, 1, and 3 were 0.62, surgical site (two), bloodstream (two), or soft tissues (one).
0.57, and 0.53, respectively, on the basis of the white blood-cell Causative microorganisms were found in the blood in four count and 0.56, 0.59, and 0.56, respectively, on the basis of patients (Staphylococcus aureus [two patients], Escherichia coli the C-reactive protein level. Procalcitonin level showed the [one], and Streptococcus pneumoniae [one]), in intraoperative highest diagnostic accuracy, with an area under the receiver- specimens in three patients (Staphylococcus aureus [three pa- operating characteristic curve of 0.62, 0.62, and 0.71 on days tients]), and in urine in seventeen patients (Escherichia coli 0, 1, and 3.
[sixteen], Klebsiella species [two]; both strains were found in As demonstrated in Figures 1-A, 1-B, and 1-C, on all one patient). In twenty-one patients, no microorganisms were three postoperative days, the white blood-cell count and C- identified. In most patients with noninfectious fever, a definite reactive protein level were similar for both groups, with ex- cause of the fever could not be established. However, presumed ception of the white blood-cell count on day 0, which was causes were postoperative fever due to a prolonged and com- higher in patients with an infectious fever. In contrast, pro- plicated operative procedure (seven patients), adverse drug calcitonin values were consistently higher in patients with in- reaction (four), gout or underlying rheumatologic disease fection compared with those without infection on day 0 (p = (four), and fever due to resorption of a large hematoma (three).
0.04), day 1 (p = 0.07), and day 3 (p = 0.003).

White blood-cell count (Fig. 1-A), C-reactive protein level (Fig. 1-B), and procalcitonin values (Fig. 1-C) in patients with noninfectious fever (left side, light gray) and infectious fever (right side, dark gray) on day 0, day 1, and day 3 after fever onset. Squares denote median values. Boxes represent 25th and 75th percentiles, with the horizontal line denoting the median. Circles represent outliers, and whiskers indicate the range not including the outliers.

Table III shows sensitivity, specificity, and positive and lying infection on all postoperative days, with odds ratios of 2.1 negative likelihood ratios of procalcitonin values at different (95% confidence interval, 1.1 to 4.1) on day 0, 2.2 (95% con- cutoff values on days 0, 1, and 3, respectively.
fidence interval, 1.0 to 4.75) on day 1, and 3.4 (95% confidence In a univariate logistic regression analysis, only pro- interval, 1.3 to 9.1) on day 3. When the postoperative time point calcitonin values, but not C-reactive protein level and white of fever onset was added in a multivariate regression model, blood-cell count values, were significant predictors for under- procalcitonin concentrations were still independent predictors TABLE III Sensitivity, Specificity, and Positive and Negative Likelihood Ratio of Procalcitonin on Days 0, 1, and 3 After the Onset of Fever with Use of Different Cutoff Values

Regression fit of procalcitonin concentrations and postoperative day in patients with noninfectious fever (Fig. 2-A) and infectious fever (Fig. 2-B).
for underlying infection, with odds ratios of 2.3 (95% confi- different postoperative days rather than absolute values had a dence interval, 1.1 to 4.4; p = 0.02), 2.3 (95% confidence in- better diagnostic potential to differentiate infectious from terval, 1.1 to 5.2; p = 0.04), and 3.3 (95% confidence interval, noninfectious fever. The relative decrease of procalcitonin in 1.2 to 9.0; p = 0.02) on days 0, 1, and 3, respectively. Again, patients with an infectious fever was significantly more pro- C-reactive protein levels and white blood-cell counts did not nounced compared with that in patients with a noninfectious significantly correlate in the multivariate analysis with under- fever, but the overall diagnostic accuracy of relative procalci- lying infection (data not shown). Additionally, we investigated tonin changes as assessed in receiver-operating characteristic whether relative changes in the concentration of procalcitonin, analysis was not as high compared with absolute values. For C-reactive protein level, and white blood-cell count among the C-reactive protein level and white blood-cell count, no sig- TABLE IV Procalcitonin Values According to Underlying Infections in Forty-five Patients with Fever of Infectious Origin Median No. of Days Median Procalcitonin Value (Interquartile Range) (ng/mL) (Interquartile Range) Respiratory tract *Includes joint infection (four patients), surgical site infection (two), bloodstream infection (two), and cellulitis (one).
nificant difference was detected between groups (data not tion; therefore, they are of limited clinical utility. A fracture itself and the inflammatory reaction caused by the fracture surgery Table IV shows procalcitonin values for various types of may stimulate the production of cytokines, leading to a non- infection, namely respiratory tract infections, urinary tract specific increase of these commonly used markers of inflam- infections, and other infections including joint infection (four mation. Thus, there is an unmet need for specific markers of patients), wound infection (two), primary bloodstream in- infection after surgical procedures. Only one small study has fection (two), and cellulitis (one). Procalcitonin values at the addressed the procalcitonin kinetics in twenty-one patients after fever onset were lower in patients with respiratory and urinary surgery for a peritrochanteric hip fracture31. The authors con- tract infections than in patients with other infections, partic- cluded that procalcitonin was less affected by the orthopaedic ularly in the four patients with positive blood cultures (the procedure and thus was superior to other infection parameters.
mean concentration of procalcitonin was 1.8, 0.91, and 1.37 on Our study confirms in a larger cohort of patients after various days 0, 1, and 3, respectively).
orthopaedic procedures that procalcitonin distinguishes infec- Procalcitonin concentrations correlated negatively with tious from noninfectious causes of fever more reliably than the onset of fever after surgery in patients with noninfectious C-reactive protein level and white blood-cell count.
fever (correlation coefficient R2 = 0.82; p < 0.0001), whereas no The exact mechanisms underlying procalcitonin in- correlation was found for patients with an infectious fever duction during or after surgery are unknown. Infection and (correlation coefficient R2 = 0.03; p = 0.86) (Figs. 2-A and 2-B).
bacterial endotoxins are stimuli for the induction of pro-calcitonin32. Endotoxin liberation or bacterial translocation within the intestine to various degrees has been reported after This study demonstrates that a single serum procalcitonin differenttypes of surgery33. However, invitroand invivodata level has moderate diagnostic accuracy in predicting un- have shown that a number of other stimuli may also induce derlying infection in patients with a new onset of fever during procalcitonin by promoting different proinflammatory cyto- the early period after orthopaedic surgery. The course of kines. On a transcriptional level, a stimulatory effect on mes- procalcitonin levels is different in a fever of infectious origin senger RNA (mRNA) production of procalcitonin has been compared with fever of noninfectious origin, and thus diag- reported for different proinflammatory cytokines including nostic. Commonly used blood markers such as white blood- tumor necrosis factor (TNF)-a, interleukin (IL)-6, and IL-1b, cell count and C-reactive protein levels were similar for all which may be upregulated as a result of fractures, tissue patients and were not at all helpful in discriminating between damage, and the surgical procedure5,32,34,35. Thus, there is a true infection and the nonspecific systemic inflammatory re- broad range of possible stimuli that might contribute to pro- sponse due to surgical stress and/or underlying trauma.
calcitonin induction after orthopaedic surgery. The nonspe- Infections in the postoperative course after orthopaedic cific induction of procalcitonin production by trauma or tissue surgery can lead to prolonged hospitalization, increased injury, however, seems to be lower compared with a specific morbidity and mortality, and high costs1-3. Timely adminis- induction by bacterial infections. The return of procalcitonin tration of adequate antibiotic therapy is an important factor to levels to normal within a few days after an uncomplicated reduce morbidity and mortality in patients with postoperative postoperative course can be explained by the physiological infections, and thus a thorough clinical examination and di- half-life of procalcitonin of eighteen to twenty-four hours in agnostic workup is mandatory. In patients with a new onset of the absence of further inducing stimuli for procalcitonin fever after an orthopaedic procedure, various laboratory pa- production35. Since the present study was limited to the eval- rameters are frequently used in the routine setting to differen- uation of the clinical usefulness of procalcitonin, we did not tiate infectious from noninfectious causes. However, parameters evaluate other cytokines that could help to better understand such as C-reactive protein level and white blood-cell count may these mechanisms.
be misleading since they are increased in all patients in the We used receiver-operating characteristic curves evalu- postoperative period and are not specific for underlying infec- ating the sensitivities and specificities at any given procalcito- nin cutoff point to compare the accuracy of procalcitonin to With an observational design and a moderate sample size, diagnose infection in patients with postoperative febrile epi- this pilot study has limitations and requires validation in a larger sodes. The findings of this study are consistent with reports population. As a consequence of the limited power of this study, from other surgical settings showing a nonspecific procalci- the confidence intervals of our receiver-operating characteristic tonin increase in patients postoperatively, depending on the analysis overlap considerably. Therefore, we refrained from extent of surgical stress and inflammation18-22. Thus, it is im- performing additional subgroup analyses on the various surgical portant to recognize that so-called normal procalcitonin values procedures or types of infections. Importantly, because of in patients after surgery are not within the normal range for multiple comparisons performed in the analysis of this study, healthy subjects, and cutoff ranges must be adapted accord- p values in the range of 0.01 to 0.05 should be viewed cautiously, ingly. At a cutoff of 0.1 ng/mL, procalcitonin had a reasonable given the attendant higher risk of a chance association. How- sensitivity of between 85% and 91% to exclude an infection ever, the clinical and diagnostic workup and the blood sampling (Table III). Conversely, in patients with procalcitonin con- of patients were standardized, and the investigators were blin- centrations above the cutoff of 0.5 ng/mL, the likelihood for ded with respect to procalcitonin results. We used the clinical underlying bacterial infection was high, especially when this evaluation of the patients based on a comprehensive diagnostic increase persisted for more than two days. In this context, the and microbiological workup and the assessment by an infec- course of procalcitonin values over time rather than a single tious diseases consultant as our diagnostic ‘ gold standard,' value should be considered to diagnose an infectious etiology.
which remains at present the best available method to establish Our study suggests that, in patients who are febrile with pro- the presence or absence of an infectious disease37,38. Only in- calcitonin values of <0.1 ng/mL, antibiotics can be initially terventional studies, in which antibiotic therapy is guided on the withheld if no obvious clinical focus of infection is present and basis of predefined procalcitonin cutoff ranges, have the po- the patient is in good general health. However, these patients tential to resolve this dilemma11. In this context, our results should be reassessed the next day with a thorough clinical regarding the procalcitonin kinetics in patients without an in- examination and repeat blood analysis. In contrast, in patients fection and the sensitivity and specificity of procalcitonin at with procalcitonin values of >0.5 ng/mL, a rapid initiation of various cutoff values and on different postoperative days are of antibiotics may be warranted. The use of an algorithm for importance to the rational design of potential future interven- patients with respiratory tract infections, in which the initia- tional studies. These studies should not only address the safety tion or continuation of antibiotics was discouraged if pro- and efficacy of serum procalcitonin levels for antibiotic stew- calcitonin was <0.1 ng/mL or £0.25 ng/mL, respectively, and ardship but also should establish cost-effectiveness by consid- was encouraged if procalcitonin was >0.5 ng/mL or >0.25 ng/mL, ering the costs of procalcitonin measurement (US$10 to $30 per has been shown to markedly reduce the initiation and duration sample) and the potential savings in the consumption of other of antibiotic therapy 6-9,27.
health-care resources. Additionally, it would be interesting to In accordance with other studies, C-reactive protein investigate the time course of other cytokines and biomarkers concentrations were increased about twentyfold to fortyfold in (e.g., IL-6) following other surgical procedures.
all patients after surgery, and levels remained high throughout Infection is far too complex a process to be reliably di- the three-day observation period18-22. In a previous study, agnosed by means of a specific cutoff value for any single bio- procalcitonin kinetics were studied in a cohort of patients marker, particularly in patients who have recently undergone undergoing elective cardiac surgery18. The investigators re- operative procedures. However, this study demonstrates that the ported that, in the presence of fever, procalcitonin was a reli- likelihood for a bacterial infection in patients presenting with able marker for infection and more relevant than C-reactive fever in the postoperative course increases gradually with in- protein for the diagnosis of postoperative infection. In the creasing serum levels of procalcitonin. As an adjunct to clinical postoperative course following major neurosurgery, procalci- and microbiological parameters, serum procalcitonin levels may tonin levels, in contrast to C-reactive protein level and white serve as a diagnostic surrogate marker for helping to differen- blood-cell count, showed a less pronounced nonspecific in- tiate infectious from noninfectious causes of early postoperative crease21. Similarly, in patients who underwent major cancer fever after orthopaedic procedures. Early identification of surgery, elevated procalcitonin and IL-6 levels were reported to patients with postoperative infections is of great importance be early markers of postoperative sepsis when associated with in order to establish effective antibiotic therapy. Conversely, systemic inflammatory response syndrome, whereas C-reactive procalcitonin may help us to avoid unnecessary antimicrobial protein levels were not36. The routine use of C-reactive protein treatment in patients with noninfectious causes of fever. n to diagnose infection in patients presenting with fever after NOTE: The authors thank the nursing staff and the surgeons of the Departments of Orthopaedic surgery is motivated by its low cost and easy availability and by Surgery and Traumatology for their continuous support during the study.
historical practice rather than on the basis of the evidence.
However, the reliability of C-reactive protein is hampered by aprotracted response with late peak levels and a low specificityin patients with systemic inflammatory response syndrome, Sabina Hunziker, MD whereas procalcitonin is more specific for distinguishing in- Thomas H¨ugle, MD fectious from noninfectious febrile episodes.
Katrin Schuchardt, MD Isabelle Groeschl, MD University Hospital Basel, Philipp Schuetz, MD Spitalstrasse 21, CH-4031 Basel, Switzerland Urs Eriksson, MDDepartment of Internal Medicine, Andrej Trampuz, MD University Hospital Basel, Infectious Diseases Service, Department of Medicine, University Hospital and University of Lausanne, Rue du Bugnon 46,CH-1011 Lausanne, Department of Orthopaedic Surgery, E-mail address: [email protected] 1. Garibaldi RA, Brodine S, Matsumiya S, Coleman M. Evidence for the non-infectious neutropenic patients with hematological malignancies: procalcitonin and IL-6 are more etiology of early postoperative fever. Infect Control. 1985;6:273-7.
reliable than C-reactive protein. Eur J Clin Microbiol Infect Dis. 2004;23:539-44.
2. Than P, Malovics I. [Significance of postoperative fever after hip prosthesis im- 17. Haimi-Cohen Y, Vellozzi EM, Rubin LG. Initial concentration of Staphylococcus plantation]. Z Orthop Ihre Grenzgeb. 2000;138:430-5. German.
epidermidis in simulated pediatric blood cultures correlates with time to positiveresults with the automated, continuously monitored BACTEC blood culture system.
3. Gaynes RP, Culver DH, Horan TC, Edwards JR, Richards C, Tolson JS. Surgical J Clin Microbiol. 2002;40:898-901.
site infection (SSI) rates in the United States, 1992-1998: the National NosocomialInfections Surveillance System basic SSI risk index. Clin Infect Dis. 2001;33 Suppl 18. Aouifi A, Piriou V, Bastien O, Blanc P, Bouvier H, Evans R, C´ elard M, Vandenesch F, Rousson R, Lehot JJ. Usefulness of procalcitonin for diagnosis of infection incardiac surgical patients. Crit Care Med. 2000;28:3171-6.
4. Laffey JG, Boylan JF, Cheng DC. The systemic inflammatory response to car-diac surgery: implications for the anesthesiologist. Anesthesiology. 2002;97: 19. Carboni GL, Fahrner R, Gazdhar A, Printzen G, Schmid RA, Hoksch B. Com- parison of procalcitonin and CrP in the postoperative course after lung decortication.
Eur J Cardiothorac Surg. 2008;33:777-80.
5. Gurlich R, Maruna P, Cermak J. [Use of procalcitonin in surgery]. Rozhl Chir.
1999;78:292-4. Czech.
20. Dorge H, Schondube FA, Dorge P, Seipelt R, Voss M, Messmer BJ. Procalcitoninis a valuable prognostic marker in cardiac surgery but not specific for infection.
6. Stolz D, Christ-Crain M, Bingisser R, Leuppi J, Miedinger D, M¨ uller C, Huber P, Thorac Cardiovasc Surg. 2003;51:322-6.
uller B, Tamm M. Antibiotic treatment of exacerbations of COPD: a randomized, controlled trial comparing procalcitonin-guidance with standard therapy. Chest.
21. Laifer G, Wasner M, Sendi P, Graber P, Gratzl O, Huber P, Fluckiger U, Zimmerli W. Dynamics of serum procalcitonin in patients after major neurosurgery. Clin Mi-crobiol Infect. 2005;11:679-81.
7. Schuetz P, Christ-Crain M, Wolbers M, Schild U, Thomann R, Falconnier C,Widmer I, Neidert S, Blum CA, Sch¨ onenberger R, Henzen C, Bregenzer T, Hoess C, 22. Oberhofer D, Rumenjak V, Lazic J, Vucic N. [Inflammatory indicators in pa- Krause M, Bucher HC, Zimmerli W, M¨ uller B; ProHOSP Study Group. Procalcitonin tients after surgery of the large intestine]. Acta Med Croatica. 2006;60:429-33.
guided antibiotic therapy and hospitalization in patients with lower respiratory tract infections: a prospective, multicenter, randomized controlled trial. BMC Health Serv 23. Sponholz C, Sakr Y, Reinhart K, Brunkhorst F. Diagnostic value and prognostic Res. 2007;7:102.
implications of serum procalcitonin after cardiac surgery: a systematic review of the 8. Christ-Crain M, Stolz D, Bingisser R, M¨ uller C, Miedinger D, Huber PR, Zimmerli literature. Crit Care. 2006;10:R145.
W, Harbarth S, Tamm M, M¨ uller B. Procalcitonin guidance of antibiotic therapy in 24. Meisner M, Tschaikowsky K, Hutzler A, Schick C, Schuttler J. Postoperative community-acquired pneumonia: a randomized trial. Am J Respir Crit Care Med.
plasma concentrations of procalcitonin after different types of surgery. 1998; 9. Christ-Crain M, Jaccard-Stolz D, Bingisser R, Gencay MM, Huber PR, Tamm M, 25. Luyt CE, Gu´ erin V, Combes A, Trouillet JL, Ayed SB, Bernard M, Gibert C, uller B. Effect of procalcitonin-guided treatment on antibiotic use and outcome in Chastre J. Procalcitonin kinetics as a prognostic marker of ventilator-associated lower respiratory tract infections: cluster-randomised, single-blinded intervention pneumonia. Am J Respir Crit Care Med. 2005;171:48-53.
trial. Lancet. 2004; 363:600-7.
26. Novotny A, Emmanuel K, Matevossian E, Kriner M, Ulm K, Bartels H, Holzmann 10. Briel M, Schuetz P, Mueller B, Young J, Schild U, Nusbaumer C, P´ B, Weighardt H, Siewert JR. Use of procalcitonin for early prediction of lethal out- HC, Christ-Crain M. Procalcitonin-guided antibiotic use vs a standard approach come of postoperative sepsis. Am J Surg. 2007;194:35-9.
for acute respiratory tract infections in primary care. Arch Intern Med. 2008;168:2000-8.
27. 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, 11. Muller B, Schuetz P, Trampuz A. Circulating biomarkers as surrogates for Bregenzer T, Hoess C, Krause M, Bucher HC, Zimmerli W, Mueller B; ProHOSP Study bloodstream infections. Int J Antimicrob Agents. 2007;30 Suppl 1:S16-23.
Group. Effect of procalcitonin-based guidelines vs standard guidelines on antibiotic 12. Christ-Crain M, Muller B. Procalcitonin in bacterial infections—hype, hope, use in lower respiratory tract infections: the ProHOSP randomized controlled trial.
more or less? Swiss Med Wkly. 2005;135:451-60.
13. Schuetz P, Christ-Crain M, Muller B. Biomarkers to improve diagnostic and 28. Thorpe TC, Wilson ML, Turner JE, DiGuiseppi JL, Willert M, Mirrett S, Reller LB.
prognostic accuracy in systemic infections. Curr Opin Crit Care. 2007;13:578- BacT/Alert: an automated colorimetric microbial detection system. J Clin Microbiol.
14. Simon L, Gauvin F, Amre DK, Saint-Louis P, Lacroix J. Serum procalcitonin and 29. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for C-reactive protein levels as markers of bacterial infection: a systematic review and nosocomial infections, 1988. Am J Infect Control. 1988;16:128-40. Erratum in: Am meta-analysis. Clin Infect Dis. 2004;39:206-17. Erratum in: Clin Infect Dis. 2005; J Infect Control. 1988;16:177.
30. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two 15. de Bont ES, Vellenga E, Swaanenburg J, Kamps W. Procalcitonin: a diagnostic or more correlated receiver operating characteristic curves: a nonparametric ap- marker of bacterial infection in neutropenic cancer patients with fever? Infection.
proach. Biometrics. 1988;44:837-45.
31. Yasmin D, Bulut G, Yildiz M. [Can procalcitonin be used for the diagnosis and 16. von Lilienfeld-Toal M, Dietrich MP, Glasmacher A, Lehmann L, Breig P, Hahn C, follow-up of postoperative complications after fracture surgery?]. Acta Orthop Schmidt-Wolf IG, Marklein G, Schroeder S, Stuber F. Markers of bacteremia in febrile Traumatol Turc. 2006;40:15-21. Turkish.
32. Muller B, White JC, Nylen ES, Snider RH, Becker KL, Habener JF. Ubiquitous 36. Mokart D, Merlin M, Sannini A, Brun JP, Delpero JR, Houvenaeghel G, Moutardier expression of the calcitonin-I gene in multiple tissues in response to sepsis. J Clin V, Blache JL. Procalcitonin, interleukin 6 and systemic inflammatory response Endocrinol Metab. 2001;86:396-404.
syndrome (SIRS): early markers of postoperative sepsis after major surgery. BrJ Anaesth. 2005;94:767-73.
33. MacFie J, O'Boyle C, Mitchell CJ, Buckley PM, Johnstone D, Sudworth P.
Gut origin of sepsis: a prospective study investigating associations between 37. Bachmann LM, Juni P, Reichenbach S, Ziswiler HR, Kessels AG, Vogelin E.
bacterial translocation, gastric microflora, and septic morbidity. Gut.
Consequences of different diagnostic ‘‘gold standards'' in test accuracy research: carpal tunnel syndrome as an example. Int J Epidemiol. 2005;34:953-5.
34. Meisner M. Pathobiochemistry and clinical use of procalcitonin. Clin Chim Acta.
38. Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig LM, Moher D, Rennie D, de Vet HC, Lijmer JG; Standards for Reporting of Diagnostic Accuracy.
35. Maruna P, Nedelnikova K, Gurlich R. Physiology and genetics of procalcitonin.
The STARD statement for reporting studies of diagnostic accuracy: explanation and Physiol Res. 2000;49 Suppl 1:S57-61.
elaboration. Ann Intern Med. 2003;138:W1-12.

Source: http://www.ortoped.fmed.edu.uy/Documentos/JBJS_2010_Thevalueofserumprocalcitonin.pdf