- Email: [email protected]
Procalcitonin: a new marker for diagnosis of acute rejection and bacterial infection in patients after heart and lung transplantation
Transplant Immunology 1998; 6: 235-241
Procalcitonin: a new marker for diagnosis of acute rejection and bacterial infection in patients after heart and lung transplantation S Hammera, F Meisner: P Dirschedlb, G Habela, P Fraunbergerc, B Meiserd, B Reichardt and C Hammera “Institute of Surgical Research, bIBE, cInstitute of Clinical Chemistry, dDepartment of Cardiothoracic Surgery, Klinikum Groflhadern, LM- University,Munich Received 6 October 1998; revised manuscript accepted 15 October 1998
Abstract: The aim of the study was to investigate the reliability of procalcitonin (PCT), a new potential marker for detection of bacterial, fungal and protoxoal infections, in order to differentiate these from viral infections and early rejections in heart, heart-lung and lung transplanted patients. PCI is a propeptide of calcitonin with unknown origin which is not detectable in plasma of healthy subjects. It increases rapidly and signitlcantly under severe microbial infections. Met/w& PCI plasma levels were measured using an immuno-luminescence assay. C-reactive protein and white blood cells were quantified to validate the PCT values. Results: Increased levels of PCT were found in all transplant patients with bacterial, timgal and protozeal infections. The magnitude of the values were clearly associated with the severity of the infection. Trauma of operation or inflammatoty events such as viral infections and rejections did not trigger PCT-production. The release of PCT did not depend on the type of pathogens even though Aspergillum resulted in the highest levels measured. Sensitivity, specificity and prognostic value of PCT for systemic infections were higher than of the other parameters investigated. Conclusion: PCT is a highly specific analyte which shows significant diagnostic validities when nonviral infections are compared with rejection episodes. PCI discriminates between inflammatory events such as rejection or viral infections and nonviral-infections including bacterial, fungal and protozoa1 infections. The half-life of PCI is 24 h indicating clearly a competent antibiotic treatment. Unnecessary antibiotic therapy can be avoided due to the early exclusion of bacterial and fungal infections.
Introduction Despite improvements in short-term outcome after heart and lung transplantation, patients are lost due to various complications in the early postoperative phase. The two major postoperative incidents are acute rejections (AR) and
Address for correspondence: C Hammer, Institute of Surgical Klinikum Grobhadem, LM-University, Munich, Research, Marchioninistr. 15, Munich, Germany. E-mail: 83177 [email protected] 0 Amold
1998
infections. Many parameters are currently used for differential diagnosis. Most mediators of inflammatory events such as cytokines, interleukins, neopterin and C-reactive protein (CRP) indicate, however, not only immunological reactions
during acute rejection, but also during viral infections, vascular occlusive diseases, malignancies, trauma and of course bacterial, fungal and protozoal infections. Recipients of solid organ transplants are at special risk for infections due to the heavy immunosuppression. ‘Ransplant recipients have an increased susceptibility to all types of infectious material, especially opportunistic organisms. 0966_3274(98)TI242OA
236
S Hammer et al.
Approximately two-thirds of transplant recipients experience at least one episode of infection during the first year after transplantation. Between 10% and 35% of heart and lung transplant patients suffer from infections in the first postoperative weeks. The largest proportion of germs are bacteria with 47-55%. Fungi and protozoa are responsible for 12-74% of infections mainly in the lung. Infections are a major cause (12%) of death in the post-transplant period. The interaction between the immunosuppressive therapy and the environmental exposure determines the risk of infection. Procalcitonin (PO, a species-specific propeptide of calcitonin, is a glycoprotein of 116 amino acids with a molecular weight of 13 kD. Its origin and function are unknown. Gene structure and gene-locus are described by Le Moullec et al.’ PCT was never induced in vitro neither by allogeneic nor by lectin or endotoxin stimulation of peripheral blood cell populations.’ PCT is a species-specific molecule therefore data are only available from humans.3 Investigations performed with synthetic PCT have to be regarded with care, since it is unknown whether synthetic or biological materials have a different tertiary structure or even different post-translational modifications.4 In in vivo studies PCT was found after endotoxin injections in healthy volunteers. Subcutaneous injection of GM-CSF increased the numbers of white blood cells and levels of interleukins but had no effect on PCT (our own observation). PCT selectively responds to severe bacterial , parasitic and fungal infections with systemic manifestation. Thus PCT promises to be a new marker with cytokine character which might develop to a specific indicator of nonviral infections in transplant patients.
discharge (control); group B, local infection; group C, generalized/multiple infection; group D, sepsis. Infections were assessed on a daily basis by positive serology, IgG or IgM antibodies, positive smear or positive blood cultures. Infections were regarded as cured after 7 days of effective antibiotic therapy. Included were those bacterial infections which were treated specifically according to the bacterial spectrum. Local infections were determined as peripheral colonization, infection of mucous membranes or locally limited, insignificant infections. Multiple or generalized infections were defined when different or the same germs were found in one or more organs. As sepsis we defined infections with positive blood cultures and severe clinical symptoms. Multiple, generalized infections and sepsis were comprised under systemic infections. Onset of rejection episodes was designated by histopathological diagnosis from endomyocardial or open lung biopsies and graded according to the criteria of the International Society for Heart and Lung ‘Bansplantation (ISHLT). Start of AR was specified as positive finding at the day of biopsy. A 3-day steroid bolus therapy determined the duration of rejection. StatIetIcaI evaluation For statistical evaluation only values of the day of diagnosis of rejection or infection were included. The results were expressed as mean values with standard error of the mean. The significance of the differences was assessed using the Mann-Whitney U-test. Proportions were compared applying the chi-square test with Yate’s correction. Probabilities of less than 0.05 were accepted as significant. Meaeurement of procalcitonin
Objectives The aim of the study was to elaborate specificity and sensitivity of PCT, to compare this molecule with other markers and to test its capability and reliability to differentiate between rejections and infections due to bacteria and fungi in patients after transplantation of thoracic parenchymal organs.
Material and methods PCT was studied retrospectively in 78 heart and lung transplanted patients, 58 males and 20 females with ages ranging from 1 to 68 years (mean 52 years). Indications for heart transplantation (n = 57) were dilatative cardiomyopathy (n = 35) ischaemic cardiomyopathy (n = 11) and others (n = 22). Lung transplantation, unilateral (n = 11) and bilateral (n = 7) was performed in 18 patients. Seven patients were transplanted due to fibrosis of the lung, four due to emphysema. Five patients suffered from cystic fibrosis and cystic lung disease. Because of chronic rejection one patient was retransplanted. Three heart-lung transplantations were performed in patients suffering from Eisenmenger’s disease (n = 2) and thrombosis of the pulmonary artery. The patients were divided into five groups to differentiate between acute rejections and infections. Group 1: no acute rejection (AR), no infection (control), group 2: AR, group 3: AR and infection, group 4: local infections, group 5: systemic infections (generalized/multiple infections and sepsis); and into four groups to ascertain the severity of infection: Group A, Zkansplant Immunology 1998; 6: 235-241
Blood was collected daily at 8 am. The separated EDTA-plasma was stored at -80°C if not used immediately. PCT was measured by the commercial specific and ultrasensitive immunoluminometric assay. This assay uses two monoclonal antibodies, one directed against residues around 96-106 of PCT as capture antibody and one directed against the residues 70-76 as tracer antibody. The sequence 70-76 is part of the calcitonin molecule, whereas the 96-106 sequence is part of the katacaltin molecule. Synthetic PCT was used as standard. This assay which is specific for PCT has a limit of detection of less than 1 pg/ml. Interand intra-assay variations at both low and high concentrations were less than 7%. In short: 20 pl of plasma are pipetted into the ‘coated tube system’. The 100 tubes (2 x 50) sitting in a holder are coated with a surplus of antigen-specific monoclonal anti-katacalzin antibody. A luminescence labelled antibody directed against the calcitonin part of PCT (250 pl) is now added. The tubes are incubated and agitated for 2 h at room temperature. The excess of the tracer is discarded and the tubes are washed five times. In a luminometer the remaining luminescence signal of the tracer bound to the tube walls is measured using the LUMItest kit reagents (BRAHMS Diagnostica, Berlin, Germany) as a complete diagnostic kit (LUMItest Procalcitonin) and a Luminometer (Behring Diagnostica, Marburg, Germany). Magnitude of the luminescence signal is proportional to the PCI concentration of the specimen.
Procalcitonin: a new marker for diagnosis of acute rejection and bacterial infection
Results PCT, CRP and white blood cells (WEK) were measured in order to differentiate between infections and rejections (Figures 14). The mean value of PCT of 59 patients tested showing neither rejection nor infection was 0.3 + 0.3 ng/ml and was assigned as control. A group of 19 transplanted patients suffering from rejection episodes of all degrees had simihu values: the mean 0.2 f 0.2 ng/rnl. Slightly increased levels were found when infections accompanied rejections (mean 0.6 + 0.6 rig/ml). Most of these infections were of local character. Local infections only resulted again in lower levels with a mean of 0.6 f 0.3 q/ml. In contrast systemic infections induced high values of 10.5 + 18.2 q/ml. There was no difference between PCT values of patients with rejection or no rejection. The differences between rejection and local infection were sign&ant @ < 0.01). PCT of patients with systemic infection was exceedingly higher than that of patients with rejection (p c 0.0001) (Figure 1).
At a cut-off point of ~0.8 ng/ml PCT sensitivity for rejection was lOO%, specificity 82% and predictive value 89%. The alpha error wasp < 0.0001. At a cut-off point for systemic infections of >0.5 pg/ml all three parameters reached 89%. CRP was found to be elevated in all groups. Even transplant patients without any signs of inflammation had a mean of 3.8 + 5.1 mg/dl which exceeds the normal value of less than 0.5 mg/dl. As soon as nonviral infections occurred the values increased to a level of 6.8 f 5.4 mgJd1 independently of whether the infections were local or systemic (Figure 2). Of absolutely no differential value were the numbers of peripheral WBC. In all groups the mean number of leucocytes was elevated and ranged between 12.0 and 16.0 x 10’ cells/I (Figure 3). PCT, CRP and WRC were also measured in order to ascertain the severity of infection (Figures 4-6). As a control group we used values of the day of discharge from the hospital. In this group PCI had a mean value of 0.3 + 0.5 ng/ml. Local infections resulted in
, r
100 I I
237
,
*.
* ’
.
ns ’
l
. I
10 .:
.
~___________________i_ no AR
AR
AR+ infect
local infect
syst.infect
Figure 1 PCT values in patients with and without rejections or infections; ns = not significant, the asterisk represents significancep < 0.01. Normal values < 0.5 nghnl (line).
100 I
.. :
0.11 Figure 2
no AR
. t
AR
AR+ infect
local infect
syst.infect
CRP values in patients with and without rejections or infections. Normal values < 0.5 mg/dl (line).
Tmnsplant Immurwlog~ 1998; 6: 235-241
238
S Hammer et al.
lOO-
OJ
no AR Figure 3
local infect
AR+ infect
AR
systinfect
WBC values in patients with and without rejections or infections. Normal values 4-10 x lo9 cells/l (lines).
I
I
*
’
* .
I
*
1 - .
II
.A.
I
*’
.
. . .
_____________________________________________________
.
. .
i_____.___.____._._.______~______________________________________. .
:
.
A .
0.1 4
1 discharge
Ngure 4
t
1 local infect
:. genJmult. infect
sepsis
Association of PCT values with severity of infections. The asterisk represents significantp < 0.01. Normal values c 0.5 n&l
0.6 + 0.6 ng/ml while multi or generalized infections showed a further increase to 7.3 +- 17.0 q/ml. The highest levels were reached in the sepsis group with 22.4 f 17.8 @ml. The PCT levels at time of discharge were significantly different to all infection groups with values ofp < 0.01 (local infection),p < 0.0001 (generalized!multi-infection),p < O.CKlOl (sepsis). The fhulings of the group with local infection were significantly different from those of the multiple/generalized infections and from those of sepsis @ < O.OOl,p < 0.0001) (Figure 4). High CRP values were found in all groups. Even at time of discharge 2.8 -C 4.0 mg/dl of CRP were seen (normal values co.5 mg/dl). Local and multiple/generalized infections showed an almost identical mean of 6.8 + 6.2 and 6.4 + 4.4 mg/dl, respectively. Sepsis patients developed values of 8.3 f 8.7 mgidl (Figure 7). WBC were 9.5 + 2.6 x lo9 cells/l in the control group and therefore normal. Local and multiple/ eneralized infections increased the number to 13.8 + 6.2 x 10$ cells/l and 15.4 -C 7.5 x lo9 cells/l, respectively. In the sepsis patient the numbers Transplant Immunology 1998; 6: 235-241
(lure).
mounted to 17.9 -C9.3 g/I. A differential diagnosis with this parameter cannot be achieved (Figure 6). Pathogens such as gram-positive and gram-negative bacteria, Candida, Aspergillum and combined infections were associated with PCI values. The assumption that the release of PCT depends on the type of pathogens could not be verified. Even though Aspergillum in some cases resulted in very high PCT levels (Figure 7).
Discussion PCT is a new and innovative parameter for the diagnosis of severe bacterial and fungal infections. It is a reliable indicator of inflammatory activities during sepsis and multi-organ failure. PCI’ is the prohormone of calcitonin with a molecular weight of 13 kD but lacks hormonal activity.’ Its biological function and organ of origin is still unknown. C-cells of the thyroid gland and other neuroendocrine celIs are supposed to produce PCT
Procalcitonin: a new marker for diagnosis of acute rejection and bacterial infedion
239
-. discharge
local infect
gen./mult. infect
sepsis
Figure 5 Associationof CRP values with severityof infections. Normal values < 0.5 @ml (line).
11 discharge Figure
local infect
gen./mult. infect
sepsis
6 Association of WBC values with severity of infections. Normal values 4-10 x lo9 ceils/l (lines).
Elevated levels were detected in patients suffering from medullaxy thyroid carcinoma, but PCT was also found in thyreodectomized patients.6 Injection of bacterial endotoxin in healthy volunteers led within 24 h to PCT levels exceeding the baseline of
normal levels in transplant patients with viral diseases and also in HIV patients even in terminal disease.” Peripheral colonization, wound healing, fungal stomatitis and encapsulated processes like abscesses could not be shown to induce PCT release.5 Systemic infections in general resulted in high levels of Pa, initially used to dia ose sepsis and septicaemia espe cially in new-born infants. !T These beneficial characteristics allow to differentiate septic shock from cardiogenic shock,13 toxic from biliary pancreatitis14 and infectious from noninfectious ARlXi5 Huber et al. describes a failure of PCT in some immunodeficient patients with severe fungal infection. The explanation given was that chronic diseases and immunosuppression downregulate or exhaust PCT production.16 In our cohorts of transplanted patients chemical immunosuppression including azathioprin, cyclosporin and tacrolimus as well as steroids and steroid bolus theray during rejection episodes had no impact on PCT behaviour. ’ Eberhard et al. observed that murine monoclonal anti-CD3-lymphocytic antibodies (Okt 3), which
MCI S Hammer et al.
100,
. . .
mixed infections
. .
. . . ,
.
10 --
. . A
%
I
:
-
F
t .
.
1 -:
: .
. I .
. *
:
0.1 4 gr~.beict.
grarmegbact.
Candida
Aspergilus
Type of pathogens Figure 7
------I. -
&fungi
n= 16, Gl u=26,G2 n=49,G3 n=15,G4
Different behaviour of PCT in transplant patients (schematic)
induce the ‘interleukin storm’ in transplant patients increase PCT significantly.‘a This specsc unresponsiveness to rejection mechanisms on one hand and conservative immunosuppression on the other hand is a major advantage of PCI’ for the monitoring of transplant patients. Neither acute nor chronic rejections of thoracic and liver grafts were accompanied by PCT increase.1g Thus levels of PCT exceeding values of 0.8 @ml (cut-off point) are concentrations which prove an infection of systemic character with a sensitivity of 82%. Four courses of PCT could be identified in transplanted patients (Figure 8). First, patients who never exceeded normal vahms and were therefore always free of severe nonviral infection (Gl, n = 16). A second group of patients started with normal values to develop high levels of PCT during postoperative treatment due to acquired infections on the ICU. Specific antibiotic treatment safely terminated the infections and discontinued PCT production ideally indicated by the half-life of 1 day (G2, n = 26). A third group with high levels of PCT Tmnplant
bacteria
gratmeghct.
Association of type of pathogens with levels of PCT
PCT
FIgwe 8
grarcpos. 8
Immunology
1998; 6: 235-241
after transplantation developed normal values within a short time due to the appropriate empiric antibiotic treatment (Q, II = 49). The most interesting group was group four which started with significant amounts of PCT which dropped for a few days under antibiotic treatment and then increased again after a plateau period. Four out of five of these patients died from irreversible Aspergilhun infections (G4, II = 15): Another advantage of PCT was found in children after bone marrow transplantation or meningitis. Here PCT allows differentiation whether clinical symptoms including ‘fever of unknown origin’ are due to bacteria or other inflammatory mechanisms. This diagnosis prevented unnecessary myelotoxic antibiotic treatment in case of nonbacterial and nonfungal iIlftXtiOIIS.m
A further diagnostic advantage of PCT as compared to other sophisticated methods is its uncomplicated and quick testing performance. After less than 3 h reliable values are established. Plasma samples stored at -20°C are stable for PCT for at least half a year.
Procalcitonin: a new marker for diagnosis of acute rejection and bacterial infection
IL6 also allows differentiation between systemic infections and rejection,*l elevates during sepsis,5 correlates with the severity of infection and predicts the outcome of the patient.** The disadvantage of IL6 as compared to PCT is its increase during viral infections, trauma, major operations” and other inflammatory events like autoimmune diseases.‘* Its short half-life in vivo and in vitro reduces its relevance for clinical routine application. CRP and leucocytes were both markers which increased during trauma and surgery and showed major overlapping during viral and bacterial infections and rejection episodes.ls The slow increase (24 h) and decrease of CRP in comparison to IL6 (1 h) and PCT (2-4 h) in peripheral blood reduces its clinical, prognostic and therapeutic value. Severity and dynamic of infections is not heralded by CRP and WEE numbers. The minor differences are not sufficient to be used for transplant patients during intensive care.’ Another advantage of PCT for clinical purposes is its ability to predict the outcome of patients and to demonstrate a typical course if the infection is not successfully treated. Beier was able to verify the prognostic properties of PCT233’4 Despite of these convincing characteristics PCT should not be regarded as the only reliable parameter for infections in transplant patients in intensive care units. In some cases peripheral colonization, infected wound healing, fungal stomatitis and encapsulated processes like abscesses were not heralded by PCT? A combination with other markers is still capable of improving the treatment and thus the clinical situation of patients after transplantation.
References Le Moullec JM, Jullienne A, Chenais J et al. The complete sequence of pre-pro-calcitonin. FEBS 1984; 167: 93-97. Staehler M, iiberfuhr P, Reichart B, Hammer C. Differentialdiagnostik der AbstoOungsreaktion und Infektion bei herztransplantierten Patienten: Neue Wege mit Zytokinen und Prokalzitonin als Marker. Transplant Medizin 1997; 9: 44-50. Gendrel D, Bohuon C. Procalcitonin, a marker of bacterial infection. Infection 1997; 25: 133-34 Meisner M. PCT und Score Systeme, 3rd Advisory Board and Investigator Meeting [Abstracts], Berlin, 1998. Karzai W, Oberhoffer M, Meier-Hellmann A, Reinhart K. Procalcitonin - a new indicator of the systemic response to severe infections. Infection 1997; 25: 329-34. Assicot M, Gendrel D, Carsin H, Raymond J, Guilbaud J, Bohuon C. High serum procalcitonin concentrations in patients with sepsis and infection. Lancet 1993; 341: 515-18. Dandona P, Nix D, Wilson MF et al. Procalcitonin increase after endotoxin injection in normal subjects. JClin Endocrin Metaboi 1994; 79: 1606-608. Brunkhorst FM, Forycki ZF, Wagner J. Release and kinetics of procalcitonin (PCT) after gram negative bacterial injection in a healthy subject [Abstract]. Shock 1997; 7 (suppl): 493.
Transplant Immunology
1998; 6: 235-241
241
9 Gerard Y, Hober D, Petitjean Set al. High serum procalcitonin level in a four year old liver transplant recipient with a disseminated candid&is. Znfect 1995; 23: 310-11. 10 Reith HB, Lehmkuhl P, Beier W, Hijgy B. Procalcitonin- ein progInfektionsparameter bei der Peritonitis. Chir nostischer Gustroenteroll995; 11 (suppl2): 47-50 11 Gerard Y, Hober D, Assicot M et al. Procalcitonin as a marker of bacterial sepsis in patients infected with HIV-l. JZnfection 1997; 35: 4146. 12 Gendrel D, Raymond J, Assicot M et al. Procalcitonin as a marker for the early diagnosis of neonatal infection. J Ped 1996; 128: 570-73. 13 De Werra I, Jaccard C, Corradin SB et al. Cytokines, nitrite/ nitrate, soluble tumour necrosis factor receptors, and procalcitonin concentrations: Comparisons in patients with septic shock, cardiogenie shock, and bacterial pneumonia. Crit Care Med 1997; 25: 607-13. 14 Brunkhorst FM, Forycki ZF, Wagner J. Friihe Identitiierung der bililren akuten Pankreatitis durch Procalcitonin-Immunreaktivitlt vorllufige Ergebnisse. C&r Gustroenteroll995,ll (suppl2): 42-46. 15 Brunkhorst FM, Forycki ZF, Wagner J. Discrimination of infectious and non-infectious etiologies of the adult respiratory distress syndrome (ARDS) with procalcitonin immunoreactivity [Abstract]. Clin Zntens Care 1995; 6: 3. 16 Huber W, Schweigart U, Bottermann P. Failure of PCT to indicate severe fungal infection in two immunodeficient patients. Znfection 1997; 25: 377-78. 17 Staehler M, Hammer C, Meiser B, Reichart B. A new marker for differential diagnosis of acute rejection and bacterial infection in heart transplantation. TransplantProc 1997; 29: 584-85. 18 Eberhard OK, Haubitz M, Brunkhorst FM, Kliem V, Koch KM, Brunkhorst R. Usefulness of procalcitonin for differentiation between activity of systemic autoimmune disease and invasive bacterial infection. A& Rheum 1997; 40: 1250-56. 19 Meisner M, Tschaikowsky K, Schmidt J, Schiittler J. Procalcitonin (PCT) - indications for a new diagnostic parameter of severe bacterial infection and sepsis in transplantation, immunosuppression and cardiac assist devices. Cardiovusc Engineer 1996; 1: 67-76. 20 Gendrel D, Raymond J, Assicot M et al. Measurement of procalcitonin levels in children with bacterial and viral meningitis. Clin Infect Dis 1997; 24: 1240-42. 21 Fraunberger P, Pfeiffer M, Haller M et al. Cytokine and cytokinereceptor profiles after liver and heart transplantation. Transplant Proc 1995; 27: 2023-27. 22 Gramm HJ, Dollinger P, Beier W. Procalcitonin - ein neuer Marker der inflammatorischen Wirtsantwort. Longitudinalstudien bei Patienten mit Sepsis und Peritonitis. Chir Gastroenterol 1995; 11 (suppl2): 51-54. 23 Beier W. Procalcitonin (Prom): ein innovativer Inflammationsparameter mit prognostischen Eigenschaften, An&h Zntensivmed 1994; 36: XVIII. 24 Reith HB, MittelkGtter U, Debus ETS, Lang J, Thiede A. Procalcitonin (PCT) immunoreactivity in critical ill patients on a surgical ICU. In: Faist E. ed. The immune consequencesof tmuma, shock and sepsis. Bologna: Monduzzi Editore, 1997; 673-77.
Procalcitonin: a new marker for diagnosis of acute rejection and bacterial infection in patients after heart and lung transplantation S Hammera, F Meisner: P Dirschedlb, G Habela, P Fraunbergerc, B Meiserd, B Reichardt and C Hammera “Institute of Surgical Research, bIBE, cInstitute of Clinical Chemistry, dDepartment of Cardiothoracic Surgery, Klinikum Groflhadern, LM- University,Munich Received 6 October 1998; revised manuscript accepted 15 October 1998
Abstract: The aim of the study was to investigate the reliability of procalcitonin (PCT), a new potential marker for detection of bacterial, fungal and protoxoal infections, in order to differentiate these from viral infections and early rejections in heart, heart-lung and lung transplanted patients. PCI is a propeptide of calcitonin with unknown origin which is not detectable in plasma of healthy subjects. It increases rapidly and signitlcantly under severe microbial infections. Met/w& PCI plasma levels were measured using an immuno-luminescence assay. C-reactive protein and white blood cells were quantified to validate the PCT values. Results: Increased levels of PCT were found in all transplant patients with bacterial, timgal and protozeal infections. The magnitude of the values were clearly associated with the severity of the infection. Trauma of operation or inflammatoty events such as viral infections and rejections did not trigger PCT-production. The release of PCT did not depend on the type of pathogens even though Aspergillum resulted in the highest levels measured. Sensitivity, specificity and prognostic value of PCT for systemic infections were higher than of the other parameters investigated. Conclusion: PCT is a highly specific analyte which shows significant diagnostic validities when nonviral infections are compared with rejection episodes. PCI discriminates between inflammatory events such as rejection or viral infections and nonviral-infections including bacterial, fungal and protozoa1 infections. The half-life of PCI is 24 h indicating clearly a competent antibiotic treatment. Unnecessary antibiotic therapy can be avoided due to the early exclusion of bacterial and fungal infections.
Introduction Despite improvements in short-term outcome after heart and lung transplantation, patients are lost due to various complications in the early postoperative phase. The two major postoperative incidents are acute rejections (AR) and
Address for correspondence: C Hammer, Institute of Surgical Klinikum Grobhadem, LM-University, Munich, Research, Marchioninistr. 15, Munich, Germany. E-mail: 83177 [email protected] 0 Amold
1998
infections. Many parameters are currently used for differential diagnosis. Most mediators of inflammatory events such as cytokines, interleukins, neopterin and C-reactive protein (CRP) indicate, however, not only immunological reactions
during acute rejection, but also during viral infections, vascular occlusive diseases, malignancies, trauma and of course bacterial, fungal and protozoal infections. Recipients of solid organ transplants are at special risk for infections due to the heavy immunosuppression. ‘Ransplant recipients have an increased susceptibility to all types of infectious material, especially opportunistic organisms. 0966_3274(98)TI242OA
236
S Hammer et al.
Approximately two-thirds of transplant recipients experience at least one episode of infection during the first year after transplantation. Between 10% and 35% of heart and lung transplant patients suffer from infections in the first postoperative weeks. The largest proportion of germs are bacteria with 47-55%. Fungi and protozoa are responsible for 12-74% of infections mainly in the lung. Infections are a major cause (12%) of death in the post-transplant period. The interaction between the immunosuppressive therapy and the environmental exposure determines the risk of infection. Procalcitonin (PO, a species-specific propeptide of calcitonin, is a glycoprotein of 116 amino acids with a molecular weight of 13 kD. Its origin and function are unknown. Gene structure and gene-locus are described by Le Moullec et al.’ PCT was never induced in vitro neither by allogeneic nor by lectin or endotoxin stimulation of peripheral blood cell populations.’ PCT is a species-specific molecule therefore data are only available from humans.3 Investigations performed with synthetic PCT have to be regarded with care, since it is unknown whether synthetic or biological materials have a different tertiary structure or even different post-translational modifications.4 In in vivo studies PCT was found after endotoxin injections in healthy volunteers. Subcutaneous injection of GM-CSF increased the numbers of white blood cells and levels of interleukins but had no effect on PCT (our own observation). PCT selectively responds to severe bacterial , parasitic and fungal infections with systemic manifestation. Thus PCT promises to be a new marker with cytokine character which might develop to a specific indicator of nonviral infections in transplant patients.
discharge (control); group B, local infection; group C, generalized/multiple infection; group D, sepsis. Infections were assessed on a daily basis by positive serology, IgG or IgM antibodies, positive smear or positive blood cultures. Infections were regarded as cured after 7 days of effective antibiotic therapy. Included were those bacterial infections which were treated specifically according to the bacterial spectrum. Local infections were determined as peripheral colonization, infection of mucous membranes or locally limited, insignificant infections. Multiple or generalized infections were defined when different or the same germs were found in one or more organs. As sepsis we defined infections with positive blood cultures and severe clinical symptoms. Multiple, generalized infections and sepsis were comprised under systemic infections. Onset of rejection episodes was designated by histopathological diagnosis from endomyocardial or open lung biopsies and graded according to the criteria of the International Society for Heart and Lung ‘Bansplantation (ISHLT). Start of AR was specified as positive finding at the day of biopsy. A 3-day steroid bolus therapy determined the duration of rejection. StatIetIcaI evaluation For statistical evaluation only values of the day of diagnosis of rejection or infection were included. The results were expressed as mean values with standard error of the mean. The significance of the differences was assessed using the Mann-Whitney U-test. Proportions were compared applying the chi-square test with Yate’s correction. Probabilities of less than 0.05 were accepted as significant. Meaeurement of procalcitonin
Objectives The aim of the study was to elaborate specificity and sensitivity of PCT, to compare this molecule with other markers and to test its capability and reliability to differentiate between rejections and infections due to bacteria and fungi in patients after transplantation of thoracic parenchymal organs.
Material and methods PCT was studied retrospectively in 78 heart and lung transplanted patients, 58 males and 20 females with ages ranging from 1 to 68 years (mean 52 years). Indications for heart transplantation (n = 57) were dilatative cardiomyopathy (n = 35) ischaemic cardiomyopathy (n = 11) and others (n = 22). Lung transplantation, unilateral (n = 11) and bilateral (n = 7) was performed in 18 patients. Seven patients were transplanted due to fibrosis of the lung, four due to emphysema. Five patients suffered from cystic fibrosis and cystic lung disease. Because of chronic rejection one patient was retransplanted. Three heart-lung transplantations were performed in patients suffering from Eisenmenger’s disease (n = 2) and thrombosis of the pulmonary artery. The patients were divided into five groups to differentiate between acute rejections and infections. Group 1: no acute rejection (AR), no infection (control), group 2: AR, group 3: AR and infection, group 4: local infections, group 5: systemic infections (generalized/multiple infections and sepsis); and into four groups to ascertain the severity of infection: Group A, Zkansplant Immunology 1998; 6: 235-241
Blood was collected daily at 8 am. The separated EDTA-plasma was stored at -80°C if not used immediately. PCT was measured by the commercial specific and ultrasensitive immunoluminometric assay. This assay uses two monoclonal antibodies, one directed against residues around 96-106 of PCT as capture antibody and one directed against the residues 70-76 as tracer antibody. The sequence 70-76 is part of the calcitonin molecule, whereas the 96-106 sequence is part of the katacaltin molecule. Synthetic PCT was used as standard. This assay which is specific for PCT has a limit of detection of less than 1 pg/ml. Interand intra-assay variations at both low and high concentrations were less than 7%. In short: 20 pl of plasma are pipetted into the ‘coated tube system’. The 100 tubes (2 x 50) sitting in a holder are coated with a surplus of antigen-specific monoclonal anti-katacalzin antibody. A luminescence labelled antibody directed against the calcitonin part of PCT (250 pl) is now added. The tubes are incubated and agitated for 2 h at room temperature. The excess of the tracer is discarded and the tubes are washed five times. In a luminometer the remaining luminescence signal of the tracer bound to the tube walls is measured using the LUMItest kit reagents (BRAHMS Diagnostica, Berlin, Germany) as a complete diagnostic kit (LUMItest Procalcitonin) and a Luminometer (Behring Diagnostica, Marburg, Germany). Magnitude of the luminescence signal is proportional to the PCI concentration of the specimen.
Procalcitonin: a new marker for diagnosis of acute rejection and bacterial infection
Results PCT, CRP and white blood cells (WEK) were measured in order to differentiate between infections and rejections (Figures 14). The mean value of PCT of 59 patients tested showing neither rejection nor infection was 0.3 + 0.3 ng/ml and was assigned as control. A group of 19 transplanted patients suffering from rejection episodes of all degrees had simihu values: the mean 0.2 f 0.2 ng/rnl. Slightly increased levels were found when infections accompanied rejections (mean 0.6 + 0.6 rig/ml). Most of these infections were of local character. Local infections only resulted again in lower levels with a mean of 0.6 f 0.3 q/ml. In contrast systemic infections induced high values of 10.5 + 18.2 q/ml. There was no difference between PCT values of patients with rejection or no rejection. The differences between rejection and local infection were sign&ant @ < 0.01). PCT of patients with systemic infection was exceedingly higher than that of patients with rejection (p c 0.0001) (Figure 1).
At a cut-off point of ~0.8 ng/ml PCT sensitivity for rejection was lOO%, specificity 82% and predictive value 89%. The alpha error wasp < 0.0001. At a cut-off point for systemic infections of >0.5 pg/ml all three parameters reached 89%. CRP was found to be elevated in all groups. Even transplant patients without any signs of inflammation had a mean of 3.8 + 5.1 mg/dl which exceeds the normal value of less than 0.5 mg/dl. As soon as nonviral infections occurred the values increased to a level of 6.8 f 5.4 mgJd1 independently of whether the infections were local or systemic (Figure 2). Of absolutely no differential value were the numbers of peripheral WBC. In all groups the mean number of leucocytes was elevated and ranged between 12.0 and 16.0 x 10’ cells/I (Figure 3). PCT, CRP and WRC were also measured in order to ascertain the severity of infection (Figures 4-6). As a control group we used values of the day of discharge from the hospital. In this group PCI had a mean value of 0.3 + 0.5 ng/ml. Local infections resulted in
, r
100 I I
237
,
*.
* ’
.
ns ’
l
. I
10 .:
.
~___________________i_ no AR
AR
AR+ infect
local infect
syst.infect
Figure 1 PCT values in patients with and without rejections or infections; ns = not significant, the asterisk represents significancep < 0.01. Normal values < 0.5 nghnl (line).
100 I
.. :
0.11 Figure 2
no AR
. t
AR
AR+ infect
local infect
syst.infect
CRP values in patients with and without rejections or infections. Normal values < 0.5 mg/dl (line).
Tmnsplant Immurwlog~ 1998; 6: 235-241
238
S Hammer et al.
lOO-
OJ
no AR Figure 3
local infect
AR+ infect
AR
systinfect
WBC values in patients with and without rejections or infections. Normal values 4-10 x lo9 cells/l (lines).
I
I
*
’
* .
I
*
1 - .
II
.A.
I
*’
.
. . .
_____________________________________________________
.
. .
i_____.___.____._._.______~______________________________________. .
:
.
A .
0.1 4
1 discharge
Ngure 4
t
1 local infect
:. genJmult. infect
sepsis
Association of PCT values with severity of infections. The asterisk represents significantp < 0.01. Normal values c 0.5 n&l
0.6 + 0.6 ng/ml while multi or generalized infections showed a further increase to 7.3 +- 17.0 q/ml. The highest levels were reached in the sepsis group with 22.4 f 17.8 @ml. The PCT levels at time of discharge were significantly different to all infection groups with values ofp < 0.01 (local infection),p < 0.0001 (generalized!multi-infection),p < O.CKlOl (sepsis). The fhulings of the group with local infection were significantly different from those of the multiple/generalized infections and from those of sepsis @ < O.OOl,p < 0.0001) (Figure 4). High CRP values were found in all groups. Even at time of discharge 2.8 -C 4.0 mg/dl of CRP were seen (normal values co.5 mg/dl). Local and multiple/generalized infections showed an almost identical mean of 6.8 + 6.2 and 6.4 + 4.4 mg/dl, respectively. Sepsis patients developed values of 8.3 f 8.7 mgidl (Figure 7). WBC were 9.5 + 2.6 x lo9 cells/l in the control group and therefore normal. Local and multiple/ eneralized infections increased the number to 13.8 + 6.2 x 10$ cells/l and 15.4 -C 7.5 x lo9 cells/l, respectively. In the sepsis patient the numbers Transplant Immunology 1998; 6: 235-241
(lure).
mounted to 17.9 -C9.3 g/I. A differential diagnosis with this parameter cannot be achieved (Figure 6). Pathogens such as gram-positive and gram-negative bacteria, Candida, Aspergillum and combined infections were associated with PCI values. The assumption that the release of PCT depends on the type of pathogens could not be verified. Even though Aspergillum in some cases resulted in very high PCT levels (Figure 7).
Discussion PCT is a new and innovative parameter for the diagnosis of severe bacterial and fungal infections. It is a reliable indicator of inflammatory activities during sepsis and multi-organ failure. PCI’ is the prohormone of calcitonin with a molecular weight of 13 kD but lacks hormonal activity.’ Its biological function and organ of origin is still unknown. C-cells of the thyroid gland and other neuroendocrine celIs are supposed to produce PCT
Procalcitonin: a new marker for diagnosis of acute rejection and bacterial infedion
239
-. discharge
local infect
gen./mult. infect
sepsis
Figure 5 Associationof CRP values with severityof infections. Normal values < 0.5 @ml (line).
11 discharge Figure
local infect
gen./mult. infect
sepsis
6 Association of WBC values with severity of infections. Normal values 4-10 x lo9 ceils/l (lines).
Elevated levels were detected in patients suffering from medullaxy thyroid carcinoma, but PCT was also found in thyreodectomized patients.6 Injection of bacterial endotoxin in healthy volunteers led within 24 h to PCT levels exceeding the baseline of
normal levels in transplant patients with viral diseases and also in HIV patients even in terminal disease.” Peripheral colonization, wound healing, fungal stomatitis and encapsulated processes like abscesses could not be shown to induce PCT release.5 Systemic infections in general resulted in high levels of Pa, initially used to dia ose sepsis and septicaemia espe cially in new-born infants. !T These beneficial characteristics allow to differentiate septic shock from cardiogenic shock,13 toxic from biliary pancreatitis14 and infectious from noninfectious ARlXi5 Huber et al. describes a failure of PCT in some immunodeficient patients with severe fungal infection. The explanation given was that chronic diseases and immunosuppression downregulate or exhaust PCT production.16 In our cohorts of transplanted patients chemical immunosuppression including azathioprin, cyclosporin and tacrolimus as well as steroids and steroid bolus theray during rejection episodes had no impact on PCT behaviour. ’ Eberhard et al. observed that murine monoclonal anti-CD3-lymphocytic antibodies (Okt 3), which
MCI S Hammer et al.
100,
. . .
mixed infections
. .
. . . ,
.
10 --
. . A
%
I
:
-
F
t .
.
1 -:
: .
. I .
. *
:
0.1 4 gr~.beict.
grarmegbact.
Candida
Aspergilus
Type of pathogens Figure 7
------I. -
&fungi
n= 16, Gl u=26,G2 n=49,G3 n=15,G4
Different behaviour of PCT in transplant patients (schematic)
induce the ‘interleukin storm’ in transplant patients increase PCT significantly.‘a This specsc unresponsiveness to rejection mechanisms on one hand and conservative immunosuppression on the other hand is a major advantage of PCI’ for the monitoring of transplant patients. Neither acute nor chronic rejections of thoracic and liver grafts were accompanied by PCT increase.1g Thus levels of PCT exceeding values of 0.8 @ml (cut-off point) are concentrations which prove an infection of systemic character with a sensitivity of 82%. Four courses of PCT could be identified in transplanted patients (Figure 8). First, patients who never exceeded normal vahms and were therefore always free of severe nonviral infection (Gl, n = 16). A second group of patients started with normal values to develop high levels of PCT during postoperative treatment due to acquired infections on the ICU. Specific antibiotic treatment safely terminated the infections and discontinued PCT production ideally indicated by the half-life of 1 day (G2, n = 26). A third group with high levels of PCT Tmnplant
bacteria
gratmeghct.
Association of type of pathogens with levels of PCT
PCT
FIgwe 8
grarcpos. 8
Immunology
1998; 6: 235-241
after transplantation developed normal values within a short time due to the appropriate empiric antibiotic treatment (Q, II = 49). The most interesting group was group four which started with significant amounts of PCT which dropped for a few days under antibiotic treatment and then increased again after a plateau period. Four out of five of these patients died from irreversible Aspergilhun infections (G4, II = 15): Another advantage of PCT was found in children after bone marrow transplantation or meningitis. Here PCT allows differentiation whether clinical symptoms including ‘fever of unknown origin’ are due to bacteria or other inflammatory mechanisms. This diagnosis prevented unnecessary myelotoxic antibiotic treatment in case of nonbacterial and nonfungal iIlftXtiOIIS.m
A further diagnostic advantage of PCT as compared to other sophisticated methods is its uncomplicated and quick testing performance. After less than 3 h reliable values are established. Plasma samples stored at -20°C are stable for PCT for at least half a year.
Procalcitonin: a new marker for diagnosis of acute rejection and bacterial infection
IL6 also allows differentiation between systemic infections and rejection,*l elevates during sepsis,5 correlates with the severity of infection and predicts the outcome of the patient.** The disadvantage of IL6 as compared to PCT is its increase during viral infections, trauma, major operations” and other inflammatory events like autoimmune diseases.‘* Its short half-life in vivo and in vitro reduces its relevance for clinical routine application. CRP and leucocytes were both markers which increased during trauma and surgery and showed major overlapping during viral and bacterial infections and rejection episodes.ls The slow increase (24 h) and decrease of CRP in comparison to IL6 (1 h) and PCT (2-4 h) in peripheral blood reduces its clinical, prognostic and therapeutic value. Severity and dynamic of infections is not heralded by CRP and WEE numbers. The minor differences are not sufficient to be used for transplant patients during intensive care.’ Another advantage of PCT for clinical purposes is its ability to predict the outcome of patients and to demonstrate a typical course if the infection is not successfully treated. Beier was able to verify the prognostic properties of PCT233’4 Despite of these convincing characteristics PCT should not be regarded as the only reliable parameter for infections in transplant patients in intensive care units. In some cases peripheral colonization, infected wound healing, fungal stomatitis and encapsulated processes like abscesses were not heralded by PCT? A combination with other markers is still capable of improving the treatment and thus the clinical situation of patients after transplantation.
References Le Moullec JM, Jullienne A, Chenais J et al. The complete sequence of pre-pro-calcitonin. FEBS 1984; 167: 93-97. Staehler M, iiberfuhr P, Reichart B, Hammer C. Differentialdiagnostik der AbstoOungsreaktion und Infektion bei herztransplantierten Patienten: Neue Wege mit Zytokinen und Prokalzitonin als Marker. Transplant Medizin 1997; 9: 44-50. Gendrel D, Bohuon C. Procalcitonin, a marker of bacterial infection. Infection 1997; 25: 133-34 Meisner M. PCT und Score Systeme, 3rd Advisory Board and Investigator Meeting [Abstracts], Berlin, 1998. Karzai W, Oberhoffer M, Meier-Hellmann A, Reinhart K. Procalcitonin - a new indicator of the systemic response to severe infections. Infection 1997; 25: 329-34. Assicot M, Gendrel D, Carsin H, Raymond J, Guilbaud J, Bohuon C. High serum procalcitonin concentrations in patients with sepsis and infection. Lancet 1993; 341: 515-18. Dandona P, Nix D, Wilson MF et al. Procalcitonin increase after endotoxin injection in normal subjects. JClin Endocrin Metaboi 1994; 79: 1606-608. Brunkhorst FM, Forycki ZF, Wagner J. Release and kinetics of procalcitonin (PCT) after gram negative bacterial injection in a healthy subject [Abstract]. Shock 1997; 7 (suppl): 493.
Transplant Immunology
1998; 6: 235-241
241
9 Gerard Y, Hober D, Petitjean Set al. High serum procalcitonin level in a four year old liver transplant recipient with a disseminated candid&is. Znfect 1995; 23: 310-11. 10 Reith HB, Lehmkuhl P, Beier W, Hijgy B. Procalcitonin- ein progInfektionsparameter bei der Peritonitis. Chir nostischer Gustroenteroll995; 11 (suppl2): 47-50 11 Gerard Y, Hober D, Assicot M et al. Procalcitonin as a marker of bacterial sepsis in patients infected with HIV-l. JZnfection 1997; 35: 4146. 12 Gendrel D, Raymond J, Assicot M et al. Procalcitonin as a marker for the early diagnosis of neonatal infection. J Ped 1996; 128: 570-73. 13 De Werra I, Jaccard C, Corradin SB et al. Cytokines, nitrite/ nitrate, soluble tumour necrosis factor receptors, and procalcitonin concentrations: Comparisons in patients with septic shock, cardiogenie shock, and bacterial pneumonia. Crit Care Med 1997; 25: 607-13. 14 Brunkhorst FM, Forycki ZF, Wagner J. Friihe Identitiierung der bililren akuten Pankreatitis durch Procalcitonin-Immunreaktivitlt vorllufige Ergebnisse. C&r Gustroenteroll995,ll (suppl2): 42-46. 15 Brunkhorst FM, Forycki ZF, Wagner J. Discrimination of infectious and non-infectious etiologies of the adult respiratory distress syndrome (ARDS) with procalcitonin immunoreactivity [Abstract]. Clin Zntens Care 1995; 6: 3. 16 Huber W, Schweigart U, Bottermann P. Failure of PCT to indicate severe fungal infection in two immunodeficient patients. Znfection 1997; 25: 377-78. 17 Staehler M, Hammer C, Meiser B, Reichart B. A new marker for differential diagnosis of acute rejection and bacterial infection in heart transplantation. TransplantProc 1997; 29: 584-85. 18 Eberhard OK, Haubitz M, Brunkhorst FM, Kliem V, Koch KM, Brunkhorst R. Usefulness of procalcitonin for differentiation between activity of systemic autoimmune disease and invasive bacterial infection. A& Rheum 1997; 40: 1250-56. 19 Meisner M, Tschaikowsky K, Schmidt J, Schiittler J. Procalcitonin (PCT) - indications for a new diagnostic parameter of severe bacterial infection and sepsis in transplantation, immunosuppression and cardiac assist devices. Cardiovusc Engineer 1996; 1: 67-76. 20 Gendrel D, Raymond J, Assicot M et al. Measurement of procalcitonin levels in children with bacterial and viral meningitis. Clin Infect Dis 1997; 24: 1240-42. 21 Fraunberger P, Pfeiffer M, Haller M et al. Cytokine and cytokinereceptor profiles after liver and heart transplantation. Transplant Proc 1995; 27: 2023-27. 22 Gramm HJ, Dollinger P, Beier W. Procalcitonin - ein neuer Marker der inflammatorischen Wirtsantwort. Longitudinalstudien bei Patienten mit Sepsis und Peritonitis. Chir Gastroenterol 1995; 11 (suppl2): 51-54. 23 Beier W. Procalcitonin (Prom): ein innovativer Inflammationsparameter mit prognostischen Eigenschaften, An&h Zntensivmed 1994; 36: XVIII. 24 Reith HB, MittelkGtter U, Debus ETS, Lang J, Thiede A. Procalcitonin (PCT) immunoreactivity in critical ill patients on a surgical ICU. In: Faist E. ed. The immune consequencesof tmuma, shock and sepsis. Bologna: Monduzzi Editore, 1997; 673-77.