Kinetics of Interleukin-6, Procalcitonin, and C-Reactive Protein After Pediatric Liver Transplantation R. Zant*, M. Melter, B. Knoppke, M. Ameres, and J. Kunkel KUNO Children’s University Hospital, Regensburg, Germany
ABSTRACT In the early phase after pediatric liver transplantation (pLT) several concomitant factors may reduce the performance of established sepsis markers. To date, their clinical interpretation is hindered by a lack of information on their postoperative kinetics. To gather more information on the postoperative course and their changes in bacterial sepsis, we prospectively studied C-reactive protein (CRP), interleukin-6 (IL-6), and procalcitonin (PCT) on 9 perioperative days in 25 consecutive pLTs. After an initial postoperative peak, IL-6 and CRP levels significantly re-increased in patients with bacterial sepsis (P < .001). In contrast, PCT had very high postoperative levels; therefore severe infection was a comparatively inferior trigger for PCT elevation compared with the initial operation. The area under the receiver operating characteristic curve to diagnose postoperative sepsis for PCT was only 0.52, compared with 0.95 for IL-6 and 0.89 for CRP. None of the studied biomarkers were depressed by poor graft function. In conclusion, PCT performs poorly as a biomarker for sepsis in the early phase after pLT. With a rapid decline of initially elevated levels, IL-6 provides the best kinetics for detection of postoperative bacterial sepsis.
B
ACTERIAL sepsis is a frequent problem following pediatric liver transplantation (pLT) [1]. Associated with surgical interventions, blood loss, and immunosuppression, overwhelming sepsis is one of the leading causes of death in the early postoperative period [2,3]. To reduce morbidity and mortality in these immunocompromised patients, an early diagnosis of sepsis is crucial. In pediatric patients, interleukin-6 (IL-6), procalcitonin (PCT), and Creactive protein (CRP) have proved to be reliable markers for the postoperative detection of severe infection. However, interpretation of biomarkers in the early phase after pLT is notably hampered by a severe systemic inflammatory response syndrome (SIRS) caused by transplant surgery, initial liver insufficiency, repeated operations, immunosuppression, and high protein losses, eg, via abdominal drains. To the best of our knowledge, to date there are no data available about the kinetics of IL-6 and PCT in the critical early phase after pLT, which further impedes their clinical interpretation. The aim of the present study was to describe the postoperative kinetics of IL6, PCT, and CRP levels and their changes in sepsis. Moreover, we assessed their correlation with early graft function and rejection.
PATIENTS AND METHODS From September 2010 to March 2012, CRP, IL-6, and PCT were prospectively measured in 22 consecutive pediatric patients who underwent 25 liver transplantations at the KUNO Children’s University Hospital. All patients were included in this prospective cohort study, which was approved by the local Institutional Review Board (10-160-0239). The study is in accordance with the Helsinki Declaration of 1975 (as revised in 1983). IL-6, CRP, and PCT were measured in all patients on the following days: the day before pLT, the day of transplantation (after admission to the pediatric intensive care unit) and the next 7 consecutive postoperative days (PODs) or up to the date of early retransplantation if this occurred within a week. Bacterial sepsis was defined as SIRS, according to the criteria established by the International Pediatric Sepsis Consensus Conference, a clinical suspicion of bacterial infection with the need to modify the existing antibiotic therapy, and detection of bacteria in either a blood culture, bile, or ascites, or full SIRS score without evidence of a nonbacterial infection [4]. After the first septic day, ongoing sepsis was diagnosed on
*Address correspondence to Robert Zant, KUNO Children’s University Hospital Regensburg, Franz-Josef-Strauss-Allee 11, 93053 Regensburg, Germany. E-mail: [email protected]
ª 2014 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
0041-1345/14 http://dx.doi.org/10.1016/j.transproceed.2014.08.048
Transplantation Proceedings, 46, 3507e3510 (2014)
3507
3508 the days following as long as the SIRS criteria were met (septic episode). Additionally, for comparative purposes, Pediatric Risk of Mortality (PRISM) III scores were calculated [5]. Patients’ mean international normalized ratio (INR) values and mean bilirubin levels at POD 5 were used to assess early graft function. Immunosuppression was started with intraoperative, intravenous administration of prednisolone (300 mg/m2 body surface area [BSA], with a maximum of 500 mg). Thereafter, all patients received immunosuppressive induction therapy with basiliximab (35 kg body weight [BW]: 10 mg; >35 kg bw: 20 mg) on PODs 0 and 4. Baseline immunosuppression in all patients consisted of cyclosporine and prednisolone. Cyclosporine was started with 50 mg/m2 BSA every 12 hours. Thereafter, cyclosporine dosing was adjusted to a maintenance trough level of 130e170 ng/mL. The standard dose for prednisolone was 15 mg/m2 BSA. Indications for high-dose prednisolone treatment (60 mg/m2 BSA) were acute liver failure, cystic fibrosis, autoimmune hepatitis, and others with a high immunologic risk in the recipient (eg, retransplantation). Standard prophylactic antibiotic treatment during the study period consisted of piperacillin-sulbactam. This regimen was altered according to the patient’s infectious status.
Assays Blood samples were taken from either the arterial or the central venous line. Quantitative measurements of PCT were performed with the use of the Advia Centaur Brahms PCT test kit (Brahms, Hennigsdorf, Germany). The analytic sensitivity of the assay was 0.02 ng/mL (range, 0.02e75 ng/mL). CRP was determined with the use of the Flex reagent cartridge/Dimension Vista System (Siemens Healthcare Diagnostic Products, Marburg, Germany), which is based on a particle-enhanced nephelometric immunoassay technique. The analytic assay sensitivity was 2.9 mg/L. IL-6 was measured with the use of the enzyme-labeled, chemiluminescent sequential immunometric Immulite 2000 IL-6 assay (Siemens Medical Solutions Diagnostics). The test’s analytic sensitivity was 2 pg/mL.
Statistical Analysis The evaluation of the 3 sepsis markers (IL-6, CRP, and PCT) was performed by calculating receiver operating characteristic (ROC) curves for all days (n ¼ 217). The Youden index was calculated to determine the optimal cutoff values. The Mann-Whitney test was used to compare concentrations of septic and nonseptic days and to compare concentrations of days with and without rejection. Potential associations between biomarker levels and early graft function were tested with linear regression. Data are presented as median with 0.1e0.9 quantiles. All analyses were performed to PAWS statistics 18 software (SPSS Inc., Chicago).
RESULTS
Median age of the study population was 2 years with a range from 19 days to 16 years. The indications for pLT are summarized in Table 1. Living donations (n ¼ 11) were performed as left lateral (n ¼ 10) or right (n ¼ 1) segment transplantation. Deceased liver transplantations were either performed as whole (n ¼ 5) or left lateral lobe split liver transplantation (n ¼ 9). In 10 patients, prednisolone was administered in a higher dose of 60 mg/m2 BSA. In 6
ZANT, MELTER, KNOPPKE ET AL Table 1. Indications for Pediatric Liver Transplantation Etiology
n
Biliary atresia Cystic fibroses Progressive familial intrahepatic cholestasis Type 1 Type 2 Acute liver failure Acute graft failure Chronic graft failure Crigler-Najjar syndrome type 1 Budd-Chiari syndrome Mitochondrial disorder Complex 1 defect Complex 1 and 4 defect Glycogenosis type IV Autoimmune hepatitis Liver failure of unknown origin
8 2 2 1 1 2 3 1 1 1 2 1 1 1 1 1
transplantations, the standard prophylactic antibiotic treatment was altered according to their infectious status. Within the study period, 2 patients were treated for graft rejection on POD 7. In 1 patient, treatment was based on clinical and laboratory findings. Because of clinical instability, liver biopsy was not obtained. In the 2nd patient, rejection was confirmed histologically. Both patients responded well to treatment. Neither of these 2 patients was assessed as septic. In these 2 patients, levels of IL-6, PCT, and CRP were not significantly elevated at the onset of rejection compared with patients without rejection on POD 7. Poor graft function was not associated with depressed levels of IL-6, PCT, and CRP. In contrast, in our study population, high PCT levels on POD 1 were significantly associated with increased INR (P < .05) and bilirubin values on POD 5 (P < .05). Also, poor postoperative PCT clearance with high levels on POD 5 were associated with high INR (P ¼ .001) and bilirubin levels (P ¼ .001) on POD 5. For IL-6 and CRP, neither peak levels nor levels on POD 5 correlated with parameters of graft function (INR and bilirubin) on POD 5. In 21 of 25 consecutive liver transplantations, SIRS was diagnosed (110 of 217 evaluated days). The criteria for bacterial sepsis were met in 4 patients on 11 days. Staphylococcus hominis (blood culture) and Enterococcus faecium (ascites, blood culture, bile) were identified as causative organisms in 3 patients. On POD 2, after primary recovery from pLT, 1 patient presented a rapid clinical deterioration with severe SIRS (full score) and further clinical signs of an infection. Despite the finding of putrid ascites intraoperatively, an infectious agent was not isolated under antibiotic therapy with meropenem and vancomycin. Median PRISM-III score on the day of admission was 10 (1/16), with a steady decline to POD 7. Septic episodes were more often associated with PRISMIII scores >13 (27% vs 16%). Within the study period, no patient died of bacterial infection.
KINETICS OF IL-6, PROCALCITONIN, AND CRP
3509
Table 2. Area Under the Receiver Operating Characteristic Curve (AUC), Cutoff, Sensitivity, Specificity, Positive (PPV) and Negative (NPV) Predictive Values, and Youden Index for Interleukin-6 (IL-6), C-Reactive Protein (CRP), and Procalcitonin (PCT) Calculated for All Evaluated Days (n [ 217) Diagnostic Test AUC
IL-6 CRP PCT
Cutoff
0.95 239 pg/mL 0.89 93 mg/L 0.52 1.1 ng/mL
Sensitivity Specificity PPV NPV (%) (%) (%) (%) Youden
100 82 82
89 91 34
34 100 56 99 5.5 94
0.89 0.73 0.16
Levels of IL-6, PCT, and CRP in Nonseptic Patients
In nonseptic patients, the median value of IL-6 was 14 pg/ mL (
ABSTRACT In the early phase after pediatric liver transplantation (pLT) several concomitant factors may reduce the performance of established sepsis markers. To date, their clinical interpretation is hindered by a lack of information on their postoperative kinetics. To gather more information on the postoperative course and their changes in bacterial sepsis, we prospectively studied C-reactive protein (CRP), interleukin-6 (IL-6), and procalcitonin (PCT) on 9 perioperative days in 25 consecutive pLTs. After an initial postoperative peak, IL-6 and CRP levels significantly re-increased in patients with bacterial sepsis (P < .001). In contrast, PCT had very high postoperative levels; therefore severe infection was a comparatively inferior trigger for PCT elevation compared with the initial operation. The area under the receiver operating characteristic curve to diagnose postoperative sepsis for PCT was only 0.52, compared with 0.95 for IL-6 and 0.89 for CRP. None of the studied biomarkers were depressed by poor graft function. In conclusion, PCT performs poorly as a biomarker for sepsis in the early phase after pLT. With a rapid decline of initially elevated levels, IL-6 provides the best kinetics for detection of postoperative bacterial sepsis.
B
ACTERIAL sepsis is a frequent problem following pediatric liver transplantation (pLT) [1]. Associated with surgical interventions, blood loss, and immunosuppression, overwhelming sepsis is one of the leading causes of death in the early postoperative period [2,3]. To reduce morbidity and mortality in these immunocompromised patients, an early diagnosis of sepsis is crucial. In pediatric patients, interleukin-6 (IL-6), procalcitonin (PCT), and Creactive protein (CRP) have proved to be reliable markers for the postoperative detection of severe infection. However, interpretation of biomarkers in the early phase after pLT is notably hampered by a severe systemic inflammatory response syndrome (SIRS) caused by transplant surgery, initial liver insufficiency, repeated operations, immunosuppression, and high protein losses, eg, via abdominal drains. To the best of our knowledge, to date there are no data available about the kinetics of IL-6 and PCT in the critical early phase after pLT, which further impedes their clinical interpretation. The aim of the present study was to describe the postoperative kinetics of IL6, PCT, and CRP levels and their changes in sepsis. Moreover, we assessed their correlation with early graft function and rejection.
PATIENTS AND METHODS From September 2010 to March 2012, CRP, IL-6, and PCT were prospectively measured in 22 consecutive pediatric patients who underwent 25 liver transplantations at the KUNO Children’s University Hospital. All patients were included in this prospective cohort study, which was approved by the local Institutional Review Board (10-160-0239). The study is in accordance with the Helsinki Declaration of 1975 (as revised in 1983). IL-6, CRP, and PCT were measured in all patients on the following days: the day before pLT, the day of transplantation (after admission to the pediatric intensive care unit) and the next 7 consecutive postoperative days (PODs) or up to the date of early retransplantation if this occurred within a week. Bacterial sepsis was defined as SIRS, according to the criteria established by the International Pediatric Sepsis Consensus Conference, a clinical suspicion of bacterial infection with the need to modify the existing antibiotic therapy, and detection of bacteria in either a blood culture, bile, or ascites, or full SIRS score without evidence of a nonbacterial infection [4]. After the first septic day, ongoing sepsis was diagnosed on
*Address correspondence to Robert Zant, KUNO Children’s University Hospital Regensburg, Franz-Josef-Strauss-Allee 11, 93053 Regensburg, Germany. E-mail: [email protected]
ª 2014 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710
0041-1345/14 http://dx.doi.org/10.1016/j.transproceed.2014.08.048
Transplantation Proceedings, 46, 3507e3510 (2014)
3507
3508 the days following as long as the SIRS criteria were met (septic episode). Additionally, for comparative purposes, Pediatric Risk of Mortality (PRISM) III scores were calculated [5]. Patients’ mean international normalized ratio (INR) values and mean bilirubin levels at POD 5 were used to assess early graft function. Immunosuppression was started with intraoperative, intravenous administration of prednisolone (300 mg/m2 body surface area [BSA], with a maximum of 500 mg). Thereafter, all patients received immunosuppressive induction therapy with basiliximab (35 kg body weight [BW]: 10 mg; >35 kg bw: 20 mg) on PODs 0 and 4. Baseline immunosuppression in all patients consisted of cyclosporine and prednisolone. Cyclosporine was started with 50 mg/m2 BSA every 12 hours. Thereafter, cyclosporine dosing was adjusted to a maintenance trough level of 130e170 ng/mL. The standard dose for prednisolone was 15 mg/m2 BSA. Indications for high-dose prednisolone treatment (60 mg/m2 BSA) were acute liver failure, cystic fibrosis, autoimmune hepatitis, and others with a high immunologic risk in the recipient (eg, retransplantation). Standard prophylactic antibiotic treatment during the study period consisted of piperacillin-sulbactam. This regimen was altered according to the patient’s infectious status.
Assays Blood samples were taken from either the arterial or the central venous line. Quantitative measurements of PCT were performed with the use of the Advia Centaur Brahms PCT test kit (Brahms, Hennigsdorf, Germany). The analytic sensitivity of the assay was 0.02 ng/mL (range, 0.02e75 ng/mL). CRP was determined with the use of the Flex reagent cartridge/Dimension Vista System (Siemens Healthcare Diagnostic Products, Marburg, Germany), which is based on a particle-enhanced nephelometric immunoassay technique. The analytic assay sensitivity was 2.9 mg/L. IL-6 was measured with the use of the enzyme-labeled, chemiluminescent sequential immunometric Immulite 2000 IL-6 assay (Siemens Medical Solutions Diagnostics). The test’s analytic sensitivity was 2 pg/mL.
Statistical Analysis The evaluation of the 3 sepsis markers (IL-6, CRP, and PCT) was performed by calculating receiver operating characteristic (ROC) curves for all days (n ¼ 217). The Youden index was calculated to determine the optimal cutoff values. The Mann-Whitney test was used to compare concentrations of septic and nonseptic days and to compare concentrations of days with and without rejection. Potential associations between biomarker levels and early graft function were tested with linear regression. Data are presented as median with 0.1e0.9 quantiles. All analyses were performed to PAWS statistics 18 software (SPSS Inc., Chicago).
RESULTS
Median age of the study population was 2 years with a range from 19 days to 16 years. The indications for pLT are summarized in Table 1. Living donations (n ¼ 11) were performed as left lateral (n ¼ 10) or right (n ¼ 1) segment transplantation. Deceased liver transplantations were either performed as whole (n ¼ 5) or left lateral lobe split liver transplantation (n ¼ 9). In 10 patients, prednisolone was administered in a higher dose of 60 mg/m2 BSA. In 6
ZANT, MELTER, KNOPPKE ET AL Table 1. Indications for Pediatric Liver Transplantation Etiology
n
Biliary atresia Cystic fibroses Progressive familial intrahepatic cholestasis Type 1 Type 2 Acute liver failure Acute graft failure Chronic graft failure Crigler-Najjar syndrome type 1 Budd-Chiari syndrome Mitochondrial disorder Complex 1 defect Complex 1 and 4 defect Glycogenosis type IV Autoimmune hepatitis Liver failure of unknown origin
8 2 2 1 1 2 3 1 1 1 2 1 1 1 1 1
transplantations, the standard prophylactic antibiotic treatment was altered according to their infectious status. Within the study period, 2 patients were treated for graft rejection on POD 7. In 1 patient, treatment was based on clinical and laboratory findings. Because of clinical instability, liver biopsy was not obtained. In the 2nd patient, rejection was confirmed histologically. Both patients responded well to treatment. Neither of these 2 patients was assessed as septic. In these 2 patients, levels of IL-6, PCT, and CRP were not significantly elevated at the onset of rejection compared with patients without rejection on POD 7. Poor graft function was not associated with depressed levels of IL-6, PCT, and CRP. In contrast, in our study population, high PCT levels on POD 1 were significantly associated with increased INR (P < .05) and bilirubin values on POD 5 (P < .05). Also, poor postoperative PCT clearance with high levels on POD 5 were associated with high INR (P ¼ .001) and bilirubin levels (P ¼ .001) on POD 5. For IL-6 and CRP, neither peak levels nor levels on POD 5 correlated with parameters of graft function (INR and bilirubin) on POD 5. In 21 of 25 consecutive liver transplantations, SIRS was diagnosed (110 of 217 evaluated days). The criteria for bacterial sepsis were met in 4 patients on 11 days. Staphylococcus hominis (blood culture) and Enterococcus faecium (ascites, blood culture, bile) were identified as causative organisms in 3 patients. On POD 2, after primary recovery from pLT, 1 patient presented a rapid clinical deterioration with severe SIRS (full score) and further clinical signs of an infection. Despite the finding of putrid ascites intraoperatively, an infectious agent was not isolated under antibiotic therapy with meropenem and vancomycin. Median PRISM-III score on the day of admission was 10 (1/16), with a steady decline to POD 7. Septic episodes were more often associated with PRISMIII scores >13 (27% vs 16%). Within the study period, no patient died of bacterial infection.
KINETICS OF IL-6, PROCALCITONIN, AND CRP
3509
Table 2. Area Under the Receiver Operating Characteristic Curve (AUC), Cutoff, Sensitivity, Specificity, Positive (PPV) and Negative (NPV) Predictive Values, and Youden Index for Interleukin-6 (IL-6), C-Reactive Protein (CRP), and Procalcitonin (PCT) Calculated for All Evaluated Days (n [ 217) Diagnostic Test AUC
IL-6 CRP PCT
Cutoff
0.95 239 pg/mL 0.89 93 mg/L 0.52 1.1 ng/mL
Sensitivity Specificity PPV NPV (%) (%) (%) (%) Youden
100 82 82
89 91 34
34 100 56 99 5.5 94
0.89 0.73 0.16
Levels of IL-6, PCT, and CRP in Nonseptic Patients
In nonseptic patients, the median value of IL-6 was 14 pg/ mL (
