Research Article For reprint orders, please contact: [email protected]
Should procalcitonin be measured routinely in acute decompensated heart failure?
Aim: To elucidate the prognostic role of procalcitonin (PCT) in patients with acute decompensated heart failure (ADHF) without clinical signs of infection at admission. Materials & Methods: Serial measurements of PCT and NT-proBNP were performed in 168 patients, aged 68 ± 10 years with ADHF followed by 3-month outcome evaluation. Results: Cox regression analysis demonstrated significant predictive value of baseline PCT for all-cause death/hospitalization (area under the curve: 0.67; p = 0.013) at 90th day. The patients with persistently elevated PCT or with an increase during the first 72 h of hospitalization had the worst prognosis (p = 0.0002). Conclusion: Baseline and serial in-hospital measurements of PCT have significant prognostic properties for 3-month all-cause mortality/hospitalization in patients with ADHF without clinical signs of infection at admission. Keywords: ADHF • NT-proBNP • procalcitonin • prognosis • serial measurements
Heart failure (HF) is a major burden on the health systems. Episodes of acute decompensated heart failure (ADHF) are associated with significant short- and long-term morbidity and mortality. Patients hospitalized for heart failure had a threefold greater risk for all-cause mortality compared with patients who did not have a heart failure hospitalization after adjustment for known baseline predictors of death [1] . Thus, there is a need for earlier diagnosis, better risk stratification and cost-effective treatment to cut rates of HF hospitalization and to improve outcome. Biomarkers strive to undertake such tasks with evident fast development of biomarker research in this domain [2] . In recent years, procalcitonin (PCT) has emerged as a promising marker for the diagnosis of bacterial infections because higher levels are found in severe bacterial infections than in viral infections and nonspecific inflammatory diseases [3,4] . Thus, PCT may be used to support clinical decision making for the initiation and discontinuation of antibiotic therapy [5] . Few studies have investigated the role of PCT in patients with ADHF, although there was
10.2217/BMM.15.29 © 2015 Future Medicine Ltd
a slightly increased concentration of PCT in patients with severe acute heart failure, the results were not conclusive [6–8] . HF patients are characterized to have altered gut permeability, owing to bowel congestion or ischemia, which leads to invasion of bacterial endotoxins known to induce PCT secretion [9,10] . Serial in-hospital measurements of PCT have never been previously studied for its prognostic capacity in patients with ADHF without overt clinical manifestation of infection at hospital admission. The aim of the present study was therefore to perform a comparative evaluation of circulating serial in-hospital measurements of PCT and NTproBNP as predictors of 3-month all-cause mortality/hospitalizations in patients with ADHF without clinical sign of infection at admission. We have also investigated the relationship between PCT concentration, antibiotic treatment and outcome in patients with ADHF.
Goran Loncar*,†,1, Verena Tscholl†,2,3, Elvis Tahirovic2, Nikola Sekularac1, Almuth Marx4, Danilo Obradovic2, Jovan Veskovic2, Mitja Lainscak5,6, Stephan von Haehling7,8, Frank Edelmann2, Aleksandra Arandjelovic1,9, Svetlana Apostolovic10, Dragana Stanojevic10, Burkert Pieske2, Tobias Trippel‡,2 & Hans-Dirk Dungen‡,2 Department of Cardiology, Clinical Hospital Zvezdara, Belgrade, Serbia 2 Department of Cardiology, Campus Virchow, Charité Universitätsmedizin Berlin, Berlin, Germany 3 Department of Cardiology, Campus Benjamin Franklin, Charité Universitätsmedizin Berlin, Berlin, Germany 4 Marx-Statistics, Nuremberg, Germany 5 Departments of Cardiology & Research & Education, General Hospital Celje, Celje, Slovenia 6 Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia 7 Department of Cardiology, Applied Cachexia Research, Charité Medical School, Berlin, Germany 8 Center for Cardiovascular Research (CCR), Charité Medical School, Campus Mitte, Berlin, Germany 9 School of Medicine, University of Belgrade, Serbia 10 Clinic for Cardiovascular Diseases, Clinical Center Nis, Serbia *Author for correspondence: Tel.: +381 64 145 50 13 Fax: +381 11 380 94 76 [email protected] † G Loncar and V Tscholl share the first authorship ‡ T Trippel and HD Dungen share the last authorship 1
Materials & methods This multicenter, international (Serbia: 9 sites; Germany: 1 site), observational, pro-
Biomark. Med. (2015) 9(7), 651–659
part of
ISSN 1752-0363
651
Research Article Loncar, Tscholl, Tahirovic et al. spective study enrolled 185 patients with ADHF between June 2011 and June 2012. The diagnosis of ADHF was diagnosed according to the European Society of Cardiology criteria [11] . Patients were excluded if they had any of the following: temperature >38°C, sepsis or active infection, acute coronary syndrome, current or planned ultrafiltration, hemofiltration or dialysis, acute myocarditis or hypertrophic obstructive, restrictive or constrictive cardiomyopathy, malignant disease with a life expectancy of less than 2 years and autoimmune disease. After exclusion of 17 patients (15 patients withdrew informed consent, one patient excluded due to violation of inclusion criteria – not hospitalized for ADHF, one patient lost to follow-up), the final analysis included 168 patients. In Serbia 158 patients were enrolled, while in Germany ten patients. Patients were included within 24 h after hospitalization with outcome evaluation at 3-month follow-up. Clinical variables and blood samples were undertaken from every patient at baseline, after 72 h of hospital admission, at hospital discharge and at 3-months follow-up. To be discharged from the hospital, patients had to be in a recompensated dry state according to the judgment of treating physician. During follow-up, patients were treated according to HF guidelines [11] . The clinical check-up and treatment optimization was performed during visits at 14 and 42 days after discharge. The primary end point was defined as allcause mortality/ all-cause hospitalizations at 3-month follow-up. The study was conducted in accordance to the Declaration of Helsinki and approved by the ethics committee of all participating institutions. All participants gave written informed consent. Biomarker assessment
Blood samples were taken from an antecubital vein. The samples were centrifuged, transferred into tubes and frozen at -80°C until it was assayed. Samples were processed by personnel blinded from any patient data. PCT was measured with a time-resolved, amplified cryptate emission assay on the KRYPTOR System (B.R.A.H.M.S GmbH, Hennigsdorf, Germany). The PCT-sensitive KRYPTOR assay has a detection limit of 0.02 ng/ml and a functional assay sensitivity of 0.06 ng/ml [12] . The interassay coefficient of variation for concentration >0.3 ng/ml is 0.14, events = 3 of 5
10
PCT (BL) > 0.14, PCT (72 h) ≤ 0.14, events = 3 of 10 PCT (BL) > 0.14, PCT (72 h) > 0.14, events = 15 of 32
0 0
30
60
90
Time (days) Figure 4. Kaplan–Meier analysis of event-free survival according to the change of procalcitonin levels after 72 h of hospitalization.
In patients with ADHF it is obviously important to identify those with the highest risk in order to improve outcome. In our study, PCT had a prognostic value for 3-month mortality/hospitalization, both at admission and at hospital discharge, identifying ADHF patients at higher risk for future adverse events, suggesting the need for more attention and accuracy in the management of such patients. Furthermore, the prognostic value of PCT was comparable to NT-proBNP, being an established prognostic marker in HF patients. In line with our results, baseline PCT was significantly associated with 90-day all-cause mortality for patients with ADHF [20] . In the recently published paper, Naffaa et al. did not find significant prognostic value of PCT for death in ADHF on univariate Cox proportional hazards model, neither when they were forced into the multivariate model [21] . The potential difference with the results of our study may be explained by demographic differences of study populations (our study group is younger with higher percentage of males) in addition to much smaller sample size of only 33 patients in the study of Naffaa et al. Our results showed that serial in-hospital measurements of PCT in patients with ADHF provide additional prognostic information on the top of the first measurement, and at which time a reevaluation could be clinically useful. Change of PCT concentration after 3 days of treatment demonstrated additional prognostic information, in other words, patients with persistently increased or increasing PCT values had the worst 3-month outcome. However, we demonstrated that the use of combined biomarkers, in
656
Biomark. Med. (2015) 9(7)
other words, addition of baseline PCT to baseline NTproBNP does not add new prognostic information. Pucariello et al. [22] investigated serial changes of PCT levels in patients with cardiogenic shock following STelevation myocardial infarction with no clinical sign of infection. A progressive decrease of PCT concentration was observed in patients who survived, whereas the lack of changes in PCT values was documented in patients who died. Indeed, in a group of unselected critically ill patients, Jensen et al. [23] observed that increase of PCT was an independent predictor of 90-day survival and that the PCT day-by-day changes was able to identify critically ill patients at a higher risk of mortality. We found that almost every fifth patient in our study cohort without clinically overt infection had increased PCT concentration at baseline. Among them 86% were not treated with antibiotics having almost double rate of all-cause mortality/hospitalization at 3 months follow-up compared with the group treated with antibiotics without statistically valid conclusion due to the small sample size of antibiotic group. Maisel et al. in the BACH trial showed that PCT may aid in the decision to administer antibiotic therapy to patients presenting with dyspnea to emergency departments (35% of patients were affected by ADHF) in which clinical uncertainty exists regarding a superimposed bacterial infection [20] . Patients with a diagnosis of ADHF and an elevated PCT concentration (>0.21 ng/ml) had a worse outcome if not treated with antibiotics, while patients with low PCT levels (
Should procalcitonin be measured routinely in acute decompensated heart failure?
Aim: To elucidate the prognostic role of procalcitonin (PCT) in patients with acute decompensated heart failure (ADHF) without clinical signs of infection at admission. Materials & Methods: Serial measurements of PCT and NT-proBNP were performed in 168 patients, aged 68 ± 10 years with ADHF followed by 3-month outcome evaluation. Results: Cox regression analysis demonstrated significant predictive value of baseline PCT for all-cause death/hospitalization (area under the curve: 0.67; p = 0.013) at 90th day. The patients with persistently elevated PCT or with an increase during the first 72 h of hospitalization had the worst prognosis (p = 0.0002). Conclusion: Baseline and serial in-hospital measurements of PCT have significant prognostic properties for 3-month all-cause mortality/hospitalization in patients with ADHF without clinical signs of infection at admission. Keywords: ADHF • NT-proBNP • procalcitonin • prognosis • serial measurements
Heart failure (HF) is a major burden on the health systems. Episodes of acute decompensated heart failure (ADHF) are associated with significant short- and long-term morbidity and mortality. Patients hospitalized for heart failure had a threefold greater risk for all-cause mortality compared with patients who did not have a heart failure hospitalization after adjustment for known baseline predictors of death [1] . Thus, there is a need for earlier diagnosis, better risk stratification and cost-effective treatment to cut rates of HF hospitalization and to improve outcome. Biomarkers strive to undertake such tasks with evident fast development of biomarker research in this domain [2] . In recent years, procalcitonin (PCT) has emerged as a promising marker for the diagnosis of bacterial infections because higher levels are found in severe bacterial infections than in viral infections and nonspecific inflammatory diseases [3,4] . Thus, PCT may be used to support clinical decision making for the initiation and discontinuation of antibiotic therapy [5] . Few studies have investigated the role of PCT in patients with ADHF, although there was
10.2217/BMM.15.29 © 2015 Future Medicine Ltd
a slightly increased concentration of PCT in patients with severe acute heart failure, the results were not conclusive [6–8] . HF patients are characterized to have altered gut permeability, owing to bowel congestion or ischemia, which leads to invasion of bacterial endotoxins known to induce PCT secretion [9,10] . Serial in-hospital measurements of PCT have never been previously studied for its prognostic capacity in patients with ADHF without overt clinical manifestation of infection at hospital admission. The aim of the present study was therefore to perform a comparative evaluation of circulating serial in-hospital measurements of PCT and NTproBNP as predictors of 3-month all-cause mortality/hospitalizations in patients with ADHF without clinical sign of infection at admission. We have also investigated the relationship between PCT concentration, antibiotic treatment and outcome in patients with ADHF.
Goran Loncar*,†,1, Verena Tscholl†,2,3, Elvis Tahirovic2, Nikola Sekularac1, Almuth Marx4, Danilo Obradovic2, Jovan Veskovic2, Mitja Lainscak5,6, Stephan von Haehling7,8, Frank Edelmann2, Aleksandra Arandjelovic1,9, Svetlana Apostolovic10, Dragana Stanojevic10, Burkert Pieske2, Tobias Trippel‡,2 & Hans-Dirk Dungen‡,2 Department of Cardiology, Clinical Hospital Zvezdara, Belgrade, Serbia 2 Department of Cardiology, Campus Virchow, Charité Universitätsmedizin Berlin, Berlin, Germany 3 Department of Cardiology, Campus Benjamin Franklin, Charité Universitätsmedizin Berlin, Berlin, Germany 4 Marx-Statistics, Nuremberg, Germany 5 Departments of Cardiology & Research & Education, General Hospital Celje, Celje, Slovenia 6 Faculty of Medicine, University of Ljubljana, Ljubljana, Slovenia 7 Department of Cardiology, Applied Cachexia Research, Charité Medical School, Berlin, Germany 8 Center for Cardiovascular Research (CCR), Charité Medical School, Campus Mitte, Berlin, Germany 9 School of Medicine, University of Belgrade, Serbia 10 Clinic for Cardiovascular Diseases, Clinical Center Nis, Serbia *Author for correspondence: Tel.: +381 64 145 50 13 Fax: +381 11 380 94 76 [email protected] † G Loncar and V Tscholl share the first authorship ‡ T Trippel and HD Dungen share the last authorship 1
Materials & methods This multicenter, international (Serbia: 9 sites; Germany: 1 site), observational, pro-
Biomark. Med. (2015) 9(7), 651–659
part of
ISSN 1752-0363
651
Research Article Loncar, Tscholl, Tahirovic et al. spective study enrolled 185 patients with ADHF between June 2011 and June 2012. The diagnosis of ADHF was diagnosed according to the European Society of Cardiology criteria [11] . Patients were excluded if they had any of the following: temperature >38°C, sepsis or active infection, acute coronary syndrome, current or planned ultrafiltration, hemofiltration or dialysis, acute myocarditis or hypertrophic obstructive, restrictive or constrictive cardiomyopathy, malignant disease with a life expectancy of less than 2 years and autoimmune disease. After exclusion of 17 patients (15 patients withdrew informed consent, one patient excluded due to violation of inclusion criteria – not hospitalized for ADHF, one patient lost to follow-up), the final analysis included 168 patients. In Serbia 158 patients were enrolled, while in Germany ten patients. Patients were included within 24 h after hospitalization with outcome evaluation at 3-month follow-up. Clinical variables and blood samples were undertaken from every patient at baseline, after 72 h of hospital admission, at hospital discharge and at 3-months follow-up. To be discharged from the hospital, patients had to be in a recompensated dry state according to the judgment of treating physician. During follow-up, patients were treated according to HF guidelines [11] . The clinical check-up and treatment optimization was performed during visits at 14 and 42 days after discharge. The primary end point was defined as allcause mortality/ all-cause hospitalizations at 3-month follow-up. The study was conducted in accordance to the Declaration of Helsinki and approved by the ethics committee of all participating institutions. All participants gave written informed consent. Biomarker assessment
Blood samples were taken from an antecubital vein. The samples were centrifuged, transferred into tubes and frozen at -80°C until it was assayed. Samples were processed by personnel blinded from any patient data. PCT was measured with a time-resolved, amplified cryptate emission assay on the KRYPTOR System (B.R.A.H.M.S GmbH, Hennigsdorf, Germany). The PCT-sensitive KRYPTOR assay has a detection limit of 0.02 ng/ml and a functional assay sensitivity of 0.06 ng/ml [12] . The interassay coefficient of variation for concentration >0.3 ng/ml is 0.14, events = 3 of 5
10
PCT (BL) > 0.14, PCT (72 h) ≤ 0.14, events = 3 of 10 PCT (BL) > 0.14, PCT (72 h) > 0.14, events = 15 of 32
0 0
30
60
90
Time (days) Figure 4. Kaplan–Meier analysis of event-free survival according to the change of procalcitonin levels after 72 h of hospitalization.
In patients with ADHF it is obviously important to identify those with the highest risk in order to improve outcome. In our study, PCT had a prognostic value for 3-month mortality/hospitalization, both at admission and at hospital discharge, identifying ADHF patients at higher risk for future adverse events, suggesting the need for more attention and accuracy in the management of such patients. Furthermore, the prognostic value of PCT was comparable to NT-proBNP, being an established prognostic marker in HF patients. In line with our results, baseline PCT was significantly associated with 90-day all-cause mortality for patients with ADHF [20] . In the recently published paper, Naffaa et al. did not find significant prognostic value of PCT for death in ADHF on univariate Cox proportional hazards model, neither when they were forced into the multivariate model [21] . The potential difference with the results of our study may be explained by demographic differences of study populations (our study group is younger with higher percentage of males) in addition to much smaller sample size of only 33 patients in the study of Naffaa et al. Our results showed that serial in-hospital measurements of PCT in patients with ADHF provide additional prognostic information on the top of the first measurement, and at which time a reevaluation could be clinically useful. Change of PCT concentration after 3 days of treatment demonstrated additional prognostic information, in other words, patients with persistently increased or increasing PCT values had the worst 3-month outcome. However, we demonstrated that the use of combined biomarkers, in
656
Biomark. Med. (2015) 9(7)
other words, addition of baseline PCT to baseline NTproBNP does not add new prognostic information. Pucariello et al. [22] investigated serial changes of PCT levels in patients with cardiogenic shock following STelevation myocardial infarction with no clinical sign of infection. A progressive decrease of PCT concentration was observed in patients who survived, whereas the lack of changes in PCT values was documented in patients who died. Indeed, in a group of unselected critically ill patients, Jensen et al. [23] observed that increase of PCT was an independent predictor of 90-day survival and that the PCT day-by-day changes was able to identify critically ill patients at a higher risk of mortality. We found that almost every fifth patient in our study cohort without clinically overt infection had increased PCT concentration at baseline. Among them 86% were not treated with antibiotics having almost double rate of all-cause mortality/hospitalization at 3 months follow-up compared with the group treated with antibiotics without statistically valid conclusion due to the small sample size of antibiotic group. Maisel et al. in the BACH trial showed that PCT may aid in the decision to administer antibiotic therapy to patients presenting with dyspnea to emergency departments (35% of patients were affected by ADHF) in which clinical uncertainty exists regarding a superimposed bacterial infection [20] . Patients with a diagnosis of ADHF and an elevated PCT concentration (>0.21 ng/ml) had a worse outcome if not treated with antibiotics, while patients with low PCT levels (
