- Email: [email protected]
Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury
J Infect Chemother xxx (2015) 1e7
Contents lists available at ScienceDirect
Journal of Infection and Chemotherapy journal homepage: http://www.elsevier.com/locate/jic
Original article
Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury Yoshihiko Nakamura*, Akira Murai, Mariko Mizunuma, Daiki Ohta, Yasumasa Kawano, Norihiko Matsumoto, Takeshi Nishida, Hiroyasu Ishikura Department of Emergency and Critical Care Medicine, Faculty of Medicine, Fukuoka University, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 29 August 2014 Received in revised form 18 November 2014 Accepted 1 December 2014 Available online xxx
Introduction: There are few investigations regarding the relationships between procalcitonin (PCT) and the acute kidney injury (AKI) in the diagnosis of sepsis. The purpose of this study was to clarify the diagnostic accuracy of the use of PCT levels in patients with or without AKI. Methods: This study was conducted as a single-center retrospective study. We enrolled 393 patients in whom PCT were measured on admission. We grouped the patients into non-AKI and AKI, and those with AKI were classified according to the RIFLE criteria (Risk, Injury, Failure). The patients in each group were further classified into the sepsis and the non-sepsis group. We subsequently investigated the diagnostic accuracy of the PCT for detecting sepsis in these groups. Results: The levels of PCT were significantly higher in the sepsis group than in the non-sepsis group among the non-AKI and each AKI patients (p < 0.0001). The diagnostic accuracy of the PCT for detecting sepsis was determined according to a ROC analysis; AUC value was 0.958 in the non-AKI group, in the Risk, Injury and Failure groups were 0.888 and 0.917, 0.857, respectively. AUC value for non-AKI group was significantly different from that of Failure group (p < 0.05). Conclusions: In Failure AKI patients, the diagnostic accuracy of the PCT level is significantly lower than non-AKI patients. It is therefore suggested that we should be careful in using PCT value to diagnose sepsis in patients with Failure under RIFLE criteria. © 2014, Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Keywords: Procalcitonin Acute kidney injury Sepsis Diagnosis
1. Introduction Sepsis caused by bacteria is the major cause of death in intensive care unit (ICU) patients. Rapid initiation of the correct treatment is crucial important for improving sepsis condition [1,2]. Additionally, the most recent international sepsis guidelines entitled “Surviving Sepsis Campaign: International Guidelines for Management of Severe Sepsis and Septic Shock: 2012” recommend early diagnosis and treatment of sepsis to avoid multiple organ failure and other adverse outcomes [3]. In treatment of sepsis antibiotics have
* Corresponding author. Department of Emergency and Critical Care Medicine, Faculty of Medicine, Fukuoka University, 7-45-1 Nanakuma, Jonan-ku, Fukuoka 814-0180, Japan. Tel.: þ81 92 801 1011; fax: þ81 92 862 8330. E-mail addresses: [email protected] (Y. Nakamura), akiramurai.fuk@gmail. com (A. Murai), [email protected] (M. Mizunuma), [email protected] (D. Ohta), [email protected] (Y. Kawano), [email protected] (N. Matsumoto), [email protected] (T. Nishida), ishikurah@fukuoka-u. ac.jp (H. Ishikura).
already clear benefits, however potential complications from their inappropriate or prolonged using are also well known. Inappropriate or prolonged using of antibiotics lead to multidrug-resistant bacterial strains, allergic reactions, antibiotic-related colitis, and other adverse events [4e6]. From such situations it is emphasized that the early and accurate detection of sepsis are very important. Since the definition of systemic inflammatory response syndrome (SIRS) was proposed by the American College of Chest Physicians/Society of Critical Care Medicine (ACCP/SCCM) in 1991 [7], many clinical trials on sepsis diagnosis and treatment have been conducted. And sepsis was defined when patients met the criteria for SIRS and an infectious source was documented or strongly suspected based on clinical presentation. Blood culture is frequently used as the “gold standard” diagnostic method for sepsis. However, it usually takes 3e7 days to obtain blood culture results and frequently yields low positive results or low sensitivity [8]. Therefore, the general practical medical treatment used for sepsis is based on the doctor's own experience (empiric therapy).
http://dx.doi.org/10.1016/j.jiac.2014.12.001 1341-321X/© 2014, Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Nakamura Y, et al., Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury, J Infect Chemother (2015), http://dx.doi.org/10.1016/j.jiac.2014.12.001
2
Y. Nakamura et al. / J Infect Chemother xxx (2015) 1e7
Fig. 1. Study design.
Many trials have identified potential biomarkers. More than 170 biomarkers have been studied for use in evaluation of sepsis [9]. In 1993, procalcitonin (PCT) was first described as a marker elevated in bacterial infections [10]. In infectious conditions, PCT is released from nearly all tissues including lung, liver, kidney, pancreas, spleen, colon, and adipose tissues [11]. Currently, PCT is recognized as one of the suitable markers for diagnosis of sepsis or severe sepsis. In comparison to other markers which have traditionally been reported, PCT gives a high rate of specificity for sepsis diagnosis [12]. However, the concentration of PCT in the human blood is elevated in various conditions, such as in severe trauma, surgical invasive procedures, and critical burn injury, which leads to SIRS. So it is necessary to be aware of false-positive results [13]. In addition, it has been reported that renal function is a major determinant of PCT levels and thus different thresholds should be applied according to renal function impairment [14]. To the best of our knowledge, there are few studies investigating the relationship between PCT and acute kidney injury (AKI). So, the purpose of this study is to clarify the accuracy of diagnosing sepsis using the PCT levels according to AKI severity. 2. Materials and methods This is a retrospective study conducted at Fukuoka University Hospital, which is a 915-bed academic center with 34-bed ICU. The
ICU has an average admission of about 800 patients per year. The ICU admissions typically include 60% outpatients (including those transferred from the emergency department), 30% transferred from other hospitals, and 10% in-house patients (not including postoperated and newborn patients). The ratio of adult and pediatric (<16 years) patients was 93:7. The study was approved by the institutional review board of Fukuoka University Hospital according to the Declaration of Helsinki. The informed consent was waived in view of the retrospective and anonymous nature of the study. 2.1. Study population All the 2582 patients admitted to the ICU during the 3 years period (June 2010 through May 2013) were enrolled. Exclusion criteria were lack of PCT measurement on admission, cardiopulmonary arrest on arrival, age < 18, end-stage renal disease, local infection (excluded by sepsis criteria [7]), etiologic agent unknown, and patients with Mycobacterium tuberculosis, virus, fungus, Legionella pneumophila infection. In this study, we excluded patients with non-bacterial infection and those whose etiologic agent unknown, because PCT is known to be a specific biomarker for bacterial infection only. Moreover we classified the patients' condition of acute renal injury (AKI) using the RIFLE (Risk Injury Failure
Table 1 RIFLE criteria [11].
Risk Injury Failure Loss ESKD
GFR criteria
Urine output criteria
Increased serum creatinine 1.5 or GFR decrease >25% Increased serum creatinine 2 or GFR decrease >50% Increased serum creatinine 3 or GFR decrease >75%, serum creatinine S4 mg/dL (acute rise >0.5 mg/dL) Persistent ARF ¼ complete loss of kidney function > 4 weeks End stage kidney disease (>3 months)
Urine output <0.5 ml/kg/h 6 h Urine output <0.5 ml/kg/h 12 h Urine output <0.3 ml/kg/h 24hr or Anuria 12 h
RIFLE, Risk of renal dysfunction, Injury to the kidney, Failure of kidney function, Loss of kidney function and End-stage kidney disease; GFR, Glomerular Filtration Rate; ARF Acute Renal Failure.
Please cite this article in press as: Nakamura Y, et al., Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury, J Infect Chemother (2015), http://dx.doi.org/10.1016/j.jiac.2014.12.001
Y. Nakamura et al. / J Infect Chemother xxx (2015) 1e7
Loss of kidney function and End-stage kidney disease) criteria [15]. Next we assigned the patients either into a sepsis group or a nonsepsis group in each AKI stage and analyzed the accuracy of diagnosing sepsis based on the level of PCT. Study protocol was shown in Fig. 1.
3
2.5. Estimated glomerular filtration rate (eGFR) We calculated eGFR by using a Japanese equation [16]. Male: eGFR (mL/min/1.73 m2) ¼ 194 (Serum Cr)1.094 (Age)0.287. Female: eGFR (mL/min/1.73 m2) ¼ 194 (Serum Cr)1.094 (Age)0.287 0.739.
2.2. PCT measurement 2.6. Statistical analysis PCT concentrations were measured by Elecsys BRAHMS PCT assay (Roche Diagnostics, Japan). Serum sample was collected with EDTA-2K as an anticoagulant using a conventional blood collection tube (TERUMO, Japan).
2.3. Classification of acute kidney injury We used the RIFLE criteria [15] (Table 1) to determine the level of AKI and patients were classified according to the severity of AKI within 24 h after admission. RIFLE criteria consist of blood creatinine (Cr) and monitoring of 24 h urine output to classify patient's state of renal injury. In this study, baseline Cr was determined based on patient's available record prior to admission or the lowest level of Cr while under hospitalization. Urine output was measured within 24 h from admission. We used glomerular filtration rate (GFR) criteria or urine output criteria, whichever gives a more severe classification. All patients were evaluated using GFR criteria and urine output criteria.
2.4. Sepsis criteria Sepsis diagnosis (include severe sepsis and septic shock) was defined by ACCP/SCCM Consensus Conference Committee [7].
Table 2 Patients' clinical diagnosis. Sepsis or non-sepsis
Disease
n
Total
Sepsis
Pneumonia Peritonitis Skin and bone and soft tissue infection Urinary tract infection Bacteremia Biliary tract infection Brain abcess Endocarditis Enterocolitis Infected aneurysm Intra-abdominal abcess Liver abcess Phlebitis Upper respiratory tract infection Stroke Trauma Metabolism, Endocrine, Allergic disease Aortic disease Poisoning Heart failure Ischemic heart disease Epilepsy Gastro intestinal bleeding Interstitial pneumonitis Malignancy Heat stroke Liver disease Pancreatitis Burn Others
36 36 17 6 6 1 1 1 1 1 1 1 1 1 75 64 19 15 13 13 11 9 9 7 5 5 4 4 4 26
110
Non-sepsis
283
Comparisons between two groups were performed using the Wilcoxon test. Results are presented as the median (interquartile range: IQR). The levels of PCT according to RIFLE criteria were compared by KruskaleWallis and Bonferroni adjusted Wilcoxon rank sum post-hoc tests. Receiver operating characteristic (ROC) analysis was performed to evaluate the ability of PCT to discriminate between the groups. The Youden index was used to identify the cutoff points. Correlations were evaluated by the Spearman's rank test. Differences at p < 0.05 were considered to be statistically significant. For analysis, the commercially available software JMP® version 10 (SAS institute Japan, Tokyo) and MedCalc® version 14 (MedCalc software bvba, Belgium) were used. 3. Results 3.1. Enrolled patients We have registered 2582 patients from June 2010 to May 2013, and we excluded 2189 according to the exclusion criteria. Thus, a total of 393 patients (225 men and 168 women) were included in this analysis (Fig. 1). The median age of the patients was 65 (53e76) years. Clinical diagnosis of patients is shown in Table 2. The number of sepsis patients is 110 (sepsis, n ¼ 21; severe sepsis, n ¼ 31; septic shock, n ¼ 58). The microbiological examination results in sepsis patients are shown in Table 3. We evaluated 150 patients with AKI and 243 patients without AKI. The AKI group includes 48 patients who met the Risk, 46 who met the Injury, 56 who met the Failure category (Fig. 1). 3.2. PCT level of each AKI groups and severity of sepsis The median and IQR levels of PCT were 0.04 (0.02e0.11) ng/mL in the non-AKI with non-sepsis group, 2.71 (0.50e15.4) ng/mL in the non-AKI with sepsis group. And the levels of PCT were 0.10 (0.04e1.00) ng/mL in the Risk with non-sepsis group, 9.68 (0.74e34.61) ng/mL with sepsis group, 0.23 (0.05e0.79) ng/mL in the Injury with non-sepsis group, 17.10 (1.69e41.64) ng/mL with sepsis group, 1.75 (0.29e3.79) ng/mL in the Failure with non-sepsis group, 11.64 (3.69e98.50) ng/mL in the sepsis group. The PCT levels in the sepsis groups were significantly higher than the non-sepsis group in non-AKI, Risk, Injury, Failure (p < 0.0001, p < 0.0001, p < 0.0001, p < 0.0001, respectively) (Fig. 2). The PCT level of Failure was significantly higher than the level of non-AKI whether sepsis
Table 3 Results of microbiological examination in sepsis patients. Etiologic agent
n
Gram Gram Gram Gram Gram Total
52 30 25 2 1 110
negative rods positive coccus positive coccus and Gram negative rods positive and Fungus positive coccus and Gram negative rods and Fungus
Please cite this article in press as: Nakamura Y, et al., Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury, J Infect Chemother (2015), http://dx.doi.org/10.1016/j.jiac.2014.12.001
4
Y. Nakamura et al. / J Infect Chemother xxx (2015) 1e7 Table 5 Comparison of levels of PCT under each sepsis severity. Sepsis sevirity Sepsis PCT (ng/mL)
Severe sepsis
Septic shock
p value
2.48 (0.54e10.36) 2.94 (0.95e21.9) 17.86 (4.63e67.36)#$ <0.001
Data are presented as medians (interquartile range). Groups were compared by KruskaleWallis and Bonferroni adjusted Wilcoxon rank sum Post-hoc tests. PCT; procalcitonin. #, vs. Sepsis Bonferroni adjusted p < 0.01. $, vs. Severe sepsis Bonferroni adjusted p < 0.01.
negative correlation between PCT and eGFR in non-sepsis and sepsis patients (Fig. 4). 4. Discussion
Fig. 2. Comparisons of the levels of procalcitonin between non-sepsis and sepsis groups for each acute kidney injury severity. AKI, acute kidney injury. Groups were compared by Wilcoxon test.
patients or not (Table 4). In sepsis group, the PCT level of patients with septic shock was significantly higher than in those patients with sepsis and severe sepsis (Table 5).
3.3. Accuracy of sepsis diagnosis As for the accuracy of diagnosing sepsis based on the level of PCT in the ROC analysis, the area under the curve (AUC) was 0.958 [95% confidence interval (CI): 0.923e0.978] in the non-AKI, 0.888 (95% CI: 0.765e0.950) in the Risk, 0.917 (95% CI: 0.900e0.968) in the Injury, and 0.857 (95% CI: 0.730e0.930) in the Failure (Fig. 3). AUC value for non-AKI was significantly different from that of Failure (p < 0.05). The optimal cutoff value for the PCT levels for diagnosing sepsis was 0.28 ng/mL (sensitivity: 90.0%, specificity: 88.7%) in the non-AKI group. In addition, the optimal cutoff values for the PCT levels for diagnosing sepsis in the Risk, Injury and Failure groups were 0.42 (sensitivity: 95.0%, specificity: 64.8%), 0.57 (sensitivity: 96.4%, specificity: 77.2%), and 7.13 (sensitivity: 65.6%, specificity: 91.7%) ng/mL, respectively (Table 6).
3.4. Correlation between PCT and Cr, eGFR There was significant positive correlation between PCT and Cr in non-sepsis and sepsis patients. In addition, there is significant
The most common causes of AKI in critically ill patients are sepsis [17]. So it is necessary that we need to diagnose sepsis in AKI patients the fastest way possible. Recently Mehanic et al. [18] reported that the PCT is a reliable marker that contributes to the early diagnosis of invasive bacterial infections and evaluation of their severity and prognosis. However, status of kidney or liver may contribute to PCT clearance from the plasma, and severe dysfunction of these organs could influence the rate of PCT elimination. Because the molecular weight of PCT is low, approximately 13 kDa, it was suggested that at least partial renal elimination of this protein is possible. It also implies that plasma elimination during renal dysfunction may be delayed [19]. So, severe dysfunction of these organs may affect the rate of PCT elimination and the concentration of PCT level in plasma. This study demonstrates that the diagnostic accuracy of the PCT level of AKI patients with Failure patients is significantly lower than of non-AKI patients. In addition, there were significant positive correlation between PCT and Cr, and negative correlation between PCT and eGFR between each patient (see Fig. 4). Furthermore, the optimal cutoff value of PCT for detecting sepsis is higher in Failure patients (see Table 6). These results suggest that kidney is one of the responsible organs in eliminating PCT from blood. Therefore, it may not be a reliable indicator of sepsis in patients with a more advanced form of AKI, such as those classified under RIFLE criteria as having failure of kidney function. On the other hand, in septic shock patients, the level of PCT was significantly higher than that in the sepsis and severe sepsis patients (see Table 5). This result suggests that the cause of the rising levels of PCT in sepsis patients was not only AKI but also sepsis severity. In this study, there were 3 patients under the non-sepsis group with abnormal PCT levels (>10 ng/mL). One patient has severe trauma, another has rhabdomyolysis and the other was suffering from alcoholic ketoacidosis.
Table 4 Comparison of levels of PCT under each AKI severity. A. Non-sepsis patients. n AKI classification PCT (ng/mL)
213 non-AKI 0.04 (0.02e0.11)
28 Risk$ 0.10 (0.04e1.00)
18 Injury$ 0.23 (0.05e0.79)
24 Failure*y# 1.75 (0.29e3.79)
p value <0.0001
30 non-AKI 2.71 (0.50e15.4)
20 Risk 9.68 (0.74e34.61)
28 Injury 17.10 (1.69e41.65)
32 Failure$ 11.64 (3.69e98.50)
p value <0.01
B. Sepsis patients. n AKI classification PCT (ng/mL)
Data are presented as medians (interquartile range). Groups were compared by KruskaleWallis and Bonferroni adjusted Wilcoxon rank sum Post-hoc tests. PCT; procalcitonin, AKI; acute kidney injury. *, vs. non-AKI Bonferroni adjusted p < 0.001. y, vs. Risk Bonferroni adjusted p < 0.05. $, vs. non-AKI Bonferroni adjusted p < 0.01. #, vs. Injury Bonferroni adjusted p < 0.05.
Please cite this article in press as: Nakamura Y, et al., Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury, J Infect Chemother (2015), http://dx.doi.org/10.1016/j.jiac.2014.12.001
Y. Nakamura et al. / J Infect Chemother xxx (2015) 1e7
5
Fig. 3. Receiver operating characteristic (ROC) curves of procalcitonin in patients with non-sepsis and sepsis groups for each acute kidney injury severity.
PCT elevation may be caused by stress of trauma. And rhabdomyolysis and alcoholic ketoacidosis patients, we could not deny the possibility complicated with an infectious disease. But in these two cases bacterial culture were negative. Ricci et al. [20] reported that oliguria and RIFLE were the most frequent criteria used to define acute kidney injury, and they considered RIFLE criteria might address common uncertainties about AKI definition. In addition, Uchino et al. [21] reported that the RIFLE criteria useful in predicting hospital mortality. Additionally,
Monti et al. [22] reported that the first criteria to initiation renal replacement therapy (RRT) was the RIFLE criteria in 38% (Failure: 70%, Injury: 25%, Risk: 22%) in ICU. This result suggested that ICU patients of Failure according to RIFLE criteria patient were need to RRT. Our report is the first report to investigate the relationship between RIFLE criteria and PCT. et al. [23] reported that plasma levels of PCT were Opatrna significantly higher in patients undergoing peritoneal dialysis (PD) (median 0.33 ng/mL) compared with healthy volunteers (0.18 ng/ mL) (p < 0.001). Whereas Steinbach et al. [24] reported that PCT levels were significantly elevated in PD patients but not in patients with chronic renal failure or in those patients under hemodialysis (HD). Rosenthal et al. [25] reported that in patients under the HD treatment, when determined immediately before HD, PCT demonstrated a sensitivity of 89%, a specificity of 81%, indicating severe infection or sepsis (cutoff value 1.5 ng/mL). And they concluded that serum PCT is an accurate indicator of infection and sepsis in patients under the intermittent HD. With the exception of PD patients' PCT levels were not significantly affected by renal diseases or treatments but were markedly elevated in the presence of infections.
Table 6 Optimal cutoff values of PCT by the presence or absence of sepsis in each acute kidney injury classification.
Fig. 4. Correlations made with the Spearman's rank correlation of procalcitonin levels with non-sepsis patients of Cr (A) (r ¼ 0.328, p < 0.0001) and estimated glomerular filtration rate (B) (r ¼ 0.325, p < 0.0001). Correlations made with the Spearman's rank correlation of procalcitonin level with sepsis patients of creatinine (C) (r ¼ 0.280, p < 0.01) and estimated glomerular filtration rate (D) (r ¼ 0.297, p < 0.01). Cr, creatinine; eGFR, estimated glomerular filtration rate.
AKI classification
PCT optimal cutoff value (ng/mL)
Sensitivity (%)
Specificity (%)
Non-AKI Risk Injury Failure
0.28 0.42 0.57 7.13
90.0 95.0 96.4 65.6
88.7 64.8 77.2 91.7
AKI, acute kidney injury; PCT, Procalcitonin.
Please cite this article in press as: Nakamura Y, et al., Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury, J Infect Chemother (2015), http://dx.doi.org/10.1016/j.jiac.2014.12.001
6
Y. Nakamura et al. / J Infect Chemother xxx (2015) 1e7
Amour et al. [14] reported that in after major aortic surgery, when ROC analysis was performed in the non-renal dysfunction group and renal dysfunction group, the diagnostic accuracy of PCT was not significantly different between these two groups but the optimal cutoff value was significant different (non-renal dysfunction 0.81 ng/mL vs renal dysfunction 2.57 ng/mL, p < 0.05). And they concluded that PCT is a valuable marker of bacterial infections after major aortic surgery, but renal function is a major determinant of PCT levels. Thus different threshold should be applied according to renal function impairment. On the other hand, Meisener et al. [19] reported that renal elimination of PCT is not a major mechanism for PCT removal from the plasma. Although the plasma elimination rate may be prolonged up to 30e50% in some patients with renal dysfunction, clinical diagnostic decisions may not be severely influenced by this moderate prolongation of PCT elimination. And they concluded that PCT can be used diagnostically in patients with renal failure as well as in those with normal renal function. But our results showed that the diagnostic accuracy of the PCT levels for detecting sepsis decreased in the Failure group. In addition, our data indicated that PCT optimal cutoff values were different in Failure group as compared with non-AKI groups. Moreover, our study demonstrated that there were slightly significant positive correlation between PCT and Cr, and significant negative correlation between PCT and eGFR between non-sepsis and sepsis patients. These findings hereby suggest that the PCT elimination pathway involves the kidney and therefore is dependent on the renal function. Some limitations in our study deserve consideration. First, this study is retrospective study involving small sample in a single center. Second, we did not examine other sepsis biomarkers (such as Interleukin-6 and triggering receptor on myeloid cells-1 and sCD14 subtype; Presepsin [26]). Third, we could not include most of the patients due to lack of data on admission for PCT determination. Fourth, we examined the levels of PCT only upon admission at the ICU. Therefore, patients who were transferred from other departments, time-lag in evaluating their PCT levels may exist. In conclusion, our study demonstrated that in Failure AKI patients, the diagnostic accuracy of the PCT level is significantly lower than non-AKI patients. It is therefore suggested that we should be careful in using PCT level to diagnose sepsis in patients with Failure under RIFLE criteria. But this study only involved small sample of AKI patients. Further investigation is needed to clarify the thresholds in AKI patients. Competing interests The authors declare that they have no competing interests. Acknowledgments We sincerely thank Ms. Kanae Misumi of the Department of Emergency and Critical Care Medicine, Faculty of Medicine, Fukuoka University for her help in data encoding. Abbreviations ICU intensive care unit SIRS systemic inflammatory response syndrome ACCP/SCCM Chest Physicians/Society of Critical Care Medicine PCT procalcitonin AKI acute kidney injury RIFLE Risk Injury Failure Loss of kidney function and End-stage kidney disease Cr creatinine eGFR estimated glomerular filtration rate
IQR ROC AUC CI RRT PD HD
interquartile range receiver operating characteristic area under the curve confidence interval renal replacement therapy peritoneal dialysis hemodialysis
References [1] Garnacho-Montero J, Garcia-Garmendia JL, Barrero-Almodovar A, JimenezJimenez FJ, Perez-Paredes C, Ortiz-Leyba C. Impact of adequate empirical antibiotic therapy on the outcome of patients admitted to the intensive care unit with sepsis. Crit Care Med 2003;31:2742e51. [2] Alberti C, Brun-Buisson C, Burchardi H, Martin C, Goodman S, Artigas A, et al. Epidemiology of sepsis and infection in ICU patients from an international multicentre cohort study. Intensive Care Med 2002;28:108e21. [3] Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock. Crit Care Med 2012;2013(41):580e637. [4] Chastre J, Wolff M, Fagon JY, Chevret S, Thomas F, Wermert D, et al. Comparison of 8 vs 15 days of antibiotic therapy for ventilator-associated pneumonia in adults: a randomized trial. J Am Med Assoc 2003;290: 2588e98. [5] Singh N, Rogers P, Atwood CW, Wagener MM, Yu VL. Short-course empiric antibiotic therapy for patients with pulmonary infiltrates in the intensive care unit. A proposed solution for indiscriminate antibiotic prescription. Am J Respir Crit Care Med 2000;162:505e11. [6] Yu VL, Singh N. Excessive antimicrobial usage causes measurable harm to patients with suspected ventilator-associated pneumonia. Intensive Care Med 2004;30:735e8. [7] Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest 1992;101: 1644e55. [8] Rangel-Frausto MS, Pittet D, Costigan M, Hwang T, Davis CS, Wenzel RP. The natural history of the systemic inflammatory response syndrome (SIRS). A prospective study. J Am Med Assoc 1995;273:117e23. [9] Pierrakos C, Vincent JL. Sepsis biomarkers: a review. Crit Care 2010;14:R15. [10] Assicot M, Gendrel D, Carsin H, Raymond J, Guilbaud J, Bohuon C. High serum procalcitonin concentrations in patients with sepsis and infection. Lancet 1993;341:515e8. [11] Cho SY, Choi JH. Biomarkers of sepsis. Infect Chemother 2014;46:1e12. [12] Herzum I, Renz H. Inflammatory markers in SIRS, sepsis and septic shock. Curr Med Chem 2008;15:581e7. [13] Christ-Crain M, Muller B. Procalcitonin in bacterial infections e hype, hope, more or less? Swiss Med Wkly 2005;135:451e60. [14] Amour J, Birenbaum A, Langeron O, Le Manach Y, Bertrand M, Coriat P, et al. Influence of renal dysfunction on the accuracy of procalcitonin for the diagnosis of postoperative infection after vascular surgery. Crit Care Med 2008;36: 1147e54. [15] Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P. Acute Dialysis Quality Initiative w. Acute renal failure e definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 2004;8:R204e12. [16] Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, et al. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis 2009;53: 982e92. [17] Murugan R, Kellum JA. Acute kidney injury: what's the prognosis? Nat Rev Nephrol 2011;7:209e17. [18] Mehanic S, Baljic R. The importance of serum procalcitonin in diagnosis and treatment of serious bacterial infections and sepsis. Mater Sociomed 2013;25: 277e81. [19] Meisner M, Lohs T, Huettemann E, Schmidt J, Hueller M, Reinhart K. The plasma elimination rate and urinary secretion of procalcitonin in patients with normal and impaired renal function. Eur J Anaesthesiol 2001;18: 79e87. [20] Ricci Z, Ronco C, D'Amico G, De Felice R, Rossi S, Bolgan I, et al. Practice patterns in the management of acute renal failure in the critically ill patient: an international survey. Nephrol Dial Transplant 2006;21: 690e6. [21] Uchino S, Bellomo R, Goldsmith D, Bates S, Ronco C. An assessment of the RIFLE criteria for acute renal failure in hospitalized patients. Crit Care Med 2006;34:1913e7. [22] Monti G, Herrera M, Kindgen-Milles D, Marinho A, Cruz D, Mariano F, et al. The DOse REsponse multicentre international Collaborative Initiative (DO-REMI). Contrib Nephrol 2007;156:434e43. [23] Opatrna S, Klaboch J, Opatrny Jr K, Holubec L, Tomsu M, Sefrna F, et al. Procalcitonin levels in peritoneal dialysis patients. Perit Dial Int 2005;25:470e2.
Please cite this article in press as: Nakamura Y, et al., Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury, J Infect Chemother (2015), http://dx.doi.org/10.1016/j.jiac.2014.12.001
Y. Nakamura et al. / J Infect Chemother xxx (2015) 1e7 [24] Steinbach G, Bolke E, Grunert A, Storck M, Orth K. Procalcitonin in patients with acute and chronic renal insufficiency. Wien Klin Wochenschr 2004;116: 849e53. [25] Herget-Rosenthal S, Marggraf G, Pietruck F, Husing J, Strupat M, Philipp T, et al. Procalcitonin for accurate detection of infection in haemodialysis. Nephrol Dial Transplant 2001;16:975e9.
7
[26] Shirakawa K, Naitou K, Hirose J, Takahashi T, Furusako S. Presepsin (sCD14ST): development and evaluation of one-step ELISA with a new standard that is similar to the form of presepsin in septic patients. Clin Chem Lab Med 2011;49:937e9.
Please cite this article in press as: Nakamura Y, et al., Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury, J Infect Chemother (2015), http://dx.doi.org/10.1016/j.jiac.2014.12.001
Contents lists available at ScienceDirect
Journal of Infection and Chemotherapy journal homepage: http://www.elsevier.com/locate/jic
Original article
Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury Yoshihiko Nakamura*, Akira Murai, Mariko Mizunuma, Daiki Ohta, Yasumasa Kawano, Norihiko Matsumoto, Takeshi Nishida, Hiroyasu Ishikura Department of Emergency and Critical Care Medicine, Faculty of Medicine, Fukuoka University, Japan
a r t i c l e i n f o
a b s t r a c t
Article history: Received 29 August 2014 Received in revised form 18 November 2014 Accepted 1 December 2014 Available online xxx
Introduction: There are few investigations regarding the relationships between procalcitonin (PCT) and the acute kidney injury (AKI) in the diagnosis of sepsis. The purpose of this study was to clarify the diagnostic accuracy of the use of PCT levels in patients with or without AKI. Methods: This study was conducted as a single-center retrospective study. We enrolled 393 patients in whom PCT were measured on admission. We grouped the patients into non-AKI and AKI, and those with AKI were classified according to the RIFLE criteria (Risk, Injury, Failure). The patients in each group were further classified into the sepsis and the non-sepsis group. We subsequently investigated the diagnostic accuracy of the PCT for detecting sepsis in these groups. Results: The levels of PCT were significantly higher in the sepsis group than in the non-sepsis group among the non-AKI and each AKI patients (p < 0.0001). The diagnostic accuracy of the PCT for detecting sepsis was determined according to a ROC analysis; AUC value was 0.958 in the non-AKI group, in the Risk, Injury and Failure groups were 0.888 and 0.917, 0.857, respectively. AUC value for non-AKI group was significantly different from that of Failure group (p < 0.05). Conclusions: In Failure AKI patients, the diagnostic accuracy of the PCT level is significantly lower than non-AKI patients. It is therefore suggested that we should be careful in using PCT value to diagnose sepsis in patients with Failure under RIFLE criteria. © 2014, Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Keywords: Procalcitonin Acute kidney injury Sepsis Diagnosis
1. Introduction Sepsis caused by bacteria is the major cause of death in intensive care unit (ICU) patients. Rapid initiation of the correct treatment is crucial important for improving sepsis condition [1,2]. Additionally, the most recent international sepsis guidelines entitled “Surviving Sepsis Campaign: International Guidelines for Management of Severe Sepsis and Septic Shock: 2012” recommend early diagnosis and treatment of sepsis to avoid multiple organ failure and other adverse outcomes [3]. In treatment of sepsis antibiotics have
* Corresponding author. Department of Emergency and Critical Care Medicine, Faculty of Medicine, Fukuoka University, 7-45-1 Nanakuma, Jonan-ku, Fukuoka 814-0180, Japan. Tel.: þ81 92 801 1011; fax: þ81 92 862 8330. E-mail addresses: [email protected] (Y. Nakamura), akiramurai.fuk@gmail. com (A. Murai), [email protected] (M. Mizunuma), [email protected] (D. Ohta), [email protected] (Y. Kawano), [email protected] (N. Matsumoto), [email protected] (T. Nishida), ishikurah@fukuoka-u. ac.jp (H. Ishikura).
already clear benefits, however potential complications from their inappropriate or prolonged using are also well known. Inappropriate or prolonged using of antibiotics lead to multidrug-resistant bacterial strains, allergic reactions, antibiotic-related colitis, and other adverse events [4e6]. From such situations it is emphasized that the early and accurate detection of sepsis are very important. Since the definition of systemic inflammatory response syndrome (SIRS) was proposed by the American College of Chest Physicians/Society of Critical Care Medicine (ACCP/SCCM) in 1991 [7], many clinical trials on sepsis diagnosis and treatment have been conducted. And sepsis was defined when patients met the criteria for SIRS and an infectious source was documented or strongly suspected based on clinical presentation. Blood culture is frequently used as the “gold standard” diagnostic method for sepsis. However, it usually takes 3e7 days to obtain blood culture results and frequently yields low positive results or low sensitivity [8]. Therefore, the general practical medical treatment used for sepsis is based on the doctor's own experience (empiric therapy).
http://dx.doi.org/10.1016/j.jiac.2014.12.001 1341-321X/© 2014, Japanese Society of Chemotherapy and The Japanese Association for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
Please cite this article in press as: Nakamura Y, et al., Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury, J Infect Chemother (2015), http://dx.doi.org/10.1016/j.jiac.2014.12.001
2
Y. Nakamura et al. / J Infect Chemother xxx (2015) 1e7
Fig. 1. Study design.
Many trials have identified potential biomarkers. More than 170 biomarkers have been studied for use in evaluation of sepsis [9]. In 1993, procalcitonin (PCT) was first described as a marker elevated in bacterial infections [10]. In infectious conditions, PCT is released from nearly all tissues including lung, liver, kidney, pancreas, spleen, colon, and adipose tissues [11]. Currently, PCT is recognized as one of the suitable markers for diagnosis of sepsis or severe sepsis. In comparison to other markers which have traditionally been reported, PCT gives a high rate of specificity for sepsis diagnosis [12]. However, the concentration of PCT in the human blood is elevated in various conditions, such as in severe trauma, surgical invasive procedures, and critical burn injury, which leads to SIRS. So it is necessary to be aware of false-positive results [13]. In addition, it has been reported that renal function is a major determinant of PCT levels and thus different thresholds should be applied according to renal function impairment [14]. To the best of our knowledge, there are few studies investigating the relationship between PCT and acute kidney injury (AKI). So, the purpose of this study is to clarify the accuracy of diagnosing sepsis using the PCT levels according to AKI severity. 2. Materials and methods This is a retrospective study conducted at Fukuoka University Hospital, which is a 915-bed academic center with 34-bed ICU. The
ICU has an average admission of about 800 patients per year. The ICU admissions typically include 60% outpatients (including those transferred from the emergency department), 30% transferred from other hospitals, and 10% in-house patients (not including postoperated and newborn patients). The ratio of adult and pediatric (<16 years) patients was 93:7. The study was approved by the institutional review board of Fukuoka University Hospital according to the Declaration of Helsinki. The informed consent was waived in view of the retrospective and anonymous nature of the study. 2.1. Study population All the 2582 patients admitted to the ICU during the 3 years period (June 2010 through May 2013) were enrolled. Exclusion criteria were lack of PCT measurement on admission, cardiopulmonary arrest on arrival, age < 18, end-stage renal disease, local infection (excluded by sepsis criteria [7]), etiologic agent unknown, and patients with Mycobacterium tuberculosis, virus, fungus, Legionella pneumophila infection. In this study, we excluded patients with non-bacterial infection and those whose etiologic agent unknown, because PCT is known to be a specific biomarker for bacterial infection only. Moreover we classified the patients' condition of acute renal injury (AKI) using the RIFLE (Risk Injury Failure
Table 1 RIFLE criteria [11].
Risk Injury Failure Loss ESKD
GFR criteria
Urine output criteria
Increased serum creatinine 1.5 or GFR decrease >25% Increased serum creatinine 2 or GFR decrease >50% Increased serum creatinine 3 or GFR decrease >75%, serum creatinine S4 mg/dL (acute rise >0.5 mg/dL) Persistent ARF ¼ complete loss of kidney function > 4 weeks End stage kidney disease (>3 months)
Urine output <0.5 ml/kg/h 6 h Urine output <0.5 ml/kg/h 12 h Urine output <0.3 ml/kg/h 24hr or Anuria 12 h
RIFLE, Risk of renal dysfunction, Injury to the kidney, Failure of kidney function, Loss of kidney function and End-stage kidney disease; GFR, Glomerular Filtration Rate; ARF Acute Renal Failure.
Please cite this article in press as: Nakamura Y, et al., Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury, J Infect Chemother (2015), http://dx.doi.org/10.1016/j.jiac.2014.12.001
Y. Nakamura et al. / J Infect Chemother xxx (2015) 1e7
Loss of kidney function and End-stage kidney disease) criteria [15]. Next we assigned the patients either into a sepsis group or a nonsepsis group in each AKI stage and analyzed the accuracy of diagnosing sepsis based on the level of PCT. Study protocol was shown in Fig. 1.
3
2.5. Estimated glomerular filtration rate (eGFR) We calculated eGFR by using a Japanese equation [16]. Male: eGFR (mL/min/1.73 m2) ¼ 194 (Serum Cr)1.094 (Age)0.287. Female: eGFR (mL/min/1.73 m2) ¼ 194 (Serum Cr)1.094 (Age)0.287 0.739.
2.2. PCT measurement 2.6. Statistical analysis PCT concentrations were measured by Elecsys BRAHMS PCT assay (Roche Diagnostics, Japan). Serum sample was collected with EDTA-2K as an anticoagulant using a conventional blood collection tube (TERUMO, Japan).
2.3. Classification of acute kidney injury We used the RIFLE criteria [15] (Table 1) to determine the level of AKI and patients were classified according to the severity of AKI within 24 h after admission. RIFLE criteria consist of blood creatinine (Cr) and monitoring of 24 h urine output to classify patient's state of renal injury. In this study, baseline Cr was determined based on patient's available record prior to admission or the lowest level of Cr while under hospitalization. Urine output was measured within 24 h from admission. We used glomerular filtration rate (GFR) criteria or urine output criteria, whichever gives a more severe classification. All patients were evaluated using GFR criteria and urine output criteria.
2.4. Sepsis criteria Sepsis diagnosis (include severe sepsis and septic shock) was defined by ACCP/SCCM Consensus Conference Committee [7].
Table 2 Patients' clinical diagnosis. Sepsis or non-sepsis
Disease
n
Total
Sepsis
Pneumonia Peritonitis Skin and bone and soft tissue infection Urinary tract infection Bacteremia Biliary tract infection Brain abcess Endocarditis Enterocolitis Infected aneurysm Intra-abdominal abcess Liver abcess Phlebitis Upper respiratory tract infection Stroke Trauma Metabolism, Endocrine, Allergic disease Aortic disease Poisoning Heart failure Ischemic heart disease Epilepsy Gastro intestinal bleeding Interstitial pneumonitis Malignancy Heat stroke Liver disease Pancreatitis Burn Others
36 36 17 6 6 1 1 1 1 1 1 1 1 1 75 64 19 15 13 13 11 9 9 7 5 5 4 4 4 26
110
Non-sepsis
283
Comparisons between two groups were performed using the Wilcoxon test. Results are presented as the median (interquartile range: IQR). The levels of PCT according to RIFLE criteria were compared by KruskaleWallis and Bonferroni adjusted Wilcoxon rank sum post-hoc tests. Receiver operating characteristic (ROC) analysis was performed to evaluate the ability of PCT to discriminate between the groups. The Youden index was used to identify the cutoff points. Correlations were evaluated by the Spearman's rank test. Differences at p < 0.05 were considered to be statistically significant. For analysis, the commercially available software JMP® version 10 (SAS institute Japan, Tokyo) and MedCalc® version 14 (MedCalc software bvba, Belgium) were used. 3. Results 3.1. Enrolled patients We have registered 2582 patients from June 2010 to May 2013, and we excluded 2189 according to the exclusion criteria. Thus, a total of 393 patients (225 men and 168 women) were included in this analysis (Fig. 1). The median age of the patients was 65 (53e76) years. Clinical diagnosis of patients is shown in Table 2. The number of sepsis patients is 110 (sepsis, n ¼ 21; severe sepsis, n ¼ 31; septic shock, n ¼ 58). The microbiological examination results in sepsis patients are shown in Table 3. We evaluated 150 patients with AKI and 243 patients without AKI. The AKI group includes 48 patients who met the Risk, 46 who met the Injury, 56 who met the Failure category (Fig. 1). 3.2. PCT level of each AKI groups and severity of sepsis The median and IQR levels of PCT were 0.04 (0.02e0.11) ng/mL in the non-AKI with non-sepsis group, 2.71 (0.50e15.4) ng/mL in the non-AKI with sepsis group. And the levels of PCT were 0.10 (0.04e1.00) ng/mL in the Risk with non-sepsis group, 9.68 (0.74e34.61) ng/mL with sepsis group, 0.23 (0.05e0.79) ng/mL in the Injury with non-sepsis group, 17.10 (1.69e41.64) ng/mL with sepsis group, 1.75 (0.29e3.79) ng/mL in the Failure with non-sepsis group, 11.64 (3.69e98.50) ng/mL in the sepsis group. The PCT levels in the sepsis groups were significantly higher than the non-sepsis group in non-AKI, Risk, Injury, Failure (p < 0.0001, p < 0.0001, p < 0.0001, p < 0.0001, respectively) (Fig. 2). The PCT level of Failure was significantly higher than the level of non-AKI whether sepsis
Table 3 Results of microbiological examination in sepsis patients. Etiologic agent
n
Gram Gram Gram Gram Gram Total
52 30 25 2 1 110
negative rods positive coccus positive coccus and Gram negative rods positive and Fungus positive coccus and Gram negative rods and Fungus
Please cite this article in press as: Nakamura Y, et al., Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury, J Infect Chemother (2015), http://dx.doi.org/10.1016/j.jiac.2014.12.001
4
Y. Nakamura et al. / J Infect Chemother xxx (2015) 1e7 Table 5 Comparison of levels of PCT under each sepsis severity. Sepsis sevirity Sepsis PCT (ng/mL)
Severe sepsis
Septic shock
p value
2.48 (0.54e10.36) 2.94 (0.95e21.9) 17.86 (4.63e67.36)#$ <0.001
Data are presented as medians (interquartile range). Groups were compared by KruskaleWallis and Bonferroni adjusted Wilcoxon rank sum Post-hoc tests. PCT; procalcitonin. #, vs. Sepsis Bonferroni adjusted p < 0.01. $, vs. Severe sepsis Bonferroni adjusted p < 0.01.
negative correlation between PCT and eGFR in non-sepsis and sepsis patients (Fig. 4). 4. Discussion
Fig. 2. Comparisons of the levels of procalcitonin between non-sepsis and sepsis groups for each acute kidney injury severity. AKI, acute kidney injury. Groups were compared by Wilcoxon test.
patients or not (Table 4). In sepsis group, the PCT level of patients with septic shock was significantly higher than in those patients with sepsis and severe sepsis (Table 5).
3.3. Accuracy of sepsis diagnosis As for the accuracy of diagnosing sepsis based on the level of PCT in the ROC analysis, the area under the curve (AUC) was 0.958 [95% confidence interval (CI): 0.923e0.978] in the non-AKI, 0.888 (95% CI: 0.765e0.950) in the Risk, 0.917 (95% CI: 0.900e0.968) in the Injury, and 0.857 (95% CI: 0.730e0.930) in the Failure (Fig. 3). AUC value for non-AKI was significantly different from that of Failure (p < 0.05). The optimal cutoff value for the PCT levels for diagnosing sepsis was 0.28 ng/mL (sensitivity: 90.0%, specificity: 88.7%) in the non-AKI group. In addition, the optimal cutoff values for the PCT levels for diagnosing sepsis in the Risk, Injury and Failure groups were 0.42 (sensitivity: 95.0%, specificity: 64.8%), 0.57 (sensitivity: 96.4%, specificity: 77.2%), and 7.13 (sensitivity: 65.6%, specificity: 91.7%) ng/mL, respectively (Table 6).
3.4. Correlation between PCT and Cr, eGFR There was significant positive correlation between PCT and Cr in non-sepsis and sepsis patients. In addition, there is significant
The most common causes of AKI in critically ill patients are sepsis [17]. So it is necessary that we need to diagnose sepsis in AKI patients the fastest way possible. Recently Mehanic et al. [18] reported that the PCT is a reliable marker that contributes to the early diagnosis of invasive bacterial infections and evaluation of their severity and prognosis. However, status of kidney or liver may contribute to PCT clearance from the plasma, and severe dysfunction of these organs could influence the rate of PCT elimination. Because the molecular weight of PCT is low, approximately 13 kDa, it was suggested that at least partial renal elimination of this protein is possible. It also implies that plasma elimination during renal dysfunction may be delayed [19]. So, severe dysfunction of these organs may affect the rate of PCT elimination and the concentration of PCT level in plasma. This study demonstrates that the diagnostic accuracy of the PCT level of AKI patients with Failure patients is significantly lower than of non-AKI patients. In addition, there were significant positive correlation between PCT and Cr, and negative correlation between PCT and eGFR between each patient (see Fig. 4). Furthermore, the optimal cutoff value of PCT for detecting sepsis is higher in Failure patients (see Table 6). These results suggest that kidney is one of the responsible organs in eliminating PCT from blood. Therefore, it may not be a reliable indicator of sepsis in patients with a more advanced form of AKI, such as those classified under RIFLE criteria as having failure of kidney function. On the other hand, in septic shock patients, the level of PCT was significantly higher than that in the sepsis and severe sepsis patients (see Table 5). This result suggests that the cause of the rising levels of PCT in sepsis patients was not only AKI but also sepsis severity. In this study, there were 3 patients under the non-sepsis group with abnormal PCT levels (>10 ng/mL). One patient has severe trauma, another has rhabdomyolysis and the other was suffering from alcoholic ketoacidosis.
Table 4 Comparison of levels of PCT under each AKI severity. A. Non-sepsis patients. n AKI classification PCT (ng/mL)
213 non-AKI 0.04 (0.02e0.11)
28 Risk$ 0.10 (0.04e1.00)
18 Injury$ 0.23 (0.05e0.79)
24 Failure*y# 1.75 (0.29e3.79)
p value <0.0001
30 non-AKI 2.71 (0.50e15.4)
20 Risk 9.68 (0.74e34.61)
28 Injury 17.10 (1.69e41.65)
32 Failure$ 11.64 (3.69e98.50)
p value <0.01
B. Sepsis patients. n AKI classification PCT (ng/mL)
Data are presented as medians (interquartile range). Groups were compared by KruskaleWallis and Bonferroni adjusted Wilcoxon rank sum Post-hoc tests. PCT; procalcitonin, AKI; acute kidney injury. *, vs. non-AKI Bonferroni adjusted p < 0.001. y, vs. Risk Bonferroni adjusted p < 0.05. $, vs. non-AKI Bonferroni adjusted p < 0.01. #, vs. Injury Bonferroni adjusted p < 0.05.
Please cite this article in press as: Nakamura Y, et al., Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury, J Infect Chemother (2015), http://dx.doi.org/10.1016/j.jiac.2014.12.001
Y. Nakamura et al. / J Infect Chemother xxx (2015) 1e7
5
Fig. 3. Receiver operating characteristic (ROC) curves of procalcitonin in patients with non-sepsis and sepsis groups for each acute kidney injury severity.
PCT elevation may be caused by stress of trauma. And rhabdomyolysis and alcoholic ketoacidosis patients, we could not deny the possibility complicated with an infectious disease. But in these two cases bacterial culture were negative. Ricci et al. [20] reported that oliguria and RIFLE were the most frequent criteria used to define acute kidney injury, and they considered RIFLE criteria might address common uncertainties about AKI definition. In addition, Uchino et al. [21] reported that the RIFLE criteria useful in predicting hospital mortality. Additionally,
Monti et al. [22] reported that the first criteria to initiation renal replacement therapy (RRT) was the RIFLE criteria in 38% (Failure: 70%, Injury: 25%, Risk: 22%) in ICU. This result suggested that ICU patients of Failure according to RIFLE criteria patient were need to RRT. Our report is the first report to investigate the relationship between RIFLE criteria and PCT. et al. [23] reported that plasma levels of PCT were Opatrna significantly higher in patients undergoing peritoneal dialysis (PD) (median 0.33 ng/mL) compared with healthy volunteers (0.18 ng/ mL) (p < 0.001). Whereas Steinbach et al. [24] reported that PCT levels were significantly elevated in PD patients but not in patients with chronic renal failure or in those patients under hemodialysis (HD). Rosenthal et al. [25] reported that in patients under the HD treatment, when determined immediately before HD, PCT demonstrated a sensitivity of 89%, a specificity of 81%, indicating severe infection or sepsis (cutoff value 1.5 ng/mL). And they concluded that serum PCT is an accurate indicator of infection and sepsis in patients under the intermittent HD. With the exception of PD patients' PCT levels were not significantly affected by renal diseases or treatments but were markedly elevated in the presence of infections.
Table 6 Optimal cutoff values of PCT by the presence or absence of sepsis in each acute kidney injury classification.
Fig. 4. Correlations made with the Spearman's rank correlation of procalcitonin levels with non-sepsis patients of Cr (A) (r ¼ 0.328, p < 0.0001) and estimated glomerular filtration rate (B) (r ¼ 0.325, p < 0.0001). Correlations made with the Spearman's rank correlation of procalcitonin level with sepsis patients of creatinine (C) (r ¼ 0.280, p < 0.01) and estimated glomerular filtration rate (D) (r ¼ 0.297, p < 0.01). Cr, creatinine; eGFR, estimated glomerular filtration rate.
AKI classification
PCT optimal cutoff value (ng/mL)
Sensitivity (%)
Specificity (%)
Non-AKI Risk Injury Failure
0.28 0.42 0.57 7.13
90.0 95.0 96.4 65.6
88.7 64.8 77.2 91.7
AKI, acute kidney injury; PCT, Procalcitonin.
Please cite this article in press as: Nakamura Y, et al., Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury, J Infect Chemother (2015), http://dx.doi.org/10.1016/j.jiac.2014.12.001
6
Y. Nakamura et al. / J Infect Chemother xxx (2015) 1e7
Amour et al. [14] reported that in after major aortic surgery, when ROC analysis was performed in the non-renal dysfunction group and renal dysfunction group, the diagnostic accuracy of PCT was not significantly different between these two groups but the optimal cutoff value was significant different (non-renal dysfunction 0.81 ng/mL vs renal dysfunction 2.57 ng/mL, p < 0.05). And they concluded that PCT is a valuable marker of bacterial infections after major aortic surgery, but renal function is a major determinant of PCT levels. Thus different threshold should be applied according to renal function impairment. On the other hand, Meisener et al. [19] reported that renal elimination of PCT is not a major mechanism for PCT removal from the plasma. Although the plasma elimination rate may be prolonged up to 30e50% in some patients with renal dysfunction, clinical diagnostic decisions may not be severely influenced by this moderate prolongation of PCT elimination. And they concluded that PCT can be used diagnostically in patients with renal failure as well as in those with normal renal function. But our results showed that the diagnostic accuracy of the PCT levels for detecting sepsis decreased in the Failure group. In addition, our data indicated that PCT optimal cutoff values were different in Failure group as compared with non-AKI groups. Moreover, our study demonstrated that there were slightly significant positive correlation between PCT and Cr, and significant negative correlation between PCT and eGFR between non-sepsis and sepsis patients. These findings hereby suggest that the PCT elimination pathway involves the kidney and therefore is dependent on the renal function. Some limitations in our study deserve consideration. First, this study is retrospective study involving small sample in a single center. Second, we did not examine other sepsis biomarkers (such as Interleukin-6 and triggering receptor on myeloid cells-1 and sCD14 subtype; Presepsin [26]). Third, we could not include most of the patients due to lack of data on admission for PCT determination. Fourth, we examined the levels of PCT only upon admission at the ICU. Therefore, patients who were transferred from other departments, time-lag in evaluating their PCT levels may exist. In conclusion, our study demonstrated that in Failure AKI patients, the diagnostic accuracy of the PCT level is significantly lower than non-AKI patients. It is therefore suggested that we should be careful in using PCT level to diagnose sepsis in patients with Failure under RIFLE criteria. But this study only involved small sample of AKI patients. Further investigation is needed to clarify the thresholds in AKI patients. Competing interests The authors declare that they have no competing interests. Acknowledgments We sincerely thank Ms. Kanae Misumi of the Department of Emergency and Critical Care Medicine, Faculty of Medicine, Fukuoka University for her help in data encoding. Abbreviations ICU intensive care unit SIRS systemic inflammatory response syndrome ACCP/SCCM Chest Physicians/Society of Critical Care Medicine PCT procalcitonin AKI acute kidney injury RIFLE Risk Injury Failure Loss of kidney function and End-stage kidney disease Cr creatinine eGFR estimated glomerular filtration rate
IQR ROC AUC CI RRT PD HD
interquartile range receiver operating characteristic area under the curve confidence interval renal replacement therapy peritoneal dialysis hemodialysis
References [1] Garnacho-Montero J, Garcia-Garmendia JL, Barrero-Almodovar A, JimenezJimenez FJ, Perez-Paredes C, Ortiz-Leyba C. Impact of adequate empirical antibiotic therapy on the outcome of patients admitted to the intensive care unit with sepsis. Crit Care Med 2003;31:2742e51. [2] Alberti C, Brun-Buisson C, Burchardi H, Martin C, Goodman S, Artigas A, et al. Epidemiology of sepsis and infection in ICU patients from an international multicentre cohort study. Intensive Care Med 2002;28:108e21. [3] Dellinger RP, Levy MM, Rhodes A, Annane D, Gerlach H, Opal SM, et al. Surviving sepsis campaign: international guidelines for management of severe sepsis and septic shock. Crit Care Med 2012;2013(41):580e637. [4] Chastre J, Wolff M, Fagon JY, Chevret S, Thomas F, Wermert D, et al. Comparison of 8 vs 15 days of antibiotic therapy for ventilator-associated pneumonia in adults: a randomized trial. J Am Med Assoc 2003;290: 2588e98. [5] Singh N, Rogers P, Atwood CW, Wagener MM, Yu VL. Short-course empiric antibiotic therapy for patients with pulmonary infiltrates in the intensive care unit. A proposed solution for indiscriminate antibiotic prescription. Am J Respir Crit Care Med 2000;162:505e11. [6] Yu VL, Singh N. Excessive antimicrobial usage causes measurable harm to patients with suspected ventilator-associated pneumonia. Intensive Care Med 2004;30:735e8. [7] Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, et al. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest 1992;101: 1644e55. [8] Rangel-Frausto MS, Pittet D, Costigan M, Hwang T, Davis CS, Wenzel RP. The natural history of the systemic inflammatory response syndrome (SIRS). A prospective study. J Am Med Assoc 1995;273:117e23. [9] Pierrakos C, Vincent JL. Sepsis biomarkers: a review. Crit Care 2010;14:R15. [10] Assicot M, Gendrel D, Carsin H, Raymond J, Guilbaud J, Bohuon C. High serum procalcitonin concentrations in patients with sepsis and infection. Lancet 1993;341:515e8. [11] Cho SY, Choi JH. Biomarkers of sepsis. Infect Chemother 2014;46:1e12. [12] Herzum I, Renz H. Inflammatory markers in SIRS, sepsis and septic shock. Curr Med Chem 2008;15:581e7. [13] Christ-Crain M, Muller B. Procalcitonin in bacterial infections e hype, hope, more or less? Swiss Med Wkly 2005;135:451e60. [14] Amour J, Birenbaum A, Langeron O, Le Manach Y, Bertrand M, Coriat P, et al. Influence of renal dysfunction on the accuracy of procalcitonin for the diagnosis of postoperative infection after vascular surgery. Crit Care Med 2008;36: 1147e54. [15] Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P. Acute Dialysis Quality Initiative w. Acute renal failure e definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 2004;8:R204e12. [16] Matsuo S, Imai E, Horio M, Yasuda Y, Tomita K, Nitta K, et al. Revised equations for estimated GFR from serum creatinine in Japan. Am J Kidney Dis 2009;53: 982e92. [17] Murugan R, Kellum JA. Acute kidney injury: what's the prognosis? Nat Rev Nephrol 2011;7:209e17. [18] Mehanic S, Baljic R. The importance of serum procalcitonin in diagnosis and treatment of serious bacterial infections and sepsis. Mater Sociomed 2013;25: 277e81. [19] Meisner M, Lohs T, Huettemann E, Schmidt J, Hueller M, Reinhart K. The plasma elimination rate and urinary secretion of procalcitonin in patients with normal and impaired renal function. Eur J Anaesthesiol 2001;18: 79e87. [20] Ricci Z, Ronco C, D'Amico G, De Felice R, Rossi S, Bolgan I, et al. Practice patterns in the management of acute renal failure in the critically ill patient: an international survey. Nephrol Dial Transplant 2006;21: 690e6. [21] Uchino S, Bellomo R, Goldsmith D, Bates S, Ronco C. An assessment of the RIFLE criteria for acute renal failure in hospitalized patients. Crit Care Med 2006;34:1913e7. [22] Monti G, Herrera M, Kindgen-Milles D, Marinho A, Cruz D, Mariano F, et al. The DOse REsponse multicentre international Collaborative Initiative (DO-REMI). Contrib Nephrol 2007;156:434e43. [23] Opatrna S, Klaboch J, Opatrny Jr K, Holubec L, Tomsu M, Sefrna F, et al. Procalcitonin levels in peritoneal dialysis patients. Perit Dial Int 2005;25:470e2.
Please cite this article in press as: Nakamura Y, et al., Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury, J Infect Chemother (2015), http://dx.doi.org/10.1016/j.jiac.2014.12.001
Y. Nakamura et al. / J Infect Chemother xxx (2015) 1e7 [24] Steinbach G, Bolke E, Grunert A, Storck M, Orth K. Procalcitonin in patients with acute and chronic renal insufficiency. Wien Klin Wochenschr 2004;116: 849e53. [25] Herget-Rosenthal S, Marggraf G, Pietruck F, Husing J, Strupat M, Philipp T, et al. Procalcitonin for accurate detection of infection in haemodialysis. Nephrol Dial Transplant 2001;16:975e9.
7
[26] Shirakawa K, Naitou K, Hirose J, Takahashi T, Furusako S. Presepsin (sCD14ST): development and evaluation of one-step ELISA with a new standard that is similar to the form of presepsin in septic patients. Clin Chem Lab Med 2011;49:937e9.
Please cite this article in press as: Nakamura Y, et al., Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury, J Infect Chemother (2015), http://dx.doi.org/10.1016/j.jiac.2014.12.001