Next Article in Journal
Genetic Variation in CCL18 Gene Influences CCL18 Expression and Correlates with Survival in Idiopathic Pulmonary Fibrosis: Part A
Next Article in Special Issue
Molecular Mechanisms of AKI in the Elderly: From Animal Models to Therapeutic Intervention
Previous Article in Journal
Benzodiazepines Reduce Relapse and Recurrence Rates in Patients with Psychotic Depression
Previous Article in Special Issue
Liver Transplant Patients with High Levels of Preoperative Serum Ammonia Are at Increased Risk for Postoperative Acute Kidney Injury: A Retrospective Study
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
JCM
Volume 9
Issue 6
10.3390/jcm9061939
jcm-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Independent Predictive Ability of Procalcitonin of Acute Kidney Injury among Critically Ill Patients

by
Ya-Ting Huang
1,
Min-Yu Lai
1,
Wei-Chih Kan
2,3 and
Chih-Chung Shiao
4,5,*
1
Department of Nursing, Camillians Saint Mary’s Hospital Luodong. No. 160, Zhongheng S. Rd., Luodong, Yilan 26546, Taiwan
2
Department of Nephrology, Department of Internal medicine, Chi-Mei Medical Center, Yongkang District, Tainan City 710, Taiwan
3
Department of Biological Science and Technology, Chung-Hwa University of Medical Technology, Rende District, Tainan City 717, Taiwan
4
Division of Nephrology, Department of Internal Medicine, Camillians Saint Mary’s Hospital Luodong. No. 160, Zhongheng S. Rd., Luodong, Yilan 26546, Taiwan
5
Saint Mary’s Junior College of Medicine, Nursing and Management, No.100, Ln. 265, Sec. 2, Sanxing Rd., Sanxing Township, Yilan County 266, Taiwan
*
Author to whom correspondence should be addressed.
J. Clin. Med. 2020, 9(6), 1939; https://doi.org/10.3390/jcm9061939
Submission received: 12 May 2020 / Revised: 17 June 2020 / Accepted: 19 June 2020 / Published: 21 June 2020
(This article belongs to the Special Issue New Pathogenetic Mechanisms and Therapeutic Challenges in Acute Kidney Injury)
Browse Figures
Versions Notes

Abstract

:
It is unclear whether serum procalcitonin (PCT) levels rise in patients with acute kidney injury (AKI), and it is also unclear whether the elevation of PCT levels in this setting is independent of the existence of infection and impaired renal clearance. We conducted a retrospective study in a regional teaching hospital in Taiwan to evaluate the AKI-predictive ability of serum PCT among critically ill patients. We enrolled 330 patients (mean age, 70.5 ± 16.4 years; 57.0% men) who were admitted to the intensive care unit (ICU) from 1 July 2016, to 31 December 2016, and who had serum PCT measurement performed within 24 h after ICU admission. We used the generalized additive model and generalized linear model to evaluate the association of serum PCT levels and renal function variables. In addition, we used the multivariate logistic regression method to demonstrate serum PCT level as an independent predictor of AKI in both the non-infected patients (odds ratio (OR) = 1.38, 95% confidence interval (CI) = 1.12–1.71, p = 0.003) and the infected patients (OR = 1.23, 95% CI = 1.03–1.46, p = 0.020). In conclusion, serum PCT level at ICU admission is an independent predictor of developing AKI irrespective of infection among critically ill patients.

1. Introduction

Acute kidney injury (AKI) is a common but complex disorder characterized by a rapid deterioration of kidney function. AKI affects about 2 to 7% of hospitalized patients and 13 to 78% of critically ill patients [1,2,3,4], with high morbidity and mortality [5]. Over the decades, advancements in therapies for AKI have been limited, and the prognoses in AKI patients remain unsatisfactory [6]. A crucial strategy for resolving this frustrating problem is to find promising biomarkers that could early identify the onset, location, type, etiology, and severity of kidney injury [7].
Procalcitonin (PCT) is a 116-amino acid peptide that behaves as a precursor in calcium homeostasis. In the inflammatory or infectious states, serum PCT level rises rapidly within 3–4 h after onset, reaches a peak level within 6–24 h, and starts to decline after 24 h and return to a normal level by the 5th day [8,9]. As a result, PCT has been proposed as a promising marker for microbial infection and sepsis [10], as well as a useful marker to guide antibiotics therapy in critical patients with severe sepsis and septic shock [11].
PCT has ever been disclosed as a predictor of AKI in different clinical settings irrespective of the existence of infection [4,12]; nevertheless, contradicting reports appeared in the literature [3,13,14]. Besides, the existing knowledge shows that the elevated serum PCT levels in the infection or inflammatory status might be related to the impaired renal clearance of PCT in the chronic kidney disease (CKD) setting [15]. Furthermore, AKI often develops following severe infection [16,17], and is associated with a pro-inflammatory nature and a reduced waste clearance. However, it is still not clarified whether the serum PCT level elevates in AKI patients, or whether the PCT level in this situation is independent of the existence of infection and impaired renal clearance. This study aimed to prove the hypothesis that serum PCT level is an independent predictor of the development of AKI, and its predictive ability is independent of infection and impaired renal function among critically ill patients.

2. Materials and Methods

2.1. Study Design and Participants Selection

This study was reviewed and approved by the Institutional Review Board (IRB) of Saint Mary’s Hospital Luodong (# SMHIRB_105012). The study design conformed to the ethical guidelines and the Helsinki Declaration of 1975 and its revised version in 2013. The study was performed following the study protocol and relevant guidelines. The need for written informed consent was waived by the above IRB because there was neither breach of privacy nor interference with clinical practice. The data were analyzed anonymously.
The retrospective study was conducted in a regional teaching hospital in Taiwan. The inclusion criteria included adult patients who admitted to the intensive care unit (ICU) during the period from 1 July 2016, to 31 December 2016, and had serum PCT levels measured within 24 h after ICU admission. The exclusion criteria included patients less than 18 years of age, and those who had been exposed to surgeries or trauma within seven days before serum PCT measurement. We only took the data of the first hospitalization for analysis in the patients who had more than one hospitalization. After obtaining the basic and clinical data, we evaluated the association of serum PCT levels with AKI, residual renal function, and infection.

2.2. Measurements

The demographic data, comorbidities, clinical variables including vital signs, laboratory tests, the support of mechanical ventilators, or noninvasive positive pressure ventilators (NIPPV), the experience with cardiopulmonary resuscitation (CPR), and vasopressors support at ICU admission, as well as 30-day mortality were obtained from medical records. Several severity scores, including the acute physiology and chronic health evaluation II (APACHE II) and the sequential organ failure assessment (SOFA) score [18] at ICU admission, were calculated.
The AKI was diagnosed according to the Kidney Disease: Improving Global Outcomes (KDIGO) criteria [19]. The baseline serum creatinine (SCr) was defined as the latest SCr before the index admission. For the patients whose SCr level before admission is not available, the baseline SCr was calculated using the Modification of Diet in Renal Disease (MDRD) formula [20]. The peak SCr was defined by the highest SCr level within seven days after ICU admission. The SCr ratio denoted the ratio of peak SCr to baseline SCr, while the delta SCr presented the difference between peak SCr and baseline SCr. The estimated glomerular filtration rate (eGFR) was calculated by the MDRD formula [20]. Two physicians established the diagnosis of infection according to the laboratory tests (such as white blood cell count, urine routine, C-reactive protein), image study, culture results, and clinical presentations.

2.3. Quantitative Measurement of Biomarkers

All the laboratory tests were performed in the central laboratory of the hospital. The serum PCT levels were measured using the sandwich principle of an automatic electrochemiluminescent immunoassay (COBAS E411, ROCHE, Basel, Switzerland) with the analytical measurement range of 0.02–100 ng/mL and detection limit of <0.02 ng/mL.

2.4. Statistical Analysis

We used Scientific Package for Social Science (PASW Statistics for Windows, Version 22.0, Chicago, IL, USA: SPSS Inc.) and R 3.6.3 (R Foundation for Statistical Computing, Vienna, Austria, accessed https://www.r-project.org/) software for statistical analyses, and took a p ≤ 0.05 as statistically significant in all statistical analyses.
Categorical variables were expressed as case number (percentage) and compared using the chi-square test. Continuous variables experienced normality evaluation using the Kolmogorov–Smirnov test and the Shapiro–Wilk normality test [21]. The continuous variables with normal distribution were reported as mean ± standard deviation (SD) and compared using an independent t-test. The continuous variables with abnormal distribution were expressed as median (range) and compared using an independent t-test after log transformation [22] and confirmation as normal distribution by QQ plot. The Kruskal–Wallis test was used to compare the serum PCT level among patients with different AKI stages, while the Mann–Whitney U test was used to compare the serum PCT level between two groups with small sample sizes and unequal case numbers in the two groups. Furthermore, the analysis of covariance (ANCOVA) was performed using mixed linear models, with statistical control for the effects of baseline renal function and severity score, to compare the PCT levels among patients with different AKI stages and infection states.
We used the generalized additive model (GAM) and generalized linear model (GLM) to evaluate the association between the serum PCT levels and some variables (such as SCr ratio, delta_SCr, and eGFR). The trends of lines were plotted by linear regression line with a LOESS (locally estimated scatterplot smoothing) curve, which is a nonparametric technique using locally weighted regression to fit a smooth curve through points in a scatter plot.
The conditional backward stepwise model of multivariate logistic regression method was used to investigate the odds ratio (OR), 95% confidence interval (CI) and p-value of the independent predictors of AKI after testing collinearity statistics. Multicollinearity was defined as a variance inflation factor (VIF) value of more than 4.0 or tolerance of less than 0.2 [23]. We put relevant variables expressing demographic information, baseline renal function, and severity score along with serum PCT levels into the multivariate logistic regression to determine the independent predictors for AKI. Hosmer and Lemeshow goodness-of-fit tests were used for calibration of the model.
By using the simple logistic regression to evaluate the association of AKI and the log-transformed PCT levels, we obtained an OR of 1.412. The power subsequently calculated by G-Power of the logistic regression method in our study was 0.99 when setting α as 0.05 and OR as 1.412.

3. Results

During the enrollment period, a total of 745 patients were extracted from the hospital database. After excluding 415 patients, we enrolled a total of 330 patients (mean age, 70.5 ± 16.4 years; 57.0% men). Among them, the baseline SCr values of 46 patients were estimated by the MDRD formula due to insufficient data. The GAM plots disclosed that the serum PCT levels started to increase with a SCr ratio of 1.5 (Figure 1A) and a delta SCr of 0.3 mg/dL (Figure 1B), which together indicated stage 1 AKI. Accordingly, we categorized the patients into the AKI group (n = 127, 38.5%) and the non-AKI group (n = 203, 61.5%) by the existence of AKI (stage 1 and more advanced stages) within seven days after ICU admission.

3.1. Basic Characteristics and Clinical Variables of the Two Groups

The basic demographic information, comorbid diseases, and most of the clinical variables at ICU admission were not significantly different between the AKI group and the non-AKI group. The only statistical differences were that the AKI group had a higher proportion of pneumonia (21.3% vs. 12.3%, p = 0.030) and a higher SOFA score (8.1 ± 3.6 vs. 6.2 ± 3.8 points). Besides, the AKI group had higher 30-day mortality than the non-AKI group (31.5% vs. 20.2%, p = 0.020). (Table 1) As for the laboratory data, the AKI group had higher serum PCT levels (2.3 (0.05, 234.6) vs. 0.5 (0.02, 242.8) ng/mL), blood urea nitrogen (BUN) (54.4 (11.0, 205.0) vs. 23.4 (5.3, 210.7) mmol/L), SCr (2.6 (0.4, 18.2) vs. 1.0 (0.3, 15.9) mmol/L), alanine aminotransferase (ALT) (36.0 (3.0, 1891.0) vs. 21.0 (1.0, 709.0) units/L, p = 0.002), potassium (4.4 ± 1.2 vs. 4.1 ± 1.0 mEq/L, p = 0.031), but lower eGFR (23.5 (1.8, 557.8) vs. 65.9 (1.3, 382.3) mL/min/1.73 m2) and bicarbonate (HCO3) (17.5 ± 7.4 vs. 21.2 ± 7.8 mEq/L) than the non-AKI group. (Table 2) (All p < 0.001 unless otherwise denoted).

3.2. The Association among Infection, Acute Kidney Injury and Impaired Renal Function

Regardless of the existence of infection, the serum PCT level had an increasing trend along with the worsening severity of AKI (denoted by increasing SCr ratio, Figure S1A) and the worsening residual renal function (denoted by decreasing eGFR, Figure S1B) within the clinically relevant range. The serum PCT levels were significantly higher in the infection group than in the non-infection group in the above two figures. (Both p < 0.001) Moreover, the PCT levels of infected patients were above 0.5 ng/mL, the cut-point indicative of systemic infection [3]. (Figure S1A,B).
Furthermore, an increasing trend of the PCT levels, along with the worsening residual renal function, could be seen in patients regardless of the existence of AKI. The serum PCT levels of the AKI group were significantly higher than those of the non-AKI group at the same corresponding eGFR levels. (p = 0.005) The finding could be interpreted as higher serum PCT levels in AKI patients than CKD patients (presented as non-AKI patients with eGFR < 60 mL/min/1.73 m2) at the same eGFR levels. (Figure 2A).
Then, we further compared the PCT levels among four groups stratified by the presence of AKI and infection. We found that the infection (+)/AKI (+) group had the highest serum PCT levels, followed along with the subsequently decreasing PCT levels in the infection (+)/AKI (−) group, the infection (−)/AKI (+) group and the infection (−)/AKI (−) group. The differences in serum PCT levels between any two groups were statistically significant, except the difference between the infection (−)/AKI (+) group and the infection (−)/AKI (−) group. (Figure 2B).
On the other hand, the median serum PCT levels were significantly higher in the AKI (+) group than in the AKI (−) group irrespective of the presence of infection. Additionally, the serum PCT levels were significantly higher in the infection (+) group than in the infection (−) group, irrespective of the presence of AKI. (All p < 0.001) (Figure S2).
Regarding the different stages of AKI, we found that the serum PCT levels significantly and stepwise increased from the non-AKI group to the AKI stage 3 group in both infection (+) and infection (−) groups. Whereas the infection (+) group had significantly higher serum PCT levels than the infection (−) group at most of the AKI stages except stage 3. (Figure 3) Furthermore, we performed ANCOVA with log-transformed PCT as the measure of interest (dependent variable), AKI stages and infection as fixed factors, and baseline eGFR and SOFA score as covariates. The ANCOVA confirmed that with the control for baseline renal function and disease severity, serum PCT levels were statistically different among patients with different AKI stages irrespective of infection, and between patients with and without infection irrespective of the AKI stages. (Both p < 0.001) (Table S1).

3.3. The Predictive Ability of Serum Procalcitonin for Acute Kidney Injury

In the multivariate logistic regression analysis including log-transformed PCT levels along with age, gender, baseline eGFR, and SOFA score at ICU admission for adjustment, we demonstrated that the serum PCT level behaved as an independent factor for predicting AKI in the whole cohort (OR = 1.27, 95% CI = 1.12–1.43), including the non-infection patients (OR = 1.38, 95% CI = 1.12–1.71, p = 0.003) and the infection group (OR = 1.23, 95% CI = 1.03–1.46, p = 0.020). (Table 3) Hosmer and Lemeshow goodness-of-fit tests confirmed the calibration of the models of the whole cohort (chi-square = 8.44, p = 0.394), the non-infection patients (chi-square = 6.52, p = 0.588) and the infection patients (chi-square = 5.28, p = 0.734). (All p < 0.001 unless otherwise denoted).

4. Discussion

To the best of our knowledge, the current study is among the first few studies demonstrating the complex association of serum PT level with AKI, infection and impaired residual renal function. The strength of the current study is the comparisons of the serum PCT levels between different settings at the same residual renal function. The study had the following findings: (1) The existence of infection, the worsening residual renal function and the increasing severity of AKI respectively associated with increased serum PCT levels. (Figure 2 and Figure 3, Supplementary Figures S1 and S2) (2) The influence on serum PT levels of AKI was more significant than CKD with the same residual renal function. (Figure 2A) (3) The serum PCT level was an independent predictor for the development of AKI regardless of the existence of an infection.

4.1. Influence on Procalcitonin: Infection, Residual Renal Function and Acute Kidney Injury

The PCT is produced and converted to calcitonin within the thyroid C cells before releasing into the circulation, resulting in a very low (0.05 ng/mL) serum PCT levels in healthy subjects [24]. In the inflammatory state, the calcitonin production process is independent of the above regulations [24], causing a variously-degreed increase in calcitonin content in various organs [25]. Thus PCT has been recognized as a useful marker for detecting infection/inflammatory, guiding antibiotic therapy, and predicting the prognoses in infected patients [10,11].
The association between PCT levels and residual renal function found in the current study was also consistent with the existing knowledge. PCT is demonstrated to be eliminated through the urine by kidneys [26], and the renal clearance of PCT significantly reduced in parallel to the decrease in eGFR [27,28]. As a result, elevated PCT levels could be seen in CKD patients [29].
To date, serum PCT levels have been found to elevate in AKI patients compared to those without AKI in various clinical settings [30,31]. The current study not only had findings consistent with the existing evidence but also further demonstrated the increasing PCT levels along with the increasing AKI severity irrespective of the existence of infection (Figure 3 and Table S1). The potential explanations for the association between PCT and AKI are as follows: (1) The direct cytotoxic effect of PCT on mesangial cells, which damages the kidney [32]. (2) The role of PCT as a yet unknown factor in the pathogenesis of AKI [4]. (3) The pro-inflammatory nature of AKI and the situation causing AKI might induce the synthesis of PCT. (4) The reduced clearance of PCT early in AKI before a significant elevation in SCr [4]. Besides, higher PCT was significantly associated with a higher level of an AKI marker, neutrophil gelatinase-associated lipocalin, an increased APACHE II score [33], and a reduced chance of AKI recovery [34]. Since AKI is considered an entity associated with inflammatory and disease severity [35], the above association provides a relevant pathophysiological implication between PCT and AKI.

4.2. Influence on Procalcitonin: Acute Kidney Injury vs. Infection

Regarding the complex influences on PCT levels of infection and AKI, diverse results exist. Although some investigation reported that the diagnostic accuracy of PCT for bacterial infection is not influenced by the existence of AKI [36], other studies disclosed that the infection-diagnostic ability of PCT was influenced in severe AKI [26] or even mild AKI [37].
In line with the previous studies [4,13,14,33,36], we disclosed that the presence of either infection or AKI was associated with an increased serum PCT level compared to those without the corresponding entity, while the infection had a higher substantial effect than AKI on PCT levels. This finding was clearly shown in the figures presenting with median PCT levels (Figure S2 and Figure 3). As a strength of the current study, we further compared PCT levels along with the worsening eGFR of four groups stratified by infection and AKI. The findings were almost consistent with the results of comparisons of median PCT levels. The only exception was that the higher PCT levels of the infection (−)/AKI (+) group compared to the infection (−)/AKI (−) group was not statistically significant (p = 0.059), and this insignificancy might be related to some unidentified bias that occurred in the range of eGFR 30–50 mL/min/1.73 m2. (Figure 2B).
Moreover, we found that the influence of infection and AKI on the PCT level could be synergistic (Figure 3 and Figure S2). The potential explanations for the synergistic effect include: (1) Both infection and AKI are associated with inflammatory/pro-inflammatory status which causes more PCT releasing into the circulation; (2) AKI associates with an impaired renal clearance of PCT from the circulation, which indirectly keeps the serum PCT levels high.

4.3. Influence on Procalcitonin: Acute Kidney Injury vs. Chronic Kidney Disease

In the current study, we found that AKI patients had significantly higher serum PCT levels than CKD patients at the same corresponding eGFR levels. Since the renal clearance ability of PCT decreased in both AKI and CKD, these results might reflect the higher amount of PCT production induced by the more inflammatory characteristic of AKI when compared to CKD.

4.4. Serum Procalcitonin as a Predictor for Acute Kidney Injury

The serum PCT level was found in the current study as an independent predictor for AKI in both the infected patients and non-infected patients. The finding was supported by the work of Heredia-Rodríguez et al. [37], which disclosed that AKI patients had significantly higher PCT levels than non-AKI patients irrespective of the presence of sepsis among cardiac surgical patients. The impact of PCT level on the AKI patients with infection when compared to that on the AKI patients without infection could be found in other investigations [12,31]. However, contradicting results exist, with some studies reporting that the serum PCT level failed to predict AKI occurrence in patients with sepsis [4] or influenza infection [13,14].

4.5. Limitations

There are several potential limitations in the current study. First, as a single-centered retrospective study, it was subject to bias. However, the finding in the current study that indicate that serum PCT significantly elevated since stage 1 AKI provides an essential clue for categorizing participants in the further prospective study to compare the exact influences of AKI and infection on serum PT levels. Second, the enrolled patients in the current study were mainly medical patients, and were highly selected by the criteria “existence of serum PCT measurements within 24 h after ICU admission.” The findings in the population might not be suitable to apply to other clinical settings, such as surgical or trauma patients or those who are not critically ill. Third, the current study evaluated the association between one measurement of serum PCT at initial admission and the development of AKI within seven days of admission. The serial changes of serum PCT levels during management in the hospitalization, although maybe relevant, were not taken into consideration. Fourth, it is controversial to present the residual renal function using eGFR calculated by the MDRD formula in AKI patients. We decided to do so because, as of yet, there is no other method available to compare residual renal function between patients with and without AKI.
Further multicentered, prospective researches may be warranted to investigate the predictive and prognostic values of PCT for AKI patients. Regarding the application of the PCT, “the PCT levels at different time points”, “various cut-points of PCT for different patient settings” or “using the change (in values or percentage) between different time points” may be potential strategies to determine the association among PCT, AKI, and infection.

5. Conclusions

The current study demonstrated that AKI had a more substantial influence on elevating serum PCT levels than CKD at the same residual renal function, and the serum PCT level of critically ill patients at ICU admission is an independent predictor for the development of AKI within the coming seven days, irrespective of infection among.

Supplementary Materials

The following are available online at https://www.mdpi.com/2077-0383/9/6/1939/s1, Figure S1. Comparisons of PCT levels between two groups stratified by infection in plots with (A) serum creatinine ratio and (B) eGFR as X-axis. Abbreviations: eGFR = estimated glomerular filtration rate; PCT = procalcitonin, Figure S2. Median levels of serum PCT among four groups stratified by AKI and infection. Note: Data were expressed as median (range). *** and ### denote p < 0.001 in the comparisons of serum PCT between AKI (+) and AKI (−) groups and between infection (+) and infection (−) groups, respectively. The statistical analyses were made using independent t-test. Abbreviations: AKI = acute kidney injury; PCT = procalcitonin. Table S1. Comparisons of PCT levels among patients with different AKI stages and infection states. Note: The data were presented as “PCT levels (95% confidence interval)”. The PCT levels were transformed back from the log form which had been used for analyses. The analysis of covariance was performed using mixed linear models with log-transformed PCT as the measure of interest (dependent variable), AKI states and infection as fixed factors, and baseline estimated glomerular filtration rate and sequential organ failure assessment score as covariates. 1 denotes overall comparisons among AKI stages. 2 denotes overall comparisons between infection group and non-infection group. Abbreviations: AKI = acute kidney injury, PCT = procalcitonin.

Author Contributions

Conceptualization, Y.-T.H. and C.-C.S.; data curation, Y.-T.H. and M.-Y.L.; formal analysis, Y.-T.H.; funding acquisition, Y.-T.H.; investigation, W.-C.K. and C.-C.S.; methodology, C.-C.S. and M.-Y.L.; resources, M.-Y.L.; supervision, C.-C.S.; validation, W.-C.K.; visualization, W.-C.K.; writing—original draft, Y.-T.H., C.-C.S. and M.-Y.L.; writing—review and editing, W.-C.K. and C.-C.S. All authors have read and agreed to the published version of the manuscript.

Funding

The research was supported by the Saint Mary’s Hospital Research Fund (#SMHRF-105009) and the Ministry of Science and Technology (MOST) of the Republic of China (Taiwan) (MOST 106-2321-B-182-002, 105-2314-B-002-045, Most 109-2321-B-182-001).

Acknowledgments

The authors thank the staff in Camillians Saint Mary’s Hospital Luodong and the Taiwan Clinical Trial Consortium (TCTC) for assisting the study. The authors also give thanks to Swei-Hsiung Tsung and Miss Simone Yu for providing English-editing of the manuscript.

Conflicts of Interest

The authors declare no conflicts of interest. The funder had no role in study design, data collection, and analysis, decision to publish, or the preparation of the manuscript.

References

  1. Ostermann, M.; Joannidis, M. Acute kidney injury 2016: Diagnosis and diagnostic workup. Crit. Care 2016, 20, 299. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  2. Mas-Font, S.; Ros-Martinez, J.; Perez-Calvo, C.; Villa-Diaz, P.; Aldunate-Calvo, S.; Moreno-Clari, E. Prevention of acute kidney injury in Intensive Care Units. Med. Intensiva 2017, 41, 116–126. [Google Scholar] [CrossRef] [PubMed]
  3. Chun, K.; Chung, W.; Kim, A.J.; Kim, H.; Ro, H.; Chang, J.H.; Lee, H.H.; Jung, J.Y. Association between acute kidney injury and serum procalcitonin levels and their diagnostic usefulness in critically ill patients. Sci. Rep. 2019, 9, 4777. [Google Scholar] [CrossRef] [PubMed]
  4. Jeeha, R.; Skinner, D.L.; De Vasconcellos, K.; Magula, N.P. Serum procalcitonin levels predict acute kidney injury in critically ill patients. Nephrology (Carlton) 2018, 23, 1090–1095. [Google Scholar] [CrossRef]
  5. Pannu, N.; James, M.; Hemmelgarn, B.; Klarenbach, S. Association between AKI, recovery of renal function, and long-term outcomes after hospital discharge. Clin. J. Am. Soc. Nephrol. 2013, 8, 194–202. [Google Scholar] [CrossRef]
  6. Druml, W. Systemic consequences of acute kidney injury. Curr. Opin. Crit. Care 2014, 20, 613–619. [Google Scholar] [CrossRef]
  7. Malhotra, R.; Siew, E.D. Biomarkers for the Early Detection and Prognosis of Acute Kidney Injury. Clin. J. Am. Soc. Nephrol. 2017, 12, 149–173. [Google Scholar] [CrossRef]
  8. Kibe, S.; Adams, K.; Barlow, G. Diagnostic and prognostic biomarkers of sepsis in critical care. J. Antimicrob. Chemother. 2011, 66 (Suppl. 2), ii33–ii40. [Google Scholar] [CrossRef] [Green Version]
  9. Poddar, B.; Gurjar, M.; Singh, S.; Aggarwal, A.; Singh, R.; Azim, A.; Baronia, A. Procalcitonin kinetics as a prognostic marker in severe sepsis/septic shock. Indian J. Crit. Care Med. Peer-Rev. Off. Publ. Indian Soc. Crit. Care Med. 2015, 19, 140–146. [Google Scholar]
  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, 515–518. [Google Scholar] [CrossRef]
  11. Prkno, A.; Wacker, C.; Brunkhorst, F.M.; Schlattmann, P. Procalcitonin-guided therapy in intensive care unit patients with severe sepsis and septic shock—A systematic review and meta-analysis. Crit. Care 2013, 17, R291. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  12. Nie, X.; Wu, B.; He, Y.; Huang, X.; Dai, Z.; Miao, Q.; Song, H.; Luo, T.; Gao, B.; Wang, L.; et al. Serum procalcitonin predicts development of acute kidney injury in patients with suspected infection. Clin. Chem. Lab. Med. 2013, 51, 1655–1661. [Google Scholar] [CrossRef] [PubMed]
  13. Cruz-Lagunas, A.; Jimenez-Alvarez, L.; Ramirez, G.; Mendoza-Milla, C.; Garcia-Sancho, M.C.; Avila-Moreno, F.; Zamudio, P.; Urrea, F.; Ortiz-Quintero, B.; Campos-Toscuento, V.L.; et al. Obesity and pro-inflammatory mediators are associated with acute kidney injury in patients with A/H1N1 influenza and acute respiratory distress syndrome. Exp. Mol. Pathol. 2014, 97, 453–457. [Google Scholar] [CrossRef] [PubMed]
  14. Rodriguez, A.; Reyes, L.F.; Monclou, J.; Suberviola, B.; Bodi, M.; Sirgo, G.; Sole-Violan, J.; Guardiola, J.; Barahona, D.; Diaz, E.; et al. Relationship between acute kidney injury and serum procalcitonin (PCT) concentration in critically ill patients with influenza infection. Med. Intensiva 2018, 42, 399–408. [Google Scholar] [CrossRef] [PubMed]
  15. Wu, S.C.; Liang, C.X.; Zhang, Y.L.; Hu, W.P. Elevated serum procalcitonin level in patients with chronic kidney disease without infection: A case-control study. J. Clin. Lab. Anal. 2020, 34, e23065. [Google Scholar] [CrossRef]
  16. Bellomo, R.; Kellum, J.A.; Ronco, C.; Wald, R.; Martensson, J.; Maiden, M.; Bagshaw, S.M.; Glassford, N.J.; Lankadeva, Y.; Vaara, S.T.; et al. Acute kidney injury in sepsis. Intensive Care Med. 2017, 43, 816–828. [Google Scholar] [CrossRef] [Green Version]
  17. Gómez, H.; Kellum, J.A. Sepsis-induced acute kidney injury. Curr. Opin. Crit. Care 2016, 22, 546–553. [Google Scholar]
  18. Vincent, J.L.; Moreno, R.; Takala, J.; Willatts, S.; De Mendonca, A.; Bruining, H.; Reinhart, C.K.; Suter, P.M.; Thijs, L.G. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med. 1996, 22, 707–710. [Google Scholar] [CrossRef]
  19. Palevsky, P.M.; Liu, K.D.; Brophy, P.D.; Chawla, L.S.; Parikh, C.R.; Thakar, C.V.; Tolwani, A.J.; Waikar, S.S.; Weisbord, S.D. KDOQI US commentary on the 2012 KDIGO clinical practice guideline for acute kidney injury. Am. J. Kidney Dis. 2013, 61, 649–672. [Google Scholar] [CrossRef]
  20. Bellomo, R.; Ronco, C.; Kellum, J.A.; Mehta, R.L.; Palevsky, P. Acute Dialysis Quality Initiative w: Acute renal failure—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, R204–R212. [Google Scholar] [CrossRef] [Green Version]
  21. Henderson, A.R. Testing experimental data for univariate normality. Clin. Chim. Acta Int. J. Clin. Chem. 2006, 366, 112–129. [Google Scholar] [CrossRef] [PubMed]
  22. Curran-Everett, D. Explorations in statistics: The log transformation. Adv. Physiol. Educ. 2018, 42, 343–347. [Google Scholar] [CrossRef] [PubMed]
  23. Hair, J.F., Jr.; Black, W.C.; Babin, B.J.; Anderson, R.E. Multivariate Data Analysis, 8th ed.; Prentice Hall: Upper Saddle River, NJ, USA, 2019. [Google Scholar]
  24. Vijayan, A.L.; Vanimaya Ravindran, S.; Saikant, R.; Lakshmi, S.; Kartik, R.; Manoj, G. Procalcitonin: A promising diagnostic marker for sepsis and antibiotic therapy. J. Intensive Care 2017, 5, 51. [Google Scholar] [CrossRef] [PubMed]
  25. Muller, B.; White, J.C.; Nylen, E.S.; Snider, R.H.; Becker, K.L.; Habener, J.F. Ubiquitous expression of the calcitonin-i gene in multiple tissues in response to sepsis. J. Clin. Endocrinol. Metab. 2001, 86, 396–404. [Google Scholar] [CrossRef] [PubMed]
  26. Nakamura, Y.; Murai, A.; Mizunuma, M.; Ohta, D.; Kawano, Y.; Matsumoto, N.; Nishida, T.; Ishikura, H. Potential use of procalcitonin as biomarker for bacterial sepsis in patients with or without acute kidney injury. J. Infect. Chemother. 2015, 21, 257–263. [Google Scholar] [CrossRef]
  27. 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, 79–87. [Google Scholar] [CrossRef]
  28. Herget-Rosenthal, S.; Klein, T.; Marggraf, G.; Hirsch, T.; Jakob, H.G.; Philipp, T.; Kribben, A. Modulation and source of procalcitonin in reduced renal function and renal replacement therapy. Scand. J. Immunol. 2005, 61, 180–186. [Google Scholar] [CrossRef]
  29. Lavin-Gomez, B.A.; Palomar-Fontanet, R.; Gago-Fraile, M.; Quintanar-Lartundo, J.A.; Gomez-Palomo, E.; Gonzalez-Lamuno, D.; Garcia-Unzueta, M.T.; Arias-Rodriguez, M.A.; Gomez-Gerique, J.A. Inflammation markers, chronic kidney disease, and renal replacement therapy. Adv. Perit. Dial. 2011, 27, 33–37. [Google Scholar] [PubMed]
  30. Clementi, A.; Brocca, A.; Virzi, G.M.; de Cal, M.; Giavarina, D.; Carta, M.; Mucino-Bermejo, M.J.; Hinna Danesi, T.; Salvador, L.; Ronco, C. Procalcitonin and Interleukin-6 Levels: Are They Useful Biomarkers in Cardiac Surgery Patients? Blood Purif. 2017, 43, 290–297. [Google Scholar] [CrossRef] [PubMed]
  31. Huang, H.L.; Nie, X.; Cai, B.; Tang, J.T.; He, Y.; Miao, Q.; Song, H.L.; Luo, T.X.; Gao, B.X.; Wang, L.L.; et al. Procalcitonin levels predict acute kidney injury and prognosis in acute pancreatitis: A prospective study. PLoS ONE 2013, 8, e82250. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  32. Araujo, M.; Doi, S.Q.; Palant, C.E.; Nylen, E.S.; Becker, K.L. Procalcitonin induced cytotoxicity and apoptosis in mesangial cells: Implications for septic renal injury. Inflamm. Res. 2013, 62, 887–894. [Google Scholar] [CrossRef] [PubMed]
  33. Takahashi, G.; Shibata, S.; Fukui, Y.; Okamura, Y.; Inoue, Y. Diagnostic accuracy of procalcitonin and presepsin for infectious disease in patients with acute kidney injury. Diagn. Microbiol. Infect. Dis. 2016, 86, 205–210. [Google Scholar] [CrossRef] [PubMed]
  34. Itenov, T.S.; Jensen, J.U.; Ostrowski, S.R.; Johansson, P.I.; Thormar, K.M.; Lundgren, J.D.; Bestle, M.H.; “Procalcitonin and Survival Study” Study Group. Endothelial Damage Signals Refractory Acute Kidney Injury in Critically Ill Patients. Shock 2017, 47, 696–701. [Google Scholar] [CrossRef] [PubMed]
  35. Ba Aqeel, S.H.; Sanchez, A.; Batlle, D. Angiotensinogen as a biomarker of acute kidney injury. Clin. Kidney J. 2017, 10, 759–768. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  36. Amour, J.; Birenbaum, A.; Langeron, O.; Le Manach, Y.; Bertrand, M.; Coriat, P.; Riou, B.; Bernard, M.; Hausfater, P. Influence of renal dysfunction on the accuracy of procalcitonin for the diagnosis of postoperative infection after vascular surgery. Crit. Care Med. 2008, 36, 1147–1154. [Google Scholar] [CrossRef]
  37. Heredia-Rodriguez, M.; Bustamante-Munguira, J.; Fierro, I.; Lorenzo, M.; Jorge-Monjas, P.; Gomez-Sanchez, E.; Alvarez, F.J.; Bergese, S.D.; Eiros, J.M.; Bermejo-Martin, J.F.; et al. Procalcitonin cannot be used as a biomarker of infection in heart surgery patients with acute kidney injury. J. Crit. Care 2016, 33, 233–239. [Google Scholar] [CrossRef]
Jcm 09 01939 g001 550
Figure 1. Generalized additive model plots showing the trends of PCT changes along with the increase in (A) SCr ratio, and (B) delta SCr. Abbreviations: SCr = serum creatinine; PCT = procalcitonin.
Figure 1. Generalized additive model plots showing the trends of PCT changes along with the increase in (A) SCr ratio, and (B) delta SCr. Abbreviations: SCr = serum creatinine; PCT = procalcitonin.
Jcm 09 01939 g001
Jcm 09 01939 g002 550
Figure 2. Comparisons of PCT levels among groups stratified by (A) AKI, and (B) AKI and infection. Abbreviations: AKI = acute kidney injury; eGFR = estimated glomerular filtration rate; inf = infection; PCT = procalcitonin.
Figure 2. Comparisons of PCT levels among groups stratified by (A) AKI, and (B) AKI and infection. Abbreviations: AKI = acute kidney injury; eGFR = estimated glomerular filtration rate; inf = infection; PCT = procalcitonin.
Jcm 09 01939 g002
Jcm 09 01939 g003 550
Figure 3. Median PCT levels among groups stratified by AKI stages and infection. Note: Data are expressed as median (range). *** denotes p < 0.001 in the comparisons of serum PCT levels among patients with different AKI stages using the Kruskal–Wallis test. #, ##, ### denote p < 0.05, <0.01, <0.001, respectively, in the comparison of serum PCT levels between infection (+) group and infection (−) group using the Mann–Whitney U test. Abbreviations: AKI = acute kidney injury; PCT = procalcitonin.
Figure 3. Median PCT levels among groups stratified by AKI stages and infection. Note: Data are expressed as median (range). *** denotes p < 0.001 in the comparisons of serum PCT levels among patients with different AKI stages using the Kruskal–Wallis test. #, ##, ### denote p < 0.05, <0.01, <0.001, respectively, in the comparison of serum PCT levels between infection (+) group and infection (−) group using the Mann–Whitney U test. Abbreviations: AKI = acute kidney injury; PCT = procalcitonin.
Jcm 09 01939 g003
Table
Table 1. Comparisons of basic characteristics and clinical variables of the two groups.
Table 1. Comparisons of basic characteristics and clinical variables of the two groups.
Total
(n = 330)		Non-AKI Group
(n = 203)		AKI Group
(n = 127)		p-Value
Basic demographic data
Age70.5 ± 16.470.5 ± 16.570.6 ± 16.30.933
Gender, male188 (57.0%)119 (58.6%)69 (54.3%)0.444
Smoker 72 (21.8%)42 (20.7%)30 (23.6%)0.752
Undertaking oral antibiotics19 (5.8%)13 (6.4%)6 (4.7%)0.524
Body mass index21.9 ± 5.722.1 ± 5.521.5 ± 5.90.351
Comorbidities
Coronal artery disease73 (22.1%)48 (23.6%)25 (19.7%)0.399
Congestive heart failure47 (14.2%)29 (14.3%)18 (14.2%)0.977
Peripheral artery occlusive disease8 (2.4%)5 (2.5%)3 (2.4%)0.954
Cerebral vascular accident104 (31.5%)61 (30%)43 (33.9%)0.469
Chronic lung disease 82 (24.8%)51 (25.1%)31 (24.4%)0.884
Chronic kidney disease 98 (29.7%)56 (27.6%)42 (33.1%)0.289
Diabetes mellitus 131(39.7%)80 (39.4%)51 (40.2%)0.892
Cancer 44 (13.3%)24 (11.8%)20 (15.7%)0.307
Liver cirrhosis28 (8.5%)16 (7.9%)12 (9.4%)0.619
Hypertension 189 (57.3%)115 (56.7%)74 (58.3%)0.773
Charlson’s score, points3.8 ± 2.63.7 ± 2.63.9 ± 2.60.460
Culture-proven infection173 (52.4%)103 (50.7%)70 (55.1%)0.438
Infection source
Pneumonia52 (15.8%)25 (12.3%)27 (21.3%)0.030
Urinary tract infection65 (19.7%)40 (19.7%)25 (19.7%)0.997
Bloodstream infection72 (21.8%)38 (18.7%)34 (26.8%)0.085
Skin infection13 (3.9%)9 (4.4%)4 (3.1%)0.560
Other source45 (13.6%)28 (13.8%)17 (13.4%)0.916
Clinical variables at ICU admission
Body temperature, °C36.5 ± 1.236.6 ± 1.136.4 ± 1.20.135
Heart rate, beat/min103 ± 24.4101.9 ± 24.9104.9 ± 23.50.282
Respiratory rate, breath/min25.1 ± 9.124.7 ± 9.525.6 ± 8.60.363
Mean arterial pressure, mmHg89.0 ± 25.589.9 ± 25.587.5 ± 25.60.415
Glasgow coma scale, points10.3 ± 4.510.1 ± 4.610.7 ± 4.40.280
APACHE II, points20.8 ± 8.220.1 ± 8.421.9 ± 7.90.051
SOFA score, points6.9 ± 3.86.2 ± 3.88.1 ± 3.6<0.001
With ventilator 99 (30.0%)63 (31.0%)36(28.3%)0.604
With NIPPV88 (26.7%)53 (26.1%)35(27.6%)0.772
With vasopressor112 (33.9%)61 (30.0%)51(40.2%)0.059
Underwent CPR29 (8.8%)17(8.4%)12(9.4%)0.737
30-days mortality81 (24.5%)41 (20.2%)40 (31.5%)0.020
Note: Continuous variables with normal distribution and categorical variables were expressed as mean ± standard deviation and n (percentage), respectively. Statistical analyses were performed using independent t-test for the continuous variables with normal distribution, or chi-square test for categorical variables. Abbreviations: AKI = acute kidney injury, APACHE II = acute physiology and chronic health evaluation II, CPR = cardiopulmonary resuscitation, ICU = intensive care unit, NIPPV = noninvasive positive pressure ventilator, SOFA = sequential organ failure assessment.
Table
Table 2. Comparisons of laboratory data of the two groups.
Table 2. Comparisons of laboratory data of the two groups.
Total
(n = 330)		Non-AKI Group
(n = 203)		AKI Group
(n = 127)		p-Value
Procalcitonin, ng/mL0.8 (0.02, 242.8)0.5 (0.02, 242.8)2.3 (0.05, 234.6)<0.001
White blood cell, ×103/mL13.5 ± 8.712.8 ± 7.414.6 ± 10.40.057
Neutrophil/ Lymphocyte ratio8.3 (0.2, 95.8)7.0 (0.2, 95.8)10.4 (0.2, 91.8)0.068
Hemoglobin, g/dL11.1 ± 2.911.2 ± 3.010.8 ± 2.80.321
Platelet, ×103/mL216.7 ± 114.0219.8 ± 116.6211.6 ± 110.10.527
Blood urea nitrogen, mmol/L32.2 (5.3, 210.7)23.4 (5.3, 210.7)54.4 (11.0, 205.0)<0.001
sCr, mmol/L1.5 (0.3, 18.2)1.0 (0.3, 15.9)2.6 (0.4, 18.2)<0.001
eGFR, ml/min/1.73 m244.2 (1.3, 557.8)65.9 (1.3, 382.3)23.5 (1.8, 557.8)<0.001
AST, units/L32.0 (3.4, 2236.0)29.0 (9.0, 2236.0)41.0 (3.4, 1027.0)0.067
ALT, units/L25.0 (1.0, 1891.0)21.0 (1.0, 709.0)36.0 (3.0, 1891.0)0.002
Sodium, mmol/L136.8 ± 9.2136.7 ± 8.3137.0 ± 10.60.806
Potassium, mEq/L4.2 ± 1.14.1 ± 1.04.4 ± 1.20.031
Calcium, mEq/L8.3 ± 1.08.3 ± 1.08.3 ± 1.10.967
PH7.3 ± 0.17.4 ± 0.17.3 ± 0.10.512
HCO3, mEq/L19.8 ± 7.821.2 ± 7.817.5 ± 7.4<0.001
Glucose, mg/dl225.1 ± 170.9209.6 ± 132.9249.8 ± 216.70.062
Albumin, mg/dl3.1 ± 0.63.1 ± 0.63.0 ± 0.60.144
Bililubin (total), mg/dl0.9 (0.0, 42.5)0.9 (0.1, 15.0)0.9 (0.0, 42.5)0.210
Baseline SCr, mmol/L 1.0 (0.2, 11.2)1.0 (0.2, 11.2)1.0 (0.2, 10.6)0.056
Delta SCr, mmol/L0.9 ± 2.20.0 ± 0.92.4 ± 2.7<0.001
Ratio of SCr1.8 ± 1.61.0 ± 0.33.0 ± 2.1<0.001
Note: Continuous variables with normal or abnormal distribution were expressed as mean ± standard deviation or median (range), respectively. Statistical analyses were performed using independent t-test for the continuous variables with normal distribution, or independent t-test after log transformation for the continuous variables with the abnormal distribution. Abbreviations: AKI = acute kidney injury, eGFR = estimated glomerular filtration rate, ALT = alanine aminotransferase, AST = aspartate aminotransferase, HCO3 = bicarbonate, SCr, serum creatinine.
Table
Table 3. The independent role of procalcitonin in predicting acute kidney injury.
Table 3. The independent role of procalcitonin in predicting acute kidney injury.
Multivariate Logistic Regression
Odds Ratio95% Confidence Intervalp-Value
Total cohort (n = 330)1.271.12–1.43<0.001
Non-infection group (n = 157)1.381.12–1.710.003
Infection group (n = 173)1.231.03–1.460.020
Note: All the three sets of multivariate analyses adjusted to age, gender, baseline estimated glomerular filtration rate, and sequential organ failure assessment score at intensive care unit admission. Procalcitonin was analyzed after log transformation.

Share and Cite

MDPI and ACS Style

Huang, Y.-T.; Lai, M.-Y.; Kan, W.-C.; Shiao, C.-C. Independent Predictive Ability of Procalcitonin of Acute Kidney Injury among Critically Ill Patients. J. Clin. Med. 2020, 9, 1939. https://doi.org/10.3390/jcm9061939

AMA Style

Huang Y-T, Lai M-Y, Kan W-C, Shiao C-C. Independent Predictive Ability of Procalcitonin of Acute Kidney Injury among Critically Ill Patients. Journal of Clinical Medicine. 2020; 9(6):1939. https://doi.org/10.3390/jcm9061939

Chicago/Turabian Style

Huang, Ya-Ting, Min-Yu Lai, Wei-Chih Kan, and Chih-Chung Shiao. 2020. "Independent Predictive Ability of Procalcitonin of Acute Kidney Injury among Critically Ill Patients" Journal of Clinical Medicine 9, no. 6: 1939. https://doi.org/10.3390/jcm9061939

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop