Abstract
Background
The C-reactive protein-to-platelet ratio (CPR) is a composite marker reflecting the interplay between inflammation and coagulation. This study aimed to investigate the association between CPR and the development of acute kidney injury (AKI) in postcardiac arrest (CA) patients.
Methods
We conducted a secondary analysis of previously published data from a cohort of cardiac arrest survivors admitted to the ICU. We employed univariate and multivariate logistic regression models, restricted cubic spline analysis, and subgroup analysis to elucidate the relationship between CPR and AKI. Receiver operating characteristic (ROC) curve analysis and decision curve analysis were performed to describe the discriminative ability and net benefit of Ln-CPR for AKI in an exploratory manner.
Results
A total of 309 patients were analyzed, of whom 163 (52.8%) developed CA-AKI. Multivariable analysis revealed that each 1-unit increase in Ln-CPR (log-transformed) was associated with 1.53 times the odds of CA-AKI (OR: 1.53, 95% CI: 1.24–1.90). Additionally, in the fully adjusted model, the highest Ln-CPR level was associated with higher odds of AKI compared to the lowest level (OR: 3.93, 95% CI: 1.78–8.68). Restricted cubic spline analysis demonstrated a linear association between Ln-CPR and the development of CA-AKI. The area under the ROC curve for Ln-CPR was 0.629 (95% CI: 0.567–0.690, p < 0.001). Exploratory decision curve analysis suggested only a marginal net benefit within a limited threshold probability range.
Conclusion
Elevated CPR was independently associated with an increased risk of AKI in ICU-admitted cardiac arrest survivors.
Supplementary Information
The online version contains supplementary material available at https://doi.org/10.1186/s12872-026-06246-1.
Keywords: Acute kidney injury, Cardiac arrest, C-reactive protein, Platelet
Introduction
Cardiac arrest (CA) is a life-threatening event with a low survival rate and generally poor prognosis, and is among the leading causes of mortality worldwide [1]. Despite successful resuscitation from cardiac arrest, the whole-body ischemia‒reperfusion response can trigger a cascade of pathophysiological processes that affect vital organs, including the brain, lungs, heart, and kidneys [2]. Acute kidney injury (AKI) is a prevalent early complication following cardiac arrest (CA), afflicting approximately 40%-50% of CA survivors [3, 4]. The development of AKI postcardiac arrest is significantly associated with an increased risk of mortality and poor neurological outcomes [5]. A large meta-analysis revealed that, compared with no AKI, post-AKI is associated with a doubled risk of mortality at both short-term and long-term follow-ups [3]. Therefore, early identification of patients at high risk for CA-associated AKI is critical to guide clinical management and improve prognostication.
Ischemia-reperfusion injury triggers a series of harmful responses, including endothelial injury, systemic inflammation, platelet activation, and microcirculatory dysfunction [6]. Within the kidney, hypoxia first damages the endothelium, initiating a strong inflammatory response. This triggers the release of proinflammatory cytokines and the influx of leukocytes, which increase vascular leakage and worsen tubular damage [7, 8]. Simultaneously, the injured endothelium promotes platelet aggregation. Combined with leukocyte adhesion and vascular congestion, this blocks microcirculatory perfusion and drives renal failure [9]. Furthermore, this pro-coagulation microenvironment disrupts endothelial junctions and encourages communication with tubular and dendritic cells, further fueling the inflammatory response [9]. Thus, coagulation and inflammation synergistically potentiate renal tissue damage.
C-reactive protein, an acute-phase reactant primarily produced by the liver in response to inflammatory stimuli, is widely recognized as an inflammatory biomarker [10, 11]. It has also been identified as an independent predictor of AKI for patients undergoing CABG [12]. Platelets serve as dual mediators of haemostasis and inflammation, driving the pathogenesis of AKI [13]. Accumulating evidence indicates that a reduced platelet count is significantly correlated with an elevated risk of AKI in neonates [14], patients with hemorrhagic shock [15], and those who have undergone CABG [16]. Given that both inflammatory and coagulation disturbances contribute concurrently to CA-AKI development, biomarkers assessing either pathway in isolation may incompletely characterize a patient’s risk. We therefore explored whether CPR, a composite marker of inflammatory and coagulopathic status [17], is associated with the risk of AKI in a cohort of cardiac arrest survivors admitted to the ICU.
Method
Study population
This study leveraged data from a previous ICU study at Erasme Hospital, Brussels, which was deposited in the Dryad Digital Repository (https://doi.org/10.5061/dryad.qv6fp83). The original investigation included comatose patients with Glasgow Coma Scale scores < 9 resulting from either in-hospital or out-of-hospital cardiac arrest (CA) between January 2007 and December 2015 [18]. Ethical approval was obtained from the local committee (P2017/264), with a waiver of informed consent due to the retrospective nature of the study. Survivors of CA received a standardized post resuscitation protocol involving targeted temperature management (TTM), deep sedation, shivering control, advanced hemodynamic monitoring, and neurological evaluation, as previously described [19]. This study conducted a secondary analysis on the dataset, applying refined exclusion criteria. Patients with incomplete liver function data or those who died within 24 h of ICU admission were excluded. For this analysis, we further excluded patients with a history of chronic renal failure, those with missing C-reactive protein or platelet count data, and those with incomplete AKI diagnosis following cardiac arrest. The process of selecting the study population is detailed in Fig. 1.
Fig. 1.
Flowchart of the study population with inclusion and exclusion criteria. Ln-CPR, log-transformed C-reactive protein to platelet ratio
Data collection
Patient demographics, chronic conditions, and arrest details were meticulously recorded. Demographics included age, sex, and weight. The chronic conditions included heart failure, hypertension, coronary artery disease, diabetes, chronic pulmonary obstructive disease/asthma, neurological disorders, chronic renal failure, and liver cirrhosis. Information on long-term corticosteroid or anticoagulant use was also collected. Arrest characteristics, location (in-hospital vs. out-of-hospital), witness status, bystander cardiopulmonary resuscitation, cardiac etiology, nonshockable rhythm, time to return of spontaneous circulation (ROSC), and epinephrine dosage. Blood samples were collected on ICU admission and once daily until ICU discharge. Our analysis was restricted to the following baseline biomarkers on admission: arterial blood gas parameters (pH, PaO2, PaCO2), platelet count, creatinine, C-reactive protein, and glucose. ICU interventions such as targeted temperature management (TTM) and continuous renal replacement therapy (CRRT) during the ICU stay were documented. All-cause ICU and in-hospital mortality—defined as death during the ICU stay and the entire hospitalization period, respectively—were recorded.
The CPR was calculated using C-reactive protein and platelet count data collected on ICU admission. Based on a previous study [17], CPR was determined using the following formula: 
.
Definition of acute kidney injury
Acute kidney injury was defined according to the Acute Kidney Injury Network (AKIN) criteria as stated in the original study and its relevant study [18, 19]. These criteria include an absolute increase in serum creatinine of ≥ 0.3 mg/dl within 48 h, a relative increase to 1.5 times the baseline level within 7 days, and/or a decrease in urine output to less than 0.5 ml/kg/hour for more than 6 h [20].
Statistical analysis
Due to the skewed, non-normal distribution of CPR, values were log-transformed (Ln-CPR) to approximate normality. Patients were categorized into Ln-CPR tertiles according to the 33rd and 66th percentiles (2.07 and 3.41, respectively).
Continuous variables were presented as mean ± standard deviation for normally distributed data or median (interquartile range, IQR) for non-normally distributed data, as determined by the Kolmogorov-Smirnov test. Group comparisons were performed using one-way ANOVA for normally distributed variables and the Kruskal-Wallis H test for non-normally distributed variables. Categorical variables are presented as counts and percentages, with group comparisons made via chi-square tests or Fisher’s exact tests when appropriate.
To identify potential risk factors for the development of AKI following cardiac arrest, univariate logistic regression analysis was initially conducted. A multivariable logistic regression model was subsequently employed to assess the independent association of Ln-CPR (as a continuous and categorical variable) with AKI. This analysis was stratified into four models to control for confounding variables: Model 1, unadjusted; Model 2, adjusted for demographic covariates (age, sex, and weight); Model 3, additionally adjusted for noncardiac etiology, time to ROSC, total epinephrine dose, creatinine, and glucose, which were selected based on a > 10% change in the effect estimate; Model 4, which included all Model 3 covariates plus additional variables with a univariate P-value < 0.1 (pH, corticosteroids, and coronary artery disease). Multicollinearity was assessed using variance inflation factors (VIF), and model fit was evaluated using the Hosmer-Lemeshow goodness-of-fit test.
The potential nonlinear relationship between Ln-CPR and AKI incidence post-cardiac arrest was explored via restricted cubic spline analysis.
The discriminative ability of Ln-CPR, C-reactive protein, and platelet counts for cardiac arrest-associated AKI was explored descriptively via receiver operating characteristic (ROC) curves, with the area under the curve (AUC) calculated for each marker. DeLong’s test was further used to compare the AUCs of Ln-CPR with those of C-reactive protein and platelet count, respectively. Furthermore, exploratory decision curve analysis (DCA) was conducted to describe the net benefit of Ln-CPR relative to C-reactive protein and platelet count alone for CA-AKI.
Interaction and stratified analyses were conducted to investigate whether the relationship between Ln-CPR and the risk of AKI varied across different subgroups.
All the statistical analyses were conducted via R statistical software, version 4.2.3. Statistical significance was set at a p-value of less than 0.05.
Results
Baseline characteristics of the study participants
Among the 435 patients assessed for eligibility, 61 were excluded because of early death or missing data, resulting in 374 cases for the prior study. Further exclusion of 65 patients with chronic renal failure or incomplete data reduced the final analysis cohort to 309 patients (Fig. 1). The study cohort was predominantly male, accounting for 227 (73.4%) of the patients. The overall incidence of CA-AKI was 52.8% (163 patients), and 12% (37 patients) required CRRT during their ICU stay.
The baseline characteristics of post-CA patients are detailed in Table 1 and were stratified by tertiles of admission Ln-CPR values: low (-1.47 to 2.07), middle (2.10 to 3.41), and high (3.42 to 7.98). No significant differences in demographic characteristics or comorbidities were observed among the three groups, with the exception of corticosteroid use and liver cirrhosis, which were more prevalent in the high Ln-CPR group. Admission laboratory findings revealed that patients with high Ln-CPR values were more likely to exhibit elevated C-reactive protein and reduced platelet counts, glucose, and PaO2 than were those with low and middle Ln-CPR values. Regarding patient outcomes, there were no significant differences in ICU stay, ICU mortality, or hospital mortality among the groups. However, patients in the high Ln-CPR group more frequently developed AKI and were more likely to receive CRRT during their ICU stay.
Table 1.
Baseline characteristics of patients stratified by tertile of the Ln-CPR
| Variables | Ln-CPR tertiles | p value | ||
|---|---|---|---|---|
| Low (n = 103) | Middle(n = 103) | High(n = 103) | ||
| DEMOGRAPHICS | ||||
| Age, years | 57.0 [50.5,70.0] | 62.0 [52.0,74.0] | 63.0 [51.5,73.0] | 0.2 |
| Male, n(%) | 83 (80.6%) | 72 (69.9%) | 72 (69.9%) | 0.134 |
| Weight, kg | 77.0 [70.0,84.5] | 80.0 [67.0,85.5] | 75.0 [70.0,85.0] | 0.968 |
| ARREST CHARACTERISTICS | ||||
| Witnessed CA, n(%) | 85 (82.5) | 86 (83.5) | 91 (88.3) | 0.459 |
| Bystander cardiopulmonary resuscitation, n(%) | 59 (57.3) | 68 (66.0) | 77 (74.8) | 0.03 |
| Out of hospital, n(%) | 87 (84.5) | 68 (66.0) | 37 (35.9) | < 0.001 |
| Noncardiac etiology, n(%) | 26 (25.2) | 43 (41.7) | 53 (51.5) | 0.001 |
| Nonshockable rhythm, n(%) | 42 (40.8) | 58 (56.3) | 71 (68.9) | < 0.001 |
| Time to ROSC, min | 15.0 [9.5,24.5] | 18.0 [10.0,30.5] | 12.0 [5.5,19.5] | 0.001 |
| Epinephrine total dose, mg | 3.0 [1.5, 6.0] | 4.0 [2.0, 6.0] | 3.0 [2.0, 4.5] | 0.192 |
| TTM, n(%) | 100 (97.1) | 91 (88.3) | 87 (84.5) | 0.008 |
| COMORBIDITIES | ||||
| Chronic heart failure, n(%) | 22 (21.4) | 14 (13.6) | 22 (21.4) | 0.257 |
| Hypertension, n(%) | 38 (36.9) | 37 (35.9) | 45 (43.7) | 0.46 |
| Coronary artery disease, n(%) | 50 (48.5) | 39 (37.9) | 35 (34.0) | 0.087 |
| Diabetes, n(%) | 14 (13.6) | 28 (27.2) | 24 (23.3) | 0.05 |
| COPD/asthma, n(%) | 13 (12.6) | 21 (20.4) | 19 (18.4) | 0.306 |
| Neurological disease, n(%) | 12 (11.7) | 18 (17.5) | 14 (13.6) | 0.476 |
| Liver cirrhosis, n(%) | 2 ( 1.9) | 1 ( 1.0) | 10 ( 9.7) | 0.003 |
| Corticosteroids, n(%) | 14 (13.6) | 20 (19.4) | 30 (29.1) | 0.021 |
| Chronic anticoagulation, n(%) | 11 (10.7) | 23 (22.3) | 19 (18.4) | 0.078 |
| BLOOD SAMPLE ON ADMISSION | ||||
| C-reactive protein, mg/L | 5.0 [2.0,10.0] | 33.0 [24.5,49.5] | 95.0 [65.5,150.0] | < 0.001 |
| Creatinine, mg/dl | 1.0 [0.9,1.3] | 1.1 [0.9, 1.4] | 1.1 [0.8,1.6] | 0.345 |
| Platelet counts, /mm3 | 235.0 [171.0,316.0] | 211.0 [160.5,252.0] | 154.0 [83.5,206.5] | < 0.001 |
| Glucose, mg/dl | 231.0 [185.0,305.5] | 224.0 [174.5,313.5] | 175.0 [142.0,259.0] | 0.001 |
| pH | 7.3 [7.2,7.4] | 7.3 [7.2,7.4] | 7.3 [7.2,7.4] | 0.592 |
| PaO2,mmHg | 134.0 [90.5,191.5] | 113.0 [85.5,181.0] | 105.0 [82.0,160.0] | 0.047 |
| PaCO2,mmHg | 38.0 [33.0,44.5] | 36.0 [32.0,43.0] | 39.0 [33.0,44.0] | 0.207 |
| CPR | 2.6 [0.9, 5.1] | 16.9 [12.5,22.1] | 63.2 [42.0,99.4] | < 0.001 |
| Ln-CPR | 0.73 ± 0.99 | 2.81 ± 0.37 | 4.34 ± 0.88 | < 0.001 |
| EVENTS | ||||
| AKI, n(%) | 41 (39.8) | 55 (53.4) | 67 (65.0) | 0.001 |
| CRRT, n (%) | 4 (3.9) | 10 (9.7) | 23 (22.3) | < 0.001 |
| ICU length of stay, days | 4.0 [2.5, 8.0] | 4.0 [2.0,8.0] | 5.0 [2.0,11.0] | 0.475 |
| ICU mortality, n(%) | 46 (44.7) | 55 (53.4) | 57 (55.3) | 0.263 |
| Hospital mortality, n(%) | 48 (46.6) | 63 (61.2) | 60 (58.3) | 0.084 |
CA Cardiac arrest, ROSC Return of spontaneous circulation, TTM Targeted temperature management, ICU Intensive care unit, COPD Chronic obstructive pulmonary disease, CPR C-reactive protein-to-platelet ratio, Ln-CPR Log-transformed C-reactive protein-to-platelet ratio, AKI Acute kidney injury, CRRT Continuous renal replacement therapy
Relationship between CPR and CA-AKI
The results of the univariate logistic regression analysis presented in Table 2 revealed several factors significantly associated with the development of AKI: high level of Ln-CPR, time to ROSC, total epinephrine dosage, corticosteroid use, creatinine levels at admission, and glucose levels. Patients in the high Ln-CPR group had a 2.81-fold greater risk of AKI than those in the low Ln-CPR group did (OR: 2.81, 95% CI: 1.60–4.95, p < 0.001).
Table 2.
Univariate logistic regression analysis for risk factors of CA-AKI development
| Variables | CA-AKI | ||
|---|---|---|---|
| OR | 95%CI | p value | |
| Ln-CPR groups | |||
| Low | reference | ||
| Middle | 1.73 | 1.00-3.01 | 0.051 |
| High | 2.81 | 1.60–4.95 | < 0.001 |
| Age, year | 1.01 | 0.99–1.02 | 0.423 |
| Male | 1.42 | 0.85–2.36 | 0.176 |
| Weight, kg | 1.01 | 1.00-1.03 | 0.137 |
| Witnessed arrest | 1.47 | 0.78–2.74 | 0.231 |
| Bystander cardiopulmonary resuscitation | 1.08 | 0.68–1.74 | 0.738 |
| Out of hospital | 0.75 | 0.47–1.19 | 0.215 |
| Noncardiac etiology | 0.63 | 0.40-1.00 | 0.052 |
| Nonshockable rhythm | 1.16 | 0.74–1.82 | 0.522 |
| Time to ROSC, min | 1.02 | 1.01–1.04 | 0.006 |
| Epinephrine total dose, mg | 1.1 | 1.03–1.18 | 0.006 |
| TTM | 0.68 | 0.32–1.45 | 0.318 |
| Chronic heart failure | 1.46 | 0.82–2.62 | 0.2 |
| Hypertension | 1.1 | 0.69–1.74 | 0.691 |
| Coronary artery disease | 1.51 | 0.95–2.39 | 0.078 |
| Diabetes | 1.01 | 0.59–1.75 | 0.959 |
| COPD/asthma | 0.84 | 0.46–1.51 | 0.554 |
| Neurological disease | 0.71 | 0.37–1.35 | 0.297 |
| Liver cirrhosis | 1.46 | 0.47–4.55 | 0.519 |
| Corticosteroids | 2.8 | 1.53–5.09 | < 0.001 |
| Chronic Anticoagulation | 1.32 | 0.73–2.41 | 0.359 |
| Creatinine, mg dl− 1 | 19.28 | 8.39–44.30 | < 0.001 |
| Glucose, mg dl− 1 | 1 | 1.00–1.00 | 0.007 |
| pH | 0.19 | 0.03–1.14 | 0.069 |
| PaO2,mmHg | 1 | 1.00–1.00 | 0.63 |
| PaCO2,mmHg | 0.99 | 0.97–1.01 | 0.218 |
Ln-CPR Log-transformed C-reactive protein-to-platelet ratio, ROSC Return of spontaneous circulation, TTM Targeted temperature management, COPD Chronic obstructive pulmonary disease
Table 3 displays the outcomes of the multivariate logistic regression analysis. Ln-CPR was significantly associated with AKI risk across all models. When analyzed continuously, each 1-unit increase in Ln-CPR was associated with an OR of 1.53(95% CI: 1.24–1.90) in Model 4. In the fully adjusted categorical analyses, the high Ln-CPR group exhibited higher odds of AKI than the low group when dichotomized (OR: 2.22, 95% CI: 1.21–4.05). Furthermore, compared to the lowest categories, the highest categories had significantly increased odds of AKI in both tertile (OR: 3.93, 95% CI: 1.78–8.68) and quartile analyses (Q4 vs. Q1: OR: 5.52, 95% CI: 2.11–14.45). There was a significant linear trend between increasing Ln-CPR and the risk of AKI both in tertiles and quartiles (P for trend < 0.001). The multivariable model showed no evidence of multicollinearity (all VIF < 3; Supplementary Table S1). The Hosmer-Lemeshow test indicated adequate model fit (P = 0.995).
Table 3.
Multivariate logistic regression analysis of the relationship between Ln-CPR and CA-AKI
| Exposure | Model 1 | Model 2 | Model 3 | Model 4 | ||||
|---|---|---|---|---|---|---|---|---|
| OR(95%CI) | p value | OR(95%CI) | p value | OR(95%CI) | p value | OR(95%CI) | p value | |
| Ln-CPR | 1.35 (1.17, 1.56) | <0.0001 | 1.36 (1.17, 1.58) | <0.0001 | 1.54 (1.25, 1.90) | < 0.0001 | 1.53 (1.24, 1.90) | < 0.0001 |
| Ln-CPR dichotomization | ||||||||
| Low | 1(ref) | 1(ref) | 1(ref) | 1(ref) | ||||
| High | 1.90 (1.21, 2.98) | 0.0055 | 1.97 (1.24, 3.11) | 0.0039 | 2.29 (1.27, 4.13) | 0.0059 | 2.22 (1.21, 4.05) | 0.0097 |
| Ln-CPR tertiles | ||||||||
| Low | 1(ref) | 1(ref) | 1(ref) | 1(ref) | ||||
| Middle | 1.73 (1.00, 3.01) | 0.0513 | 1.79 (1.02, 3.14) | 0.0439 | 1.73 (0.84, 3.55) | 0.1349 | 1.71 (0.82, 3.57) | 0.1552 |
| High | 2.81 (1.60, 4.95) | 0.0003 | 2.92 (1.64, 5.21) | 0.0003 | 4.17 (1.92, 9.03) | 0.0003 | 3.93 (1.78, 8.68) | 0.0007 |
| p for trend | 0.0003 | 0.0003 | 0.0003 | 0.0008 | ||||
| Ln-CPR quartiles | ||||||||
| Q1 | 1(ref) | 1(ref) | 1(ref) | 1(ref) | ||||
| Q2 | 1.99 (1.05, 3.80) | 0.0361 | 1.87 (0.96, 3.65) | 0.0662 | 2.20 (0.97, 5.00) | 0.0601 | 2.13 (0.92, 4.95) | 0.0794 |
| Q3 | 2.10 (1.10, 4.01) | 0.0243 | 2.12 (1.10, 4.08) | 0.0248 | 2.54 (1.11, 5.82) | 0.0273 | 2.49 (1.07, 5.83) | 0.0351 |
| Q4 | 3.50 (1.81, 6.78) | 0.0002 | 3.50 (1.78, 6.88) | 0.0003 | 5.77 (2.30, 14.51) | 0.0002 | 5.52 (2.11, 14.45) | 0.0005 |
| p for trend | 0.0003 | 0.0003 | 0.0003 | 0.0007 | ||||
Model 1: adjusted for none
Model 2: adjusted for male, age, and weight
Model 3: adjusted for male, age, weight, noncardiac etiology, time to ROSC, epinephrine total dose, creatinine, and glucose
Model 4: adjusted for male, age, weight, noncardiac etiology, time to ROSC, epinephrine total dose, creatinine, glucose, pH, corticosteroids, and coronary artery disease
Ln-CPR Log-transformed C-reactive protein-to-platelet ratio
A restricted cubic splines regression model (Fig. 2) indicated that the risk for AKI increased linearly with increasing Ln-CPR (P for nonlinearity = 0.509).
Fig. 2.
Restricted cubic splines curves for the relationship between Ln-CPR and CA-AKI. Ln-CPR, log-transformed C-reactive protein-to-platelet ratio
Exploratory ROC and decision curve analyses
The ROC curve analysis revealed that the AUCs for CPR, C-reactive protein, and platelet count were 0.629 (95% CI 0.567–0.690), 0.614 (95% CI 0.551–0.677), and 0.578 (95% CI 0.514–0.641), respectively. The AUCs of CPR, C-reactive protein and platelet count are shown in Fig. 3. Furthermore, DeLong’s test revealed no significant difference in the AUC between CPR and C-reactive protein (p = 0.2464) or between CPR and platelet count (p = 0.1487).
Fig. 3.
ROC curves of the Ln-CPR, C-reactive protein, and platelet count for CA-AKI. Ln-CPR, log-transformed C-reactive protein-to-platelet ratio; PLT, platelet; CRP, C-reactive protein
Exploratory decision curve analysis suggested a marginal and clinically uncertain net benefit difference between Ln-CPR and C-reactive protein or platelet count alone within a limited threshold probability range of 0.6 to 0.8 (Fig. 4).
Fig. 4.
Exploratory decision curve analysis of CPR, C-reactive protein, and PLT for CA-AKI. Ln-CPR, log-transformed C-reactive protein-to-platelet ratio; PLT, platelet; CRP, C-reactive protein
Subgroup analysis
We performed exploratory subgroup analyses to examine whether the association between the Ln-CPR dichotomous group and CA-AKI risk differed across various clinical subgroups. Subgroup analyses were stratified by gender, hypertension, diabetes, coronary artery disease, chronic heart failure, bystander cardiopulmonary resuscitation, out-of-hospital cardiac arrest, non-cardiac etiology, and non-shockable rhythm (Fig. 5). There were no statistically significant differences in the strength of the association between Ln-CPR and the outcome among different clinical subgroups (all P for interaction > 0.05).
Fig. 5.
Forest plot investigating the association between Ln-CPR and CA-AKI in different subgroups
Discussion
In this secondary analysis of 309 cardiac arrest survivors admitted to the ICU, we found that elevated CPR at admission was independently associated with an increased risk of AKI. This association was linear, consistent across subgroups, and persisted after adjustment for demographic and clinical covariates. However, CPR demonstrated only modest discriminative ability and did not significantly outperform C-reactive protein or platelet count alone. These findings should be regarded as hypothesis-generating rather than evidence for clinical application.
CPR is a composite marker that reflects the interplay between systemic inflammation and coagulopathic status. Elevated CPR has been observed across diverse critical-illness settings characterized by inflammatory and coagulopathic activation. In neonatal sepsis, CPR has been associated with disease severity and culture-positive infection [17, 21]; in patients with pyogenic liver abscess, it has shown utility in predicting sepsis risk [22]. Similarly, other combined inflammatory and platelet-derived indices, such as the neutrophil-lymphocyte-platelet ratio, have been associated with AKI following cardiovascular and major abdominal surgery [23, 24]. In the present study, we found that elevated CPR was associated with an increased risk of AKI following cardiac arrest. These findings collectively suggest that the association between CPR and AKI likely reflects a general inflammatory-coagulopathic state common to sepsis, major surgery, and postcardiac arrest syndrome, rather than a pathophysiology unique to cardiac arrest-associated AKI.
The underlying mechanism linking CPR to CA-AKI remains unclear. Cardiac arrest triggers whole-body ischemia-reperfusion injury, which initiates a systemic inflammatory response, characterized by cytokine release, endothelial activation, and coagulation cascade activation [25, 26]. Within this cascade, C-reactive protein and platelets engage in a bidirectional interaction. C-reactive protein is rapidly released from the liver during inflammation [27, 28], and can upregulate P-selectin expression on platelets, promoting platelet activation and adhesion [29]. Conversely, activated platelets can dissociate pentameric C-reactive protein into its monomeric isoform, which exerts potent proinflammatory and prothrombotic effects [29, 30]. This reciprocal interplay results in elevated C-reactive protein and reduced platelet counts, forming the basis of CPR. Although our study demonstrated an association between elevated CPR and increased CA-AKI risk, no direct causal link has been established between CPR and renal tubular necrosis or functional impairment. Elevated CPR likely reflects the nonspecific systemic inflammatory and coagulopathic response characteristic of postcardiac arrest syndrome, rather than representing a distinct pathophysiological mechanism specific to CA-AKI.
Kidney-specific biomarkers such as NGAL, cystatin C, KIM-1, and the urinary [TIMP-2]•[IGFBP7] index have been evaluated for AKI prediction after cardiac arrest [31–33]. They reflect renal tubular injury or stress more directly than CPR, while these markers remain limited by availability, cost, and heterogeneous performance. In contrast, C-reactive protein and platelet count are universally available and inexpensive. However, the discriminative performance of CPR for CA-AKI was modest, with no statistically significant improvement over C-reactive protein alone or platelet count alone by DeLong’s test. Decision curve analysis indicated a marginally higher net benefit for CPR compared with its individual components, but this was confined to a narrow threshold probability range of 0.6–0.8, the clinical relevance of which is uncertain. These findings do not support the use of CPR as a clinically useful predictive biomarker for CA-AKI. Whether CPR could contribute to multimodal risk assessment frameworks, in combination with clinical variables and kidney-specific biomarkers, warrants investigation in prospective studies specifically designed for biomarker evaluation.
Limitations
Given the secondary nature of our analysis using a limited dataset from a prior publication, several important limitations warrant discussion. First, the dataset lacked several potentially important variables, including infection diagnoses and inflammatory status prior to arrest. These unmeasured confounders may substantially influence both C-reactive protein and platelet count, independent of postcardiac arrest pathophysiology. Second, patients in the highest Ln-CPR tertile differed substantially from those in lower tertiles with respect to arrest characteristics (lower rates of out-of-hospital arrest, higher rates of non-cardiac etiology and non-shockable rhythm) and comorbidities (higher prevalence of liver cirrhosis and corticosteroid use). These baseline imbalances raise the concern that admission CPR may partly reflect pre-existing systemic inflammation, coagulopathy, and critical illness severity rather than a cardiac arrest-specific pathophysiological process. Consequently, the specificity of CPR for CA-AKI remains uncertain. Third, baseline creatinine was defined as the first serum creatinine measurement obtained upon ICU admission. Because AKI was defined partly based on subsequent creatinine changes within 48 h, some patients may have already had early AKI at the time of this measurement. This temporal overlap creates uncertainty regarding whether the association between admission CPR and AKI reflects a true predictive relationship or concurrent pathophysiology. Fourth, the diagnosis of AKI was based on the AKIN criteria rather than the current KDIGO guidelines, which may limit comparability with more recent studies. Due to limitations in the original dataset, the exact timing of AKI onset, AKI staging, and the diagnostic criteria for chronic renal failure could not be determined. Fifth, the study was conducted at a single center in Europe between 2007 and 2015. Postcardiac arrest care, particularly target temperature management strategies and resuscitation guidelines, has evolved substantially during this period. Whether these findings remain applicable to contemporary practice is uncertain. Future prospective studies, specifically designed to evaluate renal biomarkers in postcardiac arrest patients and incorporating comprehensive clinical data, are needed to validate these observations.
Conclusion
In conclusion, our study demonstrated that CPR levels are significantly elevated in CA survivors who develop AKI. Furthermore, after adjusting for potential confounders, CPR remained independently associated with the incidence of CA-AKI.
Supplementary Information
Acknowledgements
Not applicable.
Clinical trial number
Not applicable.
Abbreviations
- AKI
Acute kidney injury
- AKIN
Acute Kidney Injury Network
- AUC
Area under the curve
- CA
Cardiac arrest
- CABG
Coronary artery bypass grafting
- CPR
C-reactive protein to platelet ratio
- COPD
Chronic obstructive pulmonary disease
- ICU
Intensive care unit
- Ln-CPR
Log-transformed C-reactive protein to platelet ratio
- PCI
Percutaneous coronary intervention
- ROSC
Time to return of spontaneous circulation
- ROC
Receiver operating characteristic
- TTM
Target temperature management
- CRRT
Continuous renal replacement therapy
Authors’ contributions
Conceptualization and writing-review & editing by CJZ; Methodology and writing—original draft by YJL; Formal analysis by LMC; Project administration and writing-review & editing by LYL; Investigation and Data curation by BCC. All authors reviewed the results and approved the final version of the manuscript.
Funding
Not applicable.
Data availability
The datasets generated and/or analysed during the current study are available in the Dryad Digital Repository (https://doi.org/10.5061/dryad.qv6fp83).
Declarations
Ethics approval and consent to participate
The local ethical committee (Comite´ d’Ethique Hospitalo Facultaire Erasme-ULB) approved the study (P2017/264).
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Lin L, Chen L, Jiang Y, Gao R, Wu Z, Lv W, Xie Y. Construction and validation of a risk prediction model for acute kidne y injury in patients after cardiac arrest. Ren Fail. 2023;45(2):2285865. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Dalessio L. Post-Cardiac Arrest Syndrome. AACN Adv Crit Care. 2020;31(4):383–93. [DOI] [PubMed] [Google Scholar]
- 3.Prasitlumkum N, Cheungpasitporn W, Sato R, Chokesuwattanaskul R, Thongprayoon C, Patlolla SH, Bathini T, Mao MA, Rab ST, Kashani K, et al. Acute kidney injury and cardiac arrest in the modern era: an updated systematic review and meta-analysis. Hosp Pract (1995). 2021;49(4):280–91. [DOI] [PubMed] [Google Scholar]
- 4.Sandroni C, Dell’anna AM, Tujjar O, Geri G, Cariou A, Taccone FS. Acute kidney injury after cardiac arrest: a systematic review and meta-analysis of clinical studies. Minerva Anestesiol. 2016;82(9):989–99. [PubMed] [Google Scholar]
- 5.Dutta A, Hari KJ, Azizian J, Masmoudi Y, Khalid F, Kowal JL, Ahmad MI, Majeed M, Macdonald L, Sunkara P, et al. Incidence, Predictors, and Prognosis of Acute Kidney Injury Among Cardiac Arrest Survivors. J Intensive Care Med. 2021;36(5):550–6. [DOI] [PubMed] [Google Scholar]
- 6.Lazzarin T, Tonon CR, Martins D, Fávero EL Jr., Baumgratz TD, Pereira FWL, Pinheiro VR, Ballarin RS, Queiroz DAR, Azevedo PS et al. Post-Cardiac Arrest: Mechanisms, Management, and Future Perspectives. J Clin Med. 2022;12(1):259–84. [DOI] [PMC free article] [PubMed]
- 7.Ostermann M, Liu K. Pathophysiology of AKI. Best Pract Res Clin Anaesthesiol. 2017;31(3):305–14. [DOI] [PubMed] [Google Scholar]
- 8.Ortega-Trejo JA, Bobadilla NA. Is Renal Ischemic Preconditioning an Alternative to Ameliorate the Short- and Long-Term Consequences of Acute Kidney Injury? Int J Mol Sci. 2023;24(9):8345. [DOI] [PMC free article] [PubMed]
- 9.Sharfuddin AA, Molitoris BA. Pathophysiology of ischemic acute kidney injury. Nat Rev Nephrol. 2011;7(4):189–200. [DOI] [PubMed] [Google Scholar]
- 10.Li J, Chen J, Lan HY, Tang Y. Role of C-Reactive Protein in Kidney Diseases. Kidney Dis (Basel). 2023;9(2):73–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Tang Y, Mak SK, Xu AP, Lan HY. Role of C-reactive protein in the pathogenesis of acute kidney injury. Nephrol (Carlton). 2018;23(Suppl 4):50–2. [DOI] [PubMed] [Google Scholar]
- 12.Han SS, Kim DK, Kim S, Chin HJ, Chae DW, Na KY. C-Reactive Protein Predicts Acute Kidney Injury and Death After Coronary Artery Bypass Grafting. Ann Thorac Surg. 2017;104(3):804–10. [DOI] [PubMed] [Google Scholar]
- 13.Jansen MPB, Florquin S, Roelofs J. The role of platelets in acute kidney injury. Nat Rev Nephrol. 2018;14(7):457–71. [DOI] [PubMed] [Google Scholar]
- 14.Lyu D, Fu S. Association between platelet count and neonatal acute kidney injury: a cohort study using the medical information mart for intensive care III database. J Matern Fetal Neonatal Med. 2024;37(1):2379910. [DOI] [PubMed] [Google Scholar]
- 15.Wu M, Luan YY, Lu JF, Li H, Zhan HC, Chen YH, Zhang F, Tian YY, Yang ZL, Yao YM, et al. Platelet count as a new biomarker for acute kidney injury induced by hemorrhagic shock. Platelets. 2020;31(1):94–102. [DOI] [PubMed] [Google Scholar]
- 16.Hui S, Yu C. Platelet Counts, Acute Kidney Injury, and Mortality after Coronary Artery Bypass Grafting Surgery. Anesthesiology. 2016;125(2):437–8. [DOI] [PubMed] [Google Scholar]
- 17.Li X, Li T, Wang J, Feng Y, Ren C, Xu Z, Yang J, Zhang Q, An C. Clinical Value of C-Reactive Protein/Platelet Ratio in Neonatal Sepsis: A Cross-Sectional Study. J Inflamm Res. 2021;14:5123–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Iesu E, Franchi F, Zama Cavicchi F, Pozzebon S, Fontana V, Mendoza M, Nobile L, Scolletta S, Vincent JL, Creteur J, et al. Acute liver dysfunction after cardiac arrest. PLoS ONE. 2018;13(11):e0206655. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tujjar O, Mineo G, Dell’Anna A, Poyatos-Robles B, Donadello K, Scolletta S, Vincent JL, Taccone FS. Acute kidney injury after cardiac arrest. Crit Care. 2015;19(1):169. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Mehta RL, Kellum JA, Shah SV, Molitoris BA, Ronco C, Warnock DG, Levin A. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care. 2007;11(2):R31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Aggarwal S, Sangle AL, Siddiqui MS, Haseeb M, Engade MB. An Observational Study on C-Reactive Protein to Platelet Ratio in Neonatal Sepsis. Cureus. 2024;16(6):e62230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Li S, Yu S, Qin J, Peng M, Qian J, Zhou P. Prognostic value of platelet count-related ratios on admission in patients with pyogenic liver abscess. BMC Infect Dis. 2022;22(1):636. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Koo CH, Eun Jung D, Park YS, Bae J, Cho YJ, Kim WH, Bahk JH. Neutrophil, Lymphocyte, and Platelet Counts and Acute Kidney Injury After Cardiovascular Surgery. J Cardiothorac Vasc Anesth. 2018;32(1):212–22. [DOI] [PubMed] [Google Scholar]
- 24.Gameiro J, Fonseca JA, Dias JM, Milho J, Rosa R, Jorge S, Lopes JA. Neutrophil, lymphocyte and platelet ratio as a predictor of postoperative acute kidney injury in major abdominal surgery. BMC Nephrol. 2018;19(1):320. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Adrie C, Adib-Conquy M, Laurent I, Monchi M, Vinsonneau C, Fitting C, Fraisse F, Dinh-Xuan AT, Carli P, Spaulding C, et al. Successful cardiopulmonary resuscitation after cardiac arrest as a sepsis-like syndrome. Circulation. 2002;106(5):562–8. [DOI] [PubMed] [Google Scholar]
- 26.Kim YW, Cha KC, Cha YS, Kim OH, Jung WJ, Kim TH, Han BK, Kim H, Lee KH, Choi E, et al. Shock duration after resuscitation is associated with occurrence of post-cardiac arrest acute kidney injury. J Korean Med Sci. 2015;30(6):802–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.VanDevanter DR, Heltshe SL, Skalland M, West NE, Sanders DB, Goss CH, Flume PA. C-reactive protein (CRP) as a biomarker of pulmonary exacerbation presentation and treatment response. J Cyst Fibros. 2022;21(4):588–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Zhou HH, Tang YL, Xu TH, Cheng B. C-reactive protein: structure, function, regulation, and role in clinical diseases. Front Immunol. 2024;15:1425168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Siennicka A. Association between microvesicles bearing monomeric C-reactive protein and platelet reactivity. Relationship with low response to antiplatelet drugs? J Physiol Pharmacol. 2021;72(1):3–12. [DOI] [PubMed]
- 30.Eisenhardt SU, Habersberger J, Murphy A, Chen YC, Woollard KJ, Bassler N, Qian H, von Zur Muhlen C, Hagemeyer CE, Ahrens I, et al. Dissociation of pentameric to monomeric C-reactive protein on activated platelets localizes inflammation to atherosclerotic plaques. Circ Res. 2009;105(2):128–37. [DOI] [PubMed] [Google Scholar]
- 31.Berlin N, Pawar RD, Liu X, Balaji L, Morton AC, Silverman J, Li F, Issa MS, Roessler LL, Holmberg MJ, et al. Kidney-specific biomarkers for predicting acute kidney injury following cardiac arrest. Resuscitation. 2023;190:109911. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Adler C, Heller T, Schregel F, Hagmann H, Hellmich M, Adler J, Reuter H. TIMP-2/IGFBP7 predicts acute kidney injury in out-of-hospital cardiac arrest survivors. Crit Care. 2018;22(1):126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Titeca-Beauport D, Daubin D, Chelly J, Zerbib Y, Brault C, Diouf M, Slama M, Vinsonneau C, Klouche K, Maizel J. The urine biomarkers TIMP2 and IGFBP7 can identify patients who will experience severe acute kidney injury following a cardiac arrest: A prospective multicentre study. Resuscitation. 2019;141:104–10. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated and/or analysed during the current study are available in the Dryad Digital Repository (https://doi.org/10.5061/dryad.qv6fp83).