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

Machine learning-based model to predict severe acute kidney injury after total aortic arch replacement for acute type A aortic dissection

Background

Severe acute kidney injury (AKI) after total aortic arch replacement (TAAR) is related to adverse outcomes in patients with acute type A aortic dissection (ATAAD). However, the early prediction of severe AKI remains a challenge. This study aimed to develop a novel model to predict severe AKI after TAAR in ATAAD patients using machine learning algorithms.

Methods

A total of 572 ATAAD patients undergoing TAAR were enrolled in this retrospective study, and randomly divided into a training set (70 %) and a validation set (30 %). Lasso regression, support vector machine-recursive feature elimination and random forest algorithms were used to screen indicators for severe AKI (defined as AKI stage III) in the training set, respectively. Then the intersection indicators were selected to construct models through artificial neural network (ANN) and logistic regression. The AUC-ROC curve was employed to ascertain the prediction efficacy of the ANN and logistic regression models.

Results

The incidence of severe AKI after TAAR was 22.9 % among ATAAD patients. The intersection predictors identified by different machine learning algorithms were baseline serum creatinine and ICU admission variables, including serum cystatin C, procalcitonin, aspartate transaminase, platelet, lactic dehydrogenase, urine N-acetyl-β-d-glucosidase and Acute Physiology and Chronic Health Evaluation II score. The ANN model showed a higher AUC-ROC than logistic regression (0.938 vs 0.908, p < 0.05). Furthermore, the ANN model could predict 89.1 % of severe AKI cases beforehand. In the validation set, the superior performance of the ANN model was further confirmed in terms of discrimination ability (AUC = 0.916), calibration curve analysis and decision curve analysis.

Conclusion

This study developed a novel and reliable clinical prediction model for severe AKI after TAAR in ATAAD patients using machine learning algorithms. Importantly, the ANN model showed a higher predictive ability for severe AKI than logistic regression.

Sepsis and Acute Kidney Injury: A Review Focusing on the Bidirectional Interplay

Although sepsis and acute kidney injury (AKI) have a bidirectional interplay, the pathophysiological mechanisms between AKI and sepsis are not clarified and worthy of a comprehensive and updated review. The primary pathophysiology of sepsis-associated AKI (SA-AKI) includes inflammatory cascade, macrovascular and microvascular dysfunction, cell cycle arrest, and apoptosis. The pathophysiology of sepsis following AKI contains fluid overload, hyperinflammatory state, immunosuppression, and infection associated with kidney replacement therapy and catheter cannulation. The preventive strategies for SA-AKI are non-specific, mainly focusing on infection control and preventing further kidney insults. On the other hand, the preventive strategies for sepsis following AKI might focus on decreasing some metabolites, cytokines, or molecules harmful to our immunity, supplementing vitamin D3 for its immunomodulation effect, and avoiding fluid overload and unnecessary catheter cannulation. To date, several limitations persistently prohibit the understanding of the bidirectional pathophysiologies. Conducting studies, such as the Kidney Precision Medicine Project, to investigate human kidney tissue and establishing parameters or scores better to determine the occurrence timing of sepsis and AKI and the definition of SA-AKI might be the prospects to unveil the mystery and improve the prognoses of AKI patients.

Clinical Features and Outcomes of COVID-19 Patients with Acute Kidney Injury and Acute Kidney Injury on Chronic Kidney Disease

COVID-19 emerged in Wuhan in December 2019 and soon became a worldwide pandemic. We collected and analyzed the data from 1077 patients with COVID-19 who were admitted to the west campus of Wuhan Union Hospital from January 16 to April 16, 2020. Sixty (5.6%) of the 1077 COVID-19 patients were diagnosed with acute kidney injury (AKI) during hospitalization, and 18 of them (30%) had AKI on chronic kidney disease (AKI/CKD). COVID-19 patients with AKI had a worse prognosis, with higher intensive care unit (ICU) admission (28.3%) and fatality (65%) rates than patients without AKI (3.4% and 10.7%, respectively). Among the COVID-19 patients, AKI was more likely to occur in male patients, the elderly, patients with more severe disease states and those with comorbidities (such as hypertension, diabetes, coronary heart disease (CHD), chronic obstructive pulmonary disease (COPD) and CKD). COVID-19 patients with AKI were more likely to develop respiratory failure, gastrointestinal bleeding, acute liver injury, acute myocardial injury, heart failure, acute respiratory distress syndrome (ARDS), cerebrovascular accident, and disseminated intravascular coagulation (DIC) than those without AKI. Compared with patients without AKI, COVID-19 patients with AKI had lower platelet counts, lymphocyte counts, albumin levels and serum calcium levels but had elevated leukocyte counts, neutrophil counts and serum potassium levels. Inflammatory indicators, such as C-reactive protein (CRP), interleukin-6 (IL-6), and procalcitonin (PCT), were significantly higher in patients with AKI than in those without AKI. COVID-19 patients with AKI also exhibited a longer prothrombin time (PT), a longer activated partial thromboplastin time (APTT), and a higher D-dimer level than those without AKI. Survival analysis revealed that COVID-19 patients with AKI had a reduced survival rate compared with those without AKI. Furthermore, COVID-19 patients with AKI/CKD had a lower survival rate than those with AKI or CKD only. Multiple logistic regression indicated that the predictors of AKI in COVID-19 patients included complications, such as respiratory failure and acute myocardial injury, and higher creatinine and PCT levels during hospitalization.

Predictive Ability of Procalcitonin for Acute Kidney Injury: A Narrative Review Focusing on the Interference of Infection

Acute kidney injury (AKI) is a common yet complicated clinical entity with high morbidity and mortality. An essential strategy to improve AKI patients’ prognoses is finding optimal biomarkers to identify AKI in a timely manner. Procalcitonin (PCT), a well-recognized biomarker for diagnosing infection and guiding antibiotics therapy, has been proposed to predict AKI development and recovery in many clinical settings. The current review provides comprehensive and updated information from relevant studies to evaluate PCT’s AKI-predictive ability and the influence of infection on this predictive ability. PCT has demonstrated optimal predictive ability for AKI in various populations irrespective of infection. However, the predictive ability seems to be blunted by infection since infection and inflammation have a more potent influence than AKI on PCT elevation. We furthermore explain the complicated association between elevated PCT levels and AKI in infection and inflammation situations and recommend directions for further investigations to clarify the essential issue. In conclusion, although conflicting data exist, serum PCT level is a potential biomarker for predicting AKI in many clinical settings regardless of infection. Nevertheless, further studies are warranted to clarify the association between PCT, infection, and AKI and to confirm the utilization of PCT for AKI prediction.

Analysis of the diagnostic capabilities of urinary neutrophil gelatinase-associated lipocalin and serum procalcitonin for acute kidney injury at the early stage of critical care intensive care unit admission

BACKGROUND

Acute kidney injury (AKI) is a disease that negatively affects patient prognosis and requires early diagnosis and treatment. Biomarkers that predict AKI are needed for early diagnosis of this disease.

METHODS

We compared the AKI group and the non-AKI group in patients who were admitted to our critical care intensive care unit (ICU) and conducted a comparative study focusing on urinary neutrophil gelatinase-associated lipocalin (U-NGAL) and serum procalcitonin (PCT).

RESULTS

Seventy-one out of 106 ICU inpatients were diagnosed with AKI in accordance with the Kidney Disease: Improving Global Outcomes (KDIGO) criteria. Among the patients who were diagnosed with AKI stages 1 to 3, 94.4% of all patients reached the maximum stage by day 5 after admission. Comparing the non-AKI group and AKI stage 1 to 3 on days 1 to 3 after admission, U-NGAL and PCT levels in the stage 3 group were significantly higher than those in the non-AKI group. Additionally, in receiver operating characteristic curve (ROC) analysis on days 1-3 after admission, U-NGAL and PCT levels can be used as biomarkers for the diagnosis of AKI, and in particular, AKI stage 3 can be predicted and diagnosed with high accuracy. U-NGAL and PCT levels were also significantly higher in AKI due to sepsis and acute pancreatitis and due to sepsis, respectively.

CONCLUSIONS

Measuring U-NGAL and PCT levels as biomarkers for AKI may further improve the accuracy of AKI diagnosis in critical care ICU.

A risk score based on procalcitonin for predicting acute kidney injury in COVID-19 patients

Background

Acute kidney injury (AKI) has been reported developing commonly in coronavirus disease 2019 (COVID‐19) patients and could increase the risk of poor outcomes in these patients. We design this study to explore the value of serum procalcitonin (PCT) on predicting AKI and construct risk score for predicting AKI in COVID‐19 patients.

Methods

Patients diagnosed with COVID‐19 and hospitalized in Renmin Hospital of Wuhan University between January 30 and February 24, 2020, were included. The least absolute shrinkage and selection operator (LASSO) regression was performed to identify the strongest predictors of AKI. Multivariate logistic regression analysis was conducted to find independent risk factors for AKI and construct risk score using odds ratio (OR) value of those risk factors. Receiver operating characteristics (ROC) curves were plotted, and area under the ROC curve (AUC) value was calculated to evaluate the predictive value of single PCT level and the constructed risk score.

Results

Among 389 included COVID‐19 patients, 28 (7.2%) patients developed AKI. LASSO regression showed hypertension, saturation of arterial oxygen (SaO₂), PCT, and blood urea nitrogen (BUN) were the strongest predictors for AKI. After multivariate logistic regression analysis, only SaO₂ (<0.001), PCT (p = 0.004), and BUN (p = 0.005) were independently associated with development of AKI in COVID‐19 patients. The AUC of single PCT and constructed risk score was 0. 881 and 0.928, respectively.

Conclusion

PCT level is correlated with AKI in COVID‐19 patients. The efficient risk score consisted of SaO₂, PCT, and BUN is readily accessible for physicians to evaluate the possibility of AKI in COVID‐19 patients.

Galectin-3 in septic acute kidney injury: a translational study

BACKGROUND

Galectin-3 (Gal-3) is a pleiotropic glycan-binding protein shown to be involved in sepsis and acute kidney injury (AKI). However, its role has never been elucidated in sepsis-associated AKI (S-AKI). We aimed to explore Gal-3's role and its potential utility as a therapeutic target in S-AKI.

METHODS

In 57 patients admitted to the intensive care unit (ICU) with sepsis, serum Gal-3 was examined as a predictor of ICU mortality and development of AKI. In a rat model of S-AKI induced by cecal ligation and puncture (CLP), 7-day mortality and serum Gal-3, Interleukin-6 (IL-6), and creatinine were examined at 2, 8, and 24 hours (h) post-CLP. Two experimental groups received the Gal-3 inhibitor modified citrus pectin (P-MCP) at 400 mg/kg/day and 1200 mg/kg/day, while the control group received water only (n = 18 in each group).

RESULTS

Among 57 patients, 27 developed AKI and 8 died in the ICU. Serum Gal-3 was an independent predictor of AKI (OR = 1.2 [95% CI 1.1-1.4], p = 0.01) and ICU mortality (OR = 1.4 [95% CI 1.1-2.2], p = 0.04) before and after controlling for age, AKI, and acute physiology and chronic health evaluation (APACHE II) score. In the CLP rat experiment, serum Gal-3 peaked earlier than IL-6. Serum Gal-3 was significantly lower in both P-MCP groups compared to control at 2 h post-CLP (400 mg: p = 0.003; 1200 mg: p = 0.002), and IL-6 was significantly lower in both P-MCP groups at all time points with a maximum difference at 24 h post-CLP (400 mg: p = 0.015; 1200 mg: p = 0.02). In the Gal-3 inhibitor groups, 7-day mortality was significantly reduced from 61% in the control group to 28% (400 mg P-MCP: p = 0.03) and 22% (1200 mg P-MCP: p = 0.001). Rates of AKI per RIFLE criteria were significantly reduced from 89% in the control group to 44% in both P-MCP groups (400 mg: p = 0.007; 1200 mg: p = 0.007).

CONCLUSIONS

This translational study demonstrates the importance of Gal-3 in the pathogenesis of S-AKI, and its potential utility as a therapeutic target.

PD-1 and PD-L1 Expression on Circulating Lymphocytes as a Marker of Epstein-Barr Virus Reactivation-Associated Proliferative Glomerulonephritis

Alterations to the programmed cell death protein-1 (PD-1) pathway were previously shown to be involved in a poorer prognosis for patients with proliferative glomerulonephritis (PGN). Here, we investigated the association between several infectious agents and the expression of PD-1 and its ligand (PD-L1) on T and B lymphocytes in patients with PGN and nonproliferative glomerulonephritis (NPGN). A cohort of 45 newly-diagnosed patients (23 with PGN and 22 with NPGN) and 20 healthy volunteers was enrolled. The percentage of peripheral blood mononuclear cells expressing PD-1 and PD-L1 antigens was determined by flow cytometry. We found PD-1 and PD-L1 expression on T and B lymphocytes was higher in PGN patients than in NPGN patients and controls. We also found that reactivation of the Epstein-Barr virus (EBV) correlated with the expression of PD-1/PD-L1 antigens in patients with PGN. Further receiver operating characteristic analysis indicated that PD-1 expression could distinguish EBV-positive PGN patients from those with NPGN or healthy controls. The use of PD-1 expression as a non-invasive marker of PGN should be further investigated.