AKI Clinical Research Progress in 2022 II

Diagnosis and biomarkers of AKI

Application of biomarkers to specific patient populations or outcomes

Conroy's team in the United States and Angeli's team in Italy explored the performance of urinary neutrophil gelatinase-associated lipocalin (uNAGL) in predicting the occurrence and prognosis of AKI in children with sickle cell anemia and liver cirrhosis, respectively. The Conroy team recruited 250 children with sickle cell anemia in Uganda and used the NGAL dipstick method to detect the uNAGL levels of the patients. The results found that the uNAGL levels gradually increased with the severity of AKI. Children with high-risk uNGAL levels had a 2.28-fold (95% CI, 1.61-3.23) increased risk of AKI after adjusting for age and sex. uNAGL also showed good predictive power in terms of mortality risk, with an area under the ROC curve >0.85. Angeli's team recruited 162 patients with liver cirrhosis who developed AKI and followed them up to liver transplantation, death, or 90 days. The study found that uNAGL detection can help clinically identify the causes of AKI (prerenal, acute tubular necrosis, hepatorenal syndrome, mixed type), especially in the prediction of acute tubular necrosis AKI showed high discrimination ability (AUC-ROC =0.854). Higher uNAGL levels (>220ng/ml) were independently associated with terlipressin or albumin hyporesponsiveness and poor prognosis in patients.

Click to cistanche deserticola for kidney disease

A study published in the AJKD journal included 500 patients with AKI to evaluate the efficacy of 11 serum biomarkers in predicting the risk of new or worsening chronic kidney disease (CKD) 3 years after the onset of AKI. Results showed that soluble tumor necrosis factor receptor (sTNFR) 1 (sTNFR) 1, sTNFR 2, cystatin C, NAGL, 3-month eGFR, and urinary albumin-to-creatinine ratio were independently associated with renal disease progression. Construction of a multivariate model including sTNFR1, sTNFR2, cystatin C, and eGFR differentiated patients with renal disease progression and non-progression after AKI (AUC-ROC, 0.79 [95% CI, 0.70-0.83]).

A meta-analysis published in Critical Care compared the accuracy of different biomarkers in predicting hospital-acquired acute kidney injury, including 110 articles and 38,725 patients. RESULTS: The uNAGL-creatinine ratio, uNAGL, and serum NAGL had the highest diagnostic accuracy for the risk of hospital-acquired AKI.

Application of machine learning methods in early diagnosis and prognosis assessment of AKI

Machine learning is a powerful weapon for building predictive models. As an important form of artificial intelligence, it has unprecedented advantages in the medical field under the background of the era of big data. A review published in Theranostics in 2022 summarizes previous research on machine learning methods used to predict the occurrence of AKI. Various machine learning methods, including random forests, support vector machines, and decision trees, have shown good predictive performance.

Machine learning methods are also widely used to predict poor prognosis in AKI. The research team of the University of Texas Southwestern Medical Center conducted a multi-center cohort study of 9587 patients with AKI 3 days before ICU admission, using 4 machine learning methods (logistic regression, support vector machine, random forest, xgboost ) for feature selection and building a predictive model. The results showed that the machine learning method was superior to the SOFA score in predicting in-hospital death, and was superior to the AKI severity staging in the KDIGO guidelines in predicting major adverse renal events (MAKE).

Single-cell sequencing technology is expected to achieve clinical translation of AKI prediction

High-throughput single-cell sequencing is one of the hottest technologies in the biomedical field in recent years. With its extremely high resolution, it can accurately analyze cell composition information, combined with high-throughput sequencing methods, and further elucidate the genes of individual cells' Structure and gene expression status. The advantage of single-cell sequencing technology is that it can avoid the homogenization of mixed samples to cover up the heterogeneity of single cells, and can deeply explore the cell state under healthy physiological and disease abnormalities at the single-cell level. At present, more and more basic studies have revealed the molecular mechanism and intervention targets behind the occurrence and outcome of AKI through single-cell sequencing technology. Urine single-cell transcriptomics also has the potential to become a non-invasive method for evaluating AKI. The team of Professor Philipp Enghard from the University of Berlin School of Medicine in Germany performed single-cell sequencing on living cells in 40 urine samples from 32 AKI patients. The analysis showed that there are three types of living cells in the urine of AKI patients: renal parenchymal cells, urethral cells Epithelial cells, and immune cells. Most renal parenchymal cells are tubular epithelial cells (TECs) in different states of injury. Further grouping of these TECs found that normal TECs accounted for only a minority, and other TECs could be divided into TECs in a state of injury and pro-inflammatory ability, TECs in a state of oxidative stress, TECs in a proliferative state, and TECs with progenitor cells according to their transcriptome characteristics. Characteristic TECs. These results indicate that urine live-cell single-cell transcriptomics can reflect kidney damage, and provide a strong basis for the clinical transformation of urine single-cell sequencing technology.

Non-invasive technology and imaging technology play an important role in the field of AKI

The research team at King's College London conducted a single-center prospective longitudinal observational study, including 50 patients with septic shock and 10 healthy volunteers on days 0, 1, 2, and 4 after ICU admission. Contrast-enhanced ultrasonography (CEUS) was used to evaluate renal cortical perfusion every day. At the same time, transthoracic echocardiography was used to measure cardiac output, velocity time integral and vessel diameter were used to calculate renal artery blood flow, and a hand-held video microscope was used to evaluate sublingual microcirculation. To explore the mechanism of AKI in patients with sepsis. We found that renal cortical hypoperfusion is a persistent feature of critically ill septic patients who develop AKI, but is not caused by decreased renal vascular blood flow or cardiac output. Cortical hypoperfusion was not associated with changes in sublingual microcirculation.

The medical-industrial intersection has injected new impetus into solving clinical diagnosis problems and has become a research hotspot in recent years. A review published in ACS Sensors in 2022 illustrates the critical significance of biosensors in the detection and quantification of novel AKI biomarkers to improve clinical diagnostic interventions. Not only are novel diagnostic platforms and their fabrication processes introduced, but strategies to improve the effectiveness of biosensors are also proposed. The application of the combination of medicine and engineering in the field of early diagnosis of AKI is full of potential.

Prevention of AKI

Contrast-related AKI is a common complication of coronary angiography and PCI, characterized by high costs and poor long-term prognosis. A stepwise group randomized clinical trial published in JAMA this year explored the effect of the multifactorial intervention (pre-intervention training, clinical decision support during PCI, quarterly audit) on AKI after coronary angiography or PCI (predicted risk > 5%) for prevention. The results showed that the time-adjusted incidence of AKI decreased from 8.6% before the intervention (control group) to 7.2% during the intervention (OR = 0.72 [95% CI 0.56-0.93]; P = 0.01). This study introduces a novel and comprehensive clinical decision-support intervention that improves outcomes in high-risk patients undergoing cardiac catheterization. This innovative model can be applied in different medical institutions and professional fields. With future iterations of clinical decision support, focusing on clinical outcomes and fully integrating into clinical workflows can facilitate more efficient and reliable care.

Sodium-glucose cotransporter 2 (SGL-T2) inhibitors have been found to have numerous benefits for patients with type 2 diabetes. However, whether SGL T2 inhibitors increase the risk of acute kidney injury remains unclear. A cohort study from this year's AJKD examined the relationship between hospitalization for AKI and prior use of SGL T2 inhibitors versus the use of dipeptidyl peptidase 4 (DPP-4) inhibitors or glucagon-like peptide 1 receptor in older patients with type 2 diabetes. Agonist (GLP-1RA) relevance. The researchers used propensity scores to match 68,130 and 71,477 new patients with SGL-T2 inhibitors to new patients with DPP-4 inhibitors or GLP-1RAs, respectively. The risk of AKI was found to be lower in the SGL T2 inhibitor group than in the DPP-4 inhibitor group (HR=0.71, 95%CI, 0.65-0.76) or the GLP-1RA group (HR=0.81, 95%CI, 0.75-0.87). This study guides the clinical use of hypoglycemic drugs for elderly patients with type 2 diabetes. Compared with DPP-4 inhibitors or GLP-1RA, taking SGL T2 inhibitors can reduce the risk of hospitalization due to AKI.

AKI treatment

Regarding RRT in patients with AKI, there are two authoritative reviews this year. Among them, Gaudry S et al. mainly emphasized the timing of RRT, the indifference of the three modes of RRT (IHD, PIRRT/CRRT), and the dose of CRRT, all of which point to the possible complications and trauma caused by CRRT. Wald R et al described the current best RRT method from the aspects of timing, mode, anticoagulation, electrolyte management, fluid management, stop timing, etc., and looked forward to the areas that are still uncertain.

Patients with chronic kidney disease are more likely to be complicated with AKI, and their renal reserve function is poor, so they may benefit from early RRT. But on the other hand, dialysis dependence is more likely to occur. The timing of AKI in this population is unclear. A subgroup analysis of the START-AKI study attempted to answer whether early RRT was beneficial. The study included 1121 AKI patients with previous renal function results, 38.5% had CKD (defined as eGFR≤59mL/min/1.73m2), and found that the 90-day mortality rate of the previous CKD population was higher (47% vs. 40 %, p<0.001). Patients in the CKD population who started RRT earlier had similar 90-day mortality rates but higher rates of dialysis dependence (14% vs. 8%; aOR, 1.89; 95% CI, 1.05–3.43). It is proposed that AKI occurs in the CKD population, and early initiation of RRT will lead to a higher rate of dialysis dependence.

Prognosis of AKI

In recent years, the incidence of AKI requiring dialysis is increasing. The study by Lee et al. showed that the overall mortality rate of critically ill patients with AKI requiring dialysis has decreased from 2009 to 2018, but the overall level is still around 50%, and surviving patient's Reliance on dialysis has increased over time, from 36.8% in 2009 to 43.9% in 2018.

The STOP-COVID research team published a multi-center cohort study to explore the clinical factors of renal recovery in the COVID-19 combined with the AKI-KRT population, reviewing 4221 patients (876 people in the intensive care unit of 68 American hospitals) who received treatment for new crown infection AKI-KRT occurred after ICU and entered the "AKI-KRT subgroup"). Studies have shown that more severe AKI is associated with higher in-hospital mortality and worse renal function at discharge. Among patients who developed AKI-KRT, nearly two-thirds died. Of the patients who survived hospital discharge, about two-thirds recovered kidney function and were discharged without the need for dialysis. Poor baseline renal function and decreased urine output were associated with nonrecovery of renal function. Identifying these predictors is important to assess the prognosis of these critically ill patients and has important implications for the clinical care of patients with COVID-19 infection.

How the progression to AKD after AKI is not fully understood. A 2010-2018 community-based cohort study in the UK included 56,906 patients. The study showed that 33% of patients developed AKD, and 27% of AKD patients did not recover on the 90th day after the diagnosis of AKI. Progression to AKD was associated with an increased risk of 1-year mortality (HR=1.20) and new-onset CKD (HR=2.21) compared with recovery from early AKI. This study demonstrates the importance of early diagnosis and management of AKI and avoiding progression to AKD.

for more information: Ali.ma@wecistanche.com