Knowledge-aware multi-center clinical dataset adaptation: Problem, method, and application

It has great potential to integrate medical knowledge and electronic health record data for diagnosis prediction. However, present studies only utilized information from knowledge graphs, omitting potentially significant global graph structural features. In this study, we proposed a knowledge and data integrating modeling approach to reconstruct patient electronic health record data with graph structure and use medical knowledge as internal information of patient data to build a risk prediction model for acute kidney injury in patients with heart failure based on graph neural networks. Experimental results based on the MIMIC III data showed that the method proposed was superior to other baseline models in predicting the risk of acute kidney injury in heart failure patients, with an accuracy of 0.725 and an F1 score of 0.755. This study provides a novel approach to the disease risk prediction models that integrates medical knowledge and data.

(1) Background: This single-center retrospective study aimed to evaluate whether sodium-glucose cotransporter-2 inhibitors (SGLT2-i) therapy may have a nephroprotective effect to prevent contrast-induced acute kidney injury (CI-AKI) in patients with heart failure (HF) undergoing iodinated contrast medium (ICM) invasive procedures. (2) Methods: The population was stratified into SGLT2-i users and SGLT2-i non-users according to the chronic treatment with gliflozins. The primary endpoint was CI-AKI incidence during hospitalization. Secondary endpoints were all-cause mortality and the need for continuous renal replacement therapy (CRRT). (3) Results: In total, 86 patients on SGLT2-i and 179 patients not on SGLT2-i were enrolled. The incidence of CI-AKI in the gliflozin group was lower than in the non-user group (9.3 vs. 27.3%, p < 0.001), and these results were confirmed after propensity matching analysis. Multivariable logistic regression showed that only SGLT2-i treatment was an independent preventive factor for CI-AKI (OR: 0.41, 95% CI: 0.16-0.90, p = 0.045). The need for CRRT was reported only in five patients in the non-SGLT2-i-user group compared to zero patients in the gliflozin group (p = 0.05). (4) Conclusions: SGLT2-i therapy was associated with a lower risk of CI-AKI in patients with HF undergoing ICM invasive procedures.

Although hypoalbuminemia at admission is a risk for acute kidney injury (AKI) and mortality in patients with acute decompensated heart failure (ADHF), the clinical significance of decreased serum albumin levels (DAL) during ADHF therapy has not been elucidated. This study aimed to evaluate whether DAL was associated with AKI, and whether intravenous atrial natriuretic peptide (ANP) administration, which provides an effective treatment for ADHF but promotes albumin extravasation, was associated with DAL and AKI. A total of 231 consecutive patients with ADHF were enrolled. AKI was defined as ≥0.3mg/dl absolute or 1.5-fold increase in serum creatinine levels within 48h. AKI occurred in 73 (32%) of the 231 patients during ADHF therapy. The median value of decreases in serum albumin levels was 0.3g/dl at 7days after admission. When DAL was defined as ≥0.3g/dl decrease in serum albumin levels, DAL occurred in 113 patients, and was independently associated with AKI. Of the 231 patients, 73 (32%) were treated with intravenous ANP. DAL occurred more frequently in patients receiving ANP than in those not receiving ANP (77 vs. 36%, p < 0.001), and ANP was independently associated with DAL. The incidence of AKI was higher in patients receiving ANP than in those not receiving ANP (48 vs. 24%, p < 0.001). ANP was independently associated with AKI. In conclusion, DAL is associated with AKI. Intravenous ANP administration may be one of the promoting factors of DAL, which leads to AKI, indicating a possible novel mechanism of AKI.

The aim of our study was to assess the prevalence of contrast-induced acute kidney injury (CI-AKI) in patients with stable coronary artery disease (CAD) receiving optimal medical treatment with indications to coronary angiography and intraarterial administration of contrast agents. 1023 patients with stable CAD were included in the open prospective observational cohort study. The CI-AKI was defined as a rise in serum creatinine 25% from baseline. The mean age of the study group was 61.710.1 years; 72.4% were males and 84.4% had arterial hypertension. A multiple logistic regression model of prediction of CI-AKI was created. CI-AKI developed in 132 (12.9%) of the patients. The multiple logistic regression model included gender, BMI, weight, age, heart failure, diabetes mellitus, arterial hypertension, anemia, hyperuricemia, proteinuria and baseline serum creatinine. Area under the curve for the model was 0.749 (95% confidence interval 0.7030,795;p0.0001). When trying to build a prognostic model, including baseline GFR and contrast volume, the model lost significance and the AUC diminished. The CI-AKI remains quite a common kidney injury developing in patients with stable CAD undergoing percutaneous interventions. Several risk factors need to be assessed very carefully before any intervention requiring intraarterial contrast media administration especially in patients with comorbidities.

To investigate serum cystatin C level as an early biomarker for predicting acute kidney injury (AKI) in patients with acute-on-chronic liver failure (ACLF). Fifty-six consecutive patients with hepatitis B virus-related ACLF who had normal serum creatinine (Cr) level (< 1.2 mg/dL in men, or < 1.1 mg/dL in women) were enrolled in the Liver Failure Treatment and Research Center of Beijing 302 Hospital between August 2011 and October 2012. Thirty patients with chronic hepatitis B (CHB) and 30 healthy controls in the same study period were also included. Measurement of serum cystatin C (CysC) was performed by a particle-enhanced immunonephelometry assay using the BN Prospec nephelometer system. The ACLF patients were followed during their hospitalization period. In the ACLF group, serum level of CysC was 1.1 ± 0.4 mg/L, which was significantly higher (P < 0.01) than those in the healthy controls (0.6 ± 0.3 mg/L) and CHB patients (0.7 ± 0.2 mg/L). During the hospitalization period, eight ACLF patients developed AKI. Logistic regression analysis indicated that CysC level was an independent risk factor for AKI development (odds ratio = 1.8; 95%CI: 1.4-2.3, P = 0.021). The cutoff value of serum CysC for prediction of AKI in ACLF patients was 1.21 mg/L. The baseline CysC-based estimated glomerular filtration rate (eGFR(CysC)) was significantly lower than the creatinine-based eGFR (eGFR(CG) and eGFR(MDRD)) in ACLF patients with AKI, suggesting that baseline eGFR(CysC) represented early renal function in ACLF patients while the Cr levels were still within the normal ranges. Serum CysC provides early prediction of renal dysfunction in ACLF patients with a normal serum Cr level.

To examine the relationship between Neutrophil-Lymphocyte Ratio (NLR) and Acute Kidney Injury (AKI) in patients undergoing noncardiac surgery, and subgroup analysis was performed for different types of non-cardiac surgery. The present retrospective cohort study included 10,159 adult patients who underwent major noncardiac surgery at Nanfang Hospital, Southern Medical University, between 2008 and 2018. Postoperative AKI was defined as an increase in serum creatinine level of at least 0.3mg/dl within 48h, or 1.5 times higher than baseline within 7days postoperatively according to the Kidney Disease Improving Global Outcome. The correlation between preoperative NLR and postoperative AKI was determined by stepwise multivariate logistic regression analysis, and the predictive value of NLR was evaluated by the receiver operating characteristics curve (ROC) analysis. Four hundred and eighty-five (4.77%) patients developed AKI postoperatively. Preoperative NLR was independently associated with postoperative AKI in all patients undergoing non-cardiac surgery (Odds ratio [OR], 1.03; 95% confidence interval [CI], 1.00-1.06). The optimal cut-off value of NLR was 2.12 according ROC analysis. The OR and 95% CI of AKI for NLR > 2.12 was 1.48 (1.21-1.81) compared with NLR ≤ 2.12. In addition, the positive association was mainly shown in patients undergone digestive system surgery with a cut-off value of 2.12 but not in neurological and musculoskeletal system surgeries. The present study confirmed the association of preoperative NLR with postoperative AKI in digestive system surgical patients. A NLR value of 2.12 may be a useful cut-off to evaluate the risk of AKI.

Background and purposeWe conducted a systematic review and meta-analysis to assess the incidence of acute kidney injury (AKI) in patients undergoing CT angiography (CTA) and CT perfusion (CTP) for acute...

Risk factors associated with acute kidney injury in patients with traumatic brain injury: A systematic review and meta-analysis.

This multi-institute retrospective cohort study investigated the association between preoperative anemia and postoperative acute kidney injury (AKI) in patients undergoing metabolic and bariatric surgery (MBS). Using the TriNetX research network, we identified adult patients who underwent MBS between January 2010 and December 2024. Patients were categorized as anemic or non-anemic based on preoperative hemoglobin levels (hemoglobin < 12 g/dL for females and < 13 g/dL for males). Propensity score matching was performed to balance the baseline characteristics. The primary outcome was AKI within 30 days of MBS. The secondary outcomes included infections, intensive care unit admission, surgical complications, and deep vein thrombosis. Analyses were conducted separately for female and male patients, with additional sensitivity analyses for the female subgroups. After propensity score matching, preoperative anemia was significantly associated with an increased risk of postoperative AKI in females (1.51% vs. 0.71%, odds ratio (OR) 2.14, 95% confidence interval (CI) 1.53-2.98, p < 0.0001) (n = 7144 pairs). This association persisted in all sensitivity analyses. Even mild anemia (hemoglobin 10-12 g/dL) was associated with a higher AKI risk (OR 1.62, p = 0.004) in female patients. In males (n = 971 pairs), anemia similarly increased the AKI risk (4.94% vs. 2.68%, OR 1.89, 95% CI 1.16-3.07, p = 0.009). No significant differences were observed in secondary outcomes for either sex. Preoperative anemia is independently associated with an increased risk of postoperative AKI in both female and male patients undergoing MBS, suggesting that preoperative anemia screening and management could be important for reducing postoperative kidney complications in this population.

The development of acute kidney injury (AKI) in patients during hospitalization worsens the prognosis. The study aimed to estimate the prognostic value of AKI in patients with ST-segment elevation myocardial infarction (MI) in prospective follow-up study. A prospective follow-up of 12 months included 268 patients with ST-segment elevation MI who underwent percutaneous coronary intervention. The incidence of AKI was assessed and its effect on the outcome of ST-segment elevation MI during the observed period was traced. AKI was diagnosed according to Kidney Disease: Improving Global Outcomes recommendations. AKI was recorded in 20.5%. The incidence of AKI increased with decrease in the level of glomerular filtration rate on admission (P = 0.047) and an increase in the stage of acute heart failure according to Killip on admission (P = 0.006). The development of AKI was associated with an increase in-hospital mortality, increasing with a worsening stage of AKI (P<0.001), the same pattern was characteristic for death from cardiovascular causes during follow-up. AKI increases the risk of death from cardiovascular causes by four times [odds ratio 95% 4.400 (1.928-10.040), P <0.001]. The multivariate analysis revealed risk factors for the development of AKI: age >75 years (P = 0.006), female (P = 0.008), Killip >I (P = 0.045). AKI is common in patients with ST-segment elevation MI and is associated with a poor prognosis. The development of AKI is associated with an increase mortality, increasing with a worsening stage of AKI.

Objective: To analyze the influencing factors of acute kidney injury (AKI) in patients after cardiac surgery using levosimendan or dobutamine, and explore the effect of positive inotropic drugs on AKI. Methods: The clinical data of 417 patients undergoing cardiac surgery from January to June 2018 in Beijing Anzhen Hospital and treated with levosimendan or dobutamine during perioperative period were retrospectively reviewed and collected. Patients were divided into AKI group and non-AKI group according to whether AKI occurred. Univariate logistic regression analysis was used to analyze the factors related to the occurrence of AKI. The statistically significant factors (P<0.05) were further included in the multivariate logistic regression analysis. Results: Totally, 417 patients were enrolled in the study, with a mean age of (58.2±10.4) years old and a male rate of 65.0% (n=271), and the AKI incidence rate was 25.2% (105/417). Univariate logistic regression analysis showed that male, chronic kidney disease, high serum creatinine level in preoperative period, aortic obstruction time ≥ 120 minutes and extracorporeal circulation time ≥ 120 minutes were risk factors for AKI (all P<0.05). Vasodilator and levosimendan treatment during perioperative period were protective factors (P<0.05). Multivariate logistic regression analysis showed that chronic kidney disease (OR=17.291, 95%CI: 4.335-68.960, P<0.001) and high serum creatinine level (OR=1.097, 95%CI: 1.074-1.121, P<0.001) in preoperative period were independent risk factors for AKI. Perioperative application of levosimendan (OR=0.533, 95%CI: 0.288-0.984, P=0.044) was an independent protective factor. Conclusions: Risk factors for AKI after cardiac surgery include chronic kidney disease and high serum creatinine level in preoperative period. The use of levosimendan during preoperative period has the potential effect to protect against AKI.

There is limited information on the incidence of acute kidney injury (AKI) in patients with traumatic brain injury (TBI) although AKI may contribute to morbidity and mortality. We investigated the incidence of AKI in patients with moderate and severe TBI and the association of AKI with risk factors and outcomes in these patients.We studied all TBI patients over 16 years of age admitted to the two designated trauma hospitals in the state of Victoria, Australia from 1 January to 31 December 2008. Patients were included if they had head trauma and presented with a Glasgow coma scale (GCS) <13. Prospectively collected data from the hospital trauma registries, ICUs, and pathology databases were analyzed retrospectively. Risk injury failure loss end (RIFLE) criteria were used to categorize renal function.The incidence of AKI was 9.2% (19/207). Patients who developed AKI were older, had higher severity of illness scores, and a lower GCS. Overall 42.1% of these patients died in hospital compared with 18.1% in patients without AKI. In univariable linear regression analysis, age, severity of illness, and admitting hospital were associated with AKI. After multivariable logistic regression, the occurrence of AKI was associated with age (p < 0.001) and higher APACHE III scores (p = 0.016).AKI is relatively common even in patients with TBI. Its association with age and APACHE III scores helps identify patients at higher risk of AKI.

Previous research revealed a positive correlation between endothelial cell injury (indicated by albuminuria) and obstructive sleep apnea (OSA). However, little else has been revealed about acute kidney injury (AKI) in patients with OSA. This prospective study recruited consecutive patients undergoing overnight polysomnography for evaluation of sleep apnea. Patients in whom any major disease or recent infection had been previously diagnosed were excluded. Ultimately, data from 75 patients with apnea-hypopnea indices of 5 or more were analyzed. Baseline values for the urinary albumin-creatinine ratio (UACR), serum levels for three markers of AKI (cystatin C, neutrophil gelatinase-associated lipocalin [NGAL], interleukin-18 [IL-18]), and polysomnography data were recorded and analyzed. Patients then were followed for 6 months of continuous positive airway pressure (CPAP) treatment. At baseline, UACRs were greater in patients with more severe OSA (P = .005, r = .329). All three serum markers of AKI (cystatin C, NGAL, and IL-18) studied were positively correlated with OSA severity, and two (cystatin C and IL-18) were positively correlated with the frequency of oxygen desaturation during sleep. However, none of the AKI markers had positive correlations with UACR. After 6 months of CPAP treatment, UACR and IL-18 were decreased significantly in patients with good adherence. Albuminuria and levels of three serum markers of AKI (cystatin C, NGAL, IL-18) were positively correlated with OSA severity, and good adherence with CPAP treatment decreased albuminuria and interleukin-18 levels. These results may provide additional tools for assessing early renal injury in patients with OSA.

The potential nephrotoxicity of intravenous iodinated contrast media is a major concern for acute ischaemic stroke imaging evaluation. This study aimed to assess the incidence of acute kidney injury after intravenous iodinated contrast media exposure in acute ischaemic stroke patients. We conducted a retrospective cohort analysis between January 2012 and July 2018 to select adult patients admitted to the emergency department with acute ischaemic stroke. The exposed patients received a uniform intravenous dose of low osmolar non-ionic iodinated contrast media, as part of the imaging protocol for acute ischaemic stroke. The unexposed patients underwent a non-enhanced cranial computed tomography scan. Acute kidney injury was defined according to the Kidney Disease Improving Global Outcomes criteria, limited to the first 72 hours. A total of 161 and 105 patients were included in the exposed and unexposed groups, respectively. The median age was 72.8 years (interquartile range 20), 53% were men and 97% were white. Demographic and baseline characteristics were similar between the groups. The incidence of acute kidney injury between exposed (n = 10, 6.2%) and unexposed (n = 1, 1%) groups (P = 0.073) was similar and contrast exposure was not a significant predictor of acute kidney injury. Intravenous iodinated contrast media exposure during acute ischaemic stroke imaging protocols is not an independent predictor of acute kidney injury in patients with normal or near-normal renal function. Studies with larger sample sizes would help to clarify if patients with both cardiovascular risk factors and impaired renal function could benefit from prophylactic measures.

The Addition of a Nonsteroidal Anti-inflammatory Drug in Local Infiltration Analgesia During Total Knee Arthroplasty Increases the Risk of Acute Kidney Injury in Patients Who Have Renal Impairment: A Propensity-Matched Retrospective Cohort Study