Development and Validation of a Model for Predicting the Risk of Acute Kidney Injury Associated With Contrast Volume Levels During Percutaneous Coronary Intervention

Possibilities of Azilsartan Medoxomil for Preparation for Planned Percutaneous Coronary Intervention in Patients With Type 2 Diabetes Mellitus

AIM

To evaluate the efficacy and safety of azilsartan medoxomil for preoperative preparation and improving the long-term prognosis of elective percutaneous coronary intervention (PCI) in patients with ischemic heart disease (IHD), arterial hypertension (AH), and type 2 diabetes mellitus (DM).

MATERIAL AND METHODS

The study sample included patients with type 2 DM referred for elective PCI who had poor blood pressure (BP) control according to 24-hour BP monitoring (24-BPM) (mean daily systolic BP ≥130 mmHg, mean daily diastolic BP ≥80 mmHg). The data were collected from 2018 through 2020. A total of 75 patients was included and distributed by simple randomization into two groups: group 1 (main, n=37) received azilsartan medoxomil as an antihypertensive drug at a dose of 40 mg/day (previously prescribed angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers (ARB) were discontinued); group 2 (control, n=38) continued on their previous antihypertensive therapy. The follow-up period was 6 months. During each of 5 consecutive follow-up visits, the patient was examined, 24-BPM was recorded, and urinary markers of renal dysfunction (glomerular filtration rate, GFR; neutrophil gelatinase-associated lipocalin, NGAL; urine albumin-creatinine ratio, UACR; kidney injury molecule, KIM-1; and interleukin-18, IL-18) were measured.

RESULTS

During the azilsartan treatment, GFR decreased by 7.4%, while in the control group, it decreased by 18.9% (p<0.001). For 6 months of follow-up, no changes in the NGAL concentration were found in the main group, while the NGAL concentration in the control group increased by 12.9%. With azilsartan, there was a decrease in the urinary concentration of IL-18 (16.9%), while in patients of the control group, IL-18 increased (7.14%). Proteinuria progressed in both groups, which was expectable given the presence of DM; however, in patients receiving azilsartan, the UACR value increased by 37.5%, while in patients of the control group, it increased by 96.15%. These differences were statistically significant. No statistically significant differences were found in the concentrations of cystatin C and KIM-1.

CONCLUSION

This study demonstrated two important facts: the possibility for diagnosing contrast-induced acute kidney injury (CI-AKI) using new, more sensitive markers of kidney damage, which is important for assessing the effectiveness of prevention, and the possibility of using ARBs, in particular azilsartan, for the prevention of CI-AKI in patients with IHD in combination with AH and DM.

[Artificial intelligence and acute kidney injury]

Digitalization is increasingly finding its way into intensive care units and with it artificial intelligence (AI) for critically ill patients. One promising area for the use of AI is in the field of acute kidney injury (AKI). The use of AI is primarily focused on the prediction of AKI, but further approaches are also being used to classify existing AKI into different phenotypes. Different AI models are used for prediction. The area under the receiver operating characteristic curve values (AUROC) achieved with these models vary and are influenced by several factors, such as the prediction time and the definition of AKI. Most models have an AUROC between 0.650 and 0.900, with lower values for predictions further into the future and when applying Acute Kidney Injury Network (AKIN) instead of KDIGO criteria. Classification into phenotypes already makes it possible to categorize patients into groups with different risks of mortality or requirement of renal replacement therapy (RRT), but the etiologies or therapeutic consequences derived from this are still lacking. However, all the models suffer from AI-specific shortcomings. The use of large databases does not make it possible to promptly include recent changes in therapy and the implementation of new biomarkers in a relevant proportion. For this reason, serum creatinine and urinary output, with their known limitations, dominate current AI models for prediction impairing the performance of the current models. On the other hand, the increasingly complex models no longer allow physicians to understand the basis on which the warning of a threatening AKI is calculated and subsequent initiation of therapy should take place. The successful use of AIs in routine clinical practice will be highly determined by the trust of the physicians in the systems and overcoming the aforementioned weaknesses. However, the clinician will remain irreplaceable as the decisive authority for critically ill patients by combining measurable and nonmeasurable parameters.

Predictive value of systemic immune-inflammation index combined with N-terminal pro-brain natriuretic peptide for contrast-induced acute kidney injury in patients with STEMI after primary PCI

OBJECTIVE

To investigate the relationship between the incidence of contrast-induced acute kidney injury (CI-AKI) after emergency percutaneous coronary intervention (PCI) and preoperative systemic immune-inflammation index (SII) and N-terminal pro-brain natriuretic peptide (NT-proBNP) levels in patients with acute ST-segment elevation myocardial infarction (STEMI), and to further analyze the predictive value of the combination of SII and NT-proBNP for CI-AKI.

METHODS

The clinical data of 1543 patients with STEMI who underwent emergency PCI in our hospital from February 2019 to December 2022 were retrospectively analyzed. All patients were divided into training cohort (n = 1085) and validation cohort (n = 287) according to chronological order. The training cohort was divided into CI-AKI (n = 95) and non-CI-AKI (n = 990) groups according to the 2018 European Society of Urogenital Radiology definition of CI-AKI. Multivariate Logistic regression analysis was used to determine the independent risk factors for CI-AKI. Restricted cubic spline (RCS) was used to explore the relationship between SII, NT-proBNP, and the risk of CI-AKI. The receiver operating characteristic (ROC) curve was used to evaluate the predictive value of SII, NT-proBNP, and their combination in CI-AKI.

RESULTS

The incidence of CI-AKI was 8.8% (95/1085). Multivariate logistic regression analysis showed that SII, NT-proBNP, age, baseline creatinine, fasting blood glucose, and diuretics were independent risk factors for CI-AKI. RCS analysis showed that SII > 1084.97 × 109/L and NT-proBNP > 296.12 pg/mL were positively associated with the incidence of CI-AKI. ROC curve analysis showed that the area under the curve of SII and NT-proBNP combined detection in predicting CI-AKI was 0.726 (95% CI 0.698-0.752, P < 0.001), the sensitivity was 60.0%, and the specificity was 77.7%, which were superior to the detection of SII or NT-proBNP alone.

CONCLUSION

Preprocedural high SII and NT-proBNP are independent risk factors for CI-AKI after emergency PCI in patients with STEMI. The combined detection of SII and NT-proBNP can more accurately predict CI-AKI risk than the single detection.

Development of real-time individualized risk prediction models for contrast associated acute kidney injury and 30-day dialysis after contrast enhanced computed tomography

PURPOSE

This study aimed to develop preprocedural real-time artificial intelligence (AI)-based systems for predicting individualized risks of contrast-associated acute kidney injury (CA-AKI) and dialysis requirement within 30 days following contrast-enhanced computed tomography (CECT).

METHOD

This single-center, retrospective study analyzed adult patients from emergency or in-patient departments who underwent CECT; 18,895 patients were included after excluding those who were already on dialysis, had stage V chronic kidney disease, or had missing data regarding serum creatinine levels within 7 days before and after CECT. Clinical parameters, laboratory data, medication exposure, and comorbid diseases were selected as predictive features. The patients were randomly divided into model training and testing groups at a 7:3 ratio. Logistic regression (LR) and random forest (RF) were employed to create prediction models, which were evaluated using receiver operating characteristic curves.

RESULTS

The incidence rates of CA-AKI and dialysis within 30 days post-CECT were 6.69% and 0.98%, respectively. For CA-AKI prediction, LR and RF exhibited similar performance, with areas under curve (AUCs) of 0.769 and 0.757, respectively. For 30-day dialysis prediction, LR (AUC, 0.863) and RF (AUC, 0.872) also exhibited similar performance. Relative to eGFR-alone, the LR and RF models produced significantly higher AUCs for CA-AKI prediction (LR vs. eGFR alone, 0.769 vs. 0.626, p < 0.001) and 30-day dialysis prediction (RF vs. eGFR alone, 0.872 vs. 0.738, p < 0.001).

CONCLUSIONS

The proposed AI prediction models significantly outperformed eGFR-alone for predicting the CA-AKI and 30-day dialysis risks of emergency department and hospitalized patients who underwent CECT.

Contrast media-induced nephropathy in Tunisia: prospective case-control study with cardio-nephrological monitoring

Introduction

vascular opacification using iodinated contrast media (ICM) is often the primary diagnostic and therapeutic approach. However, the risk of post-injection nephrotoxicity of ICM is significantly higher in patients with underlying nephropathy. This study aimed to determine the incidence of Contrast Media Induced Nephropathy (CMIN) and identify predictive factors for its occurrence in patients from a cardiology department.

Methods

our prospective study involved 158 patients who underwent coronary angiography or angioplasty at the cardiology department between December 2017 and May 2018. Two types of ICM were used in our study: Iopromide and Iohexol. All patients received either physiological serum (9‰) or bicarbonate serum (14‰) intravenously for hydration. We defined impaired renal function as an increase in creatinine ranging from 10 to 26 µmol/L, while CMIN was defined as an increase in serum creatinine exceeding 26.5 µmol/L. We investigated the factors associated with CMIN using logistic regression analysis.

Results

the mean age of our patients was 60 ± 11 years (range: 29-82), with a predominance of men 63.9% (n=101). The most common cardiovascular risk factors were tobacco (36.1%, n = 57), diabetes (48.1%, n =76), hypertension (55%, n = 87). Pre-procedural creatinine averaged 81.1 ± 47.3 µmol / L with extremes ranging from 39 to 600 µmol / L. The median Mehran risk score was 3.2 (range: 0- 15). The interventional cardiology act consisted of coronary angiography in 86.2% (n=136) of cases, coronary angioplasty in 2.5% (n=4) of cases. We used iohexol and iopromide in 57.6% (n=91) and 42.4% (n=67) of cases, respectively. The overall incidence of CMIN was 9.5% (n=9). The multivariable regression analysis identified 4 risk factors independently linked to the occurrence of CMIN which were Pre-existing renal failure (OR: 6.05, 95%CI [1.23-29.62], p = 0.026), anemia (OR: 0.043, CI [1.03-8.96], p = 0.043), the toxic dose of PC (OR: 4.7, CI [1.28-17.7], p=0.02), and at a Mehran score = 11 (OR: 3.7, CI [0.88-15.6], p=0.036).

Conclusion

the most effective approach for CMIN is prevention, which focuses on addressing modifiable risk factors to minimize the risk especially in patients with pre-existing renal failure.

Clinical Outcomes for Acute Kidney Injury in Acute Myocardial Infarction Patients after Intra-Aortic Balloon Pump Implantation: A Single-Center Observational Study

Background

Acute kidney injury (AKI) is common after cardiac interventional procedures. The prevalence and clinical outcome of AKI in patients with acute myocardial infarction (AMI) after undergoing intra-aortic balloon pump (IABP) implantation remains unknown. The aim of this study was to investigate the incidence, risk factors, and prognosis of AKI in specific patient populations.

Methods

We retrospectively reviewed 319 patients with AMI between January 2017 and December 2021 and who had successfully received IABP implantation. The diagnostic and staging criteria used for AKI were based on guidelines from "Kidney Disease Improving Global Outcomes". The composite endpoint included all-cause mortality, recurrent myocardial infarction, rehospitalization for heart failure, and target vessel revascularization.

Results

A total of 139 patients (43.6%) developed AKI after receiving IABP implantation. These patients showed a higher incidence of major adverse cardiovascular events (hazard ratio [HR]: 1.55, 95% confidence interval [CI]: 1.06-2.26, p = 0.022) and an increased risk of all-cause mortality (HR: 1.62, 95% CI: 1.07-2.44, p = 0.019). Multivariable regression models found that antibiotic use (odds ratio [OR]: 2.07, 95% CI: 1.14-3.74, p = 0.016), duration of IABP use (OR: 1.24, 95% CI: 1.11-1.39, p < 0.001) and initial serum creatinine (SCr) (OR: 1.01, 95% CI: 1.0-1.01, p = 0.01) were independent risk factors for AKI, whereas emergency percutaneous coronary intervention was a protective factor (OR: 0.35, 95% CI: 0.18-0.69, p = 0.003).

Conclusions

AMI patients who received IABP implantation are at high risk of AKI. Close monitoring of these patients is critical, including the assessment of renal function before and after IABP implantation. Additional preventive measures are needed to reduce the risk of AKI in these patients.

Machine learning for acute kidney injury: Changing the traditional disease prediction mode

Acute kidney injury (AKI) is a serious clinical comorbidity with clear short-term and long-term prognostic implications for inpatients. The diversity of risk factors for AKI has been recognized in previous studies, and a series of predictive models have been developed using traditional statistical methods in conjunction with its preventability, but they have failed to meet the expectations in limited clinical applications, the rapid spread of electronic health records and artificial intelligence machine learning technology has brought new hope for the construction of AKI prediction models. In this article, we systematically review the definition and classification of machine learning methods, modeling ideas and evaluation methods, and the characteristics and current status of modeling studies. According to the modeling objectives, we subdivided them into critical care medical setting models, all medical environment models, special surgery models, special disease models, and special nephrotoxin exposure models. As the first review article to comprehensively summarize and analyze machine learning prediction models for AKI, we aim to objectively describe the advantages and disadvantages of machine learning approaches to modeling, and help other researchers more quickly and intuitively understand the current status of modeling research, inspire ideas and learn from experience, so as to guide and stimulate more research and more in-depth exploration in the future, which will ultimately provide greater help to improve the overall status of AKI diagnosis and treatment.

Management of Acute Coronary Syndrome in the Older Adult Population: A Scientific Statement From the American Heart Association

Diagnostic and therapeutic advances during the past decades have substantially improved health outcomes for patients with acute coronary syndrome. Both age-related physiological changes and accumulated cardiovascular risk factors increase the susceptibility to acute coronary syndrome over a lifetime. Compared with younger patients, outcomes for acute coronary syndrome in the large and growing demographic of older adults are relatively worse. Increased atherosclerotic plaque burden and complexity of anatomic disease, compounded by age-related cardiovascular and noncardiovascular comorbid conditions, contribute to the worse prognosis observed in older individuals. Geriatric syndromes, including frailty, multimorbidity, impaired cognitive and physical function, polypharmacy, and other complexities of care, can undermine the therapeutic efficacy of guidelines-based treatments and the resiliency of older adults to survive and recover, as well. In this American Heart Association scientific statement, we (1) review age-related physiological changes that predispose to acute coronary syndrome and management complexity; (2) describe the influence of commonly encountered geriatric syndromes on cardiovascular disease outcomes; and (3) recommend age-appropriate and guideline-concordant revascularization and acute coronary syndrome management strategies, including transitions of care, the use of cardiac rehabilitation, palliative care services, and holistic approaches. The primacy of individualized risk assessment and patient-centered care decision-making is highlighted throughout.

Effect of Clinical Decision Support With Audit and Feedback on Prevention of Acute Kidney Injury in Patients Undergoing Coronary Angiography: A Randomized Clinical Trial

IMPORTANCE

Contrast-associated acute kidney injury (AKI) is a common complication of coronary angiography and percutaneous coronary intervention (PCI) that has been associated with high costs and adverse long-term outcomes.

OBJECTIVE

To determine whether a multifaceted intervention is effective for the prevention of AKI after coronary angiography or PCI.

DESIGN, SETTING, AND PARTICIPANTS

A stepped-wedge, cluster randomized clinical trial was conducted in Alberta, Canada, that included all invasive cardiologists at 3 cardiac catheterization laboratories who were randomized to various start dates for the intervention between January 2018 and September 2019. Eligible patients were aged 18 years or older who underwent nonemergency coronary angiography, PCI, or both; who were not undergoing dialysis; and who had a predicted AKI risk of greater than 5%. Thirty-four physicians performed 7820 procedures among 7106 patients who met the inclusion criteria. Participant follow-up ended in November 2020.

INTERVENTIONS

During the intervention period, cardiologists received educational outreach, computerized clinical decision support on contrast volume and hemodynamic-guided intravenous fluid targets, and audit and feedback. During the control (preintervention) period, cardiologists provided usual care and did not receive the intervention.

MAIN OUTCOMES AND MEASURES

The primary outcome was AKI. There were 12 secondary outcomes, including contrast volume, intravenous fluid administration, and major adverse cardiovascular and kidney events. The analyses were conducted using time-adjusted models.

RESULTS

Of the 34 participating cardiologists who were divided into 8 clusters by practice group and center, the intervention group included 31 who performed 4327 procedures among 4032 patients (mean age, 70.3 [SD, 10.7] years; 1384 were women [32.0%]) and the control group included 34 who performed 3493 procedures among 3251 patients (mean age, 70.2 [SD, 10.8] years; 1151 were women [33.0%]). The incidence of AKI was 7.2% (310 events after 4327 procedures) during the intervention period and 8.6% (299 events after 3493 procedures) during the control period (between-group difference, -2.3% [95% CI, -0.6% to -4.1%]; odds ratio [OR], 0.72 [95% CI, 0.56 to 0.93]; P = .01). Of 12 prespecified secondary outcomes, 8 showed no significant difference. The proportion of procedures in which excessive contrast volumes were used was reduced to 38.1% during the intervention period from 51.7% during the control period (between-group difference, -12.0% [95% CI, -14.4% to -9.4%]; OR, 0.77 [95% CI, 0.65 to 0.90]; P = .002). The proportion of procedures in eligible patients in whom insufficient intravenous fluid was given was reduced to 60.8% during the intervention period from 75.1% during the control period (between-group difference, -15.8% [95% CI, -19.7% to -12.0%]; OR, 0.68 [95% CI, 0.53 to 0.87]; P = .002). There were no significant between-group differences in major adverse cardiovascular events or major adverse kidney events.

CONCLUSIONS AND RELEVANCE

Among cardiologists randomized to an intervention including clinical decision support with audit and feedback, patients undergoing coronary procedures during the intervention period were less likely to develop AKI compared with those treated during the control period, with a time-adjusted absolute risk reduction of 2.3%. Whether this intervention would show efficacy outside this study setting requires further investigation.

TRIAL REGISTRATION

ClinicalTrials.gov Identifier: NCT03453996.

Development and Validation of a Personalized Model With Transfer Learning for Acute Kidney Injury Risk Estimation Using Electronic Health Records

IMPORTANCE

Acute kidney injury (AKI) is a heterogeneous syndrome prevalent among hospitalized patients. Personalized risk estimation and risk factor identification may allow effective intervention and improved outcomes.

OBJECTIVE

To develop and validate personalized AKI risk estimation models using electronic health records (EHRs), examine whether personalized models were beneficial in comparison with global and subgroup models, and assess the heterogeneity of risk factors and their outcomes in different subpopulations.

DESIGN, SETTING, AND PARTICIPANTS

This diagnostic study analyzed EHR data from 1 tertiary care hospital and used machine learning and logistic regression to develop and validate global, subgroup, and personalized risk estimation models. Transfer learning was implemented to enhance the personalized model. Predictor outcomes across subpopulations were analyzed, and metaregression was used to explore predictor interactions. Adults who were hospitalized for 2 or more days from November 1, 2007, to December 31, 2016, were included in the analysis. Patients with moderate or severe kidney dysfunction at admission were excluded. Data were analyzed between August 28, 2019, and May 8, 2022.

EXPOSURES

Clinical and laboratory variables in the EHR.

MAIN OUTCOMES AND MEASURES

The main outcome was AKI of any severity, and AKI was defined using the Kidney Disease: Improving Global Outcomes serum creatinine criteria. Performance of the models was measured with area under the receiver operating characteristic curve (AUROC), area under the precision-recall curve, and calibration.

RESULTS

The study cohort comprised 76 957 inpatient encounters. Patients had a mean (SD) age of 55.5 (17.4) years and included 42 159 men (54.8%). The personalized model with transfer learning outperformed the global model for AKI estimation in terms of AUROC among general inpatients (0.78 [95% CI, 0.77-0.79] vs 0.76 [95% CI, 0.75-0.76]; P < .001) and across the high-risk subgroups (0.79 [95% CI, 0.78-0.80] vs 0.75 [95% CI, 0.74-0.77]; P < .001) and low-risk subgroups (0.74 [95% CI, 0.73-0.75] vs 0.71 [95% CI, 0.70-0.72]; P < .001). The AUROC improvement reached 0.13 for the high-risk subgroups, such as those undergoing liver transplant and cardiac surgery. Moreover, the personalized model with transfer learning performed better than or comparably with the best published models in well-studied AKI subgroups. Predictor outcomes varied significantly between patients, and interaction analysis uncovered modifiers of the predictor outcomes.

CONCLUSIONS AND RELEVANCE

Results of this study demonstrated that a personalized modeling with transfer learning is an improved AKI risk estimation approach that can be used across diverse patient subgroups. Risk factor heterogeneity and interactions at the individual level highlighted the need for agile, personalized care.

Effect of contrast administration on the renal function of predialysis patients undergoing fistuloplasty

OBJECTIVE

The aim of this study was to investigate if administration of iodinated contrast during endovascular interventions in arteriovenous fistula (AVF) in patients not requiring dialysis (predialysis patients) (1) negatively affects their renal function and (2) if oral hydration has a protective effect.

METHODS

All pre-dialysis patients who underwent endovascular interventions in AVF between August 2010 and April 2019 were included in the study. During the procedures, 35-50mL of Iodixanol were administered. A pre-hydration protocol was introduced in March 2015. Data were grouped before and after this date. The difference between pre- and post-contrast estimated glomerular filtration rate (eGFR) and the difference between the eGFR of hydrated and non-hydrated groups were calculated.

RESULTS

Eighty four patients who underwent 151 procedures were included in the study. In 60.3% of procedures a mean decrease of 1.35 mL/min/1.73 m² in eGFR was noted (95%CI 1.02, 1.69), while in 35.1% there was a mean increase of 1.06 mL/min/1.73 m² (95%CI 0.84, 1.28). The mean difference between pre- and post-procedure eGFR was -0.44 mL/min/1.73 m² (95% CI -0.72, -0.16) (p=.002). Oral hydration was associated with a smaller mean change in eGFR of -0.32 mL/min/1.73 m² (95%CI -0.62, -0.03) compared to the non-hydrated group with mean change of -0.47 mL/min/1.73 m² (95% CI -0.91, -0.03), but this was not statistically significant (p=.586).

CONCLUSIONS

This study demonstrates that administration of up to 50mL of iodinated contrast for endovascular interventions in AVF in predialysis patients has minimal adverse effect on the eGFR with questionable clinical significance. In addition, oral hydration before and after the procedure has only a mild protective effect against a decrease in eGFR.

A Biomarker‐Enhanced Model for Prediction of Acute Kidney Injury and Cardiovascular Risk Following Angiographic Procedures: CASABLANCA AKI Prediction Substudy

Background

The 2020 Acute Disease Quality Initiative Consensus provided recommendations on novel acute kidney injury biomarkers. In this study, we sought to assess the added value of novel kidney biomarkers to a clinical score in the CASABLANCA (Catheter Sampled Blood Archive in Cardiovascular Diseases) study.

Methods and Results

We evaluated individuals undergoing coronary and/or peripheral angiography and added 4 candidate biomarkers for acute kidney injury (kidney injury molecule‐1, interleukin‐18, osteopontin, and cystatin C) to a previously described contrast‐associated acute kidney injury (CA‐AKI) risk score. Participants were categorized into integer score groups based on the risk assigned by the biomarker‐enhanced CA‐AKI model. Risk for incident cardiorenal outcomes during a median 3.7 years of follow‐up was assessed. Of 1114 participants studied, 55 (4.94%) developed CA‐AKI. In adjusted models, neither kidney injury molecule‐1 nor interleukin‐18 improved discrimination for CA‐AKI; addition of osteopontin and cystatin C to the CA‐AKI clinical model significantly increased the c‐statistic from 0.69 to 0.73 (P for change <0.001) and resulted in a Net Reclassification Index of 59.4. Considering those with the lowest CA‐AKI integer score as a reference, the intermediate, high‐risk, and very‐high‐risk groups were associated with adverse cardiorenal outcomes. The corresponding hazard ratios of the very‐high‐risk group were 3.39 (95% CI, 2.14–5.38) for nonprocedural acute kidney injury, 5.58 (95% CI, 3.23–9.63) for incident chronic kidney disease, 6.21 (95% CI, 3.67–10.47) for myocardial infarction, and 8.94 (95% CI, 4.83–16.53) for all‐cause mortality.

Conclusions

A biomarker‐enhanced risk model significantly improves the prediction of CA‐AKI beyond clinical variables alone and may stratify the risk of future cardiorenal outcomes.

Registration

URL: https://www.clinicaltrials.gov; Unique identifier: NCT00842868.

Long-Term Clinical Impact of Contrast-Associated Acute Kidney Injury Following PCI: An ADAPT-DES Substudy

OBJECTIVES

This study sought to determine correlates and consequences of contrast-associated acute kidney injury (CA-AKI) on clinical outcomes in patients with or without pre-existing chronic kidney disease (CKD).

BACKGROUND

The incidence and impact of CA-AKI on clinical outcomes during contemporary percutaneous coronary intervention (PCI) are not fully defined.

METHODS

The ADAPT-DES (Assessment of Dual AntiPlatelet Therapy With Drug Eluting Stents) study was a prospective, multicenter registry of 8,582 patients treated with ≥1 drug-eluting stent(s). CA-AKI was defined as a post-PCI increase in serum creatinine of >0.5 mg/dL or a relative increase of ≥25% compared with pre-PCI. CKD was defined as estimated glomerular filtration rate <60 mL/min/1.73 m². The primary endpoint was the 2-year rate of net adverse clinical events (NACE): All-cause mortality, myocardial infarction (MI), definite or probable stent thrombosis, or major bleeding.

RESULTS

Of 7287 (85%) patients with evaluable data, 476 (6.5%) developed CA-AKI. In a multivariable model, older age, female sex, Caucasian race, congestive heart failure, diabetes, hypertension, CKD, presentation with ST-segment elevation MI, Killip class II to IV, radial access, intra-aortic balloon pump use, hypotension, and number of stents were independent predictors of CA-AKI. The 2-year NACE rate was higher in patients with CA-AKI (adjusted HR: 1.88; 95% CI: 1.42-2.49), as was each component of NACE (all-cause mortality, HR: 1.77; 95% CI: 1.22-2.55; MI, HR: 1.67; 95% CI: 1.18-2.36; definite/probable stent thrombosis, HR: 1.71; 95% CI: 1.10-2.65; and major bleeding, HR: 1.38; 95% CI: 1.06-1.80). Compared with the CA-AKI-/CKD- group, the CA-AKI+/CKD- (HR: 1.83; 95% CI: 1.33-2.52), CA-AKI-/CKD+ (HR: 1.56; 95% CI: 1.15-2.13), CA-AKI+/CKD+ (HR: 3.29; 95% CI: 1.92-5.67), and maintenance dialysis (HR: 2.67; 95% CI: 1.65-4.31) groups were at higher risk of NACE.

CONCLUSIONS

CA-AKI was relatively common after contemporary PCI and was associated with increased 2-year rates of NACE. Patients with pre-existing CKD were at particularly high risk for NACE after CA-AKI.

Unsupervised clustering reveals phenotypes of AKI in ICU COVID-19 patients

Background

Acute Kidney Injury (AKI) is a very frequent condition, occurring in about one in three patients admitted to an intensive care unit (ICU). AKI is a syndrome defined as a sudden decrease in glomerular filtration rate. However, this unified definition does not reflect the various mechanisms involved in AKI pathophysiology, each with its own characteristics and sensitivity to therapy. In this study, we aimed at developing an innovative machine learning based method able to subphenotype AKI according to its pattern of risk factors.

Methods

We adopted a three-step pipeline of analyses. First, we looked for factors associated with AKI using a generalized additive model. Second, we calculated the importance of each identified AKI related factor in the estimated AKI risk to find the main risk factor for AKI, at the single patient level. Lastly, we clusterized AKI patients according to their profile of risk factors and compared the clinical characteristics and outcome of every cluster. We applied this method to a cohort of severe COVID-19 patients hospitalized in the ICU of the Geneva University Hospitals.

Results

Among the 248 patients analyzed, we found 7 factors associated with AKI development. Using the individual expression of these factors, we identified three groups of AKI patients, based on the use of Lopinavir/Ritonavir, baseline eGFR, use of dexamethasone and AKI severity. The three clusters expressed distinct characteristics in terms of AKI severity and recovery, metabolic patterns and hospital mortality.

Conclusion

We propose here a new method to phenotype AKI patients according to their most important individual risk factors for AKI development. When applied to an ICU cohort of COVID-19 patients, we were able to differentiate three groups of patients. Each expressed specific AKI characteristics and outcomes, which probably reflect a distinct pathophysiology.

Does Artificial Intelligence Make Clinical Decision Better? A Review of Artificial Intelligence and Machine Learning in Acute Kidney Injury Prediction

Acute kidney injury (AKI) is a common complication of hospitalization that greatly and negatively affects the short-term and long-term outcomes of patients. Current guidelines use serum creatinine level and urine output rate for defining AKI and as the staging criteria of AKI. However, because they are not sensitive or specific markers of AKI, clinicians find it difficult to predict the occurrence of AKI and prescribe timely treatment. Advances in computing technology have led to the recent use of machine learning and artificial intelligence in AKI prediction, recent research reported that by using electronic health records (EHR) the AKI prediction via machine-learning models can reach AUROC over 0.80, in some studies even reach 0.93. Our review begins with the background and history of the definition of AKI, and the evolution of AKI risk factors and prediction models is also appraised. Then, we summarize the current evidence regarding the application of e-alert systems and machine-learning models in AKI prediction.

Artificial Intelligence and Mapping a New Direction in Laboratory Medicine: A Review

BACKGROUND

Modern artificial intelligence (AI) and machine learning (ML) methods are now capable of completing tasks with performance characteristics that are comparable to those of expert human operators. As a result, many areas throughout healthcare are incorporating these technologies, including in vitro diagnostics and, more broadly, laboratory medicine. However, there are limited literature reviews of the landscape, likely future, and challenges of the application of AI/ML in laboratory medicine.

CONTENT

In this review, we begin with a brief introduction to AI and its subfield of ML. The ensuing sections describe ML systems that are currently in clinical laboratory practice or are being proposed for such use in recent literature, ML systems that use laboratory data outside the clinical laboratory, challenges to the adoption of ML, and future opportunities for ML in laboratory medicine.

SUMMARY

AI and ML have and will continue to influence the practice and scope of laboratory medicine dramatically. This has been made possible by advancements in modern computing and the widespread digitization of health information. These technologies are being rapidly developed and described, but in comparison, their implementation thus far has been modest. To spur the implementation of reliable and sophisticated ML-based technologies, we need to establish best practices further and improve our information system and communication infrastructure. The participation of the clinical laboratory community is essential to ensure that laboratory data are sufficiently available and incorporated conscientiously into robust, safe, and clinically effective ML-supported clinical diagnostics.

Patient-centered contrast thresholds to reduce acute kidney injury in high-risk patients undergoing percutaneous coronary intervention

BACKGROUND

Contrast volume used during percutaneous coronary intervention has a direct relationship with contrast-associated acute kidney injury. While several models estimate the risk of contrast-associated acute kidney injury, only the strategy of limiting contrast volume to 3 × estimated glomerular filtration rate (eGFR) gives actionable estimates of safe contrast volume doses. However, this method does not consider other patient characteristics associated with risk, such as age, diabetes or heart failure.

METHODS

Using the National Cardiovascular Data Registry acute kidney injury risk model, we developed a novel strategy to define safe contrast limits by entering a contrast term into the model and using it to meet specific (eg, 10%) relative risk reductions. We then estimated acute kidney injury rates when our patient-centered model-derived thresholds were and were not exceeded using data from CathPCI version 5 between April 2018 and June 2019. We repeated the same analysis in a sub-set of patients who received ≤3 × eGFR contrast.

RESULTS

After excluding patients on hemodialysis, below average risk (<7%), missing data and multiple percutaneous coronary interventions, our final analytical cohort included 141,133 patients at high risk for acute kidney injury. The rate of acute kidney injury was 10.0% when the contrast thresholds derived from our patient-centered model were met and 18.2% when they were exceeded (P < .001). In patients who received contrast ≤3 × eGFR (n = 82,318), contrast-associated acute kidney injury rate was 9.8% when the contrast thresholds derived from our patient centered model were met and 14.5% when they were exceeded (P < .001).

CONCLUSIONS

A novel strategy for developing personalized contrast volume thresholds, provides actionable information for providers that could decrease rates of contrast-associated acute kidney injury. This strategy needs further prospective testing to assess efficacy in improving patient outcomes.

Refining Safe Contrast Limits for Preventing Acute Kidney Injury After Percutaneous Coronary Intervention

Background

Contrast‐associated acute kidney injury (CA‐AKI) is associated with substantial morbidity and may be prevented by using less contrast during percutaneous coronary intervention (PCI). However, tools for determining safe contrast volumes are limited. We developed risk models to tailor safe contrast volume limits during PCI.

Methods and Results

Using data from all PCIs performed at 18 hospitals from January 2015 to March 2018, we developed logistic regression models for predicting CA‐AKI, including simpler models (“pragmatic full,” “pragmatic minimum”) using only predictors easily derivable from electronic health records. We prospectively validated these models using PCI data from April 2018 to December 2018 and compared them to preexisting safe contrast models using the area under the receiver operating characteristic curve (AUC). The model derivation data set included 20 579 PCIs with 2102 CA‐AKI cases. When applying models to the separate validation data set (5423 PCIs, 488 CA‐AKI cases), prior safe contrast limits (5*Weight/Creatinine, 2*CreatinineClearance) were poor measures of safety with accuracies of 53.7% and 56.6% in predicting CA‐AKI, respectively. The full, pragmatic full, and pragmatic minimum models performed significantly better (accuracy, 73.1%, 69.3%, 66.6%; AUC, 0.80, 0.76, 0.72 versus 0.59 for 5 * Weight/Creatinine, 0.61 for 2*CreatinineClearance). We found that applying safe contrast limits could meaningfully reduce CA‐AKI risk in one‐quarter of patients.

Conclusions

Compared with preexisting equations, new multivariate models for safe contrast limits were substantially more accurate in predicting CA‐AKI and could help determine which patients benefit most from limiting contrast during PCI. Using readily available electronic health record data, these models could be implemented into electronic health records to provide actionable information for improving PCI safety.

[Contrast-induced acute kidney injury in patients with stable coronary artery disease: the most important risk factors and prevalence]

AIM

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.

MATERIALS AND METHODS

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.

RESULTS

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.

CONCLUSION

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.

Artificial Intelligence in Acute Kidney Injury Risk Prediction

Acute kidney injury (AKI) is a frequent complication in hospitalized patients, which is associated with worse short and long-term outcomes. It is crucial to develop methods to identify patients at risk for AKI and to diagnose subclinical AKI in order to improve patient outcomes. The advances in clinical informatics and the increasing availability of electronic medical records have allowed for the development of artificial intelligence predictive models of risk estimation in AKI. In this review, we discussed the progress of AKI risk prediction from risk scores to electronic alerts to machine learning methods.