Predicting Major Adverse Kidney Events among Critically Ill Adults Using the Electronic Health Record

Kidney Outcomes and Trajectories of Tubular Injury and Function in Critically Ill Patients With and Without COVID-19

IMPORTANCE

COVID-19 may injure the kidney tubules via activation of inflammatory host responses and/or direct viral infiltration. Most studies of kidney injury in COVID-19 lacked contemporaneous controls or measured kidney biomarkers at a single time point.

OBJECTIVES

To better understand mechanisms of acute kidney injury in COVID-19, we compared kidney outcomes and trajectories of tubular injury, viability, and function in prospectively enrolled critically ill adults with and without COVID-19.

DESIGN, SETTING, AND PARTICIPANTS

The COVID-19 Host Response and Outcomes study prospectively enrolled patients admitted to ICUs in Washington State with symptoms of lower respiratory tract infection, determining COVID-19 status by nucleic acid amplification on arrival.

MAIN OUTCOMES AND MEASURES

We evaluated major adverse kidney events (MAKE) defined as a doubling of serum creatinine, kidney replacement therapy, or death, in 330 patients after inverse probability weighting. In the 181 patients with available biosamples, we determined trajectories of urine kidney injury molecule-1 (KIM-1) and epithelial growth factor (EGF), and urine:plasma ratios of endogenous markers of tubular secretory clearance.

RESULTS

At ICU admission, the mean age was 55 ± 16 years; 45% required mechanical ventilation; and the mean serum creatinine concentration was 1.1 mg/dL. COVID-19 was associated with a 70% greater occurrence of MAKE (relative risk 1.70; 95% CI, 1.05-2.74) and a 741% greater occurrence of KRT (relative risk 7.41; 95% CI, 1.69-32.41). The biomarker cohort had a median of three follow-up measurements. Urine EGF, secretory clearance ratios, and estimated glomerular filtration rate (eGFR) increased over time in the COVID-19 negative group but remained unchanged in the COVID-19 positive group. In contrast, urine KIM-1 concentrations did not significantly change over the course of the study in either group.

CONCLUSIONS

Among critically ill adults, COVID-19 is associated with a more protracted course of proximal tubular dysfunction and reduced eGFR despite similar degrees of kidney injury.

Heterogeneity in the definition of major adverse kidney events: a scoping review

Acute kidney injury (AKI) is associated with persistent renal dysfunction, the receipt of dialysis, dialysis dependence, and mortality. Accordingly, the concept of major adverse kidney events (MAKE) has been adopted as an endpoint for assessing the impact of AKI. However, applied criteria or observation periods for operationalizing MAKE appear to vary across studies. To evaluate this heterogeneity for MAKE evaluation, we performed a systematic scoping review of studies that employed MAKE as an AKI endpoint. Four major academic databases were searched, and we identified 122 studies with increasing numbers over time. We found marked heterogeneity in applied criteria and observation periods for MAKE across these studies, with some even lacking a description of criteria. Moreover, 13 different observation periods were employed, with 30 days and 90 days as the most common. Persistent renal dysfunction was evaluated by estimated glomerular filtration rate (34%) or serum creatinine concentration (48%); however, 37 different definitions for this component were employed in terms of parameters, cut-off criteria, and assessment periods. The definition for the dialysis component also showed significant heterogeneity regarding assessment periods and duration of dialysis requirement (chronic vs temporary). Finally, MAKE rates could vary by 7% [interquartile range: 1.7-16.7%] with different observation periods or by 36.4% with different dialysis component definitions. Our findings revealed marked heterogeneity in MAKE definitions, particularly regarding component assessment and observation periods. Dedicated discussion is needed to establish uniform and acceptable standards to operationalize MAKE in terms of selection and applied criteria of components, observation period, and reporting criteria for future trials on AKI and related conditions.

Effect modification of dosing strategy (AUC or trough) on AKI associated with vancomycin in combination with piperacillin/tazobactam or cefepime and meropenem

Piperacillin-tazobactam (TZP), cefepime (FEP), or meropenem (MEM) and vancomycin (VAN) are commonly used in combination for sepsis. Studies have shown an increased risk of acute kidney injury (AKI) with TZP and VAN compared to FEP or MEM. VAN guidelines recommend area under the curve (AUC) monitoring over trough (Tr) to minimize the risk of AKI. We investigated the association of AKI and MAKE-30 with the two VAN monitoring strategies when used in combination with TZP or FEP/MEM. Adult patients between 2015 and 2019 with VAN > 72 hours were included. Patients with AKI prior to or within 48 hours of VAN or baseline CrCl of ≤30 mL/min were excluded. Four cohorts were defined: FEP/MEM/Tr, FEP/MEM/AUC, TZP/Tr, and TZP/AUC. A Cox Proportional Hazard Model was used to model AKI as a function of the incidence rate of at-risk days, testing monitoring strategy as a treatment effect modification. Multivariable logistic regression was used to model MAKE-30. Overall incidence of AKI was 18.6%; FEP/MEM/Tr = 115 (14.6%), FEP/MEM/AUC = 52 (14.9%), TZP/Tr = 189 (26%), and TZP/AUC = 96 (17.1%) (P < 0.001). Both drug group [(TZP; P = 0.0085)] and monitoring strategy [(Tr; P = 0.0007)] were highly associated with the development of AKI; however, the effect was not modified with interaction term [(TZP*Tr); 0.085)]. The odds of developing MAKE-30 were not different between any group and FEP/MEM/AUC. The effect of VAN/TZP on the development of AKI was not modified by the VAN monitoring strategy (AUC vs trough). MAKE-30 outcomes were not different among the four cohorts.

Current perspectives on the use of artificial intelligence in critical patient safety

Intensive Care Units (ICUs) have undergone enhancements in patient safety, and artificial intelligence (AI) emerges as a disruptive technology offering novel opportunities. While the published evidence is limited and presents methodological issues, certain areas show promise, such as decision support systems, detection of adverse events, and prescription error identification. The application of AI in safety may pursue predictive or diagnostic objectives. Implementing AI-based systems necessitates procedures to ensure secure assistance, addressing challenges including trust in such systems, biases, data quality, scalability, and ethical and confidentiality considerations. The development and application of AI demand thorough testing, encompassing retrospective data assessments, real-time validation with prospective cohorts, and efficacy demonstration in clinical trials. Algorithmic transparency and explainability are essential, with active involvement of clinical professionals being crucial in the implementation process.

Kidney Outcomes and Trajectories of Tubular Injury and Function in Critically Ill Persons with and without Coronavirus-2019

Background

Coronavirus disease-2019 (COVID-19) may injure the kidney tubules via activation of inflammatory host responses and/or direct viral infiltration. Most studies of kidney injury in COVID-19 lacked contemporaneous controls or measured kidney biomarkers at a single time point. To better understand mechanisms of AKI in COVID-19, we compared kidney outcomes and trajectories of tubular injury, viability, and function in prospectively enrolled critically ill adults with and without COVID-19.

Methods

The COVID-19 Host Response and Outcomes (CHROME) study prospectively enrolled patients admitted to intensive care units in Washington state with symptoms of lower respiratory tract infection, determining COVID-19 status by nucleic acid amplification on arrival. We evaluated major adverse kidney events (MAKE) defined as a doubling of serum creatinine, kidney replacement therapy, or death, in 330 patients after inverse probability weighting. In the 181 patients with available biosamples, we determined trajectories of urine kidney injury molecule-1 (KIM-1) and epithelial growth factor (EGF), and urine:plasma ratios of endogenous markers of tubular secretory clearance.

Results

At ICU admission, mean age was 55±16 years; 45% required mechanical ventilation; and mean serum creatinine concentration was 1.1 mg/dL. COVID-19 was associated with a 70% greater incidence of MAKE (95% CI 1.05, 2.74) and a 741% greater incidence of KRT (95% CI 1.69, 32.41). The biomarker cohort had a median of three follow-up measurements. Urine EGF, secretory clearance ratios, and eGFR increased over time in the COVID-19 negative group but remained unchanged in the COVID-19 positive group. In contrast, urine KIM-1 concentrations did not significantly change over the course of the study in either group.

Conclusions

Among critically ill adults, COVID-19 is associated with a more protracted course of proximal tubular dysfunction.

An early warning model for predicting major adverse kidney events within 30 days in sepsis patients

Background

In sepsis patients, kidney damage is among the most dangerous complications, with a high mortality rate. In addition, major adverse kidney events within 30 days (MAKE30) served as a comprehensive and unbiased clinical outcome measure for sepsis patients due to the recent shift toward targeting patient-centered renal outcomes in clinical research. However, the underlying predictive model for the prediction of MAKE30 in sepsis patients has not been reported in any study.

Methods

A cohort of 2,849 sepsis patients from the Medical Information Mart for Intensive Care (MIMIC)-IV database was selected and subsequently allocated into a training set (n = 2,137, 75%) and a validation set (n = 712, 25%) through randomization. In addition, 142 sepsis patients from the Xi'An No. 3 Hospital as an external validation group. Univariate and multivariate logistic regression analyses were conducted to ascertain the independent predictors of MAKE30. Subsequently, a nomogram was developed utilizing these predictors, with an area under curve (AUC) above 0.6. The performance of nomogram was assessed through calibration curve, receiver operating characteristics (ROC) curve, and decision curve analysis (DCA). The secondary outcome was 30-day mortality, persistent renal dysfunction (PRD), and new renal replacement therapy (RRT). MAKE30 were a composite of death, PRD, new RRT.

Results

The construction of the nomogram was based on several independent predictors (AUC above 0.6), including age, respiratory rate (RR), PaO2, lactate, and blood urea nitrogen (BUN). The predictive model demonstrated satisfactory discrimination for MAKE30, with an AUC of 0.740, 0.753, and 0.821 in the training, internal validation, and external validation cohorts, respectively. Furthermore, the simple prediction model exhibited superior predictive value compared to the SOFA model in both the training (AUC = 0.710) and validation (AUC = 0.692) cohorts. The nomogram demonstrated satisfactory calibration and clinical utility as evidenced by the calibration curve and DCA. Additionally, the predictive model exhibited excellent accuracy in forecasting 30-day mortality (AUC = 0.737), PRD (AUC = 0.639), and new RRT (AUC = 0.846) within the training dataset. Additionally, the model displayed predictive power for 30-day mortality (AUC = 0.765), PRD (AUC = 0.667), and new RRT (AUC = 0.783) in the validation set.

Conclusion

The proposed nomogram holds the potential to estimate the risk of MAKE30 promptly and efficiently in sepsis patients within the initial 24 h of admission, thereby equipping healthcare professionals with valuable insights to facilitate personalized interventions.

Inflammation and renal function decline in chronic coronary syndrome: a prospective multicenter cohort study

BACKGROUND

Renal function decline is a frequently encountered complication in patients with chronic coronary syndrome. Aside from traditional cardiovascular risk factors, the inflammatory burden emerged as the novel phenotype that compromised renal prognosis in such population.

METHODS

A cohort with chronic coronary syndrome was enrolled to investigate the association between inflammatory status and renal dysfunction. Levels of inflammatory markers, including high-sensitivity C-reactive protein (hs-CRP), tumour necrosis factor-α (TNF-α), adiponectin, matrix metalloproteinase-9, interleukin-6, lipoprotein-associated phospholipase A2, were assessed. Renal event was defined as > 25% decline in estimated glomerular filtration rate (eGFR). Inflammatory scores were calculated based on the aggregate of hs-CRP, TNF-α, and adiponectin levels.

RESULTS

Among the 850 enrolled subjects, 145 patients sustained a renal event during an averaged 3.5 years follow-up. Multivariate analysis with Cox regression suggested elevations in hs-CRP, TNF-α, and adiponectin levels were independent risk factors for the occurrence of a renal event. Whereas, Kaplan-Meier curve illustrated significant correlation between high TNF-α (P = 0.005), adiponectin (P < 0.001), but not hs-CRP (P = 0.092), and eGFR decline. The aggregative effect of these biomarkers was also distinctly correlated with renal events (score 2: P = 0.042; score 3: P < 0.001).

CONCLUSIONS

Inflammatory burden was associated with eGFR decline in patients with chronic coronary syndrome.

Predicting Long-term Outcomes in Deceased Donor Kidney Transplant Recipients Using Three Short-term Graft Characteristics

Key Points

Delayed graft function is not an ideal measure of graft function, yet is used to assess risk in kidney transplantation. We propose a model that combines it with two other measures of 90-day graft function to identify recipients at incremental risk of inferior long-term outcomes.

Background

Delayed graft function (DGF) in kidney transplant recipients is used to determine graft prognosis, make organ utilization decisions, and as an important end point in clinical trials. However, DGF is not an ideal measure of graft function. We aimed to develop and validate a model that provides incremental risk assessment for inferior patient and graft outcomes.

Methods

We included adult kidney-only deceased donor transplant recipients from 1996 to 2016. In addition to DGF, two short-term measures were used to assess risk: renal function recovery <100% (attaining half the donor's eGFR) and recipient's 90-day eGFR <30. Recipients were at no, low, moderate, or high risk if they met zero, one, two, or all criteria, respectively. Cox proportional hazard models were used to assess the independent relationship between exposure and death-censored graft failure (DCGF) and mortality.

Results

Of the 792 eligible recipients, 24.5% experienced DGF, 40.5% had renal function recovery <100%, and 6.9% had eGFR <30. Over a median follow-up of 7.3 years, the rate of DCGF was 18.7% and mortality was 25.1%. When compared with recipients at no risk, those at low, moderate, and high risk were noted to have an increase in risk of DCGF (adjusted hazard ratio [aHR], 1.53; 95% confidence interval [CI], 1.03 to 2.27; aHR, 2.84; 95% CI, 1.68 to 4.79; aHR, 15.46; 95% CI, 8.04 to 29.71) and mortality (aHR, 1.16; 95% CI, 0.84 to 1.58; aHR, 1.85; 95% CI, 1.13 to 3.07; aHR, 2.66; 95% CI, 1.19 to 5.97). When using a hierarchical approach, each additional exposure predicted the risk of DCGF better than DGF alone and 100 random bootstrap replications supported the internal validity of the risk model. In an external validation cohort deemed to be at lower risk of DCGF, similar nonsignificant trends were noted.

Conclusion

We propose a risk model that provides an incremental assessment of recipients at higher risk of adverse long-term outcomes than DGF alone. This can help advance the field of risk assessment in transplantation and inform therapeutic decision making in patients at the highest spectrum of inferior outcomes.

Longer time to recovery from acute kidney injury is associated with major adverse kidney events in patients with cirrhosis

BACKGROUND

In patients with cirrhosis and acute kidney injury (AKI), longer time to AKI-recovery may increase the risk of subsequent major-adverse-kidney-events (MAKE).

AIMS

To examine the association between timing of AKI-recovery and risk of MAKE in patients with cirrhosis.

METHODS

Hospitalised patients with cirrhosis and AKI (n = 5937) in a nationwide database were assessed for time to AKI-recovery and followed for 180-days. Timing of AKI-recovery (return of serum creatinine <0.3 mg/dL of baseline) from AKI-onset was grouped by Acute-Disease-Quality-Initiative Renal Recovery consensus: 0-2, 3-7, and >7-days. Primary outcome was MAKE at 90-180-days. MAKE is an accepted clinical endpoint in AKI and defined as the composite outcome of ≥25% decline in estimated-glomerular-filtration-rate (eGFR) compared with baseline with the development of de-novo chronic-kidney-disease (CKD) stage ≥3 or CKD progression (≥50% reduction in eGFR compared with baseline) or new haemodialysis or death. Landmark competing-risk multivariable analysis was performed to determine the independent association between timing of AKI-recovery and risk of MAKE.

RESULTS

4655 (75%) achieved AKI-recovery: 0-2 (60%), 3-7 (31%), and >7-days (9%). Cumulative-incidence of MAKE was 15%, 20%, and 29% for 0-2, 3-7, >7-days recovery groups, respectively. On adjusted multivariable competing-risk analysis, compared to 0-2-days, recovery at 3-7 and >7-days was independently associated with an increased risk for MAKE: sHR 1.45 (95% CI 1.01-2.09, p = 0.042), sHR 2.33 (95% CI 1.40-3.90, p = 0.001), respectively.

CONCLUSION

Longer time to recovery is associated with an increased risk of MAKE in patients with cirrhosis and AKI. Further research should examine interventions to shorten AKI-recovery time and its impact on subsequent outcomes.

Association between Inflammation-Based Parameters and Prognosis in Patients with Acute Kidney Injury

Background and Objectives: this study aimed to clarify the relationship between inflammation-based parameters and prognosis in patients with acute kidney injury (AKI). Materials and Methods: We analyzed the prospectively collected data of patients with AKI, who were admitted through the emergency department between March 2020 and April 2021. Their clinical characteristics, inflammation-based parameters, resolving/non-resolving AKI pattern, and major adverse kidney event (MAKE) rates were analyzed. Results: Among 177 patients, 129 (72.9%) had a resolving AKI pattern and 48 (27.1%) had a non-resolving AKI pattern. The outcome of MAKE occurred in 30 (16.9%) participants. Multivariate analyses showed that the neutrophil-to-monocyte ratio was an independent predictor of resolving AKI, and that the neutrophil-to-monocyte and neutrophil-to-lymphocyte ratios were independent predictors of MAKE occurrence. Conclusions: we demonstrated that inflammation-based parameters are valuable predictors of early recovery and MAKE occurrence in patients with AKI.

Use of Electronic Health Data for Disease Prediction: A Comprehensive Literature Review

Disease prediction has the potential to benefit stakeholders such as the government and health insurance companies. It can identify patients at risk of disease or health conditions. Clinicians can then take appropriate measures to avoid or minimize the risk and in turn, improve quality of care and avoid potential hospital admissions. Due to the recent advancement of tools and techniques for data analytics, disease risk prediction can leverage large amounts of semantic information, such as demographics, clinical diagnosis and measurements, health behaviours, laboratory results, prescriptions and care utilisation. In this regard, electronic health data can be a potential choice for developing disease prediction models. A significant number of such disease prediction models have been proposed in the literature over time utilizing large-scale electronic health databases, different methods, and healthcare variables. The goal of this comprehensive literature review was to discuss different risk prediction models that have been proposed based on electronic health data. Search terms were designed to find relevant research articles that utilized electronic health data to predict disease risks. Online scholarly databases were searched to retrieve results, which were then reviewed and compared in terms of the method used, disease type, and prediction accuracy. This paper provides a comprehensive review of the use of electronic health data for risk prediction models. A comparison of the results from different techniques for three frequently modelled diseases using electronic health data was also discussed in this study. In addition, the advantages and disadvantages of different risk prediction models, as well as their performance, were presented. Electronic health data have been widely used for disease prediction. A few modelling approaches show very high accuracy in predicting different diseases using such data. These modelling approaches have been used to inform the clinical decision process to achieve better outcomes.

Prediction Models for AKI in ICU: A Comparative Study

PURPOSE

To assess the performance of models for early prediction of acute kidney injury (AKI) in the Intensive Care Unit (ICU) setting.

PATIENTS AND METHODS

Data were collected from the Medical Information Mart for Intensive Care (MIMIC)-III database for all patients aged ≥18 years who had their serum creatinine (SCr) level measured for 72 h following ICU admission. Those with existing conditions of kidney disease upon ICU admission were excluded from our analyses. Seventeen predictor variables comprising patient demographics and physiological indicators were selected on the basis of the Kidney Disease Improving Global Outcomes (KDIGO) and medical literature. Six models from three types of methods were tested: Logistic Regression (LR), Support Vector Machines (SVM), Random Forest (RF), eXtreme Gradient Boosting (XGBoost), Light Gradient Boosting Decision Machine (LightGBM), and Convolutional Neural Network (CNN). The area under receiver operating characteristic curve (AUC), accuracy, precision, recall and F-measure (F1) were calculated for each model to evaluate performance.

RESULTS

We extracted the ICU records of 17,205 patients from MIMIC-III dataset. LightGBM had the best performance, with all evaluation indicators achieving the highest value (average AUC = 0.905, F1 = 0.897, recall = 0.836). XGBoost had the second best performance and LR, RF, SVM performed similarly (P = 0.082, 0.158 and 0.710, respectively) on AUC. The CNN model achieved the lowest score for accuracy, precision, F1 and AUC. SVM and LR had relatively low recall compared with that of the other models. The SCr level had the most significant effect on the early prediction of AKI onset in LR, RF, SVM and LightGBM.

CONCLUSION

LightGBM demonstrated the best capability for predicting AKI in the first 72 h of ICU admission. LightGBM and XGBoost showed great potential for clinical application owing to their high recall value. This study can provide references for artificial intelligence-powered clinical decision support systems for AKI early prediction in the ICU setting.

Kidney Biomarkers and Major Adverse Kidney Events in Critically Ill Patients

Background

Several biomarkers of AKI have been examined for their ability to predict AKI before serum creatinine. Few studies have focused on using kidney biomarkers to better predict major adverse kidney events (MAKE), an increasingly used composite outcome in critical care nephrology research.

Methods

Single-center prospective study collecting blood and urine samples from critically ill patients with AKI Kidney Disease Improving Global Outcomes stage 2 or above, and matched controls from a single, tertiary care intensive care unit (ICU). Samples were collected at 24-48 hours after AKI diagnosis (patients) or ICU admission (controls), 5-7 days later, and 4-6 weeks after discharge for patients with AKI. The primary outcome of interest was MAKE at hospital discharge (MAKE-DC), consisting of the composite end point of death, RRT dependence, or a decrease in estimated glomerular filtration to <75% of baseline.

Results

Serum/urinary neutrophil gelatinase-associated lipocalin (NGAL), serum/urinary cystatin C, and urinary kidney injury molecule-1 early in the AKI or ICU course were all significantly higher in patients with MAKE-DC compared with those not experiencing MAKE-DC. Additionally, serum/urinary NGAL and serum cystatin C measurements at the first time point remained significantly associated with MAKE events at 3, 6, and 12 months. Serum cystatin C, and to a lesser extent serum NGAL, significantly improved upon a logistic regression clinical prediction model of MAKE-DC (AUROC 0.94 and 0.87 versus 0.83; P=0.001 and P=0.02, respectively). Patients without MAKE-DC experienced a greater decline in serum NGAL from first to second measurement than those patients experiencing MAKE-DC.

Conclusions

Early measures of kidney biomarkers in patients who are critically ill are associated with MAKE-DC. This relationship appears to be greatest with serum NGAL and cystatin C, which display additive utility to a clinical prediction model. Trending serum NGAL may also have utility in predicting MAKE-DC.

SEA-MAKE score as a tool for predicting major adverse kidney events in critically ill patients with acute kidney injury: results from the SEA-AKI study

BACKGROUND

Acute kidney injury (AKI) is a common problem in critically ill patients and associated with high rates of morbidity and mortality. Recently, Major Adverse Kidney Events (MAKE) were introduced as important kidney endpoints. If these endpoints can be predicted, then it may help the physicians to identify high-risk patients and provide the opportunity to have targeted preventive therapy. The objective of this study was to create a simplified scoring system to predict MAKE within 28 days among AKI patients in ICU.

METHODS

This is a prospective web-based multicenter cohort study that was conducted in adults who were admitted to the ICU in 17 centers across Thailand from 2013 to 2015. A predicting score was derived from the regression equation with Receiver Operating Characteristic (ROC) analysis to evaluate the diagnostic test and produce predictive models. Internal validation was obtained using the bootstrapping method.

RESULTS

From 5071 cases, 2856 (56%) had AKI. Among those with AKI, 1749 (61%) had MAKE. Among those that have MAKE, there were 1175 (41.4%) deaths, 414 (14.4%) were on dialysis and 1154 (40.7%) had non-recovery renal function. The simplified score points of low Glasgow coma scale was 3, tachypnea was 1, vasopressor use was 1, on mechanical ventilation was 2, oliguria was 2, serum creatinine rising ≥ 3 times was 5, high blood urea nitrogen was 3, low hematocrit was 2, and thrombocytopenia was 1. The area under ROC curve for optimism corrected performance was 0.80 (0.78, 0.81). When the cut-off value was 7, the sensitivity, specificity, positive likelihood ratio, and positive predictive values were 0.75, 0.76, 3.10, and 0.84, respectively. When the scoring system was calibrated, the α intercept and β slope were 1.001 and 0, respectively.

CONCLUSIONS

SEA-MAKE scoring system is a new simplified clinical tool that can be used to predict major adverse kidney events in AKI patients. The simplicity of the scoring system is highly likely to be used in resource-limited settings. However, external validation is necessary before widespread use.

Association Between Early Recovery of Kidney Function After Acute Kidney Injury and Long-term Clinical Outcomes

IMPORTANCE

The severity of acute kidney injury (AKI) is usually determined based on the maximum serum creatinine concentration. However, the trajectory of kidney function recovery could be an additional important dimension of AKI severity.

OBJECTIVE

To assess whether the trajectory of kidney function recovery within 72 hours after AKI is associated with long-term risk of clinical outcomes.

DESIGN, SETTING, AND PARTICIPANTS

This prospective, multicenter cohort study enrolled 1538 adults with or without AKI 3 months after hospital discharge between December 1, 2009, and February 28, 2015. Statistical analyses were completed November 1, 2018. Participants with or without AKI were matched based on demographic characteristics, site, comorbidities, and prehospitalization estimated glomerular filtration rate. Participants with AKI were classified as having resolving or nonresolving AKI based on previously published definitions. Resolving AKI was defined as a decrease in serum creatinine concentration of 0.3 mg/dL or more or 25% or more from maximum in the first 72 hours after AKI diagnosis. Nonresolving AKI was defined as AKI not meeting the definition for resolving AKI.

MAIN OUTCOMES AND MEASURES

The primary outcome was a composite of major adverse kidney events (MAKE), defined as incident or progressive chronic kidney disease, long-term dialysis, or all-cause death during study follow-up.

RESULTS

Among 1538 participants (964 men; mean [SD] age, 64.6 [12.7] years), 769 (50%) had no AKI, 475 (31%) had a resolving AKI pattern, and 294 (19%) had a nonresolving AKI pattern. After a median follow-up of 4.7 years, the outcome of MAKE occurred in 550 (36%) of all participants. The adjusted hazard ratio for MAKE was higher for patients with resolving AKI (adjusted hazard ratio, 1.52; 95% CI, 1.01-2.29; P = .04) and those with nonresolving AKI (adjusted hazard ratio 2.30; 95% CI, 1.52-3.48; P < .001) compared with participants without AKI. Within the population of patients with AKI, nonresolving AKI was associated with a 51% greater risk of MAKE (95% CI, 22%-88%; P < .001) compared with resolving AKI. The higher risk of MAKE among patients with nonresolving AKI was explained by a higher risk of incident and progressive chronic kidney disease.

CONCLUSIONS AND RELEVANCE

This study suggests that the 72-hour period immediately after AKI distinguishes the risk of clinically important kidney-specific long-term outcomes. The identification of different AKI recovery patterns may improve patient risk stratification, facilitate prognostic enrichment in clinical trials, and enable recognition of patients who may benefit from nephrology consultation.

Informatics for the Modern Intensive Care Unit

Advanced informatics systems can help improve health care delivery and the environment of care for critically ill patients. However, identifying, testing, and deploying advanced informatics systems can be quite challenging. These processes often require involvement from a collaborative group of health care professionals of varied disciplines with knowledge of the complexities related to designing the modern and "smart" intensive care unit (ICU). In this article, we explore the connectivity environment within the ICU, middleware technologies to address a host of patient care initiatives, and the core informatics concepts necessary for both the design and implementation of advanced informatics systems.