Clinical Trajectories of Acute Kidney Injury in Surgical Sepsis: A Prospective Observational Study

Acute Kidney Injury Duration and 20-Year Risks of CKD and Cardiovascular Disease

Introduction

Acute kidney injury (AKI) is associated with chronic kidney disease (CKD) and cardiovascular disease (CVD); however, it is unclear whether AKI duration affects the long-term risks of CKD and CVD. Therefore, we performed a population-based cohort study examining the associations between AKI duration and CKD and CVD.

Methods

We identified patients with laboratory-recorded AKI in Denmark from 1990 through 2018. AKIs were categorized as rapid reversal AKI (≤48 hours), persistent AKI (2-7 days), and acute kidney disease (AKD) (>7 days). We estimated 20-year risks and adjusted hazard ratios (aHRs) of incident CKD and CVD.

Results

The study comprised 169,582 patients with AKI, with 100,478 and 76,838 included in the analysis of CKD and CVD, respectively. The 20-year risks of CKD were 26.3%, 29.5%, and 28.7% for rapid reversal AKI, persistent AKI, and AKD, respectively. Compared with rapid reversal AKI, aHRs were 1.13 (95% confidence interval [CI], 1.08-1.19) for persistent AKI and 1.36 (95% CI, 1.30-1.41) for AKD. Risks and rates of overall CVD were similar for rapid reversal AKI, persistent AKI, and AKD. However, persistent AKI was associated with a slightly increased aHR of heart failure (1.09; 95% CI, 1.02-1.16), and aHRs of heart failure, ischemic heart disease, and peripheral artery disease were slightly increased for AKD (1.09 [95% CI, 1.03-1.15], 1.11 [95% CI, 1.03-1.19], and 1.10 [95% CI, 1.02-1.17], respectively).

Conclusion

AKI duration was associated with development of CKD, but not overall CVD; however, rates of heart failure, ischemic heart disease, and peripheral artery disease increased slightly with AKI duration.

Sepsis-associated acute kidney injury: recent advances in enrichment strategies, sub-phenotyping and clinical trials

Acute kidney injury (AKI) often complicates sepsis and is associated with high morbidity and mortality. In recent years, several important clinical trials have improved our understanding of sepsis-associated AKI (SA-AKI) and impacted clinical care. Advances in sub-phenotyping of sepsis and AKI and clinical trial design offer unprecedented opportunities to fill gaps in knowledge and generate better evidence for improving the outcome of critically ill patients with SA-AKI. In this manuscript, we review the recent literature of clinical trials in sepsis with focus on studies that explore SA-AKI as a primary or secondary outcome. We discuss lessons learned and potential opportunities to improve the design of clinical trials and generate actionable evidence in future research. We specifically discuss the role of enrichment strategies to target populations that are most likely to derive benefit and the importance of patient-centered clinical trial endpoints and appropriate trial designs with the aim to provide guidance in designing future trials.

Persistent Acute Kidney Injury is Associated with Poor Outcomes and Increased Hospital Cost in Vascular Surgery

BACKGROUND

Postoperative acute kidney injury (AKI) is common after major surgery and is associated with increased morbidity, mortality, and cost. Additionally, there are recent studies demonstrating that time to renal recovery may have a substantial impact on clinical outcomes. We hypothesized that patients with delayed renal recovery after major vascular surgery will have increased complications, mortality, and hospital cost.

METHODS

A single-center retrospective cohort of patients undergoing nonemergent major vascular surgery between 6/1/2014 and 10/1/2020 was analyzed. Development of postoperative AKI (defined using Kidney Disease Improving Global Outcomes (KDIGO) criteria: >50% or > 0.3 mg/dl absolute increase in serum creatinine relative to reference after surgery and before discharge) was evaluated. Patients were divided into 3 groups: no AKI, rapidly reversed AKI (<48 hours), and persistent AKI (≥48 hours). Multivariable generalized linear models were used to evaluate the association between AKI groups and postoperative complications, 90-day mortality, and hospital cost.

RESULTS

A total of 1,881 patients undergoing 1,980 vascular procedures were included. Thirty five percent of patients developed postoperative AKI. Patients with persistent AKI had longer intensive care unit and hospital stays, as well as more mechanical ventilation days. In multivariable logistic regression analysis, persistent AKI was a major predictor of 90-day mortality (odds ratio 4.1, 95% confidence interval 2.4-7.1). Adjusted average cost was higher for patients with any type of AKI. The incremental cost of having any AKI ranged from $3,700 to $9,100, even after adjustment for comorbidities and other postoperative complications. The adjusted average cost for patients stratified by type of AKI was higher among patients with persistent AKI compared to those with no or rapidly reversed AKI.

CONCLUSIONS

Persistent AKI after vascular surgery is associated with increased complications, mortality, and cost. Strategies to prevent and aggressively treat AKI, specifically persistent AKI, in the perioperative setting are imperative to optimize care for this population.

The effects of timing onset and progression of AKI on the clinical outcomes in AKI patients with sepsis: a prospective multicenter cohort study

BACKGROUND

Limited studies are available concerning on the earlier identification of AKI with sepsis. The aim of the study was to identify the risk factors of AKI early which depended on the timing onset and progression of AKI and investigate the effects of timing onset and progression of AKI on clinical outcomes.

METHODS

Patients who developed sepsis during their first 48-h admission to ICU were included. The primary outcome was major adverse kidney events (MAKE) consisted of all-cause mortality, RRT-dependence, or an inability to recover to 1.5 times of the baseline creatinine value up to 30 days. We determined MAKE and in-hospital mortality by multivariable logistic regression and explored the risk factors of early persistent-AKI. C statistics were used to evaluate model fit.

RESULTS

58.7% sepsis patients developed AKI. According to the timing onset and progression of AKI, Early transient-AKI, early persistent-AKI, late transient-AKI, late persistent-AKI were identified. Clinical outcomes were quite different among subgroups. Early persistent-AKI had 3.0-fold (OR 3.04, 95% CI 1.61 - 4.62) risk of MAKE and 2.6-fold (OR 2.60, 95%CI 1.72 - 3.76) risk of in-hospital mortality increased compared with the late transients-AKI. Older age, underweight, obese, faster heart rate, lower MAP, platelet, hematocrit, pH and energy intake during the first 24 h on ICU admission could well predict the early persistent-AKI in patients with sepsis.

CONCLUSION

Four AKI subphenotypes were identified based on the timing onset and progression of AKI. Early persistent-AKI showed higher risk of major adverse kidney events and in-hospital mortality.

TRIAL REGISTRATION

This study was registered in the Chinese Clinical Trials Registry (www.chictr.org/cn) under registration number ChiCTR-ECH-13003934.

The persistent inflammation, immunosuppression, and catabolism syndrome 10 years later

With the implementation of new intensive care unit (ICU) therapies in the 1970s, multiple organ failure (MOF) emerged as a fulminant inflammatory phenotype leading to early ICU death. Over the ensuing decades, with fundamental advances in care, this syndrome has evolved into a lingering phenotype of chronic critical illness (CCI) leading to indolent late post-hospital discharge death. In 2012, the University of Florida (UF) Sepsis Critical Illness Research Center (SCIRC) coined the term Persistent Inflammation, Immunosuppression, and Catabolism Syndrome (PICS) to provide a mechanistic framework to study CCI in surgical patients. This was followed by a decade of research into PICS-CCI in surgical ICU patients in order to define the epidemiology, dysregulated immunity, and long-term outcomes after sepsis. Other focused studies were performed in trauma ICU patients and emergency department sepsis patients. Early deaths were surprisingly low (4%); 63% experienced rapid recovery. Unfortunately, 33% progressed to CCI, of which 79% had a poor post-discharge disposition and 41% were dead within one year. These patients had biomarker evidence of PICS, and these biomarkers enhanced clinical prediction models for dismal one-year outcomes. Emergency myelopoiesis appears to play a central role in the observed persistent immune dysregulation that characterizes PICS-CCI. Older patients were especially vulnerable. Disturbingly, over half of the older CCI patients were dead within one year and older CCI survivors remained severely disabled. Although CCI is less frequent (20%) after major trauma, PICS appears to be a valid concept. This review will specifically detail the epidemiology of CCI, PICS biomarkers, effect of site of infection, acute kidney injury, effect on older patients, dysfunctional high-density lipoproteins, sarcopenia/cachexia, emergency myelopoiesis, dysregulated erythropoiesis, and potential therapeutic interventions.

A review of UF SCIRC’s research efforts characterizing CCI, PICS biomarkers, effect of site of infection, acute kidney injury, effects on older patients, dysfunctional high-density lipoproteins, sarcopenia/cachexia, emergency myelopoiesis, and dysregulated erythropoiesis.

Clinical courses of acute kidney injury in hospitalized patients: a multistate analysis

Persistence of acute kidney injury (AKI) or insufficient recovery of renal function was associated with reduced long-term survival and life quality. We quantified AKI trajectories and describe transitions through progression and recovery among hospitalized patients. 245,663 encounters from 128,271 patients admitted to UF Health between 2012 and 2019 were retrospectively categorized according to the worst AKI stage experienced within 24-h periods. Multistate models were fit for describing characteristics influencing transitions towards progressed or regressed AKI, discharge, and death. Effects of age, sex, race, admission comorbidities, and prolonged intensive care unit stay (ICU) on transition rates were examined via Cox proportional hazards models. About 20% of encounters had AKI; where 66% of those with AKI had Stage 1 as their worst AKI severity during hospitalization, 18% had Stage 2, and 16% had Stage 3 AKI (12% with kidney replacement therapy (KRT) and 4% without KRT). At 3 days following Stage 1 AKI, 71.1% (70.5-71.6%) were either resolved to No AKI or discharged, while recovery proportion was 38% (37.4-38.6%) and discharge proportion was 7.1% (6.9-7.3%) following AKI Stage 2. At 14 days following Stage 1 AKI, patients with additional frail conditions stay had lower transition proportion towards No AKI or discharge states. Multistate modeling framework is a facilitating mechanism for understanding AKI clinical course and examining characteristics influencing disease process and transition rates.

Clinical Sepsis Phenotypes in Critically Ill Patients

Sepsis, defined as the life-threatening dysregulated host response to an infection leading to organ dysfunction, is considered as one of the leading causes of mortality worldwide, especially in intensive care units (ICU). Moreover, sepsis remains an enigmatic clinical syndrome, with complex pathophysiology incompletely understood and a great heterogeneity both in terms of clinical expression, patient response to currently available therapeutic interventions and outcomes. This heterogeneity proves to be a major obstacle in our quest to deliver improved treatment in septic critical care patients; thus, identification of clinical phenotypes is absolutely necessary. Although this might be seen as an extremely difficult task, nowadays, artificial intelligence and machine learning techniques can be recruited to quantify similarities between individuals within sepsis population and differentiate them into distinct phenotypes regarding not only temperature, hemodynamics or type of organ dysfunction, but also fluid status/responsiveness, trajectories in ICU and outcome. Hopefully, we will eventually manage to determine both the subgroup of septic patients that will benefit from a therapeutic intervention and the correct timing of applying the intervention during the disease process.

Serial Urinary C-C Motif Chemokine Ligand 14 and Risk of Persistent Severe Acute Kidney Injury

To assess the added prognostic value of serial monitoring of urinary C-C motif chemokine ligand 14 (uCCL14) over that of single measurements, which have been shown to be prognostic for development of persistent severe acute kidney injury (AKI) in critically ill patients.

DESIGN

Retrospective observational study.

SETTING

Data derived from two multinational ICU studies (Ruby and Sapphire).

PATIENTS

Critically ill patients with early stage 2-3 AKI.

INTERVENTIONS

None.

MEASUREMENTS AND MAIN RESULTS

We analyzed three consecutive uCCL14 measurements at 12-hour intervals after diagnosis of stage 2-3 AKI by Kidney Disease Improving Global Outcomes criteria. Primary outcome was persistent severe AKI, defined as 72 consecutive hours of stage 3 AKI, death, or receipt of dialysis prior to 72 hours. uCCL14 was measured using the NEPHROCLEAR uCCL14 Test on the Astute 140 Meter (Astute Medical, San Diego, CA). Based on predefined, validated cutoffs, we categorized uCCL14 as: low (≤ 1.3 ng/mL), medium (> 1.3 to ≤ 13 ng/mL), or high (> 13 ng/mL). Seventy-five of 417 patients with three consecutive uCCL14 measurements developed persistent severe AKI. Initial uCCL14 category strongly correlated with primary endpoint and, in most cases (66%), uCCL14 category was unchanged over the first 24 hours. Compared with no change and accounting for baseline category, decrease in category was associated with decreased odds of persistent severe AKI (odds ratio [OR], 0.20; 95% CI, 0.08-0.45; p < 0.001) and an increase in category with increased odds (OR, 4.04; 95% CI, 1.75-9.46; p = 0.001).

CONCLUSIONS

In one-third of patients with moderate to severe AKI uCCL14 risk category altered over three serial measurements and such changes were associated with altered risk for persistent severe AKI. Serial CCL-14 measurement may detect progression or resolution of underlying kidney pathology and help refine AKI prognosis.

Development and Validation of Machine Learning Models for Real-Time Mortality Prediction in Critically Ill Patients With Sepsis-Associated Acute Kidney Injury

Background

Sepsis-associated acute kidney injury (SA-AKI) is common in critically ill patients, which is associated with significantly increased mortality. Existing mortality prediction tools showed insufficient predictive power or failed to reflect patients' dynamic clinical evolution. Therefore, the study aimed to develop and validate machine learning-based models for real-time mortality prediction in critically ill patients with SA-AKI.

Methods

The multi-center retrospective study included patients from two distinct databases. A total of 12,132 SA-AKI patients from the Medical Information Mart for Intensive Care IV (MIMIC-IV) were randomly allocated to the training, validation, and internal test sets. An additional 3,741 patients from the eICU Collaborative Research Database (eICU-CRD) served as an external test set. For every 12 h during the ICU stays, the state-of-the-art eXtreme Gradient Boosting (XGBoost) algorithm was used to predict the risk of in-hospital death in the following 48, 72, and 120 h and in the first 28 days after ICU admission. Area under the receiver operating characteristic curves (AUCs) were calculated to evaluate the models' performance.

Results

The XGBoost models, based on routine clinical variables updated every 12 h, showed better performance in mortality prediction than the SOFA score and SAPS-II. The AUCs of the XGBoost models for mortality over different time periods ranged from 0.848 to 0.804 in the internal test set and from 0.818 to 0.748 in the external test set. The shapley additive explanation method provided interpretability for the XGBoost models, which improved the understanding of the association between the predictor variables and future mortality.

Conclusions

The interpretable machine learning XGBoost models showed promising performance in real-time mortality prediction in critically ill patients with SA-AKI, which are useful tools for early identification of high-risk patients and timely clinical interventions.

Early recovery status and outcomes after sepsis-associated acute kidney injury in critically ill patients

OBJECTIVES

Acute kidney injury (AKI) is one of the common complications in critically ill septic patients, which is associated with increased risks of death, cardiovascular events, and chronic renal dysfunction. The duration of AKI and the renal function recovery status after AKI onset can affect the patient prognosis. Nevertheless, it remains controversial whether early recovery status after AKI is closely related to the prognosis in patients with sepsis-associated AKI (SA-AKI). In addition, early prediction of renal function recovery after AKI is beneficial to individualized treatment decision-making and prevention of severe complications, thus improving the prognosis. At present, there is limited clinical information on how to identify SA-AKI patients at high risk of unrecovered renal function at an early stage. The study aims to investigate the association between early recovery status after SA-AKI, identify risk factors for unrecovered renal function, and to improve patients' quality of life.

METHODS

We retrospectively analyzed clinical data of septic patients who were admitted to the intensive care unit (ICU) and developed AKI within the first 48 hours after ICU admission in the Second Xiangya Hospital and the Third Xiangya Hospital of Central South University from January 2015 to March 2017. Sepsis was defined based on the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). AKI was diagnosed and staged according to the 2012 Kidney Disease: Improving Global Outcomes (KDIGO) guideline. SA-AKI patients were assigned into 3 groups including a complete recovery group, a partial recovery group, and an unrecovered group based on recovery status at Day 7 after the diagnosis of AKI. Patients' baseline characteristics were collected, including demographics, comorbidities, clinical and laboratory examination information at ICU admission, and treatment within the first 24 hours. The primary outcome of the study was the composite of death and chronic dialysis at 90 days, and secondary outcomes included length of stay in the ICU, length of stay in the hospital, and persistent renal dysfunction. Multivariate regression analysis was performed to evaluate the prognostic value of early recovery status after AKI and to determine the risk factors for unrecovered renal function after AKI. Sensitivity analysis was conducted in patients who still stayed in hospital on Day 7 after AKI diagnosis, patients without premorbid chronic kidney disease, and patients with AKI Stage 2 to 3.

RESULTS

A total of 553 SA-AKI patients were enrolled, of whom 251 (45.4%), 73 (13.2%), and 229 (41.4%) were categorized as the complete recovery group, the partial recovery group, and the unrecovered group, respectively. Compared with the complete or partial recovery group, the unrecovered group had a higher incidence of 90-day mortality (unrecovered vs partial recovery or complete recovery: 64.2% vs 26.0% or 22.7%; P<0.001) and 90-day composite outcome (unrecovered vs partial recovery or complete recovery: 65.1% vs 27.4% or 22.7%; P<0.001). The unrecovered group also had a shorter length of stay in the hospital and a larger proportion of progression into persistent renal dysfunction than the other 2 groups. After adjustment for potential confounders, patients in the unrecovered group were at an increased risk of 90-day mortality (HR=3.50, 95% CI 2.47 to 4.96, P<0.001) and 90-day composite outcome (OR=5.55, 95% CI 3.43 to 8.98, P<0.001) when compared with patients in the complete recovery group, but patients in the partial recovery group had no significant difference (P>0.05). Male sex, congestive heart failure, pneumonia, respiratory rate >20 beats per minute, anemia, hyperbilirubinemia, need for mechanical ventilation, and AKI Stage 3 were identified as independent risk factors for unrecovered renal function after AKI. The sensitivity analysis further supported that unrecovered renal function after AKI remained an independent predictor for 90-day mortality and composite outcome in the subgroups.

CONCLUSIONS

The early recovery status after AKI is closely associated with poor prognosis in critically ill patients with SA-AKI. Unrecovered renal function within the first 7 days after AKI diagnosis is an independent predictor for 90-day mortality and composite outcome. Male sex, congestive heart failure, pneumonia, tachypnea, anemia, hyperbilirubinemia, respiratory failure, and severe AKI are risk factors for unrecovered renal function after AKI. Therefore, timely assessment for the renal function in the early phase after AKI diagnosis is essential for SA-AKI patients. Furthermore, patients with unrecovered renal function after AKI need additional management in the hospital, including rigorous monitoring, avoidance of nephrotoxin, and continuous assessment for the renal function, and after discharge, including more frequent follow-up, regular outpatient consultation, and prevention of long-term adverse events.

Performance of a Standardized Clinical Assay for Urinary C-C Motif Chemokine Ligand 14 (CCL14) for Persistent Severe Acute Kidney Injury

Background

Clinical use of biomarkers requires the development of standardized assays and establishment of cutoffs. Urinary C-C motif chemokine ligand 14 (CCL14) has been validated to predict persistent severe AKI in critically ill patients with established AKI. We now report on the performance of standardized cutoffs using a clinical assay.

Methods

A second aim of the multicenter RUBY Study was to establish two cutoffs for the prediction of persistent severe AKI (defined as KDIGO stage 3 AKI for at least 72 consecutive hours). Patients who received renal replacement therapy (RRT) or died before achieving 72 hours in stage 3 AKI were also considered to have reached the end point.

Results

A cutoff value for urinary CCL14 of 1.3 ng/ml was determined to achieve high sensitivity (91%; 95% CI, 84% to 96%), and 13 ng/ml achieved high specificity (93%; 95% CI, 89% to 96%). The cutoff of 1.3 ng/ml identifies the majority (91%) of patients who developed persistent severe AKI with a negative predictive value of 92%. The cutoff at 13 ng/ml had a positive predictive value of 72% (with a negative predictive value of 75%). In multivariable adjusted analyses, a CCL14 concentration between 1.3 and 13 ng/ml had an adjusted odds ratio (aOR) of 3.82 (95% CI, 1.73 to 9.12; P=0.001) for the development of persistent severe AKI compared with those with a CCL14 ≤1.3 ng/ml, whereas a CCL14 >13 ng/ml had an aOR of 10.4 (95% CI, 3.89 to 29.9; P<0.001).

Conclusions

Using a clinical assay, these standardized cutoffs (1.3 and 13 ng/ml) allow for the identification of patients at high risk for the development of persistent severe AKI. These results have immediate utility in helping to guide AKI patient care and may facilitate future clinical trials.Clinical Trial registry name and registration number: Identification and Validation of Biomarkers of Acute Kidney Injury Recovery, NCT01868724.

Red Cell Distribution Width at Admission Predicts the Frequency of Acute Kidney Injury and 28-Day Mortality in Patients With Acute Respiratory Distress Syndrome

OBJECTIVES

To determine the association of red cell distribution width (RDW) at admission with frequency of acute kidney injury (AKI) and 28-day mortality in acute respiratory distress syndrome (ARDS) patients.

METHODS

Two hundred fifty-eight ARDS patients were investigated in retrospective and prospective studies. The primary outcome was frequency of AKI. The secondary outcome was 28-day mortality.

RESULTS

The retrospective study included 193 ARDS patients, of which 67 (34.7%) were confirmed AKI and 76 (39.4%) died within 28 days. The RDW level in the AKI group was significantly higher than in the non-AKI group ([15.15 ± 2.59]% vs. [13.95 ± 1.89]%). Increased RDW was a significant predictor of frequency of AKI (odds ratio: 1.247, 95% confidence interval [CI]: 1.044, 1.489). The area under the receiver operating characteristic curve of RDW for predicting AKI was 0.687 (95%CI: 0.610, 0.764) and the cut-off value was 14.45 (sensitivity, 56.7%; specificity, 72.8%). In addition, the proportion of patients with RDW ≥ 14.45% in the non-survival group was notably higher compared with the survival group (48.7% vs. 29.1%). Furthermore, cox regression analysis revealed that RDW ≥ 14.45% was associated with 28-day mortality (hazard ratio: 1.817, 95%CI: 1.046, 3.158), while Kaplan-Meier analysis showed patients with RDW ≥ 14.45% had a significantly lower survival rate than those with RDW < 14.45%. The prospective study, on the other hand, included 65 ARDS patients, with frequency of AKI and 28-day mortality in the RDW ≥ 14.45% group significantly higher than in RDW < 14.45%.

CONCLUSION

RDW was a significant, independent predictor for frequency of AKI and 28-day mortality in ARDS patients.

Association of persistent acute kidney injury and renal recovery with mortality in hospitalised patients

Objectives

Acute kidney injury (AKI) affects up to one-quarter of hospitalised patients and 60% of patients in the intensive care unit (ICU). We aim to understand the baseline characteristics of patients who will develop distinct AKI trajectories, determine the impact of persistent AKI and renal non-recovery on clinical outcomes, resource use, and assess the relative importance of AKI severity, duration and recovery on survival.

Methods

In this retrospective, longitudinal cohort study, 156 699 patients admitted to a quaternary care hospital between January 2012 and August 2019 were staged and classified (no AKI, rapidly reversed AKI, persistent AKI with and without renal recovery). Clinical outcomes, resource use and short-term and long-term survival adjusting for AKI severity were compared among AKI trajectories in all cohort and subcohorts with and without ICU admission.

Results

Fifty-eight per cent (31 500/54 212) had AKI that rapidly reversed within 48 hours; among patients with persistent AKI, two-thirds (14 122/22 712) did not have renal recovery by discharge. One-year mortality was significantly higher among patients with persistent AKI (35%, 7856/22 712) than patients with rapidly reversed AKI (15%, 4714/31 500) and no AKI (7%, 22 117/301 466). Persistent AKI without renal recovery was associated with approximately fivefold increased hazard rates compared with no AKI in all cohort and ICU and non-ICU subcohorts, independent of AKI severity.

Discussion

Among hospitalised, ICU and non-ICU patients, persistent AKI and the absence of renal recovery are associated with reduced long-term survival, independent of AKI severity.

Conclusions

It is essential to identify patients at risk of developing persistent AKI and no renal recovery to guide treatment-related decisions.

Machine learning for early discrimination between transient and persistent acute kidney injury in critically ill patients with sepsis

Acute kidney injury (AKI) is commonly present in critically ill patients with sepsis. Early prediction of short-term reversibility of AKI is beneficial to risk stratification and clinical treatment decision. The study sought to use machine learning methods to discriminate between transient and persistent sepsis-associated AKI. Septic patients who developed AKI within the first 48 h after ICU admission were identified from the Medical Information Mart for Intensive Care III database. AKI was classified as transient or persistent according to the Acute Disease Quality Initiative workgroup consensus. Five prediction models using logistic regression, random forest, support vector machine, artificial neural network and extreme gradient boosting were constructed, and their performance was evaluated by out-of-sample testing. A simplified risk prediction model was also derived based on logistic regression and features selected by machine learning algorithms. A total of 5984 septic patients with AKI were included, 3805 (63.6%) of whom developed persistent AKI. The artificial neural network and logistic regression models achieved the highest area under the receiver operating characteristic curve (AUC) among the five machine learning models (0.76, 95% confidence interval [CI] 0.74-0.78). The simplified 14-variable model showed adequate discrimination, with the AUC being 0.76 (95% CI 0.73-0.78). At the optimal cutoff of 0.63, the sensitivity and specificity of the simplified model were 63% and 76% respectively. In conclusion, a machine learning-based simplified prediction model including routine clinical variables could be used to differentiate between transient and persistent AKI in critically ill septic patients. An easy-to-use risk calculator can promote its widespread application in daily clinical practice.

Cardiovascular and Renal Disease in Chronic Critical Illness

With advances in critical care, patients who would have succumbed in previous eras now survive through hospital discharge. Many survivors suffer from chronic organ dysfunction and induced frailty, representing an emerging chronic critical illness (CCI) phenotype. Persistent and worsening cardiovascular and renal disease are primary drivers of the CCI phenotype and have pathophysiologic synergy, potentiating one another and generating a downward spiral of worsening disease and clinical outcomes manifest as cardio-renal syndromes. In addition to pharmacologic therapies (e.g., diuretics, beta adrenergic receptor blockers, angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, and blood pressure control), special consideration should be given to behavioral modifications that avoid the pitfalls of polypharmacy and suboptimal renal and hepatic dosing, to which CCI patients may be particularly vulnerable. Smoking cessation, dietary modifications (e.g., early high-protein nutrition and late low-sodium diets), and increased physical activity are advised. Select patients benefit from cardiac re-synchronization therapy or renal replacement therapy. Coordinated, patient-centered care bundles may improve compliance with standards of care and patient outcomes. Given the complex, heterogeneous nature of cardiovascular and renal disease in CCI and the dismal long-term outcomes, further research is needed to clarify pathophysiologic mechanisms of cardio-renal syndromes in CCI and develop targeted therapies.