Data-driven prediction of continuous renal replacement therapy survival.

Continuous renal replacement therapy (CRRT) is a form of dialysis prescribed to severely ill patients who cannot tolerate regular hemodialysis. However, as the patients are typically very ill to begin with, there is always uncertainty whether they will survive during or after CRRT treatment. Because of outcome uncertainty, a large percentage of patients treated with CRRT do not survive, utilizing scarce resources and raising false hope in patients and their families. To address these issues, we present a machine learning-based algorithm to predict short-term survival in patients being initiated on CRRT. We use information extracted from electronic health records from patients who were placed on CRRT at multiple institutions to train a model that predicts CRRT survival outcome; on a held-out test set, the model achieves an area under the receiver operating curve of 0.848 (CI = 0.822-0.870). Feature importance, error, and subgroup analyses provide insight into bias and relevant features for model prediction. Overall, we demonstrate the potential for predictive machine learning models to assist clinicians in alleviating the uncertainty of CRRT patient survival outcomes, with opportunities for future improvement through further data collection and advanced modeling.

Long-Term Mortality and Health-Related Quality of Life After Continuous Versus Intermittent Renal Replacement Therapy in ICU Survivors: A Secondary Analysis of the Quality of Life After ICU Study.

Purpose: We assessed long-term outcomes in intensive care unit (ICU) survivors with acute kidney injury (AKI) submitted to intermittent or continuous renal replacement therapy (RRT) for comparisons between groups. Methods: The multicenter prospective cohort study included 195 adult ICU survivors with an ICU stay >72 h in 10 ICUs that had at least one episode of AKI treated with intermittent RRT (IRRT) or continuous RRT (CRRT) during ICU stay. The main outcomes were mortality and health-related quality of life (HRQoL). Hospital readmissions and physical dependence were also assessed. Results: Regarding RRT, 83 (42.6%) patients received IRRT and 112 (57.4%) received CRRT. Despite the similarity regarding sociodemographic characteristics, pre-ICU state of health and type of admission between groups, the risk of death (23.5% vs 42.7%; P < .001), the prevalence of sepsis (60.7%) and acute respiratory distress syndrome (17%) were higher at ICU admission among CRRT patients. The severity of critical illness was higher among CRRT patients, regarding the need for mechanical ventilation (75.0% vs 50.6%, P = .002) and vasopressors (91.1% vs 63.9%, P < .001). One year after ICU discharge, 67 of 195 ICU survivors died (34.4%) and, after adjustment for confounders, there were no significant differences in mortality when comparing IRRT and CRTT patients (34.9% vs 33.9%; P = .590), on HRQoL in both physical (41.9% vs 42.2%; P = .926) and mental dimensions (57.6% vs 56.6%; P = .340), and on the number of hospital readmissions and physical dependence. Conclusions: Our study suggests that among ICU survivors RRT modality (IRRT vs CRRT) in the ICU does not impact long-term outcomes after ICU discharge.

Association of Estimated GFR Calculated Using Race-Free Equations With Kidney Failure and Mortality by Black vs Non-Black Race.

Importance: At a given estimated glomerular filtration rate (eGFR), individuals who are Black have higher rates of mortality and kidney failure with replacement therapy (KFRT) compared with those who are non-Black. Whether the recently adopted eGFR equations without race preserve racial differences in risk of mortality and KFRT at a given eGFR is unknown. Objective: To assess whether eGFR equations with and without race and cystatin C document racial differences in risk of KFRT and mortality in populations including Black and non-Black participants. Design, Setting, and Participants: Retrospective individual-level data analysis of 62 011 participants from 5 general population and 3 chronic kidney disease (CKD) US-based cohorts with serum creatinine, cystatin C, and follow-up for KFRT and mortality from 1988 to 2018. Exposures: Chronic Kidney Disease Epidemiology Collaboration equation with serum creatinine (eGFRcr with and without race), cystatin C (eGFRcys without race), or both markers (eGFRcr-cys without race). Main Outcomes and Measures: The prevalence of decreased eGFR at baseline and hazard ratios of KFRT and mortality in Black vs non-Black participants were calculated, adjusted for age and sex. Analyses were performed within each cohort and with random-effect meta-analyses of the models. Results: Among 62 011 participants (20 773 Black and 41 238 non-Black; mean age, 63 years; 53% women), the prevalence ratio (95% CI; percent prevalences) of eGFR less than 60 mL/min/1.73 m2 comparing Black with non-Black participants was 0.98 (95% CI, 0.93-1.03; 11% vs 12%) for eGFRcr with race, 0.95 (95% CI, 0.91-0.98; 17% vs 18%) for eGFRcys, and 1.2 (95% CI, 1.2-1.3; 13% vs 11%) for eGFRcr-cys but was 1.8 (95% CI, 1.7-1.8; 15% vs 9%) for eGFRcr without race. During a mean follow-up of 13 years, 8% and 4% of Black and non-Black participants experienced KFRT and 34% and 39% died, respectively. Decreased eGFR was associated with significantly greater risk of both outcomes for all equations. At an eGFR of 60 mL/min/1.73 m2, the hazard ratios for KFRT comparing Black with non-Black participants were 2.8 (95% CI, 1.6-4.9) for eGFRcr with race, 3.0 (95% CI, 1.5-5.8) for eGFRcys, and 2.8 (95% CI, 1.4-5.4) for eGFRcr-cys vs 1.3 (95% CI, 0.8-2.1) for eGFRcr without race. The 5-year absolute risk differences for KFRT comparing Black with non-Black participants were 1.4% (95% CI, 0.2%-2.6%) for eGFRcr with race, 1.1% (95% CI, 0.2%-1.9%) for eGFRcys, and 1.3% (95% CI, 0%-2.6%) for eGFRcr-cys vs 0.37% (95% CI, -0.32% to 1.05%) for eGFRcr without race. Similar patterns were observed for mortality. Conclusions and Relevance: In this retrospective analysis of 8 US cohorts including Black and non-Black individuals, the eGFR equation without race that included creatinine and cystatin C, but not the eGFR equation without race that included creatinine without cystatin C, demonstrated racial differences in the risk of KFRT and mortality throughout the range of eGFR. The eGFRcr-cys equation may be preferable to the eGFRcr equation without race for assessing racial differences in the risk of KFRT and mortality associated with low eGFR.

Intermittent Convective Therapies in Patients with Acute Kidney Injury: A Systematic Review with Meta-Analysis.

INTRODUCTION: In critically ill patients requiring intermittent renal replacement therapy (RRT), the benefits of convective versus diffusive clearance remain uncertain. We conducted a systematic review and meta-analysis to determine the safety, clinical efficacy, and clearance efficiency of hemofiltration (HF) and hemodiafiltration (HDF) compared to hemodialysis (HD) in patients with acute kidney injury (AKI) receiving intermittent RRT. METHOD: We searched Medline, Embase, Cochrane Library, and PROSPERO. We included clinical trials and observational studies that reported the use of intermittent HF or HDF in adult patients with AKI. The following outcomes were included: mortality, renal recovery, clearance efficacy, intradialytic hemodynamic stability, circuit loss, and inflammation modulation. RESULTS: A total of 3,169 studies were retrieved and screened. Four randomized controlled trials and 4 observational studies were included (n: 615 patients). Compared with conventional HD, intermittent convective therapies had no effect on in-hospital mortality (relative risk, 1.23; 95% confidence interval (CI), 0.76-1.99), renal recovery at 30 days (RR, 0.98; 95% CI, 0.82-1.16), time-to-renal recovery (mean difference [MD], 0.77; 95% CI, -6.56 to 8.10), and number of dialysis sessions until renal recovery (MD, -1.34; 95% CI, -3.39 to 0.72). The overall quality of included studies was low, and dialysis parameters were suboptimal for all included studies. CONCLUSION: This meta-analysis suggests that there is no significant difference in short-term mortality and renal recovery in patients with severe AKI when treated with intermittent HF or HDF compared to conventional HD. This systematic review emphasizes the need for further trials evaluating optimal convective parameters in AKI patients treated with intermittent dialysis.

Association of Acute Kidney Injury Receiving Kidney Replacement Therapy With Prognosis of Critically Ill Patients With and Without Cancer: A Retrospective Study.

OBJECTIVES: To assess outcomes of cancer patients receiving kidney replacement therapy due to acute kidney injury in ICUs and compare these with other patient groups receiving kidney replacement therapy in ICUs. DESIGN: Retrospective registry analysis. SETTING: Prospectively collected database of 296,424 ICU patients. PATIENTS: Patients with and without solid cancer with acute kidney injury necessitating kidney replacement therapy were identified and compared with those without acute kidney injury necessitating kidney replacement therapy. INTERVENTIONS: Descriptive statistics were used to ascertain prevalence of acute kidney injury necessitating kidney replacement therapy and solid cancer in ICU patients. Association of acute kidney injury necessitating kidney replacement therapy and cancer with prognosis was assessed using logistic regression analysis. To compare the attributable mortality of acute kidney injury necessitating kidney replacement therapy, 20,154 noncancer patients and 2,411 cancer patients without acute kidney injury necessitating kidney replacement therapy were matched with 12,827 noncancer patients and 1,079 cancer patients with acute kidney injury necessitating kidney replacement therapy. MEASUREMENTS AND MAIN RESULTS: Thirty-five thousand three hundred fifty-six ICU patients (11.9%) had solid cancer. Acute kidney injury necessitating kidney replacement therapy was present in 1,408 (4.0%) cancer patients and 13,637 (5.2%) noncancer patients. Crude ICU and hospital mortality was higher in the cancer group (646 [45.9%] vs 4,674 [34.3%], p < 0.001, and 787 [55.9%] vs 5,935 [43.5%], p < 0.001). In multivariable logistic regression analyses, odds ratio (95% CI) for hospital mortality was 1.73 (1.62-1.85) for cancer compared with no cancer 3.57 (3.32-3.83) for acute kidney injury necessitating kidney replacement therapy and 1.07 (0.86-1.33) for their interaction. In the matched subcohort, attributable hospital mortality of acute kidney injury necessitating kidney replacement therapy was 56.7% in noncancer patients and 48.0% in cancer patients. CONCLUSIONS: Occurrence rate of acute kidney injury necessitating kidney replacement therapy and prognosis in ICU patients with solid cancer are comparable with other ICU patient groups. In cancer, acute kidney injury necessitating kidney replacement therapy is associated with higher crude hospital mortality. However, the specific attributable mortality conveyed by acute kidney injury necessitating kidney replacement therapy is actually lower in cancer patients than in noncancer patients. Diagnosis of cancer per se does not justify withholding kidney replacement therapy.

Time to Initiation of Renal Replacement Therapy Among Critically Ill Patients With Acute Kidney Injury: A Current Systematic Review and Meta-Analysis.

OBJECTIVES: The optimal time to initiate renal replacement therapy in critically ill patients with acute kidney injury is controversial. We investigated the effect of such earlier versus later initiation of renal replacement therapy on the primary outcome of 28-day mortality and other patient-centered secondary outcomes. DESIGN: We searched MEDLINE (via PubMed), EMBASE, and Cochrane databases to July 17, 2020, and included randomized controlled trials comparing earlier versus later renal replacement therapy. SETTING: Multiple centers involved in eight trials. PATIENTS: Total of 4,588 trial participants. INTERVENTION: Two independents investigators screened and extracted data using a predefined form. We selected randomized controlled trials in critically ill adult patients with acute kidney injury and compared of earlier versus later initiation of renal replacement therapy regardless of modality. MEASUREMENTS AND MAIN RESULTS: Overall, 28-day mortality was similar between earlier and later renal replacement therapy initiation (38.43% vs 38.06%, respectively; risk ratio, 1.01; [95% CI, 0.94-1.09]; I2 = 0%). Earlier renal replacement therapy, however, shortened hospital length of stay (mean difference, -2.14 d; [95% CI, -4.13 to -0.14]) and ICU length of stay (mean difference, -1.18 d; [95% CI, -1.95 to -0.42]). In contrast, later renal replacement therapy decreased the use of renal replacement therapy (relative risk, 0.69; [95% CI, 0.58-0.82]) and lowered the risk of catheter-related blood stream infection (risk ratio, 0.50, [95% CI, 0.29-0.86). Among survivors, renal replacement therapy dependence at day 28 was similar between earlier and later renal replacement therapy initiation (risk ratio, 0.98; [95% CI, 0.66-1.40]). CONCLUSIONS: Earlier or later initiation of renal replacement therapy did not affect mortality. However, earlier renal replacement therapy was associated with significantly shorter ICU and hospital length of stay, whereas later renal replacement therapy was associated with decreased use of renal replacement therapy and decreased risk of catheter-related blood stream infection. These findings can be used to guide the management of critically ill patients with acute kidney injury.

Effect of comorbidities on survival in patients >80 years of age at onset of renal replacement therapy: data from the ERA-EDTA Registry.

BACKGROUND: The number of elderly patients on renal replacement therapy (RRT) is increasing. The survival and quality of life of these patients may be lower if they have multiple comorbidities at the onset of RRT. The aim of this study was to explore whether the effect of comorbidities on survival is similar in elderly RRT patients compared with younger ones. METHODS: Included were 9333 patients ≥80 years of age and 48 352 patients 20-79 years of age starting RRT between 2010 and 2015 from 15 national or regional registries submitting data to the European Renal Association-European Dialysis and Transplantation Association Registry. Patients were followed until death or the end of 2016. Survival was assessed by Kaplan-Meier curves and the relative risk of death associated with comorbidities was assessed by Cox regression analysis. RESULTS: Patients ≥80 years of age had a greater comorbidity burden than younger patients. However, relative risks of death associated with all studied comorbidities (diabetes, ischaemic heart disease, chronic heart failure, cerebrovascular disease, peripheral vascular disease and malignancy) were significantly lower in elderly patients compared with younger patients. Also, the increase in absolute mortality rates associated with an increasing number of comorbidities was smaller in elderly patients. CONCLUSIONS: Comorbidities are common in elderly patients who enter RRT, but the risk of death associated with comorbidities is less than in younger patients. This should be taken into account when assessing the prognosis of elderly RRT patients.

Acute Kidney Injury, Fluid Overload, and Renal Replacement Therapy Differ by Underlying Diagnosis in Neonatal Extracorporeal Support and Impact Mortality Disparately.

INTRODUCTION: We aimed to characterize acute kidney injury (AKI), fluid overload (FO), and renal replacement therapy (RRT) utilization by diagnostic categories and examine associations between these complications and mortality by category. METHODS: To test our hypotheses, we conducted a retrospective multicenter, cohort study including 446 neonates (categories: 209 with cardiac disease, 114 with congenital diaphragmatic hernia [CDH], 123 with respiratory disease) requiring extracorporeal membrane oxygenation (ECMO) between January 1, 2007, and December 31, 2011. RESULTS: AKI, FO, and RRT each varied by diagnostic category. AKI and RRT receipt were most common in those neonates with cardiac disease. Subjects with CDH had highest peak %FO (51% vs. 28% cardiac vs. 32% respiratory; p < 0.01). Hospital survival was 55% and varied by diagnostic category (45% cardiac vs. 48% CDH vs. 79% respiratory; p < 0.001). A significant interaction suggested risk of mortality differed by diagnostic category in the presence or absence of AKI. In its absence, diagnosis of CDH (vs. respiratory disease) (OR 3.04, 95% CL 1.14-8.11) independently predicted mortality. In all categories, peak %FO (OR 1.20, 95% CL 1.11-1.30) and RRT receipt (OR 2.12, 95% CL 1.20-3.73) were independently associated with mortality. DISCUSSION/CONCLUSIONS: Physiologically distinct ECMO diagnoses warrant individualized treatment strategies given variable incidence and effects of AKI, FO, and RRT by category on mortality.

The incidence and effect of resternotomy following cardiac surgery on morbidity and mortality: a 1-year national audit on behalf of the Association of Cardiothoracic Anaesthesia and Critical Care.

Over 30,000 adult cardiac operations are carried out in the UK annually. A small number of these patients need to return to theatre in the first few days after the initial surgery, but the exact proportion is unknown. The majority of these resternotomies are for bleeding or cardiac tamponade. The Association of Cardiothoracic Anaesthesia and Critical Care carried out a 1-year national audit of resternotomy in 2018. Twenty-three of the 35 centres that were eligible participated. The overall resternotomy rate (95%CI) within the period of admission for the initial operation in these centres was 3.6% (3.37-3.85). The rate varied between centres from 0.69% to 7.6%. Of the 849 patients who required resternotomy, 127 subsequently died, giving a mortality rate (95%CI) of 15.0% (12.7-17.5). In patients who underwent resternotomy, the median (IQR [range]) length of stay on ICU was 5 (2-10 [0-335]) days, and time to tracheal extubation was 20 (12-48 [0-2880]) hours. A total of 89.3% of patients who underwent resternotomy were transfused red cells, with a median (IQR [range]) of 4 (2-7 [1-1144]) units of red blood cells. The rate (95%CI) of needing renal replacement therapy was 23.4% (20.6-26.5). This UK-wide audit has demonstrated that resternotomy after cardiac surgery is associated with prolonged intensive care stay, high rates of blood transfusion, renal replacement therapy and very high mortality. Further research into this area is required to try to improve patient care and outcomes in patients who require resternotomy in the first 24 h after cardiac surgery.