Explore the vast range of titles available.

The incorporation of machine learning methods into proteomics workflows improves the identification of disease-relevant biomarkers and biological pathways. However, machine learning models, such as deep neural networks, typically suffer from lack of interpretability. Here, we present a deep learning approach to combine biological pathway analysis and biomarker identification to increase the interpretability of proteomics experiments. Our approach integrates a priori knowledge of the relationships between proteins and biological pathways and biological processes into sparse neural networks to create biologically informed neural networks. We employ these networks to differentiate between clinical subphenotypes of septic acute kidney injury and COVID-19, as well as acute respiratory distress syndrome of different aetiologies. To gain biological insight into the complex syndromes, we utilize feature attribution-methods to introspect the networks for the identification of proteins and pathways important for distinguishing between subtypes. The algorithms are implemented in a freely available open source Python-package ( https://github.com/InfectionMedicineProteomics/BINN ). Deep neural networks hold significant promise in capturing the complexity of biological systems. However, they suffer from a lack of interpretability. Here, authors present a generalizable method for developing, interpreting, and visualizing biologically informed neural networks for proteomics data.

Acute kidney injury (AKI) is a frequently encountered syndrome especially among the critically ill. Current diagnosis of AKI is based on acute deterioration of kidney function, indicated by an increase in creatinine and/or reduced urine output. However, this syndromic definition encompasses a wide variety of distinct clinical features, varying pathophysiology, etiology and risk factors, and finally very different short- and long-term outcomes. Lumping all AKI together may conceal unique pathophysiologic processes specific to certain AKI populations, and discovering these AKI subphenotypes might help to develop targeted therapies tackling unique pathophysiological processes. In this review, we discuss the concept of AKI subphenotypes, current knowledge regarding both clinical and biomarker-driven subphenotypes, interplay with AKI subphenotypes and other ICU syndromes, and potential future and clinical implications.

Background The pathophysiology of septic acute kidney injury is inadequately understood. Recently, subphenotypes for sepsis and AKI have been derived. The objective of this study was to assess whether a combination of comorbidities, baseline clinical data, and biomarkers could classify meaningful subphenotypes in septic AKI with different outcomes. Methods We performed a post hoc analysis of the prospective Finnish Acute Kidney Injury (FINNAKI) study cohort. We included patients admitted with sepsis and acute kidney injury during the first 48 h from admission to intensive care (according to Kidney Disease Improving Global Outcome criteria). Primary outcomes were 90-day mortality and renal recovery on day 5. We performed latent class analysis using 30 variables obtained on admission to classify subphenotypes. Second, we used logistic regression to assess the association of derived subphenotypes with 90-day mortality and renal recovery on day 5. Results In total, 301 patients with septic acute kidney injury were included. Based on the latent class analysis, a two-class model was chosen. Subphenotype 1 was assigned to 133 patients (44%) and subphenotype 2 to 168 patients (56%). Increased levels of inflammatory and endothelial injury markers characterized subphenotype 2. At 90 days, 29% of patients in subphenotype 1 and 41% of patients in subphenotype 2 had died. Subphenotype 2 was associated with a lower probability of short-term renal recovery and increased 90-day mortality. Conclusions In this post hoc analysis, we identified two subphenotypes of septic acute kidney injury with different clinical outcomes. Future studies are warranted to validate the suggested subphenotypes of septic acute kidney injury.

Background Acute kidney injury (AKI) is a frequent and clinically relevant problem in critically ill patients. Various randomized controlled trials (RCT) have attempted to assess potentially beneficial treatments for AKI. Different approaches to applying the Kidney Disease Improving Global Outcomes (KDIGO) criteria for AKI make a comparison of studies difficult. The objective of this study was to assess how different approaches may impact estimates of AKI incidence and whether the association between AKI and 90-day mortality varied by the approach used. Methods Consecutive acutely admitted adult intensive care patients were included in a prospective observational study. AKI was determined following the KDIGO criteria during the first 7 days of ICU admission. In this post hoc analysis, we assessed whether AKI incidence differed when applying the KDIGO criteria in 30 different possible methods, varying in (A) serum creatinine (sCr), (B) urine output (UO), and (C) the method of combining these two into an outcome, e.g., severe AKI. We assessed point estimates and 95% confidence intervals for each incidence. Univariable regression was used to assess the associations between AKI and 90-day mortality. Results A total of 1010 patients were included. Baseline creatinine was available in 449 (44%) patients. The incidence of any AKI ranged from 28% (95%CI 25–31%) to 75% (95%CI 72–77%) depending on the approach used. Methods to estimate missing baseline sCr caused a variation in AKI incidence up to 15%. Different methods of handling UO caused a variation of up to 35%. At 90 days, 263 patients (26%) had died, and all 30 variations were associated with 90-day mortality. Conclusions In this cohort of critically ill patients, AKI incidence varied from 28 to 75%, depending on the method used of applying the KDIGO criteria. A tighter adherence to KDIGO definitions is warranted to decrease the heterogeneity of AKI and increase the comparability of future studies.

Background Fluid bolus therapy is a common intervention to improve urine output. Data concerning the effect of a fluid bolus on oliguria originate mainly from observational studies and remain controversial regarding the actual benefit of such therapy. We compared the effect of a follow-up approach without fluid bolus to a 500 mL fluid bolus on urine output in hemodynamically stable critically ill patients with oliguria at least for 2 h (urine output < 0.5 mL/kg/h) in randomized setting. Methods We randomized 130 patients in 1:1 fashion to receive either (1) non-interventional follow-up (FU) for 2 h or (2) 500 mL crystalloid fluid bolus (FB) administered over 30 min. The primary outcome was the proportion of patients who doubled their urine output, defined as 2-h urine output post-randomization divided by urine output 2 h pre-randomization. The outcomes were adjusted for the stratification variables (presence of sepsis or AKI) using two-tailed regression. Obtained odds ratios were converted to risk ratios (RR) with 95% confidence intervals (CI). The between-group difference in the continuous variables was compared using mean or median regression and expressed with 95% CIs. Results Altogether 10 (15.9%) of 63 patients in the FU group and 22 (32.8%) of 67 patients in FB group doubled their urine output during the 2-h period, RR (95% CI) 0.49 (0.23–0.71), P  = 0.026. Median [IQR] change in individual urine output 2 h post-randomization compared to 2 h pre-randomization was − 7 [− 19 to 17] mL in the FU group and 19[0–53] mL in the FB group, median difference (95% CI) − 23 (− 36 to − 10) mL, P  = 0.001. Median [IQR] duration of oliguria in the FU group was 4 [2–8] h and in the FB group 2 [0–6] h, median difference (95%CI) 2 (0–4) h, P  = 0.038. Median [IQR] cumulative fluid balance on study day was lower in the FU group compared to FB group, 678 [518–1029] mL versus 1071 [822–1505] mL, respectively, median difference (95%CI) − 387 (− 635 to − 213) mL, P  < 0.001. Conclusions Follow-up approach to oliguria compared to administering a fluid bolus of 500 mL crystalloid in oliguric patients improved urine output less frequently but lead to lower cumulative fluid balance. Trial registration clinical.trials.gov, NCT02860572. Registered 9 August 2016. Graphical Abstract

Fluid therapy is a fundamental part of supportive therapy in critical care. However, it is also a suspected risk for endothelial glycocalyx degradation which is associated with poor clinical outcomes. This secondary analysis of RESPONSE randomized trial compares the effect of follow-up strategy (FU) on endothelial biomarkers to that of 500 ml crystalloid fluid bolus (FB) in oliguric, hemodynamically optimized intensive care unit (ICU) patients. 130 adult subjects were enrolled in two Finnish ICUs from January 2017 to November 2020. Blood and urine samples of 63 patients in FU group and 67 patients in FB group were collected before and after the intervention and analyzed using enzyme-linked immunosorbent assays. Single fluid bolus, given after median of 3887 ml (interquartile range 2842; 5359 ml) resuscitation fluids in the preceding 24 h, increased plasma hyaluronan concentration compared to the follow-up strategy (difference in medians 29.2 ng/ml with 95% CI [14.5ng/ml; 55.5ng/ml], P  < 0.001). No treatment effect was detected in the plasma levels of syndecan-1, , angiopoietin-2, angiopoietin receptors Tie2 and Tie1, or in soluble thrombomodulin in the adjusted median regression analysis. The increase in hyaluronan was independent of its simultaneous renal clearance but correlated moderately with the increase in endothelium-specific Tie1. The follow-up strategy did not show consistent endothelium-sparing effect but protected against hyaluronan increase. The mechanisms and consequences of hyaluronan fluctuations need further clarification. Trial registration: clinicaltrials.gov, NCT02860572. Registered 1 August 2016, https://www.clinicaltrials.gov/study/NCT02860572?term=NCT02860572&rank=1

Background Among critically ill patients with acute kidney injury (AKI), earlier initiation of renal replacement therapy (RRT) may mitigate fluid accumulation and confer better outcomes among individuals with greater fluid overload at randomization. Methods We conducted a pre-planned post hoc analysis of the STandard versus Accelerated initiation of Renal Replacement Therapy in Acute Kidney Injury (STARRT-AKI) trial. We evaluated the effect of accelerated RRT initiation on cumulative fluid balance over the course of 14 days following randomization using mixed models after censoring for death and ICU discharge. We assessed the modifying effect of baseline fluid balance on the impact of RRT initiation strategy on key clinical outcomes. Patients were categorized in quartiles of baseline fluid balance, and the effect of accelerated versus standard RRT initiation on clinical outcomes was assessed in each quartile using risk ratios (95% CI) for categorical variables and mean differences (95% CI) for continuous variables. Results Among 2927 patients in the modified intention-to-treat analysis, 2738 had available data on baseline fluid balance and 2716 (92.8%) had at least one day of fluid balance data following randomization. Over the subsequent 14 days, participants allocated to the accelerated strategy had a lower cumulative fluid balance compared to those in the standard strategy (4509 (− 728 to 11,698) versus 5646 (0 to 13,151) mL, p  = 0.03). Accelerated RRT initiation did not confer greater 90-day survival in any of the baseline fluid balance quartiles (quartile 1: RR 1.11 (95% CI 0.92 to 1.34), quartile 2: RR 1.03 (0.87 to 1.21); quartile 3: RR 1.08 (95% CI 0.91 to 1.27) and quartile 4: RR 0.87 (95% CI 0.73 to 1.03), p value for trend 0.08). Conclusions Earlier RRT initiation in critically ill patients with AKI conferred a modest attenuation of cumulative fluid balance. Nonetheless, among patients with greater fluid accumulation at randomization, accelerated RRT initiation did not have an impact on all-cause mortality. Trial registration : ClinicalTrials.gov number, https://clinicaltrials.gov/ct2/show/NCT02568722 , registered October 6, 2015.

Background Acute kidney injury (AKI) is a frequently occurring syndrome in critically ill patients and is associated with worse outcomes. Biomarkers allow early identification and therapy of AKI which may improve outcomes. Urine chitinase 3-like protein 1 (uCHI3L1) was recently identified as a promising urinary biomarker for AKI. In this multicenter study, we evaluated the diagnostic performance for AKI stage 2 or greater of uCHI3L1 in comparison with the urinary cell cycle arrest biomarkers urinary tissue inhibitor of metalloproteinases-2 (TIMP-2)•insulin-like growth factor-binding protein 7 (IGFBP7) measured by NephroCheck Risk®. Methods Post hoc laboratory study of the prospective observational FINNAKI study. Of this cohort, we included patients with stored admission urine samples and availability of serum creatinine at day 1 of admission. Patients who already had AKI stage 2 or 3 at ICU admission were excluded. AKI was defined and staged according to the KDIGO definition and staging system. The primary endpoint was AKI stage 2 or 3 at day 1. Biomarker performance was assessed by the area under the curve of the receiver operating characteristic curve (AUC). We assessed individual performance and different combinations of urine biomarkers. Results Of 660 included patients, 49 (7.4%) had AKI stages 2–3 at day 1. All urine biomarkers were increased at admission in AKI patients. All biomarkers and most combinations had AUCs < 0.700. The combination uCHI3L1•TIMP-2 was best with a fair AUC of 0.706 (0.670, 0.718). uCHI3L1 had a positive likelihood ratio (LR) of 2.25 which was comparable to that of the NephroCheck Risk® cutoff of 2.0, while the negative LR of 0.53 was comparable to that of the NephroCheck Risk® cutoff of 0.3. Conclusions We found that uCHI3L1 and NephroCheck Risk® had a comparable diagnostic performance for diagnosis of AKI stage 2 or greater within a 24-h period in this multicenter FINNAKI cohort. In contrast to initial discovery and validation studies, the diagnostic performance was poor. Possible explanations for this observation are differences in patient populations, proportion of emergency admissions, proportion of functional AKI, rate of developing AKI, and observation periods for diagnosis of AKI.

Whether positive fluid balance among patients with acute kidney injury (AKI) stems from decreased urine output, overzealous fluid administration, or both is poorly characterized. This was a post hoc analysis of the prospective multicenter observational Finnish Acute Kidney Injury study including 824 AKI and 1162 non-AKI critically ill patients. We matched 616 AKI (diagnosed during the three first intensive care unit (ICU) days) and non-AKI patients using propensity score. During the three first ICU days, AKI patients received median [IQR] of 11.4 L [8.0–15.2]L fluids and non-AKI patients 10.2 L [7.5–13.7]L, p < 0.001 while the fluid output among AKI patients was 4.7 L [3.0–7.2]L and among non-AKI patients 5.8 L [4.1–8.0]L, p < 0.001. In AKI patients, the median [IQR] cumulative fluid balance was 2.5 L [−0.2–6.0]L compared to 0.9 L [−1.4–3.6]L among non-AKI patients, p < 0.001. Among the 824 AKI patients, smaller volumes of fluid input with a multivariable OR of 0.90 (0.88–0.93) and better fluid output (multivariable OR 1.12 (1.07–1.18)) associated with enhanced change of resolution of AKI. AKI patients received more fluids albeit having lower fluid output compared to matched critically ill non-AKI patients. Smaller volumes of fluid input and higher fluid output were associated with better AKI recovery. •Acute kidney injury patients receive more fluids and have reduced fluid output.•Lower fluid input and higher fluid output associates with higher renal recovery rate.

We hypothesised that plasma concentrations of biomarkers of neutrophil activation and pro-inflammatory cytokines differ according to the phase of rapidly evolving sepsis. In an observational study, we measured heparin-binding protein (HBP), myeloperoxidase (MPO), IL-6 and IL-8 in 167 sepsis patients on intensive care unit admission. We prospectively used the emergence of the first sepsis-associated organ dysfunction (OD) as a surrogate for the sepsis phase. Fifty-five patients (of 167, 33%) developed the first OD > 1 h before, 74 (44%) within ± 1 h, and 38 (23%) > 1 h after intensive care unit admission. HBP and MPO were elevated at a median of 12 h before the first OD, remained high up to 24 h, and were not associated with sepsis phase. IL-6 and IL-8 rose and declined rapidly close to OD emergence. Elevation of neutrophil activation markers HBP and MPO was an early event in the evolution of sepsis, lasting beyond the subsidence of the pro-inflammatory cytokine reaction. Thus, as sepsis biomarkers, HBP and MPO were not as prone as IL-6 and IL-8 to the effect of sample timing.