Fluid Overload Phenotypes in Critical Illness-A Machine Learning Approach

The association of midregional pro-adrenomedullin (MR-proADM) at ICU admission and fluid overload in patients post elective cardiac surgery

Postoperative fluid overload (FO) after cardiac surgery is common and affects recovery. Predicting FO could help optimize fluid management. This post-hoc analysis of the HERACLES randomized controlled trial evaluated the predictive value of MR-proADM for FO post-cardiac surgery. MR-proADM levels were measured at four different timepoints in 33 patients undergoing elective cardiac surgery. Patients were divided into FO (> 5% weight gain) and no-FO at ICU discharge. The primary outcome was the predictive power of MR-proADM at ICU admission for FO at discharge. Secondary outcomes included the predictive value of MR-proADM for FO on day 6 post-surgery and changes over time. The association between MR-proADM and FO at ICU discharge or day 6 post-surgery was not significant (crude odds ratio (cOR): 4.3 (95% CI 0.5-40.9, p = 0.201) and cOR 1.1 (95% CI 0.04-28.3, p = 0.954)). MR-proADM levels over time did not differ significantly between patients with and without FO at ICU discharge (p = 0.803). MR-proADM at ICU admission was not associated with fluid overload at ICU discharge in patients undergoing elective cardiac surgery. MR-proADM levels over time were not significantly different between groups, although elevated levels were observed in patients with FO.

Fluid accumulation syndrome in sepsis and septic shock: pathophysiology, relevance and treatment-a comprehensive review

In this review, we aimed to comprehensively summarize current literature on pathophysiology, relevance, diagnosis and treatment of fluid accumulation in patients with sepsis/septic shock. Fluid accumulation syndrome (FAS) is defined as fluid accumulation (any degree, expressed as percentage from baseline body weight) with new onset organ-failure. Over the years, many studies have described the negative impact of FAS on clinically relevant outcomes. While the relationship between FAS and ICU outcomes is well described, uncertainty exists regarding its diagnosis, monitoring and treatment. A stepwise approach is suggested to prevent and treat FAS in patients with septic shock, including minimizing fluid intake (e.g., by limiting intravenous fluid administration and employing de-escalation whenever possible), limiting sodium and chloride administration, and maximizing fluid output (e.g., with diuretics, or renal replacement therapy). Current literature implies the need for a multi-tier, multi-modal approach to de-resuscitation, combining a restrictive fluid management regime with a standardized early active de-resuscitation, maintenance fluid reduction (avoiding fluid creep) and potentially using physical measures such as compression stockings.Trial registration: Not applicable.

Influence of fluid accumulation on major adverse kidney events in critically ill patients - an observational cohort study

BACKGROUND

Fluid accumulation (FA) is known to be associated with acute kidney injury (AKI) during intensive care unit (ICU) stay but data on mid-term renal outcome is scarce. The aim of this study was to investigate the association between FA at ICU day 3 and major adverse kidney events in the first 30 days after ICU admission (MAKE30).

METHODS

Retrospective, single-center cohort study including adult ICU patients with sufficient data to compute FA and MAKE30. We defined FA as a positive cumulative fluid balance greater than 5% of bodyweight. The association between FA and MAKE30, including its sub-components, as well as the serum creatinine trajectories during ICU stay were examined. In addition, we performed a sensitivity analysis for the stage of AKI and the presence of chronic kidney disease (CKD).

RESULTS

Out of 13,326 included patients, 1,100 (8.3%) met the FA definition. FA at ICU day 3 was significantly associated with MAKE30 (adjusted odds ratio [aOR] 1.96; 95% confidence interval [CI] 1.67-2.30; p < 0.001) and all sub-components: need for renal replacement therapy (aOR 3.83; 95%CI 3.02-4.84), persistent renal dysfunction (aOR 1.72; 95%CI 1.40-2.12), and 30-day mortality (aOR 1.70; 95%CI 1.38-2.09), p all < 0.001. The sensitivity analysis showed an association of FA with MAKE30 independent from a pre-existing CKD, but exclusively in patients with AKI stage 3. Furthermore, FA was independently associated with the creatinine trajectory over the whole observation period.

CONCLUSIONS

Fluid accumulation is significantly associated with MAKE30 in critically ill patients. This association is independent from pre-existing CKD and strongest in patients with AKI stage 3.

Unsupervised machine learning analysis to identify patterns of ICU medication use for fluid overload prediction

INTRODUCTION

Intravenous (IV) medications are a fundamental cause of fluid overload (FO) in the intensive care unit (ICU); however, the association between IV medication use (including volume), administration timing, and FO occurrence remains unclear.

METHODS

This retrospective cohort study included consecutive adults admitted to an ICU ≥72 hours with available fluid balance data. FO was defined as a positive fluid balance ≥7% of admission body weight within 72 hours of ICU admission. After reviewing medication administration record (MAR) data in three-hour periods, IV medication exposure was categorized into clusters using principal component analysis (PCA) and Restricted Boltzmann Machine (RBM). Medication regimens of patients with and without FO were compared within clusters to assess for temporal clusters associated with FO using the Wilcoxon rank sum test. Exploratory analyses of the medication cluster most associated with FO for medications frequently appearing and used in the first 24 hours was conducted.

RESULTS

FO occurred in 127/927 (13.7%) of the patients enrolled. Patients received a median (IQR) of 31 (13-65) discrete IV medication administrations over the 72-hour period. Across all 47,803 IV medication administrations, ten unique IV medication clusters were identified with 121-130 medications in each cluster. Among the ten clusters, cluster 7 had the greatest association with FO; the mean number of cluster 7 medications received was significantly greater in patients in the FO cohort compared to patients who did not experience FO (25.6 vs.10.9. p<0.0001). 51 of the 127 medications in cluster 7 (40.2%) appeared in > 5 separate 3-hour periods during the 72-hour study window. The most common cluster 7 medications included continuous infusions, antibiotics, and sedatives/analgesics. Addition of cluster 7 medications to a prediction model with APACHE II score and receipt of diuretics improved the ability for the model to predict fluid overload (AUROC 5.65, p =0.0004).

CONCLUSIONS

Using ML approaches, a unique IV medication cluster was strongly associated with FO. Incorporation of this cluster improved the ability to predict development of fluid overload in ICU patients compared with traditional prediction models. This method may be further developed into real-time clinical applications to improve early detection of adverse outcomes.

Improving mixed-integer temporal modeling by generating synthetic data using conditional generative adversarial networks: A case study of fluid overload prediction in the intensive care unit

OBJECTIVE

The challenge of mixed-integer temporal data, which is particularly prominent for medication use in the critically ill, limits the performance of predictive models. The purpose of this evaluation was to pilot test integrating synthetic data within an existing dataset of complex medication data to improve machine learning model prediction of fluid overload.

MATERIALS AND METHODS

This retrospective cohort study evaluated patients admitted to an ICU ≥ 72 h. Four machine learning algorithms to predict fluid overload after 48-72 h of ICU admission were developed using the original dataset. Then, two distinct synthetic data generation methodologies (synthetic minority over-sampling technique (SMOTE) and conditional tabular generative adversarial network (CTGAN)) were used to create synthetic data. Finally, a stacking ensemble technique designed to train a meta-learner was established. Models underwent training in three scenarios of varying qualities and quantities of datasets.

RESULTS

Training machine learning algorithms on the combined synthetic and original dataset overall increased the performance of the predictive models compared to training on the original dataset. The highest performing model was the meta-model trained on the combined dataset with 0.83 AUROC while it managed to significantly enhance the sensitivity across different training scenarios.

DISCUSSION

The integration of synthetically generated data is the first time such methods have been applied to ICU medication data and offers a promising solution to enhance the performance of machine learning models for fluid overload, which may be translated to other ICU outcomes. A meta-learner was able to make a trade-off between different performance metrics and improve the ability to identify the minority class.

Machine learning vs. traditional regression analysis for fluid overload prediction in the ICU

Fluid overload, while common in the ICU and associated with serious sequelae, is hard to predict and may be influenced by ICU medication use. Machine learning (ML) approaches may offer advantages over traditional regression techniques to predict it. We compared the ability of traditional regression techniques and different ML-based modeling approaches to identify clinically meaningful fluid overload predictors. This was a retrospective, observational cohort study of adult patients admitted to an ICU ≥ 72 h between 10/1/2015 and 10/31/2020 with available fluid balance data. Models to predict fluid overload (a positive fluid balance ≥ 10% of the admission body weight) in the 48-72 h after ICU admission were created. Potential patient and medication fluid overload predictor variables (n = 28) were collected at either baseline or 24 h after ICU admission. The optimal traditional logistic regression model was created using backward selection. Supervised, classification-based ML models were trained and optimized, including a meta-modeling approach. Area under the receiver operating characteristic (AUROC), positive predictive value (PPV), and negative predictive value (NPV) were compared between the traditional and ML fluid prediction models. A total of 49 of the 391 (12.5%) patients developed fluid overload. Among the ML models, the XGBoost model had the highest performance (AUROC 0.78, PPV 0.27, NPV 0.94) for fluid overload prediction. The XGBoost model performed similarly to the final traditional logistic regression model (AUROC 0.70; PPV 0.20, NPV 0.94). Feature importance analysis revealed severity of illness scores and medication-related data were the most important predictors of fluid overload. In the context of our study, ML and traditional models appear to perform similarly to predict fluid overload in the ICU. Baseline severity of illness and ICU medication regimen complexity are important predictors of fluid overload.

The application of artificial intelligence in the management of sepsis

Sepsis is a complex and heterogeneous syndrome that remains a serious challenge to healthcare worldwide. Patients afflicted by severe sepsis or septic shock are customarily placed under intensive care unit (ICU) supervision, where a multitude of apparatus is poised to produce high-granularity data. This reservoir of high-quality data forms the cornerstone for the integration of AI into clinical practice. However, existing reviews currently lack the inclusion of the latest advancements. This review examines the evolving integration of artificial intelligence (AI) in sepsis management. Applications of artificial intelligence include early detection, subtyping analysis, precise treatment and prognosis assessment. AI-driven early warning systems provide enhanced recognition and intervention capabilities, while profiling analyzes elucidate distinct sepsis manifestations for targeted therapy. Precision medicine harnesses the potential of artificial intelligence for pathogen identification, antibiotic selection, and fluid optimization. In conclusion, the seamless amalgamation of artificial intelligence into the domain of sepsis management heralds a transformative shift, ushering in novel prospects to elevate diagnostic precision, therapeutic efficacy, and prognostic acumen. As AI technologies develop, their impact on shaping the future of sepsis care warrants ongoing research and thoughtful implementation.

Protocolised early de-resuscitation in septic shock (REDUCE): protocol for a randomised controlled multicentre feasibility trial

BACKGROUND

Fluid overload is associated with excess mortality in septic shock. Current approaches to reduce fluid overload include restrictive administration of fluid or active removal of accumulated fluid. However, evidence on active fluid removal is scarce. The aim of this study is to assess the efficacy and feasibility of an early de-resuscitation protocol in patients with septic shock.

METHODS

All patients admitted to the intensive care unit (ICU) with a septic shock are screened, and eligible patients will be randomised in a 1:1 ratio to intervention or standard of care.

INTERVENTION

Fluid management will be performed according to the REDUCE protocol, where resuscitation fluid will be restricted to patients showing signs of poor tissue perfusion. After the lactate has peaked, the patient is deemed stable and assessed for active de-resuscitation (signs of fluid overload). The primary objective of this study is the proportion of patients with a negative cumulative fluid balance at day 3 after ICU. Secondary objectives are cumulative fluid balances throughout the ICU stay, number of patients with fluid overload, feasibility and safety outcomes and patient-centred outcomes. The primary outcome will be assessed by a logistic regression model adjusting for the stratification variables (trial site and chronic renal failure) in the intention-to-treat population.

ETHICS AND DISSEMINATION

The study was approved by the respective ethical committees (No 2020-02197). The results of the REDUCE trial will be published in an international peer-reviewed medical journal regardless of the results.

TRIAL REGISTRATION NUMBER

ClinicalTrials.gov, NCT04931485.

Cluster analysis driven by unsupervised latent feature learning of medications to identify novel pharmacophenotypes of critically ill patients

Unsupervised clustering of intensive care unit (ICU) medications may identify unique medication clusters (i.e., pharmacophenotypes) in critically ill adults. We performed an unsupervised analysis with Restricted Boltzmann Machine of 991 medications profiles of patients managed in the ICU to explore pharmacophenotypes that correlated with ICU complications (e.g., mechanical ventilation) and patient-centered outcomes (e.g., length of stay, mortality). Six unique pharmacophenotypes were observed, with unique medication profiles and clinically relevant differences in ICU complications and patient-centered outcomes. While pharmacophenotypes 2 and 4 had no statistically significant difference in ICU length of stay, duration of mechanical ventilation, or duration of vasopressor use, their mortality differed significantly (9.0% vs. 21.9%, p < 0.0001). Pharmacophenotype 4 had a mortality rate of 21.9%, compared with the rest of the pharmacophenotypes ranging from 2.5 to 9%. Phenotyping approaches have shown promise in classifying the heterogenous syndromes of critical illness to predict treatment response and guide clinical decision support systems but have never included comprehensive medication information. This first-ever machine learning approach revealed differences among empirically-derived subgroups of ICU patients that are not typically revealed by traditional classifiers. Identification of pharmacophenotypes may enable enhanced decision making to optimize treatment decisions.

Improving irregular temporal modeling by integrating synthetic data to the electronic medical record using conditional GANs: a case study of fluid overload prediction in the intensive care unit

Objective

The challenge of irregular temporal data, which is particularly prominent for medication use in the critically ill, limits the performance of predictive models. The purpose of this evaluation was to pilot test integrating synthetic data within an existing dataset of complex medication data to improve machine learning model prediction of fluid overload.

Materials and Methods

This retrospective cohort study evaluated patients admitted to an ICU ≥ 72 hours. Four machine learning algorithms to predict fluid overload after 48-72 hours of ICU admission were developed using the original dataset. Then, two distinct synthetic data generation methodologies (synthetic minority over-sampling technique (SMOTE) and conditional tabular generative adversarial network (CT-GAN)) were used to create synthetic data. Finally, a stacking ensemble technique designed to train a meta-learner was established. Models underwent training in three scenarios of varying qualities and quantities of datasets.

Results

Training machine learning algorithms on the combined synthetic and original dataset overall increased the performance of the predictive models compared to training on the original dataset. The highest performing model was the metamodel trained on the combined dataset with 0.83 AUROC while it managed to significantly enhance the sensitivity across different training scenarios.

Discussion

The integration of synthetically generated data is the first time such methods have been applied to ICU medication data and offers a promising solution to enhance the performance of machine learning models for fluid overload, which may be translated to other ICU outcomes. A meta-learner was able to make a trade-off between different performance metrics and improve the ability to identify the minority class.

Using Machine-Learning to Assess the Prognostic Value of Early Enteral Feeding Intolerance in Critically Ill Patients: A Retrospective Study

BACKGROUND

The association between gastrointestinal intolerance during early enteral nutrition (EN) and adverse clinical outcomes in critically ill patients is controversial. We aimed to assess the prognostic value of enteral feeding intolerance (EFI) markers during early ICU stays and to predict early EN failure using a machine learning (ML) approach.

METHODS

We performed a retrospective analysis of data from adult patients admitted to Beilinson Hospital ICU between January 2011 and December 2018 for more than 48 h and received EN. Clinical data, including demographics, severity scores, EFI markers, and medications, along with 72 h after admission, were analyzed by ML algorithms. Prediction performance was assessed by the area under the receiver operating characteristics (AUCROC) of a ten-fold cross-validation set.

RESULTS

The datasets comprised 1584 patients. The means of the cross-validation AUCROCs for 90-day mortality and early EN failure were 0.73 (95% CI 0.71-0.75) and 0.71 (95% CI 0.67-0.74), respectively. Gastric residual volume above 250 mL on the second day was an important component of both prediction models.

CONCLUSIONS

ML underlined the EFI markers that predict poor 90-day outcomes and early EN failure and supports early recognition of at-risk patients. Results have to be confirmed in further prospective and external validation studies.

Pharmacophenotype identification of intensive care unit medications using unsupervised cluster analysis of the ICURx common data model

BACKGROUND

Identifying patterns within ICU medication regimens may help artificial intelligence algorithms to better predict patient outcomes; however, machine learning methods incorporating medications require further development, including standardized terminology. The Common Data Model for Intensive Care Unit (ICU) Medications (CDM-ICURx) may provide important infrastructure to clinicians and researchers to support artificial intelligence analysis of medication-related outcomes and healthcare costs. Using an unsupervised cluster analysis approach in combination with this common data model, the objective of this evaluation was to identify novel patterns of medication clusters (termed 'pharmacophenotypes') correlated with ICU adverse events (e.g., fluid overload) and patient-centered outcomes (e.g., mortality).

METHODS

This was a retrospective, observational cohort study of 991 critically ill adults. To identify pharmacophenotypes, unsupervised machine learning analysis with automated feature learning using restricted Boltzmann machine and hierarchical clustering was performed on the medication administration records of each patient during the first 24 h of their ICU stay. Hierarchical agglomerative clustering was applied to identify unique patient clusters. Distributions of medications across pharmacophenotypes were described, and differences among patient clusters were compared using signed rank tests and Fisher's exact tests, as appropriate.

RESULTS

A total of 30,550 medication orders for the 991 patients were analyzed; five unique patient clusters and six unique pharmacophenotypes were identified. For patient outcomes, compared to patients in Clusters 1 and 3, patients in Cluster 5 had a significantly shorter duration of mechanical ventilation and ICU length of stay (p < 0.05); for medications, Cluster 5 had a higher distribution of Pharmacophenotype 1 and a smaller distribution of Pharmacophenotype 2, compared to Clusters 1 and 3. For outcomes, patients in Cluster 2, despite having the highest severity of illness and greatest medication regimen complexity, had the lowest overall mortality; for medications, Cluster 2 also had a comparably higher distribution of Pharmacophenotype 6.

CONCLUSION

The results of this evaluation suggest that patterns among patient clusters and medication regimens may be observed using empiric methods of unsupervised machine learning in combination with a common data model. These results have potential because while phenotyping approaches have been used to classify heterogenous syndromes in critical illness to better define treatment response, the entire medication administration record has not been incorporated in those analyses. Applying knowledge of these patterns at the bedside requires further algorithm development and clinical application but may have the future potential to be leveraged in guiding medication-related decision making to improve treatment outcomes.

Active fluid de-resuscitation in critically ill patients with septic shock: A systematic review and meta-analysis

PURPOSE

To evaluate the impact of active fluid de-resuscitation on mortality in critically ill patients with septic shock.

METHODS

A systematic search was performed on PubMed, EmBase, and the Cochrane Library databases. Trials investigating active fluid de-resuscitation and reporting data on mortality in patients with septic shock were eligible. The primary objective was the impact of active de-resuscitation in patients with septic shock on short-term mortality. Secondary outcomes were whether de-resuscitation lead to a fluid separation, and the impact of de-resuscitation on patient-centred outcomes.

RESULTS

Thirteen trials (8,030 patients) were included in the systematic review, whereof 5 randomised-controlled trials (RCTs) were included in the meta-analysis. None of the RCTs showed a reduction in mortality with active de-resuscitation measures (relative risk (RR) 1.12 [95%-CI 0.84 - 1.48]). Fluid separation was achieved by two RCTs. Evidence from non-randomised trials suggests a mortality benefit with de-resuscitation strategies and indicates a trend towards a more negative fluid balance. Patient-centred outcomes were not influenced in the RCTs, and only one non-randomised trial revealed an impact on the duration of mechanical ventilation and renal replacement requirement (RRT).

CONCLUSION

We found no evidence for superiority of active fluid de-resuscitation compared to usual care regarding mortality, fluid balance or patient-centred outcomes in patients with septic shock. Current evidence is limited by the lack of high-quality RCTs in patients with septic shock, the small sample sizes and the heterogeneity of the applied de-resuscitation techniques. In addition, validity of the majority of RCTs is compromised by their inability to achieve fluid separation.

Fluid overload and mortality in critically ill patients with severe heart failure and cardiogenic shock-An observational cohort study

OBJECTIVE

Patients with heart failure (HF) and cardiogenic shock are especially prone to the negative effects of fluid overload (FO); however, fluid resuscitation in respective patients is sometimes necessary resulting in FO. We aimed to study the association of FO at ICU discharge with 30-day mortality in patients admitted to the ICU due to severe heart failure and/or cardiogenic shock.

METHODS

Retrospective, single-center cohort study. Patients with admission diagnoses of severe HF and/or cardiogenic shock were eligible. The following exclusion criteria were applied: (I) patients younger than 16 years, (II) patients admitted to our intermediate care unit, and (III) patients with incomplete data to determine FO at ICU discharge. We used a cumulative weight-adjusted definition of fluid balance and defined more than 5% as FO. The data were analyzed by univariate and adjusted univariate logistic regression.

RESULTS

We included 2,158 patients in our analysis. 185 patients (8.6%) were fluid overloaded at ICU discharge. The mean FO in the FO group was 7.2% [interquartile range (IQR) 5.8-10%]. In patients with FO at ICU discharge, 30-day mortality was 22.7% compared to 11.7% in non-FO patients (p < 0.001). In adjusted univariate logistic regression, we did not observe any association of FO at discharge with 30-day mortality [odds ratio (OR) 1.48; 95% confidence interval (CI) 0.81-2.71, p = 0.2]. No association between FO and 30-day mortality was found in the subgroups with HF only or cardiogenic shock (all p > 0.05). Baseline lactate (adjusted OR 1.27; 95% CI 1.13-1.42; p < 0.001) and cardiac surgery at admission (adjusted OR 1.94; 95% CI 1.0-3.76; p = 0.05) were the main associated factors with FO at ICU discharge.

CONCLUSION

In patients admitted to the ICU due to severe HF and/or cardiogenic shock, FO at ICU discharge seems not to be associated with 30-day mortality.