Association of day 4 cumulative fluid balance with mortality in critically ill patients with influenza: A multicenter retrospective cohort study in Taiwan

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.

Using Body Composition Analysis for Improved Nutritional Intervention in Septic Patients: A Prospective Interventional Study

The study aimed to determine whether using body composition data acquired through bio-electrical impedance analysis (BIA) to adjust diet formulas could improve outcomes in septic patients. There were 132 septic patients in medical intensive care units enrolled in the prospective, randomized, double-blind, interventional study. For the intervention group, dietitians had access to BIA data for adjusting diet formulas according to body composition variables on days 1, 3, and 8. The patients were also stratified based on nutritional risk using the modified Nutrition Risk in Critically ill (mNUTRIC) score. Patients with intervention were more likely to achieve caloric and protein intake goals compared to the control group, especially in the low-risk group. The intervention did not significantly affect mortality, but the survival curves suggested potential benefits. The high-risk group had longer ICU stays and mechanical ventilation duration, which were mitigated by the intervention. Certain body composition variables (e.g., extracellular water to total body water ratio and phase angle) showed differences between high-risk and low-risk groups and may be related to patient outcomes. Non-invasive body composition assessment using BIA can help dietitians adjust diet formulas for critically ill septic patients. Body composition variables may be associated with sepsis outcomes, but further research with larger patient numbers is needed to confirm these findings.

A system theory based digital model for predicting the cumulative fluid balance course in intensive care patients

Background: Surgical interventions can cause severe fluid imbalances in patients undergoing cardiac surgery, affecting length of hospital stay and survival. Therefore, appropriate management of daily fluid goals is a key element of postoperative intensive care in these patients. Because fluid balance is influenced by a complex interplay of patient-, surgery- and intensive care unit (ICU)-specific factors, fluid prediction is difficult and often inaccurate. Methods: A novel system theory based digital model for cumulative fluid balance (CFB) prediction is presented using recorded patient fluid data as the sole parameter source by applying the concept of a transfer function. Using a retrospective dataset of n = 618 cardiac intensive care patients, patient-individual models were created and evaluated. RMSE analyses and error calculations were performed for reasonable combinations of model estimation periods and clinically relevant prediction horizons for CFB. Results: Our models have shown that a clinically relevant time horizon for CFB prediction with the combination of 48 h estimation time and 8-16 h prediction time achieves high accuracy. With an 8-h prediction time, nearly 50% of CFB predictions are within ±0.5 L, and 77% are still within the clinically acceptable range of ±1.0 L. Conclusion: Our study has provided a promising proof of principle and may form the basis for further efforts in the development of computational models for fluid prediction that do not require large datasets for training and validation, as is the case with machine learning or AI-based models. The adaptive transfer function approach allows estimation of CFB course on a dynamically changing patient fluid balance system by simulating the response to the current fluid management regime, providing a useful digital tool for clinicians in daily intensive care.

Association of Hypernatremia with Immune Profiles and Clinical Outcomes in Adult Intensive Care Unit Patients with Sepsis

Both hypernatremia and an abnormal immune response may increase hospital mortality in patients with sepsis. This study examined the association of hypernatremia with abnormal immune response and mortality in 520 adult patients with sepsis in an intensive care unit (ICU). We compared the mortality and ex vivo lipopolysaccharide (LPS)-induced inflammatory response differences among patients with hyponatremia, eunatremia, and hypernatremia, as well as between patients with acquired hypernatremia on ICU day 3 and those with sustained eunatremia over first three ICU days. Compared with eunatremia or hyponatremia, hypernatremia led to higher 7 day, 14 day, 28 day, and hospital mortality rates (p = 0.030, 0.009, 0.010, and 0.033, respectively). Compared with sustained eunatremia, acquired hypernatremia led to higher 7, 14, and 28 day mortality rates (p = 0.019, 0.042, and 0.028, respectively). The acquired hypernatremia group nonsignificantly trended toward increased hospital mortality (p = 0.056). Day 1 granulocyte colony-stimulating factor (G-CSF) and tumor necrosis factor (TNF) α levels were relatively low in patients with hypernatremia (p = 0.020 and 0.010, respectively) but relatively high in patients with acquired hypernatremia (p = 0.049 and 0.009, respectively). Thus, in ICU-admitted septic patients, hypernatremia on admission and in ICU-acquired hypernatremia were both associated with higher mortality. The higher mortality in patients with hypernatremia on admission was possibly related to the downregulation of G-CSF and TNF-α secretion after endotoxin stimulation. Compared to sustained eunatremia, acquired hypernatremia showed immunoparalysis at first and then hyperinflammation on day 3.

A Positive Fluid Balance in the First Week Was Associated With Increased Long-Term Mortality in Critically Ill Patients: A Retrospective Cohort Study

Introduction

Early fluid balance has been found to affect short-term mortality in critically ill patients; however, there is little knowledge regarding the association between early cumulative fluid balance (CFB) and long-term mortality. This study aims to determine the distinct association between CFB day 1-3 (CFB 1-3) and day 4-7 (CFB 4-7) and long-term mortality in critically ill patients.

Patients and Methods

This study was conducted at Taichung Veterans General Hospital, a tertiary care referral center in central Taiwan, by linking the hospital critical care data warehouse 2015-2019 and death registry data of the Taiwanese National Health Research Database. The patients followed up until deceased or the end of the study on 31 December 2019. We use the log-rank test to examine the association between CFB 1-3 and CFB 4-7 with long-term mortality and multivariable Cox regression to identify independent predictors during index admission for long-term mortality in critically ill patients.

Results

A total of 4,610 patients were evaluated. The mean age was 66.4 ± 16.4 years, where 63.8% were men. In patients without shock, a positive CFB 4-7, but not CFB 1-3, was associated with 1-year mortality, while a positive CFB 1-3 and CFB 4-7 had a consistent and excess hazard of 1-year mortality among critically ill patients with shock. The multivariate Cox proportional hazard regression model identified that CFB 1-3 and CFB 4-7 (with per 1-liter increment, HR: 1.047 and 1.094; 95% CI 1.037-1.058 and 1.080-1.108, respectively) were independently associated with high long-term mortality in critically ill patients after adjustment of relevant covariates, including disease severity and the presence of shock.

Conclusions

We found that the fluid balance in the first week, especially on days 4-7, appears to be an early predictor for long-term mortality in critically ill patients. More studies are needed to validate our findings and elucidate underlying mechanisms.

Fluid Overload Phenotypes in Critical Illness-A Machine Learning Approach

BACKGROUND

The detrimental impact of fluid overload (FO) on intensive care unit (ICU) morbidity and mortality is well known. However, research to identify subgroups of patients particularly prone to fluid overload is scarce. The aim of this cohort study was to derive "FO phenotypes" in the critically ill by using machine learning techniques.

METHODS

Retrospective single center study including adult intensive care patients with a length of stay of ≥3 days and sufficient data to compute FO. Data was analyzed by multivariable logistic regression, fast and frugal trees (FFT), classification decision trees (DT), and a random forest (RF) model.

RESULTS

Out of 1772 included patients, 387 (21.8%) met the FO definition. The random forest model had the highest area under the curve (AUC) (0.84, 95% CI 0.79-0.86), followed by multivariable logistic regression (0.81, 95% CI 0.77-0.86), FFT (0.75, 95% CI 0.69-0.79) and DT (0.73, 95% CI 0.68-0.78) to predict FO. The most important predictors identified in all models were lactate and bicarbonate at admission and postsurgical ICU admission. Sepsis/septic shock was identified as a risk factor in the MV and RF analysis.

CONCLUSION

The FO phenotypes consist of patients admitted after surgery or with sepsis/septic shock with high lactate and low bicarbonate.

Impact of Early Fluid Balance on Long-Term Mortality in Critically Ill Surgical Patients: A Retrospective Cohort Study in Central Taiwan

Fluid balance is an essential issue in critical care; however, the impact of early fluid balance on the long-term mortality in critically ill surgical patients remains unknown. This study aimed to address the impact of day 1–3 and day 4–7 fluid balance on the long-term mortality in critically ill surgical patients. We enrolled patients who were admitted to surgical intensive care units (ICUs) during 2015–2019 at a tertiary hospital in central Taiwan and retrieved date-of-death from the Taiwanese nationwide death registration profile. We used a Log-rank test and a multivariable Cox proportional hazards regression model to determine the independent mortality impact of early fluid balance. A total of 6978 patients were included for analyses (mean age: 60.9 ± 15.9 years; 63.9% of them were men). In-hospital mortality, 90-day mortality, 1-year and overall mortality was 10.3%, 15.8%, 23.8% and 31.7%, respectively. In a multivariable Cox proportional hazard regression model adjusted for relevant covariates, we found that positive cumulative day 4–7 fluid balance was independently associated with long-term mortality (aHR 1.083, 95% CI 1.062–1.105), and a similar trend was found on day 1–3 fluid balance, although to a lesser extent (aHR 1.027, 95% CI 1.011–1.043). In conclusion, the fluid balance in the first week of ICU stay, particularly day 4–7 fluid balance, may affect the long-term outcome in critically ill surgical patients.

Dynamic monitoring of kidney injury status over 3 days in the intensive care unit as a sepsis phenotype associated with hospital mortality and hyperinflammation

Background

Sepsis-associated acute kidney injury (AKI) often worsens with the deterioration of a patient's condition. Therefore, we hypothesized that monitoring AKI dynamically from day 1 to day 3 was potential to predict hospital mortality. Specifically, we explored whether monitoring AKI dynamically in the intensive care unit (ICU) could be a sepsis phenotype predictive of mortality. A new classification was established based on the change in the AKI stage from admission day 1 and day 3. We compared the hospital mortality, cytokines, and immune response pattern between each group.

Methods

We retrospectively enrolled 523 patients with sepsis, and we calculated the AKI stages on day 1 and day 3 admission to ICUs.

Among these 523 people, 388 of them were assigned to normal, improved, and deteriorated groups according to the changes in the AKI stages. 263 of which did not develop AKI on day 1 and day 3 (normal group). The AKI stage improved in 68 patients (improved group) and worsened in 57 (deteriorated group). We compared the mortality rates between the groups, and identified the relationship between the dynamic AKI status, immune response patterns, and cytokine levels.

Results

The hospital mortality rate in the deteriorated group was higher than that in the non-deteriorated group (combination of normal and improved group) (p = 0.004). Additionally, according to the Kaplan–Meier analysis, the non-deteriorated group had a distinct hospital survival curve (p = 0.004). Furthermore, both the overexpression of tumor necrosis factor-α and decreased monocyte expression of human leukocyte antigen-DR were present in the deteriorated group.

Conclusions

The deteriorated group was associated with a higher hospital mortality rate, potentially resulting from an abnormal inflammatory response. Worsening AKI in the first 3 days of ICU admission may be a sepsis phenotype predictive of hospital mortality.

Association of fluid balance trajectories with clinical outcomes in patients with septic shock: a prospective multicenter cohort study

BACKGROUND

Septic shock has a high incidence and mortality rate in Intensive Care Units (ICUs). Earlier intravenous fluid resuscitation can significantly improve outcomes in septic patients but easily leads to fluid overload (FO), which is associated with poor clinical outcomes. A single point value of fluid cannot provide enough fluid information. The aim of this study was to investigate the impact of fluid balance (FB) latent trajectories on clinical outcomes in septic patients.

METHODS

Patients were diagnosed with septic shock during the first 48 h, and sequential fluid data for the first 3 days of ICU admission were included. A group-based trajectory model (GBTM) which is designed to identify groups of individuals following similar developmental trajectories was used to identify latent subgroups of individuals following a similar progression of FB. The primary outcomes were hospital mortality, organ dysfunction, major adverse kidney events (MAKE) and severe respiratory adverse events (SRAE). We used multivariable Cox or logistic regression analysis to assess the association between FB trajectories and clinical outcomes.

RESULTS

Nine hundred eighty-six patients met the inclusion criteria and were assigned to GBTM analysis, and three latent FB trajectories were detected. 64 (6.5%), 841 (85.3%), and 81 (8.2%) patients were identified to have decreased, low, and high FB, respectively. Compared with low FB, high FB was associated with increased hospital mortality [hazard ratio (HR) 1.63, 95% confidence interval (CI) 1.22-2.17], organ dysfunction [odds ratio (OR) 2.18, 95% CI 1.22-3.42], MAKE (OR 1.80, 95% CI 1.04-2.63) and SRAE (OR 2.33, 95% CI 1.46-3.71), and decreasing FB was significantly associated with decreased MAKE (OR 0.46, 95% CI 0.29-0.79) after adjustment for potential covariates.

CONCLUSION

Latent subgroups of septic patients followed a similar FB progression. These latent fluid trajectories were associated with clinical outcomes. The decreasing FB trajectory was associated with a decreased risk of hospital mortality and MAKE.

Impact of Early Fluid Balance on 1-Year Mortality in Critically Ill Patients With Cancer: A Retrospective Study in Central Taiwan

A positive fluid balance has been found to be deleterious in critically ill patients; however, the impact of early fluid balance, particularly on long-term outcomes, in critically ill patients with cancer remains unclear. We performed this retrospective study at a tertiary-care referral hospital with 1500 beds and 6 intensive care units (ICUs) in central Taiwan, and 942 patients with cancer admitted to ICUs during 2013 to 2016 were enrolled. The primary outcome was 1-year mortality. Cancer-related data were obtained from the cancer registry, and data during ICU admissions were retrieved from the electronic medical records. The association between fluid balance, which was represented by median and interquartile range, and 1-year mortality was determined by calculating the hazard ratio (HR) with 95% confidence interval (CI) using a multivariable Cox proportional hazards regression model. The in-hospital mortality rate was 22.9% (216 of 942), and the mortality within 1 year after the index ICU admission was 38.7% (365 of 942). Compared to survivors, nonsurvivors tended to have a higher Acute Physiology and Chronic Health Evaluation II score (24.1 ± 6.9 vs 20.5 ± 6.2, P < .01), a higher age (65.0 ± 14.4 vs 61.3 ± 14.3, P < .01), a higher serum creatinine (1.5 ± 1.3 vs 1.0 ± 1.0, P < .01), and a higher cumulative day 1 to 4 fluid balance (2669, 955-5005 vs 4103, 1268-7215 mL, P < .01). Multivariable Cox proportional hazards regression analysis found that cumulative day-4 fluid balance was independently associated with 1-year mortality (adj HR 1.227, 95% CI: 1.132-1.329). A positive day 1 to 4 cumulative fluid balance was associated with shorter 1-year survival in critically ill patients with cancer. Further studies are needed to validate this association.

Examination of the association of steroids with fluid accumulation in critically ill patients, considering the possibility of biases

Glucocorticoids might have significant influence on positive fluid balance, mostly due to their mineralocorticoid effect. We assessed the association between glucocorticoid therapy and fluid balance in septic patients, in the intensive care unit (ICU). We considered two definitions of exposure: daily exposure to glucocorticoids and glucocorticoid treatment at any time. Of 945 patients, 375 were treated with glucocorticoids in the ICU. We applied four regression models. In the first, fluid balance did not differ during days with and without glucocorticoid treatment, among patients treated and not treated with glucocorticoids in the ICU. In our second model, daily fluid balance was increased in patients who were ever treated with glucocorticoids during their ICU stay compared to untreated patients. In the third model, which included only patients treated with glucocorticoids during their ICU stay, glucocorticoid treatment days were not associated with daily fluid balance. In the last model, on "steroid-free days", patients who received glucocorticoid treatment during their ICU stay had a positive fluid balance compared to those who were never treated with steroids. Despite their known mineralocorticoid activity, glucocorticoids themselves appear not to contribute substantially to fluid retention. This work highlights the importance of precise selection of variables to mitigate biases.

COVID-19-associated acute respiratory distress syndrome (CARDS): Current knowledge on pathophysiology and ICU treatment - A narrative review

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) induces coronavirus-19 disease (COVID-19) and is a major health concern. Following two SARS-CoV-2 pandemic “waves,” intensive care unit (ICU) specialists are treating a large number of COVID19-associated acute respiratory distress syndrome (ARDS) patients. From a pathophysiological perspective, prominent mechanisms of COVID19-associated ARDS (CARDS) include severe pulmonary infiltration/edema and inflammation leading to impaired alveolar homeostasis, alteration of pulmonary physiology resulting in pulmonary fibrosis, endothelial inflammation (endotheliitis), vascular thrombosis, and immune cell activation.

Although the syndrome ARDS serves as an umbrella term, distinct, i.e., CARDS-specific pathomechanisms and comorbidities can be noted (e.g., virus-induced endotheliitis associated with thromboembolism) and some aspects of CARDS can be considered ARDS “atypical.” Importantly, specific evidence-based medical interventions for CARDS (with the potential exception of corticosteroid use) are currently unavailable, limiting treatment efforts to mostly supportive ICU care.

In this article, we will discuss the underlying pulmonary pathophysiology and the clinical management of CARDS. In addition, we will outline current and potential future treatment approaches.

Phases of fluid management and the roles of human albumin solution in perioperative and critically ill patients

OBJECTIVE

Positive fluid balance is common among critically ill patients and leads to worse outcomes, particularly in sepsis, acute respiratory distress syndrome, and acute kidney injury. Restrictive fluid infusion and active removal of accumulated fluid are being studied as approaches to prevent and treat fluid overload. Use of human albumin solutions has been investigated in different phases of restrictive fluid resuscitation, and this narrative literature review was undertaken to evaluate hypoalbuminemia and the roles of human serum albumin with respect to hypovolemia and its management.

METHODS

PubMed/EMBASE search terms were: "resuscitation," "fluids," "fluid therapy," "fluid balance," "plasma volume," "colloids," "crystalloids," "albumin," "hypoalbuminemia," "starch," "saline," "balanced salt solution," "gelatin," "goal-directed therapy" (English-language, pre-January 2020). Additional papers were identified by manual searching of reference lists.

RESULTS

Restrictive fluid administration, plus early vasopressor use, may reduce fluid balance, but in some cases fluid overload cannot be entirely avoided. Deresuscitation, with fluid actively removed through diuretics or ultrafiltration, reduces duration of mechanical ventilation and intensive care unit stay. Combining hyperoncotic human albumin solution with diuretics increases hemodynamic stability and diuresis. Hyperoncotic albumin corrects hypoalbuminemia and raises colloid osmotic pressure, limiting edema formation and potentially improving endothelial function. Serum levels of albumin relative to C-reactive protein and lactate may predict which patients will benefit most from albumin therapy.

CONCLUSIONS

Hyperoncotic human albumin solution facilitates restrictive fluid therapy and the effectiveness of deresuscitative measures. Current evidence is mostly from observational studies, and more randomized trials are needed to better establish a personalized approach to fluid management.

Fluid Overload and Risk of Mortality in Critically Ill Patients

BACKGROUND

Fluid overload (FO) is a condition present in critical care units, and it is associated with clinical complications and worse outcomes for severe patients.

OBJECTIVE

The aim of this study was to verify if FO is a risk factor for mortality in critically ill patients.

METHODS

Retrospective study performed in a Brazilian intensive care unit, from January to March 2016, with patients older than 18 years and hospitalized for more than 24 hours. Demographic and clinical data, as well as fluid balance and overload, were analyzed to verify the risk factors for mortality. A logistic regression model was elaborated, and significance was set at P < .05.

RESULTS

There were 158 patients included, of which only 13 (8.2%) presented FO. Mortality was verified in individuals 30 (18.9%), of whom only 7 (23.3%) developed FO, which was lower in survivors 6 (4.9%), P = .001. In the simple regression model, the FO was significant (odds ratio [OR], 6.23; 95% confidence interval [CI], 2.04-19.53), P = .001. However, in the multiple regression model, there were significant findings only for mechanical ventilation (OR, 5.86; 95% CI, 2.10-18.12, P = .001), acute kidney injury (OR, 4.05; 95% CI, 1.53-11; P = .001), and noradrenaline (OR, 3.85; 95% CI, 1.01-9.51; P = .041); FO was not significant (OR, 3.68; 95% CI, 0.91-15.55; P = .069).

CONCLUSION

Fluid overload is higher in patients who died. Therefore, it was not considered a risk factor for mortality.

Impact of corticosteroid treatment on clinical outcomes of influenza-associated ARDS: a nationwide multicenter study

Background

Corticosteroid treatment has been widely used in the treatment of septic shock, influenza, and ARDS, although some previous studies discourage its use in severe influenza patients. This multicenter retrospective cohort study conducted in the intensive care units (ICUs) of eight medical centers across Taiwan aims to determine the real-world status of corticosteroid treatment in patients with influenza-associated acute respiratory distress syndrome (ARDS) and its impact on clinical outcomes. Between October 2015 and March 2016, consecutive ICU patients with virology-proven influenza infections who fulfilled ARDS and received invasive mechanical ventilation were enrolled. The impact of early corticosteroid treatment (≥ 200 mg hydrocortisone equivalent dose within 3 days after ICU admission, determined by a sensitivity analysis) on hospital mortality (the primary outcome) was assessed by multivariable logistic regression analysis, and further confirmed in a propensity score-matched cohort.

Results

Among the 241 patients with influenza-associated ARDS, 85 (35.3%) patients receiving early corticosteroid treatment had similar baseline characteristics, but a significantly higher hospital mortality rate than those without early corticosteroid treatment [43.5% (37/85) vs. 19.2% (30/156), p < 0.001]. Early corticosteroid treatment was independently associated with increased hospital mortality in overall patients [adjusted odds ratio (95% CI) = 5.02 (2.39–10.54), p < 0.001] and in all subgroups. Earlier treatment and higher dosing were associated with higher hospital mortality. Early corticosteroid treatment was associated with a significantly increased odds of subsequent bacteremia [adjusted odds ratio (95% CI) = 2.37 (1.01–5.56)]. The analyses using a propensity score-matched cohort showed consistent results.

Conclusions

Early corticosteroid treatment was associated with a significantly increased hospital mortality in adult patients with influenza-associated ARDS. Earlier treatment and higher dosing were associated with higher hospital mortality. Clinicians should be cautious while using corticosteroid treatment in this patient group.

Using a machine learning approach to predict mortality in critically ill influenza patients: a cross-sectional retrospective multicentre study in Taiwan

OBJECTIVES

Current mortality prediction models used in the intensive care unit (ICU) have a limited role for specific diseases such as influenza, and we aimed to establish an explainable machine learning (ML) model for predicting mortality in critically ill influenza patients using a real-world severe influenza data set.

STUDY DESIGN

A cross-sectional retrospective multicentre study in Taiwan SETTING: Eight medical centres in Taiwan.

PARTICIPANTS

A total of 336 patients requiring ICU-admission for virology-proven influenza at eight hospitals during an influenza epidemic between October 2015 and March 2016.

PRIMARY AND SECONDARY OUTCOME MEASURES

We employed extreme gradient boosting (XGBoost) to establish the prediction model, compared the performance with logistic regression (LR) and random forest (RF), demonstrated the feature importance categorised by clinical domains, and used SHapley Additive exPlanations (SHAP) for visualised interpretation.

RESULTS

The data set contained 76 features of the 336 patients with severe influenza. The severity was apparently high, as shown by the high Acute Physiology and Chronic Health Evaluation II score (22, 17 to 29) and pneumonia severity index score (118, 88 to 151). XGBoost model (area under the curve (AUC): 0.842; 95% CI 0.749 to 0.928) outperformed RF (AUC: 0.809; 95% CI 0.629 to 0.891) and LR (AUC: 0.701; 95% CI 0.573 to 0.825) for predicting 30-day mortality. To give clinicians an intuitive understanding of feature exploitation, we stratified features by the clinical domain. The cumulative feature importance in the fluid balance domain, ventilation domain, laboratory data domain, demographic and symptom domain, management domain and severity score domain was 0.253, 0.113, 0.177, 0.140, 0.152 and 0.165, respectively. We further used SHAP plots to illustrate associations between features and 30-day mortality in critically ill influenza patients.

CONCLUSIONS

We used a real-world data set and applied an ML approach, mainly XGBoost, to establish a practical and explainable mortality prediction model in critically ill influenza patients.

Incorporation of dynamic segmented neutrophil-to-monocyte ratio with leukocyte count for sepsis risk stratification

The association between sepsis and segmented neutrophil-to-monocyte (SeMo) ratio is unclear. We postulated that an increase in dynamic SeMo ratio measurement can be applied in risk stratification. This retrospective study included 727 consecutive sepsis patients in medical intensive care units (ICUs), including a subpopulation of 153 patients. According to the leukocyte (white blood cell, WBC) count on day 3 (normal range, between 4,000/µL and 12,000/µL) and delta SeMo (value of SeMo ratio on day 3 minus value of SeMo ratio on day 1; normal delta SeMo, <7), patients were grouped into 3 (delta SeMo & WBC tool). The survival lines separated significantly with hazard ratios of 1.854 (1.342–2.560) for the delta SeMo or WBC abnormal group and 2.860 (1.849–4.439) for the delta SeMo and WBC abnormal group compared to the delta SeMo and WBC normal group. Delta SeMo & WBC tool and delta sequential organ failure assessment (SOFA) tool performed better than the other tools (delta SeMo, delta WBC, day 3 WBC, and day 1 WBC). Severity in delta SeMo & WBC tool and delta SeMo tool reflected the immune dysfunction score, cytokine expression, and human leukocyte antigen D-related monocyte expression on day 1 and day 3. There was correspondence between delta SOFA and delta WBC and between delta SeMo and delta cytokine expression. Incorporation of dynamic SeMo ratio with WBC count provides risk stratification for sepsis patients admitted in the ICU.

Application of dynamic pulse pressure and vasopressor tools for predicting outcomes in patients with sepsis in intensive care units

PURPOSE

We aimed to determine whether the combination of dynamic pulse pressure and vasopressor (DPV) use is applicable for mortality risk stratification in patients with severe sepsis. We proposed the use of the DPV tool and compared it with traditional sepsis severity indices.

MATERIALS AND METHODS

All adult patients who met the sepsis criteria of the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3) between August 2013 and January 2017 were eligible for the study. Patients who expired within 3 days of admission to the intensive care unit (ICU) were excluded. The primary outcomes were 7-day and 28-day mortality.

RESULTS

The study participants included 757 consecutive adult patients. A subpopulation of 155 patients underwent immune profiling assays on days 1, 3, and 7 of ICU admission. The DPV tool had a better performance for predicting 7-day mortality (area under curve, AUC: 0.70), followed by the Sequential Organ Failure Assessment (SOFA) (AUC: 0.64), the plus pulse pressure (AUC: 0.64). For predicting 28-day mortality, the DPV tool was not inferior to the SOFA (AUC: 0.61), DPV tool (AUC: 0.59).

CONCLUSIONS

The DPV tool can be applied for 7-day and 28-day mortality risk prediction in patients with sepsis.

Machine learning for the prediction of volume responsiveness in patients with oliguric acute kidney injury in critical care

BACKGROUND AND OBJECTIVES

Excess fluid balance in acute kidney injury (AKI) may be harmful, and conversely, some patients may respond to fluid challenges. This study aimed to develop a prediction model that can be used to differentiate between volume-responsive (VR) and volume-unresponsive (VU) AKI.

METHODS

AKI patients with urine output < 0.5 ml/kg/h for the first 6 h after ICU admission and fluid intake > 5 l in the following 6 h in the US-based critical care database (Medical Information Mart for Intensive Care (MIMIC-III)) were considered. Patients who received diuretics and renal replacement on day 1 were excluded. Two predictive models, using either machine learning extreme gradient boosting (XGBoost) or logistic regression, were developed to predict urine output > 0.65 ml/kg/h during 18 h succeeding the initial 6 h for assessing oliguria. Established models were assessed by using out-of-sample validation. The whole sample was split into training and testing samples by the ratio of 3:1.

MAIN RESULTS

Of the 6682 patients included in the analysis, 2456 (36.8%) patients were volume responsive with an increase in urine output after receiving > 5 l fluid. Urinary creatinine, blood urea nitrogen (BUN), age, and albumin were the important predictors of VR. The machine learning XGBoost model outperformed the traditional logistic regression model in differentiating between the VR and VU groups (AU-ROC, 0.860; 95% CI, 0.842 to 0.878 vs. 0.728; 95% CI 0.703 to 0.753, respectively).

CONCLUSIONS

The XGBoost model was able to differentiate between patients who would and would not respond to fluid intake in urine output better than a traditional logistic regression model. This result suggests that machine learning techniques have the potential to improve the development and validation of predictive modeling in critical care research.

Incorporation of point-of-care ultrasound into morning round is associated with improvement in clinical outcomes in critically ill patients with sepsis

OBJECTIVES

Point-of-care ultrasound (POCUS) has been widely used in the intensive care unit (ICU). However, it is largely unknown whether the use of POCUS is associated with improved patient-important outcomes. The study aimed to investigate whether incorporation of POCUS during morning round on a routine basis was able to improve clinical outcomes in critically ill patients with sepsis.

DESIGN

It was a prospective observational study.

SETTING

A tertiary care emergency intensive care unit.

PATIENTS

All patients admitted to the emergency ICU from January 2016 to December 2017 were screened for potential eligibility. Sepsis was defined as infection plus signs of organ dysfunction.

INTERVENTION

The intervention group incorporated POCUS during morning round on a routine basis, and a checklist was developed to improve the compliance. The control group did not have the mandates to perform POCUS during morning round, but could use POCUS when necessary.

MEASUREMENTS

Clinical outcomes of mortality, length of stay in ICU, durations of vasopressors and mechanical ventilation were compared between the intervention and control groups. Multivariable regression model was employed to adjust for confounding factors.

MAIN RESULTS

A total of 129 subjects, including 88 in the control group and 41 in the intervention group, were included for analysis. Univariate analysis showed that the intervention group had shorter durations of mechanical ventilation (MV) (4.5 ± 1.2 vs. 5.7 ± 1.0 days; p = 0.034) and more negative fluid balance (-143 vs. 48 ml/24 h; p = 0.003) on day 3. In multivariable model, routine incorporation of POCUS was associated with lower risk of prolonged (>7 days) ICU stay (OR: 0.39, 95% CI: 0.29-0.88; p = 0.029).

CONCLUSIONS

The study showed that incorporation of POCUS during morning round on a routine basis was associated with shortened duration of MV and length of stay in ICU. The possible mechanism underlying the relationship may be via reduced fluid administration. Future randomized controlled trials are needed to validate current findings.

Ventilator Dependence Risk Score for the Prediction of Prolonged Mechanical Ventilation in Patients Who Survive Sepsis/Septic Shock with Respiratory Failure

We intended to develop a scoring system to predict mechanical ventilator dependence in patients who survive sepsis/septic shock with respiratory failure. This study evaluated 251 adult patients in medical intensive care units (ICUs) between August 2013 to October 2015, who had survived for over 21 days and received aggressive treatment. The risk factors for ventilator dependence were determined. We then constructed a ventilator dependence (VD) risk score using the identified risk factors. The ventilator dependence risk score was calculated as the sum of the following four variables after being adjusted by proportion to the beta coefficient. We assigned a history of previous stroke, a score of one point, platelet count less than 150,000/μL a score of one point, pH value less than 7.35 a score of two points, and the fraction of inspired oxygen on admission day 7 over 39% as two points. The area under the curve in the derivation group was 0.725 (p < 0.001). We then applied the VD risk score for validation on 175 patients. The area under the curve in the validation group was 0.658 (p = 0.001). VD risk score could be applied to predict prolonged mechanical ventilation in patients who survive sepsis/septic shock.