Machine learning-based prognostic model for in-hospital mortality of aortic dissection: Insights from an intensive care medicine perspective.

Objective: Aortic dissection (AD) is a severe emergency with high morbidity and mortality, necessitating strict monitoring and management. This retrospective study aimed to identify prognostic factors and establish predictive models for in-hospital mortality among AD patients in the intensive care unit (ICU). Methods: We retrieved ICU admission records of AD patients from the Medical Information Mart for Intensive Care (MIMIC)-IV critical care data set and the eICU Collaborative Research Database. Functional data analysis was further applied to estimate continuous vital sign processes, and variables associated with in-hospital mortality were identified through univariate analyses. Subsequently, we employed multivariable logistic regression and machine learning techniques, including simple decision tree, random forest (RF), and eXtreme Gradient Boosting (XGBoost) to develop prognostic models for in-hospital mortality. Results: Given 643 ICU admissions from MIMIC-IV and 501 admissions from eICU, 29 and 28 prognostic factors were identified from each database through univariate analyses, respectively. For prognostic model construction, 507 MIMIC-IV admissions were divided into 406 (80%) for training and 101 (20%) for internal validation, and 87 eICU admissions were included as an external validation group. Of the four models tested, the RF consistently exhibited the best performance among different variable subsets, boasting area under the receiver operating characteristic curves of 0.870 and 0.850. The models highlighted the mean 24-h fluid intake as the most potent prognostic factor. Conclusions: The current prognostic models effectively forecasted in-hospital mortality among AD patients, and they pinpointed noteworthy prognostic factors, including initial blood pressure upon ICU admission and mean 24-h fluid intake.

Preload responsiveness-guided fluid removal in mechanically ventilated patients with fluid overload: A comprehensive clinical-physiological study.

This study investigated fluid removal strategies for critically ill patients with fluid overload on mechanical ventilation. Traditionally, a negative fluid balance (FB) is aimed for. However, this approach can have drawbacks. Here, we compared a new approach, namely removing fluids until patients become fluid responsive (FR) to the traditional empiric negative balance approach. Twelve patients were placed in each group (n = 24). FR assessment was performed using passive leg raising (PLR). Both groups maintained stable blood pressure and heart function during fluid management. Notably, the FR group weaned from the ventilator significantly faster than negative FB group (both for a spontaneous breathing trial (14 h vs. 36 h, p = 0.031) and extubation (26 h vs. 57 h, p = 0.007); the difference in total ventilator time wasn't statistically significant (49 h vs. 62 h, p = 0.065). Additionally, FR group avoided metabolic problems like secondary alkalosis and potential hypokalemia seen in the negative FB group. FR-guided fluid-removal in fluid overloaded mechanically ventilated patients was a feasible, safe, and maybe superior strategy in facilitating weaning and disconnection from mechanical ventilation than negative FB-driven fluid removal. FR is a safe endpoint for optimizing cardiac function and preventing adverse consequences during fluid removal.

Fluid excess on intensive care unit after mechanical thrombectomy after acute ischemic stroke is associated with unfavorable neurological and functional outcomes: An observational cohort study.

INTRODUCTION: Endovascular thrombectomy stands as a pivotal component in the standard care for patients experiencing acute ischemic stroke with large vessel occlusion. Subsequent care for patients often extends to a neurological intensive care unit. While fluid management is integral to intensive care, the association between early fluid balance and neurological and functional outcomes post-thrombectomy has not yet been thoroughly investigated. METHODS: In a retrospective analysis of an observational, single-center study spanning from 2015 to 2021 at the University Medical Center Hamburg-Eppendorf, Germany, we enrolled stroke patients who underwent thrombectomy and received subsequent treatment in the ICU. Unfavorable functional and neurological outcome was defined as a mRS > 2 on day 90 after admission (mRS d90) or NIHSS > 5 at discharge, respectively. A multivariate regression model, adjusting for confounders, utilized the average fluid balance in the first 5 days to predict outcomes. Patients were dichotomized by their average fluid balance (>1 L vs <1 L) within the first 5 days, and a multivariate mRS d90 shift analysis was conducted after adjusting for covariates. RESULTS: Between 2015 and 2021, 1252 patients underwent thrombectomy, and 553 patients met the inclusion criteria (299 women [54%]). Unfavorable functional outcome was significantly associated with a higher daily average fluid balance in the first 5 days in the ICU (mRS d90 ⩽ 2: 0.3 ± 0.5 L, mRS d90 > 2: 0.7 ± 0.7 L, p = 0.02). The same association was observed for the NIHSS at discharge (NIHSS ⩽ 5: 0.3 ± 0.5 L; NIHSS > 5: 0.6 ± 0.6 L; p = 0.03). The mRS d90 shift analysis revealed significance for patients with an average fluid balance <1 L for better functional outcomes (adjusted odds ratio [AOR] 2.17; 95% confidence interval [CI] 1.54-3.07; p < 0.01). DISCUSSION: Fluid retention in post-thrombectomy stroke patients in the ICU is associated with poorer functional and neurological outcomes. Consequently, fluid retention emerges as an additional potential predictor for post-intervention stroke outcomes. Our findings provide an initial indication that preventing excessive fluid retention in stroke patients after endovascular thrombectomy could be beneficial for both functional and neurological recovery. Therefore, fluid retention might be an element to consider in optimizing fluid management for stroke patients.

Intravenous fluid therapy: an audit and discussion of improvements required for prescribers and administrators.

BACKGROUND: Foundation doctors and nurses are the clinicians most closely involved in fluid assessment, intravenous (IV) fluid prescription and administration. However, both groups report challenges regarding IV fluids. At a large NHS trust in England, adherence to the National Institute for Health and Care Excellence (NICE) guideline CG174, regarding IV fluids, was largely unknown. AIMS: To assess the baseline adherence, within the hospitals, to CG174 and identify areas for improvement. METHODS: A set of 12 audit standards were developed and used to collect data across 29 clinical areas between September 2022 and May 2023, with 255 patients receiving IV fluids at any time during their inpatient stay included. FINDINGS: For two standards target adherence of 95% was achieved, with an adherence less than 50% in most. Areas of particularly poor adherence included assessing and meeting fluid and electrolyte requirements, patient reassessment and developing IV fluid management plans. CONCLUSION: Trust baseline adherence to NICE CG174 requires improvement, particularly regarding patient assessment and reassessment, and meeting electrolyte requirements.

Ultrasound based estimate of central venous pressure: Are we any closer?

Central venous pressure (CVP) serves as a direct approximation of right atrial pressure and is influenced by factors like total blood volume, venous compliance, cardiac output, and orthostasis. Normal CVP falls within 8-12 mmHg but varies with volume status and venous compliance. Monitoring and managing disturbances in CVP are vital in patients with circulatory shock or fluid disturbances. Elevated CVP can lead to fluid accumulation in the interstitial space, impairing venous return and reducing cardiac preload. While pulmonary artery catheterization and central venous catheter obtained measurements are considered to be more accurate, they carry risk of complications and their usage has not shown clinical improvement. Ultrasound-based assessment of the internal jugular vein (IJV) offers real-time, non-invasive measurement of static and dynamic parameters for estimating CVP. IJV parameters, including diameter and ratio, has demonstrated good correlation with CVP. Despite significant advancements in non-invasive CVP measurement, a reliable tool is yet to be found. Present methods can offer reasonable guidance in assessing CVP, provided their limitations are acknowledged.

Understanding body fluid balance, dehydration and intravenous fluid therapy.

This article examines the role of water in the body, the balance of fluids in the body and the provision of intravenous (IV) fluids to patients who are dehydrated, providing a comprehensive overview of these topics for nurses. The author details various aspects of practice in IV fluid therapy, including the types of fluids used, their indications, administration and potential side effects. The article also discusses dehydration and how nurses can identify and treat this complication, which can occur as a result of many different conditions. Drawing on the relevant research, this article aims to advance nurses' knowledge of the care of patients who are dehydrated and require IV fluid therapy.

Paediatric postoperative acute kidney injury after non-cardiac surgery: Protocol for a systematic review and meta-analysis.

BACKGROUND: Paediatric acute kidney injury (AKI) is associated with significant adverse outcomes such as increased mortality, progression to chronic kidney disease and longer length of stay in hospital. Postoperative AKI is a common and recognized complication after surgery in adults. In the paediatric population, AKI postoperatively to cardiac surgery has been extensively studied. However, the incidence of postoperative AKI after non-cardiac surgery is less clear. Therefore, we aim to assess the available literature on this topic. METHODS: We will conduct a systematic review of observational and randomized controlled trials assessing the incidence of paediatric postoperative AKI after non-cardiac surgery. Pairs of reviewers will independently screen the literature and extract data and assess risk of bias from eligible studies. The databases Pubmed, Cochrane and Web of Sciences will be searched. We will conduct the review in accordance with the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines and the Grading of Recommendation, Assessment, Development and Evaluation (GRADE) approach. If sufficient homogeneity within the included trials we will conduct meta-analyses. DISCUSSION: This systematic review aims to investigate the incidence of postoperative AKI in the paediatric non-cardiac surgery population. The results of this review will provide a foundation for future research in the field of paediatric postoperative AKI.

Hemodialysis Procedures for Stable Incident and Prevalent Patients Optimize Hemodynamic Stability, Dialysis Dose, Electrolytes, and Fluid Balance.

The demographic profile of patients transitioning from chronic kidney disease to kidney replacement therapy is changing, with a higher prevalence of aging patients with multiple comorbidities such as diabetes mellitus and heart failure. Cardiovascular disease remains the leading cause of mortality in this population, exacerbated by the cardiovascular stress imposed by the HD procedure. The first year after transitioning to hemodialysis is associated with increased risks of hospitalization and mortality, particularly within the first 90-120 days, with greater vulnerability observed among the elderly. Based on data from clinics in Fresenius Medical Care Europe, Middle East, and Africa NephroCare, this review aims to optimize hemodialysis procedures to reduce mortality risk in stable incident and prevalent patients. It addresses critical aspects such as treatment duration, frequency, choice of dialysis membrane, dialysate composition, blood and dialysate flow rates, electrolyte composition, temperature control, target weight management, dialysis adequacy, and additional protocols, with a focus on mitigating prevalent intradialytic complications, particularly intradialytic hypotension prevention.

Perioperative fluid balance and early acute kidney injury after lung transplantation.

BACKGROUND: Postoperative acute kidney injury (AKI) after lung transplantation (LTx) is an important factor affecting the short-term outcomes. The focus item of transplantation centers is how to improve the incidence of AKI through optimal management during the perioperative period. OBJECTIVE: The purpose of the study is to investigate the influence of perioperative volume in the development of early AKI following LTx. METHOD: The study involved patients who had undergone LTx between October 2018 to December 2021 at China-Japan Friendship Hospital in Beijing. The patients were monitored for AKI occurring within 72 hours after LTx, as well as the renal outcomes within 30 days. The perioperative volumes were compared and analyzed to determine the impact on various clinical outcomes. RESULTS: 248 patients were enrolled in the study ultimately, with almost half of them (49.6 %) experiencing AKI. 48.8 % of AKI patients received continuous renal replacement therapy (CRRT), with 57.7 % recovered by the end of the 30-day follow-up period. A J-shaped relationship was demonstrated between perioperative volume and AKI incidence. Moreover, maintaining a positive fluid balance would increase the 30-day mortality and lead to poor renal outcomes. CONCLUSION: Perioperative volume is an independent risk factor of early AKI after LTx. Positive fluid balance increases the risk of AKI, 30-day mortality, and adverse renal prognosis. The LTx recipients may benefit from a relatively restrict fluid strategy during and after the lung transplantation.

Utility of copeptin in predicting non-pathological postoperative polyuria in patients affected by acromegaly undergoing pituitary neurosurgery.

PURPOSE: Copeptin efficiently predicts post-neurosurgical central diabetes insipidus (CDI) in patients with hypothalamic-pituitary lesions, but its role in characterizing changes in diuresis in individuals with acromegaly undergoing neurosurgery remains unexplored. Our study aimed to assess changes in postoperative fluid balance in acromegaly patients and correlate them with both copeptin and growth hormone (GH) levels. METHODS: This was a secondary analysis of a prospective study involving 15 acromegaly patients undergoing endoscopic endonasal resection at our University Hospital. Fluid balance was assessed daily, and copeptin and GH levels were evaluated preoperatively (T0), and serially on the morning of the first (T2) and second (T3) postoperative day, with an additional measurement of copeptin one hour post-extubation (T1). Patients with pre-existing or post-neurosurgical CDI were excluded from the analysis. RESULTS: Most patients (11/15) exhibited a negative fluid balance on the second postoperative day, with 4 developing polyuria. Postoperative GH levels did not differ significantly between polyuric and non-polyuric patients, but GH measured at T2 correlated significantly with negative total balance (r = -0.519, p = 0.048). Copeptin levels at T1 were significantly higher in those who developed polyuria (p = 0.013), and a copeptin value > 39.9 pmol/L at T1 showed excellent ability (Sensitivity 100%, Specificity 90.9%, p < 0.001) in predicting postoperative polyuria. Additionally, polyuric patients exhibited a higher T1 / T3 copeptin ratio (p = 0.013) and a negative fluid balance was associated with the remission of acromegaly at 12 months (p = 0.046). CONCLUSION: The early assessment of copeptin, in addition to facilitating the rapid identification of individuals at increased risk of developing CDI, could also allow the recognition of subjects with a tendency towards non-pathological polyuria in the postoperative setting, at least in individuals affected by acromegaly.

Hypocalcemia in the Immediate Postoperative Period Following Metabolic Bariatric Surgery - Hype or Harm?

PURPOSE: Hypocalcemia post-metabolic bariatric surgery (MBS) is a known long-term complication after hypoabsorptive procedures. However, data on immediate postoperative calcium are limited. Our aim was to evaluate the prevalence of hypocalcemia on the 1st postoperative day after MBS and correlate it with potential associated factors. MATERIALS AND METHODS: We analyzed data from all consecutive index MBS over 1 year. We collected data on demographics and on preoperative and postoperative values of serum calcium (TC), albumin, adjusted calcium (AC-Payne formula), magnesium, phosphorus, preoperative vitamin-D, and postoperative 24-h urine output, intravenous fluids (IVF), bolus intravenous furosemide, and creatine phosphokinase (CPK). Continuous data are expressed as means ± SD (range). Categorical data are presented as frequencies (%). Linear regression was implemented to designate potential correlations. RESULTS: The cohort included 86 patients (58.1% females). The mean preoperative TC was 9.4mg/dL ± 0.4 (8.5-10.5) and mean postoperative TC 7.8mg/dL ± 0.6 (6.3-9.3, 17.0% decrease). The mean preoperative AC was 10.1mg/dL ± 0.4 (9.2-11.2) and mean postoperative AC 8.5mg/dL ± 0.6 (7.0-10.0, 15.8% decrease). Seventy-three patients (84.8%) had abnormally low TC (< 8.5mg/dL), and 43 (50%) abnormally low AC. There was only weak correlation between postoperative TC and AC with magnesium (r = 0.258), phosphorus (r = 0.269), vitamin-D (-0.163), 24-h urine output (r = -0.168), IVF (r = -0.237), bolus furosemide (r = 0.155), and mean operative time (r = 0.010). CONCLUSIONS: In our cohort of patients, hypocalcemia was a real problem but we did not find any significant correlation with the examined factors. Further studies are warranted to validate our findings and investigate other potential correlations.

Personalized Hydration Strategy to Improve Fluid Balance and Intermittent Exercise Performance in the Heat.

Sweat rate and electrolyte losses have a large inter-individual variability. A personalized approach to hydration can overcome this issue to meet an individual's needs. This study aimed to investigate the effects of a personalized hydration strategy (PHS) on fluid balance and intermittent exercise performance. Twelve participants conducted 11 laboratory visits including a VO2max test and two 5-day trial arms under normothermic (NOR) or hyperthermic (HYP) environmental conditions. Each arm began with three days of familiarization exercise followed by two random exercise trials with either a PHS or a control (CON). Then, participants crossed over to the second arm for: NOR+PHS, NOR+CON, HYP+PHS, or HYP+CON. The PHS was prescribed according to the participants' fluid and sweat sodium losses. CON drank ad libitum of commercially-available electrolyte solution. Exercise trials consisted of two phases: (1) 45 min constant workload; (2) high-intensity intermittent exercise (HIIT) until exhaustion. Fluids were only provided in phase 1. PHS had a significantly greater fluid intake (HYP+PHS: 831.7 ± 166.4 g; NOR+PHS: 734.2 ± 144.9 g) compared to CON (HYP+CON: 369.8 ± 221.7 g; NOR+CON: 272.3 ± 143.0 g), regardless of environmental conditions (p < 0.001). HYP+CON produced the lowest sweat sodium concentration (56.2 ± 9.0 mmol/L) compared to other trials (p < 0.001). HYP+PHS had a slower elevated thirst perception and a longer HIIT (765 ± 452 s) compared to HYP+CON (548 ± 283 s, p = 0.04). Thus, PHS reinforces fluid intake and successfully optimizes hydration status, regardless of environmental conditions. PHS may be or is an important factor in preventing negative physiological consequences during high-intensity exercise in the heat.

Fluid distribution during surgery in the flat recumbent, Trendelenburg, and the reverse Trendelenburg body positions.

BACKGROUND: The distribution and elimination of infused crystalloid fluid is known to be affected by general anesthesia, but it is unclear whether changes differ depending on whether the patient is operated in the flat recumbent position, the Trendelenburg ("legs up") position, or the reverse Trendelenburg ("head up") position. METHODS: Retrospective data on hemodilution and urine output obtained during and after infusion of 1-2 L of Ringer's solution over 30-60 min were collected from 61 patients undergoing surgery under general anesthesia and 106 volunteers matched with respect to the infusion volume and infusion time. Parameters describing fluid distribution in the anesthetized and awake subjects were compared by population volume kinetic analysis. RESULTS: General anesthesia decreased the rate constant for urine output by 79% (flat recumbent), 91% (legs up) and 91% (head up), suggesting that laparoscopic surgery per se intensified the already strong anesthesia-induced fluid retention. General anesthesia also decreased the rate constant governing the return of the distributed fluid to the plasma by 32%, 15%, and 70%, respectively. These results agree with laboratory data showing a depressive effect of anesthetic drugs on lymphatic pumping, and further suggest that the "legs up" position facilitates lymphatic flow, whereas the "head up" position slows this flow. Both Trendelenburg positions increased swelling of the "third fluid space". CONCLUSIONS: General anesthesia caused retention of infused fluid with preferential distribution to the extravascular space. Both Trendelenburg positions had a modifying influence on the kinetic adaptations that agreed with the gravitational forces inflicted by tilting to body.

Skimmed, Lactose-Free Milk Ingestion Postexercise: Rehydration Effectiveness and Gastrointestinal Disturbances Versus Water and a Sports Drink in Physically Active People.

Postexercise hydration is fundamental to replace fluid loss from sweat. This study evaluated rehydration and gastrointestinal (GI) symptoms for each of three beverages: water (W), sports drink (SD), and skimmed, lactose-free milk (SLM) after moderate-intensity cycling in the heat. Sixteen college students completed three exercise sessions each to lose ≈2% of their body mass. They drank 150% of body mass loss of the drink assigned in randomized order; net fluid balance, diuresis, and GI symptoms were measured and followed up for 3 hr after completion of fluid intake. SLM showed higher fluid retention (∼69%) versus W (∼40%; p < .001); SD (∼56%) was not different from SLM or W (p > .05). Net fluid balance was higher for SLM (-0.26 kg) and SD (-0.42 kg) than W (-0.67 kg) after 3 hr (p < .001), resulting from a significantly lower diuresis with SLM. Reported GI disturbances were mild and showed no difference among drinks (p > .05) despite ingestion of W (1,992 ± 425 ml), SD (1,999 ± 429 ml), and SLM (1,993 ± 426 ml) in 90 min. In conclusion, SLM was more effective than W for postexercise rehydration, showing greater fluid retention for the 3-hr follow-up and presenting with low-intensity GI symptoms similar to those with W and SD. These results confirm that SLM is an effective option for hydration after exercise in the heat.

The methodological challenges faced when conducting hydration research in UK care homes.

BACKGROUND: The evidence base for hydration practice in care homes is underdeveloped. High-quality research is therefore needed to determine what practices support older people with dementia in drinking sufficient fluid. However, methodological developments are needed to be able to do this. AIM: To highlight the methodological issues researchers encountered during a feasibility cluster, randomised controlled trial of ThinkDrink, a hydration care guide for people with dementia living in UK care homes. DISCUSSION: This is a challenging area because of the complexity of recruitment, participation and data collection in care homes. Researchers must pay extra attention to rigour and quality in the design of their studies. There may be multiple challenges, so various strategies may be required. CONCLUSION: It is important that researchers continue to reflect on rigorous approaches to develop evidence in a crucial area of care, despite these challenges. IMPLICATIONS FOR PRACTICE: Researchers working in complex environments face a variety of challenges to complete methodologically rigorous research. It is important for researchers to be critical of research processes and data, to mitigate and overcome these challenges.

Physician awareness of fluid volume administered with intravenous antibiotics - a structured interview-based study.

Objective: Fluids administered as drug diluents with intravenous (IV) medicine constitute a substantial fraction of fluids in inpatients. Are physicians aware of fluid volumes administered with IV antibiotics for patients with suspected infections? Does addressing this lead to adjustments in 24-hour fluid administration/antibiotics? Methods: This cross-sectional interview-based study was conducted in three emergency departments. Physicians were interviewed after prescribing around-the-clock IV antibiotics for ³24 hours to patients with suspected infection. A structured interview guide assessed the physicians' awareness, considerations, and practices when prescribing IV antibiotics. The 24-hour antibiotic fluid volume was calculated. Results: We interviewed 100 physicians. The 24-hour fluid volume administered with IV antibiotics was 400mL (interquartile range: 300-400). Totally, 53% of physicians (n=53) were unaware of the fluid volume administered with IV antibiotics. 76% (n=76) did not account for the antibiotic fluid volume in the 24-hour fluid administration. 96% (n=96) indicated that they would not adjust prescribed fluids after receiving information about 24-hour antibiotic fluid volume administered for their specific patient; no comorbidities associated with fluid intolerance was primary reason for not adjusting prescribed fluids/antibiotics. 79% (n=79) opted for visibility of fluid volumes administered with IV antibiotics in the medical record. Conclusion: The majority of physicians were unaware of fluid volumes administered as a drug diluent with IV antibiotics. The majority chose not to make post-prescribing adjustments to their planned fluid administration; they regarded their patient as fluid tolerable. The physicians opted for visibility of fluid volumes administered as diluents during the prescribing process.

Identifying novel clinical phenotypes of acute respiratory distress syndrome using trajectories of daily fluid balance: a secondary analysis of randomized controlled trials.

BACKGROUND: Previously identified phenotypes of acute respiratory distress syndrome (ARDS) could not reveal the dynamic change of phenotypes over time. We aimed to identify novel clinical phenotypes in ARDS using trajectories of fluid balance, to test whether phenotypes respond differently to different treatment, and to develop a simplified model for phenotype identification. METHODS: FACTT (conservative vs liberal fluid management) trial was classified as a development cohort, joint latent class mixed models (JLCMMs) were employed to identify trajectories of fluid balance. Heterogeneity of treatment effect (HTE) for fluid management strategy across phenotypes was investigated. We also constructed a parsimonious probabilistic model using baseline data to predict the fluid trajectories in the development cohort. The trajectory groups and the probabilistic model were externally validated in EDEN (initial trophic vs full enteral feeding) trial. RESULTS: Using JLCMM, we identified two trajectory groups in the development cohort: Class 1 (n = 758, 76.4% of the cohort) had an early positive fluid balance, but achieved negative fluid balance rapidly, and Class 2 (n = 234, 24.6% of the cohort) was characterized by persistent positive fluid balance. Compared to Class 1 patients, patients in Class 2 had significantly higher 60-day mortality (53.5% vs. 17.8%, p < 0.001), and fewer ventilator-free days (0 vs. 20, p < 0.001). A significant HTE between phenotypes and fluid management strategies was observed in the FACTT. An 8-variables model was derived for phenotype assignment. CONCLUSIONS: We identified and validated two novel clinical trajectories for ARDS patients, with both prognostic and predictive enrichment. The trajectories of ARDS can be identified with simple classifier models.

Predictive value of fluid balance before extubation on its outcome in mechanically ventilated adults in the intensive care unit.

BACKGROUND: Although there are many reasons for extubation failure, maintaining negative or lower positive fluid balances 24 hours before extubation may be a key measure for successful extubation. AIM: To assess the predictive value of fluid balance before extubation and its outcome in mechanically ventilated cases in the intensive care unit (ICU). STUDY DESIGN: This retrospective cohort study involved collecting clinical data from patients undergoing mechanical ventilation in Lanzhou general adult ICU from January 2022 to December 2022. Based on extubation outcomes, the patients were divided into a successful extubation group and a failed extubation group. Their fluid balance levels 24 h before extubation were compared with analyse the predictive value of fluid balance on extubation outcomes in patients undergoing mechanical ventilation. RESULTS: In this study, clinical data from 545 patients admitted to a general adult ICU were collected. According to the inclusion and exclusion criteria, 265 (48.6%) patients were included, of which 197 (74.3%) were successfully extubated; extubation was unsuccessful in 68 (25.7%) patients. The total intake and fluid balance levels in patients in the failed extubation group 24 h before extubation were significantly higher than those in the successful extubation group, with a median of 2679.00 (2410.44-3193.50) mL versus 2435.40 (1805.04-2957.00) mL, 831.50 (26.25-1407.94) mL versus 346.00 (-163.00-941.50) mL. Receiver operating characteristic (ROC) curve analysis showed that the optimal cut-off value for predicting extubation outcomes was 497.5 mL (sensitivity 64.7%, specificity 59.4%) for fluid balance 24 h before extubation. The area under the ROC curve was 0.627 (95% confidence interval [CI] 0.547-0.707). Based on the logistic regression model, cumulative fluid balance >497.5 mL 24 h before extubation could predict its outcomes in mechanically ventilated patients in the ICU (OR = 5.591, 95% CI [2.402-13.015], p < .05). CONCLUSIONS: The fluid balance level 24 h before extubation was correlated with the outcome of extubation in mechanically ventilated patients in the ICU. The risk of extubation failure was higher when the fluid balance level was >497.5 mL. RELEVANCE TO CLINICAL PRACTICE: Tracheal intubation is a crucial life support technique for many critically ill patients, and determining the appropriate time for extubation remains a challenge for clinicians. Although there are many reasons for extubation failure, acute pulmonary oedema caused by continuous positive fluid balance and volume overload is one of the main reasons for extubation failure. Therefore, it is very important to study the relationship between fluid balance and extubation outcome to improve the prognosis of patients with invasive mechanical ventilation in ICU.

Group-Based Trajectory Modeling of Fluid Balance in Elderly Patients with Acute Ischemic Stroke: Analysis from Multicenter ICUs.

INTRODUCTION: Acute ischemic stroke (AIS) significantly contributes to severe disability and mortality among the elderly. This study aims to explore the association between longitudinal fluid balance (FB) trajectories and clinical outcomes in elderly patients with AIS. Our hypothesis posits the existence of multiple latent trajectories of FB in patients with AIS during the initial 7 days following ICU admission. METHODS: Patients (age ≥ 65 years) with AIS and continuous FB records were identified from two large databases. Group-based trajectory modeling identified latent groups with similar 7-day FB trajectories. Subsequently, multivariable logistic and quasi-Poisson regression were employed to evaluate the relationship between trajectory groups and outcomes. Additionally, nonlinear associations between maximum fluid overload (FO) and outcomes were analyzed using restricted cubic spline models. To further validate our findings, subgroup and sensitivity analysis were conducted. RESULTS: A total of 1146 eligible patients were included in this study, revealing three trajectory patterns were identified: low FB (84.8%), decreasing FB (7.2%), and high FB (7.9%). High FB emerged as an independent risk factor for in-hospital mortality. Compared with those without FO, patients with FO had a 1.57-fold increased risk of hospital mortality (adjusted odd ratio (OR) 1.57, 95% confidence interval (CI) 1.08-2.27), 2.37-fold increased risk of adverse kidney event (adjusted OR 2.37, 95% CI 1.56-3.59), and 1.33-fold increased risk of prolonged ICU stay (adjusted incidence rate ratio (IRR) 1.33, 95% CI 1.19-1.48). The risk of hospital mortality and adverse kidney event increased linearly with rising maximum FO (P for non-linearity = 0.263 and 0.563, respectively). CONCLUSION: Daily FB trajectories were associated with adverse outcomes in elderly patients with AIS. Regular assessment of daily fluid status and restriction of FO are crucial for the recovery of critically ill patients.