We determined the relationship between changes in bioelectrical impedance analysis (BIA) parameters and response of critically ill patients to fluid therapy during early postoperative period. Associations between BIA values indicating volume status of postoperative patient and clinical outcomes were also evaluated. From May 2019 to April 2020, patients who were admitted to the surgical intensive care unit (SICU) of our institution at more than 48 h after surgery were enrolled. Volume status was measured with a portable BIA device every morning for five days from SICU admission. Overhydration was defined as the case where extracellular water (ECW) ratio > 0.390 measured by BIA. Participants were daily classified into an overhydration or a normohydration group. The relationship between daily hydration status and postoperative outcome was evaluated. Most of the 190 participants showed the overhydration status in the first 48 h after surgery. The overhydration status on day 3 was significant predictor of postoperative morbidities (OR 1.182) and in-hospital mortality (OR 2.040). SOFA score was significant factor of postoperative morbidities (OR 1.163) and in-hospital mortality (OR 3.151) except for the overhydration status on day 3. Cut-off values of overhydration status by ECW ratio at day 3 for predicting postoperative morbidities and in-hospital mortality were > 0.3985 and > 0.4145, respectively. BIA would be a useful and convenient tool to assess the volume status of patients requiring intensive fluid resuscitation in early postoperative period. Overhydration status by ECW ratio on postoperative day 3 needs careful monitoring and appropriate interventions to improve clinical outcomes.

Laparoscopic surgery has been performed less frequently in the era of pancreatic cancer due to technical difficulties and concerns about oncological safety. Radical antegrade modular pancreatosplenectomy (RAMPS) is expected to be helpful to obtain a negative margin during radical lymph node dissection. We hypothesized that it would also be favorable as a laparoscopic application due to unique features. Fifteen laparoscopic RAMPS for well-selected patients with left-sided pancreatic cancer were performed from July 2011 to April 2016. Five trocars were usually used, and the operative procedures and range of dissection were similar to or the same as those of open RAMPS described by Strasberg. All medical records and follow-up data were reviewed and analyzed. All patients had pancreatic ductal adenocarcinoma. Mean operative time was 219.3 ± 53.8 min, and estimated blood loss was 250 ± 70 ml. The length of postoperative hospital stay was 6.1 ± 1.2 days, and postoperative morbidities developed in two patients (13.3%) with urinary retention. The median number of retrieved lymph nodes was 18.1 ± 6.2 and all had negative margins. Median follow-up time was 46.0 months, and the 3-year disease free survival and overall survival rates were 56.3% and 74.1%, respectively. Our early experience with laparoscopic RAMPS achieved feasible perioperative results accompanied by acceptable survival outcomes. Laparoscopic RAMPS could be a safe and oncologically feasible procedure in well-selected patients with left-sided pancreatic cancer.

Malnutrition in critically ill patients is a global concern, especially those who undergo abdominal surgery, as it is associated to higher infectious complications, prolonged hospital stays, and increased morbidity. Despite the importance of proper nutrition, guidelines remain broad, and practical implementation is often inadequate. We aimed to assess the effects of strict nutritional provision and investigate the appropriate target for nutrition support. A prospective, randomized controlled trial was conducted in critically ill patients admitted to intensive care units following abdominal surgery. The intervention group received targeted protein and calories, with consultation from a nutritional support team upon admission. In total, 181 patients in the intervention and 144 in the control group were analyzed. The intervention group demonstrated improved nutrition provision and subsequently better clinical outcomes, including a reduced 60-day mortality (4.4 versus 15.3, p = 0.001), postoperative complications (24.9 versus 47.2, p < 0.001), and in-hospital mortality (5 versus 17.4, p < 0.001). High modified nutrition risk in the critically ill scores [odds ratio (OR) = 2.658, 95% CI = 1.498–4.716] were associated with increased 60-day mortality, while active nutritional intervention (OR = 0.312, 95% CI = 0.111–0.873) was associated with lower mortality rates. Notably, the provision of targeted energy and protein alone did not exhibit a significant association with mortality outcomes.

Objective: Predicting prognosis of in-hospital patients is critical. However, it is challenging to accurately predict the life and death of certain patients at certain period. To determine whether machine learning algorithms could predict in-hospital death of critically ill patients with considerable accuracy and identify factors contributing to the prediction power.Materials and Methods: Using medical data of 1,384 patients admitted to the Surgical Intensive Care Unit (SICU) of our institution, we investigated whether machine learning algorithms could predict in-hospital death using demographic, laboratory, and other disease-related variables, and compared predictions using three different algorithmic methods. The outcome measurement was the incidence of unexpected postoperative mortality which was defined as mortality without pre-existing not-for-resuscitation order that occurred within 30 days of the surgery or within the same hospital stay as the surgery.Results: Machine learning algorithms trained with 43 variables successfully classified dead and live patients with very high accuracy. Most notably, the decision tree showed the higher classification results (Area Under the Receiver Operating Curve, AUC = 0.96) than the neural network classifier (AUC = 0.80). Further analysis provided the insight that serum albumin concentration, total prenatal nutritional intake, and peak dose of dopamine drug played an important role in predicting the mortality of SICU patients.Conclusion: Our results suggest that machine learning algorithms, especially the decision tree method, can provide information on structured and explainable decision flow and accurately predict hospital mortality in SICU hospitalized patients.

For patients undergoing abdominal surgery, malnutrition further increases the susceptibility to infection, surgical complications, and mortality. However, there is no standard tool for identifying high-risk groups of malnutrition or exact criteria for the optimal target of nutrition supply. We aimed to identify the nutritional risk in critically ill patients using modified Nutrition Risk in the Critically Ill (mNUTRIC) scores and assessing the relationship with clinical outcomes. Furthermore, we identified the ideal target of energy intake during the acute postoperative period. A prospective observational study was conducted. mNUTRIC scores and the average calories prescribed and given were calculated. To identify the high-risk group of malnutrition, receiver operating characteristic curves were plotted. The ideal target of energy adequacy and predisposing factors of 90-day mortality were assessed using multiple logistic regression analyses. A total of 206 patients were analyzed. The cutoff value for mNUTRIC score predicting 90-day mortality was 5 (Area under the curve = 0.7, 95% confidence interval (Cl) 0.606–0.795, p < 0.001). A total of 75 patients (36.4%) were classified in the high mNUTRIC group (mNUTRIC ≥ 5) and had a significantly higher postoperative complication and longer length of surgical intensive care unit stay. High mNUTRIC scores (odds ratio = 2.548, 95% CI 1.177–5.514, p = 0.018) and energy adequacy less than 50% (odds ratio = 6.427, 95% CI 1.674–24.674, p = 0.007) were associated with 90-day mortality.