Machine learning-based prognostic modeling of patients with acute heart failure receiving furosemide in intensive care units.

Purpose: This study developed machine learning models to predict in-hospital mortality, initiation of acute renal replacement therapy, and mechanical ventilation in patients with acute heart failure receiving furosemide in intensive care units. Method: An extensive database comprising static and dynamic features obtained from a Japanese hospital chain was used to construct and train the machine learning models. Results: The results revealed that the proposed machine learning models predict in-hospital mortality, initiation of acute renal replacement therapy, and mechanical ventilation with good accuracy. However, the optimal models vary depending on the predicted outcomes. The linear support vector machine classification models exhibited the highest in-hospital mortality and mechanical ventilation prediction accuracy, with the area under the receiver operating characteristic curve of 0.73 and 0.73, respectively, whereas the multi-layer neural network exhibited the highest accuracy for acute renal replacement therapy initiation prediction with an area under the receiver operating characteristic curve of 0.70. Conclusions: In conclusion, this study demonstrated that machine learning models could help predict the clinical outcomes of patients with acute heart failure receiving furosemide. However, the optimal models may differ depending on the outcome of interest.

Evaluation of adhesion barrier types in a rat hepatectomy-induced adhesion model.

BACKGROUND: Adhesion formation after hepatectomy creates problems for repeat hepatectomy. This study aimed to compare the effectiveness of a spray (AdSpray) and sheet adhesion barrier (Seprafilm) in a rat hepatectomy-induced adhesion model. METHODS: Thirty male Sprague-Dawley rats underwent partial resection of the left lateral liver lobe. They were randomly assigned to control (n = 10), AdSpray (n = 10), and Seprafilm groups. Seven days after surgery, the animals were sacrificed, and adhesions at the hepatic resection surface were blindly evaluated. RESULTS: In the control group, adhesions were formed in all 10 animals (100%), with a 69% adhesion extent (mean). In the AdSpray group, the incidence of adhesions (40%) and the adhesion extent (mean, 10%) were significantly lower than in the control group (incidence; p = 0.0147, adhesion extent; p = 0.0007). In the Seprafilm group, the incidence of adhesions was 70%. The adhesion extent of Seprafilm (mean, 30%) was significantly lower than in the control group (p = 0.0492). No significant differences were observed between the AdSpray and Seprafilm groups. As for histopathological examination, animals in the AdSpray group showed a similar healing profile to that of the control group without delayed healing and regeneration of mesothelial cells. In contrast, the Seprafilm group showed ongoing foreign body reaction to Seprafilm, and regeneration of mesothelial cells was immature at 7 days. CONCLUSIONS: Both the spray-type gel and sheet adhesion barriers significantly reduced adhesion formation after hepatectomy. The spray-type adhesion barrier caused no adverse events and induced favorable healing. These adhesion barriers may be effective in hepatectomy. Further animal studies and clinical trials are required to determine their benefits in clinical use.

Evaluation of a Spray-type, Novel Dextrin Hydrogel Adhesion Barrier Under Laparoscopic Conditions in a Porcine Uterine Horn Adhesion Model.

STUDY OBJECTIVE: To establish a porcine uterine horn adhesion model that mimicked laparoscopic procedures and use it to investigate the effect of a spray-type, novel dextrin hydrogel adhesion barrier (AdSpray; Terumo Corporation, Tokyo, Japan) on postsurgical adhesions. DESIGN: A single-blind randomized controlled trial (Canadian Task Force Classification I). SETTING: A Certified animal research facility. SUBJECTS: Sixteen female pigs. INTERVENTIONS: All animals underwent laparoscopically assisted adhesion-inducing surgery. The uterine horns and the peritoneum of the pelvic sidewall were injured. In the experimental group, AdSpray was applied to the injured site, and the handling of the sprayer was assessed. At 28 ± 1 days after surgery, animals were sacrificed, and adhesions at the injured site were evaluated. Uterine horn suture sites were examined under a light microscope to assess healing of the incised wound, the inflammatory reaction, abscess, and the foreign body reaction to the surgical suture. MEASUREMENTS AND MAIN RESULTS: The control group showed severe adhesions over the entire surface interface at the uterine horn suture sites and peritoneal resection site. Compared with the control treatment, AdSpray exhibited a higher percentage of adhesion-free sites (p < .001) and reduced the total adhesion score (p < .001). In the AdSpray group, no inflammation or abscess formation was observed on histopathological examination, and ideal healing of the suture sites was confirmed in all cases. CONCLUSION: Based on the results of the present study, the novel dextrin hydrogel shows excellent adhesion prevention and can be easily applied during laparoscopy using a dedicated sprayer.

Characterization of ESR active species on lithium chloride-modified mesoporous silica.

The surface of mesoporous silica with regular nanometer-sized pores and high surface area has been modified by metal ions or functional groups to introduce specific interactions. We found that ESR active species were formed on lithium chloride (LiCl)-modified mesoporous silica after heat treatment. The structure and the surface properties of LiCl-modified mesoporous silica were characterized by XRD, ESR, nitrogen adsorption, UV-vis-NIR, and TPD. The results suggest that the ESR active species were generated on the surface in response to heat treatment above 673 K. Moreover, it was found for the first time that LiCl-modified mesoporous silica after the heat treatment has reversible adsorption properties for hydrogen under room temperature and atmospheric pressure.