Eiji Kawamoto

Sepsis is a life-threatening syndrome characterized by dysregulated systemic inflammation and multiple organ dysfunction, yet the molecular mechanisms underlying its progression remain incompletely understood. Our previous studies demonstrated that microRNA-511-3p (miR-511-3p) is upregulated in both an experimental sepsis model and patients with sepsis, suggesting its potential role in the pathogenesis of sepsis-associated inflammation. Because patched 1 (PTCH1), a validated target of miR-511-3p, serves as a key negative regulator of hedgehog (HH) signaling, we hypothesized that miR-511-3p promotes inflammatory responses during sepsis by suppressing PTCH1 expression and consequently activating HH signaling. To investigate this mechanism, we examined multiple immune cell populations, including small intestinal epithelial cells (S-IECs), small intestinal lamina propria macrophages (SI-LPMs), and liver macrophages (Liv-Ms), in a lipopolysaccharide (LPS)-induced murine sepsis model, as these cell populations may contribute to both local intestinal and systemic inflammatory responses during sepsis. The systemic administration of a miR-511-3p inhibitor (511-Inh) significantly reduced the expression of the proinflammatory cytokines interleukin 6 (IL-6) and tumor necrosis factor (TNF), as well as Hedgehog-interacting protein (HHIP), a canonical HH target gene. The inhibition of miR-511-3p restored PTCH1 expression in the intestines, liver, and lung and increased its mRNA levels in S-IECs, SI-LPMs, and Liv-Ms. These changes were accompanied by a reduced expression of HH pathway-related genes. Furthermore, miR-511-3p inhibition attenuated sepsis-induced liver injury, as evidenced by improved histopathological findings and decreased plasma levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT). To further determine the contribution of HH signaling to sepsis-associated inflammation, septic mice were treated with vismodegib, a selective inhibitor of smoothened (SMO). The pharmacological inhibition of SMO considerably reduced the expression of Smo, Gli, Hhip, Il6, and Tnf in the small intestine, large intestine, liver, and lung, confirming that aberrant HH pathway activation amplifies inflammatory responses during sepsis. Collectively, these findings suggest that miR-511-3p drives sepsis-associated inflammation, at least in part, by suppressing PTCH1 expression and thereby promoting HH signaling. Conversely, inhibiting miR-511-3p restores PTCH1 expression, dampens HH pathway activity, and mitigates systemic inflammation and sepsis-induced organ injury. Thus, the miR-511-3p/PTCH1/HH signaling axis may represent a promising therapeutic target for sepsis.

Background: The incidence of delirium in patients with burns receiving treatment in the intensive care unit (ICU) is high, reaching up to 77%, and has been associated with increased mortality rates. Therefore, early identification of patients at high risk of delirium onset is essential for improving treatment strategies. Objective: This study aimed to create a machine learning model for predicting delirium in patients with burns during their ICU stay using patient data from the first day of ICU admission and identify predictive factors for ICU delirium in patients with burns. Methods: This study focused on 82 patients with burns aged ≥18 years who were admitted to the ICU at Mie University Hospital for ≥24 hours between January 2015 and June 2023. In total, 70 variables were measured in patients upon ICU admission and used as explanatory variables in the ICU delirium prediction model. Delirium was assessed using the Intensive Care Delirium Screening Checklist every 8 hours after ICU admission. A total of 10 different machine learning methods were used to predict ICU delirium. Multiple receiver operating characteristic curves were plotted for various machine learning models, and the area under the curve (AUC) for each was compared. In addition, the top 15 risk factors contributing to delirium onset were identified using Shapley additive explanations analysis. Results: Among the 10 machine learning models tested, logistic regression (mean AUC 0.906, SD 0.073), support vector machine (mean AUC 0.897, SD 0.056), k-nearest neighbor (mean AUC 0.894, SD 0.060), neural network (mean AUC 0.857, SD 0.058), random forest (mean AUC 0.850, SD 0.074), adaptive boosting (mean AUC 0.832, SD 0.094), gradient boosting machine (mean AUC 0.821, SD 0.074), and naïve Bayes (mean AUC 0.827, SD 0.095) demonstrated the highest accuracy in predicting ICU delirium. Specifically, 24-hour urine output (from ICU admission to 24 hours), oxygen saturation, burn area, total bilirubin level, and intubation upon ICU admission were identified as the major risk factors for delirium onset. In addition, variables, such as the proportion of white blood cell fractions, including monocytes; methemoglobin concentration; and respiratory rate, were identified as important risk factors for ICU delirium. Conclusions: This study demonstrated the ability of machine learning models trained using vital signs and blood data upon ICU admission to predict delirium in patients with burns during their ICU stay.

Background: Pain inhibits rehabilitation. In rehabilitation at medical institutions, the usefulness of virtual reality (VR) has been reported in many cases to alleviate pain. In recent years, the demand for home rehabilitation has increased. Unlike in medical situations, the patients targeted for in-home rehabilitation often have chronic pain due to physical and psychosocial factors, and the environment is not specialized for rehabilitation. However, VR might be effective for in-home rehabilitation settings. Objective: This study aims to evaluate the feasibility of applying VR to home-visit rehabilitation for homebound patients with chronic pain. Methods: This study will test the feasibility of VR applied to home-visit rehabilitation for patients with chronic pain. A single-arm pre-post comparison will be conducted to evaluate its feasibility. Screening will be conducted on patients who have given consent to participate in the study, and those who have pain that persists or recurs for more than 3 months and receive home-visit rehabilitation will be enrolled in the study. Baseline measurements will be conducted on study participants before the start of the VR intervention. VR-applied home-visit rehabilitation will be conducted once a week for a total of 10 VR interventions. The primary endpoint is the change in pain from the baseline to the tenth intervention. Pain is a subjective symptom of the study participants and will be subjectively assessed by the Numerical Rating Scale of 11 levels from 0 to 10. Pain as the primary endpoint will be measured at 3-time points per rehabilitation session: before, during, and after the rehabilitation so that changes between time points can be evaluated. Secondary endpoints are heart rate variability, range of motion of the area in the musculoskeletal system where the pain occurs, motivation for rehabilitation, catastrophic thoughts of pain, mood state, quality of life, and interviews. Assessments will be conducted at the baseline, first, fifth, and tenth interventions. After completing the clinical study (10 VR interventions), patients will continue their regular home-visit rehabilitation as usual. Results: Recruitment of participants began on February 22, 2022, and data collection is ongoing as of November 2024. The research results will be published in international peer-reviewed journals and through presentations at national and international conferences. Conclusions: This study will contribute to the development of novel rehabilitation-based solutions for homebound patients who have had difficulty obtaining adequate relief from chronic pain. Future studies will consider conducting randomized controlled trials as clinical trials to validate the efficacy of VR during home-visit rehabilitation for patients with chronic pain. International Registered Report Identifier (IRRID): DERR1-10.2196/58734

BACKGROUND The incidence of delirium in patients with burns receiving treatment in the intensive care unit (ICU) is high, reaching up to 77%, and has been associated with increased mortality rates. Therefore, early identification of patients at high risk of delirium onset is essential for improving treatment strategies. OBJECTIVE This study aimed to create a machine learning model for predicting delirium in patients with burns during their ICU stay using patient data from the first day of ICU admission and identify predictive factors for ICU delirium in patients with burns. METHODS This study focused on 82 patients with burns aged ≥18 years who were admitted to the ICU at Mie University Hospital for ≥24 hours between January 2015 and June 2023. In total, 70 variables were measured in patients upon ICU admission and used as explanatory variables in the ICU delirium prediction model. Delirium was assessed using the Intensive Care Delirium Screening Checklist every 8 hours after ICU admission. A total of 10 different machine learning methods were used to predict ICU delirium. Multiple receiver operating characteristic curves were plotted for various machine learning models, and the area under the curve (AUC) for each was compared. In addition, the top 15 risk factors contributing to delirium onset were identified using Shapley additive explanations analysis. RESULTS Among the 10 machine learning models tested, logistic regression (mean AUC 0.906, SD 0.073), support vector machine (mean AUC 0.897, SD 0.056), k-nearest neighbor (mean AUC 0.894, SD 0.060), neural network (mean AUC 0.857, SD 0.058), random forest (mean AUC 0.850, SD 0.074), adaptive boosting (mean AUC 0.832, SD 0.094), gradient boosting machine (mean AUC 0.821, SD 0.074), and naïve Bayes (mean AUC 0.827, SD 0.095) demonstrated the highest accuracy in predicting ICU delirium. Specifically, 24-hour urine output (from ICU admission to 24 hours), oxygen saturation, burn area, total bilirubin level, and intubation upon ICU admission were identified as the major risk factors for delirium onset. In addition, variables, such as the proportion of white blood cell fractions, including monocytes; methemoglobin concentration; and respiratory rate, were identified as important risk factors for ICU delirium. CONCLUSIONS This study demonstrated the ability of machine learning models trained using vital signs and blood data upon ICU admission to predict delirium in patients with burns during their ICU stay.

Background: Intensive care unit‐acquired weakness (ICU‐AW) is a syndrome characterized by a long‐term muscle weakness often observed in sepsis‐surviving patients during the chronic phase. Although ICU‐AW is independently associated with increased mortality, effective therapies have yet to be established. Programmed death‐1 (PD‐1) inhibitors have attracted attention as potential treatments for reversing immune exhaustion in sepsis; however, its impact on ICU‐AW remains to be elucidated. Here, we study how PD‐1 deficiency affects sepsis‐induced skeletal muscle dysfunction in a preclinical sepsis model.Methods: Chronic sepsis model was developed by treating wild‐type (WT) and PD‐1 knockout (KO) mice with caecal slurry, followed by resuscitation with antibiotics and saline. Mice were euthanized on days 15–17. Body weights, muscle weights, and limb muscle strengths were measured. Interleukin 13 (IL‐13) and PD‐1 expressions were examined by flow cytometry. Messenger RNA (mRNA) expressions of slow‐twitch muscles were measured by reverse transcription and quantitative polymerase chain reaction (RT‐qPCR). In an in vitro study, C2C12 myotubes were treated with lipopolysaccharide (LPS) and recombinant IL‐13 followed by gene expression measurements.Results: WT septic mice exhibited decreased muscle weight (quadriceps, P < 0.01; gastrocnemius, P < 0.05; and tibialis anterior, P < 0.01) and long‐term muscle weakness (P < 0.0001), whereas PD‐1 KO septic mice did not exhibit any reduction in muscle weights and strengths. Slow‐twitch specific mRNAs, including myoglobin (Mb), troponin I type 1 (Tnni1), and myosin heavy chain 7 (Myh7) were decreased in WT skeletal muscle (Mb, P < 0.0001; Tnni1, P < 0.05; and Myh7, P < 0.05) after sepsis induction, but mRNA expressions of Tnni1 and Myh7 were increased in PD‐1 KO septic mice (Mb, not significant; Tnni1, P < 0.0001; and Myh7, P < 0.05). Treatment of C2C12 myotube cells with LPS decreased the expression of slow‐twitch mRNAs, which was restored by IL‐13 (Mb, P < 0.0001; Tnni1, P < 0.001; and Myh7, P < 0.05). IL‐13 production was significantly higher in ILC2s compared to T cells in skeletal muscle (P < 0.05). IL‐13‐producing ILC2s in skeletal muscle were examined and found to increase in PD‐1 KO septic mice, compared with WT septic mice (P < 0.05). ILC2‐derived IL‐13 was increased by PD‐1 KO septic mice and thought to protect the muscles from experimental ICU‐AW.Conclusions: Long‐term muscle weakness in experimental ICU‐AW was ameliorated in PD‐1 KO mice. ILC2‐derived IL‐13 production in skeletal muscles was increased in PD‐1 KO mice, thereby suggesting that IL‐13 alleviates muscle weakness during sepsis. This study demonstrates the effects of PD‐1 blockade in preserving muscle strength during sepsis through an increase in ILC2‐derived IL‐13 and may be an attractive therapeutic target for sepsis‐induced ICU‐AW.

BACKGROUND Pain inhibits rehabilitation. In rehabilitation at medical institutions, the usefulness of virtual reality (VR) has been reported in many cases to alleviate pain. In recent years, the demand for home rehabilitation has increased. Unlike in medical situations, the patients targeted for in-home rehabilitation often have chronic pain due to physical and psychosocial factors, and the environment is not specialized for rehabilitation. However, VR might be effective for in-home rehabilitation settings. OBJECTIVE This study aims to evaluate the feasibility of applying VR to home-visit rehabilitation for homebound patients with chronic pain. METHODS This study will test the feasibility of VR applied to home-visit rehabilitation for patients with chronic pain. A single-arm pre-post comparison will be conducted to evaluate its feasibility. Screening will be conducted on patients who have given consent to participate in the study, and those who have pain that persists or recurs for more than 3 months and receive home-visit rehabilitation will be enrolled in the study. Baseline measurements will be conducted on study participants before the start of the VR intervention. VR-applied home-visit rehabilitation will be conducted once a week for a total of 10 VR interventions. The primary endpoint is the change in pain from the baseline to the tenth intervention. Pain is a subjective symptom of the study participants and will be subjectively assessed by the Numerical Rating Scale of 11 levels from 0 to 10. Pain as the primary endpoint will be measured at 3-time points per rehabilitation session: before, during, and after the rehabilitation so that changes between time points can be evaluated. Secondary endpoints are heart rate variability, range of motion of the area in the musculoskeletal system where the pain occurs, motivation for rehabilitation, catastrophic thoughts of pain, mood state, quality of life, and interviews. Assessments will be conducted at the baseline, first, fifth, and tenth interventions. After completing the clinical study (10 VR interventions), patients will continue their regular home-visit rehabilitation as usual. RESULTS Recruitment of participants began on February 22, 2022, and data collection is ongoing as of November 2024. The research results will be published in international peer-reviewed journals and through presentations at national and international conferences. CONCLUSIONS This study will contribute to the development of novel rehabilitation-based solutions for homebound patients who have had difficulty obtaining adequate relief from chronic pain. Future studies will consider conducting randomized controlled trials as clinical trials to validate the efficacy of VR during home-visit rehabilitation for patients with chronic pain. INTERNATIONAL REGISTERED REPORT DERR1-10.2196/58734

Background: Effective communication strategies are becoming increasingly important in intensive care units (ICUs) where patients at high risk are treated. Distributed leadership promotes effective communication among health care professionals (HCPs). Moreover, beyond facilitating patient care, it may improve well-being among HCPs by fostering teamwork. However, the impact of distributed leadership on the communication structure and well-being of HCPs remains unclear. Objective: We performed a social network analysis (SNA) to assess the characteristics of each HCP in the network, identify the number of HCP connections, analyze 4 centralities that can measure an HCP’s importance, and evaluate the impact of distributed leadership structure on the well-being and communication structure of the medical staff. Methods: Wearable sensors were used to obtain face-to-face interaction data from the ICU medical staff at Mie University Hospital, Japan. Participants wore a badge on the front of their clothing during working hours to measure the total frequency of face-to-face interactions. We collected data about the well-being of medical staff using the Center for Epidemiological Studies–Depression (CES-D) questionnaire and measured 4 centralities using SNA analysis. A CES-D questionnaire was administered during the study to measure the well-being of the HCPs. Results: Overall, 247 ICU workers participated in this clinical study for 4 weeks yearly in February 2016, 2017, and 2018. The distributed leadership structure was established within the ICU in 2017 and 2018. We compared these results with those of the traditional leadership structure used in 2016. Most face-to-face interactions in the ICU were among nurses or between nurses and other professionals. In 2016, overall, 10 nurses could perform leadership tasks, which significantly increased to 24 in 2017 (P=.046) and 20 in 2018 (P=.046). Considering the increased number of nurses who could perform leadership duties and the collaboration created within the organization, SNA in 2018 showed that the betweenness (P=.001), degree (P<.001), and closeness (P<.001) centralities significantly increased compared with those in 2016. However, the eigenvector centrality significantly decreased in 2018 compared with that in 2016 (P=.01). The CES-D scores in 2018 also significantly decreased compared with those in 2016 (P=.01). The betweenness (r=0.269; P=.02), degree (r=0.262; P=.03), and eigenvector (r=0.261; P=.03) centralities and CES-D scores were positively correlated in 2016, whereas the closeness centrality and CES-D scores were negatively correlated (r=−0.318; P=.01). In 2018, the degree (r=−0.280; P=.01) and eigenvector (r=−0.284; P=.01) centralities were negatively correlated with CES-D scores. Conclusions: Face-to-face interactions of HCPs in the ICU were measured using wearable sensors, and nurses were found to be centrally located. However, the introduction of distributed leadership created collaboration and informal leadership in the organization, altering the social network structure of HCPs and increasing organizational well-being. Trial Registration: University Hospital Medical Information Network (UMIN) UMIN000037046; https://center6.umin.ac.jp/cgi-open-bin/icdr_e/ctr_view.cgi?recptno=R000042211

Background: Shift workers are at high risk of developing sleep disorders such as shift worker sleep disorder or chronic insomnia. Cognitive behavioral therapy (CBT) is the first-line treatment for insomnia, and emerging evidence shows that internet-based CBT is highly effective with additional features such as continuous tracking and personalization. However, there are limited studies on internet-based CBT for shift workers with sleep disorders. Objective: This study aimed to evaluate the impact of a 4-week, physician-assisted, internet-delivered CBT program incorporating machine learning–based well-being prediction on the sleep duration of shift workers at high risk of sleep disorders. We evaluated these outcomes using an internet-delivered CBT app and fitness trackers in the intensive care unit. Methods: A convenience sample of 61 shift workers (mean age 32.9, SD 8.3 years) from the intensive care unit or emergency department participated in the study. Eligible participants were on a 3-shift schedule and had a Pittsburgh Sleep Quality Index score ≥5. The study comprised a 1-week baseline period, followed by a 4-week intervention period. Before the study, the participants completed questionnaires regarding the subjective evaluation of sleep, burnout syndrome, and mental health. Participants were asked to wear a commercial fitness tracker to track their daily activities, heart rate, and sleep for 5 weeks. The internet-delivered CBT program included well-being prediction, activity and sleep chart, and sleep advice. A job-based multitask and multilabel convolutional neural network–based model was used for well-being prediction. Participant-specific sleep advice was provided by sleep physicians based on daily surveys and fitness tracker data. The primary end point of this study was sleep duration. For continuous measurements (sleep duration, steps, etc), the mean baseline and week-4 intervention data were compared. The 2-tailed paired t test or Wilcoxon signed rank test was performed depending on the distribution of the data. Results: In the fourth week of intervention, the mean daily sleep duration for 7 days (6.06, SD 1.30 hours) showed a statistically significant increase compared with the baseline (5.54, SD 1.36 hours; P=.02). Subjective sleep quality, as measured by the Pittsburgh Sleep Quality Index, also showed statistically significant improvement from baseline (9.10) to after the intervention (7.84; P=.001). However, no significant improvement was found in the subjective well-being scores (all P>.05). Feature importance analysis for all 45 variables in the prediction model showed that sleep duration had the highest importance. Conclusions: The physician-assisted internet-delivered CBT program targeting shift workers with a high risk of sleep disorders showed a statistically significant increase in sleep duration as measured by wearable sensors along with subjective sleep quality. This study shows that sleep improvement programs using an app and wearable sensors are feasible and may play an important role in preventing shift work–related sleep disorders. International Registered Report Identifier (IRRID): RR2-10.2196/24799.

BACKGROUND Effective communication strategies are becoming increasingly important in intensive care units (ICUs) where patients at high risk are treated. Distributed leadership promotes effective communication among health care professionals (HCPs). Moreover, beyond facilitating patient care, it may improve well-being among HCPs by fostering teamwork. However, the impact of distributed leadership on the communication structure and well-being of HCPs remains unclear. OBJECTIVE We performed a social network analysis (SNA) to assess the characteristics of each HCP in the network, identify the number of HCP connections, analyze 4 centralities that can measure an HCP’s importance, and evaluate the impact of distributed leadership structure on the well-being and communication structure of the medical staff. METHODS Wearable sensors were used to obtain face-to-face interaction data from the ICU medical staff at Mie University Hospital, Japan. Participants wore a badge on the front of their clothing during working hours to measure the total frequency of face-to-face interactions. We collected data about the well-being of medical staff using the Center for Epidemiological Studies–Depression (CES-D) questionnaire and measured 4 centralities using SNA analysis. A CES-D questionnaire was administered during the study to measure the well-being of the HCPs. RESULTS Overall, 247 ICU workers participated in this clinical study for 4 weeks yearly in February 2016, 2017, and 2018. The distributed leadership structure was established within the ICU in 2017 and 2018. We compared these results with those of the traditional leadership structure used in 2016. Most face-to-face interactions in the ICU were among nurses or between nurses and other professionals. In 2016, overall, 10 nurses could perform leadership tasks, which significantly increased to 24 in 2017 (P=.046) and 20 in 2018 (P=.046). Considering the increased number of nurses who could perform leadership duties and the collaboration created within the organization, SNA in 2018 showed that the betweenness (P=.001), degree (P<.001), and closeness (P<.001) centralities significantly increased compared with those in 2016. However, the eigenvector centrality significantly decreased in 2018 compared with that in 2016 (P=.01). The CES-D scores in 2018 also significantly decreased compared with those in 2016 (P=.01). The betweenness (r=0.269; P=.02), degree (r=0.262; P=.03), and eigenvector (r=0.261; P=.03) centralities and CES-D scores were positively correlated in 2016, whereas the closeness centrality and CES-D scores were negatively correlated (r=−0.318; P=.01). In 2018, the degree (r=−0.280; P=.01) and eigenvector (r=−0.284; P=.01) centralities were negatively correlated with CES-D scores. CONCLUSIONS Face-to-face interactions of HCPs in the ICU were measured using wearable sensors, and nurses were found to be centrally located. However, the introduction of distributed leadership created collaboration and informal leadership in the organization, altering the social network structure of HCPs and increasing organizational well-being. CLINICALTRIAL University Hospital Medical Information Network (UMIN) UMIN000037046; https://center6.umin.ac.jp/cgi-open-bin/icdr_e/ctr_view.cgi?recptno=R000042211