Tsukuba Memorial Hospital

Sepsis mortality due to drug-resistant organisms remains high. However, the effectiveness of empiric combination therapy for sepsis caused by potential drug-resistant gram‑negative bacilli remains debated. This was a nested case-control study using Japanese nationwide data from a medical reimbursement system between 2010 and 2017. The eligibility criteria were adult patients who were admitted to the intensive care unit, with new-onset sepsis at or after admission. We enrolled patients with combination therapy and monotherapy by propensity score matching using their baseline characteristics. We developed a logistic regression model including combination therapy, severity of organ dysfunction, and interaction terms as the adjusting variables. Overall, 7084 patients were included in this study, 3542 and 3542 of whom were enrolled in the combination therapy and monotherapy groups, respectively. The logistic regression model revealed combination therapy and moderate and severe organ dysfunction were significantly associated with in-hospital mortality (odds ratio [OR] 95% confidence interval [CI]: 1.52 [1.20-1.91], OR [95% CI]: 1.78 [1.43-2.23], and OR [95% CI]: 4.22 [2.61-6.81], respectively). Interaction between combination therapy and severe organ dysfunction was statistically significant (OR [95% CI]: 0.41 [0.21-0.80], P for interaction = .009). However, adjusted in-hospital mortality based on the model was not significantly different between the combination therapy and monotherapy groups among patients with severe organ dysfunction (44% [95% CI: 38-51] vs 47% [95% CI: 41-54]). Our results did not reveal any specific effectiveness of combination therapy; however, the significant interaction between combination therapy and severe organ dysfunction may indicate the possible survival benefit of combination therapy for severe sepsis subset.

BackgroundGastric cancer remains a major cause of cancer-related morbidity and mortality. Despite advances in surgical and perioperative care, prolonged hospitalization continues to strain healthcare systems. Predicting postoperative length of stay (LOS) could support personalized care and efficient resource allocation. Japan's nationwide Diagnosis Procedure Combination (DPC) database provides real-world data for large-scale analysis, but no study has applied machine learning to predict LOS after gastrectomy.MethodsThis retrospective study included 26,097 patients who underwent gastrectomy between 2017 and 2022 at 472 hospitals in Japan. Using XGBoost, we developed a predictive model based on 1,433 admission-time variables extracted from the DPC database. Model performance was evaluated using Root Mean Squared Error (RMSE) and Mean Absolute Error (MAE) in a five-fold cross-validation. SHAP values were used to interpret feature importance.ResultsThe final model achieved an RMSE of 3.74 and MAE of 2.82 days. Key predictors of LOS included surgical procedure (laparoscopic distal gastrectomy and open total gastrectomy), designated cancer hospital, hospital size, peritoneal dissemination, and admission ADL score. SHAP analysis revealed that Laparoscopic distal gastrectomy and higher hospital volume were associated with shorter LOS, while open total gastrectomy was associated with longer LOS.ConclusionsWe developed a machine learning model that predicts postoperative length of stay with an error range of 2–4 days using admission data. This proof-of-concept study demonstrates the feasibility of predicting length of stay from admission data, showing that explainable AI can replicate intuitive patterns in surgical oncology while simultaneously identifying unexpected insights from administrative data. These findings highlight the clinical potential of explainable AI for perioperative workflow optimization.

The authors declare no conflicts of interest. No new data were generated or analyzed in support of this research. All data discussed in this article are available in the cited references.

Patients with atrial fibrillation (AF) often present with symptoms similar to acute coronary syndrome (ACS), including chest pain and elevated levels of high-sensitivity cardiac troponin (hs-cTn). The 0/1-hour algorithm using hs-cTn is a rapid diagnostic tool endorsed by the European Society of Cardiology to rule out myocardial infarction (MI). However, because its effectiveness in patients with AF remains unclear, in this study we assessed the diagnostic accuracy of the 0/1-hour algorithm in patients with and without AF presenting with chest pain in the emergency department. We conducted a secondary analysis of the DROP-ACS cohort, including 1,333 patients from Japan and Taiwan, with AF in 10.3% of cases. We examined the algorithm's negative predictive value (NPV), sensitivity, positive predictive value (PPV), and specificity for ruling MI in or out. Patients with AF were more frequently placed in the observe group (54% vs. 34.9%, P<0.05) and less often in the rule-out group (24.1% vs. 44.6%, P<0.05). The NPV and sensitivity for ruling out MI were 100%, while the PPV and specificity were lower in patients with AF (60% and 89.7%, respectively). The 0/1-hour algorithm effectively ruled out MI in patients with AF, with high safety and accuracy. However, patients with AF are more likely to be stratified into the observe group, requiring further examination for final diagnosis.

The criteria for diagnosing sepsis-induced coagulopathy (SIC) may overlap with those of Japanese Association for Acute Medicine (JAAM) disseminated intravascular coagulation (DIC). This study determined if the diagnostic criteria of SIC overlap with JAAM DIC diagnostic criteria for identifying patients with DIC according to the International Society on Thrombosis and Haemostasis (ISTH) criteria and whether the patients diagnosed with these criteria have the same prognosis.This multicenter retrospective study included patients with sepsis diagnosed using the JAAM and ISTH DIC and SIC criteria on days 1 and 4. The established ISTH DIC criteria was the reference standard for primary outcome that compared the characteristics of SIC and JAAM DIC. Secondary outcomes were multiple organ dysfunction syndrome (MODS), ventilator-free and intensive care unit-free days, and in-hospital mortality.A total of 1,438 patients were included in this study. On day 1, the JAAM DIC and SIC criteria diagnosed almost all patients with ISTH DIC (98 and 94%, respectively), predicting ISTH DIC (area under the receiver operating curve [AUC]: 0.740 versus 0.752, p = 0.523) and MODS (AUC: 0.686 versus 0.697, p = 0.546) on day 4 and progressing to ISTH DIC in the same proportion (28.6 versus 30.1%, p = 0.622). There were no differences in survival probabilities (p = 0.196) or secondary outcomes between patients diagnosed using JAAM DIC and SIC criteria on day 1.SIC and JAAM DIC diagnoses were equal among patients with sepsis, suggesting that SIC criteria add little to current DIC scoring systems.

Background Achieving a balance between the utilization and limitation of carbapenems for initial sepsis therapy is imperative, yet documentation on the use of carbapenems in sepsis treatment is limited. This study aimed to ascertain the prevalence of carbapenem use in Japanese intensive care units (ICUs) and evaluate the association between its use in the initial antibiotic therapy for sepsis and patient prognosis. Methods This study is a post hoc descriptive analysis of the Focused Outcome Research on Emergency Care for Acute Respiratory Distress Syndrome, Sepsis, and Trauma (FORECAST), a multicenter, prospective cohort study conducted in 59 ICUs in Japan from January 2016 to March 2017. This study described the rate of carbapenem use overall and in each ICU and assessed the association between carbapenem use and in-hospital mortality by generalized linear mixed effect model adjusting for patient characteristics as fixed effect confounders and the hospitals as random effect confounders. Results Out of 1140 participants, 627(55.0%) received and 513 (45.0%) did not receive carbapenems following the diagnosis of severe sepsis or septic shock. Patient severity was similar in both groups in terms of Sepsis-related Organ Failure Assessment (SOFA; 9 and 8) and Acute Physiology and Chronic Health Evaluation II (APACHE II; 23 and 22) scores. Among 48 of the 59 ICUs that registered more than three patients in the FORECAST registry, the median carbapenem utilization rate stood at 55.2% (minimum: 20.0%, maximum: 100.0%, IQR: 43.6%-67.2%). Hospital mortality rates were 25.6% and 20.5% in carbapenem recipients and non-recipients, respectively. A correlation between initial carbapenem use and increased in-hospital mortality was discerned in the unadjusted model (difference of 5.2%), but this association was not evident in the adjusted model (difference of 1.2%; 95%CI: -4.3,6.6; p=0.66). Conclusions Initial antibiotic therapy with carbapenems was noted in over half of the sepsis patients in Japanese ICUs. In-hospital mortality was not associated with the use of carbapenems.

IntroductionThe purposes of this study were to clarify whether non-small cell lung cancer (NSCLC) patients with epidermal growth factor receptor (EGFR) gene mutation have a higher incidence of cancer-associated venous and arterial thrombosis/thromboses (CATs) than EGFR-negative patients, and whether patients who develop CATs have a significantly poorer survival.Material and methodsThe relationship between EGFR mutation and the development of CATs was evaluated retrospectively in 1,891 patients with NSCLC, including 381 who were EGFR-positive. Propensity matching and logistic regression analysis were performed.ResultsAmong 1,891 patients with NSCLC, 37 patients (2.0%) developed CATs. After propensity matching, the incidence of CATs was 3.9% in EGFR-positive patients and 0.9% in EGFR-negative patients (p = 0.011). Therefore, the risk of developing CATs was higher in EGFR-positive than EGFR-negative patients. In EGFR-positive patients, the most common time to develop CATs was at the time of lung cancer diagnosis, but there were some patients who developed CATs during second-line treatment or later when EGFR-tyrosine kinase inhibitors were no longer effective. In advanced EGFR-positive patients, there was no significant difference in overall survival depending on the presence or absence of CATs, but the development of CATs impaired the long survival expected in EGFR-positive patients.ConclusionsEGFR-positive patients are currently treated with tyrosine kinase inhibitors and are expected to have a certain long-term prognosis. However, there is concern that overall survival might be shortened if CATs develop. Therefore, the possibility of developing CATs in EGFR-positive NSCLC patients must be addressed.