Severe pneumonia (SP) is a respiratory tract disease that seriously threatens human health. The herpesvirus detected in patients, especially with severe and immunodeficient diseases, is gradually attracting the attention of clinical doctors. However, little is known about the effect of herpesvirus on the prognosis of SP patients and the pulmonary microbial community. Here, we retrospectively analyzed respiratory samples from 45 SP patients detected by metagenomic next-generation sequencing (mNGS). A total of five types of herpesviruses were detected, with Human alphaherpesvirus 1 (HHV-1) in 19 patients, Human betaherpesvirus 5 (CMV) in 7 patients, Human betaherpesvirus 7 (HHV-7) in 6 patients, Human alphaherpesvirus 2 (HHV-2) in 5 patients, and Human gammaherpesvirus 4 (EBV) in 4 patients. Further analysis showed that the mortality of the herpesvirus-positive group was significantly higher than that of the negative group. The results also showed that HHV-1 was significantly associated with the prognosis of SP patients, while the other herpesviruses did not have a significant difference in patient mortality. A comparison of the microbial community characteristics of SP patients showed a significant difference in beta-diversity between herpesvirus-positive and negative groups. Species difference analysis showed that the herpesvirus-positive group was related to more conditional pathogens, such as Pneumocystis jirovecii and Burkholderia cepacia. In summary, our results suggest that the presence of herpesvirus is associated with the mortality of SP patients. Furthermore, enrichment of conditional pathogens in the respiratory tract of herpesvirus-positive SP patients may be a potential reason for the increased mortality.
Background: Complicated intra-abdominal infections (cIAIs) in the abdominal cavity or within an abdominal organ are numerous and frequent dangerous entities in the treatment of critically ill patients. Early clinical evaluation is necessary.Methods: This retrospective multicenter study included patients from 10 intensive care units (ICUs). Risk factors for the overall survival (OS) of patients with cIAI were selected using least absolute shrinkage and selection operator regression, and a nomogram was constructed subsequently. Calibration curve and receiver operating characteristic (ROC) curve were used to evaluate the calibration and discriminative ability.Results: In total, 544 patients diagnosed with cIAI were enrolled and divided into the study (n = 276) and validation (n = 268) sets. Sex, acute gastrointestinal injury, acute kidney injury, rare bacterium infection, Charlson score, and APACHE II score were identified as independent risk factors and were constructed for the nomogram. The nomogram showed marked calibration capability with a concordance index (C-index) of 0.909 and 0.831 in the study and validation set, respectively. Compared with the common clinical prognostic scoring system, the nomogram achieved the highest discrimination ability with an area under the curve (AUC) value of 0.91 and 0.83 in the study set and validation set, respectively.Conclusions: Our newly constructed nomogram provides a useful tool for risk stratification and prognosis evaluation of cIAI.
Sepsis remains a major cause of mortality in ICU patients, requiring accurate prognostic tools for optimal management. This study aimed to develop and validate an interpretable machine learning-based nomogram for predicting in-hospital mortality in sepsis patients to guide clinical decision-making. This retrospective cohort study included 407 adult sepsis patients from ICU admissions between January 2019 and December 2024. Patients were randomly divided into training (n = 284) and validation (n = 123) cohorts. A LASSO-based multivariate logistic regression was applied to construct a predictive nomogram. Seven independent predictors emerged for in-hospital mortality: age, abdominal infection, vasopressor requirement, WBC, BNP, APACHE II score, and mechanical ventilation support. The nomogram showed strong discrimination with an AUC of 0.900 in the training cohort and 0.796 in the validation cohort. Calibration curves, decision curve analysis, and SHAP interpretation confirmed good model performance and clinical utility. The novel machine learning-based nomogram provides a practical, interpretable risk assessment tool for early mortality prediction, potentially improving patient outcomes through enhanced clinical understanding and timely interventions in critically ill sepsis patients.