Prediction of fatal adverse prognosis in patients with fever-related diseases based on machine learning: A retrospective study.

BACKGROUND: Fever is the most common chief complaint of emergency patients. Early identification of patients at an increasing risk of death may avert adverse outcomes. The aim of this study was to establish an early prediction model of fatal adverse prognosis of fever patients by extracting key indicators using big data technology. METHODS: A retrospective study of patients' data was conducted using the Emergency Rescue Database of Chinese People's Liberation Army General Hospital. Patients were divided into the fatal adverse prognosis group and the good prognosis group. The commonly used clinical indicators were compared. Recursive feature elimination (RFE) method was used to determine the optimal number of the included variables. In the training model, logistic regression, random forest, adaboost and bagging were selected. We also collected the emergency room data from December 2018 to December 2019 with the same inclusion and exclusion criterion. The performance of the model was evaluated by accuracy, F1-score, precision, sensitivity and the areas under receiver operator characteristic curves (ROC-AUC). RESULTS: The accuracy of logistic regression, decision tree, adaboost and bagging was 0.951, 0.928, 0.924, and 0.924, F1-scores were 0.938, 0.933, 0.930, and 0.930, the precision was 0.943, 0.938, 0.937, and 0.937, ROC-AUC were 0.808, 0.738, 0.736, and 0.885, respectively. ROC-AUC of ten-fold cross-validation in logistic and bagging models were 0.80 and 0.87, respectively. The top six coefficients and odds ratio (OR) values of the variables in the Logistic regression were cardiac troponin T (CTnT) (coefficient=0.346, OR = 1.413), temperature (T) (coefficient=0.235, OR = 1.265), respiratory rate (RR) (coefficient= -0.206,OR = 0.814), serum kalium (K) (coefficient=0.137, OR = 1.146), pulse oxygen saturation (SPO2) (coefficient= -0.101, OR = 0.904), and albumin (ALB) (coefficient= -0.043, OR = 0.958). The weights of the top six variables in the bagging model were: CTnT, RR, lactate dehydrogenase, serum amylase, heartrate, and systolic blood pressure. CONCLUSIONS: The main clinical indicators of concern included CTnT, RR, SPO2, T, ALB and K. The bagging model and logistic regression model had better diagnostic performance comprehesively. Those may be conducive to the early identification of critical patients with fever by physicians.

Clinical analysis of 215 consecutive cases with fever of unknown origin: A cohort study.

Fever of unknown origin (FUO) is a frequently observed phenomenon in clinical practice. The present study was aimed to investigate potential causes of FUO, thereby improving clinical diagnosis of this disorder.In this retrospective study, clinical data were collected from 215 patients who were diagnosed with FUO between January 2009 and December 2010, and an 18 to 36 months follow-up visit was also performed for these patients.Among these FUO cases, the most common causes of the disease were infectious diseases (IDs) (42.3%), followed by connective tissue diseases (CTDs) (32.1%), miscellaneous (Mi) (10.7%) and neoplasm (N) (6.5%), while the causes for the other 18 cases (8.4%) were still unknown. The most common types of ID, CTD, and N were tuberculosis (16/91, 17.6%), adult onset Still disease (AOSD) (37/69, 53.6%) and non-Hodgkin lymphoma (6/14, 42.9%), respectively.IDs still represent the most common causes of FUO. Regularly intermittent fever with urinary infections and irregularly intermittent fever with infective endocarditis may be regarded as some signs in clinical diagnosis of FUO.

Perinatal outcomes and congenital abnormalities in the newborns of women affected by the 2009 pandemic influenza A (H1N1) in Beijing, China.

OBJECTIVE: To investigate the pregnancy complications, perinatal outcomes, and congenital abnormalities (CAs) that occurred in Beijing, China, when pregnant women became infected with the 2009 pandemic influenza A (H1N1) (H1N1 pdm). METHODS: Pregnancy complications, perinatal outcomes, and CAs were compared among 3 groups of pregnant women. The 23 women in group 1 were confirmed to harbor viral RNA; the 23 in group 2 had serum levels of virus-specific antibodies against H1N1 pdm, meaning that they were suspected of being infected with the virus; and the 93 in group 3 had no detectable virus-specific antibodies. RESULTS: Perinatal outcomes and pregnancy complications were not significantly different in groups 1 and 3. Higher percentages of stillbirths (12.0%) and placental disorders (13.0%) were observed in group 2 than in group 3. Many women in group 2 (62.5%) experienced symptoms of having a cold during pregnancy and most took no medication. Two cases of CA occurred in group 1, in the offspring of women infected in the second trimester. CONCLUSION: When left untreated, infection with the 2009 H1N1 pdm virus during pregnancy appears to have increased fetal mortality and morbidity. Because CAs are traumatic for all concerned, their possible association with the virus should be further evaluated.

Characteristics of pediatric scrub typhus in a new endemic region of northern China.

BACKGROUND: Scrub typhus emerged in northern China in 1986. Our objective was to document epidemiologic, clinical and laboratory features of pediatric scrub typhus in this new endemic area. METHODS: The pediatric patients diagnosed with scrub typhus during the 12-year period from 1995 through 2006 in Feixian County, Shandong province were enrolled in the study. The cases were diagnosed based on either specific antibody detection using the indirect immunofluorescent assay or detection of partial Orientia tsutsugamushi gene by polymerase chain reaction. RESULTS: Seventy pediatric scrub typhus cases were included in the study. The cases occurred from September through November. The common clinical manifestations included headache (100%), skin rash (91%), eschar (84%), lymphadenopathy (61%), and gastrointestinal signs (56%). None of the patients had neurologic involvement or thrombocytopenia. All children responded well to treatment with chloramphenicol. The patients who presented with abnormal chest radiography took significantly longer time to defervescence (Z[r] = 2.528, P = 0.011). Three strains of O. tsutsugamushi were isolated and all were identified as Kawasaki type. CONCLUSIONS: The manifestations of pediatric scrub typhus cases in the novel endemic region of northern China may be less severe than in other regions. Careful examination of skin eschars is helpful for the clinical diagnosis.

Clinical characteristics of the autumn-winter type scrub typhus cases in south of Shandong province, northern China.

BACKGROUND: Before 1986, scrub typhus was only found endemic in southern China. Because human infections typically occur in the summer, it is called "summer type". During the autumn-winter period of 1986, a new type of scrub typhus was identified in Shandong and northern Jiangsu province of northern China. This newly recognized scrub typhus was subsequently reported in many areas of northern China and was then called "autumn-winter type". However, clinical characteristics of associated cases have not been reported. METHODS: From 1995 to 2006, all suspected scrub typhus cases in five township hospitals of Feixian county, Shandong province were enrolled. Indirect immunofluorescent assay (IFA) was used as confirmatory serodiagnosis test. Polymerase chain reaction (PCR) connected with restriction fragment length polymorphism (RFLP) and sequence analyses were used for genotyping of O. tsutsugamushi DNAs. Clinical symptoms and demography of confirmed cases were analyzed. RESULTS: A total of 480 scrub typhus cases were confirmed. The cases occurred every year exclusively between September and December with a peak occurrence in October. The case numbers were relatively higher in 1995, 1996, 1997, and 2000 than in other years. 57.9% of cases were in the group aged 21-50. More cases occurred in male (56%) than in female (44%). The predominant occupational group of the cases was farmers (85.0%). Farm work was reported the primary exposure to infection in 67.7% of cases. Fever, rash, and eschar were observed in 100.0%, 90.4%, and 88.5% of cases, respectively. Eschars formed frequently on or around umbilicus, abdomen areas, and front and back of waist (34.1%) in both genders. Normal results were observed in 88.7% (WBC counts), 84.5% (PLT counts), and 89.7% (RBC counts) of cases, respectively. Observations from the five hospitals were compared and no significant differences were found. CONCLUSION: The autumn-winter type scrub typhus in northern China occurred exclusively from September to December with a peak occurrence in October, which was different from the summer type in southern China. In comparison with the summer type, complications associated with autumn-winter type scrub typhus were less severe, and abnormalities of routine hematological parameters were less obvious.