Background Tuberculosis (TB) is a great mimicker and diagnostic chameleon, and prone to be diagnosed as malignancy. Even though many reports have described the differences between pulmonary TB and lung cancer, the atypical systemic hematogenous disseminated TB (HDTB) is very rare and more confusing in clinical practice. Case presentation A 73-year-old man, HIV-negative, was hospitalized to the local county hospital because of chest pain, low-grade fever, asthenia, anorexia and weight loss for the pasting two months. The CT findings of the two lungs showed multiple round or round-like nodules of different sizes, with clear boundaries and partial fusion. The level of serum CA19–9 was significantly higher than normal, and progressively increased. There were multiple enlarged lymph nodes in the neck, mediastinum, abdominal cavity and pelvic cavity. The symptoms were diagnosed as hematogenous spread of gastrointestinal tumor in the local county hospital. However, when transferred to our provincial hospital, through comprehensive dynamic analysis, this patient was diagnosed as atypical systemic HDTB, no cancer at all. Through routine anti-TB therapy for one year, the patient was recovered very well at the follow-up of half year after withdrawal. Conclusions In the past, most TB misdiagnosis cases involved in single organ and were finally confirmed through invasive examination. This case enriched clinical experiences in the diagnosis of atypical HDTB. We encouraged clinicians to establish a dynamic thinking for diagnosis and treatment and emphasized the value of biopsy and 18F-FDG-PET in distinguishing TB and cancer.

Oxidative stress is crucial to the biology of tumors. Oxidative stress’ potential predictive significance in colorectal cancer (CRC) has not been studied; nevertheless here, we developed a forecasting model based on oxidative stress to forecast the result of CRC survival and enhance clinical judgment. The training set was chosen from the transcriptomes of 177 CRC patients in GSE17536. For validation, 65 samples of colon cancer from GSE29621 were utilized. For the purpose of choosing prognostic genes, the expression of oxidative stress-related genes (OXEGs) was found. Prognostic risk models were built using multivariate Cox regression analysis, univariate Cox regression analysis, and LASSO regression analysis. The outcomes of the western blot and transcriptome sequencing tests were finally confirmed. ATF4, CARS2, CRP, GPX1, IL1B, MAPK8, MRPL44, MTFMT, NOS1, OSGIN2, SOD2, AARS2, and FOXO3 were among the 14 OXEGs used to build prognostic characteristics. Patients with CRC were categorized into low-risk and high-risk groups according on their median risk scores. Cox regression analysis using single and multiple variables revealed that OXEG-related signals were independent risk factors for CRC. Additionally, the validation outcomes from western blotting and transcriptome sequencing demonstrated that OXEGs were differently expressed. Using 14 OXEGs, our work creates a predictive signature that may be applied to the creation of new prognostic models and the identification of possible medication candidates for the treatment of CRC.