Fever of unknown origin: A 12-year case series in Colombia.

BACKGROUND: Fever of unknown origin (FUO) is a diagnostic challenge with highly heterogeneous causes. Its etiology can change according to the studied regions, and the chance of reaching a diagnosis depends on available resources. The aim of this study is to describe the clinical characteristics, etiology and the usefulness of diagnostic aids in cases of FUO managed over 12 years in a Colombian reference center. METHODOLOGY: Single-institution retrospective case series. All cases of FUO between 2006 and 2017 were identified with the help of an electronic medical record search software. Cases of adults with fever for more than three weeks who remained undiagnosed after three days of hospitalization are described. RESULTS: Of 1,009 cases evaluated, 112 cases met the inclusion criteria (median age 43 years, 66% men). The etiologies identified were infectious (31.2%), inflammatory (20.5%), neoplastic (14.3%), and miscellaneous (2.7%) diseases. 31.2% remained without etiological diagnosis. The most frequent conditions were tuberculosis (17%), Hodgkin's lymphoma (7.1%), systemic lupus erythematosus (6.3%), disseminated histoplasmosis, and adult Still's disease. Contrast tomography and biopsies were the studies that most frequently supported or confirmed the final diagnosis. CONCLUSIONS: This series of contemporary Latin American cases suggests that the categories of FUO etiologies are similar to those reported in studies from developed countries, with tuberculosis being the most frequent cause in our setting. Our results highlight the importance of tomography-guided invasive studies in the diagnostic approach to FUO.

Risk factors associated with methicillin-resistant Staphylococcus aureus skin and soft tissue infections in hospitalized patients in Colombia.

OBJECTIVES: Methicillin-resistant Staphylococcus aureus (MRSA) skin and soft tissue infections (SSTIs) represent a major clinical problem in Colombia. The aim of this study was to evaluate the risk factors associated with MRSA SSTI in Colombia. METHODS: A multicenter cohort study with nested case-control design was performed. Patients with an SSTI with at least 48h of inpatient care were included. Patients with an MRSA SSTI were considered the case group and patients with either a non-MRSA SSTI or with an Methicillin-susceptible S. aureus (MSSA) SSTI were the control groups. A multivariate logistic regression approach was used to evaluate risk factors associated with MRSA SSTI with two different statistical models. RESULTS: A total 1134 patients were included. Cultures were positive for 498 patients, of which 52% (n=259) were Staphylococcus aureus. MRSA was confirmed in 68.3% of the S. aureus cultures. In the first model, independent risk factors for MRSA SSTI were identified as the presence of abscess (P<0.0001), cellulitis (P=0.0007), age 18-44 years (P=0.001), and previous outpatient treatment in the previous index visit (P=0.003); surgical site infection was a protective factor (P=0.008). In the second model, the main risk factor found was previous outpatient treatment in the previous index visit (P=0.013). CONCLUSIONS: Community-acquired SSTIs in Colombia are commonly caused by MRSA. Therefore, clinicians should consider MRSA when designing the initial empirical treatment for purulent SSTI in Colombia, although there seems to be low awareness of this fact.