# Epidemiological Analysis and Susceptibility Profile of Pathogens Isolated in Blood Cultures From Hospital Angeles Lindavista in Mexico City

## TL;DR

This study analyzed bacteria causing bloodstream infections in a Mexican hospital, focusing on their resistance patterns and clinical associations.

## Contribution

The paper provides a detailed epidemiological and susceptibility profile of pathogens from blood cultures in a specific hospital setting.

## Key findings

- Escherichia coli was the most common bacterium, with 11 isolates producing extended-spectrum β-lactamases.
- ESBL-producing E. coli showed high susceptibility to amikacin, cefotaxime, and tigecycline but resistance to other antibiotics.
- No statistically significant associations were found between clinical predictors and isolated pathogens.

## Abstract

Background

Bloodstream infections contribute to morbidity and mortality. A wide variety of microorganisms can be identified and recovered from these infections, and blood culture is the primary method for detecting them and determining their antimicrobial susceptibility. This helps clinicians confirm the efficacy of the initial empiric therapy or select an appropriate alternative based on the susceptibility of the isolated microorganism, thereby reducing the risks associated with antimicrobial misuse. We aimed to describe and analyze the bacteria isolated in blood culture samples, their antimicrobial resistance at Hospital Angeles Lindavista, and their association with demographic and clinical characteristics; to provide tools for the effective antimicrobial treatment of bacteremia; and to monitor the evolution of microbial resistance in this institution.

Methods

A retrospective observational study was conducted at Hospital Angeles Lindavista, including all patients with suspected sepsis, as determined by the physician’s judgement, who had at least one blood sample collected from January 2024 to December 2024. Variables examined included age, gender, clinical service, comorbidities, risk factors, diagnoses, detected pathogens, and antibiogram. The relationship between a specific bacterium and an affected organ system was analyzed using a chi-square test, followed by residual analysis. To explore potential associations between clinical variables, regression models were used. Finally, to verify the stability of the results, we performed a sensitivity analysis using only complete cases and compared these findings to those from the entire dataset. The statistical analysis was conducted using R version 4.4.1 (2024-06-14 ucrt) (R Foundation for Statistical Computing, Vienna, Austria) in RStudio (Posit Software, Boston, MA).

Results

We identified an adult population with a median age of 49 years and a predominance of men. The median hospital stay was eight days; hospitalization was the most common service, and the gastrointestinal system was most affected. Regarding risk factors, 31 patients (16%) had type 2 diabetes mellitus, and sepsis was the most frequent diagnosis (n = 18, 9%). Escherichia coli (E. coli) was the most common bacterium, isolated 17 times overall, with 11 producing extended-spectrum β-lactamases (ESBLs). These isolates showed high susceptibility to amikacin, cefotaxime, and tigecycline but exhibited marked resistance to all other antibiotics, including carbapenems. No statistically significant associations were found between any clinical predictor. It was suggested that ESBL-producing E. coli is linked to urinary tract infections; Pseudomonas aeruginosa appeared to be associated with multiorgan involvement, while Staphylococcus aureus was correlated with musculoskeletal infections.

Conclusions

The findings show that the epidemiology of bloodstream infections in our hospital remains steady, with consistent demographic patterns, a high rate of gram-negative organisms, and no clear clinical predictors of positive cultures. These results highlight the importance of microbiological surveillance, appropriate empiric antibiotic selection, and the careful interpretation of results, especially when potentially contaminating organisms are involved.