Risk Factors Associated with Failure of Linezolid Therapy in Vancomycin-Resistant Enterococcus faecium Bacteremia: A Retrospective Cohort Study in a Referral Center in Mexico

We aimed to assess the factors associated with 30-day mortality in patients with vancomycin-resistant Enterococcus faecium (VREf) bloodstream infection (BSI) who received treatment with linezolid in an 11-year retrospective cohort of patients with VREf BSI. A univariate and stepwise multivariate logistic regression analysis was performed to determine 30-day mortality factors. Moreover, a Cox proportional hazards analysis of predictor covariates of mortality was performed. Eighty patients were included in the final analysis; 42 (53%) died and 38 (47%) survived 30 days after the index bacteremia. Thirteen patients of 42 (31%) died in the first 7 days. The Acute Physiology and Chronic Health Evaluation II (APACHE II) score was significantly associated with 30-day mortality (adjusted odds ratio [aOR], 1.46; 95% confidence interval [CI]: 1.22-1.76; p < 0.001) in the multivariate analysis. Moreover, VREf BSI persisting for more than 48 hours was a strong factor related to 30-day mortality (aOR, 19.6; 95% CI: 1.46-263; p = 0.01). Adequate control of infection source showed a trend to be protective without reaching significance in the multivariate analysis (aOR, 0.19; 95% CI: 0.04-1.0; p = 0.05). The Cox proportional hazards analysis confirmed the same significant mortality predictor besides linezolid treatment within the first 48 hours as a protective factor (hazard ratio 0.46; 95% CI: 0.23-0.92, p = 0.02). Severely ill patients with high APACHE II score and persistent bacteremia have a higher risk of failure with linezolid therapy.

Risk factors and outcomes associated with vancomycin-resistant Enterococcus faecium and ampicillin-resistant Enterococcus faecalis bacteraemia: A 10-year study in a tertiary-care centre in Mexico City

OBJECTIVES

We sought to identify risk factors associated with vancomycin-resistant Enterococcus faecium (VRE) and ampicillin-resistant Enterococcus faecalis (ARE) bacteraemia, predictors of 30-day mortality, and 90-day recurrence-free survival according to resistance.

METHODS

We evaluated clinical records of patients with E. faecalis and E. faecium bacteraemia (2007-2017). We performed bivariate and multivariate logistic regression analyses to identify factors associated with VRE and ARE bacteraemia and predictors of 30-day mortality. A Kaplan-Meier estimate of 90-day recurrence-free survival was done.

RESULTS

We identified 192 and 147 E. faecium and E. faecalis bacteraemia episodes, respectively, of which 55.7% of E. faecium were VRE (94% vanA) and 12.2% of E. faecalis were ARE. Factors related to VRE bacteraemia were previous hospitalisation (aOR, 80.18, 95% CI 1.81-634), history of central venous catheter (aOR, 11.15, 95% CI 2.48-50.2) and endotracheal cannula use (aOR, 17.91, 95% CI 1.22-262.82). There was higher attributable mortality to VRE (28%, 95% CI 14-68%; P < 0.001) and ARE (10%, 95% CI 0.1-36%; P = 0.58) compared with their susceptible counterparts. APACHE II (aOR, 1.45, 95% CI 1.26-1.66) and history of chemotherapy (aOR, 3.52, 95% CI 1.09-11.39) were predictors of E. faecium bacteraemia 30-day mortality. We could not recognise any factor related to ARE bacteraemia or E. faecalis 30-day mortality.

CONCLUSION

History of hospitalisation and invasive device use were related to VRE bacteraemia. APACHE II and history of chemotherapy were predictors of mortality. We could not identify factors related to ARE or predictors of mortality.

1393. Tuberculosis (TB) After Solid-organ Transplant (SOT) and Hematopoietic Stem Cell Transplant (HSCT)

Background

Tuberculosis is an important opportunistic infection that affects transplant recipients; the risk of active infection increases significantly when compared with the general population. Most disease results from reactivation of latent infection, being extrapulmonary and disseminated disease the most common presentations. Most cases occur during the first year post-transplantation when immunosuppression is higher. We describe the clinical characteristics of patients diagnosed with TB after transplant.

Methods

Single-center, retrospective study of adult SOT and HSCT recipients in Mexico City, who developed active TB after transplant. We reviewed medical records, and collected demographic data, clinical characteristics, and outcome.

Results

We identified 16 patients with post-transplant TB; 13 SOT, and 3 HSCT recipients. The majority of SOT recipients were women (53.8%); median age was 43 years, 9 were kidney and 4 liver transplant recipients. At TB diagnosis, 84.6% of patients were on 3 immunosuppressors. Latent TB was assessed before transplant in 5 patients (38.4%), of these 3 (60%) were tuberculin skin test+, and 2 received isoniazid. Extrapulmonary disease was most common (7, 53.8%). Predominant symptoms were fever (53.8%), chills (30.8%), and diaphoresis (38.5%); six were diagnosed during the 1st year (46.2%) post-transplant; the median of time to diagnosis was 24 months after transplant. The diagnosis was made by histopathology in most cases. Twelve patients received first-line anti-TB treatment. Overall mortality was 30.8%, directly attributable to TB in 2. In the HSCT group, 2 were women; median age was 22 years, 2 allogeneic and 1 autologous transplant. One patient had been treated for latent TB before transplantation. Two developed disseminated disease. Two patients presented within 6 months after the transplant, and the other within a year. Mortality was 100%, attributable to the infection in two patients.

Conclusion

In regions with intermediate to a high prevalence of TB; post-transplant TB could result from reactivation or post-transplant exposure. Most cases occur within the first year post-transplant; clinical symptoms are nonspecific, which lead to a delay in diagnosis. Morbidity and mortality remains high.

Disclosures

All authors: No reported disclosures.

1502. Risk Factors Associated to Aeromonas spp. Infection and Outcome at a Third-level Care Hospital in Mexico

Background

Information regarding infections caused by Aeromonas spp. is limited. The aim of this study was to describe the clinical and epidemiologic characteristics of infections by Aeromonas spp. during a 7-year period at a tertiary care hospital in Mexico.

Methods

We analyzed a retrospective cohort of adults with Aeromonas spp. clinical isolates, between January 1, 2010 and August 31, 2017 from the Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán. We analyzed demographic, clinical, microbial characteristics and antimicrobial susceptibility test. Risk factors for bacteremia and mortality were identified with logistic regression; adjustment for possible confounder factors was carried out. Data were analyzed with IBM SPSS Statistics version 21. Statistical significance was defined as the value for P < 0.05.

Results

We identified 387 patients with an infection by Aeromonas spp.; the median age was 55 years and 52% were women. 94 (24.3%) patients had an autoimmune disease; 74 (19.1%) had a solid tumor (ST); 51 (13.2%) had chronic kidney disease (CKD); 43 (11.1%) had chronic liver disease (CLD); 37 (9.6%) had a hematologic malignancy (HM); 23 (5.9%) were solid-organ recipients and 20 (5.2%) HIV infected. 41.6% (n = 161) of the infections were healthcare related. Stools were the most frequent isolation site (n = 299, 77%), followed by blood cultures (n = 29, 7.4%) and abscess (n = 19, 4.9%). The most common species was A. veronii. Aminoglycosides, quinolones, carbapenems, and trimethoprim/sulfamethoxazole were the most active antibiotics in vitro. The variables associated with bacteremia were CLD OR 5.647; diabetes OR 2.376 and nosocomial acquisition OR 4.08. 30-day mortality was 5.7%, while global mortality was 10.3%. Mortality was associated with sex (male) OR 1.753; HM OR 2.741; ST OR 2.576; polymicrobial infection OR 2.445; ICU admission OR 5.175 and bacteremia OR 3.881.

Conclusion

Infections by Aeromonas spp. have increased and have a greater incidence among individuals with chronic diseases and immunosuppressive states in this setting. Mortality described in this cohort was minor than previously stated by other series.

Disclosures

All authors: No reported disclosures.

573. Enterococcal Bacteremia in a Tertiary Care Center in Mexico: A Retrospective Analysis Focus on Vancomycin-Resistant E. faecium and Ampicillin-Resistant E. faecalis

Background

The primary pathogens in genera enterococcus are E. faecalis and E. faecium, increasing acquired resistance to glycopeptides and β lactamic has done the management more challenging. We aimed to describe the risk factors for acquisition of bacteremia for vancomycin-resistant E. faecium (VRE) and ampicillin-resistant E. faecalis (ARE) and the 30-day mortality in comparison to susceptible enterococcal bloodstream infection (BSI)

Methods

From 2007- 2017 medical records of all BSI for E. faecalis and E. faecium were evaluated. Risk factor for acquisition of VRE and ARE as well as the significant variables associated with 30-day mortality for enterococcal BSI were determined by univariate and multivariate analysis. The molecular mechanism of VRE was performed by PCR

Results

There were 192 patients with E. faecium BSI of which 107(56%) patients had VRE BSI with 94% VRE strains expressing vanA gene. The index bacteremic episodes were classified as nosocomial o healthcare associated in 99%, 102(95%) had hospitalization 1 year before and 101(94%) history of use of antibiotics 3 months earlier, the multivariate analysis showed duration of the previous hospitalization >10 days (OR, 80.18; 95% CI, 1.81–634), use of central venous catheter [OR, 11.15; 95% CI, 2.48–50.2), and endotracheal cannula [OR, 17.91; 95% CI, 1.22–262) as significant associated variables. The mortality for VRE was greater than susceptible E. faecium (60% vs. 24%, P < 0.001). The only factors for 30-day mortality for E. faecium BSI in the multivariate analysis was APACHE ll score [OR,1.45; 95% CI, 1.26–1.66) and patients with chemotherapy of cancer. (OR, 3.52; 95% CI, 1.09–11.39). 147 patients had E. faecalis BSI of which 18 (11%) patients had ARE, we did not find relevant clinical differences of ARE in comparison with ampicillin-susceptible E. faecalis, neither in risk factors for acquisition of ARE nor 30-day mortality [7(39%) vs. 38(29%), P = 0.58] in uni and multivariate analysis

Conclusion

Our evaluation showed in a period of 10 years that VRE expressing vanA gene had a strong association with patients with previous nosocomial exposure. Severely ill patients and cancer patients on chemotherapy during the bacteremic episode were the variables more associated with 30-day mortality. ARE is yet of low prevalence and less known, constant surveillance about it is warranted

Disclosures

All authors: No reported disclosures.