Clinical and economic burden of community-onset multidrug-resistant infections requiring hospitalization.

OBJECTIVES: To analyze the clinical and economic burden of community-acquired (CA) or community-onset healthcare-associated (COHCA) multidrug-resistant (MDR) infections requiring hospitalization. METHODS: Case-control study. Adults admitted with CA or COHCA MDR infections were considered cases, while those admitted in the same period with non-MDR infections were controls. The matching criteria were source of infection and/or microorganism. Primary outcome was 30-day clinical failure. Secondary outcomes were 90-day and 1-year mortality, hospitalization costs and resource consumption. RESULTS: 194 patients (97 cases and 97 controls) were included. Multivariate analysis identified age (odds ratio [OR], 1.07, 95% confidence interval [CI], 1.01-1.14) and SOFA score (OR, 1.45, CI95%, 1.15-1.84) as independent predictors of 30-day clinical failure. Age (hazard ratio [HR] 1.09, 95%CI, 1.03-1.16) was the only factor associated with 90-day mortality, whereas age (HR 1.06, 95%CI, 1.03-1.09) and Charlson Index (HR 1.2, 95%CI, 1.07-1.34) were associated with 1-year mortality. MDR group showed longer hospitalization (p<0.001) and MDR hospitalization costs almost doubled those in the non-MDR group. MDR infections were associated with higher antimicrobial costs. CONCLUSIONS: Worse economic outcomes were identified with community-onset MDR infections. MDR was associated with worse clinical outcomes but mainly due to higher comorbidity of patients in MDR group, rather than multidrug resistance.

Weighting the impact of virulence on the outcome of Pseudomonas aeruginosa bloodstream infections.

OBJECTIVES: We assessed the association between the lethality of Pseudomonas aeruginosa in a Caenorhabditis elegans model and outcomes of P. aeruginosa bloodstream infections. METHODS: A total of 593 P. aeruginosa bloodstream isolates recovered from a prospective Spanish multicentre study were analysed. Clinical variables, susceptibility profiles and Type III Secretion System (TTSS) genotypes (exoU/exoS genes) were available from previous studies. A C. elegans virulence score (CEVS) was used, classifying the isolates into high (CEVS 4-5), intermediate (CEVS 3) and low (CEVS 1-2) virulence. The main outcome analysed was 30-day mortality. RESULTS: Up to 75% (446/593) of the isolates showed a high-virulence phenotype, and 17% (101/593) a low-virulence one. No association between virulence phenotype and the main outcome variable (30-day mortality) was found (29/101 (28.7%) versus 127/446 (28.5%), p 1). However, an inverse association between C. elegans virulence and multidrug-resistant and extensively drug-resistant profiles was documented (OR 0.655 (95% CI 0.571-0.751) and OR 0.523 (95% CI 0.436-0.627), p <0.001, respectively), whereas the exoU genotype was significantly more frequent among isolates showing high virulence (10/101 (9.9%) versus 112/446 (25.1%), p <0.001). Moreover, although significance was not reached, strains showing a high-virulence phenotype tended to be associated with community-acquired infections (1/101 (1%) versus 25/446 (5.6%), p 0.065), whereas low-virulence phenotypes tended to be associated with a higher illness severity (such as higher median Pitt score: 2 (1-4) versus 1 (0-3), p 0.036, or initial multiorgan dysfunction: 17/101 (16.8%) versus 41/446 (9.2%), p 0.024), with some underlying conditions (such as chronic renal failure 24/101 (23.8%) versus 59/446 (13.2%), p 0.013), and with the respiratory source of infections (17/101 (16.8%) versus 45/446 (10.1%), p 0.058). CONCLUSIONS: Our results indicate that the P. aeruginosa virulence phenotype in a C. elegans model correlates with virulence genotype (TTSS) and resistance profile, but it is a poor prognostic marker of mortality in bloodstream infections.

Impact of multidrug resistance on Pseudomonas aeruginosa ventilator-associated pneumonia outcome: predictors of early and crude mortality.

The prevalence of multidrug-resistant (MDR) Pseudomonas aeruginosa has increased over the past decade and a significant rise in these isolates in ventilator-associated pneumonia (VAP) has been observed. However, the impact of MDR on VAP outcome has not been analysed in depth. We investigated the risk factors for early and crude mortality in a retrospective study of microbiologically and clinically documented VAP. Ninety-one VAP episodes in 83 patients were included, 31 caused by susceptible P. aeruginosa and 60 by MDR strains, of which 42 (70 %) were extensively drug-resistant (XDR) P. aeruginosa. Thirteen episodes concomitantly presented P. aeruginosa bacteraemia, in seven of which the origin was the respiratory tract. Whereas susceptible P. aeruginosa episodes were more likely than MDR episodes to receive adequate empirical (68 % vs. 30 %; p < 0.001) and definitive antimicrobial therapy (96 % vs. 50 %; p < 0.001), susceptible P. aeruginosa VAP presented a trend towards early mortality (29 % vs. 15 %; p = 0.06). A logistic regression model with early mortality as the dependent variable identified multiorgan dysfunction syndrome (MODS) [odds ratio (OR) 10.4; 95 % confidence interval (CI) 1.7-63.5; p = 0.01] and inadequate antibiotic therapy (OR 4.27; 95 % CI 0.98-18.4; p = 0.052) as independent risk factors for early mortality. A similar analysis identified MODS (OR 4.31; 95 % CI 1.14-16.2; p = 0.03) as the only independent predictor of crude mortality. The severity of acute illness clinical presentation was the main predictor of mortality. Despite adequate antibiotic therapy, susceptible P. aeruginosa seems to cause major early mortality. Although adequate therapy is essential to treat VAP, the severity of acute illness is a more important factor than drug resistance.

Extensively drug-resistant Pseudomonas aeruginosa: risk of bloodstream infection in hospitalized patients.

Several studies have suggested that resistance determinants usually reduce virulence. However, their contribution to decrease bloodstream infections is unclear. Our aim was to identify risk factors of extensively drug-resistant (XDR) Pseudomonas aeruginosa (PA) bacteremia and to assess the prevalence of XDR-PA bacteremia. A retrospective study of PA bloodstream infections in our patient population with at least one clinical sample isolate due to PA (2006-2007) was carried out. A total of 2,131 patients with PA clinical samples were detected. Among 1,657 patients with susceptible-PA isolates, 95 developed PA-susceptible bacteremia. Concomitantly, among 474 patients with multidrug-resistant (MDR)-PA isolates, 265 with XDR-PA, and 209 with non-XDR MDR-PA, 43 developed XDR-PA bacteremia and 13 non-XDR MDR-PA bacteremia, respectively. Pulsed-field gel electrophoresis (PFGE) revealed the clonal nature of the two predominant XDR-PA phenotypes and genetic heterogeneity in non-XDR MDR-PA phenotypes. The proportion of XDR-PA bacteremia was higher than the proportion of bacteremia in the susceptible-PA population (16 % vs. 6 %; p < 0.001). A logistic regression model identified prior exposure to fluoroquinolones [odds ratio (OR) 2.80; 95 % confidence interval (CI) 1.02 to 7.70] as the independent variable associated with XDR-PA bacteremia. Our study suggests that XDR-PA strains have a greater ability to develop bacteremia. It remains unclear as to whether this invasive capacity depends on clonal traits or on other virulence determinants.