Mortality Risk Higher for Carbapenem-Resistant Bloodstream Infections

Carbapenem resistance (CR) is associated with increased mortality risk among patients hospitalized with bloodstream infections (BSI), according study findings published in Clinical Infectious Diseases.

Researchers conducted a prospective multicenter study in Italy from June 2018 to January 2020. Patients hospitalized with BSI caused by CR gram-negative bacilli (GNB) were observed for 30 days. The primary outcomes were 30-day mortality and attributable mortality. Factors independently associated with 30-day mortality were evaluated via multivariable regression.

The final analysis included 1276 patients, all of whom were aged 18 years and older. The patients were allocated into groups by causative pathogen, as follows:

• Carbapenem-susceptible GNB (CS-GNB; 56.7%); 
• Klebsiella pneumonia-producing CR Enterobacterales (KPC-CRE; 23.8%);
• Metallo-β-lactamases-producing CR Enterobacterales (MBL-CRE; 6.0%);
• CR Pseudomonas aeruginosa (CRPA; 4.8%); and
• CR Acinetobacter baumannii (CRAB; 8.7%). 

Hospitalization rates were higher among patients with BSI due to KPC-CRE (79.3%), MBL-CRE (77.9%), CRPA (82.0%), and CRAB (94.9%) compared with those with BSI due to CS-GNB (75.1%; P <.001). Intensive care unit (ICU) transfer was more commonly required among patients with CR-GNB compared with those with CS-GNB.

At 30 days, the overall rate of mortality was 21.6%, with the highest mortality rates observed among patients with CRAB (43.2%), followed by those with MBL-CRE (36.4%), CRPA (32.8%), and KPC-CRE (26.6%).

There were a total of 1000 and 276 survivors and non-survivors, respectively within 30 days of BSI onset. Non-survivors vs survivors were older (median age, 71 vs 66.5 years) and had higher rates of diabetes (37.3% vs 27.3%), cardiovascular disease (53.6% vs 39.3%), hospital-acquired BSI (68.1% vs 58.0%), and septic shock at BSI onset (41.3% vs 14.8%).

Factors independently associated with 30-day mortality included ICU transfer (adjusted odds ratio [aOR], 7.09; 95% CI, 3.28-17.22; P <.001), hospitalization in a medical ward (aOR, 5.23; 95% CI, 2.37-12.29; P <.001), and sepsis-related organ failure assessment (SOFA) score (aOR, 1.23; 95% CI, 1.18-1.29; P <.001).

Factors inversely associated with 30-day mortality included urinary tract as the source of infection (aOR, 0.42; 95% CI, 0.27-0.66; P =.004) and early receipt of appropriate therapy (aOR, 0.62; 95% CI, 0.43-0.89; P =.010).

Stratified by causative pathogen, the risk of 30-day mortality highest among patients with MBL-CRE aOR, 5.86; 95% CI, 2.72-12.76), followed by those with CRAB (aOR, 2.65; 95% CI, 1.52-4.61) and those with CRPA (aOR, 1.99; 95% CI, 1.48-5.95). Attributable mortality rates for patients with BSI due to KPC-CRE, MBL-CRE, CRAB, and CRPA were 5%, 35%, 16%, and 19% respectively. 

Limitations of this study include potentially limited generalizability due to the high incidence of CR-GNB in Italy, the observational design, and the inability to establish causality between antimicrobial resistance and mortality. 

According to the researchers, “Prevention and surveillance strategies should be implemented in the hospital setting to avoid unfavorable outcome attributable to CR-GNB infections.”