When treating infections caused by gram-negative pan-drug-resistant (PDR) bacteria, colistin and/or tigecycline-based combination empiric therapy is recommended, according to results of a study published in the European Journal of Clinical Microbiology & Infectious Diseases.

A significant emerging healthcare problem is pathogenic bacteria with resistance to multiple antimicrobial treatments: It is estimated to contribute a high number of extra deaths directly attributable to infection with these strains of bacteria. Further, several isolates of PDR bacteria have emerged as nonsusceptible to all agents of all antimicrobial categories. Despite this, excess mortality risk associated with PDR bacteria has only been estimated, whereas multidrug-resistant bacteria mortality risk has been significantly investigated. Few previous studies have investigated the specific clinical course and outcome of PDR bacterial infections, and results are limited as a result of small sample sizes, the inadequate definition of PDR, and the lack of description of important confounding factors including empiric therapy used.

Therefore, this retrospective, single-center cohort study investigated the clinical course, treatment pattern, prognostic factors, and outcome of patients with PDR infections, using a uniform definition for such infection. The primary outcome was in-hospital infection-related mortality, which was defined as mortality during hospitalization that could be attributed to infectious disease as either the immediate or underlying cause. Secondary outcomes included in-hospital mortality, defined as morality during hospitalization resulting from any cause, and the rate of effective empirical antimicrobial therapy.

In total, 65 patients from the University Hospital of Heraklion in Crete, Greece, who had infections caused by PDR gram-negative pathogens were included in the study. Isolates were classified as PDR if there was evidence of nonsusceptibility to all agents of all antimicrobial categories according to internally proposed criteria. The median age of the study cohort was 64 years, and the median Charlson comorbidity index was 3.0. Of the 65 PDR isolates, 31 (48%) were Klebsiella pneumoniae, 28 (43%) were Acinetobacter baumannii, and 6 (9%) were Pseudomonas aeruginosa.

Of the 34 patients who received antimicrobial therapy, results demonstrated that the use of noncolistin, nontigecycline-based combination was associated with higher infection-related in-hospital mortality rate (56%) compared with a 16% mortality rate associated with colistin combination and a 25% mortality rate linked to carbapenems and tigecycline treatment (P =.005).

Infection-related in-hospital mortality was 32%, and overall in-hospital mortality was 45%. Bivariate analysis showed that there were 5 factors that were significantly associated with infection-related in-hospital mortality: age, prior steroid use, severe sepsis at the diagnosis of PDR infection, the choice of empiric antimicrobial therapy, and the Charlson comorbidity index. Multivariate analysis showed 3 factors that were significantly associated with infection-related in-hospital mortality: empirical treatment with a noncolistin, nontigecycline combination (odd ratio [OR], 7.5; P =.008); prior steroid use (OR, 4.1; P =.049); Charlson comorbidity index (OR, 1.5; P =.030); severe sepsis at the diagnosis of PDR infection; and age. The source of PDR isolation (blood vs other) did not have an effect on in-hospital mortality (42% vs 29%, respectively; P =.341).

Related Articles

Overall, the study authors concluded that, “Our results support the use of colistin and/or tigecycline-based combinations as empirical therapy when infection due to gram-negative PDR pathogens is suspected, but prospective studies are necessary to further investigate specific antimicrobial combination treatments and their association with outcomes.”


Kofteridis DP, Andrianaki AM, Maraki S, et al. Treatment pattern, prognostic factors, and outcome in patients with infection due to pan-drug-resistant gram-negative bacteria [published online January 13, 2020]. Eur J Clin Microbiol Infect Dis. doi:10.1007/s10096-019-03784-9