Difficult to Treat: Do We Need a New Definition?

To the Editor—We read with interest the article by Kadri et al [1] in which they described a new definition of gram-negative bacteremia, that is, difficult to treat (DTR), based on susceptibility to “first-line” antibiotics, generally beta-lactams or fluoroquinolones. We concur with the need to improve on the existing multidrug-resistant (MDR), extremely drug-resistant (XDR), and pandrug-resistant definitions [2] as they do not help with actual therapeutic choices. However, DTR does not help significantly because with introduction of new antibiotics, the definition of DTR will shift/mutate/evolve, including the assessment on what constitutes first-line treatment. We also agree that the use of older, often toxic agents is not ideal, even when they are demonstrated to be active in vitro, if there are newer alternatives. We think the term “carbapenem-resistant” when appropriately applied encompasses most DTR gram-negative infections and it provides useful phenotypic characteristics that enable pathogen-directed treatment. Carbapenem-susceptible infections that may be extended-spectrum beta-lactamase (ESBL) producers and also resistant to fluoroquinolones should not be considered DTR. We were also puzzled why this examination of the Premier Database did not identify Stenotrophomonas maltophilia as an important DTR pathogen that causes bloodstream infections (BSIs) in the United States. We examined a similar Premier dataset specifically for BSIs [3] and found that S. maltophilia is the most common etiology of carbapenem-resistant gram-negative BSIs. There may be 2 reasons for this. First, the Centers for Disease Control and Prevention/European Centre for Disease Control and Prevention definitions for MDR, XDR, and Physicians Desk Reference only focused on acquired resistance, while S. maltophilia is intrinsically resistant to most beta-lactams due to the presence of chromosomal ESBL (L2) and metallo-carbapenemase (L1) [4]. Second, the algorithm that the authors used required susceptibility testing to a given antibiotic, and most clinical microbiology laboratories do not test carbapenems for S. maltophilia. In our analysis of the Premier dataset, we considered all BSI isolates of S. maltophilia as carbapenem resistant. From 43 095 gram-negative bacteremias, 3.5% were caused by carbapenem-resistant gram-negative pathogens. Stenotrophomonas maltophilia was the most common species at 32%, while Pseudomonas aeruginosa caused 25% of infections, Klebsiella pneumoniae caused 17%, Acinetobacter baumannii caused 15%, Proteus mirabilis caused 8%, and Escherichia coli caused 3%. Importantly, of all carbapenem-resistant gram-negative BSIs, carbapenem-resistant Enterobacteriaceae only accounted for 28%. Trimethoprim/sulfamethoxazole (TMP/SMX) is considered the “gold standard” [5] for the treatment of S. maltophilia, and indeed >95% are susceptible to this drug. However, TMP/SMX is not a first-line agent in bacteremia; thus, S. maltophilia, by definition, should be considered DTR. Whether or not resistance is acquired or intrinsic does not change how difficult it is to treat an infection. Recently approved new antibiotics and others likely to be approved will render the DTR definition obsolete. What we really need now are rapid diagnostics for identification and susceptibility data that provide actionable information, especially for gram-negative BSIs. Potential conflicts of interest. Both authors: No reported conflicts of interest. Both authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.