Portal de Periódicos da CAPES

Impact of antibiotics on the intestinal microbiota needs to be re-defined to optimize antibiotic usage

2017; Elsevier BV; Volume: 24; Issue: 1 Linguagem: Inglês

10.1016/j.cmi.2017.09.017

1469-0691

Étienne Ruppé, Charles Burdet, Nathalie Grall, Victoire de Lastours, François-Xavier Lescure, Antoine Andremont, Laurence Armand-Lefèvre,

Antibiotic Resistance in Bacteria

Antibiotic resistance is increasing worldwide and poses a fundamental, long-term threat to human health. Solutions tackling antibiotic resistance include the protection of our intestinal microbiota against antibiotics to prevent the acquisition, selection and overgrowth of multidrug-resistant bacteria (MDRB, here refers to extended-spectrum β-lactamase-producing Enterobacteriaceae (ESBL-PE), carbapenemase-producing Enterobacteriaceae (CPE), vancomycin-resistant enterococci (VRE) and Clostridium difficile). Hence, clinicians aim to use antibiotics that have the least impact on the intestinal microbiota, and data supporting the choice of an antibiotic over another one in this process are expected [[1]Weiss E. Zahar J.-R. Lesprit P. Ruppe E. Leone M. Chastre J. et al.Elaboration of a consensual definition of de-escalation allowing a ranking of β-lactams.Clin Microbiol Infect. 2015; 21 (649.e1–10)Google Scholar]. However, what ‘narrow’ versus ‘broad’ spectrum antibiotic really means with regards to the impact on the microbiota is far from clear. Here we have reviewed the basis of these uncertainties and tried to draw future directions for research in that field. We are inhabited by a number of bacteria that roughly equals our own number of cells [[2]Sender R. Fuchs S. Milo R. Revised estimates for the number of human and bacteria cells in the body.PLoS Biol. 2016; 14: e1002533Crossref PubMed Scopus (2279) Google Scholar], the biggest contingent being hosted in our intestinal microbiota. There is compelling evidence that the antibiotic intake impacts the intestinal microbiota (at least its distal, colonic part that is available through faeces collection), but the supporting data are scattered and rely on various protocols, mostly culture-based [[3]Sullivan A. Edlund C. Nord C.E. Effect of antimicrobial agents on the ecological balance of human microflora.Lancet Infect Dis. 2001; 1: 101-114Abstract Full Text Full Text PDF PubMed Scopus (596) Google Scholar]. Although recent advances of metagenomic sequencing have deepened our knowledge about the dynamics and functions of the intestinal microbiota, few studies have focused on the impact caused by individual antibiotics [4Grall N. Lazarevic V. Gaïa N. Couffignal C. Laouénan C. Ilic-Habensus E. et al.Unexpected persistence of extended-spectrum β-lactamase-producing Enterobacteriaceae in the faecal microbiota of hospitalised patients treated with imipenem.Int J Antimicrob Agents. 2017; https://doi.org/10.1016/j.ijantimicag.2017.02.018Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar, 5Stewardson A.J. Gaïa N. François P. Malhotra-Kumar S. Delémont C. Martinez de Tejada B. et al.Collateral damage from oral ciprofloxacin versus nitrofurantoin in outpatients with urinary tract infections: a culture-free analysis of gut microbiota.Clin Microbiol Infect. 2015; 21 (344.e1–11)Abstract Full Text Full Text PDF PubMed Scopus (87) Google Scholar, 6Dethlefsen L. Relman D.A. Incomplete recovery and individualized responses of the human distal gut microbiota to repeated antibiotic perturbation.Proc Natl Acad Sci USA. 2011; 108: 4554-4561Crossref PubMed Scopus (1523) Google Scholar, 7Fernandes M.R. Ignacio A. Rodrigues V.A.A. Groppo F.C. Cardoso A.L. Avila-Campos M.J. et al.Alterations of intestinal microbiome by antibiotic therapy in hospitalized children.Microb Drug Resist Larchmt N. 2017; 23: 56-62Crossref PubMed Scopus (7) Google Scholar], with substantial heterogeneity in their protocols with regards to DNA extraction, sequencing type (targeted or shotgun metagenomics), bioinformatic pipeline and biostatistical analyses. Altogether the assumption that one antibiotic impacts the microbiota harder than another is at best speculative, as data are either lacking or heterogeneous. Nonetheless, clinicians need some comparative, actionable data supporting the switch to antibiotics that would be less harmful to the microbiota (referred to as antibiotic de-escalation). It should be noted that the clinical spectrum of one given antibiotic and its impact on the microbiota are two distinct phenomena. The clinical spectrum refers to (i) the in vitro distribution of the minimal concentration that inhibits the growth of various bacterial species, and (ii) the pharmacokinetic data ensuring that the concentration of the antibiotic achieved in the infectious site is above the MIC for the bacterial pathogens involved in the infection. The wideness of the clinical spectrum refers to the number of bacterial species against which the antibiotic can be used in a therapeutic perspective. Conversely, the impact on the gut microbiota results from both the concentrations achieved by the antibiotic in the colon and the susceptibility of intestinal bacteria (which is usually not known). For instance, erythromycin is considered to be a narrow-spectrum antibiotic because it is only efficient against infections by susceptible Gram-positive bacteria in humans. Yet, intestinal concentrations of erythromycin are high enough (up to 600 μg/g of faeces) for the drug to be active on some intestinal Gram-negative bacteria, including Enterobacteriaceae [[8]Hartley C.L. Clements H.M. Linton K.B. Effects of cephalexin, erythromycin and clindamycin on the aerobic Gram-negative faecal flora in man.J Med Microbiol. 1978; 11: 125-135Crossref PubMed Scopus (17) Google Scholar]. On the other hand, imipenem has a very wide clinical spectrum because it is active on a large range of Gram-positive and Gram-negative bacteria, but it achieves very low intestinal concentrations [[9]Nord C.E. Kager L. Philipson A. Stiernstedt G. Effect of imipenem/cilastatin on the colonic microflora.Rev Infect Dis. 1985; 7: S432-S434Crossref PubMed Google Scholar] and its impact on the microbiota does not seems to reflect its clinical spectrum [[4]Grall N. Lazarevic V. Gaïa N. Couffignal C. Laouénan C. Ilic-Habensus E. et al.Unexpected persistence of extended-spectrum β-lactamase-producing Enterobacteriaceae in the faecal microbiota of hospitalised patients treated with imipenem.Int J Antimicrob Agents. 2017; https://doi.org/10.1016/j.ijantimicag.2017.02.018Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar]. Hence, although the clinical spectrum of old and new antibiotics has been assessed by standardized methods, their impact on the intestinal microbiota has not. This is precisely what needs to be assessed to support the clinicians' decision to de-escalate the antibiotic therapy. To that end, we first need to select markers reflecting the various aspects of the impact of antibiotics on the microbiota. From a clinical perspective, the rationale for de-escalation with regards to the intestinal microbiota is to prevent the overgrowth of MDRB (Fig. 1) that may be present in the intestinal microbiota. This can be measured by the intestinal concentrations of MDRB (expressed in colony-forming units per gram of faeces) and by their relative abundance (e.g. the ratio between the concentrations of MDRB and that of total Gram-negative bacilli) [[10]Ruppé E. Lixandru B. Cojocaru R. Büke C. Paramythiotou E. Angebault C. et al.Relative fecal abundance of extended-spectrum-β-lactamase-producing Escherichia coli strains and their occurrence in urinary tract infections in women.Antimicrob Agents Chemother. 2013; 57: 4512-4517Crossref PubMed Scopus (76) Google Scholar]. From an ecological perspective, the rationale for de-escalation is to limit the damage caused by antibiotics in the gut. Nowadays, this can be assessed by targeted-gene sequencing (such as 16S rRNA gene profiling) or shotgun metagenomics, both methods relying on next-generation sequencing. Diversity (evenness of the distribution of species) and richness (number of unique species or genes) indices can be measured in order to assess the effect of an antibiotic on the intestinal microbiota. Also, the evolution of the composition of the microbiota can be measured between samples collected within the same subjects before the antibiotic intake and at different times after. More specifically, shotgun metagenomics also allows the analysis of the resistome (i.e. the content of antibiotic resistance determinants), using dedicated databases [[11]Xavier B.B. Das A.J. Cochrane G. Ganck S.D. Kumar-Singh S. Aarestrup F.M. et al.Consolidating and exploring antibiotic resistance gene data resources.J Clin Microbiol. 2016; 54: 851-859Crossref PubMed Scopus (56) Google Scholar]. Besides, ‘culturomics’, referring to the extensive culture of samples in various conditions [[12]Lagier J.-C. Armougom F. Million M. Hugon P. Pagnier I. Robert C. et al.Microbial culturomics: paradigm shift in the human gut microbiome study.Clin Microbiol Infect. 2012; 18: 1185-1193Abstract Full Text Full Text PDF PubMed Scopus (760) Google Scholar] and that has been found to detect bacterial species missed in metagenomics [[13]Lagier J.-C. Khelaifia S. Alou M.T. Ndongo S. Dione N. Hugon P. et al.Culture of previously uncultured members of the human gut microbiota by culturomics.Nat Microbiol. 2016; 1: 16203Crossref PubMed Scopus (584) Google Scholar], should not be used in this context because no standardized protocol has been proposed. Setting up studies aimed at assessing the impact of antibiotics on the microbiota in patients is challenging because of the heterogeneity of dosing regimens, the co-administration of other antibiotics or other drugs that may also impact the microbiota, and the previous intake of antibiotics. Also, including patients for a rarely administered antibiotic in current practice may require substantial time and funding. Conversely, in settings such as the intensive care unit, the administration of antibiotics is more frequent, but antibiotics are rarely given alone. Hence, to ensure the homogeneity of participants and antibiotic regimen that will enable comparisons to be made between studies and antibiotics, we believe that studies aimed at assessing the impact of antibiotics on the microbiota should be performed on healthy volunteers. Such healthy volunteers should meet NIH criteria [[14]NIH Clinical Center: Healthy Volunteers n.d. Available at: https://clinicalcenter.nih.gov/recruit/volunteers.html (accessed 19 September 2017).Google Scholar] and in this context be devoid of MDRB. A sufficient number of healthy volunteers should be included to take into account the possible inter-individual variations in the absorption (for oral drugs) and excretion of the drugs and to obtain reliable estimates of their effect on the microbiota. However, it is not possible to inoculate healthy volunteers with MDRB for obvious ethical reasons. Hence, the assessment of the quantitative dynamics of MDRB under antibiotic exposure should be performed in animal models with an effort to mimic the intestinal concentrations observed in humans [[15]Perez F. Pultz M.J. Endimiani A. Bonomo R.A. Donskey C.J. Effect of antibiotic treatment on establishment and elimination of intestinal colonization by KPC-producing Klebsiella pneumoniae in mice.Antimicrob Agents Chemother. 2011; 55: 2585-2589Crossref PubMed Scopus (49) Google Scholar]. The animal microbiota is quite different from that of humans, so an alternative could be to culture ex vivo the faeces of human volunteers spiked with controlled inoculum of MDRB [[16]Borriello S.P. Barclay F.E. An in-vitro model of colonisation resistance to Clostridium difficile infection.J Med Microbiol. 1986; 21: 299-309Crossref PubMed Scopus (91) Google Scholar]. In such models, the evolution of the concentrations of the MDRB under exposure to the tested antibiotics could be measured. More complex will be the design of studies assessing the impact of antibiotics on the acquisition and transmission of MDRB. This will require contextual studies in specific wards thereby raising the difficulty of harmonizing the protocols and results. During the clinical development of a new antibiotic, phase I includes healthy volunteers to assess the safety of the drug and the pharmacokinetic parameters. For a limited extra-cost, metagenomic analyses of faeces and ex vivo experiments involving MDRB could be performed in the same volunteers, following a standardized protocol (see a proposition in Fig. 1) established by experts in antibiotic resistance, gut microbiota, pharmacokinetics and study design (particularly the duration of follow up and the time-points of sampling). During workshops, these experts should choose the markers and present them in an accessible and actionable fashion to the clinician (e.g. in a radar plot, Fig. 1) and build a composite ecological score allowing the ranking of antibiotics in terms of impact on the gut. This evaluation could be requested by health agencies such as the US Food and Drug Administration and the European Medicines Agency in order to release a new antibiotic. Still, this would apply to antibiotics under development, not to those currently in use. Hence, we call for funding to perform the studies which may, at last, provide the data supporting de-escalation and terminate the never-ending discussions of whether the impact of piperacillin-tazobactam on the intestinal microbiota is higher than that of ceftazidime or vice versa. These studies are urgently needed in the current context of the spread of antibiotic resistance and of the dry antibiotic pipeline. ER, CB and AA are consultants for DaVolterra. ER is also a consultant for MaaT Pharma. None.