Attributable Deaths Caused by Infections with Antibiotic-Resistant Bacteria in France

We read with interest the paper by Alessandro Cassini and colleagues1Cassini A Hogberg LD Plachouras D et al.Attributable deaths and disability-adjusted life-years caused by infections with antibiotic-resistant bacteria in the EU and the European Economic Area in 2015: a population-level modelling analysis.Lancet Infect Dis. 2019; 19: 56-66Summary Full Text Full Text PDF PubMed Scopus (1415) Google Scholar on deaths attributable to antibiotic-resistant bacterial infections in the EU in 2015, in which it was estimated that 33 110 deaths each year are caused by infections with antibiotic-resistant bacteria. The method used to provide this estimation is based on mathematical constructs developed in the EU as part of the Burden of Communicable Disease in Europe project, in the context of widespread anxiety about antibiotic resistance and the risk of present and future mortality due to antibiotic-resistant bacteria. In the report by Cassini and colleagues,1Cassini A Hogberg LD Plachouras D et al.Attributable deaths and disability-adjusted life-years caused by infections with antibiotic-resistant bacteria in the EU and the European Economic Area in 2015: a population-level modelling analysis.Lancet Infect Dis. 2019; 19: 56-66Summary Full Text Full Text PDF PubMed Scopus (1415) Google Scholar the number of deaths due to multidrug-resistant bacteria in France was estimated to be 5543 per year (out of 29 000 deaths due to bacterial infections per year). At the IHU Méditerranée Infection (Marseille, France), we examine, on a weekly basis, the microbiological results of most hospital and private laboratories in the southern region of France (which has around 5 million inhabitants), accounting for about 300 000 bacterial isolates per year.2Abat C Chaudet H Colson P Rolain J-M Raoult D Real-time microbiology laboratory surveillance system to detect abnormal events and emerging infections, Marseille, France.Emerg Infect Dis. 2015; 21: 1302-1310Crossref PubMed Scopus (33) Google Scholar, 3Huart M Bedubourg G Abat C et al.Implementation and initial analysis of a laboratory-based weekly biosurveillance system, Provence-Alpes-Côte d'Azur, France.Emerg Infect Dis. 2017; 23: 582-589Crossref PubMed Scopus (7) Google Scholar We also examine the results of antibiotic susceptibility testing of all isolated bacteria in the University Hospital of Marseille (a 3800-bed hospital),2Abat C Chaudet H Colson P Rolain J-M Raoult D Real-time microbiology laboratory surveillance system to detect abnormal events and emerging infections, Marseille, France.Emerg Infect Dis. 2015; 21: 1302-1310Crossref PubMed Scopus (33) Google Scholar as well as all deaths that occur each week in our hospitals (between 50 and 70 deaths) in patients with a positive bacterial isolate in any kind of microbiological sample in the preceding months. From these patients, we look for 30-day attributable mortality per multidrug-resistant bacteria and any infection site of these patients.4Abat C Fournier PE Jimeno MT Rolain JM Raoult D Extremely and pandrug-resistant bacteria extra-deaths: myth or reality?.Eur J Clin Microbiol Infect Dis. 2018; 37: 1687-1697Crossref PubMed Scopus (23) Google Scholar In practice, we have only reported one case of a bacterium resistant to all antibiotics tested within the last 5 years, associated with a death in intensive care unit (ICU), probably for multifactorial reasons.4Abat C Fournier PE Jimeno MT Rolain JM Raoult D Extremely and pandrug-resistant bacteria extra-deaths: myth or reality?.Eur J Clin Microbiol Infect Dis. 2018; 37: 1687-1697Crossref PubMed Scopus (23) Google Scholar, 5Abat C Rolain J-M Dubourg G Fournier P-E Chaudet H Raoult D Evaluating the clinical burden and mortality attributable to antibiotic resistance: the disparity of empirical data and simple model estimations.Clin Infect Dis. 2017; 65: S58-S63Crossref PubMed Scopus (30) Google Scholar Because of this large discrepancy between estimation and real deaths, we asked an ICU doctor (ML) associated with our research unit to conduct a short survey within Société Française d'Anesthésie et de Réanimation (the French Society of Anaesthesia and Intensive Care Medicine, which comprises more than 350 ICU practitioners and anaesthetists in France) on their practice over the past 10 years. Participants were asked if they had dealt with patient deaths directly linked to an antibiotic therapeutic impasse despite treatment adjusted to antibiotic susceptibility test results. Among the 251 respondents, 116 (46%) had seen no cases of death and 106 (42%) had seen between one and five cases over the past 10 years. Thus, among 222 (88%) of the 251 participants with more than 10 years of experience, there were probably around 45 deaths per year (IQR 22–73, calculated from the median of responses) due to an antibiotic therapeutic impasse. In France, the mortality of patients hospitalised in ICUs is around 19% (26 600 of 140 000 patients). This reflects the risk of drift without verifying and counting the number of real deaths. In France, the 2015 BURDEN study reported an estimated 12 500 deaths per year in France due to antibiotic resistance,6Colomb-Cotinat M Lacoste J Brun-Buisson C Jarlier V Coignard B Vaux S Estimating the morbidity and mortality associated with infections due to multidrug-resistant bacteria (MDRB), France, 2012.Antimicrob Resist Infect Control. 2016; 5: 56Crossref PubMed Scopus (47) Google Scholar whereas the number of deaths due to bacterial infections in all French hospitals is less than 29 000 per year. The model used in the BURDEN study predicted a number of deaths per year that is more than two times higher than the number (5543 deaths) predicted by the model used by Cassini and collegaues.1Cassini A Hogberg LD Plachouras D et al.Attributable deaths and disability-adjusted life-years caused by infections with antibiotic-resistant bacteria in the EU and the European Economic Area in 2015: a population-level modelling analysis.Lancet Infect Dis. 2019; 19: 56-66Summary Full Text Full Text PDF PubMed Scopus (1415) Google Scholar In fact, Cassini and colleagues' report does not represent true data on number of deaths. Mathematical models for estimation of deaths are disconnected from the reality of clinical practice of hospital doctors and clinical microbiologists, who look daily for deaths and antibiotic resistance. In practice, we believe that it is becoming essential to create a national account register recording the true instances of deaths related to the presence of a bacterium in a pathogenic situation resistant to all antibiotics available. This is crucial to obtain a reasonable understanding of mortality due to antibiotic resistance, which cannot be acquired by unrealistic estimations using inappropriate mathematical models based on extrapolation of multiple and non-controlled studies. We declare no competing interests. We thank the Société Française d'Anesthésie et de Réanimation and Jean Michel Constantin for editing and disseminating the survey. This work was supported by the French Government under the Investissements d'avenir (Investments for the Future) programme managed by the Agence Nationale de la Recherche (reference Méditerranée Infection 10-IAHU-03). Attributable deaths and disability-adjusted life-years caused by infections with antibiotic-resistant bacteria in the EU and the European Economic Area in 2015: a population-level modelling analysisOur results present the health burden of five types of infection with antibiotic-resistant bacteria expressed, for the first time, in DALYs. The estimated burden of infections with antibiotic-resistant bacteria in the EU and EEA is substantial compared with that of other infectious diseases, and has increased since 2007. Our burden estimates provide useful information for public health decision-makers prioritising interventions for infectious diseases. Full-Text PDF Open AccessAttributable deaths caused by infections with antibiotic-resistant bacteria in France – Authors' reply"All models are wrong, but some are useful"—George E P Box1 Full-Text PDF