International Journal of Antimicrobial Agents 27 (2006) 476–481
Multidrug resistance to antimicrobials as a predominant factor influencing patient survival Evangelos J. Giamarellos-Bourboulis, Evangelia Papadimitriou, Nearchos Galanakis, Anastasia Antonopoulou, Thomas Tsaganos, Kyriaki Kanellakopoulou, Helen Giamarellou ∗ 4th Department of Internal Medicine, Athens Medical School, Attikon University Hospital, 1 Rimini Str., 124 64 Athens, Greece Received 13 October 2005; accepted 19 December 2005
Abstract The impact of multidrug resistance to antimicrobials was assessed in a cohort of 243 patients with microbiologically documented infections by a variety of susceptible and multidrug-resistant (MDR) species. Multidrug resistance was defined as resistance to more than two antimicrobial agents of different chemical structure. Cox regression analysis was performed to define differences and the significance of any predisposing factors. Overall survival of patients infected by susceptible isolates was prolonged compared with patients infected by MDR isolates (P = 0.013). Mortality rates of infections caused by susceptible and MDR isolates were 4.87% and 16.15%, respectively (P = 0.013); the higher mortality rate for MDR isolates was more pronounced for infections by Klebsiella pneumoniae and Pseudomonas aeruginosa. Mean (± standard error (S.E.)) survival of patients infected by susceptible and MDR isolates in patients without signs of severe sepsis was 28 days and 27.29 ± 0.35 days, respectively (P = not significant). Mean (± S.E.) survival of patients with severe sepsis caused by susceptible and MDR isolates was 7.70 ± 4.62 days and 10.45 ± 2.18 days, respectively (P = 0.048). Diabetes mellitus type 2, the presence of severe sepsis and any underlying malignancy were the most important risk factors affecting survival. It is concluded that infections by MDR isolates were accompanied by higher mortality rates and decreased survival compared with infections by susceptible isolates. Diabetes mellitus type 2 and underlying malignancies were significant co-morbid conditions, whereas survival after infection by susceptible isolates was particularly decreased in the event of severe sepsis. © 2006 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. Keywords: Multiresistance; Risk factor; Survival
1. Introduction The increased prevalence of multidrug-resistant (MDR) isolates in the nosocomial setting has raised the need for new strategies in the management of nosocomial infections [1]. One dominant question that the clinician often faces nowadays is whether infections by MDR pathogens significantly affect mortality. Few studies have been performed and they are characterised by controversy. Infections by resistant isolates of Staphylococcus aureus [2] and Enterococcus faecalis [3] are accompanied by lower survival rates compared with infections by susceptible isolates of the same species, and a recent meta-analysis of 31 studies concluded that bacteraemia by methicillin-resistant S. aureus (MRSA) led to ∗
Corresponding author. Tel.: +30 210 58 31 994; fax: +30 210 53 26 446. E-mail address:
[email protected] (H. Giamarellou).
decreased survival compared with methicillin-susceptible S. aureus (MSSA) [4]. However, a recent study failed to disclose any difference in mortality rates of critically ill patients with nosocomial bacteraemias caused by either antibioticresistant or -susceptible Gram-negative pathogens [5]. The controversial results of the above studies raise the need to identify the impact of multidrug resistance on the overall survival of hospitalised patients with severe infections. The present study was designed to clarify this issue.
2. Patients and methods 2.1. Study design In a prospective study for the period January 1997 to January 2000, all patients hospitalised with either a community-
0924-8579/$ – see front matter © 2006 Elsevier B.V. and the International Society of Chemotherapy. All rights reserved. doi:10.1016/j.ijantimicag.2005.12.013
E.J. Giamarellos-Bourboulis et al. / International Journal of Antimicrobial Agents 27 (2006) 476–481
acquired or a hospital infection at the 1st Department of Propedeutic Medicine of Laikon General Hospital of Athens were followed-up. Two hundred and forty-three patients with a definite microbiological diagnosis of infection matched for age and sex were considered for further statistical analysis. The study was approved by the Ethics Committee of the Laikon General Hospital of Athens. A complete analysis was performed for each patient, comprising thorough history and physical examination, cultures of blood, urine, sputum or pus, blood gas analysis, lung radiography, and computed tomography of the lungs and upper or lower abdomen, whenever considered necessary. Survival of each patient was recorded from the time of advent of fever for a total of 28 days. The presence of the following predisposing factors was recorded: diabetes mellitus type 2, neoplasia, intake of corticosteroids, neutropenia, underlying malignancy and chronic renal failure. Neutropenia was defined as an absolute neutrophil count
