Colistin therapy for multidrug-resistant Gram-negative infection: clinical outcome and risk factors

Abstract

Multidrug-resistant (MDR) Gram-negative bacterial infection is widespread all over the world, and recently it led to the reintroduction of colistin for treatment of MDR Pseudomonas aeruginosa and Acinetobacter spp. infection [1]. According to the 2007 European antimicrobial resistance surveillance system (EARSS) consisting of 33 European countries, six countries reported carbapenem resistance rates higher than 25 % among P. aeruginosa isolates [2]. The rate of imipenem-resistant Acinetobacter reached up to 51 % in Korea according to the report of the Korean Nationwide Surveillance of Antimicrobial Resistance (KONSAR) [3]. Polymyxin E (colistin) that was developed in the 1940s is still the principal therapeutic agent for MDR Gram-negative pathogens. Because of the nephrotoxicity and neurotoxicity of colistin, it was not used for a long time [4]. Colistin has lately been considered for MDR treatment due to the emergence of MDR bacteria. However, the efficacy test results of colistin have been quite controversial. Most studies have been retrospective, and the clinical data analysis has been done without control groups [5, 6]. In this study, we evaluated the clinical efficacy of colistin in MDR bacterial infection. We conducted a retrospective hospital-based cohort study at an 850-bed hospital by reviewing the medical records of patients who received intravenous colistin for MDR bacterial infection from May 2007 to December 2010. Demographics, clinical presentation, causative organism, and outcomes were recorded. Patients were excluded if they received colistin treatment for \72 h and were under the age of 18. Types of infection were evaluated, following the United States Centers for Disease Control and Prevention (CDC) criteria. Recent invasive procedure was defined as preceding an invasive procedure within 48 h, including arterial, central venous catheter, urinary catheter, or applying mechanical ventilation before the culture results of MDR infection. MDR bacteria were defined to be resistant to all antibiotic agents such as penicillins, aminoglycosides, ampicillin/sulbactam, cephalosporins, aztreonam, carbapenems, fluoroquinolones, and tetracyclines, except colistin. Interpretation of susceptibility results was carried out according to the current Clinical and Laboratory Standards Institute guidelines. A total of 107 patients were enrolled in this study, and their demographic and clinical data are summarized in Table 1. Sixty-six patients (61.7 %) were over 64 years, and 36 patients (33.6 %) had a hospital stay of more than 125 days before MDR bacteria culture. Common underlying diseases were hypertension (63.5 %), diabetes (34.6 %), chronic obstructive pulmonary disease (24.3 %), cancer (19.6 %), chronic renal failure (CRF), and end-stage renal disease (ESRD; 12.1 % for both CRF and ESRD). More than 95 % of colistin therapy took place in ICUs. Cerebrovascular event (27.1 %) including cerebral infarction or hemorrhage was the most common cause of admission. We evaluated the clinical and microbiological responses both to colistin and to previously exposed antibiotics combined with colistin. Efficacy was evaluated as both clinical and bacteriological responses. After 3 days of colistin therapy, clinical improvement was graded as a good or poor clinical response. A good responder was defined as a patient whose symptoms were improved with bacteriologically clear pathogen within 1 week. Baseline characteristics and various risk factors were compared Y. H. Jun S. J. Jeun S. H. Kang H. J. Choi (&) Division of Infectious Disease, Department of Internal Medicine, Ewha Womans University, Mokdong Hospital, 911-1 MokDong, Yangcheon-Ku, Seoul, Korea e-mail: heechoi@ewha.ac.kr