The systematic Review of Quinolones Resistance of Escherichia coli Isolated from Urinary Tract Infections in Iran over the Last Ten Years (2001-2011)

Aims: Urinary tract infection (UTI) is one the most common bacterial infection and Escherichia coli (E. coli) has been isolated from the majority of UTI cases. On the other hand, the rate of UTIs caused by quinolones resistant E. coli is increasing worldwide. We aimed to perform a systematic review of quinolones resistance of E. coli isolated from urinary tract infections in Iran over last ten years. Review Article Akya et al.; ARRB, 6(4): 234-244, 2015; Article no.ARRB.2015.081 235 Methodology: In this study the data of 53 cross-sectional studies from 2001 to 2011 on quinolones resistance of E. coli isolated from UTIs in 22 cities of Iran was collected. The data was statistical analyzed using SPSS version 18 and Stata software. Results: Of 53 studies, 44, 41, 8 and 6 studies tested ciprofloxacin, nalidixic acid, norfloxacin and ofloxacin, respectively. Majority of studies, 31(58.5%), was done on out-patients. For susceptibility testing of isolates, 49 and 4 of studies used disk diffusion and E test method, respectively. Forty six of studies (86.8%) used mid-stream urinesample, while the rest used other methods for sampling including catheter, suprapubic aspiration and urine bag. Overall resistance for nalidixic acid, ciprofloxacin, norfloxacin and ofloxacin were 42.3%, 28.2%, 48.5% and 24.1%, respectively. The results of most studies were significantly heterogenic (P<0.001). Conclusion: Results indicated the resistance to quinolones including ciprofloxacin, norfloxacin and nalidixic acid is quite high. Given the increased resistance to these drugs, urine culture and antimicrobial susceptibility testing are essential for the treatment of UTIs.


INTRODUCTION
In medical references, E. coli is the most common agent of urinary tract infections (UTIs) in all parts of the world [1,2]. According to a study it is estimated that 130-175 million human UTIs occur every year in the world, with about 80% being caused by uropathogenic E. coli (EUPEC) strains [2]. Using new antibiotics for the treatment of bacterial infections, resistance to the drugs has emerged. Resistance to the new generations of antibiotics and subsequently the development of resistant strains has become prevalent not only among the hospital-acquired infections but also among community-acquired infections [3]. Consequently, the choice of empiric treatments for UTIs has now become challenging, since 20-50% of E. coli isolates are now resistant to the first-line of antibiotics [4]. Following the increased resistance to cotrimoxazole and ampicillin as the first line therapy, quinolones have been widely prescribed as an alternative for the treatment of UTIs [6][7][8][9][10]. These antibiotics have become some of the most frequently prescribed antimicrobial agents worldwide and have widely been used in the clinical treatment of various bacterial infections, including the UTIs caused by E. coli. However, many studies have reported the isolation of quinolone resistant strains of E. coli [3,4].
Given this emerging problem, determining the resistant rate of isolates for quinolones is essential for the better treatment of UTIs.
In different parts of Iran, researchers have reported the prevalence of quinolone resistance for E. coli isolated from urinary tract infections [6][7][8][9][10]. However, a systematic review of comprehensive information of these studies has not been performed and subsequently the picture of emerging resistance of E. coli in Iran is not available. This study was conducted to determine a national picture of in vitro susceptibilities for ciprofloxacin, nalidixic acid, norfloxacin and ofloxacin among urine isolates of E. coli. Since studies used different methods of susceptibility testing with various antibiotics, samples and the isolates (community acquired and nosocomial origin), we systematically reviewed and did metaanalysis of the collected data of these studies.

Literature Search
The literature search was performed during the period from 2001 until 2011. For the review process we followed the PRISMA recommendations (Preferred Reporting Items for Systematic Reviews and Meta-Analyses). Searched keywords were the MeSH (Medical Subject Headings) terms "Escherichia coli" OR "E. coli "AND "quinolone resistance" AND "antibiotic susceptibility pattern" AND "Iran" in several electronic databases and publishers.

Inclusion Criteria
Among all articles or abstracts found, those with the following features were included in the study:

Exclusion Criteria
Studies with at least one of the below criteria were excluded: 1. Case report studies. 2. Studies with low sample size (less than 20 cases). 3. Methods for determination of antibiotic susceptibility testing could not be found from study. 4. The origin of samples was not clear.

Duplicate publications both in English and
Persian languages (the article published later and/or with more detailed results was chosen for analysis.) 6. Duplicate publications and congress abstracts.

Critical Appraisal and Selection of Studies
Eighty resources both English and Persian languages were reviewed by three groups namely microbiologists, epidemiologists and specialist of infectious diseases. From 96 studies, 43 were excluded because of one of the above mentioned reasons and 53 studies [6-10,17-64] were selected for data extraction and analysis ( Fig. 1). The relevant data including urine sampling methods (mid-stream urinary, catheter and suprapubic aspiration), year and location of sampling, type of patients (inpatient and outpatient), methods for antibiotic susceptibility testing (DAD, MIC and E-test), types of antibiotics, the gender and age (average age) of patients was also recorded. All data were collected and were inserted into an Excel file.

Statistical Analysis
Statistical analysis of all data was performed using STATA11software. For the purpose of this study we weighted each study using inverse of variance. In addition, random effect model was used to calculate pooled summary estimate of resistance. Heterogeneity among studies was tested using chi-squared. In order to find the reason for heterogeneity, the result of each antibiotic was pooled according to different variables such as urine sampling methods (midstream urinary, catheter and suprapubic aspiration), year and location of sampling, type of patients (inpatient and outpatient), methods for antibiotic susceptibility testing (DAD, MIC and E-test), types of antibiotics, the gender and age (average age) of patients using meta-regression. Antibiotic susceptibility pattern of E. coli isolates from UTIs in Iran was pooled by forest plot using the Meta-Analyst software. Statistical heterogeneity of the results was checked using Cochrane Q-test with significance set at P<0.05. . Therefore the use of antibiotics should be supervised not only for humans but also for animal treatments.

CONCLUSION
In conclusion given the fact that resistance to quinolones has been increased in our country, rational use of this group of antibiotics is required. Given the lower resistant rate of ofloxacin, it is more effective for empirical therapy in Iran. Furthermore, the urine culture and antimicrobial susceptibility testing are essential in successful treatment of UTIs.