Evaluating the antimicrobial resistance patterns and MDR frequency in Escherichia coli strains isolated from clinical specimens taken from patients in Iran: 2017–2019


 Objectives The purpose of this study was to evaluate the antimicrobial resistance and frequency of MDR strains in E. coli isolated from patients in two hospitals of Iran. In this descriptive-analytical study, 13322 clinical specimens were collected from two teaching hospitals. The E. coli isolates were identified using standard bacteriological methods. Then, the resistance patterns of the isolates were analyzed by disk agar diffusion method according to the CLSI.Results Out of 13322 clinical samples, 964 (7.23%) isolates of E. coli were identified. In both hospitals, high resistance to ampicillin and cefalexin was presented in 621 (64.4%) and 402 (41.7%) isolates, respectively. The highest antibiotic resistance was observed in burn unit, burn ICU and burn restoration section, while all 8 strains which were isolated from the neonatal-ICU, were sensitive to all tested antibiotics except cefalexin, nitrofurantoin, nalidixic acid, and ampicillin. Also, the bacteria isolated from urine, wound, stool, and blood samples were resistant to all tested antibiotics. Increased resistance to different antibiotics in burn hospital has created increasing concern. Very high resistance to some antibiotics indicates that these drugs are misused in therapeutic centers and highlights that infection control measures should be arranged in the ICUs of our hospitals.


Introduction
Escherichia coli (E. coli) causes important infections such as sepsis, neonatal meningitis, wound infection, and urinary tract infections (UTIs) (1,2). The UTI is the second most common bacterial infection after respiratory tract infection in outpatients and inpatients. The frequency of UTI is more common in women than in men, with approximately half of women experiencing it at least once in their lifetime (3,4). UTI in infants, pregnant women, the elderly, diabetics, multiple sclerosis, and immunocompromised individuals is associated with an increased risk (5,6). The prevalence of UTIs in the developing countries is estimated to be at least 250 million cases per year (6,7). Beta-lactam antibiotics are often used to control infections caused by E. coli, however several different antibiotic resistance mechanisms have been identi ed in E. coli (8,9). Failure to control and prevention of the antibiotic resistance will result in increased resistant isolates and augmented therapeutic failure in the treatment of infections caused by these isolates, resulting in prolonged course of the disease, increasing the likelihood of transmission of disease to others, increased length of hospital stay, need for more expensive medications, more complications for treatment and increased risk of death (1). The growing use of antibiotics and the subsequent increase in antibiotic resistance is one of the major concerns in health centers around the world. On the other hand, the antibiotic susceptibility pattern of E. coli clinical isolates varies in different regions of the world, different cities of a country, and even in different hospitals of a city or region. Awareness of the antibiotic resistance pattern of bacteria in each region is essential to effective treatment of the infections, control of the nosocomial infections and reduce the prevalence of resistant organisms. The purpose of this study was to evaluate the antimicrobial resistance and frequency of MDR strains in E. coli isolated from patients.

Methods
In this descriptive-analytical study, 13322 different clinical specimens (such as ulcer, urine, sputum, blood, stool, etc.) collected from patients admitted to Bu-Ali Sina hospital (a pediatric treatment center) and Zare hospital (A burn center) in Sari city from spring 2017 to winter 2019, were evaluated to identify E. coli isolates. Clinical samples were cultured on blood agar and eosin methylene blue agar (EMB) and incubated at 37° C for 24 hours. About urine culture, the patients with clinical symptoms such as fever or chills, dysuria, decreased urine volume, frequent or intense urge to urination, pain, pyuria, etc., along with a bacterial counting of ≥ 10 5 cfu/ml in urine culture were included in this study (2). After isolation, Gram staining and conventional standard biochemical and microbiological tests were used to identi cation of the isolates (10). Also, the analytical pro le index (API) (BioMérieux, Lyon, France) was used for nal con rmation of the isolates. Next, the antibiotic resistance patterns of the isolates were evaluated by standard disk agar diffusion method (Kirby-Bauer) according to CLSI (Clinical and Laboratory Standard Institute) guidelines (11). Antibiotics used (MAST Diagnostic Co, UK) included imipenem (10 µg), meropenem (10 µg), amikacin (30 µg), ceftazidime (30 µg), nitrofurantoin (300 µg), cipro oxacin (5 µg), gentamicin (10 µg), ampicillin (10 µg), cefalexin (30 µg), cefepime (30 µg), nalidixic acid (30 µg) and tobramycin (10 µg). Statistical analysis was performed using SPSS software version 16. effective antibiotics in different units of the two studied hospitals. Resistance to both carbapenems was observed in E. coli isolated from all samples except sputum. Also, bacteria isolated from urine, wound, stool and blood samples were resistant to all tested antibiotics in this study. Moreover, one E. coli was isolated from semen culture which was sensitive to aminoglycosides (amikacin and gentamicin) (Additional le 1: Table S1). According to the results, 221 (22.9%) isolates were resistant to only one antibiotic, while 7 (0.72%) isolates were resistant to at least 9 antibiotics. In this study, 24 (2.4%), 78 (8.09%), 125 (12.9%), and 85 (8.8%) isolates were resistant to all antibiotics tested belonged the class of aminoglycosides, cephalosporins, quinolones, and carbapenems, respectively. Further information on these results is provided in (Additional le 2: Table S2). Based on statistical analysis, a signi cant relationship was observed between resistance to all antibiotics, except nalidixic acid, and variables such as hospital wards of bacterial isolation, clinical sample type, age and sex of the patients (P <0.05). Also, 241 (25%) clinical isolates of E. coli were resistant to at least 3 antibiotics from 3 different classes and were considered as MDR isolates (Additional le 3: Table S3). Discussion E. coli is the most common urinary tract pathogen (12). Moreover, most cases of E. coli infection in this study were related to the urinary system, accounting for more than half of the cases. Also, studies carried out in Iran and other neighboring and non-neighboring countries have con rmed this result (13,14). In addition, women were the most infected individuals with E. coli (70.1%), which is consistent with other studies conducted in Tehran, Isfahan and Rasht (10,15,16). On the other hand, the bacterial isolates in the present study showed the lowest antibiotic resistance rates against amikacin in both hospitals. An Iranian study performed in 2019 showed that 2.2% of the clinical isolates of E. coli were resistant to this antibiotic (17). However, other studies conducted in Iran (18), Saudi Arabia (12) and Turkey (19), detected no amikacin resistant isolates of E. coli. These results indicate the good e cacy of amikacin as one of the effective antibiotics against infections caused by E. coli. Also, the results of this study showed low antibiotic resistance, except against ampicillin and cefalexin, in E. coli isolated from Bu-Ali Sina hospital compared to burn hospital. Due to the weakened immune system of burn patients, broad-spectrum antibiotics are used to treat bacterial infections in this group, which may increase the prevalence of antibiotic resistance (20). With the exception of carbapenems, the above results are consistent with other Iranian studies conducted in 2019 (21,22). The antibiotic resistance rate of E. coli strains in Turkey was also much lower than in the present study (20). However, their research conducted on 21 urine samples, which may be the reason for lower resistance rate. Differences in antibiotic resistance rates in different regions of the world can be due to genetic differences in resistant strains, various antibiotic use policies in regional hospitals, levels of arbitrary use of antibiotics in the community and various study time (23). Among the antibiotics studied in this study, ampicillin was the least effective ones with the highest resistance rate, which was similar to the results obtained in other Iranian studies carried out in Ahvaz, Rasht and Kashan (15,22,24). These results indicate that unlike amikacin, ampicillin is not suitable for the treatment of E. coli infections in Iran. Moreover, resistance to carbapenems among isolates collected from Bu-Ali Sina hospital in this study was similar to other studies in Iran and neighboring countries (13,19,21,22,25), while the reported carbapenem resistance rates in non-neighboring countries such as Malaysia and Nigeria were 29.2% and 19.7%, respectively (26,27). This discrepancies may be due to the differences in the type of sample, as Bu-Ali Sina hospital is a specialized pediatric hospital. However, resistance to carbapenems was observed extensively in the Zare as a burn hospital in the present study, which is justi able given that the hospital is a specialized burn center. Beside, quinolones and uoroquinolones such as nalidixic acid and cipro oxacin are also used to treat UTIs. Similar studies, in Ahvaz (22) and Sanandaj (21), reported 64.2% and 78.8% cipro oxacin resistance, respectively, which is consistent with the results of the burn hospital in the present study. In other studies (19,27), the rate of quinolone resistance was higher than that of uoroquinolones, which was consistent with the results of this study. This may be due to the greater use of drugs such as nalidixic acid to treat UTIs. Numerous factors, including supplementary use of antibiotics in agriculture and veterinary medicine, also contribute to the development of antibiotic resistance (28). Moreover, one of the serious problems worldwide is the multi-drug resistance and the development of MDR strains, which is a major concern in the treatment of infections caused by E. coli (8,29). In this study, 80.9% of the urinary isolates of E. coli were MDR, which was very similar to the study performed in Ahwaz (22). In the research of Badamachi et al, (18), 54.6% of urinary isolates were MDR, while only 1 to 5-year-old children were studied. Since the prevalence of MDR E. coli strains, especially in burn patients, can increase morbidity and mortality rates (29), so their presence can be a major risk factor for the patients. Understanding the resistance rates of the clinical isolates of E. coli to common antibiotics in hospitals is essential to control the antibiotic resistance of this bacterium. According to the results of the present study and other studies, aminoglycosides are almost effective drugs against E. coli infections, which can be suggested as the selective antibiotics in the treatment of resistant strains in this area. On the other hand, caring for the elderly and the high risk patients for the infection with E. coli, such as burn and ICU patients, is of great importance, which can lead to reduced infection-related mortality and reduced rates of disease transmission.

Limitations
The present study had some limitations; Lack of information on the underlying disease status of the patients, Also, in order to achieve better results, detection of common antibiotic resistance genes is also necessary.

Consent for publication
There is no limit to the publication.

Competing interests
The author reports no con icts of interest in this work Availability of data and materials Data generated or analyzed during this study are included in this published article.

Funding Sources
This study supported by Mazandaran University of Medical Sciences, Sari, Iran. year-old