Antimicrobial susceptibility patterns among bacteria isolated from intensive care units of the largest teaching hospital at the northwest of Iran

1Applied Drug Research Center, Tabriz University of Medical Sciences, Tabriz, Iran, 2Student Research Committee, Tabriz University of Medical Sciences, Tabriz, Iran, 3Faculty of Medicine, Tabriz University of Medical Sciences, Tabriz, Iran, 4Tabriz Health Services Management Research Center, Tabriz University of Medical Sciences, Tabriz, Iran, 5Cardiovascular Research Center, Tabriz University of Medical Sciences, Tabriz, Iran


INTRODUCTION
Patients a d m i t t e d t o intensive care units (ICUs) are more vulnerable to various nosocomial infections because of their underlying illnesses and exposure to various invasive medical devices (Jatin, Mary, 2012;Archibald et al., 1997;Fridkin, 2001;Kollef, Fraser, 2001;Rhomberg et al., 2006).
Antibiotics are the mainstay tools in the management of infections.Disappointingly, a number of studies have reported inappropriate uses of antimicrobials as well as the emergence of antibiotic resistance in most parts of the world (Tunger et al., 2009;Bergman et al., 2006).Today, the emergence of multidrug-resistant (MDR) gramnegative and methicillin-resistant Staphylococcus aureus (MRSA) pathogens has become a major public health concern.Moreover, they negatively influence patients' outcomes and increase the length of hospital stays and healthcare costs (Giske et al., 2008;De Kraker et al., 2011;Resch, Wilke, Fink, 2009;Mulvey, Simor, 2009).
Many studies have also showed that controlled use of antibiotics leads to improved microorganism sensitivity to antimicrobial agents (Lee et al., 2010;Takesue et al., 2010).The emergence of resistant pathogens and their negative impact on patient survival and healthcare costs need the evaluation of antimicrobial susceptibility in each health center and hospital in order to control infection.Therefore, this study assessed antimicrobial susceptibility patterns in the ICUs of the largest teaching hospital in northwestern Iran.To the best of the authors' knowledge, this is the first report of bacterial resistance in ICUs at this location.

Study Design and Setting
A retrospective study was done on laboratory records of the positive cultures patients with nosocomial infection who were admitted at the five ICUs of Imam Reza Hospital, a 500-bed tertiary care hospital affiliated to Tabriz University of Medical Sciences during a 21-month period (March 2010 to January 2012).This study was approved in the ethic committee of the university.

Data and Sample Collection
For doing the study a questionnaire form was designed in order to record patients' data included age, sex, file number, ward, culture results, site of infection, diagnostic species, and antibiogram test results.Prescribed drugs for ICU cultures were demonstrated using the hospital information system (HIS).All samples were taken in aseptic conditions.The sources of specimen included blood, urine, sputum, bronchial washing, wound, catheter, cerebrospinal fluid (CSF), tracheal, peritoneal, pleural, and synovial fluids.

Antimicrobial Susceptibility Testing
The Kirby-Bauer disk diffusion method was used to test antimicrobial susceptibility (in Mueller-Hinton agar medium, Merck, Germany) according to the guidelines of the National Committee for Clinical Laboratory Standards (NCCLS, 1997).

Determination of Pattern of Antibiotic Prescriptions
A clinical pharmacist and two intensive care subspecialists evaluated the appropriateness of prescribed antibiotics based on result of each isolated cultures from ICU patients.
Patterns of antibiotic prescription based on culture results were divided into four groups: Group 1) Adherence with culture results, according to culture the drug is added or discontinued.Group 2) Not adherence with culture results as follows: 2-1) No change in medication; While, culture results showed resistance to antibiotics or following antibiotic therapy.We need to change the antibiotics but physician did not notice to culture results and antibiotics continued.2-2) inappropriate changes in medication: the doctor noticed that you need to change the antibiotic, but changes were not accordance to protocols and appropriate drug that affect antibiotic therapy to improve patient has not chosen.
Group 3) rational change without culture result guide: due to the lack of laboratory antibiogram test for all antibiotics, a rationale prescription of drugs by doctors was made.
Group 4) No judgment: Because there was no complete transparency in the medical record for patient, judgment in this case considered impossible by experts.
Antibiotics utilization rates in ICUs were determined by Defined Daily Dose (DDD) per 100 bed-days.

Data Analysis
Data were processed in SPSS 16. Results were shown as standard deviation f o r quantitative variables and percentage for categorical variables.

RESULTS
A total number of 556 cultures isolated from 328 patients who were admitted at the five ICUs of the hospital were included in the study.Of which, 55.5% (n=182) were males and 44.5% (n=146) were females.The mean (SD) forage of the patients was 54± 19.5 years old.Data were collected from the five ICUs of the hospital as follows: 31.1% (n=102) from brain ICU, 22.6% (n=74) neurology ICU, 28.4% (n=93) lung ICU, 16.5% (n=54) surgery ICU and 1.5% (n=5) from the general ICU.As shown, the brain ICU has the greatest number of patients.
Among Gram-positives, Staphylococcus aureus was the most frequently isolated pathogen (39.7%, n=29) followed by Staphylococcus epidermidis.Susceptibility rates of Gram-positive and Gram-negative microorganisms are shown in Table II and III respectively.Based on these findings, rate of methicillin resistance Staphylococcus aureus (MRSA) was 87.5%.However, all cases of Staphylococcus aureus and 90% of Staphylococcus epidermidis were sensitive vancomycin.
Overall susceptibility rates of the microorganisms isolated from the ICUs are also shown in table IV.The most active antimicrobials were vancomycin (93.5%) followed by amikacin (71.5%) and gentamicin (46%).The overall susceptibility of isolated microorganisms to regular important antibiotics was as follows: 36% for ciprofloxacin, 19% for imipenem, 20% for trimethoprimsulfamethoxazole, 20,5%for ceftazidime, and 12% for ceftriaxone.
The most Staphylococcus strains were observed in lung and blood samples (31% and 17.2% respectively).
Enterobacter and E.coli were the most commonly isolated pathogens from urine samples (37.7% and 36.2%respectively as well as the most frequently isolated microorganisms from lung samples were Enterobacter (52.7%) followed by Pseudomonas aeruginosa (10.2%) and Acinetobacter (9.1%).Entrobacter aerogenes was the most common strain in all samples.Entrobacter aerogenes and E.coli was the most commonly detected microorganisms in most ICUs.
Pattern of antibiotic prescription showed that only 22.5% of prescriptions were in accordance with culture results and 34.5% were not adherence with culture results.Moreover, in 38% of prescriptions logical changes were made without culture result guide.In 5%, no judgment could be made.

DISCUSSION
To the best of the authors' knowledge, this is the first report from the ICUs of the largest teaching hospital in northwestern Iran.The result of this study highlighted the important key role of rational prescription of antibiotics.This matter becomes more important when it concerns the emergence of antimicrobial resistance produced by gramnegative bacilli and MRSA strains (Giske et al., 2008; De Kraker et al., 2011;Resch, Wilke, Fink, 2009;Mulvey, Simor, 2009).Hence, the detection of antimicrobial resistance patterns in every hospital plays an important role in the management of infections.
The SENTRY Antimicrobial Surveillance Program which included 25 ICUs in Europe (1997)(1998) and North America (2001) reported S. aureusas the most common pathogen in developed countries followed by P. aeruginosa and E. coli (Streit et al., 2004;Fluit et al., 2001).However, this pattern in developing countries such as Brazil (Mendes et al., 2005), Turkey (Kucukates, 2005), Iran (Mohammadtaheri et al., 2010), and other Middle Eastern countries were documented as Pseudomonas and Acinetobacter (Aly, Al-Mousa, Al Asar, 2008;Bayram, Balci, 2006).In the present study, Enterobacter aerogenes followed by E. coli and P. aeruginosa were the organisms most frequently recovered from ICUs.In a Serbian report, Enterobacter strains were the bacteria most commonly isolated from ICUs (Djordjevic et al., 2012).
In the current study, Escherichia coli (36.2%) and Enterobacter aerogenes (52.7%) were the isolates most commonly found in urinary and respiratory tracts.These findings are in agreement with those of the study conducted by Streit et al. in North America (Streit et al., 2004).However, in a report from Oman, Escherichia coli and Pseudomonas aeruginosa were the pathogens most commonly found in the urinary and respiratory tracts of ICU patients (Al-Lawati, Crouch, Elhag, 2000).Whereas, Klebsiella and Pseudomonas were the microorganisms most commonly found in the urine of patients admitted to the ICU of Imam Khomeini Hospital in Tehran (Hadadi et al., 2008).
In the current study, 86.2% of microorganisms isolated from urine were gram-negative bacteria.This finding is in line with a 2003 NNIS study that showed the highest rates of gram-negative infections to be in urine (Gaynes, Edwards, 2005).
The most common gram-positive organisms recovered from the ICUs in this study were S. aureus followed by S. epidermidis.This result is in agreement with the findings of Mohammadtaheri et al. (2010) and Khalili et al. (2012) in their studies conducted in ICU and infectious ward settings in Iran.().
All MRSA isolates in the current study were susceptible to vancomycin; this is in line with the report from Iran and the CAN-ICU study (Mohammadtaheri et al., 2010;Khalili et al., 2012;Zhanel et al., 2008).The other active agent against MRSA was trimethoprimsulfamethoxazole (46.7%); this result is similar to that of Mohammadtaheri et al. (51.9%) (Mohammadtaheriet al., 2010), but it is far from that of the CAN-ICU study (11.7%) (Zhanel et al., 2008).
Currently, MRSA is an emerging problem associated with a higher rate of mortality and increased durations of hospitalization and healthcare costs (Giskeet al., 2008;De Kraker et al., 2011).This higher rate of MRSA should raise the concerns of health authorities and lead to a larger battle against the propagation of infections.
Recent data from Iran revealed that the resistance of microorganisms to imipenem has increased.The susceptibility rates of pathogens to imipenem in earlier studies from Iran were about 98% and 87.5% in 200687.5% in and 200887.5% in , respectively (Vessal et al., 2006;;Hadadi et al., 2008).In 2010, however, the authors of the current study reported the decline of these rates to 2-37% and 19%, respectively (Mohammadtaheri et al., 2010).Accordingly, it may be concluded that the appearance of pathogens resistant to imipenem is increasing in Iran with the passing of time and consumption of the drug.
C u r r e n t r e s u l t s a l s o s h o w e d t h a t 2 5 .8 % of Acinetobacter, 20% of Klebsiella, and 16.6% of Pseudomonas were an MDR gram-negative pathogen.In one study from Turkey from 2000 to 2002, 45.4% of Acinetobacters and 37.7% of P. aeruginosa isolates were MDR (Yaman et al., 2004).In another study conducted in Iran, 80% of Acinetobacter and 59% of Klebsiella isolates were MDR (Hadadi et al., 2008).In the Karlowsky et al. (2003) study, among Acinetobacter isolates, the MDR rate was 11.6-24.2%.
MDR gram-negative bacilli-related infections have become a major, rising problem (Thomson, Bonomo, 2005).Despite the growth of MDR strains, very few antimicrobials have been developed to combat these infections (Rubinstein, Vaughan, 2005).The growing body of evidence also proves the development of resistant gram-negative infections without adequate therapeutic options (Falagas et al., 2005).Moreover, MDR gramnegative infections are associated with increased mortality rates, extended durations of hospitalization, and higher healthcare costs (Bergman et al., 2006).
At the end, the results of the present study showed a higher rate of antimicrobial resistance patterns in the ICUs of the largest referral tertiary teaching hospital in northwestern Iran.The irrational use of antibiotics reported in this study may partially justify these results.Therefore, detecting and managing bacterial resistance should be given more attention by clinicians as an important part of any infection control program.Additionally, the development of antibiotic stewardship programs in hospitals, encouragement of rational antibiotic use, and national and local antibacterial surveillances and control programs are recommended to reduce the development of resistant strains and encourage the successful management of infections.
This study has some limitations.Because of its retrospective design, access to clinicians and a medical menu was not possible.Therefore, a clinical judgment about the administration of antibiotics in some cases was not applicable.

CONCLUSION
This study showed a higher rate of bacterial resistance in the ICUs of a tertiary teaching hospital in Iran.Due to the emergence of resistant microorganisms in ICUs and considering the critical condition of patients in such a setting, more surveillance and control of the use of antibiotics is needed to combat infections.

TABLE I -
Bacterial typing of ordered cultures Bacterial type

TABLE II -
Susceptibility rates of the most common Gram-positive cocci isolated from the ICUs

TABLE III -
Susceptibility rates of Gram-negative bacilli isolated from the ICU

TABLE IV -
Overall susceptibility rates of the microorganisms isolated from the ICUs