Clinical Response and Outcome in Patients with Multidrug Resistant Gram-negative Infections.

OBJECTIVE: In this study, frequency and antimicrobial sensitivity pattern of multidrug resistant (MDR) microorganisms were evaluated in a referral teaching hospital in Iran.
METHODS: Patients with MDR Gram-negative pathogens were followed during the course of hospitalization. Demographic data, baseline diseases, type of biological sample, isolated microorganism, type of infection, antibiotic regimen before the availability of the culture result and change in the antibiotic regimen following receiving the antibiogram results, response to the treatment regimen, and duration of hospitalization and patient's outcome were considered variables for each recruited patient.
FINDINGS: In 71% of the patients, antibiotic regimens were changed according to the antibiogram results. A carbapenem alone or plus amikacin or ciprofloxacin were selected regimens for patients with extended-spectrum beta-lactamase (ESBL) infections. For patients with probable carbapenem-resistant Enterobacteriaceae infections, a carbapenem plus colistin was the most common antibiotic regimen. Clinical response was detected in 54.5% of the patients who were treated based on the antibiogram results. Clinical response was higher in the ESBL producers (ESBL-P) than the non-ESBL-P infections (75% vs. 52%). However, this difference was not significant (P = 0.09). Most nonresponders (80%) had sepsis due to Klebsiella species. Finally, 41.9% of the patients were discharged from the hospital and 58.2% died.
CONCLUSION: Same as other countries, infections due MDR microorganisms is increasing in the recent years. This type of resistance caused poor clinical response and high rate mortality in the patients.

INTRODUCTION

The rapid spread of resistance among common pathogenic microorganisms is a serious challenge around the word. This phenomenon affects antibiotics' effectiveness and limits available options for the treatment of common infections in human.[12]

Serious infections due to beta-lactamase producing microorganisms, especially in hospitalized patients are increasing now. Several mechanisms for antibiotic resistance have been introduced in Gram-negative bacteria. Both enzymatic and nonenzymatic pathways cause resistance to third-generation cephalosporins, aminoglycosides, fluoroquinolones, and carbapenems. Antibiotic resistance occurs following mutation in chromosomal genes or by horizontal transfer of genes between different microorganisms. The main mechanism of antimicrobial resistance in Enterobacteriaceae family is transferring of plasmid encoding extended-spectrum beta-lactamase (ESBL).[123456] ESBL producers (ESBL-P) are Gram-negative microorganisms which almost always belong to the Enterobacteriaceae species. These Gram-negative bacteria secret ESBL enzyme in periplasmic space and hydrolyze the beta-lactam ring in penicillins, cephalosporins, and aztreonam. In general, carbapenems and cephamycines are resistant to this enzyme. ESBL-P pathogens can cause severe and life-threatening infections such as bacteremia, sepsis, pneumonia, and meningitis.[34] In the United States, 26,000 infections and 17,000 deaths per 2012 were due to ESBL-P species.[5]

According to the Centers for Disease Control and Prevention (CDC) report, more than 19% of healthcare-associated infections are resistant to extended-spectrum cephalosporins. In the United States, 37% of nosocomial infections were due to ESBL-P Enterobacteriaceae species. The mortality rate was 57% more common in patients with bloodstream infection caused by ESBL-P than nonproducers.[5] Prolong hospitalization, presence of invasive medical devices, receiving total parenteral nutrition, age <12 weeks, prior treatment with cephalosporins and aminoglycosides, recent surgery, and hemodialysis are defined as risk factors for colonization with ESBL-P species.[7]

Antibiotic resistance is a critical issue in developing countries. The incidence of infections due to resistant microorganisms is increasing in the recent years in Iran.[89] In this study, frequency and antimicrobial sensitivity pattern of multidrug resistant (MDR) Gram-negative microorganisms were evaluated in a referral teaching hospital in Tehran, Iran.

METHODS

This cross-sectional study was performed between December 2014 and January 2016 in Imam Khomeini Hospital, a referral teaching hospital affiliated to the Tehran University of Medical Sciences, Tehran, Iran.

Patients with nosocomial infections (acquired 48–72 following the hospital admission) were included. Biologic clinical samples including urine, cerebrospinal fluid (CSF), blood, and tracheal secretions that were referred to the central laboratory department from different wards of the hospital were analyzed according to the clinical and laboratory standard institute instructions.[10] Antimicrobial sensitivity patterns of all isolates were recognized by using standard antibiotic disks on Mueller-Hinton agar. Following antibiotic disks from HiMedia, Bioscience Company, India, was used for the primary antibiogram and ESBL screening; ciprofloxacin (5 µg), ceftriaxone (30 µg), cefotaxime (30 µg), ceftazidime (30 µg), amikacin (30 µg), ampicillin-sulbactam (10/10 µg), imipenem (10 µg), and meropenem (10 µg). After 24 h of incubation, if an inhibitory concentration zone was <25 mm for ceftriaxone, 27 mm for cefotaxime or 22 mm for ceftazidime, phenotypic confirmatory test was performed with double disk synergy test. For this test, cefotaxime/clavulanic acid (30/10 µg) and ceftazidime/clavulanic acid (30/10 µg) discs were used.[10] Increasing of ≥5 mm in the inhibition zone diameter in double synergy test versus the antibiotic tested alone was considered in favor of ESBL-P isolates. Non-ESBL isolates that were resistant to imipenem or meropenem was categorized as probable carbapenem-resistant Gram-negative microorganisms.[10]

Patients, in whom MDR Gram-negative pathogens were confirmed phenotypically, were detected and followed by the clinical pharmacists during the course of hospitalization. MDR was defined as resistance to at least three classes of antibiotics (aminoglycosides, anti-MRSA cephalosporins, antipseudomonal penicillins + beta-lactamase inhibitors, carbapenems, and nonextended spectrum cephalosporins; first and second generation cephalosporins, extended-spectrum cephalosporins; and third and fourth generation cephalosporins, cephamycins, fluoroquinolones, folate pathway inhibitors, glycylcyclines, monobactams, penicillins, penicillins + beta-lactamase inhibitors, polymyxins, phosphonic acids, phenicols, and tetracyclines).[11]

Demographic data, baseline diseases, type of biological sample, isolated microorganism, type of infection, antibiotic regimen before availability of the culture result, and change in the antibiotic regimen following receiving the antibiogram results, response to the treatment regimen, duration of hospitalization, and patient's outcome were considered variables for each recruited patient. Cultures compatible with patient clinical status were measured as true infection according to the CDC definitions for health-care associated infections.[11] Patients with positive culture without these criteria were considered as colonized.

Statistical analyses were performed by the IBM Corp. Released 2012. IBM SPSS Statistics for Windows, Version 21.0. (IBM Corp., Armonk, NY). Continuous data were expressed as a mean ± standard deviation (SD). Categorical variables were reported as percentages. Chi-square or Fisher exact test was used for comparing the categorical variables between the groups. Continuous variables were compared by the independent t-test. P < 0.05 was defined as statistically significant.

RESULTS

During the study period, fifty patients with MDR Gram-negative infections including confirmed ESBL or probably carbapenem-resistant Enterobacteriaceae (CRE) were detected. The mean ± SD of patients' age was 59.02 ± 17.96 years old and thirty (60%) of them were males. Tracheal secretions (17 [34%]), urine (15 [30%]), blood (8 [16%]), soft tissue (3 [6%]), peritoneal fluid (2 [4%]), CSF (1 [2%]), and pleural fluid (1 [2%]) were positive in the patients. Most patients were hospitalized in Intensive Care Unit (35 [70%]) and general ward (8 [16%]), followed by emergency, neurosurgery, and Coronary Care Unit wards. Klebsiella species (78%), Escherichia coli (20%), and Enterobacter cloacae (2%) were isolated microorganisms from the patients' biological samples.

Antimicrobial sensitivity pattern of the microorganisms is shown in Table 1. Most active antibiotics were carbapenems and aminoglycosides, respectively. All of isolated E. coli and E. cloacae but only 56% of isolated Klebsiellas species were sensitive to carbapenems. The result of antimicrobial susceptibility tests revealed that 30% of the isolated microorganisms were resistant to carbapenems that may be CRE. However, most of these species (86.7%) were ESBL negative. According to the double disk synergy test, 17 (34%) of all isolates were ESBL-P and others were ESBL-negative. ESBL was positive in 58.8% and 41.2% of isolated Klebsiella species and E. coli, respectively. Based on the CDC definition, the clinical condition was compatible with the isolates in 62% of all patients and 25.8% of patients with ESBL-P infections. In 71% of the cases, antibiotic regimens were changed according to the antibiogram results. A carbapenem alone or plus amikacin or ciprofloxacin were selected regimens for patients with ESBL infections. For patients with probable CRE infections, a carbapenem plus colistin was the most common antibiotic regimen. Clinical response was detected in 54.5% of the patients who were treated based on the antibiogram results. Clinical response was higher in the ESBL-P than the non-ESBL-P infections (75% vs. 52%). However, this difference was not significant (P = 0.09). Most nonresponders (80%) had sepsis due to Klebsiella species. Finally, 41.9% of the patients were discharged from the hospital and 58.2% died. Characteristics of patients with ESBL-P and probably CRE infections were summarized in Tables 2 and 3, respectively.

Table 1

Resistance pattern of isolated Gram-negative pathogens

Table 2

Characteristics of patients with extended-spectrum β-lactamase-producing infections

Table 3

Characteristics of patients with probable carbapenem resistant enterobacteriaceae infections

DISCUSSION

Inappropriate antibiotic administration and consequent increasing number of MDR pathogens including ESBL-P and CRE are a serious worldwide concern in recent years.[121314] Rapid growing of ESBL-P and CRE among community and hospitalized patients is a global threat, especially in critically ill patients.[15] Considering that only limited new antimicrobial agents have been introduced in recent years; in some situations, we do not have an effective weapon against these pathogens.[1315]

Following extensive use of cephalosporins in last years, resistance rate of Enterobacteriaceae family to these agents is increasing around the world. Cephalsporins-resistant rate of these microorganisms was 30% among 11 countries of Asia in 2010. This rate received to 87% in 2014 at Latin America.[2]

Only in limited studies, the prevalence of ESBL-P pathogens was evaluated in Iran and ranged from 43.6% in Ilam to 74% in Milad Hospital.[15] However, the average rate of ESBL-P microorganisms was 42.2% in Iran.[16] In a recent study, more than 50% of isolated microorganisms from bile specimens were ESBL-P.[17] The most isolated ESBL-P were Klebsiella species followed by E. coli.[4] In European hospitals, more than 80% of isolated E. coli and Klebsiella pneumonia were belonged to the ESBL-P category.[2]

In the present study, the frequency of ESBL-P pathogens was lower than the previous reports from our country. In a report from three hospitals of Iran, all isolated ESBL-P microorganisms were sensitive to carbapenems.[15] However, in our study, some of ESBL-P species and most of ESBL negative strains were CRE. All CRE were Klebsiella species.

To interpret the result of clinical responses, limitations of the study should be considered. The sample size of the study was small for assessment of the treatment outcome. ESBL-P pathogens were identified phenotypically but were not confirmed by the genotypic assay method. Genotypic assay is not easily available method in our hospitals and only is applied for research purpose. There are several clinical diagnostic laboratory tests for detection of ESBL-P microorganisms.[18192021] Although double disk synergy test is a common and practical method for ESBL confirmation but some isolates may be missed by this test. The sensitivity of this method could be reduced by microorganisms that show low-ESBL activity.[18] It has been shown that 13.63% of ESBL positive strains were not recognized by double disk method.[18] Therefore, some of non-ESBL strains in our study may be false negatives of the test. In this study, CRE isolates were detected based on the results of the disk diffusion method and were not confirmed based on the phenotypic and genotypic assays.

Most MDR Gram-negative strains frequently carry both carbapenemase and ESBL genes. Specific methods such as bromic acid in combination with clavulanate are recommended to unmask the underlying ESBLs among Enterobacteriaceae family with carbapenemase enzyme. However, carbapenems-hydrolyzing ability of non-ESBL species is not impossible.[22]

Pharmacokinetic parameters such as inadequate tissue penetration of antimicrobial agents can influence the clinical responses in in vivo settings.[23] Most of the recruited patients had at least one of the following severe comorbidities including malignancies, respiratory disorders, ischemic heart disease, heart failure, diabetes mellitus, renal failure, cerebrovascular accident, hepatitis, immunodeficiency, and sepsis. A high rate mortality rate among our patients may be related to these conditions.

Unfortunately like other countries, CRE prevalence in our country is increasing in the recent years. Empiric administration of carbapenems should be restricted to patients with risk factors of infections with ESBL-P bacteria and in specific clinical situations. To limit the use of last-line antibiotics such as carbapenems, availability of accurate phenotypic, and genotypic methods for detection of ESBL-P and carbapenemase strains is essential in clinical practice.

AUTHORS' CONTRIBUTION

Masoume Malekolkottab contributed in data gathering. Lida Shojaei contributed in drafting the manuscript and data gathering. Hossein Khalili contributed in data interpretation and manuscript editing. Mahsa Doumanlu performed laboratory analysis.

FINANCIAL SUPPORT AND SPONSORSHIP

Nil.

CONFLICTS OF INTEREST

There are no conflicts of interest.