Risk factors for acquisition of extended spectrum beta lactamase producing Escherichia coli and Klebsiella pneumoniae in North-Indian hospitals

Abstract

Multidrug resistance and production of extended spectrum β-lactamases (ESBLs) by enteric gram negative rods in hospitals and community continue to be a matter of scientific concern. This retrospective study was executed to assess the prevalence of ESBL-producing Escherichia coli and Klebsiella pneumoniae at two North Indian hospitals and to determine the risk factors associated with the acquisition of these organisms. A total of 346 bacterial isolates were obtained. Of these, 48.27% (n = 167) were confirmed to be ESBL producers while 51.73% (n = 179) were non ESBL-producers. Among the ESBL producers, 55.69% (n = 93) were E. coli and 44.31% (n = 74) were K. pneumoniae. ESBL producing isolates showed co-resistance to multitude of antibiotics tested. Length of hospital stay (>3 days) and previous exposure to antibiotics were found as significant risk factors (p = 0.01 and 0.02) associated with the acquisition of ESBL-producing E. coli and K. pneumoniae isolates. Imipenem and meropenem can be suggested as drugs of choice in our study.

1. Introduction

Multidrug resistance and production of extended spectrum β-lactamases (ESBLs) by enteric gram negative rods in hospitals and community continue to be a matter of scientific concern. Continued selective pressure has lead to the emergence of multiple-drug-resistant (MDR) bacteria. Multidrug resistance is an inevitable genetic response to misappropriated exposures of pathogen populations to antibiotics (Sheikh et al., 2003).

MDR bacteria are difficult to treat (Shakil et al., 2008). There is a widespread occurrence of ESBLs, particularly in the hospital environment. ESBLs exhibited resistance to third-generation cephalosporins and are now a menace to hospitalized patients globally (Paterson and Bonomo, 2005; Pfaller and Segreti, 2006).

ESBLs are bacterial enzymes that confer resistance to advanced generation cephalosporins and can lead to the therapeutic failure of these agents. ESBLs are often plasmid associated and there can be cross-species dissemination of these plasmids. Moreover, these plasmids often carry genes encoding co-resistance to other antibiotics such as aminoglycosides, tetracyclines, chloramphenicol, fluoroquinolones and sulfamethoxazole–trimethoprim, to name a few. Concomitant β-lactam and aminoglycoside resistance is increasingly being reported (Ma et al., 2009).

Escherichia coli and Klebsiella pneumoniae remain the dominant ESBL-producing organisms isolated globally (Jacoby and Munoz-Price, 2005; Fatemeh et al., 2012). Antibacterial resistance of Enterobacteriaceae, especially the emergence of MDR strains is an important clinical problem worldwide (Giske et al., 2008; Abhijit et al., 2013; Majda et al., 2013).

Resistance to a wide variety of common antimicrobials has made the proliferation of ESBL-producing strains a serious global health concern that has complicated treatment strategies. The high proportion of ESBL producers among the Enterobacteriaceae and the complex molecular epidemiology with diverse types of ESBL genes involved therein are alarming. The aim of this retrospective study was to determine the prevalence and antimicrobial susceptibility pattern of ESBL producing enteric Gram-negative bacteria and also to identify the risk factors associated with the acquisition of these organisms.

2. Methods

2.1. Sample and data collection

The study was conducted for a period of 12 months (March 2013 to February 2014) at two North Indian hospitals (Lucknow region). Both of these were multispecialty hospitals (approximately 900 bedded each). A total of 346 isolates were obtained from clinical samples during this study admitted into various units like medical, surgical, pediatric, ophthalmologic, orthopedic units as well as neonatal intensive care units. The isolates were obtained from the cultures of specimens from patients hospitalized for >48 h. The ‘length of hospital stay’ was defined as the period between the date of admission of patients to the hospital and the date of sample collection. The sources of these isolates were urine (113), blood (43), sputum (36), eye discharges (76), surface swabs (57) and body fluids (21). Demographic and other relevant details were retrieved from patient record files. All the specimens were quickly sent to the laboratory to be processed for microbial pathogens. Standard methods for isolation and identification of these bacteria were used (Cowan and Steel, 1970).

2.2. Antimicrobial susceptibility testing and ESBL detection

Antimicrobial susceptibility testing of the isolates was performed by the standard disk diffusion method as recommended by the Clinical and Laboratory Standards Institute (CSLI, 2008). Isolates for which the minimum inhibitory concentration (MIC) of ceftazidime and/or cefotaxime was >2 μg ml⁻¹, which was obtained by microbroth dilution method, were classified as probable ESBL-producers. The results were reconfirmed by using HiComb MIC test strips as per the manufacturer’s instructions (Himedia, Mumbai, India). The ESBL phenotypic confirmatory test with ceftazidime, cephotaxime, ceftriaxone and cefixime was performed for all the probable isolates by disk diffusion method on Mueller–Hinton agar plates with and without 10 μg of amoxicillin/clavulanate. A ⩾5 mm increase in the zone of diameter of third generation cephalosporins, tested in combination with amoxyclav versus its zone when tested alone was considered indicative of ESBL production. E. coli ATCC 25922 was used as ESBL negative and K. pneumoniae 700603 was used as ESBL-positive reference strain.

2.3. Statistical analysis

The association of study variables with ESBL and non-ESBL-producing E. coli and/or K. pneumoniae acquisition status was tested by using Student’s t-test or Fisher’s exact test as appropriate. The odds ratios (ORs) with 95% confidence intervals (CIs) for having ESBL associated acquisitions were calculated. A two-tailed p value of <0.05 was taken as significant. Statistical analysis was performed using SISA software.

3. Results

A total of 346 bacterial isolates were obtained from clinical samples during the 12 month study period from patients to different units of the hospital. Of these 48.27% (n = 167) were found to be ESBL producers while 51.73% (n = 179) were non ESBL producers. Among the ESBL producers, 55.69% (n = 93) were confirmed to be E. coli while 44.31% (n = 74) were identified as K. pneumoniae. Antibiotic susceptibility tests were performed for ESBL producing isolates. Imipenem and meropenem were found to be the most effective antibiotics against the ESBL-producing E. coli and K. pneumoniae strains. Zero percent resistance was found against imipenem in the case of K. pneumoniae while it was 2.5% (n = 2) in the case of E. coli. 1.08% (n = 1) resistance was observed against meropenem in the case of E. coli while it was 2.70% (n = 2) in the case of K. pneumoniae isolates. 53.76% (n = 50) E. coli isolates were found to be resistant against cefepime while 51.35% (n = 38) resistant was observed for K. pneumoniae isolates. Both E. coli and K. pneumoniae isolates were found to be resistant against amikacin. Among the third-generation cephalosporins, maximum resistance was found against Ceftriaxone, which was 78.49% (n = 73) and 79.73% (n = 59) for E. coli and K. pneumoniae isolates, respectively. The percent resistance to selected antimicrobials exclusively among the ESBL-producing isolates has been shown in Table 1.

Table 1.

Percentage of resistance to the selected antimicrobial agents among the ESBL-positive isolates.

S. no.	Antibiotics	E. colin = 93			K. pneumoniaen = 74
		S	R	%R	S	R	%R
1	Ampicillin	08	85	91.39	07	67	90.54
2	Aztreonam	14	79	84.95	05	69	93.24
3	Amikacin	25	68	68.81	30	44	59.46
4	Cefotaxime	22	71	76.34	16	58	78.38
5	Ceftazidime	33	60	64.52	24	50	67.57
6	Cefazolin	25	68	73.11	18	56	75.68
7	Cefepime	43	50	53.76	36	38	51.35
8	Cefoxitin	34	59	63.44	19	55	74.32
9	Ceftriaxone	20	73	78.49	15	59	79.73
10	Ciprofloxacin	40	53	56.99	20	54	72.97
11	Piperacillin	24	69	74.19	24	50	67.57
12	Tigecycline	36	57	61.29	21	53	71.62
13	Gentamicin	23	70	75.27	16	58	78.38
14	Imipenem	91	02	2.15	74	00	00.00
15	Meropenem	92	01	1.08	72	02	2.70

S = Sensitive, R = Resistant.

The mean age of patients harboring ESBL producing isolates was found to be 38.65 years (SD 11.32) while the same was 39.84 years (SD 12.15) for the patients harboring non-ESBL producing E. coli and K. pneumoniae strains. Mean duration of hospital stay was observed as 5.63 (SD 1.15) days among patients harboring ESBL producing organisms while the same was 2.13 (SD 0.56) days among patients with non-ESBL producing isolate acquisition. The length of hospital stay (>3 days) was found to be a significant risk factor for acquisition of ESBL producing isolates (p = 0.01). Prior exposure to beta lactams, and Aminoglycoside/fluoroquinolone antibiotics was also found as a significant risk factor for ESBL producing isolate acquisition (p = 0.04 and 0.02). Stay in ICU, presence of an invasive device and underlying illness were not found as significant risk factors in the present study. The demographic and clinical characteristics of the patients have been shown in Table 2.

Table 2.

Study characteristics in two groups of patients according to acquisition of ESBL-producing and Non-ESBL-producing E. coli and/or K. pneumoniae isolates.

Characteristics	Acquisition of ESBL Producing E. coli and/or K. pneumoniaen = 167	Acquisition of non-ESBL Producing E. coli and/or K. pneumoniaen = 179	p-Value	OR (95% CI)
Mean age	38.65 (SD 11.32)	39.84 (SD 12.15)	0.10	1.19 (0.65–3.03)
Sex
Male	116 (69.46)	95 (53.07)	–	1.00
Female	66 (39.52)	69 (38.54)	0.27	1.28 (0.83–1.97)
Length of hospital stay
⩽3 days	51	63	–	1.00
>3 days	136	96	0.01⁎	0.57 (0.36–0.90)
Mean duration of hospital stay (days)	5.63 (SD 1.15)	2.13 (SD 0.56)	1.00	3.50 (3.41–3.58)
Underlying illness
Diabetes	43 (25.75)	61 (34.08)	0.09	0.67 (0.42–1.07)
Typhoid	27 (16.17)	31 (17.32)	0.77	0.92 (0.52–1.62)
Conjunctivitis	35 (20.96)	41 (22.91)	0.66	0.94 (0.72–1.23)
Others	51 (30.54)	57 (31.84)	0.68	1.10 (0.60–1.75)
Stay in ICU	57 (34.13)	46 (25.70)	0.09	1.50 (0.94–2.38)
Non ICU	113 (67.66)	130 (72.63)	0.31	0.79 (0.50–1.25)
Ventilated	67 (40.12)	59 (32.96)	0.17	1.36 (0.88–2.11)
Non ventilated	109 (65.27)	111 (62.01)	0.53	1.15 (0.74–1.79)
Exposure to antibiotics
Beta-lactams	37 (22.16)	57 (31.84)	0.04⁎	0.61 (0.38–0.99)
Aminoglycosides/fluoroquinolones	27 (16.16)	47 (26.25)	0.02⁎	0.52 (0.22–0.92)

OR = odds ratio, CI = confidence interval, SD = standard deviation.

^⁎P value <0.05 is significant.

4. Discussion

ESBLs have become widespread in hospital as well as community settings (Pfaller and Segreti, 2006; Shakil et al., 2010). These enzymes are becoming increasingly expressed by many strains of pathogenic bacteria with a potential for dissemination. Presence of ESBLs compromises the activity of a wide spectrum of antibiotics creating major therapeutic difficulties with a significant impact on the outcome of patients. The continued emergence of ESBLs presents a serious diagnostic challenge to the clinical microbiology laboratories (Meeta et al., 2013).

In the present study, we observed that 55.69% (n = 93) of the E. coli and 44.31% (n = 74) of the K. pneumoniae isolates were ESBL producers. ESBL production was more common among the E. coli isolates as compared to the K. pneumoniae isolates which is in harmony with the finding of other studies (Tankhiwale et al., 2004; Babypadmini and Appalaraju, 2004; Umadevi et al., 2011).

The ESBL producing isolates showed co-resistance against most of the antibiotics tested, which was consistent with most of the recent findings (Fanos et al., 2007; Shakil et al., 2009; Shahla and Samaneh, 2010; Shanthi and Sekar, 2010; Majda et al., 2013). ESBL-producers are frequently resistant to advance generation cephalosporins. However, in the present study some ESBL producing isolates showed susceptibility to advance generation cephalosporins (cefotaxime, ceftazidime, and ceftriaxone). This could happen due to the incidental presence of various other mechanisms of resistance and counter resistance in a given bacterium. This was in harmony with previously published studies in which authors have reported ESBL producing isolates that were susceptible to advance generation cephalosporins (Shakil et al., 2009; Thakur et al., 2013; Ahmed et al., 2013). All ESBL producing E. coli and K. pneumoniae isolates were sensitive to imipenem and meropenem. This is in harmony with the findings of Iraj and Nilufar, 2010. The authors have suggested carbapenems as drugs of choice for the treatment of ESBL related infections (Iraj and Nilufar, 2010).

The length of hospital stay (>3 days) and prior exposure to beta lactam and aminoglycosides/fluoroquinolones antibiotics were also found as significant risk factors (p = 0.01 and 0.02, respectively) for acquisition of ESBL producing isolates. This is coherent to the finding of Shanthi and Sekar, 2010. Underlying illness, presence of an invasive device and prolonged disease were not found to be significantly associated with the acquisition of ESBL-producing E. coli and K. pneumoniae among the patients studied.

ESBL producing gram negative bacteria have been responsible for numerous outbreaks of infection worldwide and therefore, pose a challenge to efficient infection control. So it is essential to identify ESBLs as a routine in the hospitals. ESBL-detection in a hospital indicates the need for use of appropriate antibiotic prescribing practices. ESBLs are found in certain genera of the Enterobacteriaceae family including E. coli and K. pneumoniae and other bacteria like Staphylococcus aureus, Haemophilus influenza and Pseudomonas aeruginosa (Vandana and Honnavar, 2009). The occurrence of ESBL among clinical isolates varies greatly worldwide and geographically and is rapidly changing over time (Babypadmini and Appalaraju, 2004).

During the last several decades, the prevalence of MDR organisms in hospitals and medical centers has increased steadily. The prevalence of Gram negative bacteria resistant to third generation cephalosporins, fluoroquinolones, carbapenems, and aminoglycosides has also increased (Kritu et al., 2013).

ESBL-producing organisms pose unique challenges to clinical microbiologists, clinicians, infection control professionals and scientists engaged in finding new antibacterial molecules. ESBL-producing strains are usually found in those hospitals where antibiotic use is frequent and the patients are in critical condition (Shahanara et al., 2013).

Since ESBL production is more common among the nosocomial pathogens, early detection will definitely help in controlling hospital infections which are caused by this group of organisms. Enterobacteriaceae are the common isolates in most of the laboratories. Now-a-days, a majority of these isolates are MDR. The control of these MDR organisms is a therapeutic challenge. This difficulty is enhanced further by the co-existence of the resistance to β-lactams, aminoglycosides and fluoroquinolones, as observed in our study. Of all the available antimicrobial agents, carbapenems are the most effective and reliable treatment options for infections which are caused by the ESBL producing isolates (Paterson and Bonomo, 2005).

5. Conclusion

Problems associated with ESBL producing isolates include multidrug resistance, difficulty in treatment, and increased mortality of patients. Most of our study isolates were found to be resistant against four or more antibiotics. Imipenem and meropenem can be suggested as drugs of choice in our study. Length of hospital stay and prior exposure to antibiotics were found to be significant risk factors associated with ESBL-producing E. coli and/or K. pneumoniae acquisition status of patients. Therefore, restricting the use of third-generation cephalosporins, along with implementation of infection control measures, are the most effective means of controlling and decreasing the spread of ESBL producing pathogens.