The present prospective study was carried out over a period of one and half year from December 2014 to May 2016. It was approved by the Institutional Ethical Committee. A total of 1310 samples like endotracheal aspirate, pus, urine, blood, sputum, catheter tip and ear swabs were collected from patients admitted in Intensive Care Unit, medicine, surgery, obstetrics gynaecology, otorhinolaryngology, paediatric wards and burn unit. A total of 925 males and 385 females between 5-75 years of age group, who developed VAP, SSI, UTI, septicaemia, Lower Respiratory Tract Infection (LRTI) and middle ear infection after 48 hours of admission were included in the study. Those who developed these infections before 48 hours of admission were excluded from study.

Direct gram stain was done from all specimens except blood. The specimens were inoculated onto Mac-Conkey’s agar, blood agar and cystine lactose electrolyte deficient (in case of urine sample) medium and incubated at 37°C for 24 hours in 7-10% CO₂ concentration. Those organisms which showed non-lactose fermenting colonies on Mac-Conkey’s agar and not acidify the butts of Triple Sugar Iron (TSI) agar were presumptively identified as non-fermenters and confirmed by standard microbiological techniques [1]. All the isolates were tested for anti-microbial susceptibility (Hi-Media Mumbai) by Kirby-Bauer disk diffusion method on Mueller- Hinton agar [7]. Meropenem resistant isolates were proceed to Modified-Hodge test, Combined disc test and AmpC disc test for detection of Carbapenemase, Metallo-β-lactamase and AmpC-β-lactamase.

Quality control strains were Escherichia coli ATCC 25922, for Modified Hodge test- Positive control- Klebsiella pneumoniae ATCC BAA-1705 and Negative control- Klebsiella pneumoniae ATCC BAA-1706.

Lawn culture of E.coli ATCC 25922 was made from an overnight culture suspension adjusted to 0.5 McFarland standards on Mueller-Hinton agar. After drying the plate, a 10 μg meropenem disc was placed at the centre and the test strain was streaked from edge of the disc to periphery of plate. The plate was incubated at 37°C for overnight. The presence of distorted zone of inhibition (clover leaf pattern) was interpreted as positive [Table/Fig-1] result [8].

[Table/Fig-1]:

Modified Hodge test.

Test organism was inoculated on Mueller-Hinton agar and two 10μg imipenem discs were placed. 10μl solutions (750μg) of Ethylene Diamine Tetra Acetic Acid (EDTA) were added to one of them & incubated the plate at 35°C for 16-18 hours. Metallo-β-lactamase positive [Table/Fig-2] result considered, if zone of inhibition of imipenem + EDTA disc was >7mm than that of imipenem disc alone [9].

[Table/Fig-2]:

Combined disc test.

Lawn culture of E.coli ATCC 25922 was made from an overnight culture suspension adjusted to 0.5Mc Farland standards on Mueller-Hinton agar plate. A 30μg cefoxitin disc placed in centre and a blank disc (6mm diameter) which was moistened with sterile normal saline and inoculated with few colonies of test organism, was placed beside the cefoxitin disc almost touching it. The plate was incubated at 37°C for overnight. A flattening or indentation of zone of inhibition of cefoxitin in the vicinity of the disc containing test organism was interpreted as positive [Table/Fig-3] result [10].

[Table/Fig-3]:

AmpC disc test.

Among 1310 patients, 181 showed NFGNB growth, therefore, isolation rate was 13.82%. Majority of non-fermenters were isolated from endotracheal aspirate, but isolation rate was maximum in surgical ward. Predominant isolates were Pseudomonas aeruginosa and Acinetobacter baumannii [Table/Fig-4]. All the isolates showed low sensitivity to piperacillin, ticarcillin and ceftazidime, however, most effective antibiotic in the present study was colistin followed by amikacin and imipenem [Table/Fig-5]. Meropenem resistance was 44.2% (80/181). Metallo-β-lactamase and AmpC β lactamase were produced by 56.82% (25/44) and 72.22% (26/36) of meropenem resistant Pseudomonas and Acinetobacter species respectively [Table/Fig-6].

Isolation rate of non-fermenters was 13.82%, higher than the reports by Benachinmardi et al., and Bruno et al., indicating the rising trend of non-fermenter infections which may be due to survival of organisms in health care setup, disruption of normal flora by excessive use of antibiotics, therapeutic intervention, increased duration of hospital stay, use of steroids and immunosuppressive therapy [11,12]. Non-fermenters were isolated most commonly from endotracheal aspirate indicating respiratory colonization of Pseudomonas and Acinetobacter species. The commonest isolate was Pseudomonas aeruginosa, which is similar to some studies [13,14] and higher isolation rate of Acinetobacter species corresponding to other study [15].

According to sensitivity pattern, antibiotics such as colistin, amikacin and imipenem are available for treatment of non-fermenters. Sensitivity rate was more in Pseudomonas aeruginosa than other species of Pseudomonas and resistance rate was higher in Acinetobacter baumannii than Acinetobacter lwoffii in most of the antibiotics. In the present study, amikacin sensitivity was 80% for Pseudomonas aeruginosa which is contrast to the finding of Kumar R et al., who have reported the same to be 32% [16]. Imipenem resistance was 62.1% in Pseudomonas aeruginosa whereas Malini et al., detected it 6% only [14]. In our study, high percentage of resistance has been detected against beta-lactam (piperacillin, ticarcillin) and cephalosporin (ceftazidime). This leaves us with limited therapeutic options, thereby, resulting in resistance among other group of antibiotics and such resistance have a potential for rapid spread since, they are usually plasmid mediated. The present study showed higher resistance in ciprofloxacin which can be explained by the fact of non-judicious use and easy availability even without prescription. Meropenem resistance was 40% in Pseudomonas aeruginosa and 54.39% in Acinetobacter baumannii which may be up to 42.7% [17] and 90.3% [18] indicate increasing trend of carbapenem resistance in non-fomenters. A total of 56.82% Metallo beta lactamase and 72.22% AmpC beta lactamase were detected among Pseudomonas and Acinetobacter isolates. Among carbapenem resistance acquisition, AmpC-beta-lactamase mechanism is more common than Metallo beta lactamase and carbapenemase other than Metallo beta lactamase in our health care setup. Overall our study has implicated the severity of carbapenem resistant non-fermenters, which are affecting the infection control activities among hospital setup.

Numbers of samples are less. Our study was limited to the detection of carbapenemase production only among carbapenem resistant non-fermenters isolated from inpatients.

Carbapenem resistance rate was more in Acinetobacter than Pseudomonas species. Our study highlights that no single phenotypic test is sufficient for detection and differentiation of carbapenem resistance among non-fermenters. Detection of carbapenem resistance (Metallo-β-lactamase, Carbapenemases other than MBL and AmpC β lactamase) is essential because they limit treatment options and can spread to susceptible bacteria by gene transfer. Antibiotics according to susceptibility pattern and infection control measures are necessary in prevention of emergence of non-fermenters in our teaching institute. However, higher resistance to beta lactam and cephalosporin is one of the alarming sign in our region.