Analysis of the Frequency, Antibiotic Susceptibility, and Related Genes among Foodborne Pathogenic Bacteria Isolated from Hospital Refrigerators in Tehran, Iran

Abstract

Background:

Hospital refrigerators as essential food storage can be important source of food contamination. We aimed to investigate the frequency and antibiotic susceptibility of the pathogenic bacteria in three hospital refrigerators in Tehran.

Methods:

This study was performed on 254 samples, collected from 60 refrigerators of the various wards of three hospitals, A, B, and C, in Tehran, Iran from 2020 to 2021. Following isolation and identification of isolates, the antibiotic susceptibility pattern was determined. PCR-based assays were used to screen the presence of antibiotic resistance genes of resistant isolates.

Results:

From 254 collected samples, 236 samples (92.9%) were contaminated. Most strains were isolated from refrigerators with poorly cleaned, temperatures above 8 °C in non-critical wards. Most bacteria belonging to Enterobacteriaceae (68.8%), followed by Staphylococcus (11.9%), and Enterococcus (10.6%), while the frequency of non-Enterobacteriaceae isolates was 8.9%. The highest antibiotic resistant bacteria were in extended spectrum beta-lactamase (ESBL) 9.7%, vancomycin-resistant enterococci (VRE) 5.3%, methicillin-resistant S. epidermidis (MRSE) 0.4%, and methicillin-resistant S. aureus (MRSA) 0.4%, respectively. The bla_OXA-48, bla_CTX, and bcla_TEM genes were found only in 10% of Enterobacteriaceae isolates. The bla_OXA-51 gene was found in all non-Enterobacteriaceae isolates. The vanA and mecA genes were detected in antibiotic-resistant Enterococcus and Staphylococcus.

Conclusion:

Our findings suggests major concern about cross-contamination and the emergence of antibiotic-resistant isolates as a potential health threat with hospital refrigerators origin. More attention to hospital refrigerators cleaning is necessary to prevent foodborne diseases and nosocomial infections.

Introduction

Improper cleanliness of surfaces, equipments, and places for food processing and storage may potentially accelerate the distribution of pathogenic bacteria, cross-contamination, and the prevalence of food-borne diseases (1, 2). Cross-contamination is the pollution of food from other sources. A good cleaning with appropriate disinfectants can help to prevent foodborne illness from occurring through cross-contamination (1). In addition, foods can be sources of the prominent pathogens of nosocomial infections (3). Therefore, this is a very important issue that needs special attention in refrigerators of various wards of hospitals, which can be a possible source of the prevalence of foodborne nosocomial infections in hospitalized high-risk patients (4). Taking antibiotics in such patients increases the chance of developing antibiotic resistance (5).

Currently, the worldwide prevalence of multi-drug resistant (MDR) bacteria are well-recognized to be one of the most common health issues (5, 6). The methicillin-resistant Staphylococcus (MRSA, MRSE), vancomycin-resistant Enterococcus (VRE), carbapenem-resistant Klebsiella, and extended-spectrum beta - lactamases (ESBLs)-producing Enterobacteriaceae are the most important MDR bacteria leading to increased morbidity and mortality of hospitalized patients (7, 8). Mechanisms of resistance to antibiotics may be related to the combination of the enzymatic degradation, changes in membrane permeability, and modification of target proteins (9). The resistance genes play a vital role in the virulence and antimicrobial resistance (10). The presence of antimicrobial-resistant microorganisms or antibiotic resistance genes in foods, it is a potential major public health challenge (11).

Although there are some reports on the bacterial contaminations in the surfaces, equipment, and environment of hospitals focusing on the antibiotic resistance (12, 13), less attention has been paid to food contaminated by antimicrobial-resistant bacteria in the hospital refrigerators.

The present study, for first time, aimed to investigate the presence, frequency, and distribution of the foodborne antibiotic-resistant bacteria in the hospital’s refrigerators in Tehran.

Materials and Methods

Bacterial isolates

Random sampling of 60 refrigerators in the wards of three hospitals with more than 400 beds (A, B, and C located in Tehran, Iran 2020 to 2021) was performed once a week for ten consecutive months. The sampling part of hospitals was critical and non-critical wards (ICU, CCU, and surgeries). Wet swabs were taken from the walls, floors, and the inner surface of the refrigerator doors. Samples were enriched for 24 h at 37 °C in tryptic soy broth (TSB) medium and then cultured in MacConkey, blood agar, and xylose lysine deoxycholate (XLD) agar followed by incubation for 24 h at 37 °C. After observing growth, the strains were then purified by re-culturing. The doubtful colonies were identified according to the Gram staining, and standard biochemical tests such as catalase, oxidase, coagulase, mannitol, glucose and lactose fermentation, hemolysis, indole, motility, citrate, urease, amino acid decarboxylase (lysine, ornithine, arginine), and MR-VP test (14). For further analyses, isolates were cultured in Luria-Bertani (LB) broth medium and were incubated at 37 °C for 24 h and their cultures were preserved at −70 °C supplemented with 20% glycerol.

Antimicrobial susceptibility testing

Routine laboratory antibiotics (UK MAST company) including amikacin (AMK: 30 μg), ampicillin (AMP: 30μg), aztreonam (ATM: 30 μg), ciprofloxacin (CIP: 5 μg), ceftriaxone (CRO: 30 μg), ceftazidime (CAZ: 30 μg), cefoxitin (FOX: 30 μg), gentamicin (GEN: 10 μg), imipenem (IMP: 10 μg), lioezolid (LNZ: 10 μg), meropenem (MEM: 10 μg), tetracycline (TE: 10 μg), tobramycin (TN: 10 μg), and vancomycin (VAN: 30 μg) were used for antibiogram testing. Antibiotic susceptibility of antibiotics was determined by disk agar diffusion method by using Müller-Hinton-agar (MHA) medium according to Clinical and Laboratory Standards Institute (CLSI) guidelines and the results were reported as sensitive (S), intermediate (I), and resistance (R) (15).

For vancomycin, its susceptibility/resistance pattern was firstly reported using VAN disc; ≥17 mm as sensitive, 15–16 mm intermediate and ≤ 14 mm as resistant. Moreover, agar dilution method using BHI agar was also done as per CLSI (Supplement M100; 2021) guidelines. E. faecalis ATCC 29212 (susceptible) and E. faecalis ATCC 51299 (resistant) were used as control strains. Presence of >1 colony indicated presumptive vancomycin resistance. Interpretive criteria were defined as per CLSI: MIC ≤ 4 sensitive, 8–16 intermediate and ≥ 32 as resistant.

DNA Extraction

DNA extraction from bacteria was performed by the boiling method according to the described method (16) and its quality was quantified with a nanodrop.

Detection of antibiotic resistance genes by PCR

The antibiotic resistance genes including bla_TEM, bla_CTX-M, bla_SHV (ESBL) and bla_OXA-48 (carbapenemases) in Enterobacteriaceae, bla_OXA-23, bla_OXA-24, and bla_OXA-51 (class D carbapenemases) in non-Enterobacteriaceae (Acinetobacter), vanA (VRE) in Enterococcus, and mecA (MRSA, MRSE) in Staphylococcus were assayed by PCR using specific oligonucleotide primers (Table 1). Amplification was performed using a BioRad MJ MiniTM PCR system with pervious reported protocols and the specific annealing temperature for each gene (Table 1) and the band pattern was analyzed by a gel documentation system (7, 8, 17). Positive control and negative control were included in each PCR run. Positive controls for ESBL genes and vanA were respectively E. coli ATCC 35218 and E. faecalis ATCC 51299. For bla_OXA-48, class D carbapenemases, and mecA, a carbapenemase-resistant Klebsiella, Acinetobacter, and an MRSA isolate, characterized previously in our laboratory by sequencing served as positive control.

Table 1:

The list of primers, annealing temperatures, and expected amplicon sizes for molecular detection of antibiotic resistance genes of isolates.

Group of Resistance	Type of bacteria	Gene	Primers (5′–3′)	Annealing	Product size (bp)	Ref.
ESBL	Enterobacteriaceae	bla TEM	F:TCCGCTCATGAGACAATAACC	57 °C	931	(17)
			R:TTGGTCTGACAGTTACCAATGC
		bla CTX-M	F: CTTCCAGAATAAGGGAATCCC		909
			R: CCGTTTCCGCTATTACAAAC
		bla SHV	F: TGGTTATGCGTTATATTCGCC	64 °C	868
			R:GGTTAGCGTTGCCAGTGCT
Carbapenemases	Enterobacteriaceae (Klebsiella)	bla OXA-48	F: GCGTGGTTAAGGATGAACAC	30 °C	438	(18)
			R: CATCAAGTTCAACCCAACCG
	Non-Enterobacteriaceae (Acinetobacter)	bla OXA-23	F:GATCGGATTGGAGAACCAGA	52 °C	501	(19)
			R:ATTTCTGACCGCATTTCCAT
		bla OXA-24	F:GGTTAGTTGGCCCCCTTAAA		246
			R:AGTTGAGCGAAAAGGGGATT
		bla OXA-51	F:TAATGCTTTGATCGGCCTTG		353
			R:TGGATTGCACTTCATCTTGG
VRE	Enterococcus	vanA	F:GGGAAAACGACAATTGC	50 °C	732	(20)
			R:GTACAATGCGGCCGTTA
MRSA, MRSE	Staphylococcus	mecA	F:GTAGAAATGACTGAACGTCCGATAA		310	(21)
			R:CCAATTCCACATTGTTTCGGTCTAA

Statistical analysis

The results were expressed as absolute frequencies and percentages. For the statistical analyses, the statistical software SPSS version 21.0 for Windows (IBM Corp., Armonk, NY, USA) was used. Analyses were performed with using oneway analysis of variance (ANOVA) and Tukey’s test to compare the differences between the means (P<0.05). Chi-Square test was used for the significant relationship between the groups. The curves were plotted using Excel software version 2016 (Microsoft Corporation, USA).

The experiments were carried out based on guidelines of the Research Ethical Committee of Islamic Azad University-Science and Research Branch.

Results

The frequency of isolates

Among 254 samples collected from the refrigerators of three hospitals, 236 (92.9%) samples showed bacterial growth while no bacterial growth were observed in 18 (7.1%) samples. Chi-Square test showed a significant relationship between the apparent cleanliness of refrigerators and the growth of bacteria (P<0.05). Most isolated bacteria were found at temperatures above 8°C (70/74; 94.6%) and the lowest bacteria were obsreved at temperatures below 2 °C (11/13; 84.6%). The isolated bacteria from the floor of refrigerators (91/93; 97.8%) was significantly higher than their door (84/88; 95.5%) and inner walls (61/73; 83.6%) (P=0.001). In addition, 90.8% (119/131) of samples of critical wards were positive, while this rate was 95.1% (117/123) for non-critical wards. In addition, 91.9% (204/222) of samples collected from weekly cleaned refrigerators were showed the bacterial contamination.

The isolates were divided into four bacterial groups: Staphylococcus, Enterococcus, Enterobacteriaceae and non-Enterobacteriaceae. The highest isolates belonged to Enterobacteriaceae and the non-Enterobacteriaceae was the lowest isolates (Fig. 1a). The Chi-Square analysis showed a significant association between the frequency of isolated bacteria with the wards of the hospital (P<0.05). Most isolates of Staphylococcus, Enterococcus, and non-Enterobacteriaceae groups were obtained from the critical wards, while most isolates of Enterobacteriaceae were isolated from the non-critical wards (Fig. 1b). The most isolates in each bacterial group were isolated from refrigerators with temperatures 2–8 °C (Fig. 1c). In all temperature ranges (<2, 2–8, and >8), the highest frequency isolates belonged to the Enterobacteriaceae (162/236; 68.6%). The highest frequency of Staphylococcus and Enterococcus isolates were obtained from the floors, door, and inner walls of refrigerators, respectively. Most of the Enterobacteriaceae isolates were identified in the refrigerator’s door, while the frequency of the non-Enterobacteriaceae isolates was equal in different parts of the refrigerators (Fig. 1d). Among the Enterobacteriaceae isolates, E. coli, K. pneumoniae, and P. aeruginosa were the frequent isolates, whereas the Salmonella spp were not found in any of the refrigerators.

Fig. 1:

The total frequency of bacteria groups (Staphylococcus, Enterococcus, Enterobacteriaceae. and non-Enterobacteriaceae) isolated from hospitals refrigerators (a), the frequency of bacteria groups in hospitals refrigerators based on ward type (b), temperature (c), and different parts (floor, wall, door) of refrigerators (d)

Determination of antibiotic susceptibility

The antibiotic susceptibility test (Table 2) showed that 18.2% (43/236) of isolates were antibiotic-resistant and 81.8% (193/236) of the isolated bacteria were sensitive to antibiotics. Among the antibiotic-resistant isolates, the highest percentage of resistance belonged to ESBL group followed by VRE. The lowest resistance was also related to MRSE and MRSA groups (Table 3).

Table 2:

Antimicrobial resistance profile of isolates

Antibiotic	No. (%) of antibiotic resistance isolates
	Enterobacteriaceae (n=162)			Non-Enterobacteriaceae (n=21)			Staphylococcus (n=28)			Enterococcus (n=25)
	S	I	R	S	I	R	S	I	R	S	I	R
Amikacin	162 (100)	0	0	21 (100)	0	0	-	-	-	-	-	-
Ampicillin	79 (48.8)	83 (51.2)	0	17 (80.9)	4 (19.1)	0	-	-	-	-	-	-
Aztreonam	159 (98.2)	1 (0.6)	2 (1.2)	16 (76.2)	0	5 (23.9)	-	-	-	-	-	-
Ciprofloxacin	159 (98.2)	2 (1.2)	1 (0.6)	16 (76.2)	0	5 (23.9)	-	-	-	-	-	-
Ceftriaxone	161 (99.4)	0	1 (0.6)	20 (95.2)	0	1 (4.8)	-	-	-	-	-	-
Ceftazidime	156 (96.3)	0	6 (3.7)	15 (71.4)	0	6 (28.6)	-	-	-	-	-	-
Imipenem	157 (96.9)	0	5 (3.1)	15 (71.4)	0	6 (28.6)	-	-	-	-	-	-
Meropenem	158 (97.5)	0	4 (2.5)	15 (71.4)	0	6 (28.6)	-	-	-	-	-	-
Tetracycline	162 (100)	0	0	17 (80.9)	0	4 (19.1)	28 (100)	0	0	25 (100)	0	0
Gentamycin 10	162 (100)	0	0	17 (80.9)	0	4 (19.1)	28 (100)	0	0	25 (100)	0	0
Gentamycin 120	162 (100)	0	0	-	-	-	-	-	-	25 (100)	0	0
Linezolid	162 (100)	0	0	-	-	-	-	-	-	24 (96)	0	1 (4)
Vancomycin	-	-	-	-	-	-	-	-	-	13* (52)	0	12** (48)
Cefoxitin	-	-	-	-	-	-	26 (92.8)	0	2 (7.2)	-	-	-

S: sensitive, I: intermediate, R: resistance

The vancomycin date repored based on MIC results:

^* MIC ≤ 4, sensitive;

^** MIC ≥ 32, resistant

Table 3:

Antibiotic resistance profile of isolates

Antibiotic-resistance group	Total frequency (%)	Bacterial group
		Enterobacteriaceae	Non-Enterobacteriaceae	Staphylococcus	Enterococcus
MRSA	1 (0.42)	0	0	1 (3.57)	0
MRSE	1 (0.42)	0	0	1 (3.57)	0
VRE	12 (5.10)	0	0	0	12* (48)
ESBL	10 (4.23)	10 (6.18)	0	0	0
Carbapenemases	13 (5.52)		13 (61.90)	0	0
Non-resistance	193 (81.77)	148 (91.36)	6 (28.58)	26 (92.85)	13 (52)
Other antibiotic-resistance group	6 (2.54)	4 (2.46)	2 (9.52)	0	0
Total	236 (100)	162 (100)	21 (100)	28 (100)	25 (100)

^* MIC ≥ 32, resistant

Antibiotic resistance genes analysis

The antibiotic-resistant bacteria were selected to investigate the presence of resistant genes. The PCR reaction confirmed the presence of resistance genes of bla_TEM, bla_CTX-M, and bla_OXA-48 in Enterobacteriaceae group, bla_OXA-23 and bla_OXA51 in non-Enterobacteriaceae, vanA in Enterococcus, and mecA in Staphylococcus in compared with positive controls (Fig.2). The frequency of antibiotic resistance genes revealed that the bla_TEM, bla_CTX-M, and bla_OXA-48 genes in 1 of 10 resistant Enterobacteriaceae, while the bla_SHV gene was not found in them. The presence of bla_OXA-48 gene confirmed the carbapenem-resistant Klebsiella. The bla_OXA-51 and bla_OXA-23 genes were respectively identified in 13 and 2 resistant non-Enterobacteriaceae isolates, while the bla_OXA-24 gene was not found in any of the isolates. The high frequency of bla_OXA-51 genes indicated that the prevalence of carbapenemase encoding Acinetobacter. The result also showed that one of 2 isolates of the MRSA/MRSE Staphylococcus group carried the mecA gene and 4 isolates of the VRE Enterococcus group (out of 12 cases) were encoded the vanA gene. The presence of MRSA and VRE isolates containing these genes in the hospital refrigerators.

Fig. 2:

Gel electrophoresis for characterization of resistance genes of mecA (a), vanA (b), bla_OXA-48 (c), bla_TEM (d), bla_SHV (e), and bla_CTX-M (f). C+: positive control; C−: negative control

Discussion

The role of hospital refrigerators in the food contamination and the prevalence of foodborne illness or nosocomial infections are important. This investigation provided detailed evidence about the presence, frequency, and distribution of the foodborne antibiotic-resistant pathogenic bacteria in the refrigerators of three hospitals in Tehran. Among 254 samples, 236 (92.9%) samples were shown bacterial contamination, and only in 18 (7.1%) samples, no bacterial growth was observed, which was close to the results of related studies performed on the bacterial and fungal contaminations in the home and restaurant refrigerators (22, 23). Most frequent isolates belonged to Enterobacteriaceae, Staphylococcus, Enterococcus, and non-Enterobacteriaceae, respectively.

We primarily examined the relationship between bacterial contamination and the appearance of the refrigerators in different wards of the hospitals. The lowest bacterial contamination was observed in the cleaned refrigerators (16.4%), that demonstrate that the regular and weekly cleaning of the refrigerators by means of appropriate detergents and disinfectants, can effectively reduce their bacterial contamination. The low temperatures, similar to the refrigerators, is another vital factor in the prevention of the growth of pathogenic bacteria in foods. When the appliance is full, the temperature may increase to a higher range via inhibition of air circulation, which can potentially affect the microbial contamination of the refrigerator (24, 25). The current findings were consistent with this principle.

In the current study, the highest isolated bacteria belonged to the floor and door. The preservation of different materials on the floors, and not cleaning the refrigerators at short intervals can be the reasons for high contamination of the floors (26). The higher bacterial contamination (90.8%) in the critical wards (ICU and CCU) in this study may reflected the improper cleaning of refrigerators located in those parts. The high prevalence of Enterobacteriaceae in non-critical wards may be attributed to the human origin that can possibly create foods cross-contamination by coliform bacteria due to poor hygiene. In addition, the resistance of these bacteria to detergents and antibiotics can be another factor. Here, among the common foodborne bacteria, E. coli, and S. aureus were the frequent isolates, whereas Salmonella spp were not found in any of the refrigerators, which was similar to the previous studies (27–29).

Antibiotic-resistant bacteria are a major cause of hospital-acquired infections and a serious public health concern (6). A high percentage of MRSA and ESBL resistant bacteria (~65%) were reported by Taheri in Arak hospital (22). Ekrami et al reported MRSA isolates in 60% of isolates (29). In another study, the frequency of MDR-MRSA was found at 91.6% in hospital’s kitchens samples and 18.7% in ICU (30). The results of these studies were higher than the results of the current study. In the current study due to our samples only collected from hospital’s refrigerators, the frequency of antibiotic-resistant bacteria were low, whereas in the previous studies samples have collected from different sources in the hospital.

The antibiotic resistance genes are the basis of the high prevalence of the antibiotic-resistant bacterial. The frequency of bla_TEM, bla_CTX-M, and bla_OXA-48 genes (10%) in isolated Enterobacteriaceae was similar to the Tamma et al results, which observed bla_CTX-M gene in 11% of gram-negative bacteria (31). The both bla_OXA-51 and bla_O-XA-23 genes were observed in 2 isolates, suggesting the carbapenemase-resistant isolates (32, 33). The co-existence of multiple resistance genes was probably due to the presence of multiple resistance plasmids in the isolates, which can enhance antimicrobial resistance (34).

The co-resistance to vancomycin and methicillin emphasizes that a concern regarding the emergence of MDR isolates. Vancomycin and methicillin resistance can be conferred by the vanA and mecA genes (35). In a study conducted by Hussein et al, out of 109 MRSA isolates, 55 (50.4%) isolates carried the mecA gene (36). In other studies, the frequency of vanA gene was found 30% (37) and 47.4% (38). The frequency of mecA and vanA genes in these studies were higher than the present study (mecA:0.4% and vanA:5.3%). Antibiotic resistance is often achieved by horizontal transfer of resistance genes. Plasmid-mediated transfer of antibiotic resistance play a crucial role in the transfer of resistance genes between different species (39). The simultaneous presence of VRE and MRSA/MRSE isolates can possibly transmit vanA and mecA genes to other Staphylococcus or Enterococcus and their resistance (8, 40).

This study was limited by its relatively small sample size. The study collection comprised only strains isolated within one year, collected from only the refrigerators of a few hospitals in Tehran, the capital of Iran and may not necessarily be representative of the epidemiology of foodborne pathogenic bacteria in other areas of Iran. The antibiotic-resistant isolates from the refrigerator varies from one hospital to another, depending on the local pattern of antibiotic resistance, infection control activities, and the level of public health. For this reason, it is necessary to perform further studies on antibiotic-resistant bacteria isolated from refrigerators of hospitals in various regions of Iran.

Conclusion

Although the strains isolated in our study did not have high resistance to the antibiotics tested in this study, there were only a few resistant strains with refrigerators origin with a potential to acquire other resistance. The utilization of broad-spectrum antibiotics by patients admitted to the hospital can be one of the effective factors in increasing the resistance of foodborne bacteria in the hospital environment. Therefore, it is recommended that the proper and timely cleaning of hospital refrigerators to prevent the prevalence of foodborne antibiotic-resistant bacteria and nosocomial infections.

Journalism Ethics considerations

Ethical issues (Including plagiarism, informed consent, misconduct, data fabrication and/or falsification, double publication and/or submission, redundancy, etc.) have been completely observed by the authors. Since this study did not include any human or animal sample,the institutional research board did not require to obtain an ethical code.