Prevalence , antimicrobial resistance and risk factors for campylobacteriosis in Lebanon

Introduction: The rapid increase in Campylobacter strains resistant to antibiotics represents a major problem for public health. In Lebanon, campylobacteriosis is underdiagnosed since bacteria detection in stool samples is not performed routinely. This study aims to evaluate the prevalence, sources and routes of transmission, risk factors and antimicrobial susceptibility patterns of Campylobacter spp. in Lebanon. Methodology: Stool samples collected from 1000 Lebanese patients with diarrhea, and 150 meat samples taken from supermarkets and slaughterhouses were subjected to Campylobacter detection. Colonies were identified by Gram staining, oxidase and catalase activities. They were then differentiated at the species level by hippurate test and PCR. Susceptibility of Campylobacter spp. to antibiotics was studied by the disc diffusion standard method. Results: Campylobacter spp. were detected in 21.5% of stool samples; the main isolated species being C. jejuni (83.2%) and C. coli (13.9%). The highest Campylobacter infection rates were detected among children (41.8%) and during summer (31.6%). Consumption of contaminated meat and salads, and contact with animals represented the major risk factors for campylobacteriosis, with poultry carcasses and bovine cuts identified as the main bacteria reservoirs. Neither demographic determinants nor season had a major effect on the prevalence of campylobacteriosis. Erythromycin was the most active agent against Campylobacter spp. A multi-resistance rate was observed in 35.9% of isolates. Conclusions: Campylobacteriosis is a major public health concern in Lebanon. Bacteria detection in stool culture should be performed routinely to allow an early diagnosis and a better monitoring of the disease and its burden.


Introduction
Campylobacteriosis is a foodborne diarrheal illness caused by bacteria of the genus Campylobacter, with most cases caused by Campylobacter jejuni and Campylobacter coli.The most common clinical symptoms of Campylobacter infections include diarrhea, abdominal pain, fever, headache, nausea and/or vomiting [1].The incidence of campylobacteriosis has gradually increased over the last 10 years in both developed and developing countries [2].According to the European Food Safety Authority (EFSA), Campylobacter spp.are now considered to be the leading cause of bacterial gastroenteritis worldwide with higher occurrence rates of Campylobacter infections observed in children under 5 years old [3,4].
Consumption of contaminated and undercooked poultry meat has been considered as the main source of Campylobacter infections in humans [1,5,6].Raw milk and dairy products [7], contaminated water [8,9], and contact with pets and farm animals [10,11] have been also identified as possible sources for campylobacteriosis, and have contributed to outbreaks of the disease worldwide.
Although most Campylobacter infections are mild, self-limiting and usually resolve within a few days without antibiotic treatment, severe gastrointestinal disease can occur particularly in immunocompromised persons.Moreover, postinfectious complications of campylobacteriosis, including Guillain-Barré syndrome and reactive arthritis, can occur in both immunocompromised and immunocompetent persons [12].
When clinical therapy is warranted, erythromycin is considered the drug of choice.Fluoroquinolones have been also frequently used owing to their broad spectrum of activity against enteric pathogens.A rapid increase in the proportion of Campylobacter strains resistant to macrolides and fluoroquinolones has been reported in several countries worldwide, and therefore represents a major concern for public health [1,13,14].
Even though gastroenteritis is one of the leading cause of hospitalization in Lebanese infants under one year, research work on Campylobacter human infections are still very limited in Lebanon.Few studies have investigated the prevalence and antimicrobial susceptibility of Campylobacter spp.isolated from humans and poultry in Lebanon, and results have shown that Campylobacter is rare in diarrheic stools of patients compared to Salmonella and Shigella.Indeed, Campylobacter infection may be significantly underdiagnosed because its search is not part of the routine stool testing, and its culture is demanding and more difficult than that of other bacteria [15][16][17][18].Moreover, these studies highlighted the increased rates of resistance of Campylobacter isolates to antibiotics, and emphasized the need for assessing the routes of transmission of campylobacteriosis to humans and animals [15][16][17][18].These preliminary and inconclusive data warrants therefore further investigations to estimate the true prevalence of campylobacteriosis in Lebanon, and to better understand the involvement of Campylobacter species in the pathogenesis of gastroenteritis.
This work is carried out to assess the epidemiological and clinical profiles of Campylobacter spp. in stool cultures from Lebanese patients with diarrhea.It also aims to determine the antibiotic susceptibility pattern of isolated species, as well as sources of transmission and risk factors contributing to the emergence of the disease in Lebanon.

Patients
The study was conducted in eight districts belonging to the North, South and Mount Lebanon Governorates, according to the Declaration of Helsinki and in agreement with standards of the Ethical Committee of Notre Dame University Hospital.Subjects enrolled in the study were Lebanese outpatients who attended hospitals and medicals centers with three or more watery loose stools in 24-hour period at any time of day between September 2016 and August 2017.After obtaining a written informed consent from all participants, a questionnaire was filled to gather relevant information regarding demographic and clinical data, dietary intake, treatment, risk factors, etc. Patients treated with antibiotics and suffering from chronic diseases were excluded from the study.

Samples collection
Stool samples were collected from all patients enrolled in the study.A total of 150 meat samples, commonly consumed by the Lebanese population (poultry meat (n = 62), beef (n = 48), lamb (n = 25) and goat meat (n = 15)), were randomly collected from 62 supermarkets and 88 broiler slaughterhouses located in different rural areas of the various districts explored in the study.Most of these slaughterhouses had good working conditions and practices.Swabs taken aseptically from carcasses and their matching livers and cuts were performed on different working days and hours.Each sample was packed in a sterile jar containing 100 mL of buffered peptone water (Merck, Darmstadt, Germany) and immediately transported to the laboratory in a small refrigerator under 2-5°C and tested on the same day.

Campylobacter culture
The selective agar used for the isolation of Campylobacter was a commercially available preparation, charcoal cefoperazone desoxycholate agar (CCDA) (Liofilchem, Teramo, Italy) containing 32 mg/L of cefoperazone and 10 mg/L of amphotericin B. Plates were incubated microaerobically (6% O2) at 37°C for two days.Suspect colonies (Small, gray and translucent colonies (1-2 mm in diameter) with a metallic sheen and butter-like consistency) were identified to the genus level by positive oxidase and catalase reactions and a typical Gram stain appearance (curved "seagull wing shaped" or spiral shaped small motile Gram-negative rods).

Campylobacter species identification
The discrimination between Campylobacter species was based on the standardized hippurate hydrolysis test and PCR (polymerase chain reaction) technique.

Hippurate hydrolysis test
The hippurate hydrolysis activity of Campylobacter species was analyzed using the Hippurate Test Kit (Liofilchem, Teramo, Italy) according to the manufacturer's instructions.The hippurate hydrolysis test allows differentiating C. jejuni (positive test revealed by the appearance of a deep blue/violet color in 30 minutes) from all other Campylobacter species (negative test).

Molecular identification of Campylobacter species
DNA was extracted from agar plates using Insta-Gene1 matrix (Bio-Rad, California, USA) according to the manufacturer's instructions.DNA extracts were then stored at -20°C until required for PCR screening.
Campylobacter speciation was performed using PCR targeting the 16S rRNA gene for the coidentification of Campylobacter jejuni and Campylobacter coli.Both species were discriminated by amplifying the hippuricase gene found exclusively in Campylobacter jejuni and the aspartokinase gene found in Campylobacter coli, using specific primers.PCR was carried out in a total volume of 50μl, containing 100 ng genomic DNA, 1X Taq DNA polymerase buffer, 2 mM MgCl2, 0.25 mM dNTP (Solis BioDyne, Tartu, Estonia), 100 ng of each primer [19] and 0.02 U of Taq DNA polymerase (Solis BioDyne, Tartu, Estonia).Following initial denaturation at 95°C for 5 minutes, 30 amplification cycles of denaturation (95°C for 1 minute), annealing (annealing temperature for 1 minute), and elongation (72°C for 1 minute) were performed in a thermal cycler Perkin-Elmer 9600.PCR products were then visualized on a 0.8% agarose gel under UV radiation.Primers and annealing temperatures used for the different amplifications as well PCR products are summarized in Table 1.

Isolation and identification of other diarrheal pathogens
Stool samples were cultured in selenite broth and various solid selective media, namely Salmonella-Shigella (SS) agar for the isolation of Salmonella spp.and Shigella spp., as well as Sorbitol MacConkey agar, blood agar and thiosulfate citrate bile salt cholera medium for the isolation of Escherichia coli, Staphylococcus aureus and Vibrio cholera, respectively.After 24-hours' incubation at 37°C, isolates were stained by the Gram method and identified by the appropriate biochemical tests and API 20E gallery (BioMérieux, Lyon, France).Clostridium spp.toxins were detected by ELISA technique (BioMérieux, Lyon, France).The presence of parasites (Entamoeba histolytica, Giardia intestinalis, etc.) in feces was routinely determined by microscopic examination of stool samples.Cryptosporidium spp.and Cyclospora cayetanensis oocysts were detected by modified Ziehl-Neelsen (MZN).Rotavirus and adenovirus were simultaneously detected in stool samples using the chromatographic immunoassay CerTest Rotavirus + Adenovirus one step combo card test (Biotec, Zaragoza, Spain).

Statistical analysis
Statistical analyses were performed using GraphPad Prism 6 (GraphPad Software, Inc, San Diego, CA, USA).Fisher's exact test was used to compare frequencies between groups.The distribution of Campylobacter infection was correlated to the monthly average temperature using Pearson correlation test.P values less than 0.05 were considered significant.

Demographic profile of participants
During a one year period between September 2016 and August 2017, stool samples were collected from Table 1.Primers and reactions conditions used for the amplification.

Prevalence of Campylobacter infection and distribution regarding gender, age and season
Examination of stool specimens (n = 1000) showed that 21.5% of samples were detected as positive for Campylobacter spp.C. jejuni was the main species, identified in 179 patients (83.2%) followed by C. coli found in 30 patients (13.9%).Other Campylobacter species counted only for 2.8% of gastroenteritis.Biochemical identification of C. jejuni and C. coli was further confirmed by PCR analysis as previously described in Table 1.
The seasonality of campylobacteriosis was determined in this study.As shown in Figure 1, Campylobacter has a clear seasonal pattern, with a large peak seen in the summer, mainly in July and August, while the lowest rates were observed between December and February.In agreement, a significant positive correlation was observed between the frequency of Campylobacter infection and the monthly average temperature (r = 0.83; p < 0.0001).

Clinical presentation of campylobacteriosis
Patients infected with campylobacteriosis showed a vast range of clinical symptoms.Most patients (80.0%) had watery diarrhea, while a bloody or bloody-mucous diarrhea was observed in 13% and 7% of cases, respectively.The frequency and period of diarrhea

Effect of age, gender and seasonality on campylobacteriosis prevalence
The isolation rate of Campylobacter spp. with regards to gender, age and season is shown in Table 2. Frequencies were not significantly different between males and females (p = 0.54) as well as within the age groups (p = 0.28).Moreover, Campylobacter infection rates were comparable between seasons (p = 0.14).

Risk factors for campylobacteriosis
The risk factors for campylobacteriosis were also explored in this study.As shown in Table 3, the risk for Campylobacter infection increased following consumption of meat, chicken or salad (p < 0.0001) and, to a lesser extent, after contact with animals (p = 0.0068).In contrast, there was no significant correlation between consumption of non-pasteurized milk or nontreated water and the prevalence of campylobacteriosis (p = 0.74; p = 0.26 respectively).

Antimicrobial Susceptibility Pattern of Campylobacter isolates
Table 4 shows the susceptibility pattern of isolated Campylobacter spp. to antibiotics.The highest rates of sensitivity were observed for erythromycin, chloramphenicol, clarithromycin and levofloxacin.In contrast, metronidazole was the least active agent against isolates, followed by amoxicillin and ampicillin.Multiple drug resistance to two or more antimicrobial agents was observed in 35.9% of Campylobacter isolates.

Reservoirs of Campylobacter
In order to identify the main categories of meat representing the most significant reservoirs of Campylobacter in Lebanon, a total of 150 samples of the most commonly consumed raw meats were collected from supermarkets and slaughterhouses in different Lebanese regions.Campylobacter spp.were detected in 33 samples (22.0%), among which 31 (93.9%)collected from slaughterhouses and 2 only from supermarkets.C. jejuni was identified in 28 (84.8%)samples, while C. coli only counted for 15.2% of contaminated samples.Strains were isolated from 45.5% and 33.3% of raw chicken and beef samples, respectively.The lowest rates of contamination with Campylobacter were observed in lamb (15.1%) and goat meat (6.1%).C. jejuni was the most prevalent in chicken (93.3%), beef (90.1%) and goat meat (100%), while C. coli was the most frequently isolated species in lamb (60%).Meat sites that are most likely to be contaminated with Campylobacter spp.were investigated by taking random samples of broiler carcasses and their matching livers and cuts.The highest rate of Campylobacter contamination in chicken was found in carcasses (46.7%), while in beef, lamb and goat meat, Campylobacter was predominant in meat cuts (Table 5).C. jejuni was identified in all carcasses (n = 9), as well as in 2 liver samples and 17 meat cuts.In contrast, none of the carcasses showed contamination with C. coli.It was mainly identified in cuts of lamb (n = 3) and liver of chicken (n = 1) and beef (n = 1) (Table 5).

Discussion
In agreement with previous studies [4,22,23], Campylobacter was the leading cause of gastroenteritis in Lebanon, with C. jejuni being the most frequently isolated species followed by C. coli.Our study showed a much higher prevalence of Campylobacter infection than that observed by Talhouk et al. in 1998 (0.7%), as well as to the prevalence reported by Dabboussi et al. in 2012 (11.1%) who evaluated the frequency of Campylobacter infection in 90 children with diarrhea in Northern Lebanon [15].These results highlight the remarkable increase of campylobacteriosis rates in Lebanon.Therefore, the detection of Campylobacter in patients' stool culture should be part of the routine diagnosis of human gastroenteritis in Lebanon, even if it is difficult, demanding and expensive.Moreover, N: number of samples; Meat samples were randomly collected at different working days and hours from Lebanese supermarkets and broiler slaughterhouses.Swabs were taken aseptically from carcasses and their matching livers and cuts, and tested for the presence of Campylobacter spp.
when compared to reports from other countries, the prevalence was higher to those found in Mexico (15.7%) and China (14.9%), but lower to those obtained in developed countries such as the United States (26.5%),New Zealand (33%) and the Netherlands (71.4%) [4].This variation might be attributed to differences in surveillance systems and Campylobacter detection methods between developed and developing countries, as well as to demographic, geographic and period differences between studies [4,24].The distribution of Campylobacter isolates according to patient's age and gender revealed that Campylobacter infections can occur in all age groups, with highest percentages noticed in children and adults, irrespective of the gender.Results reported from previous studies were variable and sometimes contradictory.According to the surveillance study carried out by the National Reference Center for Campylobacter and Helicobacter in France, children under 5 years of age and people aged over 65 had the highest prevalence of Campylobacter infections compared to other groups [25].In contrast, a casecontrol study carried out by Friedman et al. in 2000 revealed that most populations in industrialized countries showed a bimodal age distribution with the highest incidence seen in young children and young adults [26].As for developing countries, the infection was mainly limited to children under 2 years of age [27].The highest campylobacteriosis rates observed among children may be attributed to their relatively naïve immunological system and the high frequency of exposure to Campylobacter reservoirs.Another possibility is the lower threshold for medical assessment and testing among children, thus resulting in a higher diagnosis rate in this age group.As for adults and elderly people, high rates of exposure were related to high-risk food consumption patterns and poor hygiene level, respectively.On the other hand, males were overall more affected than females with male-tofemale gender ratio of 1.15 to 1.This has shown to be a consistent finding across most populations, and is related to the increased male exposure through highrisk occupations such as farming and meat processing, differences in food and water consumption patterns, and food hygiene practices [28].
The association of Campylobacter infection with season was also investigated in the present study.As observed in other developing countries, campylobacteriosis incidence was found to be the highest during the warmest months of the year, namely in July and August [3,29].These results were further supported by the positive strong correlation between the frequency of Campylobacter isolates and the monthly average temperature (r = 0.83; p < 0.0001).The pronounced seasonality observed in our study may be attributed to the increased Campylobacter prevalence in various reservoirs and seasonal changes in human behavior that affect exposure [22,[30][31][32].
Several factors can be associated with a higher risk for campylobacteriosis.Consistent with all previous data, the major risk factor identified in the present study was the consumption of meat and salad.Indeed, subjects were 2 to 3 times more likely to develop campylobacteriosis after consuming raw or undercooked food (p < 0.0001).Our results showed that contact with animals is another important source of transmission of Campylobacter infection in Lebanon (p = 0.0068).In contrast, consuming raw milk and drinking non-treated water were not statistically associated with culture positivity for Campylobacter isolates (p = 0.74 and p = 0.26 respectively).These results can be explained by the high standards of microbiological quality applied to drinking water, as well as by the fact that the majority of the Lebanese population does not consume tap water or nonpasteurized milk.Moreover, the prevalence of Campylobacter infection did not significantly differ according to gender, age and seasons (p = 0.54, p = 0.28 and p = 0.11 respectively).These results suggest that, despite the clear seasonal pattern observed in Figure 1, neither the climate nor age and gender can be considered as major determinants for Campylobacter infection in Lebanon.
The role of meat as a major source of Campylobacter transmission to humans in Lebanon was confirmed in our study by the detection of Campylobacter spp. in 22.0% of samples collected from markets and slaughterhouses.Our finding of C. jejuni domination over C. coli agrees with other studies that establish similar relation between these two species in various types of meat samples [33,34].The majority of contaminated meat samples (93.9%) were collected from slaughterhouses rather than supermarkets.These findings suggest that transmission to human results from handling and consumption of meat that has been contaminated during slaughter or carcass processing and not during food transport or storage [18,33,35,36].As expected, the main categories of meat representing the most significant reservoirs of Campylobacter in Lebanon were chicken (45.5%), followed by beef (33.3%), lamb (15.1%) and finally goat meat (6.1%).In agreement with results observed by Fadlallah et al. in 2018 [16], the highest rate of Campylobacter contamination in poultry meat was found in carcasses (46.7%).In contrast, in beef, lamb and goat meat Campylobacter spp.were predominant in the cuts.In fact, the intestinal tract of chicken can harbor a large number of Campylobacter spp.During evisceration, the intestinal tract may leak or rupture and the contents are transferred to the skin of the carcass [16,18,35,37].Moreover, according to the research conducted by Bolton in 2015, Campylobacter spp.are particularly sensitive to desiccation.For poultry slaughter, the carcasses are water-chilled, maintaining a wet surface and facilitating the survival of Campylobacter spp.[38].
Campylobacter infections are typically self-limited.However, an antibiotic therapy is indicated in the presence of severe complications or a weakened immune system.The failure to administer appropriate antibiotics in the presence of severe complications or a weakened immune system was associated with fatal outcome in bacteremia caused by Campylobacter species [39][40][41].In this regard, the antimicrobial susceptibility patterns of Campylobacter spp.were investigated in the present study.Interestingly, C. jejuni and C. coli demonstrated comparable susceptibility profiles; a finding that was previously noted by Talhouk et al. in 1998 [17].Despite decades of use, the rate of resistance of Campylobacter to erythromycin was quite low.Indeed, due to its low cost, safety use of administration and narrow of activity, erythromycin is still considered as the optimal drug of choice for the treatment of campylobacteriosis.Chloramphenicol, clarithromycin, levofloxacin, ciprofloxacin, nalidixic acid and imipenem were also among the most effective agents against Campylobacter infections.In contrast, the highest rates of resistance were recorded to metronidazole, amoxicillin and ampicillin.The high prevalence of multi-resistant strains observed in our study was previously reported by the European Food Safety Authority [3].Is it noteworthy to mention that rates of resistance of Campylobacter spp. to ampicillin (60.7%), amoxicillin (62.3%) and clindamycin (41.9%) are remarkably higher than those reported by Talhouk et al. in 1998 (5%, 23% and 31% respectively) [17].This may be explained by the uncontrolled use of these antibiotics in human medicine, such as treating infections other than gastroenteritis, self-medication and access without prescription.It can also be attributed to their use in veterinary medicine in order to control, prevent and treat infections, and enhance animal growth.

Conclusion
Campylobacter is the leading cause of gastroenteritis in Lebanon, with C. jejuni being the most isolated species from patients' stool culture and meat samples.Even though, the highest infection rates were reported in children and during summer, neither season nor age or gender seems to be major modifying factors for campylobacteriosis in Lebanon.Consumption of contaminated meat and salads resulting from mishandling or cross-contamination remains the principal route of Campylobacter transmission to humans.The incidence of Campylobacter resistance to erythromycin is still relatively low in the Lebanese population, and thus it should always be regarded as the drug of choice in treatment of campylobacteriosis.In contrast, the highest resistance patterns were observed to the most commonly prescribed antibiotics in human and veterinary medicine.Human campylobacteriosis remains an unresolved public health problem of high importance.Detection of Campylobacter spp. in patients' stool culture should be therefore considered as a routine test in the diagnosis of gastroenteritis in order to allow early diagnosis of the disease and better monitoring of antimicrobial resistance among Lebanese patients.

Figure 1 .
Figure 1.Seasonal distribution of Campylobacter spp.isolated in patients' stool between September 2016 and August 2017.

Table 2 .
Frequency of Campylobacter infection with regard to gender, age and season.
N: number of individuals; Fisher's exact test was performed to generate p values for differences in Campylobacter spp.isolation rate with regards to gender, age and season.

Table 3 .
Risk factors for campylobacteriosis.
N: number of individuals; ** p < 0.01, **** p < 0.0001 based on Fisher's exact test for differences in Campylobacter spp.isolation rate with regards to potential risk factors.

Table 4 .
Antimicrobial susceptibility pattern of Campylobacter spp.isolates.Campylobacter susceptibility testing was determined by the disc diffusion standard method.The diameters of inhibition zones were measured and evaluated according to EUCAST 2018 recommendations for ampicillin, amoxicillin, erythromycin, ciprofloxacin, tetracycline and clarithromycin, and CLSI 2015 interpretive criteria for Enterobacteriaceae for the remaining antibiotics.

Table 5 .
Distribution of Campylobacter spp.isolates per site of isolation.