Background
Campylobacter spp. are common causes of human bacterial gastrointestinal disease worldwide in particular
C. jejuni and
C. coli are recognized as the leading cause of acute bacterial diarrhea worldwide accounting for nearly 400 million cases per year [
1,
2].
Campylobacter-associated gastroenteritis is highly prevalent in South and Southeast Asia and is commonly associated with traveler’s diarrhea (TD) [
3‐
6]. Over the last decade, recent evidence of other
Campylobacter spp. to include
C. ureolyticus and
C. concisus has highlighted the potential role of these microorganisms to cause gastroenteritis in both industrialized and developing countries [
7,
8].
Several
Campylobacter spp. (non-
C. jejuni/coli) were isolated in the early 1990s from diarrhea stool samples though the ability to culture these organisms was highly variable among laboratories [
9,
10]. Many
Campylobacter spp. require different conditions for growing due to their fastidious laboratory growth requirements in comparison to other Gram-negative organisms. These factors suggest that other
Campylobacter spp. besides
C. jejuni and
C. coli are likely underestimated as diarrhea etiologic agents as a result of inappropriate detection and isolation procedures which generally support only the growth of
C. jejuni and
C. coli [
11].
Campylobacter concisus historically has been associated as being a human oral pathogen causing gingivitis and peridontitis [
12]. More recent studies have shown that
C. concisus has been isolated from chronic, more milder diarrheal cases in both adults and children as compared to
C. jejuni [
13‐
16]. Additionally,
C. concisus has been implicated in the development of Crohn’s disease and inflammatory bowel disease (IBD) as it has been isolated from fecal samples and colonic biopsies of children with Crohn’s disease and from ulcerative colitis biopsy samples from adults with IBD [
17‐
19].
C. ureolyticus, like
C. concisus, has been detected more frequently in diarrheal cases. A recent study from Hatanaka et al. showed that
C. ureolyticus was detected in 51.9% of diarrheal stool samples from children with gastroenteritis [
20]. Similar studies conducted in Ireland and Chile have also implicated
C. ureolyticus as an emerging gastrointestinal pathogen [
21‐
23]. Like
C. concisus, C. ureolyticus has been isolated at high rates from patients with ulcertative colitis and Crohn’s Disease [
19,
24].
Currently, there are no to limited data on the prevalence of non-C. jejuni/C. coli Campylobacter spp. specifically, C. ureolyticus and C. concisus, in diarrhea cases in Thailand and Nepal particularly among travelers. In the present/current study, specific pathogen-free stool samples from travelers seeking medical care for traveler’s diarrhea and asymptomatic controls in Nepal and Thailand were assayed for the detection of C. ureolyticus and C. concisus using Campylobacter genus and species specific primers for C. ureolyticus and C. concisus. To our knowledge, this is the first report of C. ureolyticus and C. concisus in stool samples in travelers with traveler’s diarrhea and asymptomatic controls from Nepal and Thailand.
Results
A total of 61 previously determined to be pathogen-negative TD stool samples (CIWEC Travel Medicine Clinic, Nepal: cases = 36, control = 14; Bamrungrad International Hospital, Thailand: cases = 9, controls = 2) were positive for
Campylobacter 16S
rRNA (Table
1). Eighteen isolates (29.5%; 18/61) were positive only for
Campylobacter genus and were not further speciated beyond
C. concisus and
C. ureolyticus.
Table 1
C. concisus and C. ureolyticus detected as monoinfections and in mixed infections by PCR using species specific housekeeping genes in pathogen-negative traveler’s diarrhea cases and asymptomatic controls from CIWEC Travel Medicine Clinic, Kathmandu, Nepal and Bamrungrad International Hospital, Bangkok, Thailand
C. concisus: single pathogen | 24 (28.9%) | 3 (4%) | 9.8 (2.8–34.1) | 0.0003 |
C. ureolyticus: single pathogen | 4 (4.8%) | 4 (5.3%) | 0.9 (0.2–3.7) | NS |
C. concisus and C. ureolyticus mixed infection | 2 (2.4%) | 1 (1.3%) | 1.8 (0.2–20.5) | NS |
Campylobacter spp. (non C. concisus and C. ureolyticus) | 6 (7.2%) | 6 (8.0%) | 0.7 (0.2–2.0) | NS |
C. concisus: Single pathogen | 2 (7.7%) | 0 | N/A | N/A |
C. ureolyticus: Single pathogen | 1 (3.8%) | 1 (3.3%) | 1.2 (0.07–19.5) | NS |
C. concisus and C. ureolyticus mixed infection | 1 (3.8%) | 0 | N/A | N/A |
Campylobacter spp. (non C. concisus and C. ureolyticus) | 5 (19.2%) | 1 (3.3%) | 4.8 (0.5–41.6) | NS |
At the CIWEC Travel Medicine Clinic, Nepal,
C. consisus was identified significantly more often in TD cases (28.9%; 24/83) when compared to asymptomatic controls (4%; 3/75) (OR = 9.8; 95% CI 2.8–34.1;
P = 0.0003) when isolated as the sole pathogen (Table
1).
C. ureolyticus was detected in four cases (4.8%; 4/83) and four controls (5.3%; 4/75) respectively (OR = 0.90; 95% CI 0.2–3.7;
P = 0.88) when isolated as the sole pathogen.
Campylobacter concisus and C. ureolyticus mixed infections were detected in two cases (2.4%; 2/83) and in one control (1.3%; 1/75) sample (OR = 1.8; 95% CI 0.2–20.5; P = 0.62).
At the Bamrungrad International Hospital, Thailand,
C. consisus was identified in only two TD cases (7.7%; 2/26) and in none of the controls (0%; 0/30) (Table
1).
C. ureolyticus was detected in one case (3.8%; 1/26) and in one control (3.1%; 1/30) (OR = 1.16; 95% CI 0.07–19.5;
P = 0.92) in the stool samples collected at Bamrungrad.
C. concisus and
C. ureolyticus as a mixed infection were detected in one case (3.8%; 1/26) and in none of the control samples.
In Table
2, the number of
C. jejuni and
C. coli isolates detected by culture are listed in both monoinfections and mixed infections by site.
C. jejuni was detected significantly more often in the cases (7.9%; 38/480) when compared to the controls (2.9%; 6/209) at the CIWEC Travel Medicine Clinic, Nepal where
C. jejuni was the sole pathogen isolated (OR = 6.5; 95% CI 2.2–19.1;
P = 0.0007).
C. coli was detected slightly more often in the cases (1.5%; 7/480) when compared to the controls (1.4%; 3/209) though not significantly (OR = 1.0; 95% CI 0.3–3.9;
P = 0.9) when
C. coli was the sole pathogen detected. In mixed infections,
C. jejuni was detected significantly more often in the cases (8.1%; 39/480) when compared to the controls (1.9%; 4/209) (OR = 20.2; 95% CI 2.7–152.9;
P = 0.0036) (Table
2).
C. coli was detected more often in the cases (1.3%; 6/480) when compared to the controls (0.9%; 2/209) though not significantly in mixed infections (OR = 1.3; 95% CI 0.3–6.6;
P = 0.74) (Table
2).
Table 2
C. jejuni and C. coli isolated as single pathogens and from mixed infections via culture in traveler’s diarrhea cases and asymptomatic controls from CIWEC Travel Medicine Clinic, Kathmandu, Nepal and Bamrungrad International Hospital, Bangkok, Thailand
C. jejuni: single pathogen | 38 (7.9%) | 6 (2.9%) | 6.5 (2.2–19.1) | 0.0007 |
C. coli: single pathogen | 7 (1.5%) | 3 (1.4%) | 1.0 (0.3–3.9) | NS |
C. jejuni: mixed infection | 39 (8.1%) | 4 (1.9%) | 20.2 (2.7–152.9) | 0.0036 |
C. coli: mixed infection | 6 (1.3%) | 2 (0.9%) | 1.3 (0.3–6.6) | NS |
C. jejuni: single pathogen | 24 (13.9%) | 4 (2.4%) | 2.9 (1.2–6.9) | 0.01 |
C. coli: single pathogen | 4 (2.3%) | 0 | N/A | N/A |
C. jejuni: mixed infection | 19 (11%) | 1 (0.6) | 20.2 (2.7–152.9) | 0.0036 |
C. coli: mixed infection | 5 (2.9%) | 0 | N/A | N/A |
At Bamrungrad International Hospital, Thailand,
C. jejuni was detected significantly more often in the cases (13.9%; 24/173) when compared to the controls (2.4%; 4/165) when
C. jejuni was the sole pathogen isolated (OR = 2.9; 95% CI 1.2–6.9;
P = 0.01) (Table
2).
C. coli was detected only in the cases (2.3%; 4/173) when
C. coli was isolated as a single pathogen.
C. jejuni was detected significantly more often in the cases (11%; 19/173) when compared to the controls (0.6%; 1/165) (OR = 20.2; 95% CI 2.7–152.9;
P = 0.0036) in mixed infections. In mixed infections,
C. coli was detected only in the cases (2.9%; 5/173) at Bamrungrad International Hospital, Thailand.
Discussion
The prevalence of
Campylobacter spp. as etiologic agents for TD in Nepal and Thailand has been well documented, but the vast majority of these studies have focused on the detection of
C. jejuni and
C. coli as the primary
Campylobacter spp [
4,
5,
30‐
32]. More recently, other
Campylobacter spp. to include
C. concisus and
C. ureolyticus have been implicated in human gastroenteritis due to improved methods of laboratory detection. Most clinical laboratories do not regularly test for other
Campylobacter spp. such as
C. concisus and
C. ureolyticus due to both the fastidious nature of the organisms, the requirement for an enriched H
2 environment in order for growth and both species being unable to grow on the selective media that is commonly used for
C. jejuni and
C. coli [
16,
20,
33]. It is highly probable that due to these reasons, gastroenteritis caused by
C. concisus and
C. ureolyticus is highly underestimated. In TD etiologic studies,
C. concisus and
C. ureolyticus are rarely tested for and therefore the incidence of
C. concisus and
C. ureolyticus in TD is largely unknown.
In this study,
C. concisus was detected significantly more often in TD stool samples from Nepal as compared to asymptomatic controls. These stools samples were previously categorized as pathogen-negative after undergoing a comprehensive laboratory workup for enteric bacteria, viruses and parasites. In the present study, 92.3% of the
C. concisus cases were from Nepal. Similarly, 80% of the
C. ureolyticus cases were from Nepal though there were only five total cases of
C. ureolyticus detected. There are no previous published data on the prevalence of
C. concisus and
C. ureolyticus from the CIWEC Travel Medicine Clinic, Kathmandu, Nepal and at Bamrungrad International Hospital, Bangkok, Thailand as these microorganisms are not routinely tested for in the clinical laboratories at these facilities. In a previous TD case–control study conducted at the CIWEC Travel Medicine Clinic from 2001 to 2003,
Campylobacter was the most prevalent pathogen detected in cases (17%), but the
Campylobacter isolates were not speciated in this study [
4]. Multiple studies have also shown that
Campylobacter spp. is the most common pathogen in Thailand to cause acute gastroenteritis in travelers, deployed US military personnel and the indigenous population [
5,
6,
31,
34,
35]. Similar to cases from CIWEC,
C. jejuni was the most common detected pathogen from cases at Bamrungrad International Hospital. To our knowledge, this is the first report showing detection of
C. concisus and
C. ureolyticus in TD samples in a case–control setting from Nepal and Thailand and that these results strongly suggest a potential role of
C. concisus in the development of TD in travelers to Nepal.
Recent studies have shown that the prevalence of
C. concisus and
C. ureolyticus typically are equal or higher when compared to
C. jejuni and
C. coli in studies in which they are both included in the laboratory workup [
36‐
38]. In Australia, Underwood et al. showed that
C. concisus had a prevalence of 49.7% in patients with gastroenteritis as compared to
C. jejuni (5%) and that it was detected as a single agent in 78% of the
C. concisus positive cases indicating a potential role for
C. concisus in the development of gastroenteritis in these patients [
16]. In a 2 year study conducted in Denmark, Nielsen et al. showed that
C. concisus was prevalent in 26.1% of patients with gastroenteritis as compared to
C. jejuni/coli detected in 31.9% of infected patients [
15]. In the first report of
C. ureolyticus in children with gastroenteritis in Japan, Hatanaka et al. showed that
C. ureolyticus was prevalent in 51.9% of the infected children as compared to
C. jejuni/coli infecting 15.5% of the children [
20]. In a study from Southern Ireland,
C. ureolyticus was prevalent in 41% of diarrheal patients as compared to
C. jejuni at 50.7% and
C. coli at 5.7% [
28]. In the present study, only the incidence of
C. concisus in the Nepal pathogen-negative TD cases (28.9%) was higher than the incidence of both
C. jejuni (16.0%) and
C. coli (2.7%) in the Nepal TD cases that yielded at least one enteric pathogen under routine laboratory testing. It is likely that the incidence of
C. concisus and
C. ureolyticus are highly underestimated due to the fact that detection of these organisms was not included in the initial routine laboratory testing of the TD cases from both CIWEC and Bamrungrad hospital.
There were several limitations to the study. As mentioned previously, only pathogen-negative stools were selected which limited the number of samples used. Between the two sites, there were 653 cases and 374 controls (total = 1027). This resulted in a smaller sample size particularly for the TD samples from Thailand (26 cases and 30 controls) as well as an underestimation of the overall prevalence of both
C. concisus and
C. ureolyticus in all of the cases from both sites. The selection of pathogen-negative stools also inhibited our ability to address co-infections with
C. concisus and
C. ureolyticus. However, since pathogen-negative samples were solely used, we can infer from the data that those samples containing
C. concisus or
C. ureolyticus were monoinfections with no other enteric pathogens being detected previously. As both organisms have been implicated in the development of gastroenteritis, there have been recent studies to include whole genome sequencing aimed at delineating virulence factors to elucidate the pathogenic mechanisms of each organism. Both
C. concisus and
C. ureolyticus contain genes similar to
C. jejuni for adhesins, invasins and toxins that are known to play an important role in the development of acute gastroenteritis [
37,
39‐
42]. These virulence factors were not tested for in the present study. Additionally, PCR assays should have been included for the detection of
C. jejuni and
C. coli as PCR has shown to be far superior for the detection of both
C. jejuni and
C. coli when compared to culture [
43‐
46]. It is likely that the overall estimate of
C. jejuni and
C. coli by PCR would increase to those that were positive by culture. Future plans include testing all 826 samples using PCR for
C. jejuni, C. coli, C. concisus and
C. ureolyticus to better compare the prevalence of each organism using a similar molecular method of detection and in the same batch of samples.
As this is the first report of
C. concisus and
C. ureolyticus in acute gastroenteritis in travelers to Nepal and Thailand, further studies with long term clinical follow up of infected travelers should be initiated to assess the development of any long term post infectious sequelae to include IBD and ulcerative colitis that have been observed in studies examining the long term effects of
Campylobacter infection [
47,
48]. In conclusion, as it is evident that at least
C. concisus from this study is involved in the development of acute gastroenteritis in travelers to Nepal, it is imperative that more studies include both
C. concisus and
C. ureolyticus are included in the laboratory diagnostic methods to better ascertain their true prevalence as etiological agents of acute gastroenteritis. Additionally, there are limited data on the geographic distribution of
C. concisus and
C. ureolyticus in developing and developed countries and further epidemiological studies to ascertain this are warranted.
Authors’ contributions
OS, CM and LB designed the study. SA and PP served as site primary investigators respectively in Thailand and Nepal. OS and SR conducted the laboratory work and collected the data. OS, SR and BES conducted data analysis. OS and BES drafted the manuscript. All authors read and approved the final manuscript.