Evaluation of a rapid fluorescence immunoassay for detecting Campylobacter antigens in stool samples
- Open Access
- 01.12.2025
- Research
Erschienen in:
Abstract
Background
Campylobacter genus is a major cause of bacteria-induced diarrheal infectious diseases worldwide, with an increasing incidence in both high- and low-income countries [1, 2]. In addition, Campylobacter spp. can overcome the gastrointestinal barrier, leading to bacteremia. Blood stream infections by Campylobacter spp. accounts for < 1% of Campylobacter spp. but are associated with substantial mortality rates of 3–28% [3‐5]. In addition, bacteremia caused by Campylobacter can lead to complications such as infections in the joints, bones, and soft tissues, as well as vascular infections including mycotic aneurysms, endocarditis, spondylodiscitis, and meningoencephalitis [4, 6‐11]. Post-infection complications can include reactive arthritis and Guillain–Barré syndrome [12]. Immunoproliferative small intestinal disease, a type of lymphoma, has been reported in association with Campylobacter infections. Notably, Campylobacter jejuni has been found in biopsy specimens of patients with this intestinal disease; in these patients, antimicrobial therapy targeting C. jejuni has led to rapid remission of the disease [13].
According to the European Centre for Disease Control and Prevention (ECDC), the species that most frequently cause campylobacteriosis in Europe are Campylobacter jejuni and Campylobacter coli, followed by Campylobacter upsaliensis, Campylobacter lari, and Campylobacter fetus [14]. Most cases of C. upsaliensis infections have been reported from the European Union (EU), Australia, Canada, South Africa, and the United States [15]. Cats and dogs are the main reservoirs [16].
Anzeige
In 2021, 129, 960Campylobacter cases were reported in the EU, with the majority being C. jejuni (88.4%) and C. coli (10.1%) [14]. During the severe acute respiratory syndrome coronavirus 2 (SARS CoV-2) pandemic, there was an increase in C. fetus cases, with 148 reported in 2021 compared to 130 in 2020 and 122 in 2019 [14]. Campylobacter fetus infections have recently been identified as the most common cause of Campylobacter-associated bacteremia, leading to secondary tissue infections such as vascular infections and endocarditis (83%), with a mortality rate of up to 25% [10, 11, 17]. The primary reservoirs for C. fetus are cattle and sheep, and products from these animals are suspected sources of human infections [18]. In France, C. fetus recently caused an outbreak in a rehabilitation center, resulting in significant morbidity among elderly patients [19].
Antibiotic therapy for campylobacteriosis is most effective when started within the first 3 days after symptom onset; it shortens the duration of intestinal symptoms and also reduces the gut population of Campylobacter [20]. It is recommended to limit transmission in daycare centers and other places with groups of children [21]. Rapid identification of these bacteria can guide the choice of antibiotic therapy.
Stool culture to detect Campylobacter requires a minimum of 48 h and has a sensitivity ranging from 60 to 76% [22, 23]. Although the specificity of coproculture is excellent, its sensitivity is reduced for Campylobacter spp. detection. Several culture-independent diagnostic tests are available, providing faster results with better sensitivity and good specificity. Among them, molecular methods such as real-time polymerase chain reaction (qPCR) and enzyme-linked immunosorbent assays (ELISAs) require additional automation, are technically demanding, and are often validated only for C. jejuni and C. coli [23‐25]. Since the 2010s, syndromic PCR formats have become the first diagnostic test of choice for the detection of Campylobacter spp., often even replacing coproculture. Immunochromatographic tests are easier to use but have lower reported sensitivity [26‐28].
This study evaluated the analytical and clinical performance of the Sofia Campylobacter Fluorescence Immunoassay (SCFIA) for the rapid detection of Campylobacter antigens in stool specimens from patients with signs and symptoms of infectious gastroenteritis. SCFIA is a new rapid test designed for the detection of C. jejuni, C. coli, C. upsaliensis, and C. lari antigens in stool specimens. The test uses advanced immunofluorescence-based lateral flow technology to provide a rapid qualitative result within 15 min.
Anzeige
Materials and methods
SCFIA evaluation
This study was conducted between July and November 2023 at the French National Reference Centre for Campylobacter and Helicobacter (NRCCH) located in the Bacteriology Laboratory at the University Hospital of Bordeaux.
Testing for Campylobacter antigens was conducted using the SCFIA (QuidelOrtho Corp., San Diego, CA, USA), for the detection of C. jejuni, C. coli, C. upsaliensis, and C. lari antigens in stool specimens. A proprietary algorithm calculates a specimen over cut-off (S/CO) value, such that S/CO ≥ 1 indicates a positive result, and S/CO < 1 indicates a negative result. The S/CO value is an indicator of the ability of the assay to bind antigens of Campylobacter species and the antigen content of the specimen. All tests based on the SCFIA were conducted according to the manufacturer’s instructions [29]. Frozen or cooled specimens were brought to room temperature and mixed well before SCFIA testing.
A prospective evaluation was conducted using 205 fresh stool specimens sex ratio, 1.05; mean age, 37 ± 32 years) referred for testing for gastrointestinal infectious pathogens. Specimens were transported at 4 °C in Cary–Blair medium (FecalSwab, Copan, Italy) prior to testing. The samples were plated on Campylosel (bioMérieux, Marcy l’Étoile, France) and incubated for 3 days at 36 °C in jars using an Anoxomat microprocessor (Mart Microbiology, B.V. Lichtenvoorde, The Netherlands) to create a microaerobic atmosphere (80–90% N2, 5–10% CO2, and 5–10% H2). Subsequently, bacteria were identified via matrix-assisted laser desorption ionization–time of flight mass spectrometry (Bruker, 2023 library) as previously described [30]. For molecular detection, 50 µL of each sample was tested on the BD MAX Enteric Bacterial Panel, which includes targets for the detection of C. jejuni and C. coli [31, 32]. Among these 205 specimens, 173 tested negative in culture and PCR, and 32 were positive (27 C. jejuni, 4 C. coli, and 1 mixed infection of C. jejuni and C. coli). We also conducted a retrospective analysis using 94 frozen specimens collected in Cary–Blair medium between 2020 and 2021. These were aliquoted upon reception into tubes that had never previously been defrosted, and stored at − 80 °C. They all tested positive in culture and BD MAX PCR for Campylobacter (83 C. jejuni, 11 C. coli).
Positivity for Campylobacter in BD MAX PCR was considered the gold standard.
Linearity, limit of detection, and reactivity to other species
The linearity of the assay S/CO values and limit of detection (LOD) for C. jejuni, C. coli, C. upsaliensis, and C. lari were evaluated using serial dilutions. To assess reactivity to other Campylobacter species not declared by the manufacturer, C. armoricus (CCUG 73571T), C. fetus (ATCC 27374), and C. ornithocola (CECT 9147) were included in the serial dilutions because these species can also cause human gastroenteritis. The phylogenetically related species Aliarcobacter butzleri (ATCC 49616), Helicobacter cinaedi (CCUG 18818T), and Helicobacter pullorum (CCUG 33837T) were included to investigate potential cross-reactivity that could result in false-positive test results.
To establish the serial dilutions, well-characterized specimens with known species were grown on blood agar plates under microaerobic conditions. Subsequently, species identity was verified via MALDI-TOF. The grown cultures were used to create stock solutions for each species in Cary–Blair medium for a subsequent serial dilutions. The established stock solutions had the following concentrations in colony-forming units (CFUs) per mL: 3 × 108 (C. armoricus), 4.2 × 108 (C. coli), 1.2 × 108 (C. fetus), 3.9 × 108 (C. jejuni), 4.5 × 108 (C. lari), 1.7 × 108 (C. ornithocola), 4.5 × 107 (C. upsaliensis), 1.4 × 107 (A. butzleri), 5.4 × 107 (H. cinaedi), and 5.4 × 107 (H. pullorum). Stock solutions were diluted by 1:10, 1:100, 1:1000, 1:10,000, and 1:100,000. Each dilution was tested with the SCFIA to estimate the LOD of the assay. The linearity of the assay was evaluated by correlating the S/CO values with the individual dilutions.
Results
Accuracy
In the prospective evaluation of 205 specimens, the rate of positive agreement between the SCFIA and the gold standard was 96.88% (Table 1). The rate of negative agreement was 98.84%. In the retrospective evaluation, the positive agreement rate between the SCFIA and the gold standard was 97.87%. Overall, the positive agreement rate between the SCFIA and the gold standard was 97.62 (Table 1).
Table 1
Positive, negative, and overall agreement between SCFIA and the gold standard
Comparison vs. gold standard | Prospective samples | Retrospective samples | Prospective + retrospective samples | ||||
|---|---|---|---|---|---|---|---|
pos (n) | neg (n) | pos (n) | neg (n) | pos (n) | neg (n) | ||
SCFIA | pos (n) | 31 | 2 | 92 | NA | 123 | 2 |
neg (n) | 1 | 171 | 2 | NA | 3 | 171 | |
Positive agreement (95% CI) | 96.88 (82.00–99,84) | 97.87 (91.79–99.63) | 97.62 (92.67–99.38) | ||||
Negative agreement (95% CI) | 98.84 (95.45–99.80) | NA | 98.84 (95.45–99.80) | ||||
Overall agreement | 98.54 | 97.87 | 98.33 | ||||
Anzeige
Three specimens that tested positive in culture (3 C. jejuni, 1 from the prospective samples, and 2 from the retrospective samples) and PCR but negative in the SCFIA were also positive in an ELISA (RIDASCREEN Campylobacter, R-Biopharm, Darmstadt, Germany). Two specimens from the prospective samples that tested negative in culture but positive in the SCFIA were also positive in PCR (Table 1).
Linearity and reactivity to other Campylobacter species
In the serial dilutions established for different Campylobacter species, the SCFIA showed the expected reactivity to C. coli, C. jejuni, C. lari, and C. upsaliensis, as described by the manufacturer. However, it also demonstrated reactivity to C. ornithocola and C. armoricus. The last dilution that provided a positive test result (S/CO > 1) for all species was 1:10. The S/CO values for positive results at the 1:10 dilution were 2.53 for C. coli and 4.82 for C. armoricus. The bacterial concentrations for positive results at the 1:10 dilution were 4.5 × 106 CFU/mL for C. upsaliensis and 4.5 × 107 CFU/mL for C. lari. Overall, the average S/CO at the 1:10 dilution was 4.01, correlated with a concentration of 3.0 × 107 CFU/mL.
The test results for C. fetus were negative even in the undiluted specimens at the highest concentration, with S/CO values near zero, indicating that the assay antibodies do not have any affinity to bind C. fetus antigens or to detect this species even at high concentrations (Table 2).
Table 2
SCFIA S/CO and CFU/mL values of serial dilutions for Campylobacter species
Species | Dilution | CFU/mL | S/CO | Species | Dilution | CFU/mL | S/CO |
|---|---|---|---|---|---|---|---|
C. armoricus | 0 | 3.0 × 108 | 29.826 | C. jejuni | 0 | 3.9 × 108 | 27.456 |
1:10 | 3.0 × 107 | 4.815 | 1:10 | 3.9 × 107 | 4.109 | ||
1:100 | 3.0 × 106 | 0.700 | 1:100 | 3.9 × 106 | 0.425 | ||
1:1,000 | 3.0 × 105 | 0.125 | 1:1,000 | 3.9 × 105 | 0.000 | ||
1:10,000 | 3.0 × 104 | 0.0720 | 1:10,000 | 3.9 × 104 | 0.0437 | ||
C. coli | 0 | 4.2 × 108 | 19.208 | C. upsaliensis | 0 | 4.5 × 107 | 26.085 |
1:10 | 4.2 × 107 | 2.529 | 1:10 | 4.5 × 106 | 3.963 | ||
1:100 | 4.2 × 106 | 0.222 | 1:100 | 4.5 × 105 | 0.434 | ||
1:1,000 | 4.2 × 105 | 0.135 | 1:1,000 | 4.5 × 104 | 0.120 | ||
1:10,000 | 4.2 × 104 | 0.0000 | 1:10,000 | 4.5 × 103 | 0.0373 | ||
C. lari | 0 | 4.5 × 108 | 24.655 | C. fetus | 0 | 1.2 × 108 | 0.000 |
1:10 | 4.5 × 107 | 4.717 | 1:10 | 1.2 × 107 | 0.138 | ||
1:100 | 4.5 × 106 | 0.154 | 1:100 | 1.2 × 106 | 0.140 | ||
1:1,000 | 4.5 × 105 | 0.000 | 1:1,000 | 1.2 × 105 | 0.138 | ||
1:10,000 | 4.5 × 104 | 0.0000 | 1:10,000 | 1.2 × 104 | 0.1007 | ||
C. ornithocola | 0 | 1.7 × 108 | 28.297 | Average over all species and dilution series | 0 | 3.0 × 108 | 25,92 |
1:10 | 1.7 × 107 | 3.922 | 1:10 | 3.0 × 107 | 4,01 | ||
1:100 | 1.7 × 106 | 0.438 | 1:100 | 3.0 × 106 | 0,40 | ||
1:1,000 | 1.7 × 105 | 0.056 | 1:1,000 | 3.0 × 105 | 0,07 | ||
1:10,000 | 1.7 × 104 | 0.0000 | 1:10,000 | 3.0 × 104 | 0,03 |
Anzeige
The S/CO values were strongly correlated with the estimated number of bacterial copies per mL for each Campylobacter species, with correlation coefficients (R2) ranging from 0.991 for C. lari to 0.999 for C. ornithocola and C. armoricus. Across all species, the R2 was 0.9971 (Fig. 1). Given the high linear correlation observed between S/CO values measured using the SCFIA and the bacterial concentration in the dilutions for all Campylobacter species, a linear regression analysis was conducted to estimate the number of CFU/mL needed to reach a positive test result (S/CO ≥ 1). The estimated minimum concentration in CFU/mL at the test-specific S/CO cut-off of 1 was 1.47 × 106 for C. upsaliensis, while the maximum was 17 × 106 CFU/mL for C. coli (Table 3).
Fig. 1
Linear trend and correlation coefficients (R2) between Sofia Campylobacter Fluorescence Immunoassay (SCFIA) specimen cut-off (S/CO) values and the number of colony-forming units (CFU)/mL per dilution step
Table 3
Estimated minimum bacterial concentrations (CFU/mL) required in SCFIA for a positive test result (S/CO ≥ 1)
Species | Linear regression | Estimated CFU/mL for S/CO = 1 |
|---|---|---|
y = CFU/mL | ||
x = S/CO | ||
C. armoricus | y = 1 × 107 x– 4 × 106 | 6.00 × 106 |
C. coli | y = 2 × 107 x– 3 × 106 | 1.70 × 107 |
C. jejuni | y = 1 × 107 x– 4 × 106 | 6.00 × 106 |
C. lari | y = 2 × 107 x– 6 × 106 | 1.40 × 107 |
C. ornithocola | y = 6 × 106 x– 2 × 106 | 4.00 × 106 |
C. upsaliensis | y = 2 × 106 x– 534,225 | 1.47 × 106 |
All species | y = 1 × 107 x– 3 × 106 | 7.00 × 106 |
Reactivity to phylogenetically related species
The phylogenetically related species A. butzleri, H. cinaedi, and H. pullorum were included in the serial dilutions with initial concentrations of 1.4 × 107 for A. butzleri and 5.4 × 107 for both H. cinaedi and H. pullorum. The SCFIA test provided negative results for all three species, even at the highest concentration. The measured S/CO values were zero for all three species at the highest concentration, indicating that the SCFIA test has no potential to cross-react with these phylogenetically related species.
Anzeige
Discussion
The performance and value of Campylobacter antigen detection in stool samples have been described in numerous reports [33]. To our knowledge, this study presents the first independent evaluation of the performance of the SCFIA. Stool antigen tests have shown variable performance, perhaps due to intrinsic differences among the tests or the reference methods used in different studies. Culture methods are particularly known to lack sensitivity for Campylobacter detection compared to PCR methods. We assessed the clinical performance of the SCFIA using stool specimens confirmed to be positive via PCR, in both prospective and retrospective analyses.
The advantage of the SCFIA test kit is its automated reading, which eliminates operator influence. This is particularly beneficial for samples with low positivity, where immunochromatographic tests can be misinterpreted by users.
Our retrospective evaluation included 299 samples, of which 126 (42.1%) were positive for Campylobacter: 110 C. jejuni (87.3%), 15 C. coli (11.9%), and 1 for both, C. jejuni and C. coli (0.8%). This distribution closely matches data reported by the ECDC at the European level. According to the ECDC, 88.4% of confirmed infections in 2021 were caused by C. jejuni, followed by C. coli at 10.1% [14].
The linear correlation between S/CO values and approximated CFU/mL values was high across all individual serial dilutions and species, with correlation R2 values ranging from 0.991 (lowest for C. lari) to 0.999 (highest for C. armoricus and C. ornithocola). All serial dilutions included specimens with concentrations below and above the S/CO cut-off value of 1 and the LOD of the assay. The strong correlation between S/CO values and bacterial concentrations, spanning a wide range and including specimens around the assay cut-off and LOD, is a key indicator of the test’s reliability, particularly for borderline specimens. Infected humans usually excrete 106 to 109C. jejuni per gram of stool [34]. This test is therefore sufficiently sensitive to detect C. jejuni and C. coli in human stool specimens.
Anzeige
Linear regression estimated bacterial concentrations between 1.47 × 106 and 17 × 106 CFU/mL for different Campylobacter species at the S/CO cut-off of 1, with the exception of C. fetus. This result indicates that the ability of the assay to detect C. jejuni, C. coli, C. lari, C. upsaliensis, C. armoricus, and C. ornithocola is nearly equivalent. Such near-equivalence in human tests has not previously been described. For example, a study that evaluated the ProsPect Campylobacter immunoassay (Remel, Lenexa, KS, USA) reported higher detection limits for C. jejuni and C. coli than for other Campylobacter species that were also detectable [35]. The ability of some tests to detect C. upsaliensis has been previously described [36‐40]. Generally, these studies reported a 1- to 10-fold lower sensitivity for C. upsaliensis compared to C. jejuni [38, 39], which is not the case with the SCFIA.
In this study, the SCFIA showed good reactivity for C. ornithocola and C. armoricus, in addition to species listed in the manufacturer’s instructions (C. jejuni, C. coli, C. lari, and C. upsaliensis). This finding is not surprising, as C. ornithocola and C. armoricus were described in 2017 and 2019 as members of the C. lari group [41, 42] due to their strong phylogenetic relationship. Thus, the reactivity of the SCFIA was expected due to antigen similarity between these two species and C. lari. The ability of certain kits to detect C. lari has been described; for example, Kawatsu et al. [37] reported a Campylobacter immunochromatographic assay that detects a 15-kDa cell surface protein of C. jejuni, and Regnath et al. [43] reported similar results using the RIDA QUICK and RIDASCREEN Campylobacter kits (R-biopharm). The clinical significance of this finding is likely limited. Both C. ornithocola and C. armoricus are rarely detected in stool specimens from patients with gastroenteritis because these species are not detectable by the syndromic PCR formats marketed worldwide. They are also difficult to distinguish from C. lari via MALDI-TOF due to their phylogenetic proximity [41, 42]. In 2021, we detected C. ornithocola and C. armoricus in 2 and 4 of 8,709 strains sent to our reference center, respectively. In 2022, 4 of 8,971 strains tested at our reference center corresponded to C. ornithocola, whereas C. armoricus was not detected. These data indicate that infections with these two species are very rare [44].
However, we acknowledge the absence of detection of C. fetus, which is the third most common Campylobacter species isolated from campylobacteriosis specimens in France (NRCCH data available on www.cnrch.fr). Campylobacter fetus can cause invasive infections in elderly or immunocompromised patients [5]. Unfortunately, no kit on the market currently detects this species, making this detection gap common. Investigating the absence of cross-reactivity with closely related bacteria such as Aliarcobacter and enterohepatic Helicobacter in the kit was crucial in developing and evaluating a new test, as has already been achieved by other research teams [36, 37].
Conclusion
SCFIA is a rapid, accurate antigen test that could be of great utility to severely ill hospitalized patients who could benefit from early targeted antimicrobial therapy. Its capacity to detect the main species responsible for campylobacteriosis is of major interest for clinical use.
Anzeige
Acknowledgements
The authors acknowledge the support of QuidelOrtho INC in providing the fluorescent immunoassays used in this study. The manuscript was edited for proper use of English language by Textcheck (reference number: 25011405). Please direct any questions regarding this certificate or the English in the certified paper to: certified@textcheck.com.
Declarations
Ethics approval and consent to participate
All diagnostic methods are routinely performed at our institution. Therefore, informed consent was not requested from patients to use their stool samples. However, to ensure subject anonymity, all identifiable patient data were removed from the present study.
Consent for publication
Not applicable.
Competing interests
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
