Enteric Aeromonas Infection: a Common Enteric Bacterial Infection with a Novel Infection Pattern Detected in an Australian Population with Gastroenteritis

ABSTRACT Aeromonas species are emerging human enteric pathogens. However, they are currently not routinely detected in many diagnostic laboratories, and information regarding Aeromonas enteric infections detected using molecular methods is lacking. Here, we investigated the detection of Aeromonas species and four other enteric bacterial pathogens in 341,330 fecal samples from patients with gastroenteritis processed in a large Australian diagnostic laboratory between 2015 and 2019. These enteric pathogens were detected using quantitative real-time PCR (qPCR) methods. Furthermore, we compared the qPCR cycle threshold (CT) values obtained from fecal samples that tested positive for Aeromonas only by molecular detection with those of samples that tested positive by both molecular detection and bacterial isolation methods. Aeromonas species were found to be the second most common bacterial enteric pathogens among patients with gastroenteritis. We observed a unique pattern of three infection peaks for Aeromonas, which correlated with the age of the patients. Aeromonas species were the most common enteric bacterial pathogens in children younger than 18 months. Fecal samples that tested positive for Aeromonas only by molecular detection had significantly higher CT values than fecal samples that tested positive by both molecular detection and bacterial culture. In conclusion, our findings reveal that Aeromonas enteric pathogens exhibit an age-related three-peak infection pattern, distinguishing them from other enteric bacterial pathogens. Moreover, the high rate of Aeromonas enteric infection discovered in this study suggests that Aeromonas species should be routinely tested in diagnostic laboratories. Our data also show that combining qPCR with bacterial culture can enhance the detection of enteric pathogens. IMPORTANCE Aeromonas species are emerging human enteric pathogens. However, these species are currently not routinely detected in many diagnostic laboratories, and no studies have reported the detection of Aeromonas enteric infection using molecular methods. We investigated the presence of Aeromonas species and four other enteric bacterial pathogens in 341,330 fecal samples from patients with gastroenteritis using quantitative real-time PCR (qPCR) methods. Interestingly, we discovered that Aeromonas species were the second most common bacterial enteric pathogens in patients with gastroenteritis, exhibiting a novel infection pattern compared to those of other enteric pathogens. Furthermore, we found that Aeromonas species were the most prevalent enteric bacterial pathogens in children aged 6 to 18 months. Our data also revealed that qPCR methods exhibit higher sensitivity in detecting enteric pathogens compared to that of bacterial culture alone. Moreover, combining qPCR with bacterial culture enhances the detection of enteric pathogens. These findings emphasize the importance of Aeromonas species in public health.

Laboratory diagnosis of Aeromonas infections usually involves the isolation of potential Aeromonas species by bacterial cultivation using selective media developed for other enteric pathogens. Aeromonas species are then identified using various methods, such as biochemical tests or matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS). Currently, limited information is available regarding the use of molecular methods for detecting Aeromonas enteric infection in diagnostic laboratories.
In this study, we examined the detection of Aeromonas species and four other enteric pathogens by quantitative real-time PCR (qPCR) in 341,330 fecal samples from patients with gastroenteritis in a large Australian diagnostic laboratory. We also compared the cycle threshold (C T ) values obtained from fecal samples that were Aeromonas positive only by molecular detection with samples that were positive using both molecular detection and bacterial culture.

RESULTS
Aeromonas species were the second most common bacterial pathogens in patients with gastroenteritis. A total of 341,330 stool samples were analyzed between 2015 and 2019 by qPCR. Using commercially available molecular detection methods, the five common enteric bacterial pathogens were identified to the genus level. Campylobacter species were most common, with the detection rate being 512.96 per 10,000 samples. Aeromonas species were the second most common, with a detection rate of 255.15 per 10,000 samples. The detection rates for Salmonella, Yersinia, and Shigella/enteroinvasive Escherichia coli (EIEC) species were 144.52, 57.22, and 50.07 per 10,000 samples, respectively (Table 1).
Aeromonas enteric infection displayed three infection peaks related to patient age. The detection rates of Aeromonas species by age groups and gender of the patients are shown in Fig. 1. The Aeromonas detection rate in children 0 to 4 years old was high (325.48 per 10,000 samples) and formed the first peak of infection. Detection rates dropped in patients aged 5 to 19 and then increased in those aged 20 to 29 to 228.02 per 10,000 samples, forming the second small peak of infection. The detection rate continuously increased in patients over 50 years old and formed the third infection peak in patients aged over 80 years old (range, 253.64 to 459.76 per 10,000). The association between the Aeromonas detection rate and patient age was found to be significant (P , 0.001; odds ratio, 1.007; 95% confidence interval, 1.006 to 1.008). Male patients were found to have a higher detection rate than female patients (P , 0.001; odds ratio, 1.136; 95% confidence interval, 1.088 to 1.186). The number of positive results and processed samples in each age group are provided in Table 2.
The Campylobacter detection rate displayed a peak at 667.54 per 10,000 samples in young adults aged 20 to 29 years. There was an association between the Campylobacter detection rate and patient age (P , 0.001; odds ratio, 1.003; 95% confidence interval, 1.002 to 1.004). The detection rate in male patients was found to be significantly higher (P , 0.001; odds ratio, 1.752; 95% confidence interval, 1.699 to 1.807).
The detection rate for Salmonella species was found to have a peak in young children 0 to 4 years old (247.47 per 10,000 samples). The association between the Salmonella detection rate and patient age was found to be significant (P , 0.001; odds ratio, 0.987; 95% confidence interval, 0.986 to 0.988). Male patients were found to have a significantly higher detection rate than female patients (P , 0.001; odds ratio, 1.194; 95% confidence interval, 1.128 to 1.264).
For Shigella/EIEC and Yersinia species, the detection rates were observed to peak in young adults aged 20 to 29 years old. The Shigella/EIEC detection rate was found to be associated with patient age (P , 0.001; odds ratio, 0.996; 95% confidence interval, 0.994 to 0.998). The Shigella/EIEC detection rate was also found to be higher in male patients (P , 0.001; odds ratio, 1.439; 95% confidence interval, 1.307 to 1.584). The Yersinia detection rate was associated with patient age (P = 0.014; odds ratio, 1.002; 95% confidence interval, 1 to 1.004). However, patient gender was not found to be associated with the Yersinia detection rate.
Aeromonas species were the most common enteric bacterial pathogen in children younger than 18 months. Molecular detection revealed an Aeromonas enteric infection peak in children 0 to 4 years old. This infection peak was not seen in our previous report of Aeromonas enteric infection detection by bacterial culture in the Douglass Hanly Moir (DHM) laboratory during the same years. We therefore conducted further analysis by grouping these children based on 6-month age intervals. Interestingly, Aeromonas species showed a novel infection pattern compared to those of other enteric pathogens. Aeromonas enteric infections were common in children aged 6 to 24 months old. In children younger than 18 months old, Aeromonas species were the most common enteric bacterial pathogens (Fig. 2). The detection rates of Aeromonas species and four other enteric pathogens in male and female children less than 2 years old were not statistically different (P . 0.05).
Fecal samples positive for Aeromonas only by molecular detection had significantly higher mean C T values than fecal samples positive by both molecular detection and Aeromonas culture. The mean C T value of the 235 fecal samples that were positive using qPCR detection but negative using Aeromonas culture was 32.5, which was significantly higher than the mean C T value (29.03) of the 45 fecal samples that were positive by both qPCR detection and Aeromonas culture (P , 0.001; 95% confidence interval, 1.9 to 5.04).
Examination of false-negative qPCR results among samples with positive Aeromonas and other enteric bacterial culture. The false-negative qPCR results among samples with positive bacterial enteric pathogens varied among different bacterial species. Aeromonas had the highest false-negative rate (11.73%), followed by Salmonella (8.75%). The false-negative rates for Campylobacter and Shigella were low, at 1.1% and 0.1%, respectively (Table 3).

DISCUSSION
Aeromonas species are increasingly being recognized as human enteric pathogens. In this retrospective study, we investigated the presence of Aeromonas species and four other enteric bacterial pathogens detected using molecular detection methods. Our analysis involved a large number of fecal samples from patients with gastroenteritis. The findings from this study provide novel and valuable information regarding laboratory diagnosis of Aeromonas enteric infections.
Aeromonas species were the second most identified bacterial pathogens by molecular detection in patients with gastroenteritis. The most common enteric bacterial pathogen species detected were Campylobacter species (Table 1). Notably, molecular detection of Aeromonas enteric infection revealed a distinctive pattern characterized by three infection peaks, a unique feature not observed in the other four bacterial enteric pathogens tested (Fig. 1). The occurrence of Aeromonas enteric infections was predominantly observed in young children and individuals over 50 years old ( Fig. 1), suggesting a higher susceptibility to these infections during stages of relative immune weakness. Additionally, there was an increase in Aeromonas enteric infections among patients aged 20 to 29 years, possibly attributed to increased exposure to Aeromonas pathogens. Our previous research demonstrated that the majority (96%) of Aeromonas enteric infections in Australia were locally acquired, with no history of overseas travel (25). Future investigations are needed to identify the sources of Aeromonas enteric infections in Australia; effective strategies can then be implemented to reduce these infections. Further analysis of enteric infections in children aged 0 to 4 years revealed that Aeromonas species were the most common bacterial enteric pathogens in children younger than 18 months old (Fig. 2). Previous studies have also reported cases of Aeromonas-associated gastroenteritis in children. For example, a prospective study conducted by Gracey et al. in 1982 in Western Australia examined the isolation of Aeromonas species from fecal samples of 1,156 children with gastroenteritis, with an equal number of age-and sex-matched controls. The study found Aeromonas species in 10.2% of children with diarrhea and 0.6% of controls (26). The most common presentation was self-limiting diarrhea, although 25% of patients exhibited dysentery-like symptoms (26). Consistent with our findings, Gracey et al. also found that Aeromonas enteric infections were most prevalent in children less than 2 years old (26). Another study by Qamar et al. in 2016 reported that Aeromonas was a significant pathogen (22.2%) for moderate to severe diarrhea in children under 5 years old in Pakistan and Bangladesh (27). Multiple studies from various global regions have also reported Aeromonasassociated gastroenteritis in children (28)(29)(30)(31)(32)(33). The findings from our current study, along with previous research, consistently support the importance of Aeromonas species as important enteric pathogens in young children.
The molecular detection methods used in this study generally exhibited higher sensitivity in detecting enteric pathogens compared to bacterial isolation methods. However, the degree of improvement varied for different pathogens. The detection rate of Aeromonas species using the qPCR method in this study was 255.15 per 10,000 samples, representing a 4.5-fold increase compared to the detection rate by Aeromonas isolation (56.73/10,000 samples) from fecal samples of patients with gastroenteritis at DHM during the same period (25). Similarly, the molecular detection rates for Campylobacter, Salmonella, and Yersinia in this study were 512.96, 144.52, and 57.22 per 10,000 samples, respectively, representing 1.66-, 1.04-, and 30.26-fold increases compared to bacterial isolation, respectively (25). It is worth noting that not all samples that tested positive by bacterial culture yielded positive results by qPCR. The false-negative rates of qPCR among samples with positive bacterial cultures varied from 0.1 to 11.73%. These findings show the importance of utilizing both bacterial culture and qPCR to enhance the detection of enteric pathogens.
Our analysis shows that fecal samples that were positive only by molecular detection of Aeromonas species had a significantly higher C T value from qPCR than fecal samples that were positive by both molecular detection and bacterial culture. This observation indicates that samples with lower quantities of Aeromonas bacteria were more likely to yield negative results in Aeromonas culture. Currently, the isolation of Aeromonas species from fecal samples in diagnostic laboratories primarily relies on selective media designed for other enteric bacterial pathogens (25). Our results suggest that Aeromonas-specific selective media should be developed to enhance the Aeromonas isolation rate in diagnostic laboratories.
In summary, we report that Aeromonas species were the second most common bacterial enteric pathogen species in patients with gastroenteritis in an Australian population, as detected by molecular methods. Our study also reveals three distinct peaks of Aeromonas enteric infection that are novel and age related. Furthermore, we show that Aeromonas species were the most common enteric bacterial pathogen species in children younger than 18 months. Currently, Aeromonas enteric pathogens are not routinely tested in diagnostic laboratories. The high rate of Aeromonas enteric infection discovered in our study and the importance of Aeromonas affecting different patient age groups suggest that Aeromonas and Yersinia (E) species were found to peak in young adults aged 20 to 29 years. A positive association between the detection rate and patient age was observed for Aeromonas, Campylobacter, and Yersinia (P , 0.001, P , 0.001, and P = 0.014, respectively), while a negative association was observed for Salmonella and Shigella/ EIEC (P = 0.987 and P , 0.001, respectively). EIEC, enteroinvasive E. coli.

MATERIALS AND METHODS
Clinical data of enteric bacterial pathogens detected using molecular methods. The data analyzed in this study were provided by the Douglass Hanly Moir (DHM) pathology laboratory in Sydney, Australia. From 2015 to 2019, a total of 341,330 stool samples from patients with gastroenteritis were tested for bacterial enteric pathogens using qPCR methods, including Aeromonas, Campylobacter, Salmonella, Shigella EIEC, and Yersinia. Fecal samples mixed in stool transport and recovery (STAR) buffer were extracted using a MagnaPure 96 (Roche, Basel, Switzerland) instrument, and qPCR was performed on the LightCycler 480 II (Roche) system using the LightMix gastro bacteria kit manufactured by TIB Molbiol (Berlin, Germany) for the detection of Aeromonas, Campylobacter, Salmonella, Shigella/EIEC, and Yersinia enterocolitica. C T cutoff values of #37 (#38 for Aeromonas) were considered positive if associated with an amplification curve. At DHM, one fecal sample from each case was used for molecular detection of enteric pathogens.
Examination of the detection rates of Aeromonas, Campylobacter, Salmonella, Shigella/EIEC, and Yersinia species in patients with gastroenteritis. The data of detection of the five enteric bacterial pathogens using qPCR methods at the DHM laboratory over 5 years (2015 to 2019) were analyzed. The detection rate for each enteric bacterial pathogen (positive results per 10,000 samples) was calculated based on the 5-year data.
Comparison of the detection rates for Aeromonas, Campylobacter, Salmonella, Shigella/EIEC, and Yersinia species in patients of different age groups. The patients were grouped into different age groups. The number of processed fecal samples in each of the age groups and the number of positive results were recorded. Detection rates were calculated for the five enteric bacterial pathogens in each of the age groups. Logistic regression analyses were performed using a binomial generalized linear model to assess the relationship between the isolation rates of enteric pathogens and patients' age or gender. The statistical analyses were performed by using R v. 4.0.4 software and RStudio v. 1.4.1106. A P value of ,0.05 was considered statistically significant.
Comparison of qPCR C T values. At the DHM, fecal samples were tested using both molecular detection and bacterial isolation of Aeromonas. To examine whether Aeromonas bacterial numbers in fecal samples are a potential factor contributing to Aeromonas isolation positivity, we compared the C T values of Aeromonas qPCR detection from fecal samples that were only positive by molecular detection with those of fecal samples that were positive by both molecular detection and Aeromonas culture. The qPCR C T values were available for 280 Aeromonas-positive fecal samples detected in the year 2021. Of the 280 fecal samples, 45 samples were positive by both qPCR detection and bacterial cultivation; the remaining 235 fecal samples were positive only by qPCR detection but negative for Aeromonas isolation. The mean C T values of the samples of these two groups were compared using Welch's t test.
For Aeromonas cultivation from fecal samples, the presumptive Aeromonas isolates were gathered from three different types of culture plates, which included xylose lysine deoxycholate (XLD) agar, thiosulfate-citrate-bile salts-sucrose (TCBS) agar, and horse blood agar (HBA) containing an ampicillin AMO25 disc (Oxoid, Scoresby, VIC, Australia), based on the appearance of the colonies, as described previously (25). The isolates were then subjected to an oxidase test. Those that tested positive for oxidase were subsequently identified using matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS). The results were reported as "Aeromonas species," without specifying the species due to the possible error in identifying Aeromonas beyond the genus level by MALDI-TOF MS.
Examination of false-negative qPCR results among samples with positive Aeromonas and other enteric bacterial culture. Previously, we reported the detection of Aeromonas and other enteric bacterial pathogens by bacterial culture at the DHM between 2015 and 2019 (25). These culture-positive samples were also screened for enteric pathogens by qPCR methods. Here, we examined the occurrence of false-negative qPCR results among samples that tested positive by bacterial culture.
Ethics approval. The use of bacterial detection data provided by DHM for this retrospective study, which did not involve patient consent, was approved by the University of New South Wales HREAP executive (HC200755).

ACKNOWLEDGMENTS
C. Yuwono conducted the analyses of enteric bacterial pathogen isolations. M. C. Wehrhahn organized the clinical data of molecular detection of Aeromonas and other enteric bacterial pathogens. F. Liu helped with the data analysis. L. Zhang conceived and supervised the project. All authors contributed to the manuscript writing and editing. This project was supported by the Boost Award and a Faculty Research Grant (grant no. PS46772) awarded to L. Zhang by the University of New South Wales.
We declare no conflicts of interest.