Experience with PCR Testing for Enteric Bacteria and Viruses of Emergency Department Patients with Acute Gastroenteritis: Are There Implications for the Early Treatment of Clostridioides difficile Infection?

Early identification of acute gastroenteritis (AGE) pathogens via PCR may improve the management of patients presenting to the emergency department (ED). In this study, we evaluated the implementation of a testing algorithm for ED patients with AGE using the BD MAX automated PCR system. Data from 133 patients were analyzed. A total of 56 patients (42%) tested positive via PCR for at least one bacterial or viral pathogen. The median time to report PCR results was 6.17 h compared to 57.28 h for culture results for bacterial pathogens. The most common pathogen was Clostridioides difficile (n = 20, 15%). In total, 14 of the 20 C. difficile-positive patients were aged >65 years and 17 of the 20 patients (85%) were diagnosed with a clinically relevant infection based on typical symptoms and laboratory values. They received antibiotics, mostly oral vancomycin, starting a median of 11.37 h after ED admission. The introduction of PCR for the diagnosis of AGE infection in patients presenting to the ED may have the greatest impact on the rapid identification of C. difficile and the timely administration of antibiotics if necessary.


Introduction
Multiplex PCR is being increasingly used for the diagnosis of acute gastroenteritis (AGE), not only to increase diagnostic sensitivity but also to provide timely diagnosis compared to bacterial culture and conventional stepwise diagnostics [1,2].However, PCR cannot replace the isolation of bacterial pathogens by culture, which is required for serotyping and accurate characterisation of different pathotypes, e.g., Shiga toxinproducing E. coli (STEC) and Salmonella.PCR causes additional laboratory costs, and the clinical impact in terms of specific therapy is limited [1].In the outpatient setting, testing for viruses is not indicated because the results have no impact on the treatment of the patient [3].However, identification of the bacterial pathogen may be helpful in deciding whether to use an antibiotic, for example, in cases of Shigellosis, prolonged Campylobacter infection, severe symptoms, or underlying chronic disease [3,4].In the hospital setting, PCR testing for viruses and bacteria could also improve patient hygiene management.However, broad-spectrum multiplex PCR testing is not recommended for patients who have been hospitalised for more than 48 h because infectious hospital-associated AGE is associated with a small spectrum of pathogens, mainly norovirus and Clostridioides difficile [3].
The emergency department (ED) is a critical interface for timely diagnostics in a tertiary hospital.Immediate testing of patients presenting with AGE in the ED may help to improve patient management, as they need to be isolated when admitted to a ward and this may be important for timely therapeutic decisions.Many patients presenting to the ED with severe illness belong to vulnerable populations, such as the elderly or immunocompromised.For these patient groups, broad panel testing for AGE pathogens may be most relevant.This also includes children, who are known to have an increased incidence of AGE [1,5].As there is a large overlap in clinical symptoms caused by different pathogens, it is challenging to differentiate patients for more individualized testing to avoid delays in diagnostic reporting and to simplify the workflow for the emergency department staff.This problem also applies to testing for C. difficile, which can cause mild disease as well as severe pseudomembranous colitis [6].
Here we report the results of an observational study on the testing of ED patients with AGE using the enteric viral panel (EVP), enteric bacterial panel (EBP), and Cdiff PCR assays of the automated BD MAX ™ system (BD, Heidelberg, Germany).All assays were performed in parallel from the same sample, including the PCR assay as a one-step diagnostic strategy for C. difficile infection (CDI).

Hygiene Management
Our hospital hygiene management requires the isolation of patients with diarrhea until a completely negative diagnostic result has been obtained.There were two cases of patients in whom a pathogen not included in the PCR panels used here was subsequently identified.In one of these cases, Yersinia enterocolitica was isolated from the same sample and another patient tested positive for Cryptosporidium parvum in a second sample.Of note, additional BD Max PCR panels covering both pathogens are available but were not used for the initial testing due to the low incidence of these pathogens to reduce diagnostic costs.In total, 85 out of 133 patients were admitted to a hospital ward.While there was no significant difference in the length of hospital stay between PCR-negative and PCR-positive patients, the latter spent more time in single-room isolation (Table 2).This was because 24% of the hospitalized patients with a negative PCR result (12/50) but only 2.9% of patients with a positive PCR result (1/35) were initially not correctly isolated after transfer to a ward (p = 0.027, Fisher's exact test).

C. difficile
The most common pathogen detected was C. difficile, with 20 cases out of 133 patients (15%), and as expected, the majority of cases were in the older patient group (Table 3).A total of 18 out of 20 patients (90%) were hospitalized.Therefore, further data analysis on antibiotic treatment decisions focused on these cases.According to international guidelines, the diagnosis of CDI was based on the criteria of a positive laboratory test in the presence of diarrhea and/or pseudomembranous colitis, taking into account additional parameters such as fever, elevated white blood cell (WBC) count, and a 50% increase in serum creatinine level.In three patients (15%), including two children, the positive PCR result was not clinically relevant due to the absence of characteristic symptoms or symptoms of other causes (Table 3).Eleven patients (55%) were diagnosed with CDI for the first time.PCR Ct values showed a wide variation in cases of clinical relevance and did not correlate with WBC or creatinine laboratory parameters (Table 3).Specific antibiotic treatment was preferentially initiated with vancomycin, as recommended (Table 3).For cases for which the time of initiation of antibiotic therapy was available in the hospital information system, the median time from ED presentation to antibiotic administration was 11.37 h (IQR 7.22-15.25)and the median time to report a positive C. difficile toxB PCR result was 8.9 h (IQR 4.83-13.75).It should be noted that, in several cases, vancomycin was given before the PCR result was obtained.One of the hospitalized C. difficile-positive patients was not correctly isolated as recommended.For the other patients, the median time of isolation was 182.33 h (IQR 140.44-235) compared to a median hospital length of stay of 193.98 h (IQR 143.06-264.16).Fulminant cases of CDI were not observed in this study.

Discussion
The implementation of PCR into the diagnostic workflow of AGE may result in a benefit for optimizing patient management [1,4,7].In this study, the value of PCR for testing ED patients was examined.For bacterial pathogens that are conventionally diagnosed by culture, the detection rate was slightly increased and the main advantage of PCR testing was a shorter time to report positive test results.As expected, the most commonly detected enteric virus was norovirus, an important pathogen often associated with transmission between patients in close contact in hospitals leading to nosocomial infections [3].Timely diagnosis of AGE may have a therapeutic benefit by both reducing unnecessarily prescribed antibiotics and targeting a more specific therapy when needed but the impact of rapid multiplex PCR on patient hygiene management should not be overestimated [1,3,8].As PCR cannot exclude all pathogens, patients with AGE but a negative screening PCR result are isolated upon admission to a ward until all diagnostic test results are available or until 48 h after suspending diarrhea.However, errors in hygiene management, as also observed in this study, may occur, and it can be argued that staff will be better sensitized to isolate the patient for whom a positive PCR test is reported in a timely manner.
The most common pathogen detected in ED patients was C. difficile, which, as expected, was detected in higher numbers in the older patient group [6].We included PCR testing for C. difficile as a one-step diagnostic in the workflow.This did not strictly follow the preferred workflow in the diagnostic guidelines of the European Society of Clinical Microbiology and Infectious Diseases (ESCMID) and the Infectious Diseases Society of America (IDSA) which recommend a two-step diagnostic algorithm consisting of a sensitive screening test and a less sensitive confirmatory C. difficile toxin assay [9,10].The use of a PCR test alone is only a weak recommendation when there are specific institutional criteria for the submission of patient stool samples [10].Confirmatory testing with a second assay is suggested to reduce the number of clinically irrelevant results, thereby reducing overdiagnosis of CDI [11].C. difficile can cause only colonization or disease with a wide variety of symptoms, ranging from diarrhea, vomiting, fever, and abdominal pain to severe pseudomembranous colitis and even septic shock [6].The use of PCR as a stand-alone test has not been defined as an optimal approach to differentiate between colonization and infection [9].On the other hand, there are studies in which the use of one-step PCR testing results in more patients being accurately diagnosed and fewer deaths, indicating that suboptimal analytical sensitivity in two-step algorithms may miss cases that should be treated [12][13][14].It has been shown that about 70% of PCR-positive but toxin antigen-negative patients were found to have probable or possible CDI [13].It is therefore not surprising that the guidelines also state that a positive PCR screening test in combination with a negative toxin antigen test needs to be clinically evaluated.However, this strategy may lead to excessive repeat testing.In this study, the decision was made to include C. difficile in the PCR panel as a one-step testing strategy in order to reduce the workload and spare time.C. difficile testing is not routinely recommended for children as standard diagnostics but was included in this study in order to simplify the workflow [10,15].Although low PCR Ct values have been reported to be associated with clinically relevant infections, no Ct cut-off values can be defined to distinguish colonization from infection [16,17].In addition to clinical symptoms, elevated WBC count and serum creatinine levels may be considered in the diagnosis of CDI as they may indicate the development of severe infection [18].In this study, there was no correlation between low PCR Ct values and both WBC count and serum creatinine, and all parameters should be used together to assess clinical relevance.
The ED is critical for the early identification of CDI in patients presenting with AGE [18,19].Reassessment of admitted patients in the following days may not be the optimal method for CDI diagnosis and the timely initiation of antibiotic treatment to prevent the development of severe or fulminant disease [18].As shown here, the majority of ED patients with a positive PCR test result were diagnosed with a clinically relevant infection that required treatment.Based on the information available in the patient records, approximately 69% of the patients had no recent history of CDI and the result was inter-preted as a first diagnosis.The most commonly used antibiotic was oral vancomycin [20].In our hospital-specific guidelines, fidaxomicin is preferentially recommended for the treatment of patients at high risk of recurrent infection [20].
In conclusion, this study shows that the implementation of PCR testing for ED patients with AGE has the greatest impact on the early detection of CDI in the elderly patient population.As CDI can result in severe disease associated with high mortality and the risk of readmission in patients with underlying disease or high age, C. difficile diagnosis based on PCR testing alone in the context of clinical symptoms is acceptable for therapeutic decisions.Timely reporting of positive PCR results may also increase attention to basic hygiene measures like hand hygiene and pathogen-specific measures such as isolation and personal protective equipment.

PCR Testing and Microbiology Diagnostic Workflows
The samples were unpreserved stool samples collected from patients with acute gastroenteritis who presented to the ED of the Jena University Hospital between July 2019 and January 2022.Samples were transported to the microbiology laboratory via the central pneumatic tube system and immediately tested with the BD MAX ® EBP, EVP, and Cdiff assays during the normal working day between 7 a.m. and 5 p.m.Samples arriving after 5 p.m. were analyzed in the morning of the next day.Test results were reported by telephone to the ED as soon as possible between 7 a.m. and 8 p.m.
For bacterial cultures, samples were inoculated on Hektoen agar, salmonella-shigella agar, Butzler agar (Thermo Fisher Scientific, Wesel, Germany), a selenite broth (BD) at 37 • C, and a second selenite broth at 30 • C. Cultures were incubated for two days with daily visits for the growth of suspect colonies.Brilliance E. coli/coliform chromogenic agar (Thermo Fischer Scientific, Wesel, Germany) and GN broth (BD) were used for the isolation of STEC.Diagnosis on Y. enterocolitica, which is not included as a target in the BD MAX ™ EBP, was performed through inoculation on CIN (Thermo Fisher Scientific) agar incubated at 30 • C. The following methods were used to identify bacterial pathogens in the culture: Vitek MS (bioMeriéux, Nürtingen, Germany), eazyplex ® SalmoTyper LAMP assay (Amplex Diagnostics, Gars-Bahnhof, Germany), and seroagglutination (Sifin Diagnostics, Berlin, Germany) for Salmonella spp.; API E (bioMeriéux) and seroagglutination for Shigella spp.; eazyplex ® EHEC complete LAMP assay (Amplex Diagnostics) for STEC; Vitek MS for Campylobacter spp.; and Vitek MS and seroagglutination for Y. entercolitica.
All assays were performed according to the manufacturer's protocols.

Data Analysis
The turnaround times of PCR and culture microbiology diagnosis were calculated using the times recorded in the laboratory information system when sample testing was requested, and the results were reported to the ED.Laboratory data (WBC and creatinine) of patients who tested positive for C. difficile were obtained from the laboratory information systems.Medical records were reviewed to identify patients' symptoms and comorbidities, to determine when antibiotic therapy was changed or started after admission, and to calculate the duration of single-room isolation.Fisher's exact and Mann-Whitney U tests for statistical analysis were performed using the Statistics Kingdom online tool (www.statskingdom.com).

Table 2 .
Isolation of hospitalized patients.