Multiplex Molecular Stool Testing Rarely Impacts Antimicrobial Treatment Decisions More Than Three Days After Admission

Background Acute diarrheal illness in the United States is a significant cause of healthcare utilization and hospitalizations. For patients who develop diarrhea while hospitalized, testing for pathogens other than Clostridium difficile (C. difficile) using conventional stool testing is low yield. Newer testing modalities for infectious diarrhea such as the multiplex molecular stool testing provide an improved detection rate and a faster turn-around time compared to conventional stool testing. Methods We retrospectively examined the use of a multiplex molecular stool test at our institution for all hospital encounters over a two-year period to determine which organisms were identified ≤ 3 days and > 3 days after admission. Results A total of 2032 patients underwent multiplex molecular stool testing during the study period, with 1698 (83.6%) performed ≤ 3 days and 334 (16.3%) > 3 days after admission. An enteric non-C. difficile pathogen was identified more frequently when patients were tested ≤ 3 days after admission (350, 20.6%) as compared to > 3 days after admission (38, 11.4%, p<0.0001). Excluding coinfections, C. difficile was identified more frequently when patients were tested > 3 days after admission (64, 20.3%) versus another organism (30, 9.0%) (p<0.0001). Of those patients with a non-C. difficile pathogen identified > 3 days after admission, a bacterial pathogen amenable to treatment was only identified in 6% (21) of patients. Conclusion Multiplex molecular stool testing for patients tested > 3 days after admission is a low yield of information that could guide antimicrobial treatment decisions, and C. difficile testing is more useful in this clinical situation.


Introduction
Acute diarrheal illness is a significant cause of morbidity and mortality worldwide, with the most common infectious etiologies being Norovirus, Campylobacter spp., and non-typhoidal Salmonella infections [1]. In resource-rich countries, such as the United States, diarrhea illness is a rare cause of death but is a significant source of outpatient care utilization and hospitalizations [2]. In the US, diagnostic testing for communityonset diarrhea should be performed in patients with fever, bloody or mucoid stools, severe symptoms, or an immunocompromised state who would benefit from antimicrobial therapy [3].
Conversely, for patients who develop diarrhea while hospitalized, testing for infectious etiologies via stool culture for enteric pathogens is low value, as well as costly [4,5]. Prior studies with conventional stool testing methods have shown that in patients who develop diarrhea while hospitalized, testing for pathogens other than Clostridium difficile (C. difficile) is low yield when performed > 3 days after admission [6], which has led to the adoption of the "3-day rule" by many laboratories.
Conventional testing methods are increasingly being replaced by multiplex molecular stool testing, which provides an improved detection rate for multiple pathogens simultaneously with a high sensitivity and a rapid turnaround time [7][8][9]. However, it is not clear if the "3-day rule" should be applied to this new test. Prior studies on this topic revealed the percent positivity rates by molecular testing for stool samples 1 2 3 collected 72 hours after admission ranged from 6% to 14.6% [10][11].
Our institution began using multiplex molecular stool testing in place of routine stool culture for the diagnosis of suspected infectious diarrhea on January 5, 2015. This test identifies 22 different bacterial, viral, and parasitic stool pathogens, including C. difficile. In this study, we retrospectively examined the use of a multiplex molecular stool test for all hospital encounters over a two-year period to determine which organisms were identified in patients who are tested after being hospitalized ≤ 3 days and > 3 days. We also aimed to evaluate if prior guidance regarding a "3-day rule" is applicable in the molecular testing age.

Statistical analysis
Continuous variables were described using the median and interquartile range (IQR), and categorical variables were described with frequency and percentage. When excluding coinfections, the Chi-Square test was used as patients could only be placed into one group. When including co-infections, McNemar's test was used as patients could have tested positive for both enteric non-C. difficile pathogens and C. difficile. All statistical analyses were conducted using SAS 9.3 (SAS Institute, Cary, NC, USA).

Results
The study sample consisted of 2032 patients, of which the median age was 64 (IQR 51-76) years, and 854 were males (42%). The median number of days into the patient's hospital stay that the multiplex molecular stool test was performed was 1 (range 0-63) day. In total, 703 (34.6%) patients tested positive for any pathogen: 315 (15.5%) patients tested positive for C. difficile alone, 333 (16.4%) patients tested positive for at least one pathogen other than C. difficile, and 55 (2.7%) patients tested positive for both C. difficile and at least one other pathogen ( Table 1).

TABLE 1: Patient Demographics and Test Results
Data are n(%) unless otherwise stated; percentages might not add up to 100% due to rounding.

C. difficile, Clostridium difficile
The frequency and percentage of patients who tested positive for all pathogens included in the multiplex molecular stool assay, as well as the breakdown of the organism identified as compared to the day of testing (≤ 3 days versus > 3 days after admission), is shown in Table 2. An enteric, non-C. difficile pathogen was identified in 350 patients (20.6%) when patients were tested ≤ 3 days after admission as compared to > 3 days after admission (38, 11.4%, p<.0001) ( Table 2).

Discussion
This retrospective study reviewed the use of multiplex molecular stool testing performed ≤ 3 and > 3 days after admission at a large academic institution over a two-year period. Clinical decision support was not instituted with the transition from conventional stool testing to multiplex molecular stool testing and subsequently, it was noted that providers were using this test indiscriminately for all hospitalized patients with concern for infectious diarrhea.
Most patients had stool testing performed early in the admission (median 1, range 0-63); however, 334 (16.3%) of patients had multiplex molecular stool testing performed > 3 days after admission. For patients tested > 3 days after admission, testing was negative in 232 (69.5%), consistent with previous findings that nosocomial diarrhea is often noninfectious in etiology [12].
Patients were more likely to be diagnosed with an enteric non-C. difficile pathogen if tested ≤ 3 days after admission (350, 20.6%) as compared to > 3 days after admission (38, 11.4%, p<.0001). This is consistent with previous data showing that stool testing for infectious pathogens is higher yield earlier in admission. Importantly, after excluding coinfections, patients who had multiplex molecular stool test performed > 3 days after admission were more likely to have C. difficile identified (64, 20.3%) versus another organism (30, 9.0%) ( Table 1, p<0.0001). After a case review of those 30 patients tested > 3 days after admission who a pathogen other than C. difficile identified, only 21 (6%) were bacterial causes potential amenable to treatment, and of those, only 7 (2%) patients received treatment.
This strength of this study is that it is a large study reviewing the real-life application of a novel testing modality. This study provides actionable data for other institutions transitioning to multiplex molecular stool testing by supporting the implementation of clinical decision support for a "3-day rule" to discourage molecular stool testing > 3 days after admission. The limitations of this study include its retrospective nature and that it was performed at a single healthcare institution.

Conclusions
Hospitalized patients with diarrhea tested > 3 days after admission using multiplex molecular stool testing were more likely to have C. difficile identified versus another enteric organism. Clinical decision support is advised to guide providers to preferentially perform C. difficile testing for patients hospitalized > 3 days as opposed to multiplex molecular stool testing.  (HRP-103), which can be found by navigating to the IRB Library within the IRB system. Sincerely, Leroy Kromis, PharmD IRB Chair. Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue. Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.