Rapid Molecular Detection of Gastrointestinal Pathogens and Its Role in Antimicrobial Stewardship

ABSTRACT

We aimed to detect the etiological agents of acute diarrhea by a molecular gastrointestinal pathogen test (MGPT) and to assess the impact of MGPT on antimicrobial stewardship programs (ASP). This is a prospective observational study and was conducted between 1 January 2015 and 30 June 2017. We included consequent patients who had acute diarrhea. At the end of 2015, we implemented ASP in acute diarrhea cases and compared the outcomes in the pre-ASP and post-ASP periods. An FDA-cleared multiplexed gastrointestinal PCR panel system, the BioFire FilmArray (Idaho Technology, Salt Lake City, UT), which detects 20 pathogens in stool, was used. In 499 out of 699 patients (71%), at least one pathogen was detected. Among 314 adults with positive MGPT, 101 (32%) enteropathogenic Escherichia coli (EPEC), 71 (23%) enteroaggregative E. coli (EAEC), 68 (22%) enterotoxigenic E. coli (ETEC), 55 (18%) Shiga toxin-producing E. coli (STEC) (17%) Norovirus, 48 (15%) Campylobacter, 21 (7%) Salmonella, and 20 (6%) Clostridium difficile strains were detected. Among 185 children, 55 (30%) EPEC, 37 (20%) C. difficile, 32 (17%) Norovirus, 29 (16%) EAEC, 22 (12%) STEC, 21 (11%) ETEC, 21 (11%) Campylobacter, 20 (11%) Salmonella, and 16 (5%) Rotavirus strains were detected. Inappropriate antibiotic use decreased in the post-ASP period compared with the pre-ASP period among inpatients (42.9% and 25.8%, respectively; P = 0.023). Using MGPT in clinical practice significantly decreased the unnecessary use of antibiotics. Detection of high rates of C. difficile in children and Salmonella spp., as well as relatively high rates of Campylobacter spp., which were hard to isolate by routine stool culture, were remarkable.

INTRODUCTION

Acute gastroenteritis (AGE) causes considerable morbidity and mortality in all over the world, with an estimated 1.4 million deaths every year (1, 2). In recent studies, enteropathogenic Escherichia coli (EPEC), enterotoxigenic E. coli (ETEC), Clostridium difficile, enteroaggregative E. coli (EAEC), Norovirus, and Campylobacter strains were the most commonly detected pathogens in etiology of acute diarrhea (3–6). The lack of detection of the etiologic agents in AGE may lead to inappropriate antibiotic use.

The increasing availability of molecular tests might identify pathogens that could have been difficult to recognize using traditional methods and that could be a crucial component of antimicrobial stewardship programs (ASP) (7–9). The clinical use of molecular gastrointestinal pathogen test (MGPT) in respiratory infections was implemented, but not yet for diarrhea cases (7, 10). According to the Infectious Diseases Society of America (IDSA) guideline for the diagnosis and management of infectious diarrhea, MGPT was placed in detection of outbreaks in public health, but its use in routine clinical practice was not detailed (11). However, the contribution and place of MGPT in AGE remain to be defined. Additionally, detection of the pathogens could be added to ASP, which could evolve the health care provider behaviors and clinical practices (12). In this study, we aimed to determine the place of MGPT in AGE and to assess its impact in ASP.

MATERIALS AND METHODS

Study design and population.

This is a prospective observational study and was conducted between 1 January 2015 and 30 June 2017. We designed our study according to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) study (13).

We included consequent patients who had acute watery diarrhea and fever for more than 72 h or acute bloody diarrhea, and MGPT was performed in a 265-bed private hospital in Istanbul, Turkey. After the introduction of the molecular rapid test in our hospital, its use increased each year. At the end of 2015, we implemented ASP, including educational meetings and activities among prescribers, and we compared outcomes pre-ASP and post-ASP. The pre-ASP period was between 1 January 2015 and 31 December 2015, and the post-ASP period was between 1 January 2016 and 30 June 2017.

Information about clinical features of the patients, demographic features, chronic disorders, and detailed information about antibiotic use during hospitalization were collected by the medical records. Laboratory tests, including stool analysis, stool culture, blood culture, blood electrolyte levels, complete blood count, and C-reactive protein (CRP) levels, were studied according to patient clinical condition.

Definitions.

WHO defines diarrhea as the passage of three or more loose or liquid stools per day, or more frequently than is normal for the individual (11). Diarrhea is classified as inflammatory or noninflammatory according to the pathophysiology. Inflammatory diarrhea is generally caused by invasive or toxin-producing bacteria. Common pathogens are Salmonella, Shigella, Shiga toxin-producing E. coli, enteroinvasive E. coli (EIEC), C. difficile, Yersinia, and Entamoeba histolytica. Noninflammatory diarrhea is usually viral but can be bacterial or parasitic. ETEC, Staphylococcus aureus, Rotavirus, Norovirus, Giardia, Cryptosporidium, and Vibrio cholerae are the common pathogens in etiology (14, 15).

Unnecessary antibiotic use based on the MGPT was defined as inappropriate. Inappropriate antibiotic use duration was calculated after the detection of a pathogen that does not require an antibiotic.

Diagnosis.

For molecular detection and identification of gastrointestinal pathogens, an FDA-cleared multiplexed gastrointestinal PCR panel system, the BioFire FilmArray (Idaho Technology, Salt Lake City, UT), which detects 20 pathogens (Campylobacter species [C. jejuni/C. coli/C. upsaliensis], Clostridium difficile toxin A/B, Plesiomonas shigelloides, Salmonella species, Vibrio species [V. parahaemolyticus, V. vulnificus, and V. cholerae], Yersinia species, EAEC, EPEC, ETEC, Shiga toxin-producing E coli O157, Shigella/enteroinvasive E. coli [EIEC], Cryptosporidium species, Cyclospora cayetanensis, Entamoeba histolytica, Giardia lamblia [G. intestinalis and G. duodenalis], Adenovirus F 40/41, Astrovirus, Norovirus GI/GII, Rotavirus A, and Sapovirus), was used according to the manufacturer's instructions. The MGPT gave results within 3 h, which were immediately reported to the physician. Stool microscopy and routine stool culture were also used for diagnosis if needed.

Data analysis.

In statistical analysis, a chi-square test for a comparison of categorical variables and t test for continuous variables were used. STATA 11 (TX, USA) was used for analysis, and a P value of <0.05 was set as the significance level. The institutional review board of Koç University approved the study.

RESULTS

In total, 699 patients were included. In 499 (71%) patients, at least one pathogen was detected, and 176 out of 499 (36%) were inpatients (Fig. 1). Among 314 adults with positive MGPT, 101 (32%) EPEC, 71 (23%) EAEC, 68 (22%) ETEC, 55 (18%) STEC, 53 (17%) Norovirus, 48 (15%) Campylobacter, 21 (7%) Salmonella, and 20 (6%) C. difficile strains were detected (Fig. 2). Among 185 children, 55 (30%) EPEC, 37 (20%) C. difficile, 32 (17%) Norovirus, 29 (16%) EAEC, 22 (12%) STEC, 21 (11%) ETEC, 21 (11%) Campylobacter, 20 (11%) Salmonella, and 16 (5%) Rotavirus strains were detected (Fig. 2). The rates of multiple pathogens detected by MGPT were 42.4% in adults (133 out of 314 patients) and 36.8% in children (68 out of 185 patients).

FIG 1.

Study population.

FIG 2.

The most commonly detected pathogens among the patients with acute gastroenteritis. y axis represents number of each pathogen detected.

We performed the microscopic examination of the stools that included patients infected with one pathogen. While erythrocytes were detected in 51% and leukocytes were detected in 73% of the stool microscopy images in inflammatory gastroenteritis, in noninflammatory gastroenteritis, these rates were 14% and 49%, respectively.

Stool cultures of 26 Salmonella-positive patients (with MGPT) were performed, and Salmonella was grown in 7 (27%) of them.

Among hospitalized patients with positive MGPT, inappropriate antibiotic use was detected in 24 out of 56 (42.9%) patients pre-ASP and in 31 out of 120 patients (25.8%) post-ASP (P = 0.023). In subgroup analysis, inappropriate antibiotic use decreased both in children and adults in the post-ASP period compared with the pre-ASP period, but due to the low number of patients in subgroup analysis, the difference was not statistically significant (decreased from 41% to 20% in children and from 44.1% to 29.3% in adults). The mean duration of inappropriate antibiotic use was 6.4 days in the pre-ASP term and 7.2 days in the post-ASP term (P > 0.5). The most commonly used inappropriate antibiotics were quinolones, nitroimidazoles, and third-generation cephalosporins.

DISCUSSION

Diarrhea is one of the leading causes of antibiotic consumption. Because of the high rate of inappropriate antibiotic use despite the availability of the MGPT, we implemented an ASP that included training sessions about the diagnosis and management of diarrhea at the end of 2015. Then we compared outcomes in the pre-ASP and pre-ASP periods. In our study, in 43% of patients, antibiotic treatment was maintained inappropriately in the pre-ASP period. After ASP, inappropriate antibiotic use decreased to 26% (P = 0.023). Despite the fact that it is still high, the antibiotic use rate may continue to decrease with continuous training activities and by the increase in experience of health care providers in terms of MGPT.

In our study, the most commonly detected pathogens in acute diarrhea were EPEC, EAEC, ETEC, Norovirus, STEC, and Campylobacter species. The high rates of C. difficile in children and Salmonella spp., and relatively high rates of Campylobacter spp., which was hard to detect by routine stool culture and frequently require antibiotic treatment, were also remarkable findings. The parasites, including Entamoeba histolytica, were rarely detected. In a multicenter point-prevalence study of community-acquired diarrhea from Europe, of the 709 samples, 268 samples yielded only one pathogen, and 116 samples yielded multiple organisms by the FilmArray gastrointestinal (GI) panel. The most commonly detected pathogens were strains of EPEC, Campylobacter, toxigenic C. difficile, EAEC, Norovirus, and ETEC, similar to what was found in our study. Similarly, they also did not detect Entamoeba histolytica or Vibrio species (5). In a study by Kharo et al., among 76 positive samples, EPEC was the most commonly detected pathogen, followed by toxigenic C. difficile, Sapovirus, EAEC, Norovirus, Campylobacter, and ETEC (6).

Stool microscopy is generally used as a first-line diagnostic tool in an initial evaluation of patients with acute diarrhea. In our study, although leukocytes and erythrocytes were seen more commonly in inflammatory gastroenteritis than in noninflammatory gastroenteritis, they were not sufficient to discriminate inflammatory and noninflammatory gastroenteritis. Stool culture, another conventional diagnostic test, only detected 27% of Salmonella infections, which means that routine stool culture fails to detect each three of four cases in our study population, but it could be easier to detect if the stool culture was enriched.

C. difficile represents one of the most critical threats categorized by the Centers for Disease Control and Prevention (16). In our study, not in adults but among children, C. difficile was found to be one of the most commonly detected pathogens. ASP is crucial for decreasing C. difficile infections by reducing antibiotic consumption. In recent studies, including a meta-analysis, it was shown that the incidence of C. difficile infections among inpatients decreased by implementing ASP (17–19).

One of the strongest point of our study was having the opportunity to use MGPT whenever needed, so that we reached to high number of cases. Additionally, we had the chance to obtain an MGPT result within 3 h, so we were able to decide whether to start an antibiotic in an early stage of management. One of the limitations of this study was being a single center. Additionally, the detection of pathogens using MGPT does not mean that the detected pathogen is the cause of diarrhea, but this is a general issue for molecular diagnostic tests, and the test results should be interpreted carefully.

In this study, we clearly demonstrated the pathogen distribution in acute diarrhea with a large sample size, and we also showed beneficial effect of the MGPT by a significant decrease in unnecessary use of antibiotics, so that MGPT could be beneficial in ASP in acute diarrhea.

In conclusion, using MGPT in clinical practice significantly decreased the inappropriate use of antibiotics. The high rates of C. difficile in children and of Salmonella spp., and relatively high rates of Campylobacter spp., which was hard to isolate by routine stool culture, were remarkable.