By Carol A. Kemper, MD, FACP

Dr. Kemper reports no financial relationships relevant to this field of study.

Utility of GI Multiplex Assay

SOURCE: Cybulski RJ Jr, Bateman AC, Bourassa L, et al. Clinical impact of a multiplex gastrointestinal PCR panel in patients with acute gastroenteritis. Clin Infect Dis 2018 April 25; doi:10.1093/cid/ciy357.

A nine-month prospective study was launched in 2016-2017 to compare the results of the new BioFire Gastrointestinal FilmArrayTM with conventional laboratory techniques in persons with acute diarrhea. A total of 1,887 consecutive stool samples obtained from both the inpatient and outpatient setting were tested in parallel using the FilmArray and stool culture. Multiple media were used, including blood agar, MacConkey, MacConkey-sorbitol, Salmonella-Shigella selective media, and Campylobacter selective agars, as well as enhanced media for recovery of Vibrio spp. Additional conventional testing included ova and parasite, Giardia antigen, Cryptosporidium/Cyclospora modified acid-fast smear, and a viral gastroenteritis PCR panel. Patients with Clostridium difficile detected by either FilmArray or conventional methods were excluded from the final analysis.

The FilmArray successfully identified one or more pathogens in 669 specimens (35.3%) compared with 113 specimens (6%) using conventional culture techniques. Of these 669 specimens, 155 (8.2%) were bacterial pathogens, generally isolated in culture — resulting in a 37% greater recovery of conventional bacterial pathogens by FilmArray compared with conventional methods. Of these, significantly more STEC, Plesiomonas shigelloides, and Yersinia enterocolitica were identified by PCR.

Organisms not readily captured in culture also more frequently identified by PCR included Campylobacter spp. and Shigella/Enteroinvasive Escherichia coli (which are indistinguishable by this array). Interestingly, the most common pathogen identified by FilmArray was C. difficile (25.5%). FilmArray also provided improved identification of non-bacterial pathogens, including E. histolytica and Giardia lamblia, when compared with standard ova and parasite examination (1.4% vs. 0.3%).

Co-infections also were much more readily identified using FilmArray, with multiple positive targets identified in 184 specimens (27.5% of all positive specimens). Remarkably, 38 of these involved organisms that might be isolated in culture, but only four of them were detected by culture. The remaining 115 co-infections were due to pathogens not generally isolated in culture.

Fourteen specimens were positive by culture but negative by FilmArray; 12 of these were organisms not included in the current FilmArray. These included Aeromonas (eight cases), Campylobacter hyointestinalis (three cases), and one case each of Helicobacter pullorum, Salmonella enterica (a non-typhoid species), and Campylobacter jejuni.

Although all specimens received between 11 p.m. and 7 a.m. were set up the following morning, the median turn-around time for FilmArray results still was only 18 hours compared with 47 hours using conventional culture techniques (P < 0.0001). The median time to initiation of appropriate therapy for organisms detected by FilmArray was 22 hours vs. 72 hours for conventional methods (P < 0.0001). Sixty percent of patients with positive stool specimens were treated with targeted therapy, indicating that physicians generally found the results clinically significant. Once results were available, treatment was discontinued in eight of nine cases identified as STEC.

The use of gastrointestinal multiplex assays for patients with acute diarrhea provides much higher yield and more quickly than conventional techniques — at least when it’s performed in-house. Thus, the FilmAssay required less labor and improved laboratory flow, but also importantly it allowed for directed rather than empiric therapy much more quickly an important stewardship goal. However, the clinical applicability of this newer technology is still in the learning phase, as a recent case so sharply illustrated. An American returning from India was hospitalized recently at our facility with severe watery diarrhea and profound hypotension and dehydration (he had been working in the slums). He had been hospitalized in India and treated with multiple different antibiotics, including parenteral therapy, before returning to the United States. His diarrhea continued, unabated, and a GI multiplex assay obtained in the outpatient setting was positive for Shigella, Giardia, and cholera. He received ciprofloxacin and tinidazole for five days, with some improvement, and then relapsed, with 15-20 watery stools a day.

Clinically, it certainly seemed like it could be relapsed cholera, but despite repeated courses of treatment? He required admission over the Memorial Day weekend, and all cultures and conventional assays were negative, including cultures for Vibrio, Shigella, Shigella toxin, and C. difficile. A repeat multiplex assay was still positive for cholera, Shigella, and Giardia. We were left with the perplexing task of sorting out whether this was drug-resistant cholera with relapse (and negative cultures), Shigella (with negative cultures and a negative toxin test), or possible Giardia (of course, complicated by the fact that stool O&P are now in the age of PCR, perversely, send-outs and the results delayed by five days or longer over the Memorial Day weekend), or was this something else? I don’t know who was more frustrated — the patient or us — although, granted, he was the one with unrestrained diarrhea. In the end, the stool was positive for Giardia trophozoites and he improved with nitazoxanide.

And as one of our internists explained to a very frustrated patient, the rest of the PCR results were just “footprints in the sand.”


‘The World Is Covered by a Thin Layer of Feces’

SOURCE: Janezic S, Mlakar S, Rupnik M. Dissemination of Clostridium difficile spores between environment and households: Dog paws and shoes. Zoonoses Public Health 2018, April 23: doi:10.1111/zph.12475. [Epub ahead of print].

This quote is the best line I’ve ever heard — by Lucy Tompkins, MD, Stanford (my infectious disease attending many years ago).

This smart little study examines the risk of acquiring Clostridium difficile when walking the dog (literally). The researchers examined 20 households in Eastern Slovenia with a pet dog. Five were urban households and 15 were rural. Samples from the shoes, household slippers, and dog paws were collected within 30 minutes of walking the dog or the owner returning from a walk. Duplicate samples were permitted in households with two dogs. All samples were submitted for PCR ribotyping and toxinotyping, as well as culture.

Ninety samples were collected from 20 households, including 25 from dog paws, 44 from shoes (both the right and the left), and 21 from household slippers. Of these, remarkably, C. difficile was detected on 31 of 90 specimens (34%) from 14 of the households (70%). C. difficile was isolated from 43% of shoes, 28% of slippers, and 24% of dog paws. Altogether, 465 C. difficile isolates were obtained and sequenced, revealing 13 different ribotypes. Half were PCR ribotype 014/020, which was found in 18 different samples collected in eight different homes. Five of these 13 different ribotypes were toxigenic.

This study fits nicely with earlier work in New York City, which found that sand boxes and dog play areas often are contaminated with C. difficile. Basically, C. difficile is all around us. But it does make me wonder about the risk of spreading C. difficile in the hospital on my shoes. We don’t allow dogs in isolation rooms for this reason, but could my feet be a vector while I perform my daily rounds?