ArticlePDF Available

748. The Changing Epidemiology of Clostridioides difficile Infection and the NAP1/027 Strain in Two Quebec Hospitals

Authors:

Abstract and Figures

Background In 2003, many hospitals in Québec, Canada experienced an increase in the incidence of healthcare-associated C. difficile infection (HA-CDI) associated with increased morbidity and mortality. This increase was associated with the dissemination of the NAP1/027 strain. The objective of this study was to describe the epidemiology of HA-CDI in two tertiary care hospitals based in Montréal from 2003 to 2019. Methods Surveillance for HA-CDI was performed using standard definitions from 2003 to 2019 at the Montreal General Hospital (MGH) and Royal Victoria Hospital (RVH), in Montréal, Québec. C. difficile was isolated from stool specimens using standard methods. Pulsed field gel electrophoresis and ribotyping were performed to determine genotype. Antibiotic utilization and infection control interventions implemented over the same time period were reviewed. Results A total of 4314 cases of CDAD were identified during the study period: 2295 at the RVH and 2019 at the MGH. The incidence decreased from 29.5 to 5.9 cases per 10,000 patient-days between 2003 and 2019 at the RVH and from 23.8 to 3.9 cases per 10,000 patient-days at the MGH. Of the 124 isolates available for genotyping in 2003, 112 were NAP1 (90.3%) compared to 5 out of 53 (9.4%) in 2019. Fluoroquinolone utilization decreased from 230 to 139 DDDs per 1,000 patient-days between 2003 and 2019, whereas total antibiotic utilization increased from 1296 to 1550 DDDs per 1,000 patient-days. Infection Control interventions included empirically placing patients with diarrhea on precautions, intensified cleaning measures, formal antibiotic stewardship, introduction of a real-time PCR C. difficile test in June 2010, and a move to a facility with only single rooms at the RVH in April 2015. Incidence of HA-CDI at the RVH and MGH and antibiotic utilization between 2003 and 2019 Conclusion An important change in HA-CDI epidemiology was observed in two Canadian tertiary care hospitals based in Montréal between 2003 and 2019. There was a significant decrease in incidence of HA-CDI and a genotype shift from a predominance of NAP1 strains to non-NAP1 strains. Utilization of fluoroquinolones, to which the NAP1 strain is resistant, concurrently decreased. Infection control interventions targeting isolation, diagnosis, disinfection, and antibiotic stewardship have contributed to the major observed reduction in HA-CDI incidence. Disclosures All Authors: No reported disclosures
Content may be subject to copyright.
Abstracts • OFID 2021:8 (Suppl 1) • S471
endonuclease analysis (REA) strain typing was performed on the recovered CD
isolates.
Results. Toxin testing was positive in 19/50 (38%) cases. Compared to stool tox-
in-negative cases, toxin-positive cases were older (95% vs. 71% were age 65, p=0.06),
more likely to have a history of CDI (37% vs. 23%, p=0.34), and have 1 CDI episodes
within 6months (37% vs. 19%, p =0.26). Treatment for CDI was more common in
patients who had a positive toxin text. (95% vs 61%, p= 0.009). Among the 38 patients
that received treatment, 33 received vancomycin (87%) and 8 patients (21%) had rCDI
at 30days. Of the 8 patients with rCDI, 2 were re-admitted to the hospital for CDI.
e average PCR cycle threshold was lower in the toxin-positive stools compared to
toxin-negative stools (24.46 and 29.96, p< 0.001; Fig. 1)e endemic REA group Y
was the most common CD strain recovered (30%) and the previously epidemic and
virulent REA group BI strain was recovered in 11% of the cases.
Conclusion. CDI cases diagnosed by positive stool PCR and positive toxin tests
had more typical risk factors for CDI, a lower PCR cycle threshold and were more likely
to have been treated for CDI. Outcomes were similar in this setting where infection
with the virulent BI strain was uncommon.
Disclosures. Stuart Johnson, MD, Acurx Pharmaceuticals (Advisor or Review
Panel member)Bio-K+ (Advisor or Review Panel member)Ferring Pharmaceutical
(Advisor or Review Panel member)
747. Association of Clostridiodes difficile Infection Incidence With Renewed Vigor
in Infection Prevention Practices With the Onset of the COVID-19 Pandemic
AhmedA.Khan, MD1; SanaWa qa r, MD2; 1Southern Illinois University School of
Medicine, Springeld, IL; 2Southern Illinois University, Springeld, IL
Session: P-36. HAI: C.difficile
Background. Clostridioides dicile is the leading cause of hospital associated
infections. In 2017 it lead to an estimated 223,900 cases, 12,800 deaths and $1 billion
in attributable healthcare costs.[1] Judicious use of antibiotics and good hand hygiene
practices form the cornerstone of prevention. During the COVID-19 pandemic there
has been a focus on infection control practices such as hand hygiene, which would also
lead to decreased incidence of other contagious infections such as C.dicile diarrhea.
Methods. We looked at the incidence of C.dicile infection in a tertiary care hos-
pital, 1year before and 1year aer the start of the COVID-19 pandemic. We looked at
the absolute number of hospital associated C.dicile infections and the rate per 1000
patient days. e testing methodology changed during the time of the study. Initially it
included NAAT for C.dicile, however in March of 2020 the testing strategy included
testing for GDH antigen and toxin A/B to dierentiate between infection and asymp-
tomatic colonization.
Results. From January 1st and December 31st 2019 there were a total of 182 C.dif-
cile infections with a rate of 1.29% per 1000 patient days. Between January 1st and
December 31st 2020 there were a total of 51 C.dicile infections with a rate of 0.39%
per 1000 patient days. ere was an absolute risk reduction of 0.9% and relative risk re-
duction of 69.7%. Hand hygiene audits did not show a dierence in adherence between
the two periods, with a compliance rate of 98% for both.
Conclusion. Our data suggests that there was a substantial reduction in C.di-
cile infection rate aer widespread knowledge of COVID-19 and implementation of
enhanced infection prevention strategies. ese included frequent reminders of hand
washing, gowning and social distancing to name some. is information was conveyed
in the form of widely disseminated signs in highly visible areas, frequent reminders
electronically and in person between sta and providers. ere are limitations in our
study, which include diculty in longitudinally assessing the extent to which patient
care providers adhered to infection prevention strategies and a change in testing
strategy for C.dicile diagnosis during this time.
Disclosures. Al l Authors: No reported disclosures
748. The Changing Epidemiology of Clostridioides difficile Infection and the
NAP1/027 Strain in Two Quebec Hospitals
SandrineCouture, M.D.,C.M.1; CharlesFrenette, MD, FRCPC1; RowinAlfaro, B.
Sc1; LorneSchweitzer, MD1; IanSchiller, MSc1; NancyDoherty, College Diploma1;
RahulNanda, M.D.,C.M.1; YvesLongtin, MD2; Danielirion, PharmD1;
VivianLoo, MD, M.Sc.3; 1McGill University Health Centre, Montreal, Quebec,
Canada; 2Jewish General Hospital, Montreal, Montreal, QC, Canada; 3McGill
University, Montreal, Quebec, Canada
Session: P-36. HAI: C.difficile
Background. In 2003, many hospitals in Québec, Canada experienced an in-
crease in the incidence of healthcare-associated C.dicile infection (HA-CDI) asso-
ciated with increased morbidity and mortality. is increase was associated with the
dissemination of the NAP1/027 strain. e objective of this study was to describe
the epidemiology of HA-CDI in two tertiary care hospitals based in Montréal from
2003 to 2019.
Methods. Surveillance for HA-CDI was performed using standard denitions
from 2003 to 2019 at the Montreal General Hospital (MGH) and Royal Victoria
Hospital (RVH), in Montréal, Québec. C.dicile was isolated from stool specimens
using standard methods. Pulsed eld gel electrophoresis and ribotyping were per-
formed to determine genotype. Antibiotic utilization and infection control interven-
tions implemented over the same time period were reviewed.
Results. A total of 4314 cases of CDAD were identied during the study period:
2295 at the RVH and 2019 at the MGH. e incidence decreased from 29.5 to 5.9
cases per 10,000 patient-days between 2003 and 2019 at the RVH and from 23.8 to
3.9 cases per 10,000 patient-days at the MGH. Of the 124 isolates available for gen-
otyping in 2003, 112 were NAP1 (90.3%) compared to 5 out of 53 (9.4%) in 2019.
Fluoroquinolone utilization decreased from 230 to 139 DDDs per 1,000 patient-days
between 2003 and 2019, whereas total antibiotic utilization increased from 1296 to
1550 DDDs per 1,000 patient-days. Infection Control interventions included empir-
ically placing patients with diarrhea on precautions, intensied cleaning measures,
formal antibiotic stewardship, introduction of a real-time PCR C.dicile test in June
2010, and a move to a facility with only single rooms at the RVH in April 2015.
Incidence of HA-CDI at the RVH and MGH and antibiotic utilization between
2003 and2019
Conclusion. An important change in HA-CDI epidemiology was observed in two
Canadian tertiary care hospitals based in Montréal between 2003 and 2019. ere was
a signicant decrease in incidence of HA-CDI and a genotype shi from a predomin-
ance of NAP1 strains to non-NAP1 strains. Utilization of uoroquinolones, to which
the NAP1 strain is resistant, concurrently decreased. Infection control interventions
targeting isolation, diagnosis, disinfection, and antibiotic stewardship have contrib-
uted to the major observed reduction in HA-CDI incidence.
Disclosures. Al l Authors: No reported disclosures
749. ANurse-Driven Protocol for Early Detection of Clostridiodes difficile
Infections
ShannonBeckman, RN, BSN1; JonathanChia, DO1; BethanyStibbe, MASCP1;
MonicaRykse, MPH, CHES1; MichaelS.Wa n g, MD2; 1Spectrum Health, Saint Joseph,
Michigan; 2Spectrum Health Lakeland, St Joseph, Michigan
Session: P-36. HAI: C.difficile
Background. Clostridiodes dicile infections (CDI) are a signicant cause of hos-
pital acquired infections, resulting in signicant morbidity and mortality. Early detec-
tion of CDI has been shown to reduce the spread of CDI within the hospital. As nurses
are frequently at the patients bedside, we proposed to empower the nursing sta to
assess, collect stool samples, and order C.dicile testing.
Methods. Rates of CDI were measured by our Infection Control Department.
Hospital-onset CDI (HO-CDI) was dened as a positive C.dicile PCR assay aer
3days of admission, dened as a stay of at least 3 midnights. Community-onset CDI
(CO-CDI) was dened as any case that was diagnosed in the Emergency Department
or inpatient ward < 3days of hospitalization based on stool testing as above. Nursing
was instructed and empowered to assess, collect stool specimens, and place an order
for C.dicile testing, based on the criteria of 3 loose or watery stools over 24 hours.
Nursing was also educated to not order a test if patients had received stool soeners,
enemas, or laxatives within 24 hours. e protocol was initiated in February 2019.
Results. Rates of HO-CDI increased during the intervention period, rising from 2.6
cases/10000 patient days and peaking at 17.7 cases/10000 patient days (average 6.7 vs.
12.1monthly cases per 10,000 patient days. Rates of CO-CDI did not signicantly change
(12.4 vs. 11.5monthly cases per 10000 patient days). Due to concerns of inappropriate
testing, which included testing aer laxatives, enemas, or sending specimens despite < 3
stools over 24 hours, the protocol was discontinued in June 2019. Although the HO-CDI
rate remained elevated over the next month, the rate subsequently decreased over the next
several months (12.1 vs. 8.0 cases per 10000 patient days). Overall testing also increased
over the study period (148.3 vs. 169.9 cases/per 10000 patientdays).
Figure 1 - Clostridiodes dicile rates
Downloaded from https://academic.oup.com/ofid/article/8/Supplement_1/S471/6450565 by guest on 06 December 2021
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.