The prevalence of ESBL-producing Enterobacteriaceae in a nursing home setting compared with elderly living at home: a cross-sectional comparison

In this cross-sectional comparison study we found no difference in the prevalence
of ESBL-producing Enterobacteriaceae among nursing home residents and the elderly living in their own homes (11 and 8.7 %
respectively). We found that a total of 10 % were ESBL-carriers among all of the elderly
in the study, which was higher than expected.

Our investigation has limitations. We did not perform PFGE, rep-PCR or MLST on isolates
to examine the possibility of spread of the ESBL-producing Enterobacteriaceae between the elderly. However, an analysis of the respective resistance patterns was
performed, and it showed widespread resistance to trimethoprim-sulfametoxazole and
ciprofloxacin, two widely used antibiotics against common infections like UTIs. Another
limitation is that only faecal sample per participant was collected, and an intermittent
secretion of ESBL-producing bacteria could theoretically have been missed 17]. In total 160 persons were screened, a higher number would have increased the statistical
power and perhaps the results could have been different. We don’t think so due to
the fact that the healthy elderly is travelling a lot (in Sweden) and this seems to
be of just as high importance to become rectal carrier of ESBL: s as staying at a
nursing home. However, the risk for a type II error cannot be fully excluded in this
material.

The two major reasons for not participate in the study was due to shortage of staff
and a high ratio of residents suffering from dementia. The shortage of staff was an
argument from some of the nursing homes that declined to participate, but we haven’t
performed a survey over the staffing levels in the actual nursing homes. The size
of the nursing home were not an important factor regarding participation in the study.
Rooney et al. have hypothesized that nursing homes that participate in this kind of
study are the ones that are sufficiently staffed and therefore are better equipped
to observe hygiene routines 17]. In line with this, Andersson et al. reports that lack of time is the main reason
why infection preventing routines are not followed 15], which increases the risk of spreading potential ESBL-producing bacteria via the
staff’s hands or clothes 9], 15]. Only 91 of 434 residents in the nursing homes were screened. The main reasons for
this loss was dementia diagnosis, but also severe illness and immobilization of the
residents. Immobility, comorbidity and severe illness are risk factors for colonization
with ESBL-producing Enterobacteriaceae, and not being able to include the residents with these risk factors might affect
the result 4], 8]. In the group of elderly staying at home, when we finally found them there were only
a few who declined participation so we did not perform any non-response analyzation.

Many nursing home residents per definition have risk factors for ESBL colonization
and living at a nursing home is a risk factor per se 2], 4]. In this study, the ESBL-positive residents at nursing homes had between two and
four established risk factors for colonization. Assessment of risk factors is therefore
an important instrument to find patients who could be at risk of being colonized with
ESBL-producing Enterobacteriaceae.

Nine of the 17 ESBL-positive isolates were resistant to???2 antibiotic classes in
addition to betalactam antibiotics, which may complicate the treatment of common infections.
The most common findings were resistance to ciprofloxacin and trimethoprim-sulfametoxazole,
both antibiotics commonly used against UTIs and one has to consider the choice of
antibiotics against this condition to get full effect and not to drive the resistance.

There was a female dominance in both study populations, and this corresponds partly
with the representation of gender in Sweden among the elderly: in 2013, 60 % of individuals
in the ages between 80 and 89 years 30] were female. The difference in age between the ESBL-positive in the two groups were
20 years (median age 90 and 70.5 years respectively), which in part reflects the difference
in age between the two populations, but it also suggests the co-existence of different
categories of individuals that become colonized with ESBL-producing Enterobacteriaceae, with different risk factors. Brolund et al. has shown that the median age of infected
female patients in Sweden has decreased from 62 to 52 y between 2007 and 2011, which
indicates a higher number of younger, healthier people being colonized with ESBL-producing
Enterobacteriaceae through community-acquisition, which is a shift from earlier clinical picture where
the nosocomial setting with elderly patients with comorbidity dominated 10]. Five of the six elderly living at home that were ESBL-positive had foreign travel
as risk factor, whereof four had travelled outside the Nordic countries. Foreign travel
was the only statistically significant connection to ESBL-carriage (p?=?0.017), and has in Sweden been reported to be a major risk factor for ESBL acquisition,
24].

To our knowledge there is only one earlier study of the prevalence of faecal carriage
of ESBL-producing Enterobacteriaceae in a nursing home setting in Sweden 2008, and this showed a prevalence of 3 % 15]. Other European studies have shown prevalence numbers ranging from 6.2 to 40.5 %
8], 17]. These numbers, along with our result, indicates that nursing homes may act as a
reservoir for ESBL-producing bacteria. These results should be taken in consideration
when treating these groups of patients if they develop severe sepsis or septic shock.