|
|
|
EDITORIAL COMMENTARY
Decreased Effectiveness of Metronidazole for the Treatment of Clostridium difficile Infection?1Department of Medical Microbiology, National Reference Laboratory for Clostridium difficile, Leiden University Medical Center, Leiden, The Netherlands; and 2Department of Medical Microbiology, Leeds Teaching Hospitals and University of Leeds, Leeds, United Kingdom http://www.journals.uchicago.edu/doi/abs/10.1086/588294
Received 25 February 2008; accepted 5 March 2008; electronically published 19 May 2008. |
|
Q&A
244.1
Clostridium difficile infection – are acid suppressant medicines a
risk factor?
Expiry: October
2010
Summary
¨ C.
difficile is a leading
cause of iatrogenic outbreaks of diarrhoea. There are national and local
initiatives to reduce the incidence of C.
difficile-associated disease (CDAD).
¨ Risk factors for CDAD include antibiotic
use and hospitalisation. Patients most at risk are the elderly, particularly if
they have underlying disease or immunosuppression.
¨ The use of proton pump inhibitors (PPIs)
and histamine H2 receptor antagonists (H2RAs), which
suppress gastric acid secretion, have also been suggested to be a risk factor
for the development of CDAD.
¨ A number of observational case-control and
cohort studies have explored this association. The studies have a number of
limitations, the data are conflicting and a causal link has not been
established. However, the evidence suggests that if there is an association, it
is probably stronger for PPIs and for H2RAs.
¨ For this, and other well established
reasons, it is recommended that PPIs should only be used where there is a clear
indication.
Background
Clostridium difficile
infection is a leading cause of iatrogenic outbreaks of diarrhoea. The
consequences of infection range from minor gastrointestinal disturbance to
fatal bowel inflammation. The
infection is caused by an overgrowth of C.
difficile in the colon. C. difficile
can be part of the normal gastrointestinal flora; rates of colonisation vary
from around 3% in asymptomatic healthy adults, to 20% in hospitalised patients
and elderly patients in long term care facilities.1 It is transmitted via spores that are present in
faeces and can survive for long periods in the environment. Local and national
initiatives have been introduced to control the infection.
A number of risk
factors for C. difficile-associated
disease (CDAD) have been identified of which antibiotic use (see Medicines Q&A 242.1)2 and
hospital stay are most strongly associated; others include increasing age,
severe underlying disease, non-surgical gastrointestinal procedures, presence
of a nasogastric tube and increased duration of hospital stay.3 The risk increases if more
than one of these factors is present.
Another reported
risk factor for CDAD is the use of medications that suppress gastric acid
secretion, including proton pump inhibitors (PPIs) and histamine-H2
receptor antagonists (H2RAs).3, 4 The mechanism by which
these agents might increase the risk of CDAD is unclear. It is biologically
plausible that by raising gastric pH, they may increase the burden of some
ingested pathogens, but, as C. difficile spores
have been shown to be acid-resistant in
vitro, an increase in gastric pH would not be expected to influence C. difficile infection rates.5 However, it has been
postulated that in vivo, gastric acid
may act synergistically with salivary nitrates to produce a sporicidal effect,
and that vegetative C. difficile, from
germinated spores, may be a ‘transient’ vector for infection and would be
killed at normal gastric pH.6
This Medicines Q&A summarises the results
of a systematic review and four subsequent studies that have investigated an
association between the use of anti-ulcer medicines and the incidence of CDAD.
Answer
The systematic
review included a number of observational studies evaluating the association
between PPI and H2RA use and CDAD.5 The figures cited in
the review for the number of included studies, the number of patients and the
odds ratios (OR) for the risk of CDAD vary between the abstract, the text and
the summary meta-analysis plots and this raises some doubts about the accuracy
of the review. In essence, it appears that the review included 20 studies (19
papers) of acid suppression medication and CDAD, comprising 17 case-control and
three cohort studies. All but one of the studies investigated hospital
inpatients. The exception was a study using data from the
Of 12 studies (11
papers) that evaluated the association between PPI use and CDAD, 10 studies
found a positive association (in only six was the finding statistically
significant), one a negative association and the findings of one study was
neutral. Across the 12 studies, the odds
ratio for the association with PPI use was 2.05 [1.47 to 2.85]. Of 14 studies
that evaluated H2RA therapy, eight reported a positive association
between H2RA use and CDAD with only three having statistically
significant results. The overall odds ratio for the association was 1.47 [1.06
to 2.05].
There was
significant heterogeneity between the studies included in the systematic
review. Pooling odds ratios in this circumstance is contentious and the summary
statistics must be viewed with caution. Some of the included studies were very
small and may have influenced the overall ORs to a greater extent than
warranted. Despite this and other limitations alluded to below, the authors
suggest that the data provide some evidence of a dose-response relationship,
with the association being stronger for PPIs, which are the more potent acid
suppressants, than for H2RAs.
Several further
studies have been published since data were collected for the systematic review;
four of them are described below. Three studies examined the relationship
between acid suppressants and community-associated CDAD. Two of the four
studies found a positive association between CDAD and PPI use but a less strong
association, if any, with H2RAs. The remaining two studies found no
association with PPI use.
The first study
retrospectively matched 94 inpatients who had developed CDAD during
hospitalisation with matched controls. It found a significant association
between PPI use and CDAD with an OR of 3.6 [1.7 to 8.3], p<0.001).7 Too
few patients were exposed to H2RAs to confirm or reject an
association.
The second study
involved 836 patients hospitalised for community-associated CDAD; each patient
was matched to 10 controls.8 Cases were more likely than controls to
have used a PPI or H2RA in the 45 days prior to hospital admission
(OR 1.5 [1.2 to 1.8]). Risk was greater for PPIs (OR 1.6 [1.3 to 2.0]) than for
H2RAs (OR 1.4 [0.9 to 2.2]).
In a further
study, 1,389 patients hospitalised with CDAD within 60 days of receiving
antibiotics were matched by age and antibiotic use to 12,303 controls who were
not hospitalised with CDAD. PPI use by case patients was categorised as current
(within 90 days) recent (91 to 180 days) or remote (>180days). Cases were no more likely than controls to be
classed as current users of PPIs (OR 0.9 [0.8 to 1.1]).9
In the remaining
study, PPI and H2RA use was assessed in 61 patients with
community-associated C. difficile
infection. Prior exposure to anti-ulcer medicines was not significantly more
common in cases than matched controls but odds ratios were not reported.10
In summary, the
evidence for an association between acid suppressant medication and CDAD is
conflicting. It is derived from observational studies, of which a number were
quite small and several were conducted retrospectively. The limitations of such
studies include confounding and bias and a causal link between acid suppressant
use and CDAD has not been proven. However, given the challenge that C. difficile infection presents, a
possible biologically plausible explanation for an association between acid
suppressant use and CDAD, the, albeit flawed, evidence of an association
particularly for PPI use, and current concerns about overuse of PPIs,11 there is no shortage of good reasons to
critically review PPI prescribing.
National
recommendations for the prevention and management of C. difficile infection advise that PPIs should only be used when
there is a clear clinical indication.12 Although restricting
antimicrobial prescribing and improving hygiene are two key strategies in
preventing CDAD, restricting PPI use may be an additional useful strategy.
Limitations
Available data are
inadequate to establish a causal relationship between anti-ulcer medicines and C. difficile infection.
Disclaimer
· Medicines Q&As are intended for
healthcare professionals and reflect
· Each Medicines Q&A relates only to the
clinical scenario described.
· Medicines Q&As are believed to
accurately reflect the medical literature at the time of writing.
· See www.nelm.nhs.uk
for full disclaimer.
References
1. Job ML
and Jacobs NF. Drug-induced Clostridium
difficile-associated disease. Drug Safety 1997; 17(1): 37-46.
2. UK Medicines information (UKMi). Medicines Q&A
242.1: Clostridium difficile
infection – which antimicrobials are implicated? October 2008. Available online
at:
Quality Assurance
Prepared
by
Karoline
Date Prepared
October 2008
Checked by
Christine
Date of check
October 2008
Search strategy
· MIDatabank [clostridium difficile].
· Embase 1980-date [Clostridium difficile (MeSH,
major) + Antiulcer agent (MeSH, exp, major).
· Medline 1950-date [Clostridium infection
(MeSH, major) + Anti-ulcer agents (MeSH, exp, major).
· Health Protection Agency, www.hpa.org.uk [clostridium].
· Department of Health www.dh.gov.uk [clostridium].
· National electronic Library for Medicines
[clostridium].
Download: QA244_C_difficile_-_are_acid_suppressant_medicines_a_risk_factor.doc |