Is over-use of proton pump inhibitors fuelling the current epidemic of Clostridium difficile-associated diarrhoea?

doi:10.1016/j.jhin.2008.04.023

Journal of Hospital Infection

Volume 70, Issue 1, September 2008, Pages 1-6

https://doi.org/10.1016/j.jhin.2008.04.023 Get rights and content

Summary

Many developed countries have seen an increase in cases of Clostridium difficile-associated diarrhoea (CDAD) in recent years. This has occurred despite heightened awareness of the risks of broad-spectrum antibiotics, overall reduction in antibiotic use and increased focus on hospital hygiene. Some of the increase is due to the introduction of new hypervirulent strains, but it predates the description of these. The epidemic coincides with increased use of proton pump inhibitors (PPIs), much of which is inappropriate according to UK and other national guidelines. Gastric acid is a key host defence against other gastrointestinal infections and epidemiological and animal studies have demonstrated a positive association between incident CDAD and PPI use. An association with recurrence of CDAD after initially successful treatment has also been found. Vegetative C. difficile cells are rapidly killed at normal gastric pH, but survive at the pH found in patients taking PPI. It has recently been shown that vegetative organisms survive long enough on moist surfaces for transmission between patients to occur. We conclude that restricting PPI use to patients with an appropriate indication would reduce unnecessary expenditure on these agents, and might be an additional means of controlling the current epidemic of CDAD.

Introduction

The UK, North America and many European countries are undergoing an epidemic of Clostridium difficile-associated diarrhoea (CDAD). This has occurred despite improvement in hospital hygiene and better control of antibiotic prescribing.

Improved case ascertainment, hypervirulent strains and an ageing population have contributed to this, but may not fully explain the considerable increase in cases seen in recent years.

The epidemic coincides with a major increase in use of proton pump inhibitors (PPIs). Inappropriate use, defined by the National Institute for Health and Clinical Excellence (NICE) guidelines in 2000, has been found in 67% of UK National Health Service (NHS) inpatients.1, 2 Similar results based on their own national guidance have been found in Australia, Ireland and the USA.³ Gastric acid is a key immune defence against gastrointestinal infection, and a possible association with CDAD was first described in 2003.⁴ Since then, numerous studies of hospital and community cases have found an association with PPI use; these have recently been systematically reviewed, with a pooled odds ratio of 1.94 (1.37–2.75).⁵ General practioner and hospital prescribers of PPIs may be unaware of this possible adverse effect and, as many prescriptions are unnecessary, an opportunity exists to minimise this risk factor. This could add to conventional control measures and potentially reduce CDAD-associated morbidity and mortality.

Section snippets

Methods

We conducted a literature search of Medline, PubMed and Google Scholar in 2007, using the search terms ‘Clostridium difficile’, ‘proton pump inhibitor’, ‘gastric acid’. Relevant references were retrieved and reference lists searched manually. The search was repeated in February 2008.

Epidemiology

Since the early 2000s, CDAD has increased in many developed countries, with well-documented outbreaks in Canada, the USA and many European countries.6, 7, 8 In the UK, a voluntary reporting system between 1990 and 2004, and mandatory reporting of all cases aged >65 years since 2004, has demonstrated a clear increase in notified cases (Figure 1). A report from the Office of National Statistics showed annual increases in mention of C. difficile in death certificates between 2000 and 2004.⁹

In the

Association between PPI use and CDAD

An association between hypochlorhydria and CDAD was first postulated in 1982, long before PPIs were developed.¹² The association is biologically plausible, is consistent with the increased risk of post-pyloric compared with gastric enteral feeding, and may be part of the explanation for increased incidence in the elderly. A retrospective study of hospital inpatients showed a positive association with PPI use in 2003, with an odds ratio (OR) of 2.5 (95% CI: 1.5–4.2).⁴ Since then numerous studies

Why is this association unrecognised?

Despite general acceptance that PPIs are a risk factor for gastrointestinal infection with pathogens such as salmonella and campylobacter, the risk of CDAD has had no discernible impact on prescribing patterns. This may be due to two widely held but over-simplified beliefs, first that C. difficile is acid resistant, and second that only spores are relevant to transmission. It is certainly the case that spores are acid resistant when incubated in vitro at normal gastric pH; in vivo, however,

Should PPI prescribing be modified on the strength of current evidence?

PPIs are one of the most widely prescribed groups of drugs in the UK. They are regarded as being extremely safe, are highly effective at relieving dyspepsia and are heavily promoted by the pharmaceutical industry. Consequently, their use has more than doubled in the last five years. NICE first published guidelines on their use in 2000, and this was replaced with a guide to management of dyspepsia in primary care in 2005.³⁰ These guidelines give clear indications for appropriate PPI use in

Conclusion: prescribers of PPIs should follow NICE guidelines

Even if the association between PPI use and CDAD is coincidental, following NICE prescribing guidance would save the NHS at least £100 million each year.³ If the association is causal, thousands of cases and hundreds of unnecessary deaths might be avoided in the future. Until the potential association between PPIs and CDAD is confirmed or refuted, prescribers should follow NICE guidance, and avoid unnecessary use of these potent and potentially dangerous drugs.

References (31)

R. Cunningham et al.
Proton pump inhibitors as a risk factor for Clostridium difficile diarrhoea
J Hosp Infect
(2003)
E.J. Kuijper et al.
Emergence of Clostridium difficile-associated disease in North America and Europe
Clin Microbiol Infect
(2006)
L. Gurian et al.
Influence of gastrointestinal secretions on Clostridium difficile infection
Gastroenterology
(1982)
NICE
Guidance on the use of proton pump inhibitors (PPI) in the treatment of dyspepsia
(July 2000)
N.M. Walker et al.
An evaluation of the use of proton pump inhibitors
Pharm World Sci
(2001)
I. Forgacs et al.
Overprescribing proton pump inhibitors
Br Med J
(2008)
J. Leonard et al.
Systematic review of the risk of enteric infection in patients taking acid suppression
Am J Gastroenterol
(2007)
J. Pépin et al.
Clostridium difficile-associated diarrhea in a region of Quebec from 1991 to 2003: a changing pattern of disease severity
Can Med Assoc J
(2004)
D.B. Blossom et al.
The challenges posed by reemerging Clostridium difficile infection
Clin Infect Dis
(2007)
Office for National Statistics
Deaths involving Clostridium difficile: England and Wales, 1999–2004
Health Stat Q
(2006)

D. Reeves

The 2005 Garrod lecture: the changing access of patients to antibiotics – for better or worse?

J Antimicrob Chemother

(2007)

B. Cookson

Hypervirulent strains of Clostridium difficile

Postgrad Med J

(2007)

E. Dubberke et al.

Clostridium difficile associated disease in a setting of endemicity: identification of novel risk factors

Clin Infect Dis

(2007)

S. Dial et al.

Risk of Clostridium difficile diarrhea among hospital inpatients prescribed proton pump inhibitors: cohort and case-control studies

Can Med Assoc J

(2004)

A. Akhtar et al.

Increasing incidence of Clostridium difficile-associated diarrhea in African-American and Hispanic patients: association with the use of proton pump inhibitor therapy

J Natl Med Assoc

(2007)

Cited by (49)

Proton Pump Inhibitors in cancer patients: How useful they are? A review of the most common indications for their use
2017, Critical Reviews in Oncology/Hematology
Proton-Pump Inhibitors (PPIs) are commonly prescribed in the general population and in cancer patients. A supposed role in the prevention of gastric mucosal damage apparently justify their use in patients undergoing cytotoxic chemotherapy, steroids and radiotherapy on the gastro-duodenal region. They are frequently given also to patients admitted to Intensive Care Units, for the prevention of stress-related gastric ulcers. The evidence about these use of gastroprotection is reviewed. In the majority of the cases the prescription of PPIs is not justified. In two circumstances (chemotherapy and stress-related gastric disease) randomized studies have shown a protective action of PPIs although this effect did not translate into the reduction of serious clinical consequences. PPIs are not free of toxic effects that are acknowledged by an expanding literature. Also the interaction with anticancer drugs is a potential source of unwanted consequences.
Predictors of clinical events occurring during hospital stay among elderly patients admitted to medical wards in Italy
2016, European Journal of Internal Medicine
Clinical events occurring during hospital stay are independent predictors of prolonged hospitalization, in-hospital mortality and readmission among elderly patients admitted to medical wards.
To identify predictors of intercurrent clinical events (ICE) during hospital stay among the main demographic, functional and clinical characteristics assessed at hospital admission in a multicenter sample of elderly inpatients in Italy.
This observational prospective cohort study was conducted in 66 internal and geriatric medicine hospital wards in 2010. It enrolled 1267 inpatients aged 65 years or older living at home before hospitalization. Multivariable Poisson regression analyses were employed to identify the most common ICEs as well as their independent predictors.
During the hospital stay 427 patients (33.7%) experienced at least one ICE. The most common ICEs were urinary tract infections, pneumonia, anemia, arrhythmias and fluid electrolyte disorders. After correction for age, sex, comorbidity, cognitive impairment and functional dependence, independent predictors of any ICE were: being a bladder catheter holder (RR [risk ratio] 1.86, 95% CI 1.52–2.27), being on treatment at home with a proton pump inhibitor (PPI) (RR 1.25, 95% CI 1.03–1.53), with immunosuppressant therapy (RR 2.10, 95% CI 1.24–3.56), and body temperature at admission (RR 1.19, 95% CI 1.06–1.33).
Four clinical characteristics, easily assessable at admission, may be useful to identify elderly inpatients at a higher risk for developing ICEs during hospital stay. Furthermore three of these predictors are modifiable factors, thus interventions reducing the use of catheter, PPI and immunosuppressants may result in reduction of ICEs.
Clostridium difficile infection after cardiac surgery: Prevalence, morbidity, mortality, and resource utilization
2014, Journal of Thoracic and Cardiovascular Surgery
Citation Excerpt :
More than half were on a proton-pump inhibitor, which is linked to increased occurrence of CDI.27-29 Increased gastric pH allows survival of CDI spores.28,29 Other known risk factors illustrated by our study included older age1,2,8,21,22,26 and previous colonoscopy or gastrointestinal surgery1,21,28; some patients were immunocompromised.1,9,28
Despite increasing efforts to prevent infection, the prevalence of hospital-associated Clostridium difficile infections (CDI) is increasing. Heightened awareness prompted this study of the prevalence and morbidity associated with CDI after cardiac surgery.
A total of 22,952 patients underwent cardiac surgery at Cleveland Clinic from January 2005 to January 2011. CDI was diagnosed by enzyme immunoassay for toxins and, more recently, polymerase chain reaction (PCR) testing. Hospital outcomes and long-term survival were compared with those of the remaining population in propensity-matched groups.
One hundred forty-five patients (0.63%) tested positive for CDI at a median of 9 days postoperatively, 135 by enzyme immunoassay and 11 by PCR. Its prevalence more than doubled over the study period. Seventy-seven patients (48%) were transfers from outside hospitals. Seventy-three patients (50%) were exposed preoperatively to antibiotics and 79 (56%) to proton-pump inhibitors. Patients with CDI had more baseline comorbidities, more reoperations, and received more blood products than patients who did not have CDI. Presenting symptoms included diarrhea (107; 75%), distended abdomen (48; 34%), and abdominal pain (27; 19%). All were treated with metronidazole or vancomycin. Sixteen patients (11%) died in hospital, including 5 of 10 who developed toxic colitis; 3 of 4 undergoing total colectomy survived. Among matched patients, those with CDI had more septicemia (P < .0001), renal failure (P = .0002), reoperations (P < .0001), prolonged postoperative ventilation (P < .0001), longer hospital stay (P < .0001), and lower 3-year survival, 52% versus 64% (P = .03), than patients who did not have CDI.
Although rare, the prevalence of CDI is increasing, contributing importantly to morbidity and mortality after cardiac surgery. If toxic colitis develops, mortality is high, but colectomy may be lifesaving.
Infections in the Elderly
2013, Critical Care Clinics
Citation Excerpt :
Antibiotic therapy within 6 weeks before CDAD is a strong risk factor.115 In combination with proton pump inhibitors, the risk is even higher,116 leading to a 2.5 to 3.5 higher CDAD-related mortality in elderly patients during severe infections.117,118 Clindamycin, fluoroquinolones, and cephalosporins are associated with the highest risk, although virtually all antibacterials have the potential to induce CDAD.112–114
Emergent adverse effects of proton pump inhibitors
2013, Presse Medicale
Recent literature reports of potential adverse effects (AEs) of proton pump inhibitors (PPIs), especially during long-term treatments.
To present a literature review of major AEs: digestive infections, pneumonia, bone fracture, hypomagnesemia, interstitial nephritis, gastric cancer and neutropenia.
The authors used Pubmed; articles in English or French, published between August 2006 and August 2011 were analyzed.
Two reviewers analyzed the references of title and summary to retain mainly observational studies, controlled clinical trials, meta-analyzes, case reports.
For digestive infections: observational studies have shown a link moderate to high (OR 1.4 to 8.3) with exposure to PPIs. For pneumonia: some case-control studies reported a modest significative risk (OR 1.2 to 1.6), some not. The risk appears dose dependent and greater in subjects at risk. For fractures: the majority of observational studies report a significative increase in low to moderate risk (OR 1.2 to 3.1), correlated with the dose and duration of treatment. For magnesium deficiency: rare but potentially severe, they are described in case reports. Interstitial nephritis are described in case reports and for different PPIs, suggesting a class effect. For the stomach neoplasm: if three observational studies show an increased cancer risk (OR 1.5 to 2, 3), confounding factors make the causal link uncertain. Neutropenia is reported in a clinical observation, a class effect is suggested.
One can regret the absence of controlled clinical trials; indeed the observational studies have the interest to move closer to “real life”, but often have methodological bias.
Although AEs PPIs do not call into question the usefulness of this drug class, they show the need to limit their prescribing to indications for which efficacy has been proven. Moreover, PPIs treatment must be regularly reassessed to avoid exposing patients to unnecessary risks.
La littérature récente rapporte des effets indésirables (EI) émergents des inhibiteurs de la pompe à protons (IPP), en particulier lors des traitements au long cours.
Présenter une analyse bibliographique des principaux EI émergents des IPP: infections digestives, pneumopathie, fracture osseuse, hypomagnésémie, néphrite interstitielle, cancer gastrique et neutropénie.
Les auteurs ont utilisé Pubmed ; les articles en langue anglaise ou française, publiés entre août 2006 et août 2011 ont été analysés.
Deux relecteurs ont analysé les références sur titre et résumé pour retenir principalement les études observationnelles, essais cliniques contrôlés, méta-analyses, observations cliniques.
Pour les infections digestives : des études observationnelles ont montré un lien significatif allant de modéré à fort (OR de 1,4 à 8,3) avec l’exposition aux IPP. Pour les pneumopathies : des études cas-témoins ont rapporté un lien modeste (OR de 1,2 à 1,6), d’autres non. Le risque semble dose-dépendant et plus important chez les sujets à risque. Pour les fractures : la majorité des études observationnelles rapportent une augmentation faible à modérée du risque (OR de 1,2 à 3,1), corrélée à la dose et à la durée de traitement. Pour les hypomagnésémies : rares mais potentiellement sévères, elles sont décrites dans des observations cliniques. Les néphrites interstitielles sont décrites dans des observations cliniques et pour différents IPP, suggérant un effet classe. Pour les cancers gastriques, si 3 études observationnelles montrent une augmentation du risque de cancer (OR de 1,5 à 2,3), des facteurs confondants rendent le lien de causalité incertain. La neutropénie est rapportée dans une observation clinique, un effet de classe est suggéré.
On peut regretter l’absence d’essais cliniques contrôlés ; en effet si les études observationnelles ont l’intérêt de se rapprocher plus de « la vraie vie », elles ont souvent des biais méthodologiques.
Même si les EI émergents des IPP ne remettent pas en cause l’utilité de cette classe de médicaments, ils engagent à limiter leur prescription aux indications pour lesquelles leur efficacité a été prouvée. De plus, les traitements doivent être régulièrement réévalués afin de ne pas exposer les patients à des risques inutiles.
Risk factors for Clostridium difficile infections in hospitalized patients
2011, Medicina Clinica
Identificar los factores de riesgo y estimar los efectos brutos de la infección por Clostridium difficile (ICD) adquirida en el hospital.
Estudio de casos y controles apareados por edad, sexo y fecha de ingreso. Se evaluaron factores de riesgo del paciente y de la asistencia. Se compararon las estancias hospitalarias y la mortalidad.
Se incluyeron 38 casos y 76 controles (edad media de 73 años). Los casos presentaban un peor índice de Charlson (p = 0,02), una estancia preinfección superior (mediana 10 frente a 5,5 días) y habían recibido tratamiento antibiótico previo con mayor frecuencia (89,5 frente a 40,7%) que sus controles. La albuminemia < 3,5 g/dL (odds ratio [OR] 7,1; intervalo de confianza del 95% [IC 95%] 1,4-37,0), y haber recibido cefalosporinas (OR 10,1; IC 95% 1,8-55,1), quinolonas (OR 9,4; IC 95% 1,1-41,1), o inhibidores de la bomba de protones (OR 6,6; IC 95% 1,1-41,1) se asociaron independientemente a mayor riesgo de ICD. Tanto la estancia hospitalaria total (31,5 frente a 5,5 días) como la mortalidad hospitalaria (31,6 frente a 6,6%) fueron superiores en los casos que en los controles. Los aislados de C. difficile correspondieron al toxinotipo V (PFGE NAP 8) y al 0.
El uso de inhibidores de la bomba de protones, cefalosporinas y quinolonas, y la hiponutrición aumentan el riesgo de ICD; esta se asocia a importantes efectos brutos de mortalidad y exceso de estancia.
To identify risk factors, and to estimate the crude effects attributable to hospital acquired Clostridium difficile infection (CDI).
Case-control study matched by age, gender, and admission date. Patient and healthcare risk factors were evaluated. Hospital stays and mortality were compared.
Thirty-eight cases and 76 controls were included (mean age 73 years). Cases presented worse Charlson index (P .02), higher pre-infection stay (median 10 vs. 5.5 days) and had received antibiotic treatment more frequently (89.5 vs. 40.7%) than their control counterparts. Albuminemia < 3.5 gr/dL (OR 7.1; 1.4-37), having received cephalosporins (OR 10.1; 1.8-55.1), quinolones (OR 9.4; 1.1-41.1), or proton pump inhibitors (OR 6.6; 1.1-41.1) were associated with an independent higher risk of CDI. Total hospital stay (31 vs. 5.5 days), as well as crude mortality, was higher for cases than for control patients (31.6 vs. 6.6%).
Receiving cephalosporins, quinolones and proton pump inhibitors, as well as hyponutrition, increase the risk of CDI. CDI is associated with relevant crude effects on mortality and excess of stay.

View all citing articles on Scopus

View full text

ReviewIs over-use of proton pump inhibitors fuelling the current epidemic of Clostridium difficile-associated diarrhoea?

Summary

Introduction

Section snippets

Methods

Epidemiology

Association between PPI use and CDAD

Why is this association unrecognised?

Should PPI prescribing be modified on the strength of current evidence?

Conclusion: prescribers of PPIs should follow NICE guidelines

J Hosp Infect

Clin Microbiol Infect

Gastroenterology

Guidance on the use of proton pump inhibitors (PPI) in the treatment of dyspepsia

An evaluation of the use of proton pump inhibitors

Pharm World Sci

Overprescribing proton pump inhibitors

Br Med J

Systematic review of the risk of enteric infection in patients taking acid suppression

Am J Gastroenterol

Clostridium difficile-associated diarrhea in a region of Quebec from 1991 to 2003: a changing pattern of disease severity

Can Med Assoc J

The challenges posed by reemerging Clostridium difficile infection

Clin Infect Dis

Deaths involving Clostridium difficile: England and Wales, 1999–2004

Health Stat Q

The 2005 Garrod lecture: the changing access of patients to antibiotics – for better or worse?

J Antimicrob Chemother

Hypervirulent strains of Clostridium difficile

Postgrad Med J

Clostridium difficile associated disease in a setting of endemicity: identification of novel risk factors

Clin Infect Dis

Risk of Clostridium difficile diarrhea among hospital inpatients prescribed proton pump inhibitors: cohort and case-control studies

Can Med Assoc J

Increasing incidence of Clostridium difficile-associated diarrhea in African-American and Hispanic patients: association with the use of proton pump inhibitor therapy

J Natl Med Assoc

Review
Is over-use of proton pump inhibitors fuelling the current epidemic of Clostridium difficile-associated diarrhoea?