Proton Pump Inhibitors

1. Proton Pump InhibitorsProton Pump Inhibitors
Tom MillerTom Miller

2. PPIsPPIs
Available since late 80’sAvailable since late 80’s
One of the most frequently prescribed One of the most frequently prescribed
classes of drug worldwideclasses of drug worldwide
£425m in England in 2006 (25m Rx)£425m in England in 2006 (25m Rx)
11
OverprescribedOverprescribed
2,32,3

Good efficacyGood efficacy
Safe…Safe…

3. BNFBNF
Common: Common:
–GI, headache, dizziness.GI, headache, dizziness.
Less common: Less common:
–Dry mouth, insomnia, drowsiness, malaise, blurred vision, Dry mouth, insomnia, drowsiness, malaise, blurred vision,
rash and pruritis.rash and pruritis.
Rare: Rare:
–Liver dysfunction, oedema, hypersensitivity reactions, Liver dysfunction, oedema, hypersensitivity reactions,
photosensitivity, photophobia, fever, sweating, depression, photosensitivity, photophobia, fever, sweating, depression,
interstitial nephritis, blood disorders, arthralgia, myalgia, interstitial nephritis, blood disorders, arthralgia, myalgia,
skin reactions.skin reactions.
May increase risk of gastro-intestinal infections May increase risk of gastro-intestinal infections

4. A few extra…A few extra…
C-difficileC-difficile
Hip FracturesHip Fractures
PneumoniaPneumonia
B12 deficiencyB12 deficiency

5. C-difficileC-difficile; The Evidence; The Evidence
Systematic reviewSystematic review
Cohorts and case-controlsCohorts and case-controls
Animal modelsAnimal models

8. Systematic reviewSystematic review
44
Leonard Leonard et alet al. J Am Gastroenterology 2007;102:2047-2056. J Am Gastroenterology 2007;102:2047-2056
Objective – Evaluate associations between Objective – Evaluate associations between
acid suppression and enteric infectionacid suppression and enteric infection
Data Sources – MEDLINE, EMBASE and Data Sources – MEDLINE, EMBASE and
CINAHLCINAHL
Selection – Observational studies including Selection – Observational studies including
cross-sectional, case control and cohort that cross-sectional, case control and cohort that
evaluated risk of enteric infection associated evaluated risk of enteric infection associated
with antisecretory therapy.with antisecretory therapy.

9. Data SynthesisData Synthesis
12 papers, 2948 patients with 12 papers, 2948 patients with C-diffC-diff..
Increased risk of taking antisecretory Increased risk of taking antisecretory
therapy therapy
–Pooled OR 1.94 (95% CI 1.37-2.75) Pooled OR 1.94 (95% CI 1.37-2.75)
Significant heterogeneity p=0.0006Significant heterogeneity p=0.0006
Association greater for PPI vs. H2RAAssociation greater for PPI vs. H2RA
Salmonella and Compylobacter Salmonella and Compylobacter
–OR 2.55 (95% CI 1.53-4.26) again with OR 2.55 (95% CI 1.53-4.26) again with
heterogeneity and greater for PPIsheterogeneity and greater for PPIs

10. Cohort and Case ControlCohort and Case Control
Risk of Risk of Clostridium difficileClostridium difficile diarrhoea diarrhoea
among hospital inpatients prescribed among hospital inpatients prescribed
proton pump inhibitorsproton pump inhibitors
Dial Dial et alet al. CMAJ 2004;171(1):33-8. CMAJ 2004;171(1):33-8

11. CohortCohort
1187 patients from pharmacy database 1187 patients from pharmacy database
receiving abx in an 8 month periodreceiving abx in an 8 month period
2 Medical and one Cardiothoracic sugical 2 Medical and one Cardiothoracic sugical
wardward
Analysis of meds, ward, number and type of Analysis of meds, ward, number and type of
abx and type of acid suppressive therapyabx and type of acid suppressive therapy
Positive toxin assay as recorded with Positive toxin assay as recorded with
infection controlinfection control

12. Case-controlCase-control
To control for ‘sicker patients’ with added risk To control for ‘sicker patients’ with added risk
factors for factors for C-diffC-diff
Separate hospital, same timeSeparate hospital, same time
Cases (94) were consecutive patients with history Cases (94) were consecutive patients with history
of diarrhoea and positive of diarrhoea and positive C-diffC-diff toxin assay (new toxin assay (new
patients).patients).
Controls (94) had abx but no Controls (94) had abx but no C-diff, C-diff, type matched type matched
for age, type and number of abx, Charlston co-for age, type and number of abx, Charlston co-
morbidity indexmorbidity index
Average ages 75.5 and 73.0Average ages 75.5 and 73.0
PPI exposure at least 3 daysPPI exposure at least 3 days

13. ResultsResults
Cohort Cohort
–Those taking PPIs had RR of 2.1 (1.4-Those taking PPIs had RR of 2.1 (1.4-
3.4) of developing 3.4) of developing C-diff C-diff diarrhoeadiarrhoea
–Remained significant after multivariate Remained significant after multivariate
analysis for ward type and abx numbersanalysis for ward type and abx numbers

14. Case-controlCase-control
–Unadjusted OR 3.1 (1.7-5.6), significant Unadjusted OR 3.1 (1.7-5.6), significant
after adjustment for other significant after adjustment for other significant
factorsfactors
–Extended use >6 months OR 6.9 (2.3-Extended use >6 months OR 6.9 (2.3-
20.8)20.8)
–Of 21 pts experiencing 1 or more relapse Of 21 pts experiencing 1 or more relapse
19 (90%) were Rx PPI OR 5.2 (1.1-24.6)19 (90%) were Rx PPI OR 5.2 (1.1-24.6)

15. DiscussionDiscussion
Controlled for number of abx and co-Controlled for number of abx and co-
morbiditiesmorbidities
In cohort greater effect seen with low-In cohort greater effect seen with low-
risk abxrisk abx
Time and dose relevanceTime and dose relevance
Omeprazole and Pantoprazole = class Omeprazole and Pantoprazole = class
effecteffect

16. Case-control 2Case-control 2
Proton pump inhibitor therapy is a risk Proton pump inhibitor therapy is a risk
factor of factor of Clostridium difficileClostridium difficile--
associated diarrhoeaassociated diarrhoea
Yearsley Yearsley et alet al. Aliment Pharmacol . Aliment Pharmacol
Ther 2006;24:613-619.Ther 2006;24:613-619.

17. MethodsMethods
155 patients with CDAD, 153 controls.155 patients with CDAD, 153 controls.
Controls chosen as patients on same Controls chosen as patients on same
ward with closest DOBward with closest DOB
Mean age 79.1 and 78.7Mean age 79.1 and 78.7

18. ResultsResults
PPI: OR 2.03 (1.21-3.41) p=0.004PPI: OR 2.03 (1.21-3.41) p=0.004
Abx and acid suppression OR 1.84 Abx and acid suppression OR 1.84
(1.01-3.36) p=0.046(1.01-3.36) p=0.046
5 Cases PPI alone5 Cases PPI alone

19. AnimalsAnimals
Comparitive role of antibiotics and Comparitive role of antibiotics and
proton pump inhibitor in experimental proton pump inhibitor in experimental
Clostridium difficileClostridium difficile infection in mice infection in mice
Kaur Kaur et alet al. Microbiol. Immunol. . Microbiol. Immunol.
2007;51(12):1209-1214.2007;51(12):1209-1214.

20. Human studies bedevilled by problems of Human studies bedevilled by problems of
sample size, uncertainties of how much and sample size, uncertainties of how much and
how often drugs taken, combination how often drugs taken, combination
therapies and ignorance of exposuretherapies and ignorance of exposure
Morphological findings in mice during c-diff Morphological findings in mice during c-diff
similar to human intestine and successful similar to human intestine and successful
model of c-diff infection used for model of c-diff infection used for
investigation of histological changesinvestigation of histological changes

21. MethodsMethods
4 groups4 groups
–1. Control1. Control
–2. 2. C-difficileC-difficile
–3. Ampicilln and 3. Ampicilln and C-difficileC-difficile
–4. 4. Lansoprazole andLansoprazole and C-difficile C-difficile
Assesed for C-difficile activity, MPO activity and Assesed for C-difficile activity, MPO activity and
histology.histology.
BALB/c

22. ResultsResults
Groups 1, 2 and 3 colonisedGroups 1, 2 and 3 colonised
Caecal contents of group 1 and 2 Caecal contents of group 1 and 2
negative for CDT A + Bnegative for CDT A + B
Antibiotic (c) – 100% A, 83.3% BAntibiotic (c) – 100% A, 83.3% B
PPI (d) – 83.3% A, 100 B PPI (d) – 83.3% A, 100 B

23. Antibiotic and PPI groups significantly Antibiotic and PPI groups significantly
higher colonisation than higher colonisation than C-difficileC-difficile
group (and control)group (and control)
MPO activity significantly higher in MPO activity significantly higher in
antibiotic and PPI group vs. antibiotic and PPI group vs. C-difficile C-difficile
(and control)(and control)

24. Antibiotic and PPI groups showed significant Antibiotic and PPI groups showed significant
differences for epithelial damage, oedema differences for epithelial damage, oedema
and neutrophil infiltrate in the colon against and neutrophil infiltrate in the colon against
controls and controls and C-difficile groupC-difficile group
No significant difference between antibiotic No significant difference between antibiotic
and PPI group for colonisation or histology and PPI group for colonisation or histology
(except for oedema)(except for oedema)

25. DiscussionDiscussion
C-difficleC-difficle group adequately colonised group adequately colonised
but no production of toxin or but no production of toxin or
histological change akin to adult histological change akin to adult
carrierscarriers
PPIs act as an independent factor for PPIs act as an independent factor for
C-difficile infection in experimental C-difficile infection in experimental
animalsanimals

26. SummarySummary
Cohort and case control human studies Cohort and case control human studies
show PPI use to have an increased risk of show PPI use to have an increased risk of
developing developing C-difficileC-difficile diarrhoea in those diarrhoea in those
taking antibiotics and as an independent risk taking antibiotics and as an independent risk
factor, compounded by animal model factor, compounded by animal model
evidence.evidence.
Evidence includes omeprazole, Evidence includes omeprazole,
pantoprazole and lansoprazole pantoprazole and lansoprazole  class class
effect.effect.

27. Food for thought…Food for thought…
Indication for PPIIndication for PPI
Consider reducing or stopping with Consider reducing or stopping with
antibioticsantibiotics
Acid suppressant guidelines??Acid suppressant guidelines??

28. Also in the newsAlso in the news
Use of acid suppressants increase the risk Use of acid suppressants increase the risk
of community acquired of community acquired C-difficileC-difficile by 3 fold by 3 fold
88
, ,
but is not a risk for hospitalisation in but is not a risk for hospitalisation in
community dwelling older patientscommunity dwelling older patients
99
Recent PPI increases the risk of community Recent PPI increases the risk of community
acquired pneumonia, especially in the under acquired pneumonia, especially in the under
40’s40’s
1010
PPIs mayPPIs may
1111
or may not or may not
1212
cause a decline in cause a decline in
B12 status with prolonged useB12 status with prolonged use

29. FinallyFinally
PPIs also to blame for broken hips…PPIs also to blame for broken hips…
In a nested case control study the In a nested case control study the
adjusted odds ratio for hip fracture adjusted odds ratio for hip fracture
associated with PPI use over 1 year associated with PPI use over 1 year
was 1.44 (1.30-1.59), increased in was 1.44 (1.30-1.59), increased in
long term high dosage to AOR 2.65 long term high dosage to AOR 2.65
(1.80-3.90), strength of association (1.80-3.90), strength of association
increasing with duration of therapy.increasing with duration of therapy.

30. ReferencesReferences
1.1. NHS PACT centre pages. Drugs for dyspepsia 2006.NHS PACT centre pages. Drugs for dyspepsia 2006.
2.2. Walker NM, Mc Donald. An evolution of the use of proton pump inhibitors. Pharm World Sci Walker NM, Mc Donald. An evolution of the use of proton pump inhibitors. Pharm World Sci
2001;23:116-7.2001;23:116-7.
3.3. Grube RR May DB. Stress Ulcer prophylaxis in hospitalised patients not in intensive care untis. Am Grube RR May DB. Stress Ulcer prophylaxis in hospitalised patients not in intensive care untis. Am
J Health Syst Pharm 2007;64:1396-400.J Health Syst Pharm 2007;64:1396-400.
4.4. Leonard Leonard et alet al. J Am Gastroenterology 2007;102:2047-2056. J Am Gastroenterology 2007;102:2047-2056
5.5. Dial Dial et alet al. CMAJ 2004;171(1):33-8. CMAJ 2004;171(1):33-8
6.6. Yearsley Yearsley et alet al. Aliment Pharmacol Ther 2006;24:613-619. Aliment Pharmacol Ther 2006;24:613-619
7.7. Kaur Kaur et alet al. Microbiol. Immunol. 2007;51(12):1209-1214. Microbiol. Immunol. 2007;51(12):1209-1214
8.8. Dial Dial et alet al. Use of gastric-acid suppressive agents and the risk of community-acquired . Use of gastric-acid suppressive agents and the risk of community-acquired C-diff-C-diff-
associated disease. JAMA 2005;294:2989-2995.associated disease. JAMA 2005;294:2989-2995.
9.9. PPI and hospitalisation for PPI and hospitalisation for C-diffC-diff-associated disease: A population based study. CID 2006;43:1272--associated disease: A population based study. CID 2006;43:1272-
1276.1276.
10.10. Use of PPIs and the risk of community acquired pneumonia. Gulmez Use of PPIs and the risk of community acquired pneumonia. Gulmez et alet al. Arch Intern Med. . Arch Intern Med.
2007;167:950-955.2007;167:950-955.
11.11. Do acid lowering agents affect vitamin B12 status in older adults? Dharmarajan Do acid lowering agents affect vitamin B12 status in older adults? Dharmarajan et alet al. J AM Med Dir . J AM Med Dir
Assoc. 2008;9(3):162-7.Assoc. 2008;9(3):162-7.
12.12. Long term use of proton pump inhibitors and vitamin B12 status in elderly individuals. Elzen Long term use of proton pump inhibitors and vitamin B12 status in elderly individuals. Elzen et alet al. .
Aliment Pharmacol Ther 2008;10. EPub.Aliment Pharmacol Ther 2008;10. EPub.
13.13. Yang Yang et alet al. Long term PPI therapy and risk of hip fracture. JAMA 2006;296(24):2947-2954.. Long term PPI therapy and risk of hip fracture. JAMA 2006;296(24):2947-2954.