Documente Academic
Documente Profesional
Documente Cultură
net/publication/330781139
CITATIONS READS
0 39
9 authors, including:
Chien-Chang Lee
National Taiwan University Hospital
206 PUBLICATIONS 3,175 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Wan-Ting Hsu on 16 March 2019.
Chih-Hung Wang, Cheng-Han Li, Ronan Hsieh, Cheng-Yi Fan, Tze-Chun Hsu,
Wei-Che Chang, Wan-Ting Hsu, Yu-Ya Lin & Chien-Chang Lee
To cite this article: Chih-Hung Wang, Cheng-Han Li, Ronan Hsieh, Cheng-Yi Fan, Tze-
Chun Hsu, Wei-Che Chang, Wan-Ting Hsu, Yu-Ya Lin & Chien-Chang Lee (2019): Proton
pump inhibitors therapy and the risk of pneumonia: a systematic review and meta‐analysis of
randomized controlled trials and observational studies, Expert Opinion on Drug Safety, DOI:
10.1080/14740338.2019.1577820
Article views: 43
ORIGINAL RESEARCH
Proton pump inhibitors therapy and the risk of pneumonia: a systematic review and
meta-analysis of randomized controlled trials and observational studies
Chih-Hung Wang*a,b, Cheng-Han Li*c, Ronan Hsiehd, Cheng-Yi Fanc, Tze-Chun Hsua, Wei-Che Changc, Wan-Ting Hsue,
Yu-Ya Linf and Chien-Chang Lee a,b,e
a
Department of Emergency Medicine, National Taiwan University Hospital, Taipei, Taiwan; bDepartment of Emergency Medicine, College of
Medicine, National Taiwan University, Taipei, Taiwan; cCollege of Medicine, National Taiwan University, Taipei, Taiwan; dDepartment of Medicine,
Albert Einstein Medical Center, Philadelphia, PA, USA; eDepartment of Epidemiology, Harvard School of Public Health, Boston, MA, USA;
f
Department of Pharmacy, E-Da hospital, Kaohsiung, Taiwan
CONTACT Chien-Chang Lee, cclee100@gmail.com Health Data Science Research Group, Department of Emergency Medicine, National Taiwan University
Hospital, No. 7, Chung-Shan South Road, Taipei 100, Taiwan
*
These authors contributed equally to this work.
Supplemental data for this article can be accessed here.
© 2019 Informa UK Limited, trading as Taylor & Francis Group
2 C.-H. WANG ET AL.
therapy and the risk of pneumonia. Subsequently, we further analyses; (4) Pneumonia validation: validating the diagnosis
performed sensitivity and subgroup analyses in an attempt to of pneumonia through radiographic or laboratory results; (5)
identify a potential source of heterogeneity in the interpreta- Different time periods: reducing the likelihood of exposure
tion of the meta-analysis results. misclassifications by categorizing drug-filling records into dif-
ferent time periods; (6) Sensitivity analysis: conducting sensi-
tivity analyses that took potential sources of bias into account.
2. Methods We also used Scottish Intercollegiate Guideline Network (SIGN)
2.1. Data sources and search strategy tool to evaluate each included study.
Figure 2. Summary forest plot of overall risk of pneumonia associated with proton pump inhibitor use, subdivided by study design.
EXPERT OPINION ON DRUG SAFETY 5
pneumonia (OR, 1.57; 95% CI, 1.18–2.09; I2, 46.1%). Moreover, studies and the presence of publication bias, this conclusion
there was an increased risk of pneumonia associated with pan- should be interpreted with caution. Through sensitivity and
toprazole or lansoprazole therapy (pantoprazole: OR, 1.82; 95% subgroup analyses, several possible sources of heterogeneity
CI, 1.35–2.45; I2, 1.6%; lansoprazole: OR, 1.17; 95% CI, 1.00–1.37; were identified, including patient population, study design,
I2, 0.0%) and in patients receiving PPIs for less than 7 or 30 days and interval between the initiation of PPIs and diagnosis of
(7 days: OR, 3.50; 95% CI, 2.86–4.28; I2, 0.0%; 30 days: OR, 2.49; pneumonia, suggesting the possible presence of protopathic
95% CI, 2.10–2.94; I2, 0.0%). bias in the meta-analyses of observational studies.
The funnel plot that included all studies (n = 58) in the
primary meta-analysis demonstrated clustering of studies in
favor of a positive association (Figure 3(a)). The Egger’s test 4.2. Comparisons with previous studies
confirmed asymmetry of the funnel plot (p-value, 0.01), indi- Previous meta-analyses [4–6] have indicated the significant asso-
cating the presence of publication bias. When publication bias ciation between PPI use and an increased risk of pneumonia. It
was examined according to the study design, only the funnel has also been noted that the inter-study heterogeneity is high;
plot that included observational studies was significantly however, limited by the number of included studies, the poten-
asymmetrical (p-value, 0.01; Figure 3(b)), while the funnel tial sources of heterogeneity could not be identified. These
plot that included RCTs was not (p-value, 0.30; Figure 3(c)). studies [4–6] demonstrated that newly prescribed PPI users
were associated with a higher risk of pneumonia in comparison
with chronic users. Similarly, our results indicate that even
4. Discussion a period as short as 7 days of PPI use was significantly associated
with an increased risk of pneumonia. Moreover, the interval
4.1. Main findings
between the initiation of PPIs and diagnosis of pneumonia may
The systematic review identified 65 studies examining the be a possible cause of significant heterogeneity.
relationship between PPI use and risk of pneumonia. The At intragastric pH values less than 4.0, 99.9% of the bacteria
primary meta-analysis of 58 studies, involving 7,643,982 are killed within 30 minutes [77]. It is hypothesized that PPIs
patients, indicates that PPI use may be associated with an may predispose individuals to pneumonia due to the fact that
increased risk of pneumonia (OR, 1.43; 95% CI, 1.30–1.57; I2, elevated intragastric pH may contribute to bacterial over-
95.4%). However, due to the great heterogeneity between growth in the stomach [78], which subsequently increases
6 C.-H. WANG ET AL.
(a)
(b)
(c)
Figure 3. (a) Funnel plot of all studies included in the primary meta-analysis. (b) Funnel plot of observational studies. (c) Funnel plot of randomized controlled trials.
EXPERT OPINION ON DRUG SAFETY 7
the risk of micro-aspiration of bacteria [79]. However, it has However, in comparison with observational studies, the
been reported that following initiation of 30 mg day−1 lanso- number of included patients or follow-up duration in RCTs
prazole, the median 24-hour intragastric pH value on days 1, 2 may be lower or shorter than those in observational studies.
and 7 remained slightly below 4 [80]. Therefore, it is not Additionally, in most included RCTs [15–20,22], the develop-
biologically plausible that pneumonia could occur in such ment of pneumonia was not the primary outcome and was
a short time frame. not clearly defined. Therefore, observational studies using
Furthermore, tachyphylaxis has not been reported in PPI healthcare databases could still provide valuable data for
use [81]; therefore, the suppression of gastric acid production researchers and regulatory agencies [90] to investigate the
and the elevation of intragastric pH should be more pro- safety of drugs, as long as the relevant bias could be well
nounced in patients receiving long-term PPIs. Nevertheless, controlled.
as shown in previous studies [4–6] and ours, long-term PPI Subgroup analyses found that the heterogeneity may be
use was not or much less significantly associated with an decreased if the studies were pooled according to different
increased risk of pneumonia in comparison with short-term types of PPIs. Nevertheless, the potency in suppressing gastric
use. Therefore, these results raise concerns about protopathic acid secretion has not been shown to be significantly different
bias in investigating the association between PPI use and risk among different PPIs [91,92]. Therefore, whether different
of pneumonia among observational studies. types of PPIs could lead to different risks of pneumonia should
be examined further.
4.3. Interpretation of sensitivity and subgroup analyses
Protopathic bias, or reverse causality, is a source of systematic
4.4. Study limitations
bias that occurs when exposure status may change in response
to the manifestation of a latent outcome. PPIs may be prescribed Important adverse effects may occur rarely, and accordingly,
for the early symptoms of pneumonia, such as non-specific chest systematic reviews and meta-analyses that pool harms data
discomfort, and when pneumonia is later diagnosed, PPIs may from various sources can provide greater insight than a single
fallaciously appear to cause pneumonia. Protopathic bias has RCT or observational study. However, publication bias could
been noted in research investigating antibiotics and antibiotic- raise serious problems. The main study limitation of our meta-
resistant infections [82], alcohol consumption and systemic lupus analysis results was the presence of significant publication
[83], and NSAIDs and cancer [84]. Moreover, protopathic bias has bias, which seemed to be caused by the included observa-
been noted in research regarding PPIs and gastric cancer [85] or tional studies. Meyer et al. [93], found evidence of publication
ischemic heart disease [86]. bias in gastroenterological research that abstracts with signifi-
To deal with protopathic bias, a lag-time approach may be cant findings were more likely to be published. More than half
a useful strategy in observational studies [85,87]. Alternatively, of the abstracts presented from 2011 to 2013 in Americas
restricting analysis to NSAID users may help to minimize pro- Hepato-Pancreato-Biliary Association Congresses failed to
topathic bias in the study of PPIs. For NSAID users, PPIs were reach publication [93]. Nevertheless, we searched relevant
prescribed to prevent ulcers or dyspepsia rather than for the conference abstracts and the publication bias remained
early symptoms of pneumonia. As shown in our sensitivity significant.
analyses of NSAID users, the pooled results demonstrate that Publication bias [94] could jeopardize the validity of meta-
PPI use was not statistically associated with an increased risk analyses. Studies [95,96] have demonstrated that once unpub-
of pneumonia. In comparison with the primary meta-analysis, lished data are pooled in the analysis, the conclusion may be
the inter-study heterogeneity was much lower among the very different from the original results. Most studies [94] have
studies of NSAID users. Therefore, by circumventing the pro- focused on the consequences of publication bias concerning
blem induced by protopathic bias, the pooled results of stu- RCTs. In the present study, the most common study design
dies on NSAID users appear to be less biased and more was observational; the effects of publication bias on this type
credible. of study is unclear but may be important, since the incidence
In contrast, RCTs may be less susceptible to protopathic of adverse events may be misestimated and inaccurate asso-
bias, since RCTs adopted an intention-to-treat approach to ciations between clinical variables may be assumed. This may
analyze the data, with PPI use started from the time of rando- subsequently lead to withdrawals, revocations, or suspensions
mization. Moreover, most of the included studies were obser- of marketing authorizations. Post-marketing reporting of
vational in design, which could only establish an association adverse events during real-world use of a medicinal product
rather than a causal relationship between PPI use and risk of is important in identifying new, rare, or serious adverse reac-
pneumonia. One of the indications for prescribing PPIs is tions with the potential to influence the benefit–risk balance
gastroesophageal reflux disease [88], which by itself, may of a product [97,98]. Between 2012 and 2016, meta-analyses
contribute to an increased risk of pneumonia through micro- were cited in 13 of the 18 regulatory actions within the EU
aspiration [89]. Thus, confounding by indication in observa- [99]; therefore, cautious inspection of available evidence con-
tional studies may lead to overestimated effect estimates, cerning the adverse events of a certain medication is
which could only be avoided by RCTs. In our subgroup ana- paramount.
lyses, the pooled results of RCTs demonstrate no statistically Secondly, the results of subgroup analyses were based on
significant association between PPI use and risk of pneumonia univariate analyses; accordingly, the results should be used for
and low heterogeneity. exploratory purpose rather than viewed as definitive conclusions.
8 C.-H. WANG ET AL.
4.5. Applications in clinical practice employment, consultancies, honoraria, stock ownership or options, expert
testimony, grants or patents received or pending, or royalties.
The meta-analytic results of sensitivity and subgroup analyses
suggested that the previously indicated associations [4–6]
between PPI therapy and increased risk of pneumonia may Reviewer disclosures
be biased by the protopathic bias, publication bias and con- Peer reviewers on this manuscript have no relevant financial or other
founding by indication. According to the meta-analytic results relationships to disclose.
of RCTs and observational studies focusing on NSAID users,
PPI use may not be associated with increased risk of pneumo-
nia. Concern for pneumonia should not prevent clinicians from ORCID
initiating indicated PPI therapy if otherwise indicated [100]. Chien-Chang Lee http://orcid.org/0000-0002-1243-2463
5. Conclusion References
Our primary meta-analysis demonstrates that PPI use may be Papers of special note have been highlighted as either of interest (•) or of
associated with an increased risk of pneumonia. Nevertheless, considerable interest (••) to readers.
the presence of significant between-study heterogeneity and 1. Heidelbaugh JJ, Goldberg KL, Inadomi JM. Overutilization of proton
pump inhibitors: a review of cost-effectiveness and risk [corrected].
publication bias raised concerns regarding the validity of the Am J Gastroenterol. 2009;104(Suppl 2):S27–32.
primary meta-analytic result. Protopathic bias, or reverse caus- 2. Laheij, R.J., et al. Risk of community-acquired pneumonia and use
ality, may cause overestimated association. Studies that of gastric acid-suppressive drugs. JAMA. 2004;292(16):1955–1960.
adopted a design to account for protopathic bias did not 3. Sanduleanu, S., et al. Non-Helicobacter pylori bacterial flora during
show a significant association between PPI use and risk of acid-suppressive therapy: differential findings in gastric juice and
gastric mucosa. Aliment Pharmacol Ther. 2001;15(3):379–388.
pneumonia. Therefore, concern for pneumonia should not 4. Lambert, A.A., et al. Risk of community-acquired pneumonia with
prevent clinicians from initiating indicated PPI therapy if other- outpatient proton-pump inhibitor therapy: a systematic review and
wise indicated. meta-analysis. PLoS One. 2015;10(6):e0128004.
• A previous systematic review and meta-analysis investigated
similar the same topic as the current work.
Acknowledgments 5. Eom, C.S., et al. Use of acid-suppressive drugs and risk of pneumo-
nia: a systematic review and meta-analysis. Cmaj. 2011;183
The authors would like to sincerely thank the staff of the Core Labs, the (3):310–319.
Department of Medical Research, National Taiwan University Hospital for • A previous systematic review and meta-analysis investigated
technical assistance. similar the same topic as the current work.
6. Johnstone J, Nerenberg K, Loeb M. Meta-analysis: proton pump
inhibitor use and the risk of community-acquired pneumonia.
Aliment Pharmacol Ther. 2010;31(11):1165–1177.
Author contributions
• A previous systematic review and meta-analysis investigated
Cheng-Han Li and Cheng-Yi Fan involved in the data collection and extrac- similar the same topic as the current work.
tion. Chih-Hung Wang and Cheng-Han Li performed the statistical analysis, 7. Horwitz RI, Feinstein AR. The problem of “protopathic bias” in case-
interpreted the results and wrote the draft. Tze-Chun Hsu and Wan-Ting Hsu control studies. Am J Med. 1980;68(2):255–258.
provided technical support of statistical analyses. Ronan Hsieh reviewed the 8. Filion, K.B., et al. Proton pump inhibitors and the risk of hospitalisa-
manuscript and provided critical insights. Wei-Che Chang and Yu-Ya Lin tion for community-acquired pneumonia: replicated cohort studies
helped with language editing and critically revised the manuscript. Chien- with meta-analysis. Gut. 2014;63(4):552–558.
Chang Lee obtained funding, conceive the idea, designed the study, and 9. Liberati, A., et al. The PRISMA statement for reporting systematic
guided the statistical analysis and manuscript writing, and supervise the reviews and meta-analyses of studies that evaluate health care
whole process. All authors approved the final version of the manuscript and interventions: explanation and elaboration. Ann Intern Med.
agree to be accountable for all aspects of the work. 2009;151(4):W65–94.
10. Stroup, D.F., et al. Meta-analysis of observational studies in epide-
miology: a proposal for reporting. Meta-analysis Of Observational
Data availability statement Studies in Epidemiology (MOOSE) group. JAMA. 2000;283
(15):2008–2012.
The data that support the findings of this study are available from the 11. Anglemyer A, Horvath HT, Bero L. Healthcare outcomes assessed
corresponding author, upon reasonable request. with observational study designs compared with those assessed in
randomized trials. Cochrane Database Syst Rev. 2014;4:MR000034.
12. Viechtbauer W. Bias and efficiency of meta-analytic variance esti-
Funding mators in the random-effects model. J Educ Behav Stat. 2005;30
(3):261–293.
This manuscript was supported by the Taiwan National Ministry of Science 13. DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin
and Technology (MOST 104-2314-B-002-039-MY3) and National Taiwan Trials. 1986;7(3):177–188.
University Hospital (108-S4091). 14. Kantorova, I., et al. Stress ulcer prophylaxis in critically ill patients:
a randomized controlled trial. Hepatogastroenterology. 2004;51
(57):757–761.
15. American Lung Association Asthma Clinical Research, C, et al.
Declaration of interest
Efficacy of esomeprazole for treatment of poorly controlled
The authors have no relevant affiliations or financial involvement with any asthma. N Engl J Med. 2009;360(15):1487–1499.
organization or entity with a financial interest in or financial conflict with 16. Scheiman, J.M., et al. Prevention of peptic ulcers with esomepra-
the subject matter or materials discussed in the manuscript. This includes zole in patients at risk of ulcer development treated with low-dose
EXPERT OPINION ON DRUG SAFETY 9
acetylsalicylic acid: a randomised, controlled trial (OBERON). Heart. 37. Gau, J.T., et al. Pharmacotherapy and the risk for
2011;97(10):797–802. community-acquired pneumonia. BMC Geriatr. 2010;10:45.
17. Sugano, K., et al. Lansoprazole for secondary prevention of gastric 38. Redelmeier, D.A., et al. Postoperative pneumonia in elderly patients
or duodenal ulcers associated with long-term low-dose aspirin receiving acid suppressants: a retrospective cohort analysis. BMJ.
therapy: results of a prospective, multicenter, double-blind, rando- 2010;340:c2608.
mized, double-dummy, active-controlled trial. J Gastroenterol. 39. Meijvis, S.C., et al. Microbial evaluation of proton-pump inhibitors
2011;46(6):724–735. and the risk of pneumonia. Eur Respir J. 2011;38(5):1165–1172.
18. Sugano, K., et al. Lansoprazole for secondary prevention of gastric 40. Pasina, L., et al. Prevalence and appropriateness of drug prescrip-
or duodenal ulcers associated with long-term non-steroidal tions for peptic ulcer and gastro-esophageal reflux disease in
anti-inflammatory drug (NSAID) therapy: results of a prospective, a cohort of hospitalized elderly. Eur J Intern Med. 2011;22
multicenter, double-blind, randomized, double-dummy, (2):205–210.
active-controlled trial. J Gastroenterol. 2012;47(5):540–552. 41. Ernst, P., et al. Pramipexole use and the risk of pneumonia. BMC
19. Lima, J.J., et al. Association of CYP2C19 polymorphisms and Neurol. 2012;12:113.
lansoprazole-associated respiratory adverse effects in children. 42. Hermos, J.A., et al. Risk of community-acquired pneumonia in
J Pediatr. 2013;163(3):686–691. veteran patients to whom proton pump inhibitors were
20. Estborn L, Joelson S. Frequency and time to onset of dispensed. Clin Infect Dis. 2012;54(1):33–42.
community-acquired respiratory tract infections in patients receiv- 43. Liu, C.L., et al. Angiotensin-converting enzyme inhibitor/angioten-
ing esomeprazole: a retrospective analysis of patient-level data in sin II receptor blockers and pneumonia risk among stroke patients.
placebo-controlled studies. Aliment Pharmacol Ther. 2015;42 J Hypertens. 2012;30(11):2223–2229.
(5):607–613. 44. Mullerova, H., et al. The natural history of community-acquired
21. Lin, C.C., et al. Stress ulcer prophylaxis in patients being weaned pneumonia in COPD patients: a population database analysis.
from the ventilator in a respiratory care center: a randomized con- Respir Med. 2012;106(8):1124–1133.
trol trial. J Formos Med Assoc. 2016;115(1):19–24. 45. Nielsen, A.G., et al. The impact of statin use on pneumonia risk and
22. Selvanderan, S.P., et al. Pantoprazole or Placebo for Stress Ulcer outcome: a combined population-based case-control and cohort
Prophylaxis (POP-UP): randomized double-blind exploratory study. study. Crit Care. 2012;16(4):R122.
Crit Care Med. 2016;44(10):1842–1850. 46. Jena AB, Sun E, Goldman DP. Confounding in the association of
23. Alhazzani, W., et al. Withholding pantoprazole for stress ulcer proton pump inhibitor use with risk of community-acquired
prophylaxis in critically Ill patients: a pilot randomized clinical trial pneumonia. J Gen Intern Med. 2013;28(2):223–230.
and meta-analysis. Crit Care Med. 2017;45(7):1121–1129. 47. Juthani-Mehta, M., et al. Modifiable risk factors for pneumonia
24. van de Garde, E.M., et al. Angiotensin-converting enzyme inhibitor requiring hospitalization of community-dwelling older adults: the
use and pneumonia risk in a general population. Eur Respir J. health, aging, and body composition study. J Am Geriatr Soc.
2006;27(6):1217–1222. 2013;61(7):1111–1118.
25. Gulmez, S.E., et al. Use of proton pump inhibitors and the risk of 48. Long, M.D., et al. Increased risk of pneumonia among patients with
community-acquired pneumonia: a population-based case-control inflammatory bowel disease. Am J Gastroenterol. 2013;108
study. Arch Intern Med. 2007;167(9):950–955. (2):240–248.
26. Hennessy, S., et al. Observed association between antidepressant 49. Ramsay, E.N., et al. Proton pump inhibitors and the risk of pneu-
use and pneumonia risk was confounded by comorbidity monia: a comparison of cohort and self-controlled case series
measures. J Clin Epidemiol. 2007;60(9):911–918. designs. BMC Med Res Methodol. 2013;13:82.
27. Almirall, J., et al. New evidence of risk factors for 50. Buendgens, L., et al. Administration of proton pump inhibitors in
community-acquired pneumonia: a population-based study. Eur critically ill medical patients is associated with increased risk of
Respir J. 2008;31(6):1274–1284. developing Clostridium difficile-associated diarrhea. J Crit Care.
28. Beaulieu, M., et al. Do proton-pump inhibitors increase the risk for 2014;29(4):696 e11–5.
nosocomial pneumonia in a medical intensive care unit? J Crit Care. 51. de Groot, M.C., et al. Sources of heterogeneity in case-control
2008;23(4):513–518. studies on associations between statins, ACE-inhibitors, and proton
29. Sarkar M, Hennessy S, Yang YX. Proton-pump inhibitor use and the pump inhibitors and risk of pneumonia. Eur J Epidemiol. 2014;29
risk for community-acquired pneumonia. Ann Intern Med. 2008;149 (10):767–775.
(6):391–398. 52. Herzig, S.J., et al. Acid-suppressive medication use in acute stroke
30. Herzig, S.J., et al. Acid-suppressive medication use and the risk for and hospital-acquired pneumonia. Ann Neurol. 2014;76(5):712–718.
hospital-acquired pneumonia. JAMA. 2009;301(20):2120–2128. 53. Ho, S.W., et al. Population-based cohort study on the risk of pneu-
31. Marciniak, C., et al. Examination of selected clinical factors and monia in patients with non-traumatic intracranial haemorrhage
medication use as risk factors for pneumonia during stroke reha- who use proton pump inhibitors. BMJ Open. 2014;4(11):e006710.
bilitation: a case-control study. Am J Phys Med Rehabil. 2009;88 54. Im, J.P., et al. Proton pump inhibitor use before percutaneous
(1):30–38. endoscopic gastrostomy is associated with adverse outcomes.
32. Myles, P.R., et al. Risk of community-acquired pneumonia and the Gut Liver. 2014;8(3):248–253.
use of statins, ace inhibitors and gastric acid suppressants: a 55. Kroupa, R., et al. Is there a real risk of the long term medical
population-based case-control study. Pharmacoepidemiol Drug treatment of gastroesophageal reflux disease? United European
Saf. 2009;18(4):269–275. Gastroenterol J. 2014;2(1):A264.
33. Rodriguez, L.A., et al. Acid-suppressive drugs and 56. Chen, C.H., et al. Proton pump inhibitor usage and the associated
community-acquired pneumonia. Epidemiology. 2009;20 risk of pneumonia in patients with chronic kidney disease.
(6):800–806. J Microbiol Immunol Infect. 2015;48(4):390–396.
34. Roughead, E.E., et al. Proton-pump inhibitors and the risk of anti- 57. Ho, S.W., et al. Risk of stroke-associated pneumonia with
biotic use and hospitalisation for pneumonia. Med J Aust. 2009;190 acid-suppressive drugs: a population-based cohort study.
(3):114–116. Medicine (Baltimore). 2015;94(29):e1227.
35. Dublin, S., et al. Use of proton pump inhibitors and H2 blockers and 58. Lee, S.W., et al. Proton pump inhibitors did not increase risk of
risk of pneumonia in older adults: a population-based case-control pneumonia in patients with chronic obstructive pulmonary disease.
study. Pharmacoepidemiol Drug Saf. 2010;19(8):792–802. J Clin Med Res. 2015;7(11):880–883.
36. Eurich, D.T., et al. Recurrent community-acquired pneumonia in 59. Merli, M., et al. The chronic use of beta-blockers and proton pump
patients starting acid-suppressing drugs. Am J Med. 2010;123 inhibitors may affect the rate of bacterial infections in cirrhosis.
(1):47–53. Liver Int. 2015;35(2):362–369.
10 C.-H. WANG ET AL.
60. Ozkan, M.C., et al. Use of proton pump inhibitors in hematopoietic 80. Harder, H., et al. Effect of 7-day therapy with different doses of the
stem cell transplantation does not increase the frequency of febrile proton pump inhibitor lansoprazole on the intragastric pH in
neutropenia. Eur Rev Med Pharmacol Sci. 2015;19(21):4105–4110. healthy human subjects. Scand J Gastroenterol. 1999;34
61. Terg, R., et al. Proton pump inhibitor therapy does not increase the (6):551–561.
incidence of spontaneous bacterial peritonitis in cirrhosis: 81. Tani N. [Application of proton pump inhibitor to special conditions;
a multicenter prospective study. J Hepatol. 2015;62(5):1056–1060. Zollinger-Ellison syndrome]. Nihon Rinsho. 1992;50(1):154–159.
62. Bonvicini, D., et al. Type of stress-ulcer prophylaxis and incidence of 82. Schneider-Lindner, V., et al. Antibacterial drugs and the risk of
ventilator-associated pneumonia. Intensive Care Med Exp. 2016;4. community-associated methicillin-resistant Staphylococcus aureus
63. Chatterjee, S., et al. Anticholinergic medication use and risk of in children. Arch Pediatr Adolesc Med. 2011;165(12):1107–1114.
pneumonia in elderly adults: a nested case-control study. J Am 83. Wang, J., et al. Alcohol consumption is not protective for systemic
Geriatr Soc. 2016;64(2):394–400. lupus erythematosus. Ann Rheum Dis. 2009;68(3):345–348.
64. Courson, A.Y., et al. Routine prophylaxis with proton pump inhibi- 84. Gonzalez-Perez A, Garcia Rodriguez LA, Lopez-Ridaura R. Effects of
tors and post-transplant complications in kidney transplant recipi- non-steroidal anti-inflammatory drugs on cancer sites other than
ents undergoing early corticosteroid withdrawal. Clin Transplant. the colon and rectum: a meta-analysis. BMC Cancer. 2003;3:28.
2016;30(6):694–702. 85. Tamim H, Monfared AA, LeLorier J. Application of lag-time into
65. Kawanishi, K., et al. Risk factors for aspiration pneumonia after exposure definitions to control for protopathic bias.
endoscopic hemostasis. Dig Dis Sci. 2016;61(3):835–840. Pharmacoepidemiol Drug Saf. 2007;16(3):250–258.
66. McCaleb, R.V., et al. Clinical outcomes of acid suppressive therapy 86. Ali AK. The link between proton pump inhibitors and ischemic
use in hematology/oncology patients at an academic medical heart disease could be explained by protopathic bias:
center. Ann Pharmacother. 2016;50(7):541–547. a pharmacovigilance analysis. Value Health. 2015;18(7):A373–4.
67. Othman F, Crooks CJ, Card TR. Community acquired pneumonia 87. Arfe A, Corrao G. The lag-time approach improved drug-outcome
incidence before and after proton pump inhibitor prescription: association estimates in presence of protopathic bias. J Clin
population based study. BMJ. 2016;355:i5813. Epidemiol. 2016;78:101–107.
•• This is an important cohort study using advanced statistical 88. Jacobson, B.C., et al. Who is using chronic acid suppression therapy
approach to deal with possible confounding factors. and why? Am J Gastroenterol. 2003;98(1):51–58.
68. Sasabuchi, Y., et al. Risks and benefits of stress ulcer prophylaxis for 89. Gaude GS. Pulmonary manifestations of gastroesophageal reflux
patients with severe sepsis. Crit Care Med. 2016;44(7):e464–9. disease. Ann Thorac Med. 2009;4(3):115–123.
69. Six, S., et al. Hyperoxemia as a risk factor for ventilator-associated 90. Rodriguez EM, Staffa JA, Graham DJ. The role of databases in drug
pneumonia. Crit Care. 2016;20(1):195. postmarketing surveillance. Pharmacoepidemiol Drug Saf. 2001;10
70. Vukadinović, D., et al. Factors associated with early treatment fail- (5):407–410.
ure in adult hospitalized patients with community-acquired 91. Calabrese, C., et al. Ninety-six-hour wireless oesophageal pH mon-
pneumonia. Vojnosanit Pregl. 2017;74(9):803–813. itoring following proton pump inhibitor administration in NERD
71. Arai, N., et al. Histamine H2-blocker and proton pump inhibitor use patients. Aliment Pharmacol Ther. 2008;28(2):250–255.
and the risk of pneumonia in acute stroke: a retrospective analysis 92. Hellstrom PM, Vitols S. The choice of proton pump inhibitor: does it
on susceptible patients. PLoS One. 2017;12(1):e0169300. matter? Basic Clin Pharmacol Toxicol. 2004;94(3):106–111.
72. Bang UC, Bendtsen F. The use of proton pump inhibitors is not 93. Meyer, C., et al. Is publication bias present in gastroenterological
associated with an increased risk of infections in a cohort of research? An analysis of abstracts presented at an annual congress.
patients with alcoholic cirrhosis from danish nationwide registers. PeerJ. 2018;6:e4995.
Hepatology. 2017;66:286A. 94. Golder, S., et al. Reporting of adverse events in published and
73. Ho, S.W., et al. Association of proton pump inhibitors usage with unpublished studies of health care interventions: a systematic
risk of pneumonia in dementia patients. J Am Geriatr Soc. 2017;65 review. PLoS Med. 2016;13(9):e1002127.
(7):1441–1447. 95. Doshi P, Jefferson T, Del Mar C. The imperative to share clinical
74. Hsu, W.T., et al. Risk of pneumonia in patients with gastroesopha- study reports: recommendations from the Tamiflu experience. PLoS
geal reflux disease: a population-based cohort study. PLoS One. Med. 2012;9(4):e1001201.
2017;12(8):e0183808. 96. Wieseler, B., et al. Access to regulatory data from the european
75. Rane, P., et al. Risk of pneumonia associated with proton pump medicines agency: the times they are a-changing. Syst Rev.
inhibitor use among community dwelling adults with gastroeso- 2012;1:50.
phageal reflux disease. Value Health. 2017;20(5):A179–A180. 97. Alves C, Macedo AF, Marques FB. Sources of information used by
76. Zirk-Sadowski, J., et al. Risks of longer term proton pump inhibitor regulatory agencies on the generation of drug safety alerts. Eur
exposure in 228,752 older adults. Eur Geriatric Med. 2017;8:S29–S30. J Clin Pharmacol. 2013;69(12):2083–2094.
77. Giannella RA, Broitman SA, Zamcheck N. Gastric acid barrier to 98. Inacio P, Cavaco A, Airaksinen M. The value of patient reporting to
ingested microorganisms in man: studies in vivo and in vitro. Gut. the pharmacovigilance system: a systematic review. Br J Clin
1972;13(4):251–256. Pharmacol. 2017;83(2):227–246.
78. Torres, A., et al. Stomach as a source of colonization of the respira- 99. Lane S, Lynn E, Shakir S. Investigation assessing the publicly avail-
tory tract during mechanical ventilation: association with able evidence supporting postmarketing withdrawals, revocations
ventilator-associated pneumonia. Eur Respir J. 1996;9 and suspensions of marketing authorisations in the EU since 2012.
(8):1729–1735. BMJ Open. 2018;8(1):e019759.
79. Williams C, McColl KE. Review article: proton pump inhibitors and 100. Savarino V, Dulbecco P, Savarino E. Are proton pump inhibitors
bacterial overgrowth. Aliment Pharmacol Ther. 2006;23(1):3–10. really so dangerous? Dig Liver Dis. 2016;48(8):851–859.