DR MAHYAR ETMINAN (Orcid ID : 0000-0003-4628-6270)

Accepted Article
Article type : Brief Report

Title: Risk of Gynecomastia with Users of Proton Pump

Inhibitors

Running Title: Risk of Gynecomastia with Proton Pump Inhibitor Use

Bonnie He1, Bruce Carleton2-4, Mahyar Etminan5

1
Department of Medicine, Faculty of Medicine, University of British Columbia, Vancouver,
Canada
2
Division of Translational Therapeutics, Department of Pediatrics, University of British
Columbia, Vancouver, Canada
3
British Columbia Children’s Hospital Research Institute, University of British Columbia,
Vancouver, Canada
4
Pharmaceutical Outcomes Program, British Columbia Children’s Hospital, Vancouver,
Canada
5
Department of Ophthalmology and Visual Sciences and Pharmacology, University of British
Columbia, Vancouver, Canada

Corresponding Author:
Mahyar Etminan
Assistant Professor of Ophthalmology and Visual Sciences| Faculty of Medicine
Associate Member, Department of Anesthesia, Pharmacology and Therapeutics
The University of British Columbia |The Eye Care Center
Room 323-2550 Willow Street, Vancouver BC, V5Z 3N9
Phone 604-875-4725 | Fax 604-875-4663
Email: etminanm@mail.ubc.ca
ORCID: 0000-0003-4628-6270

ACKNOWLEDGEMENTS
The study was funded by the therapeutic evaluation unit of the British Columbia Provincial
Health Services Authority.

This article has been accepted for publication and undergone full peer review but has not
been through the copyediting, typesetting, pagination and proofreading process, which may
lead to differences between this version and the Version of Record. Please cite this article as
doi: 10.1002/phar.2245
This article is protected by copyright. All rights reserved.
 CONFLICT OF INTEREST
Bonnie He, Bruce Carleton and Mahyar Etminan have no conflicts of interest.
Accepted Article
ABSTRACT

Introduction: Proton pump inhibitors (PPI) are commonly prescribed for many
gastrointestinal diseases. A number of case reports have linked PPIs to gynecomastia in men,
but large epidemiologic studies are lacking.

Objective: To quantify the risk of gynecomastia with PPIs in male patients

Methods: Using the PharMetrics Plus™ health claims database from the United States, a
retrospective cohort study of new PPI users and new amoxicillin users from 2006 to 2016
was conducted. Diagnosis of gynecomastia was identified by the International Classification
for Diseases, 9th edition (ICD-9) and 10th edition (ICD-10) codes. Cases were defined as
patients with two codes for gynecomastia within 90 days, with the first code as the event code.
Hazard ratios (HRs) were computed by adjusting for alcoholic cirrhosis, hyperthyroidism,
testicular cancer, Klinefelter syndrome, and obesity; as well as the use of ketoconazole,
risperidone, spironolactone, and androgen deprivation therapy. A sensitivity analysis defining
exposure with two PPI prescriptions was also undertaken.

Results: There were 389 cases of gynecomastia diagnosed among 220,791 new PPI users,
and 996 gynecomastia cases diagnosed among 837,740 new amoxicillin users. The crude HR
for PPI use compared to amoxicillin use was 1.70 (95% confidence interval (CI): 1.461-
1.976). The adjusted HR for the sensitivity analysis was 1.299 (95% CI: 1.146-1.473). The
adjusted HR was 1.4795 (95% CI: 1.2431-1.7609) for patients over 50 years old and 1.324
(95% CI: 1.1133-1.5745) for patients 50 years old or younger.

Conclusion: This large retrospective cohort study suggests that patients who used PPIs are at
higher risk of developing gynecomastia. Clinicians may want to convey this information to
male patients who require long-term PPI therapy.

Key Words: Proton pump inhibitors; gynecomastia; retrospective cohort study

Proton-pump inhibitors (PPIs), used in treating symptoms related to gastroesophageal reflux

disease, are one of the most widely prescribed classes of drugs in the United States.(1) Proton

pump inhibitors (PPIs) block gastric acid secretion by binding irreversibly to hydrogen-

potassium adenosine triphosphatase enzymes found on gastric parietal cells.(2) Several

randomized controlled trials evaluating PPIs have demonstrated their efficacy in the

management of various gastrointestinal disorders including erosive esophagitis, peptic ulcer

disease, gastroesophageal reflux disease, and non-ulcer dyspepsia.(2, 3) But recent studies

This article is protected by copyright. All rights reserved.

 have found numerous adverse effects associated with long-term PPI use, including
Accepted Article hypomagnesemia, kidney disease, pneumonia, bone fractures, and Clostridium difficile

infection.(2) One adverse effect of PPIs that has been largely understudied is gynecomastia,

an enlargement of breast tissue in males.(4) Although gynecomastia is generally

asymptomatic and often not noted by patients, common complaints in individuals who

experience gynecomastia include social anxiety from aesthetic concerns and discomfort from

local pain and tenderness.(5) In particular, as treatment for gastric acid-related diseases in

children has risen substantially in recent years,(6) the psychological burden of gynecomastia

is significant and can threaten self-esteem and self-identity.(7) This is especially true for

adolescents during the pubertal period in which the gender-incongruent experience can result

in severe emotional consequences including depression, social withdrawal, bulimia-anorexia,

school behavioral problems, and even suicide.(8)

Unfortunately, only a few studies examining the pathological causes of gynecomastia have

been conducted, most of which have been case reports with a small population of patients.(9-

12) Furthermore, there is discrepancy on whether gynecomastia occurs secondary to PPI use

as one pharmacoepidemiologic study found no clinical association with PPIs and

development of gynecomastia,(13) whereas a more recent study from the Spanish

Pharmacovigilance System identified 24 cases of PPI-related gynecomastia.(14) With the

increasing use of PPIs and limited information on their adverse effects, it is imperative for

both patients and clinicians to be made aware of any associations between gynecomastia and

PPIs. In light of the sparse evidence, we conducted a retrospective cohort study of middle-

aged men in the United States to quantify the risk of gynecomastia secondary to PPI use.

This article is protected by copyright. All rights reserved.

 MATERIALS AND METHODS
Accepted Article Data sources

PharMetrics Plus™, a large private health claims database from the United States, was used

as the main data source for this study. This database captures health claims for more than 150

million unique enrollees, with fully adjudicated medical and pharmacy claims. Medical

claims data captures all physician visits including medical diagnoses through the

International Classification of Diseases, 9th and 10th editions (ICD-9 and 10) codes. The

ICD-9 codes were available from 2006 to 2014 while the ICD-10 codes were available from

2015 to 2016. All prescription drug data includes information on the drug name, the dose

quantity dispensed, and the days of medication supplied. The database has good

representation of all geographic areas of the United States and has been used in previous

pharmacoepidemiologic studies.(15, 16) We had access to a random sample of 9,053,240

patients from 2006 to 2016.

Study Design and Cohort Description

We used a new users retrospective cohort study design and identified new users of PPIs and

new users of amoxicillin from 2006 to 2016. To avoid immortal time bias, we avoided using

a random sample from the cohort as the comparator group and selected instead a comparator

group that was also taking a drug not known to be associated with gynecomastia. As such,

amoxicillin users were chosen as the control group since amoxicillin has not been associated

with gynecomastia. A 1-year period prior to the index date was used to ensure incident drug

use and ascertain covariate information. The date of the first PPI or amoxicillin prescription

was deemed the index date. Each cohort member was followed to the first diagnosis of

gynecomastia (identified through ICD-9 611.1 or ICD-10 N62 codes), a switch from a PPI or

amoxicillin (or vice versa), or termination of the follow-up period. A gynecomastia case was

This article is protected by copyright. All rights reserved.

 required to have had at least two codes for gynecomastia with the second code documented
Accepted Article within 90 days of the first code.

Statistical analysis

Descriptive statistics were used to examine covariate distribution between PPI and

amoxicillin users. A Cox Proportional Hazards model was constructed to compute an

adjusted hazards ratio (HR) for the following variables that may have increased the risk of

gynecomastia: alcoholic cirrhosis, hyperthyroidism, testicular cancer, Klinefelter syndrome,

and obesity; as well as the use of ketoconazole, risperidone, spironolactone, and androgen

deprivation therapy within the last 12 months. Results were also stratified by age as the risk

of gynecomastia increases after the age of 50 years old.(17) We also undertook a sensitivity

analysis to examine the risk of gynecomastia among those who used two prescriptions for

either a PPI or amoxicillin. This was done to ensure maximum exposure and prevent any

possible misclassification that may have occurred if patients took only one prescription and

did not renew subsequent prescriptions (exposure misclassification). Use of two prescriptions

was defined as having received a second prescription within 40 days of dispensing the first

prescription. The 40-day period was computed by considering a 30-day prescription renewal

period with an extra 10 days of grace period to allow time for renewal. Finally, the rate ratio

was also computed for boys (aged 10-19 years).

This article is protected by copyright. All rights reserved.

 RESULTS
Accepted Article There were 220,791 new PPI users and 837,740 new amoxicillin users with a mean age of

47.6 (standard deviation (SD) 16.0) years old and 30.7 (SD 21.3) years old, respectively

(Table 1). There were 389 incident cases of gynecomastia were identified among PPI users

and 996 incident gynecomastia cases were diagnosed among amoxicillin users. The follow-

up and total observation time for PPI users was 2.9 years (SD 2.2 years) and 630,998 person-

years, respectively, whereas amoxicillin users had a follow-up time of 3.1 years (SD 2.4

years) and total observation time of 2,603,129 person-years. The crude HR for PPI use

compared to amoxicillin use was 1.70 (95% confidence interval [CI]: 1.461-1.976). The

number needed to harm is 1746. Stratifying for age, the adjusted HR for PPI and amoxicillin

users who were over 50 years old was 1.4795 (95% CI: 1.2431-1.7609). The adjusted HR for

PPI and amoxicillin users who were 50 years old or younger was 1.324 (95% CI: 1.1133-

1.5745). The adjusted HR for the sensitivity analysis was 1.299 (95% CI; 1.146-1.473). The

adjusted HR for children under the age of 20 years old was 1.83 (95% CI: 1.22-2.79).

Compared to amoxicillin users, PPI users had a higher utilization of ketoconazole and

spironolactone (Table 1).

DISCUSSION

This study is the first large epidemiologic study to quantify the risk of gynecomastia related

to PPI use. Due to its versatility in treating a spectrum of gastric diseases, PPIs are one of the

most frequently prescribed therapies in clinical practice and are also often used as over-the-

counter medications. However, there is mounting evidence that PPIs are associated with

various side effects, including acute and chronic kidney disease, hypomagnesemia,

Clostridium difficile, pneumonia, bone fractures, and even gastric cancer.(18-24)

Gynecomastia, another adverse effect related to PPI use, has wide-ranging psychosocial

This article is protected by copyright. All rights reserved.

 consequences and can result in severe distress, particularly among adolescent males.(25) In
Accepted Article our study, we found that users of PPIs were 30% more likely to develop gynecomastia

compared to users of amoxicillin, however the absolute risk is small, as the number needed to

harm was 1746. Nevertheless, as there are approximately 14.9 million patients who are

prescribed PPIs in the United States,(26) and assuming that half (7.5 million) of the patients

receiving PPIs are men, this means that there are potentially 4295 male patients with

gynecomastia. As such, clinicians should therefore discuss with patients the potential risks of

gynecomastia and its impact on social and emotional quality of life prior to prescribing PPIs,

particularly with patients who are more vulnerable to psychosocial distress (i.e., adolescents,

have anxiety, etc.).

The etiology of gynecomastia is unclear, but it is believed to be attributed to an excess of

estrogens, deficiency of androgens, or an altered estrogen-testosterone ratio.(27) Omeprazole

at high doses has been observed to inhibit cytochrome P450 3A4, a liver enzyme responsible

for the metabolism of estradiol.(28, 29) This results in an increase of estradiol levels, and

therefore an elevated estrogen-androgen ratio, that triggers the development of gynecomastia.

Various case reports have found that the time course from the initiation of PPI use to the

onset of gynecomastia is approximately 3 months.(10, 11)

In order to account for physiological changes that occur with aging, we performed an age-

matched analysis by stratifying the cohorts of PPI users and amoxicillin users by age (greater

than 50 years old vs. 50 years old or younger). The strengths in our study include having a

sufficiently large sample size to generate a meaningful statistical analysis in assessing the

link between gynecomastia and PPI use and allowing for a 1-year period prior to the index

date to ensure incident use. One limitation of our findings is that we did not account for other

This article is protected by copyright. All rights reserved.

 factors that are known to be associated with gynecomastia such as race, anabolic steroid
Accepted Article abuse, substantial weight loss, chronic kidney disease, and hereditary gynecomastia.(5) We

also did not have access to medical information – specifically the diagnosis of gynecomastia

and body mass index. Although we did adjust for obesity in our study, it is possible there may

still be some confounding residues from obesity as some patients with obesity may not have

been coded for this condition. However, the requirement for at least two physician visit codes

for gynecomastia provided extra reassurance that patients with gynecomastia did return for

follow up.

In conclusion, the results of this large retrospective cohort study suggest an increase in the

risk of gynecomastia among male patients who use PPIs. Clinicians should be cognizant of

this adverse event when prescribing PPIs, especially to young men or boys who may be

negatively affected by the social, mental, and emotional impacts of gynecomastia related to

PPI use.

REFERENCES

1. Kantor ED, Rehm CD, Haas JS, Chan AT, Giovannucci EL. Trends in prescription drug
use among adults in the united states from 1999-2012. JAMA 2015:314:1818-30.

2. Nehra AK, Alexander JA, Loftus CG, Nehra V. Proton pump inhibitors: Review of
emerging concerns. Mayo Clin Proc 2018 Feb;93(2):240-6.

3. Vaezi MF, Yang YX, Howden CW. Complications of proton pump inhibitor therapy.
Gastroenterology 2017:153:35-48.

4. Nuttall FQ, Warrier RS, Gannon MC. Gynecomastia and drugs: A critical evaluation of the
literature. Eur J Clin Pharmacol 2015:71:569-78.

5. Narula HS, Carlson HE. Gynaecomastia—pathophysiology, diagnosis and treatment.

Nature Reviews Endocrinology 2014:10:684-94.

This article is protected by copyright. All rights reserved.

 6. Ward RM, Kearns GL. Proton pump inhibitors in pediatrics: Mechanism of action,
pharmacokinetics, pharmacogenetics, and pharmacodynamics. Paediatr Drugs 2013:15:119-
Accepted Article 31.

7. Sollie M. Management of gynecomastia—changes in psychological aspects after surgery—

a systematic review. Gland Surgery 2018:7:S70-S76.

8. Kinsella CJ, Landfair A, Rottgers SA, et al. The psychological burden of idiopathic
adolescent gynecomastia. Plast Reconstr Surg 2012:129:1-7

9. Gynecomastia during omeprazole therapy. N Engl J Med 1991:324:635.

10. Lindquist M, Edwards IR. Endocrine adverse effects of omeprazole. BMJ: British
Medical Journal 1992:305:451-2.

11. Convens C, Verhelst J, Mahler C. Painful gynaecomastia during omeprazole therapy. The
Lancet 1991:338:1153

12. Thompson DF, Carter JR. Drug-induced gynecomastia. Pharmacotherapy: The Journal of
Human Pharmacology and Drug Therapy 1993:13:37-45.

13. García Rodríguez LA, Jick H. Risk of gynaecomastia associated with cimetidine,
omeprazole, and other antiulcer drugs. BMJ: British Medical Journal 1994:308:503-6.

14. Carvajal A, Macias D, Gutierrez A, et al. Gynaecomastia associated with proton pump
inhibitors: A case series from the Spanish pharmacovigilance system. Drug Saf 2007:30:527-
31.

15. Skeldon SC, Carleton B, Brophy JM, Sodhi M, Etminan M. Statin medications and the
risk of gynecomastia. Clin Endocrinol 2018:89:470-3.

16. Etminan M, Carleton B, Brophy JM. Risperidone and risk of gynecomastia in young men.
J Child Adolesc Psychopharmacol 2015:25:671-3.

17. Niewoehner CB, Nuttal FQ. Gynecomastia in a hospitalized male population. Am J Med
1984:77:633-8.

18. Zhou B, Huang Y, Li H, Sun W, Liu J. Proton-pump inhibitors and risk of fractures: An
update meta-analysis. Osteoporos Int 2016:27(1):339-47.

19. Eom CS, Jeon CY, Lim JW, et al. Use of acid-suppressive drugs and risk of pneumonia:
A systematic review and meta-analysis. CMAJ 2011:83:310-9.

20. Kwok CS, Arthur AK, Anibueze CI, et al. Risk of clostridium difficile infection with acid
suppressing drugs and antibiotics: Meta-analysis. Am J Gastroenterol 2012:107:1011-9.

21. Cheungpasitporn W, Thongprayoon C, Kittanamongkolchai W, et al. Proton pump

inhibitors linked to hypomagnesemia: A systematic review and meta-analysis of
observational studies. Ren Fail 2015:37:1237-41.

This article is protected by copyright. All rights reserved.

 22. Antoniou T, Macdonald EM, Hollands S, Gomes T, Mamdani MM, Garg AX, Paterson
JM, Juurlink DN. Proton pump inhibitors and the risk of acute kidney injury in older patients:
Accepted Article A population-based cohort study. CMAJ Open 2015:166-71.

23. Lazarus B, Chen Y, Wilson FP, et a. Proton pump inhibitor use and the risk of chronic
kidney disease. JAMA Internal Medicine 2016:176:238-246

24. Brusselaers N, Wahlin K, Engstrand L, Lagergren J. Maintenance therapy with proton

pump inhibitors and risk of gastric cancer: A nationwide population-based cohort study in
Sweden. BMJ Open 2017;7.

25. Nuzzi LC, Cerrato FE, Erikson CR, Webb ML, Rosen H, Walsh EM, DiVasta AD,
Greene AK, Labow BI. Psychosocial impact of adolescent gynecomastia: A prospective case-
control study. Plast Reconstr Surg 2013:131: 890–896.

26. Lam JR, Schneider JL, Quesenberry CP, Corley DA. Proton pump inhibitor and
histamine-2 receptor antagonist use and iron deficiency. Gastroenterology 2017:152(4):821-
829.

27. Johnson RE, Murad MH. Gynecomastia: Pathophysiology, Evaluation, and Management.
May Clinic Proceedings 2009.84:1010-15.

28. Satoh T, Munakata H, Fujita K, et al. Studies on the interactions between drug and
estrogen. II. on the inhibitory effect of 29 drugs reported to induce gynecomastia on the
oxidation of estradiol at C-2 or C-17. Biol Pharm Bull 2003:695-700.

29. Satoh T, Fujita KI, Munakata H, et al. Studies on the interactions between drugs and
estrogen: Analytical method for prediction system of gynecomastia induced by drugs on the
inhibitory metabolism of estradiol using Escherichia coli coexpressing human CYP3A4 with
human NADPH-cytochrome P450 reductase. Anal Biochem 20002;86:179-86.

This article is protected by copyright. All rights reserved.

 TABLES
Accepted Article Table 1. Baseline Demographics and Covariates for Amoxicillin and PPI Users

Variables Amoxicillin PPI

(N=837,740) (N=220,791)

Age: mean (SD) 30.7 (21.3) 47.6 (16.0)

Follow-up Time (years): mean (SD) 3.1 (2.4) 2.9 (2.2)

Baseline comorbidity n (%) n (%)

Alcoholic cirrhosis 1891 (0.23%) 2410 (1.09%)

Hyperthyroidism 3061 (0.37%) 2009 (0.91%)

Testicular cancer 692 (0.08%) 370 (0.17%)

Klinefelter syndrome 119 (0.01%) 44 (0.02%)

Obesity 51,551 (6.15%) 25,291 (11.45%)

Other medications used

within last 12 months

Ketoconazole 18,978 (2.27%) 7582 (3.43%)

Risperidone 2888 (0.34%) 1102 (0.50%)

Spironolactone 3169 (0.38%) 2617 (1.19%)

H2 Blocker 0 0

ADT 316 (0.04%) 233 (0.11%)

(SD: standard deviation; N: total number of users; n = total number of cases; H2 blocker = histamine-2 receptor antagonist; ADT =
androgen deprivation therapy)

This article is protected by copyright. All rights reserved.