P2034 - Association of Proton Pump Inhibitor Use with All-cause and Cause-specific Mortality

Monday, October 25, 2021

10:30 AM – 4:15 PM PT

Location: Shoreline Exhibit Hall

Has Audio

Presenting Author(s)

Chun-Han Lo, MD, MPH

Massachusetts General Hospital
Boston, MA, United States

Chun-Han Lo, MD, MPH¹, Peiyun Ni, MD¹, Yan Yan, MD, PhD², Wenjie Ma, MBBS, PhD¹, Amit D. Joshi, MBBS, PhD¹, Long H. Nguyen, MD, MS¹, Raaj S. Mehta, MD³, Paul Lochhead, MD, MPH, PhD¹, Mingyang Song, MBBS⁴, Gary C. Curhan, MD, ScD⁵, Yin Cao, PhD⁶, Andrew T. Chan, MD, MPH¹
¹Massachusetts General Hospital, Boston, MA; ²Washington University Medical School, St. Louis, MO; ³Massachusetts General Hospital, Cambridge, MA; ⁴Harvard T.H. Chan School of Public Health, Boston, MA; ⁵Brigham & Women's Hospital/Harvard Medical School, Boston, MA; ⁶Washington University School of Medicine in St. Louis, St. Louis, MO

Introduction: The use of proton pump inhibitors (PPIs) has rapidly increased in the past two decades. Concerns about the regular use of PPIs contributing to mortality have been raised. We aimed to investigate the association between long-term use of PPIs and the risk of death.

Methods: We performed a prospective, new-user study of two nationwide cohorts of health professionals in the United States—the Nurses’ Health Study (2004-2018) and the Health Professionals Follow-up Study (2004-2018). We utilized marginal structural Cox models with adjustment for time-dependent confounding to calculate the hazard ratios (HRs) and 95% confidence intervals (CIs) for death according to PPI use and a lag-time approach to control for confounding by indication (protopathic bias).

Results: Among 38,539 women and 15,653 men followed up for 602,648 person-years and a median of 13.8 years, we documented 9,950 deaths, including 2,245 deaths from cancer, 2,202 from cardiovascular diseases, and 5,503 deaths from other causes. Compared to nonusers of PPI, new PPI users had a significantly higher risk of all-cause mortality and mortality due to cancer, respiratory diseases, digestive diseases, and renal diseases. After applying a lag-time period of up to six years, the associations were attenuated and no longer statistically significant (all-cause: hazard ratio [HR], 1.08; 95% confidence interval [CI], 0.98-1.20; cancer: HR, 1.07; 95% CI, 0.83-1.36; respiratory diseases: HR, 1.22; 95% CI, 0.88-1.68; and digestive diseases: HR, 0.91; 95% CI, 0.49-1.69) with the exception of mortality due to renal diseases (HR, 2.65; 95% CI, 1.52-4.64). Longer duration of PPI use did not confer a higher risk of all-cause mortality and mortality due to cancer, respiratory diseases, and digestive diseases but was associated with death due to renal disease.

Discussion: Regular use of PPIs was not associated with a higher risk of all-cause and most cause-specific mortality after accounting for potential confounding by indication. Given that PPI use has been linked to an increased risk of acute kidney injury and chronic kidney disease, the observed association between PPI use and death from renal diseases requires further confirmation.

	Lag-time period
	No lag		2 years		4 years		6 years
Cause of death	No. of events	HR (95% CI)^a	No. of events	HR (95% CI)^a	No. of events	HR (95% CI)^a	No. of events	HR (95% CI)^a
All
PPI nonusers	9038	1 (reference)	9767	1 (reference)	9496	1 (reference)	8531	1 (reference)
PPI users	912	1.29 (1.13-1.46)	877	1.11 (1.00-1.23)	790	1.10 (1.00-1.21)	632	1.08 (0.98-1.20)
Cancer
PPI nonusers	2027	1 (reference)	1953	1 (reference)	1644	1 (reference)	1222	1 (reference)
PPI users	218	1.42 (1.19-1.69)	176	1.27 (1.07-1.51)	118	1.10 (0.90-1.34)	80	1.07 (0.83-1.36)
Cardiovascular diseases
PPI nonusers	2027	1 (reference)	2136	1 (reference)	2029	1 (reference)	1766	1 (reference)
PPI users	175	1.07 (0.85-1.35)	163	0.98 (0.80-1.20)	142	0.93 (0.76-1.13)	113	0.97 (0.77-1.22)
Others^b
PPI nonusers	4984	1 (reference)	5673	1 (reference)	5823	1 (reference)	5543	1 (reference)
PPI users	519	1.31 (1.13-1.51)	538	1.11 (0.98-1.26)	530	1.15 (1.03-1.29)	439	1.12 (1.00-1.26)
Respiratory diseases
PPI nonusers	785	1 (reference)	810	1 (reference)	741	1 (reference)	654	1 (reference)
PPI users	95	1.63 (1.26-2.10)	89	1.39 (1.08-1.80)	83	1.51 (1.16-1.97)	56	1.22 (0.88-1.68)
Digestive diseases
PPI nonusers	208	1 (reference)	211	1 (reference)	185	1 (reference)	159	1 (reference)
PPI users	29	1.61 (1.02-2.54)	26	1.40 (0.87-2.27)	24	1.78 (1.03-3.10)	14	0.91 (0.49-1.69)
Renal diseases
PPI nonusers	114	1 (reference)	126	1 (reference)	129	1 (reference)	111	1 (reference)
PPI users	24	2.98 (1.82-4.88)	22	2.66 (1.61-4.40)	18	2.08 (1.23-3.51)	18	2.65 (1.52-4.64)
Neurologic diseases
PPI nonusers	1025	1 (reference)	1423	1 (reference)	1643	1 (reference)	1661	1 (reference)
PPI users	85	1.17 (0.90-1.53)	96	0.90 (0.69-1.16)	108	0.87 (0.69-1.09)	100	0.86 (0.67-1.11)
Infection
PPI nonusers	224	1 (reference)	237	1 (reference)	222	1 (reference)	200	1 (reference)
PPI users	24	1.05 (0.63-1.74)	26	1.19 (0.75-1.90)	23	1.42 (0.89-2.27)	12	0.99 (0.51-1.93)
Abbreviations: CI, confidence interval; H2RA, histamine-2 receptor antagonist; HR, hazard ratio; NSAID, nonsteroidal anti-inflammatory drug; PPI, proton pump inhibitor. ^a Weights for the marginal structural Cox model were computed using baseline and time-varying variables, including age, cohort, questionnaire cycle, smoking status, body mass index, physical activity, Alternate Healthy Eating Index-2010, alcohol intake, regular NSAIDs/aspirin use, H2RA use in the past, history of cancer, myocardial infarction, stroke, gastroesophageal reflux disease, Barrett's esophagus, peptic ulcer disease, and gastrointestinal bleeding. ^b Others include respiratory diseases, digestive diseases, renal diseases, neurologic diseases, infection, and other less common causes.

Table: Table. Hazard Ratios (95% Confidence Intervals) for All-cause and Cause-specific Mortality According to Proton Pump Inhibitor Use

Disclosures:
Chun-Han Lo indicated no relevant financial relationships.
Peiyun Ni indicated no relevant financial relationships.
Yan Yan indicated no relevant financial relationships.
Wenjie Ma indicated no relevant financial relationships.
Amit Joshi indicated no relevant financial relationships.
Long Nguyen indicated no relevant financial relationships.
Raaj Mehta indicated no relevant financial relationships.
Paul Lochhead indicated no relevant financial relationships.
Mingyang Song indicated no relevant financial relationships.
Gary Curhan: OM1, Inc – Employee, Stockholder/Ownership Interest (excluding diversified mutual funds). UpToDate – Other Financial or Material Support, Author, Section Editor.
Yin Cao: .
Andrew Chan: Bayer Pharma AG – Consultant. Boehringer Ingelheim – Consultant. Pfizer Inc. – Consultant.

Chun-Han Lo, MD, MPH¹, Peiyun Ni, MD¹, Yan Yan, MD, PhD², Wenjie Ma, MBBS, PhD¹, Amit D. Joshi, MBBS, PhD¹, Long H. Nguyen, MD, MS¹, Raaj S. Mehta, MD³, Paul Lochhead, MD, MPH, PhD¹, Mingyang Song, MBBS⁴, Gary C. Curhan, MD, ScD⁵, Yin Cao, PhD⁶, Andrew T. Chan, MD, MPH¹. P2034 - Association of Proton Pump Inhibitor Use with All-cause and Cause-specific Mortality, ACG 2021 Annual Scientific Meeting Abstracts. Las Vegas, Nevada: American College of Gastroenterology.