Sage Journals: Discover world-class research

Abstract

French

Background:

Kidney transplant recipients are commonly prescribed proton-pump inhibitors (PPIs), but due to concern for polypharmacy, chronic use should be limited.

Objective:

The objective was to describe PPI use in kidney transplant recipients beyond their first year of transplant to better inform and support deprescribing initiatives.

Design:

We conducted a retrospective, population-based cohort study using linked health care databases.

Setting:

This study was conducted in Alberta, Canada.

Patients:

We included all prevalent adult, kidney-only transplant recipients between April 2008 and December 2017 who received their transplant between May 2002 and December 2017.

Measurements:

The primary outcome was ongoing or new PPI use and patterns of use, including frequency and duration of therapy, and assessment of indication for PPI use.

Methods:

We ascertained baseline characteristics, covariate information, and outcome data from the Alberta Kidney Disease Network (AKDN). We compared recipients with evidence of a PPI prescription in the 3 months prior to study entry to those with a histamine-2-receptor antagonist (H2Ra) fill and those with neither.

Results:

We identified 1823 kidney transplant recipients, of whom 868 (48%) were on a PPI, 215 (12%) were on an H2Ra, and 740 (41%) were on neither at baseline. Over a median follow-up of 5.4 years (interquartile range [IQR] = 2.6-9.3), there were almost 45 000 unique PPI prescriptions dispensed, the majority (80%) of which were filled by initial PPI users. Recipients who were on a PPI at baseline would spend 91% (IQR = 70-98) of their graft survival time on a PPI in follow-up, and nephrologists were the main prescribers. We identified an indication for ongoing PPI use in 54% of recipients with the most common indication being concurrent antiplatelet use (26%).

Limitations:

Our kidney transplant recipients have access to universal health care coverage which may limit generalizability. We identified common gastrointestinal indications for PPI use but did not include rare conditions due to concerns about the validity of diagnostic codes. In addition, symptoms suggestive of reflux may not be well coded as the focus of follow-up visits is more likely to focus on kidney transplant.

Conclusions:

Many kidney transplant recipients are prescribed a PPI at, or beyond, the 1-year post-transplant date and are likely to stay on a PPI in follow-up. Almost half of the recipients in our study did not have an identifiable indication for ongoing PPI use. Nephrologists frequently prescribe PPIs to kidney transplant recipients and should be involved in deprescribing initiatives to reduce polypharmacy.

Keywords

Alberta Choosing Wisely epidemiology kidney transplantation population-based research

Introduction

Proton-pump inhibitors (PPIs) are frequently prescribed medications in the ambulatory care setting (8%-10% of all prescriptions).^1,2 The Canadian Association of Gastroenterology recommends that chronic PPI use (defined as >8 weeks) be limited to patients with specific indications including prior gastrointestinal bleed (GIB) or concurrent non–steroidal anti-inflammatory drug (NSAID) use.³ Choosing Wisely Canada suggests discontinuation of PPI should be trialed at least once per year due to concerns of adverse events.^3-5 As with the general population, long-term PPI use in kidney transplant recipients has been linked to complications, such as hip fractures, kidney dysfunction, and mortality.⁵

In the kidney transplant recipient population, PPIs are often prescribed immediately post-transplant for prophylaxis against peptic ulcer disease caused by high-dose prednisone or dyspepsia due to mycophenolate acid. In addition to potential complications of chronic PPI use, increased pill burden in this population is associated with non-adherence and poor quality of life.^6,7 Therefore, reassessment of PPI use beyond the first year of transplant, when immunosuppression dosage is typically at its lowest, may be considered to minimize complications and reduce pill burden. To better inform and support deprescribing initiatives, we performed a retrospective population–based study to quantify and characterize ongoing PPI use in kidney transplant recipients beyond their first year of transplant.

Methods

Design and Setting

We conducted a retrospective population–based cohort study using linked health care databases within the Alberta Kidney Disease Network (AKDN), which incorporates data from Alberta Health, the provincial health ministry.⁸ Over 99% of Alberta residents are registered with Alberta Health and have universal access to hospital care and physician services. We followed guidelines for the reporting of observational studies (Supplemental Table S1) and a protocol approved by the research ethics boards at the University of Alberta and the University of Calgary, with a waiver of patient consent.

Data Sources

We ascertained baseline characteristics, covariate information, and outcome data from the AKDN records (Supplemental Table S2). The Alberta Health database contains information on demographic data, vital statistics, and diagnostic and procedural information for inpatient and outpatient physician services. We identified kidney transplant recipients from the Alberta Kidney Care (AKC) North and South databases, which provide care to all patients in the province treated with kidney replacement therapy, including dialysis and kidney transplantation. The Pharmaceutical Information Network (PIN) captures prescription drug information on all medications dispensed in Alberta since January 1, 2008, and includes information on drug dosage, day supply, and prescribing physician specialty. The PIN database was used to identify prescription fills for PPIs and histamine-2-receptor antagonists (H2Ras) as well as other relevant medications including immunosuppression, antiplatelets, anticoagulants, and NSAIDs (Supplemental Table S3). These databases have been previously used for research on health services and health outcomes in the kidney transplant recipient population.^9-13

Population

We included all prevalent adult, kidney-only transplant recipients between April 1, 2008 and December 31, 2017 who received their transplant between May 1, 2002 and December 31, 2017 in Alberta, Canada (accrual period). We excluded pediatric recipients (<18 years old) and recipients of a previous non-kidney organ transplant or a simultaneous multi-organ transplant, including kidney-pancreas. Recipients were excluded if they died, emigrated from the province, or experienced graft failure (defined as return to maintenance dialysis) within the first year of their transplant or prior to April 1, 2008. Thus, to be included in the study, recipients must have survived to April 1, 2008 or the 1-year post-transplant date with a functioning graft, whichever came later (designated the index date). Recipients were followed from the index date until the first of death, graft failure (defined as return to maintenance dialysis or re-transplantation), emigration, or end of study period (March 31, 2019). The last eligible transplant date for cohort entry was December 31, 2017 for a last designated index date of December 31, 2018. In Alberta, the rate of emigration is <1%.^14,15 The cohort creation is shown in Supplemental Figure S1, and a schematic of the study design is presented in Supplemental Figure S2.

Proton-Pump Inhibitor/Histamine-2-Receptor Antagonist Prescriptions

We identified transplant recipients who had evidence of a PPI or H2Ra prescription in the 3 months prior to and including their index date. For most recipients, immunosuppression around the first year of transplant, including prednisone, should be at the lowest dose and so PPI and H2Ra use may no longer be indicated. If a patient had been prescribed a PPI prior to this, but was not on a PPI at their index date, they were not considered to be on a PPI.¹⁶ Proton-pump inhibitors of interest included those commonly prescribed in Canada: pantoprazole, omeprazole, lansoprazole, rabeprazole, esomeprazole, and dexlansoprazole. Chronic PPI use was defined as a prescription fill of greater than 8 weeks in duration.³ Histamine-2-receptor antagonists prescribed in Canada included ranitidine, cimetidine, famotidine, and nizatidine. We excluded recipients who had evidence of both PPI and H2Ra prescription fills in the 3 months before their index date to compare mutually exclusive groups (n = 17).

Baseline Characteristics

Baseline characteristics were obtained at the index date. Demographic data such as age and sex were determined from the Alberta Health administrative database. While specific race/ethnic groups cannot be identified in this database, more than 75% of the Alberta population is white.¹⁷ Postal codes were linked to the Canadian Census using the Postal Code Conversion file to determine median neighborhood household income quintile (level 5 being the highest) as well as rural vs urban location of residence. Baseline characteristics were complete except for income and residence location (<1% of recipients). Those with missing data for income were re-categorized into the middle (third) quintile and those with missing data for location of residence were re-categorized as urban.

We used the AKC North and South program databases for transplant-related data such as pre-transplant dialysis modality, dialysis duration, and time since transplant. Graft function at the index date was determined by calculating the estimated glomerular filtration rate (eGFR, based on the 2009 Chronic Kidney Disease Epidemiology Collaboration equation) using the mean of all outpatient serum creatinine measurements in the 3 months prior to and including the index date.¹⁸ Albuminuria was defined by all outpatient random spot urine measurements in the 3 months prior to and including the index date (albumin-creatinine ratio [ACR], protein-creatinine ratio [PCR], or urine dipstick). If multiple albuminuria measurements were available, the median value was calculated. Albuminuria was categorized based on the KDIGO (Kidney Disease: Improving Global Outcomes) Chronic Kidney Disease (CKD) definition as normal/mild (A1: ACR <3 mg/mmol, PCR <150 mg/g, or dipstick negative or trace), moderate (A2: ACR 3-30 mg/mmol, PCR 150-500 mg/g, or dipstick 1+), or heavy (A3: ACR >30 mg/mmol, PCR >500 mg/g, dipstick ≥2+), as previously described.⁹

Comorbidities were identified by the presence of one or more diagnostic codes from April 1994 up to and including the index date, using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and International Statistical Classification of Diseases, Tenth Revision (ICD-10) coding algorithms for physician claims and hospitalization data.^19-21 Comorbidities were defined using validated algorithms, whenever possible (Supplemental Table S2). We identified comorbidities for which PPI use may be indicated including gastroesophageal reflux disease (GERD), peptic ulcer disease, Helicobacter pylori infection, prior upper GIB, and concurrent NSAID, antiplatelet, or anticoagulant use.^22-24

For the index prescription fill, we reported the type of PPI or H2Ra used, the daily dose prescribed, and the prescribing physician specialties. Proton-pump inhibitor dose was categorized as low, standard, or high dose (Supplemental Table S4).²⁵ High-dose PPI use is often indicated in the setting of recent upper GIB and H pylori infection.²⁵ Finally, the PIN database was used to identify relevant baseline medication use, including immunosuppressive agents, antiplatelet, anticoagulant, NSAIDs, and cardiac medications (Supplemental Table S3).

Outcomes

The primary outcome was evidence of a PPI prescription fill in follow-up. At the prescription level, we described patterns of PPI use including the total number of PPI prescriptions in follow-up, type of PPI prescribed, the daily dose prescribed, the median duration of each prescription, the number of chronic PPI prescriptions, and the specialty of the physician prescriber. At the recipient level, we assessed the median number of PPI prescriptions per recipient and how many recipients were using PPI chronically. To assess overall exposure duration, we determined the percentage of graft survival time in follow-up that the recipient was on a PPI defined as: ([total days supplied of PPI]/[total days of graft survival in follow-up]) × 100%. Based on the last prescription fill in follow-up, we also determined the proportion of recipients who had an identifiable gastrointestinal indication for ongoing PPI use. We assessed for the concurrent use of medications in the 3 months prior to the last prescription in follow-up, the presence of GERD, peptic ulcer disease, and/or H pylori infection in the 3 years prior, and any history of upper GIB since 1994.

Statistical Analysis

Descriptive variables were expressed as counts with percentages and medians with interquartile ranges (IQRs), where appropriate. We compared baseline characteristics of recipients with an index prescription for a PPI to those with an H2Ra and neither using χ² and Kruskal-Wallis test, where appropriate. Statistical analyses were performed using R version 4.1.2 (R-project.org). A P-value of <.05 was used to define statistical significance.

Results

Baseline Characteristics

There were 1823 prevalent adult kidney transplant recipients included in the study. Of these, 868 (48%) were on a PPI, 215 (12%) were on an H2Ra, and 740 (41%) were on neither therapy at their index date. The patient characteristics are shown in Table 1. The median age was 53 years (IQR = 41-63) and 36% were female. Median baseline eGFR was 59 mL/min/1.73 m² (IQR = 46-72), and this was not significantly different between the 3 groups. At baseline, there was an indication for PPI use in 55% of the PPI group (P < .001). The recipients on a PPI were older, had more comorbidities, and more likely to have an indication for PPI use compared with the recipients on neither drug at the index date (P < .001). An identifiable indication did not differ between the PPI and H2Ra groups (P = .6). Most initial PPI prescriptions were for pantoprazole (65%), were standard dose (68%), and prescribed by nephrologists (64%) (Supplemental Table S5). All recipients in the H2Ra group were prescribed ranitidine (100%). The PPI group was more likely to have a higher median number of prescriptions compared with the H2Ra and neither groups (18 vs 14 vs 7, respectively, P ≤ .001).

Table 1.

Baseline Characteristics of Kidney Transplant Recipients With a Prescription for a PPI, H2Ra, or Neither at the Index Date.

Characteristic	Total cohort (N = 1823)	PPI user (n = 868)	H2Ra user (n = 215)	Neither (n = 740)	P
Demographics
Age, y	53.3 [41.0-62.5]	55.8 [44.3-63.9]	53.0 [39.9-61.4]	50.2 [37.8-60.7]	<.001
≥65	343 (19)	198 (23)	41 (19)	104 (14)	<.001
Female	660 (36)	306 (35)	76 (35)	278 (38)	.6
Socio-economic status
Lowest quintile	440 (24)	227 (26)	41 (19)	172 (23)	.03
Second quintile	376 (21)	170 (20)	45 (21)	161 (22)
Middle quintile	368 (20)	188 (22)	49 (23)	131 (18)
Fourth quintile	322 (18)	131 (15)	48 (22)	143 (19)
Highest quintile	317 (17)	152 (18)	32 (15)	133 (18)
Urban residence	1647 (90)	778 (90)	199 (93)	670 (91)	.4
Distance to transplant center, km	22.5 [13.1-109.0]	22.5 [13.5-133.1]	24.0 [13.4-60.2]	22.4 [12.5-97.8]	.3
≤50	1230 (67)	570 (66)	150 (70)	510 (69)	.3
50.1-150	182 (10)	90 (10)	22 (10)	70 (9)
150.1-300	180 (10)	93 (11)	25 (12)	62 (8)
>300	231 (13)	115 (13)	18 (8)	98 (13)
Transplant-related characteristics
Pre-transplant modality
Hemodialysis	1058 (58)	552 (64)	119 (55)	387 (52)	<.001
Peritoneal dialysis	459 (25)	206 (24)	62 (29)	191 (26)
Pre-emptive	186 (10)	75 (9)	29 (13)	82 (11)
Missing	120 (7)	35 (4)	5 (2)	80 (11)
Dialysis duration, y	2.4 [1.3-3.9]	2.7 [1.6-4.0]	2.5 [1.2-4.6]	2.0 [1.1-3.3]	<.001
Previous kidney transplant	104 (6)	56 (6)	7 (3)	41 (6)	.2
Transplant era
2002-2006	491 (27)	158 (18)	43 (20)	290 (39)	<.001
2007-2011	532 (29)	286 (33)	60 (28)	186 (25)
2012-2017	800 (44)	424 (49)	112 (52)	264 (36)
Transplant to index date, y	1.0 [1.0-1.5]	1.0 [1.0-1.0]	1.0 [1.0-1.0]	1.0 [1.0-2.8]	<.001
Index date
2008-2010	807 (44)	322 (37)	77 (36)	408 (55)	<.001
2011-2013	329 (18)	186 (21)	34 (16)	109 (15)
2014-2018	687 (38)	360 (41)	104 (48)	223 (30)
Index eGFR,^a mL/min/1.73 m²	58.9 [46.1-72.0]	58.1 [44.3-72.5]	58.3 [48.6-72.6]	59.8 [47.3-71.2]	.3
>90	121 (7)	65 (7)	14 (7)	42 (6)	.2
60-89	717 (39)	333 (38)	83 (39)	301 (41)
30-59	804 (44)	398 (46)	103 (48)	303 (41)
15-29	104 (6)	58 (7)	6 (3)	40 (5)
<15	14 (1)	6 (1)	0	8 (1)
Missing	63 (3)	8 (1)	9 (4)	46 (6)
Albuminuria^a
Normal/mild	1473 (81)	733 (84)	171 (80)	569 (77)	<.001
Moderate	167 (9)	76 (9)	25 (12)	66 (9)
Severe	90 (5)	40 (5)	8 (4)	42 (6)
Missing	93 (5)	19 (2)	11 (5)	63 (9)
Comorbidities^b
Hypertension	1741 (96)	843 (97)	208 (97)	690 (93)	<.001
Diabetes mellitus	676 (37)	375 (43)	75 (35)	226 (31)	<.001
Myocardial infarction	80 (4)	52 (6)	4 (2)	24 (3)	.004
PCI or CABG	144 (8)	100 (12)	9 (4)	35 (5)	<.001
Heart failure	356 (20)	200 (23)	38 (18)	118 (16)	.001
Atrial fibrillation	131 (7)	80 (9)	10 (5)	41 (6)	.005
Stroke or TIA	230 (13)	135 (16)	19 (9)	76 (10)	.001
PVD	269 (15)	165 (19)	34 (16)	70 (9)	<.001
GERD	331 (18)	214 (25)	23 (11)	94 (13)	<.001
Peptic ulcer disease	160 (9)	115 (13)	8 (4)	37 (5)	<.001
Helicobacter pylori	186 (10)	105 (12)	19 (9)	62 (8)	.04
Liver disease	119 (7)	61 (7)	11 (5)	47 (6)	.6
Chronic pulmonary disease	420 (23)	246 (28)	43 (20)	131 (18)	<.001
Dementia	18 (1)	11 (1)	2 (1)	5 (1)	.5
Cancer	285 (16)	151 (17)	35 (16)	99 (13)	.08
Indication for PPI prescription at baseline^c
GERD	151 (8)	106 (12)	7 (3)	38 (5)	<.001
Peptic ulcer disease	46 (3)	34 (4)	2 (1)	10 (1)	.001
Helicobacter pylori	111 (6)	61 (7)	12 (6)	38 (5)	.3
Prior upper GIB	211 (12)	138 (16)	20 (9)	53 (7)	<.001
Single/dual antiplatelet	369 (20)	215 (25)	78 (36)	76 (10)	<.001
Anticoagulant	93 (5)	60 (7)	6 (3)	27 (4)	.003
NSAIDs	25 (1)	16 (2)	2 (1)	7 (1)	.3
High-dose prednisone (>10 mg daily)	110 (6)	54 (6)	19 (9)	37 (5)	.1
Any of the above	807 (44)	474 (55)	113 (53)	220 (30)	<.001
Medications^d
Median number of prescriptions	12 [7-22]	18 [11-29]	14 [10-24]	7 [3-12]	<.001
Immunosuppression use
Prednisone	1218 (67)	696 (80)	164 (76)	358 (48)	<.001
Cyclosporine	68 (4)	42 (5)	2 (1)	24 (3)	.02
Tacrolimus	1247 (68)	617 (71)	159 (74)	471 (64)	.001
MMF	545 (30)	281 (32)	26 (12)	238 (32)	<.001
MPA	648 (36)	330 (38)	122 (57)	196 (26)	<.001
Azathioprine	110 (6)	70 (8)	9 (4)	31 (4)	.002
Sirolimus	119 (7)	46 (5)	25 (12)	48 (6)	.003
Other medications
Single antiplatelet	357 (20)	208 (24)	76 (35)	73 (10)	<.001
Dual antiplatelet^e	12 (1)	7 (1)	2 (1)	3 (0.4)	.5
Combination therapy	16 (1)	9 (1)	1 (0.5)	6 (1)	.7
ACEi/ARB	640 (35)	345 (40)	85 (40)	210 (28)	<.001
Beta-blocker	590 (32)	355 (41)	84 (39)	151 (20)	<.001
Calcium channel blocker	680 (37)	377 (43)	92 (43)	211 (29)	<.001
Diuretic	300 (16)	186 (21)	28 (13)	86 (12)	<.001
Statin	710 (39)	398 (46)	110 (51)	202 (27)	<.001
Fibrate	23 (1)	15 (2)	3 (1)	5 (1)	.2
Ezetimibe	43 (2)	23 (3)	8 (4)	12 (2)	.1

Note. Data are presented as number (%) or median [interquartile range]. The index date is the later of either April 1, 2008 or the 1-year post-transplant date. PPI = prescribed proton-pump inhibitor; H2Ra = histamine-2-receptor antagonist; eGFR = estimated glomerular filtration rate; PCI = percutaneous coronary intervention; CABG = coronary artery bypass graft; TIA = transient ischemic attack; PVD = peripheral vascular disease; GERD = gastroesophageal reflux disease; GIB = gastrointestinal bleed; NSAID = non–steroidal anti-inflammatory drug; MMF = mycophenolate mofetil; MPA = mycophenolic acid; ACEi = angiotensin-converting enzyme inhibitor; ARB = angiotensin II receptor blocker; ACR = albumin-creatinine ratio.

Median values calculated using mean eGFR and median albuminuria (ACR, PCR, or urine dipstick) for each recipient, using all outpatient measurements in the 3-month look-back window prior to and including the index date.

Assessed by the presence of diagnostic or procedural codes, based on validated algorithms, where applicable, from April 1994 up to and including the index date (Supplemental Table S2).

Assessed on the index date with a look-back window of 3 years for GERD, peptic ulcer disease, and Helicobacter pylori, since 1994 for any history of upper GIB, and 3 months for concurrent medication use.

Medications were identified as any prescription filled during the 3-month look-back window prior to and including the index date.

Dual antiplatelet use was defined as evidence of either a combination pill with 2 antiplatelet formulations (eg, ASA-dipyridamole tablet) or prescriptions for 2 different antiplatelets in the 3-month look-back window prior to and including the index date. Combination therapy was defined as evidence of a prescription fill for ≥1 antiplatelet and ≥1 anticoagulant in the 3-month look-back window prior to and including the index date.

PPI Use in Follow-up: Prescription-Level Analysis

The median follow-up time was 5.4 years (IQR = 2.6-9.3). Overall, there were 44 917 unique PPI prescriptions in follow-up with 35 730 (80%) being prescribed to the recipients on a PPI at baseline (Table 2). Pantoprazole was the most prescribed PPI, and nephrologists were the main prescribers of PPIs in all 3 groups.

Table 2.

PPI Prescriptions in Follow-Up in Kidney Transplant Recipients on a PPI, H2Ra, or Neither at Baseline.

Prescription-level analysis	PPI user (n = 35 730)	H2Ra user (n = 1208)	Neither (n = 7979)
Type of PPI
Pantoprazole	22 208 (62)	819 (68)	4090 (51)
Omeprazole	9997 (28)	242 (20)	2822 (35)
Lansoprazole	2239 (6)	104 (9)	413 (5)
Rabeprazole	761 (2)	38 (3)	264 (3)
Esomeprazole	467 (1)	5 (0.4)	377 (5)
Dexlansoprazole	58 (0.2)	0	13 (0.2)
Dose category^a
Low dose	134 (0.4)	3 (0.2)	64 (1)
Standard dose	24 733 (69)	730 (60)	5075 (64)
High dose	10 863 (30)	475 (39)	2840 (36)
Duration
Median days supplied	28 [14-30]	28 [14-30]	30 [14-90]
Chronic prescriptions^b	8251 (23)	212 (18)	2969 (37)
Prescriber
General practice	5477 (15)	413 (34)	1284 (16)
Nephrologist	23 795 (67)	542 (45)	4938 (62)
Gastroenterologist	96 (0.3)	5 (0.4)	259 (3)
Surgery^c	760 (2)	32 (3)	230 (3)
Other	678 (2)	64 (5)	145 (2)
Missing	4924 (14)	152 (13)	1123 (14)

Note. Data are presented as N (%) or median [interquartile range]. Exposure category determined by either PPI, H2Ra, or Neither prescribed in the 3-month look-back window prior to and including the index date. PPI = proton-pump inhibitor; H2Ra = histamine-2-receptor antagonist.

Defined by the National Institute for Health Care and Excellence guidelines (Supplemental Table S4).

Defined as prescriptions with a duration ≥8 weeks.

Includes general surgery, vascular surgery, and urology.

PPI Use in Follow-up: Recipient-Level Analysis

Kidney transplant recipients who were on a PPI at baseline were more likely to be prescribed a PPI in follow-up compared with those who were on an H2Ra or neither (96% vs 27% vs 46%, respectively) (Table 3 and Figure 1). These recipients also had a higher median number of PPI prescriptions (26 vs 7 vs 10, respectively, P < .001). The PPI group spent a significantly higher percentage of their graft survival time in follow-up on a PPI compared to the H2Ra and neither groups (91% vs 22% vs 38%, respectively, P < .001). We identified an indication for ongoing PPI use in 54% of recipients with the most common indication being concurrent antiplatelet use (26%). In follow-up, 341 (46%) recipients in the neither group were prescribed a PPI with an identifiable indication being present in 54% of recipients.

Table 3.

Recipient-Level Analysis of PPI Use in Follow-Up in Kidney Transplant Recipients on a PPI, H2Ra, or Neither at Baseline.

Recipient-level analysis	PPI user (N = 829) (96%)	H2Ra user (N = 58) (27%)	Neither (N = 341) (46%)	P
Median number of PPI prescriptions per recipient	26 [11-51]	7 [3-21]	10 [3-29]	<.001
≥3 PPI prescriptions	779 (94)	44 (76)	257 (75)	<.001
Chronic prescriptions^a	580 (70)	33 (57)	246 (72)	.06
Percent time in follow-up on PPI^b	91 [70-98]	22 [7-43]	38 [8-74]	<.001
Indication for PPI prescription in follow-up^c
GERD	70 (8)	6 (10)	32 (9)	.8
Peptic ulcer disease	17 (2)	1 (2)	6 (2)	.9
Helicobacter pylori	48 (6)	5 (9)	13 (4)	.2
Prior upper GIB	153 (18)	8 (14)	49 (14)	.2
Antiplatelet	214 (26)	18 (31)	80 (23)	.4
Anticoagulant	105 (13)	5 (9)	33 (10)	.3
NSAID	22 (3)	0	14 (4)	.2
High-dose prednisone (>10 mg daily)	28 (3)	5 (9)	29 (9)	<.001
Any of the above	447 (54)	38 (66)	184 (54)	.2

Note. Data are presented as N (%) or median [interquartile range]. Exposure category determined by either PPI, H2Ra, or neither prescribed in the 3-month look-back window prior to and including the index date. All patients included in this table were patients who had at least 1 PPI prescription in follow-up. PPI = proton-pump inhibitor; H2Ra = histamine-2-receptor antagonist; GERD =gastroesophageal reflux disease; GIB = gastrointestinal bleed; NSAID = non–steroidal anti-inflammatory drug.

Defined as prescriptions with a duration ≥8 weeks.

Percentage of graft survival time that the recipient was on a PPI (defined as: total days supplied of PPI/total days of graft survival × 100%) from index date until end of follow-up.

We assessed for the concurrent use of medications in the 3 months prior to the last prescription in follow-up, the presence of GERD, peptic ulcer disease, concurrent medication use and/or Helicobacter pylori infection in the 3 years prior, and any history of upper GIB since 1994.

Figure 1.

PPI use in kidney transplant recipients by exposure group at the index date.

Discussion

In our cohort of 1823 kidney transplant recipients, almost half were prescribed a PPI at baseline and most of these recipients continued their PPI. Proton-pump inhibitor users spent 91% of their graft survival time in follow-up on a PPI over a median of 5 years. Despite this, an indication for prescription was only identified in about half of recipients. We found that nephrologists were the main prescribers of PPIs.

In our cohort, 48% of kidney transplant recipients were prescribed a PPI at baseline which is similar to other studies (45%-54%).^26,27 This likely reflects the degree of comorbidity in transplant recipients as only 8%-10% of the general population are prescribed PPIs.¹ The recipients in the PPI group were more likely to have cardiac and vascular comorbidities compared with the recipients in the H2Ra and neither groups. This may explain why the most common indication was concurrent antiplatelet prescription. Kidney transplant recipients are also at higher risk of GIB. Previously, Sood et al²⁸ reported that the 3-year cumulative incidence rate of upper GIB for kidney transplant recipients was 1.6% compared with 0.2% in the general population in Ontario, Canada. Thus, it is not surprising that many kidney transplant recipients are prescribed PPIs given their cardiovascular comorbidities and risk of GIB.

Kidney transplant recipients are vulnerable to polypharmacy given their underlying multi-morbidity.²⁷ In our cohort, the median number of prescriptions at baseline was 12 (IQR = 7-22), which was similar to other studies.²⁷ In 1 paper, kidney transplant patients who had ≥10 medications self-reported lower scores of physical functioning, social function, energy/fatigue, and higher pain using the Kidney Disease Quality of Life-Short Form questionnaire.⁶ Lower self-reported quality of life has been independently associated with increased mortality in stable kidney transplant recipients.²⁹ In addition, other adverse outcomes have been associated with long-term PPI use in recipients including hip fractures and mortality.⁵ Conflicting evidence for the association with PPI use and acute rejection and/or worsening graft function exist, and studies with longer follow-up are needed.^5,30 Regardless, it is reasonable to reduce polypharmacy in this high-risk population with reassessment of ongoing PPI use as recommended by national campaigns, such as Choosing Wisely Canada.³

In the dialysis population, which has a similarly high prevalence of PPI use, deprescribing initiatives have been shown to be effective.^27,31,32 In a Canadian study by McIntyre et al,³² 67% of patients without an identifiable indication for PPI were able to discontinue therapy with 75% staying off PPI after 6 months. However, deprescribing PPIs can be challenging as it may lead to rebound gastric hypersecretion resulting in recurrent symptoms and/or bleeding ulcers in select patients.³³ Another Canadian study found that only 30% of patients with kidney failure who were prescribed a PPI had an approved indication.³⁴ Unfortunately, almost half of the group who had their PPI discontinued were restarted within 8 weeks due to symptoms and/or GIB.³⁴ In addition, there is limited evidence that deprescribing initiatives actually lead to improved clinical outcomes. Systematic reviews in the elderly general population have suggested that deprescribing initiatives reduce all-cause mortality, but these were based on studies with low-to-moderate quality of evidence.^35,36 The potential benefits of deprescribing unnecessary medications include reduced pill burden, improved quality of life, lower risk of drug interactions, and limited cost to the patient and health care system. In our study, nephrologists were the main prescribers of PPIs and may benefit from validated deprescribing tools to reduce PPI use in kidney transplant recipients.

Our study has several strengths including the use of linked health care databases to study PPI prescriptions in almost 2000 kidney transplant recipients over a follow-up of 5 years. Whereas prior studies have presumed indefinite use and relied on patient self-report or 1-time prescription fill,^15,37 we captured all prescriptions dispensed provincially. We were able to calculate the graft survival time in follow-up spent on a PPI and attempted to identify indications for chronic PPI use to inform potential deprescribing initiatives. However, there are limitations worth noting. Our recipients have access to universal health care coverage which may limit the generalizability of our results in regions with different health care delivery and practices. We captured provincial prescription fills but did not have data on the indication for new or ongoing use. In addition to this, we were not able to account for over-the-counter medication use, such as for H2Ra or NSAID, which may have led to misclassification of the exposure groups and an underestimation of the indication for PPI use. While we attempted to identify common gastrointestinal indications for PPI use, we did not include rare conditions and Barrett’s esophagitis due to concerns about the validity of the diagnostic codes.³⁸ Given the retrospective design, there may have been confounding variables that we were not able to account for. In addition, symptoms suggestive of reflux may not be well coded as the focus of follow-up visits is more likely to focus on kidney transplant. A prospective study would have provided more granularity for PPI indication but would have been time-consuming and costly to conduct.

Conclusions

In conclusion, 48% of kidney transplant recipients are prescribed a PPI at or beyond the 1-year post-transplant date and those who are taking a PPI are likely to spend most of their follow-up and graft survival time on a PPI. Only half of PPI prescriptions in all groups were associated with an identifiable gastroenterological indication. Targeted approaches to reduce pill burden and unnecessary PPI prescriptions may improve quality of life and reduce medication non-adherence in this high-risk population. Future research is needed to determine whether deprescribing initiatives lead to improved outcomes in kidney transplant recipients.

List of Abbreviations

ACR, albumin-to-creatinine ratio; AKDN, Alberta Kidney Disease Network; AKC, Alberta Kidney Care; CKD, chronic kidney disease; eGFR, estimated glomerular filtration rate; GERD, gastroesophageal reflux disease; GIB, gastrointestinal bleed; H2Ra, histamine-2-receptor antagonist; ICD-9-CM, International Classification of Diseases, Ninth Revision, Clinical Modification; ICD-10, International Statistical Classification of Diseases, Tenth Revision; IQR, interquartile range; KDIGO, Kidney Disease: Improving Global Outcomes; NSAID, non–steroidal anti-inflammatory drug; PCR, protein-creatinine ratio; PIN, Pharmaceutical Information Network; PPI, proton-pump inhibitor.

Supplemental Material

sj-docx-1-cjk-10.1177_20543581241228723 – Supplemental material for Proton Pump Inhibitors Use in Kidney Transplant Recipients: A Population-Based Study

Supplemental material, sj-docx-1-cjk-10.1177_20543581241228723 for Proton Pump Inhibitors Use in Kidney Transplant Recipients: A Population-Based Study by James Kiberd, Robert R. Quinn, Pietro Ravani, Krista L. Lentine, Alix Clarke, Rachel Jeong, Labib Faruque and Ngan N. Lam in Canadian Journal of Kidney Health and Disease

Footnotes

Acknowledgements

This study is based in part on data provided by Alberta Health and Alberta Health Services. The interpretation and conclusions contained herein are those of the researchers and do not necessarily represent the views of the Government of Alberta or Alberta Health Services. Neither the Government of Alberta nor Alberta Health or Alberta Health Services express any opinion in relation to this study.

Ethics Approval and Consent to Participate

This study was approved by the research ethics boards at the University of Alberta and the University of Calgary with waiver of patient consent.

Consent for Publication

The authors have consented to publication of this manuscript.

Author Contributions

J.K. and N.N.L. participated in research design. A.C. participated in data analysis. J.K. drafted and revised the manuscript. All authors were involved in data interpretation and final approval of the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

James Kiberd

Robert R. Quinn

Pietro Ravani

Alix Clarke

Ngan N. Lam

Data Availability

The authors are not able to make their data set available to other researchers due to our contractual arrangements with the provincial health ministry (Alberta Health), who is the data custodian. Researchers may make requests to obtain a similar data set at .

Supplemental Material

Supplemental material for this article is available online.

References

Rotman

Bishop

. Proton pump inhibitor use in the U.S. ambulatory setting, 2002-2009. PLoS ONE. 2013;8(2):e56060. doi:10.1371/journal.pone.0056060.

Battistella

Jandoc

McArthur

Garg

. A province-wide, cross-sectional study of demographics and medication use of patients in hemodialysis units across Ontario. Can J Kidney Health Dis. 2018;5:2054358118760832. doi:10.1177/2054358118760832.

Choosing Wisely Canada. Toolkit: Bye-bye, PPI. https://choosingwiselycanada.org/perspective/ppi-toolkit/. Published 2019. Accessed March 20, 2023.

Lyu

Jorgenson

Hansen

Djamali

Astor

. Proton pump inhibitors, but not H2-receptor antagonists, are associated with incident fractures among kidney transplant recipients. Transplantation. 2020;104(12):2609-2615. doi:10.1097/TP.0000000000003178.

Boonpheng

Thongprayoon

Bathini

Sharma

Mao

Cheungpasitporn

. Proton pump inhibitors and adverse effects in kidney transplant recipients: a meta-analysis. World J Transplant. 2019;9(2):35-47. doi:10.5500/wjt.v9.i2.35.

Woźniak

Kolonko

Chudek

Nowak

Farnik

Więcek

. Influence of polypharmacy on the quality of life in stable kidney transplant recipients. Transplant Proc. 2018;50(6):1896-1899. doi:10.1016/j.transproceed.2018.02.128.

Marienne

Laville

Caillard

, et al. Evaluation of changes over time in the drug burden and medication regimen complexity in ESRD patients before and after renal transplantation. Kidney Int Rep. 2021;6(1):128-137. doi:10.1016/j.ekir.2020.10.011.

Hemmelgarn

Clement

Manns

, et al. Overview of the Alberta Kidney Disease Network. BMC Nephrol. 2009;10:30. doi:10.1186/1471-2369-10-30.

Lam

Tonelli

Lentine

, et al. Albuminuria and posttransplant chronic kidney disease stage predict transplant outcomes. Kidney Int. 2017;92(2):470-478. doi:10.1016/j.kint.2017.01.028.

10.

Jeong

Quinn

Lentine

, et al. Outcomes following macrolide use in kidney transplant recipients. Can J Kidney Health Dis. 2019;6:2054358119830706. doi:10.1177/2054358119830706.

11.

Lam

Boyne

Quinn

, et al. Mortality and morbidity in kidney transplant recipients with a failing graft: a matched cohort study. Can J Kidney Health Dis. 2020;7:2054358120908677. doi:10.1177/2054358120908677.

12.

Lam

Klarenbach

Quinn

, et al. Renal function, albuminuria, and the risk of cardiovascular events after kidney transplantation. Transplant Direct. 2018;4(10):e389. doi:10.1097/TXD.0000000000000828.

13.

Jeong

Lentine

Quinn

, et al. NSAID prescriptions in kidney transplant recipients. Clin Transplant. 2021;35(9):e14405. doi:10.1111/ctr.14405.

14.

Statistics Canada. Estimates of the components of international migration, by age and sex, annual. https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1710001401. Published 2017. Accessed May 10, 2021.

15.

Statistics Canada. Estimates of the components of interprovincial migration, by age and sex, annual. https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=1710001501. Published 2017. Accessed May 10, 2021.

16.

Douwes

Gomes-Neto

Eisenga

, et al. Chronic use of proton-pump inhibitors and iron status in renal transplant recipients. J Clin Med. 2019;8:1382. doi:10.3390/jcm8091382.

17.

Statistics Canada. Census profile, 2016 census. https://www12.statcan.gc.ca/census-recensement/2016/dp-pd/prof/details/page.cfm?Lang=E&Geo1=PR&Code1=48&Geo2=PR&Code2=01&SearchText=Alberta&SearchType=Begins&SearchPR=01&B1=Visibleminority&TABID=1&type=1. Published 2017. Accessed May 10, 2021.

18.

Levey

Stevens

Schmid

, et al. A new equation to estimate glomerular filtration rate. Ann Intern Med. 2009;150(9):604-612. doi:10.7326/0003-4819-150-9-200905050-00006.

19.

Quan

Saunders

, et al. Assessing validity of ICD-9-CM and ICD-10 administrative data in recording clinical conditions in a unique dually coded database. Health Serv Res. 2008;43(4):1424-1441. doi:10.1111/j.1475-6773.2007.00822.x.

20.

Tonelli

Wiebe

Fortin

, et al. Methods for identifying 30 chronic conditions: application to administrative data Healthcare Information Systems. BMC Med Inform Decis Mak. 2015;15:31. doi:10.1186/s12911-015-0155-5.

21.

Quan

Khan

Hemmelgarn

, et al. Validation of a case definition to define hypertension using administrative data. Hypertension. 2009;54(6):1423-1428. doi:10.1161/HYPERTENSIONAHA.109.139279.

22.

Arnason

Wells

van Walraven

Forster

. Accuracy of coding for possible warfarin complications in hospital discharge abstracts. Thromb Res. 2006;118(2):253-262. doi:10.1016/j.thromres.2005.06.015.

23.

Lopushinsky

Covarrubia

Rabeneck

Austin

Urbach

. Accuracy of administrative health data for the diagnosis of upper gastrointestinal diseases. Surg Endosc. 2007;21(10):1733-1737. doi:10.1007/s00464-006-9136-1.

24.

Thirumurthi

Desilva

Castillo

Richardson

Abraham

. Identification of Helicobacter pylori infected patients, using administrative data. Aliment Pharmacol Ther. 2008;28:1309-1316. doi:10.1111/j.1365-2036.2008.03845.x.

25.

National Institute for Health and Care Excellence. Gastro-oesophageal reflux disease and dyspepsia in adults: investigation and management (Clinical guideline). www.nice.org.uk/guidance/cg184. Published September 3, 2014. Accessed May 10, 2021.

26.

January

Progar

Nesselhauf

Hagopian

Malone

. Choice of acid suppressant therapy and long-term graft outcomes after kidney transplantation. Pharmacotherapy. 2020;40(11):1082-1088. doi:10.1002/phar.2470.

27.

Oosten

MJM

Van Logtenberg

SJJ

Hemmelder

, et al. Polypharmacy and medication use in patients with chronic kidney disease with and without kidney replacement therapy compared with matched controls. Clin Kidney J. 2021;14(12):2497-2523. doi:10.1093/ckj/sfab120.

28.

Sood

Garg

Bota

, et al. Risk of major hemorrhage after kidney transplantation. Am J Nephrol. 2015;41(1):73-80. doi:10.1159/000371902.

29.

Molnar-Varga

Molnar

Szeifert

, et al. Health-related quality of life and clinical outcomes in kidney transplant recipients. Am J Kidney Dis. 2011;58(3):444-452. doi:10.1053/j.ajkd.2011.03.028.

30.

Flothow

DJG

Suwelack

Pavenstädt

Schütte-Nütgen

Reuter

. The effect of proton pump inhibitor use on renal function in kidney transplanted patients. J Clin Med. 2020;9(1):258. doi:10.3390/jcm9010258.

31.

Strid

Simrén

Björnsson

. Overuse of acid suppressant drugs in patients with chronic renal failure. Nephrol Dial Transplant. 2003;18(3):570-575. doi:10.1093/ndt/18.3.570.

32.

McIntyre

McQuillan

Bell

Battistella

. Targeted deprescribing in an outpatient hemodialysis unit: a quality improvement study to decrease polypharmacy. Am J Kidney Dis. 2017;70(5):611-618. doi:10.1053/j.ajkd.2017.02.374.

33.

Al-Aly

Maddukuri

Xie

. Proton pump inhibitors and the kidney: implications of current evidence for clinical practice and when and how to deprescribe. Am J Kidney Dis. 2020;75(4):497-507. doi:10.1053/j.ajkd.2019.07.012.

34.

Czikk

Parpia

Roberts

Jain

Zimmerman

. De-prescribing proton pump inhibitors in patients with end stage kidney disease: a quality improvement project. Can J Kidney Health Dis. 2022;9:20543581221106244.

35.

Kua

Mak

VSL

Huey Lee

. Health outcomes of deprescribing interventions among older residents in nursing homes: a systematic review and meta-analysis. J Am Med Dir Assoc. 2019;20(3):362-372. doi:10.1016/j.jamda.2018.10.026.

36.

Bloomfield

Greer

Linsky

, et al. Deprescribing for community-dwelling older adults: a systematic review and meta-analysis. J Gen Intern Med. 2020;35(11):3323-3332. doi:10.1007/s11606-020-06089-2.

37.

Lyu

Hansen

Jorgenson

Astor

. Associations between proton pump inhibitor and histamine-2 receptor antagonist and bone mineral density among kidney transplant recipients. Am J Nephrol. 2020;51:433-441. doi:10.1159/000507470.

38.

Jacobson

Gerson

. The inaccuracy of ICD-9-CM Code 530.2 for identifying patients with Barrett’s esophagus. Dis Esophagus. 2014;23(1):1-7. doi:10.1111/j.1442-2050.2007.00800.x.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.13 MB