Abstract
Objectives:
Prophylactic acid suppression with proton pump inhibitors or H2 receptor antagonists is often administered after kidney transplantation. The Association of proton pump inhibitors or H2 receptor antagonists with acute rejection, hypomagnesemia, and graft loss in kidney transplant recipients is not well established.
Material and Methods:
We performed a retrospective cohort study of 302 kidney transplant recipients at one center (57% male; mean age 35.5±11.2 years) with more than 6 months post-transplant follow-up. Recipients were grouped according to gastric acid prophylaxis: only proton pump inhibitors (n=179), only H2 receptor antagonists (n=42), proton pump inhibitors and H2 receptor antagonists (n=55), and nonusers (n=26). The primary outcome was biopsy-proven acute rejection. Graft loss and hypomagnesemia were defined as secondary outcomes.
Results:
Nonusers were younger and mostly under steroid-free immunosuppression compared to other study groups (p=0.030 and p=0.009, respectively). The primary outcome was similar across study groups (p=0.266). Kaplan-Meier analyses also demonstrated similar 10-year graft survival rates: 95.5% for proton pump inhibitors, 97.6% for H2 receptor antagonists, 100% for proton pump inhibitors/H2 receptor antagonists, and 96.2% for nonusers (p=0.275).
Conclusions:
The use of proton pump inhibitors is not associated with acute rejection or graft loss but may cause mild hypomagnesemia in kidney transplant recipients.
Keywords: acute rejection, H2 receptor antagonists, hypomagnesemia, kidney transplantation, proton pump inhibitors
Introduction
Proton pump inhibitors (PPIs) and H2-receptor antagonists (H2RAs) are frequently used after kidney transplantation for prophylaxis or treatment of gastroesophageal reflux disease, dyspepsia, or peptic ulcer disease1. Although the favorable safety profile of these agents led them to become some of the most frequently used drugs, prolonged exposure has been associated with impaired kidney function1, hypomagnesemia2, and other complications, including dementia in the general population. Kidney transplant recipients often have reduced glomerular filtration rates (GFR) compared to the general population and are particularly vulnerable to the nephrotoxic adverse effects of medications.
The mechanism of PPI-induced hypomagnesemia is still uncertain. However, low urine magnesium (Mg) and fractional Mg excretion show a defect in intestinal absorption or increased losses in the gut3,4. The loss of function in TRPM6 due to high intestinal pH may be responsible for PPI-related hypomagnesemia5. One study found that the long-term use of H2RAs was also associated with hypomagnesemia6. Kidney transplant recipients are particularly vulnerable to co-medications that increase the risk of hypomagnesemia because calcineurin inhibitors (CNIs), a mainstay of transplant immunosuppression, are associated with lower serum magnesium levels3. Hypomagnesemia is, in turn, associated with adverse clinical outcomes, including an increased risk of cardiovascular morbidity and mortality7 in the general population, as well as associated with new-onset diabetes after transplantation (NODAT)8.
Kidney transplant recipients receive drugs with narrow therapeutic indices, such as CNIs, mammalian target of rapamycin inhibitors (mTORi) and mycophenolic acid (MPA) derivates. Interactions of PPI and H2RA with these drugs can lead to significant clinical consequences. For example, in some pharmacokinetic studies, PPIs have reduced mycophenolate mofetil (MMF) absorption by suppressing gastric acidification9. In both randomized controlled10 and observational studies11, low serum MPA levels were associated with an increased risk of acute rejection and overall poor allograft outcome. It is also uncertain whether poor allograft survival in kidney transplant recipients receiving PPI is caused by PPI-associated acute interstitial nephritis (AIN)12.
To advance our understanding of the clinical outcomes associated with the use of PPI and/or H2RA in kidney transplantation, we conducted a retrospective cohort study at one center. We compared outcomes including biopsy-proven acute rejection (BPAR), hypomagnesemia and allograft loss in kidney transplant recipients who receive PPI and/or H2RA, compared with no gastric acid prophylaxis.
Materials and Methods
Patients and Study Design
This research was approved by the ethical committee of the Istanbul University School of Medicine Clinical Studies Board (IRB approval number 2011/483–480), complied with the Declaration of Helsinki, and registered with ClinicalTrials.gov (NCT03123796). All patients enrolled in the study provided written informed consent to extract their medical data into the center’s research database.
Patients who underwent kidney transplantation at a tertiary care center between 2000 and 2012 were included in this retrospective, single-center cohort study. Kidney transplant recipients at least 18 years of age who were followed up for longer than 6 months were initially enrolled. Patients who used any form of gastric acid prophylaxis (PPI and/or H2RA) for less than 6 months or had no adequate information regarding the use of these agents were excluded. Also, patients with multi-organ transplantation and systemic severe illnesses (i.e., cancer, overt congestive heart failure, active opportunistic infections) were not included in the study.
In total, 302 kidney transplant recipients [171 (57%) men; 154 (51%) from deceased donors, mean age 35.5±11.2 years] were enrolled. PPI and H2RA use for gastric acid prophylaxis were defined as the use of lansoprazole 30 mg daily or equivalent dose of other PPIs, use of famotidine 40 mg daily, or equivalent dose of other H2RAs, respectively. Kidney transplant recipients were grouped based on their PPI and/or H2RA intake: Only PPI (n=179), only H2RA (n=42), used PPI and H2RA (PPI/H2RA) (n=55), and nonuser groups (n=26). Recipient and donor data (demographic, clinical, and immunologic) were retrieved from medical records, and last follow-up was in January 2017.
Definition of Immunosuppressive Regimens
Induction therapy (ATG Fresenius, 2 mg/kg/day, for 3 to 7 days) was used in all kidney transplant recipients from deceased donors. Patients were categorized based on induction immunosuppressive regimens into three groups: Antithymocyte globulin (ATG), interleukin‐2 receptor blocking antibodies (IL2rAb), and no induction treatment. Induction use in the data is recorded as a binary indication (given or not), but the dose and duration of treatment information are unavailable.
All patients received intraoperative methylprednisolone bolus injection at a dosage of 500 mg, and afterwards were treated by triple maintenance immunosuppressive regimen including a CNI (cyclosporine or tacrolimus), an antiproliferative drug [azathioprine (AZA) or MPA derivates] and prednisolone. Calcineurin inhibitors were initiated 2 days and antiproliferatives one day before living-related and unrelated donor transplantations. Target blood levels of cyclosporine (C0) and tacrolimus after transplantation were 200–300 ng/mL and 8–12 ng/mL for the first three months, and 50–150 ng/mL and 4–8 ng/mL for subsequent months, respectively. MMF and AZA were administered at a dosage of 2 g/day (1440 mg/day for mycophenolate sodium) and 1.5 mg/kg/day, respectively. On postoperative day 1, patients received methylprednisolone beginning with a dose of 120 mg daily, with a rapid taper and reaching to the maintenance dose of 10 mg daily within the first month and 5 mg daily within the first year. Alterations were made in treatment strategies per immunologic risk and post-transplant complications, if necessary.
Maintenance immunosuppressive regimen was defined at 3 months after kidney transplantation. If the maintenance treatment was altered during the follow-up after the first 3 months, the immunosuppressive treatment regimen at the last follow-up was recorded as maintenance treatment. There were no HLA identical transplantations. For some patients, CNIs were decreased, stopped or switched to mTOR inhibitors because of the side effects of CNIs and infections in the long term.
Follow-up Principles
Patients were initially followed at the transplantation clinic at weekly intervals after surgery, and follow-up intervals were increased to one month and then three months. Laboratory data, including levels of serum creatinine, albumin, Mg, tacrolimus, and cyclosporine trough levels, along with urinalysis and complete blood count, were retrieved from patients’ charts. Estimated GFR (eGFR) was calculated using the serum creatinine-based Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation13. Proteinuria was measured with protein to creatinine ratio (UPCR) in spot urine specimens. Serum Mg levels were determined using standard laboratory methods in our center. Mg levels were taken into consideration after at least a 6-month use of any form of gastric acid prophylaxis. For analytic purposes, the mean value of three consecutive measurements of serum Mg between posttransplant 6–24 months was calculated, and hypomagnesemia was defined as a mean Mg level of <0.70 mmol/L. The follow-up period was considered as the time interval between kidney transplantation and the last outpatient visit, graft failure, or death.
Study Outcomes
The primary outcome was the incidence of BPAR. Standard indication for graft biopsy in our center is a ≥25% rise in serum creatinine and/or new-onset ≥1 g/g proteinuria with no apparent cause. Secondary outcomes were graft loss and hypomagnesemia. Graft loss was defined as the return to dialysis, re-transplantation, or allograft removal. Last visit eGFR was also analyzed as an exploratory outcome.
Statistical Analyses
Results are reported as the mean±SD when normally distributed or as the median (interquartile range [IQR]) otherwise. Comparisons of continuous variables between the groups were evaluated by using analysis of variance (ANOVA) or Kruskal-Wallis tests, where appropriate. Differences in the proportions of different patient groups were compared by the chi-squared test or Fisher’s exact test. Logistic regression analyses were performed to delineate predictors of BPAR and hypomagnesemia, which were reported as odds ratio (ORs) and 95% confidence intervals (CIs). Multivariate Cox regression analysis was carried out to determine predictors of graft loss, and results were described as hazard ratios (HRs) and 95% CIs. In regression, variables found to have an effect on the outcomes (p-value of 0.2 or less) were included in multivariable analyses. Statistical analyses were performed using SPSS statistical software (SPSS version 21.0, IBM Corp., Armonk, NY, USA). Kaplan-Meier curves were generated using MedCalc for Windows (MedCalc version 19.0, MedCalc Software, Ostend, Belgium). A p-value of 0.05 or less was considered to be statistically significant.
Results
Overall Characteristics of Patients
Over the study period, 302 kidney transplant recipients (131 women, 171 men) were followed up for a median of 109 (IQR 82–155) months. Median follow-up durations were different among study groups: 91.0 (73–112) months for PPI group, 163.5 (133.7–192) months for H2RA group, 168 (132–233) months for PPI/H2RA group, and 118 (70.5–252.2) months for nonuser group (p<0.001). The mean age of the cohort was 35.5±11.2 years.
Among the cohort, use of PPIs, H2RAs, PPI/H2RAs, and nonusers were identified in 179 (59.3%), 42 (13.9%), 55 (18.2%), and 26 (8.6%) patients, respectively. Patients who received PPIs were younger than those who received H2RAs and combined therapy but similar in age to those with nonusers: 44.4±10.9 in the PPI group, 49.5±10.5 in the H2RA group, 50.9±9.9 in the PPI/H2RA group and 42.9±11.7 in the nonuser group (p=0.030). Deceased donor kidney transplantation rate was lower in the nonuser group [(5), 19.2%] than the PPI, H2RA, and PPI/H2RA groups [90 (50.3%), 22 (52.4%) and 37 (67.3%), respectively; p=0.001]. Donor age, donor sex, primary kidney disease, HLA mismatches, panel reactive antibody (PRA) levels, CNI, and diuretic use were similar among all groups. MMF use was less common in PPI/H2RA group as compared to other groups (p=0.001), while the use of corticosteroids was lower in the nonuser group (p=0.009). Baseline demographic, clinical, and laboratory characteristics of patients are shown in Table 1.
Table 1.
Baseline demographic, clinical, and laboratory characteristics of patients.
| Characteristics | All patients (n=302) |
PPI (n=179) |
H2RA (n=42) |
PPI/H2RA (n=55) |
Nonusers (n= 26) |
P value |
|---|---|---|---|---|---|---|
| General characteristics | ||||||
| Female sex, n (%) | 131 (43.4) | 80 (44.7) | 15 (35.7) | 25 (45.5) | 11 (42.3) | 0.74 |
| Age, years, mean±SD | 46.2±11.1 | 44.4± 10.9 | 49.5±10.5 | 50.9±9.9 | 42.9±11.7 | 0.030 |
| Deceased donor, n (%) | 153 (50.7) | 90 (50.3) | 22 (52.4) | 37 (67.3) | 5 (19.2) | 0.001 |
| Donor age, mean±SD | 40.76±13.5 | 42.52± 13.8 | 37.71±11.9 | 35.24±12.8 | 45.2±11.2 | 0.30 |
| Female donor sex, n (%) | 104 (34.4) | 61 (34.1) | 15 (35.7) | 16 (29.1) | 12 (46.2) | 0.51 |
| Follow-up period (months), median (IQR) | 109 (82–155) |
91 (73–112) |
163.5 (133.7–192) | 168.0 (132–233) |
118 (70.5–252.2) |
<0.001 |
| Primary kidney disease, n (%) | ||||||
| Diabetic nephropathy | 11 (3.6) | 6 (3.4) | 3 (7.1) | 0 (0.0) | 2 (7.7) | 0.50 |
| Hypertensive nephropathy | 15 (5.0) | 7 (3.9%) | 2 (4.8%) | 5 (9.1) | 1 (3.8) | |
| Chronic glomerulonephritis | 61 (20.1) | 53 (29.6) | 9 (21.4) | 11 (20.0) | 8 (30.8) | |
| Chronic pyelonephritis | 22 (7.3) | 15 (8.4) | 4 (9.5) | 1 (1.8) | 2 (7.7) | |
| Polycystic kidney disease | 5 (1.7) | 3 (1.7) | 1 (2.4) | 1 (1.8) | 0 (0) | |
| Amyloidosis | 6 (3.4) | 6 (3.4) | 0 (0) | 5 (9.1) | 2 (7.7) | |
| Unknown | 99 (32.8) | 56 (31.3) | 16 (38.1) | 19 (34.6) | 8 (30.8) | |
| Others | 56 (18.5) | 33 (18.4) | 7 (16.7) | 13 (23.6) | 3 (11.5) | |
| Number of HLA mismatches, n (%) | ||||||
| 0 | 14 (4.6) | 9 (5) | 2 (4.8) | 2 (3.6) | 1 (3.8) | 0.85 |
| 0–5 | 281 (93.0) | 167 (93.3) | 38 (90.5) | 51 (92.7) | 25 (96.2) | |
| 6 | 7 (2.3) | 3 (1.7) | 2 (4.8) | 2 (3.6) | 0 (0) | |
| Pre-transplant PRA ≥10%, n (%) | 9 (3) | 7 (3.9) | 1 (2.4) | 0 (0) | 1 (3.8) | 0.50 |
| Immunosuppressive medications, n (%) | ||||||
| CNIs | 260 (86.1) | 159 (88.8) | 33 (78.6) | 46 (83.6) | 22 (84.6) | 0.33 |
| Mycophenolic acid derivatives | 233 (77.2) | 148 (82.7) | 33 (78.6) | 31 (56.4) | 21 (80.8) | 0.001 |
| Steroids | 286 (94.7) | 172 (96) | 41 (97.6) | 52 (94.5) | 21 (80.8) | 0.009 |
| Diuretic use | 13 (4.3) | 7 (3.9) | 2 (4.8) | 3 (5.5) | 1 (3.8) | 0.96 |
Abbreviations: CNIs, calcineurin inhibitors; HLA, human leukocyte antigen; H2RA, H2-receptor antagonists; IQR, interquartile range; PPIs, proton pump inhibitors; PRA, panel reactive antibodies; SD, standard deviation.
Study Outcomes
The primary outcome (incidence of BPAR) was similar across study groups (16.2% in PPI, 4.8% in H2RA, 12.7% in PPI/H2RA, and 11.5% in nonuser groups; p=0.266). Overall, 13 patients experienced graft loss over 54 (49.5–155) months. Graft loss rates were also similar across study groups (6.1% in PPI, 2.4% in H2RA, 0% in PPI/H2RA, and 3.8% in nonuser groups; p=0.227). Kaplan-Meier analyses revealed that 10-year graft survival rates were 95.5% in PPI, 97.6% in H2RA, 100% in PPI/H2RA, and 96.2% in nonuser groups (p=0.275 with log-rank test) (Figure 1).
Figure 1.
Ten-year graft survival rates were 95.5% in PPI, 97.6% in H2RA, 100% in PPI/H2RA, and 96.2% in nonuser groups (p=0.275 with log-rank test) (H2RA: H2 receptor antagonist, PPI: proton pump inhibitor)
Hypomagnesemia was more common in the PPI group (38.5%) as compared to H2RA (19%), PPI/H2RA (25.5%), and nonuser (30.8%) groups; however, this was not statistically significant (p=0.053). Mean serum Mg levels were similar between groups (p=0.135) (Table 2). Last visit eGFR was 66.9 (45.5–83.2) ml/min/1.73m2 in the PPI group, 62.5 (55.4–77.9) ml/min/1.73m2 in the H2RA group, 57.4 (40.9–71.5) ml/min/1.73m2 in the PPI/H2RA group, and 59.9 (40.8–77.5) ml/min/1.73m2 in the nonuser group (p=0.051).
Table 2.
Laboratory parameters and study outcomes in various groups
| Outcomes | PPI (n=179) |
H2RA (n=42) |
PPI/H2RA (n=55) |
Nonusers (n=26) |
P value |
|---|---|---|---|---|---|
| BPAR, n (%) | 29 (16.2) | 2 (4.8) | 7 (12.7) | 3 (11.5) | 0.266 |
| Graft loss, n (%) | 11 (6.1) | 1 (2.4) | 0 (0) | 1 (3.8) | 0.227 |
| Hypomagnesemia*, n (%) | 69 (38.5) | 8 (19.0) | 14 (25.5) | 8 (30.8) | 0.053 |
| Mg (mmol/l), mean±SD | 0.72±0.11 | 0.76±0.09 | 0.73±0.06 | 0.72±0.073 | 0.135 |
| Last eGFR (ml/min/1.73m2), median (IQR) | 66.9 (45.5–83.2) |
62.5 (55.4–77.9) |
57.4 (40.9–71.5) |
59.9 (40.8–77.5) |
0.051 |
Hypomagnesemia was defined as mean Mg level of <0.70 mmol/L
Abbreviations: BPAR, biopsy-proven acute rejection; eGFR, estimated glomerular filtration rate; H2RA, H2-receptor antagonists; IQR, interquartile range; Mg, magnesium; PPIs, proton pump inhibitors; SD, standard deviation.
Predictors of Primary and Secondary Outcomes
Logistic regression analyses of all patients revealed that only CNI-based immunosuppressive treatment predicted BPAR (OR: 0.347, 95% CI 0.148–0.811, p=0.015) (Table 3), whereas no variable predicted hypomagnesemia (Table 4). In a univariate analysis of all patients, donor age (HR 1.043, 95% CI 1.000–1.089, p=0.049) and the number of HLA mismatches (HR 1.682, 95% CI 1.002–2.823, p=0.049) predicted graft loss. However, these parameters did not predict the graft loss in multivariate Cox regression analysis (Table 5).
Table 3.
Univariate logistic regression analyses regarding biopsy-proven acute rejection in all patients.
| Predictors | Univariate Analysis | Multivariate Analysis | ||
|---|---|---|---|---|
| Odds Ratio (95% CIs) | P value | Odds Ratio (95% CIs) | P value | |
| Recipient age | 1.014 (0.984–1.044) | 0.366 | ||
| Donor age | 1.017 (0.902–1.042) | 0.182 | 1.013 (0.984–1.042) | 0.380 |
| Recipient sex (female) | 0.638 (0.320–1.272) | 0.202 | ||
| Donor sex (female) | 0.867 (0.428–1.756) | 0.693 | ||
| Number of HLA mismatches | 1.210 (0.909–1.609) | 0.191 | 1.215 (0.910–1.620) | 0.186 |
| Donor type (living) | 0.603 (0.306–1.188) | 0.144 | 0.633 (0.302–1.325) | 0.225 |
| PPI use | 2.339 (0.881–6.214) | 0.088 | 2.381 (0.825–6.866) | 0.108 |
| H2RA use | 0.547 (0.250–1.196) | 0.130 | 0.696 (0.288–1.683) | 0.421 |
| Primary kidney disease | ||||
| Diabetic nephropathy | 0.628 (0.078–5.036) | 0.661 | ||
| Hypertensive nephropathy | 2.457 (0.744–8.119) | 0.140 | 2.768 (0.795–9.640) | 0.110 |
| Chronic glomerulonephritis | 1.151 (0.557–2.382) | 0.704 | ||
| Chronic pyelonephritis | 0.618 (0.139–2.748) | 0.527 | ||
| Amyloidosis | 1.166 (0.249–5.458) | 0.846 | ||
| CNI-based immunosuppression | 0.368 (0.167–0.807) | 0.013 | 0.347 (0.148–0.811) | 0.015 |
| Steroid free immunosuppression | 2.439 (0.313–18.977) | 0.394 | ||
Abbreviations: CI, confidence interval; CNI, calcineurin inhibitor; H2RA, H2 receptor antagonists; HLA, human leukocyte antigen; PPI, proton pump inhibitor.
Table 4.
Univariate and multivariate logistic regression analyses regarding hypomagnesemia in all patients.
| Predictors | Univariate Analysis | Multivariate Analysis | ||
|---|---|---|---|---|
| Odds Ratio (95% CIs) | P value | Odds Ratio (95% CIs) | P value | |
| Recipient age | 0.944 (0.581–1.534) | 0.815 | ||
| Donor age | 1.005 (0.987–1.023) | 0.583 | ||
| Recipient sex | 0.944 (0.581–1.534) | 0.815 | ||
| Donor sex | 1.211 (0.734–2.000) | 0.454 | ||
| Number of HLA mismatches | 1.092 (0.893–1.336) | 0.393 | ||
| PPI use | 1.833 (0.986–3.407) | 0.055 | 0.701 (0.362–1.359) | 0.293 |
| H2RA use | 0.490 (0.282–0.853) | 0.012 | 1.772 (0.984–3.190) | 0.057 |
| Diuretic use | 1.806 (0.591–5.525 | 0.300 | ||
| Primary kidney disease | ||||
| Diabetic nephropathy | 0.444 (0.094–2.097) | 0.306 | ||
| Hypertensive nephropathy | 1.027 (0.341–3.089) | 0.963 | ||
| Chronic glomerulonephritis | 0.887 (0.513–1.534) | 0.668 | ||
| Chronic pyelonephritis | 0.754 (0.286–1.990) | 0.569 | ||
| Amyloidosis | 1.297 (0.413–4.071) | 0.656 | ||
| CNI-based immunosuppression | 2.288 (1.017–5.151) | 0.045 | 2.102 (0.925–4.777) | 0.076 |
| Steroid free immunosuppression | 0.610 (0.220–1.688) | 0.341 | ||
Abbreviations: CI, confidence interval; CNI, calcineurin inhibitor; H2RA, H2 receptor antagonists; HLA, human leukocyte antigen; PPI, proton pump inhibitor.
Table 5.
Univariate and multivariate Cox regression analyses regarding graft loss in all patients.
| Predictors | Univariate Analysis | Multivariate Analysis | ||
|---|---|---|---|---|
| Hazard Ratio (95% CIs) | P value | Hazard Ratio (95% CIs) | P value | |
| Recipient age | 0.982 (0.931–1.036) | 0.517 | ||
| Donor age | 1.043 (1.000–1.089) | 0.049 | 1.024 (0.979–1.072) | 0.293 |
| Recipient sex | 1.321 (0.439–3.980) | 0.620 | ||
| Donor sex | 1.252 (0.408–3.836) | 0.694 | ||
| Number of HLA mismatches | 1.682 (1.002–2.823) | 0.049 | 1.475 (0.870–2.503) | 0.149 |
| Donor type | 1.311 (0.416–4.136) | 0.644 | ||
| PPI use | 2.425 (0.524–11.220) | 0.257 | ||
| H2RA use | 0.166 (0.021–1.294) | 0.087 | 0.219 (0.027–1.768) | 0.154 |
| Chronic glomerulonephritis as primary kidney disease | 2.446 (0.797–7.508) | 0.118 | 2.310 (0.735–7.262) | 0.152 |
Abbreviations: CI, confidence interval; H2RA, H2 receptor antagonist; HLA, human leukocyte antigen; PPI, proton pump inhibitor.
Discussion
In this retrospective cohort study, we found that the risks of BPAR and graft loss were similar across patients with and without gastric acid prophylaxis, and hypomagnesemia was slightly increased in kidney transplant recipients who received PPIs. This work adds to the existing literature, in which some prior studies14,15 have found a possible increased risk of acute rejection with PPI use while others have not16,17.
There is a plausible biological mechanism for an association between PPIs and kidney allograft rejection. PPIs may reduce exposure to MPAs through decreased MMF dissolution at higher gastric pH levels. Reduced serum MPA levels can increase rejection rates10,18,19. In vitro studies have shown that MMF tablets completely dissolve at pH 4.0, but only 47% and 13% of the tablet dissolve at pH 5 and 7, respectively20. Although their potencies and duration of action are different, all PPIs have been shown to increase gastric pH levels to above 4.020,21. Therefore, this drug-drug interaction is considered a class effect15. Nevertheless, we found no relationship between PPI use and rejection.
We also did not demonstrate any association between use of H2RA and rejection. Several studies show that even after 5 days of treatment, tolerance to the effects of H2RAs develops and the pharmacological ability to inhibit gastrin secretion reduces22,23. This may explain why H2RA use was not found to be associated with acute rejection. Our center does not routinely perform MPA therapeutic drug monitoring and gastric pH. Therefore, we could not confirm the reduction of MPA exposure due to the co-administration of our patients with PPI.
A few publications have examined associations between PPI use and outcomes among kidney transplant recipients. In a study comparing 125 patients taking pantoprazole with 77 patients using ranitidine, no significant difference was found between the two groups regarding BPAR frequency16. In a comparison of 213 kidney transplant recipients receiving PPIs versus 390 kidney transplant recipients on ranitidine by Knorr et al., BPAR in the first year post-transplant was similar in both groups15. However, PPI intake and rejection rates were associated with African American patients. In another study, BPAR was similar among 183 patients using PPI and 339 patients using H2RAs16. A recently published single-center retrospective analysis of 455 kidney transplant recipients found no significant relationship between PPI use and BPAR over 3.3 years of follow-up24, and a recently published meta-analysis of 6786 kidney transplant recipients revealed similar findings12. Our study is different from these previous reports with well-established control groups such as H2RA, PPI/H2RA, and nonuser groups, and follow-up time is quite longer than these studies.
Conflicting results have been reported in studies examining the relationship between PPI use and hypomagnesemia in kidney transplant recipients. A cohort study of 512 patients found no significant association between PPI use and hypomagnesemia5. On the other hand, in a recent study with 686 stable outpatient kidney transplant recipients, PPI use was associated with lower Mg values and 24-hour urinary Mg excretion. More patients with hypomagnesemia were found in the group using PPI25. A meta-analysis showed a similar relationship between the risk of hypomagnesemia and the use of PPI in kidney transplant recipients12. Our study did not find any relationship between PPI use and hypomagnesemia in the multivariate analysis, even though there was a trend through hypomagnesemia in patients using PPI. This is not quite consistent with studies in the general population and kidney transplant recipients that reported hypomagnesemia related to PPI use26; however, our finding may have been an underestimation considering the number of enrolled patients. Polypharmacy is an increasing problem in kidney transplant recipients. Although PPIs have been used extensively to prevent gastrointestinal complaints and complications of immunosuppressive drugs, particularly corticosteroid therapy, Food and Drug Administration (FDA) guidelines do not recommend PPI use with this indication27. Furthermore, almost two-thirds of these drugs are unnecessarily prescribed28. A precise decision should be made considering the risk-benefit ratio for each patient who is planning to start gastric acid prophylaxis.
Our study has several limitations. First, it has a single-centered retrospective design with an imbalance of patient characteristics across the study groups. A cause-effect relationship cannot be established. Second, donor-specific antibodies, which are known to be closely related to rejections, have not been regularly monitored after transplantation. Third, graft survival rates could have been overestimated due to selection bias. However, the main strength of our study is the long follow-up period. In addition, the study included different groups according to gastric acid prophylaxis.
In conclusion, using PPI in kidney transplant recipients is not associated with BPAR and graft loss but may be associated with mild hypomagnesemia. Further prospective multicenter studies are needed to reveal the outcomes of PPI use in kidney transplant recipients.
Acknoweldgements
K.L.L. was supported by the Mid-America Transplant/Jane A. Beckman Endowed Chair in Transplantation and National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) grant R01DK102981.
Footnotes
Disclosures
Outside of this work, K.L.L. received consulting fees from CareDx and speaker honoaria from Sanofi.
References
- 1.Hess M, Hoenderop J, Bindels R, Drenth J. Systematic review: hypomagnesaemia induced by proton pump inhibition. Alimentary pharmacology & therapeutics. 2012;36(5):405–413. [DOI] [PubMed] [Google Scholar]
- 2.Cheungpasitporn W, Thongprayoon C, Kittanamongkolchai W, et al. Proton pump inhibitors linked to hypomagnesemia: a systematic review and meta-analysis of observational studies. Renal failure. 2015;37(7):1237–1241. [DOI] [PubMed] [Google Scholar]
- 3.Hoorn EJ, van der Hoek J, Rob A, Kuipers EJ, Bolwerk C, Zietse R. A case series of proton pump inhibitor–induced hypomagnesemia. American journal of kidney diseases. 2010;56(1):112–116. [DOI] [PubMed] [Google Scholar]
- 4.Cundy T, Dissanayake A. Severe hypomagnesaemia in long‐term users of proton‐pump inhibitors. Clinical endocrinology. 2008;69(2):338–341. [DOI] [PubMed] [Google Scholar]
- 5.Van Ende C, Van Laecke S, Marechal C, et al. Proton-pump inhibitors do not influence serum magnesium levels in renal transplant recipients. Journal of nephrology. 2014;27(6):707–711. [DOI] [PubMed] [Google Scholar]
- 6.Markovits N, Loebstein R, Halkin H, et al. The association of proton pump inhibitors and hypomagnesemia in the community setting. The Journal of Clinical Pharmacology. 2014;54(8):889–895. [DOI] [PubMed] [Google Scholar]
- 7.Joosten MM, Gansevoort RT, Mukamal KJ, et al. Urinary magnesium excretion and risk of hypertension: the prevention of renal and vascular end-stage disease study. Hypertension. 2013;61(6):1161–1167. [DOI] [PubMed] [Google Scholar]
- 8.Van Laecke S, Van Biesen W, Verbeke F, De Bacquer D, Peeters P, Vanholder R. Posttransplantation hypomagnesemia and its relation with immunosuppression as predictors of new‐onset diabetes after transplantation. American Journal of Transplantation. 2009;9(9):2140–2149. [DOI] [PubMed] [Google Scholar]
- 9.Schaier M, Scholl C, Scharpf D, et al. Proton pump inhibitors interfere with the immunosuppressive potency of mycophenolate mofetil. Rheumatology. 2010;49(11):2061–2067. [DOI] [PubMed] [Google Scholar]
- 10.Van Gelder T, Hilbrands L, Vanrenterghem Y, et al. A randomized double-blind, multicenter plasma concentration controlled study of the safety and efficacy of oral mycophenolate mofetil for the prevention of acute rejection after kidney transplantation1. Transplantation. 1999;68(2):261–266. [DOI] [PubMed] [Google Scholar]
- 11.Van Gelder T, Silva HT, de Fijter JW, et al. Renal transplant patients at high risk of acute rejection benefit from adequate exposure to mycophenolic acid. Transplantation. 2010;89(5):595–599. [DOI] [PubMed] [Google Scholar]
- 12.Boonpheng B, Thongprayoon C, Bathini T, Sharma K, Mao MA, Cheungpasitporn W. Proton pump inhibitors and adverse effects in kidney transplant recipients: A meta-analysis. World journal of transplantation. 2019;9(2):35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Levey AS, Stevens LA, Schmid CH, et al. A new equation to estimate glomerular filtration rate. Annals of internal medicine. 2009;150(9):604–612. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Courson AY, Lee JR, Aull MJ, Lee JH, Kapur S, McDermott JK. Routine prophylaxis with proton pump inhibitors and post‐transplant complications in kidney transplant recipients undergoing early corticosteroid withdrawal. Clinical Transplantation. 2016;30(6):694–702. [DOI] [PubMed] [Google Scholar]
- 15.Knorr JP, Sjeime M, Braitman LE, Jawa P, Zaki R, Ortiz J. Concomitant proton pump inhibitors with mycophenolate mofetil and the risk of rejection in kidney transplant recipients. Transplantation. 2014;97(5):518–524. [DOI] [PubMed] [Google Scholar]
- 16.van Boekel GA, Kerkhofs CH, van de Logt F, Hilbrands LB. Proton pump inhibitors do not increase the risk of acute rejection. The Netherlands journal of medicine. Feb 2014;72(2):86–90. [PubMed] [Google Scholar]
- 17.Rouse GE, Hardinger K, Tsapepas D, Tichy EM. A comparison of histamine receptor antagonists versus proton pump inhibitor gastrointestinal ulcer prophylaxis in kidney transplant recipients. Progress in Transplantation. 2017;27(1):4–9. [DOI] [PubMed] [Google Scholar]
- 18.Cooper M, Deering KL, Slakey DP, et al. Comparing outcomes associated with dose manipulations of enteric-coated mycophenolate sodium versus mycophenolate mofetil in renal transplant recipients. Transplantation. Aug 27 2009;88(4):514–20. doi: 10.1097/TP.0b013e3181b0e65e [DOI] [PubMed] [Google Scholar]
- 19.Langone A, Doria C, Greenstein S, et al. Does reduction in mycophenolic acid dose compromise efficacy regardless of tacrolimus exposure level? An analysis of prospective data from the M ycophenolic R enal T ransplant (MORE) R egistry. Clinical transplantation. 2013;27(1):15–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Kees MG, Steinke T, Moritz S, et al. Omeprazole impairs the absorption of mycophenolate mofetil but not of enteric‐coated mycophenolate sodium in healthy volunteers. The Journal of Clinical Pharmacology. 2012;52(8):1265–1272. [DOI] [PubMed] [Google Scholar]
- 21.Miner P Jr, Katz PO, Chen Y, Sostek M. Gastric acid control with esomeprazole, lansoprazole, omeprazole, pantoprazole, and rabeprazole: a five-way crossover study. The American journal of gastroenterology. 2003;98(12):2616–2620. [DOI] [PubMed] [Google Scholar]
- 22.Lachman L, Howden CW. Twenty-four-hour intragastric pH: tolerance within 5 days of continuous ranitidine administration. The American journal of gastroenterology. 2000;95(1):57–61. [DOI] [PubMed] [Google Scholar]
- 23.Prichard P, Jones D, Yeomans N, Mihaly G, Smallwood R, Louis W. The effectiveness of ranitidine in reducing gastric acid‐secretion decreases with continued therapy. British journal of clinical pharmacology. 1986;22(6):663–668. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Flothow DJG, Suwelack B, Pavenstädt H, Schütte-Nütgen K, Reuter S. The Effect of Proton Pump Inhibitor Use on Renal Function in Kidney Transplanted Patients. J Clin Med. Jan 18 2020;9(1)doi: 10.3390/jcm9010258 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Douwes RM, Gomes-Neto AW, Schutten JC, et al. Proton-Pump Inhibitors and Hypomagnesaemia in Kidney Transplant Recipients. Journal of Clinical Medicine. 2019;8(12):2162. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Nochaiwong S, Ruengorn C, Awiphan R, et al. The association between proton pump inhibitor use and the risk of adverse kidney outcomes: a systematic review and meta-analysis. Nephrology, dialysis, transplantation : official publication of the European Dialysis and Transplant Association - European Renal Association. Feb 1 2018;33(2):331–342. doi: 10.1093/ndt/gfw470 [DOI] [PubMed] [Google Scholar]
- 27.Scarpignato C, Gatta L, Zullo A, Blandizzi C. Effective and safe proton pump inhibitor therapy in acid-related diseases - A position paper addressing benefits and potential harms of acid suppression. BMC medicine. Nov 9 2016;14(1):179. doi: 10.1186/s12916-016-0718-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Heidelbaugh JJ, Goldberg KL, Inadomi JM. Magnitude and economic effect of overuse of antisecretory therapy in the ambulatory care setting. The American journal of managed care. 2010;16(9):e228–34. [PubMed] [Google Scholar]