Advertisement
American Journal of Kidney Diseases

Proton Pump Inhibitor Use and Risk of Hip Fracture in Kidney Transplant Recipients

Published:November 17, 2016DOI:https://doi.org/10.1053/j.ajkd.2016.09.019

      Background

      Posttransplantation bone disease is a significant problem, with few well-evidenced therapeutic options. Proton pump inhibitors (PPIs) are associated with hip fracture in the general population and are widely prescribed for kidney transplant recipients.

      Study Design

      A case-control study.

      Setting & Participants

      From the US Renal Data System, we identified from diagnoses and procedures 231 kidney transplant recipients with a first hip fracture. Cases were matched at the hip fracture index date with 15,575 controls on age, sex, race, and transplantation year.

      Predictor

      PPI use.

      Outcomes

      First hip fracture.

      Results

      In the year prior to the index date, a PPI was prescribed to 65.4% of cases and 57.4% of controls. Additionally, in 34.6% of cases and 28.9% of controls, a PPI was prescribed for >80% of the year preceding the index date (higher PPI users). Unadjusted ORs of hip fracture associated with any and higher PPI use were 1.55 (95% CI, 1.18-2.05) and 1.65 (95% CI, 1.2-2.27), respectively. When adjusted for baseline demographic, clinical, and pharmacologic covariables, any and higher PPI use remained associated with hip fracture, with ORs of 1.39 (95% CI, 1.04-1.84) and 1.41 (95% CI, 1.02-1.95), respectively.

      Limitations

      Potential residual confounding through either incorrectly ascertained or unavailable confounders; cohort limited to Medicare beneficiaries receiving low-income subsidy.

      Conclusions

      In summary, PPI use was associated with hip fracture risk in the US kidney transplant population.

      Index Words

      Hip fracture is associated with increased mortality, decreased mobility, and loss of independence.
      • Hannan E.L.
      • Magaziner J.
      • Wang J.J.
      • et al.
      Mortality and locomotion 6 months after hospitalization for hip fracture: risk factors and risk-adjusted hospital outcomes.
      • Magaziner J.
      • Hawkes W.
      • Hebel J.R.
      • et al.
      Recovery from hip fracture in eight areas of function.
      • Mossey J.M.
      • Mutran E.
      • Knott K.
      • Craik R.
      Determinants of recovery 12 months after hip fracture: the importance of psychosocial factors.
      Patients with end-stage renal disease have a considerably elevated risk for hip fracture compared with the general population,
      • Alem A.M.
      • Sherrard D.J.
      • Gillen D.L.
      • et al.
      Increased risk of hip fracture among patients with end-stage renal disease.
      and the years immediately following kidney transplantation are an especially high-risk period.
      • Ball A.M.
      • Gillen D.L.
      • Sherrard D.
      • et al.
      Risk of hip fracture among dialysis and renal transplant recipients.
      Several factors likely contribute to the risk for hip fracture following successful kidney transplantation and include pre-existing chronic kidney disease−associated mineral bone disease, corticosteroid exposure, and osteoporosis.
      • Kalantar-Zadeh K.
      • Molnar M.Z.
      • Kovesdy C.P.
      • Mucsi I.
      • Bunnapradist S.
      Management of mineral and bone disorder after kidney transplantation.
      Many transplantation centers prescribe proton pump inhibitors (PPIs) as peptic ulcer prophylaxis early posttransplantation. However, a significant proportion of kidney transplant recipients remain on PPI therapy beyond the immediate posttransplantation period. Omeprazole alone is the 7th, 6th, and 5th most prescribed drug in years 1, 2, and 3 posttransplantation, respectively.

      2013 Atlas of CKD & ESRD. http://www.usrds.org/2013/pdf/v2_ch7_13.pdf. Accessed June 20, 2016.

      PPI use has been linked to increased fracture risk in the general population.
      • Khalili H.
      • Huang E.S.
      • Jacobson B.C.
      • Camargo C.A.
      • Feskanich D.
      • Chan A.T.
      Use of proton pump inhibitors and risk of hip fracture in relation to dietary and lifestyle factors: a prospective cohort study.
      • Yang Y.X.
      • Lewis J.D.
      • Epstein S.
      • Metz D.C.
      Long-term proton pump inhibitor therapy and risk of hip fracture.
      However, to our knowledge, no evidence exists specifically linking PPI use and hip fracture risk in the kidney transplantation population. In the present study, we challenged the null hypothesis of no association between PPI use and posttransplantation hip fracture in a contemporary cohort of US kidney transplant recipients.

      Methods

      Source Population and Study Design

      The source population was defined as all first-time kidney transplant recipients recorded and contributing person-time to the US Renal Data System (USRDS) while having a functioning kidney transplant January 1, 2007, to December 31, 2011. The USRDS contains detailed data from the United Network of Organ Sharing and all Medicare claims for eligible patients insured through this federal program. Therein, we conducted this study using a retrospective nested matched case-control design. Hip fracture events were identified using a primary inpatient International Classification of Diseases, Ninth Revision (ICD-9) diagnosis code of 820.xx or 821.xx. We further required that patients have a claim for ICD-9 procedure codes 78.55, 79.15, 79.35, 81.52, 81.51, 79.05, 79.25, or 81.40 within 7 days of hip fracture diagnosis (Table S1, available as online supplementary material). For each case, the index date was defined as the date of hip fracture diagnosis. Controls were matched (without a fixed matching ratio) with at least 1 case at the index date on transplantation vintage (year of transplantation), age (within 3 years), sex, and race (non–African American vs African American). We used this approach in order to be parsimonious and avoid overfitting in our multivariable models because we could not expect to have large numbers of events. Matching on transplantation vintage, age, sex, and race controls for these variables, but precludes estimation of the associations of these factors with the event of interest.
      • Winkelmayer W.C.
      • Glynn R.J.
      • Levin R.
      • Mittleman M.A.
      • Pliskin J.S.
      • Avorn J.
      Late nephrologist referral and access to renal transplantation.
      However, associations of these factors with hip fracture risk are well established and hence it was not an objective of our study to confirm these associations, but rather to eliminate confounding of the association of interest by these factors.
      • Naylor K.L.
      • Li A.H.
      • Lam N.N.
      • Hodsman A.B.
      • Jamal S.A.
      • Garg A.X.
      Fracture risk in kidney transplant recipients: a systematic review.
      • Farmer M.E.
      • White L.R.
      • Brody J.A.
      • Bailey K.R.
      Race and sex differences in hip fracture incidence.
      Controls were eligible to subsequently become cases (ie, conditional risk set matching). We excluded patients with a history of nonkidney solid-organ transplantation or recorded history of prior hip fracture. We required that all cases and controls have Medicare Parts A and B as their primary payer and Medicare Part D prescription drug coverage with the low-income subsidy for at least 1 year prior to the index date.

      Exposure of Interest

      PPI use was the exposure of interest. We identified PPI use in the year prior to the index date using Medicare Part D prescription claims. We defined PPI exposure in 3 different ways: (1) any PPI prescription claim in the 365 days preceding the index date; (2) lesser use, defined as pharmacy-dispensed pills covering <80% of the 365 days preceding the index date; and (3) higher use, defined as pharmacy-dispensed pills covering ≥80% of the 365 days preceding the index date.

      Covariables

      Patient characteristics were abstracted from the USRDS for each patient and included age, sex, race, body mass index, dialysis vintage prior to transplantation, time since transplantation, living (vs deceased) donor transplant, pretransplantation panel-reactive antibody titers (≤80% vs >80%), and acute rejection episodes. We identified the following comorbid conditions that have been shown to be associated with fracture risk (via changes in bone metabolism or fall risk) based on in- and outpatient claims in the year prior to the index date: diabetes mellitus, cardiovascular disease, cerebrovascular disease, arrhythmia, and rheumatologic disease (Table S1
      • Sennerby U.
      • Melhus H.
      • Gedeborg R.
      • et al.
      Cardiovascular diseases and risk of hip fracture.
      • Lipscombe L.L.
      • Jamal S.A.
      • Booth G.L.
      • Hawker G.A.
      The risk of hip fractures in older individuals with diabetes: a population-based study.
      • van Staa T.P.
      • Geusens P.
      • Bijlsma J.W.
      • Leufkens H.G.
      • Cooper C.
      Clinical assessment of the long-term risk of fracture in patients with rheumatoid arthritis.
      ). We used Medicare Parts B and D claims to identify relevant immunosuppression use in the year prior to index: (1) tacrolimus, (2) cyclosporine, (3) mycophenolate mofetil or mycophenolic acid, (4) azathioprine, (5) mammalian target of rapamycin (mTOR) inhibitor (sirolimus or everolimus), and (6) corticosteroid use. We also identified bisphosphonate use in the year prior to the index date using Medicare Part D prescription data because prescription of bisphosphonates may occur in particular high-risk patients and is often accompanied by PPI coprescription.

      Statistical Analysis

      Differences between the case and control groups were evaluated using univariate conditional logistic models after controlling for the case-control matched groups (Table 1). To assess for case-control imbalance, we computed the mean difference between cases and controls within each case-control group for continuous variables and calculated the average proportion of discrepant cases for binary variables (Table S2).
      Table 1Characteristics of Hip Fracture Cases and Their Matched Controls
      VariableCases (n = 231)Controls (n = 15,575)P
      Obtained from a univariate conditional logistic regression model using a complete-case analysis. P value for male sex and African American race cannot be computed because these variables were hard matched.
      Matched
       Age, y
      Controls were matched to ±3 years.
      51.8 ± 12.951.2 ± 10.40.09
       Male sex108 (46.8)9,308 (59.8)
       African American race26 (11.3)1,126 (7.2)
       Time since Tx, y
      Controls were matched to ±1 year.
      6.9 ± 5.34.6 ± 4.00.7
      Not matched
       Hispanic ethnicity42 (19.0)5,143 (33.4)0.0001
      Missing10 (4.3)184 (1.2)
       Body mass index, kg/m226.0 ± 5.027.3 ± 5.20.008
      Missing31 (13.4)1,417 (9.1)
       Time since ESRD diagnosis, y9.9 ± 5.28.1 ± 4.40.1
       Comorbid conditions, recorded history of
      Diabetes mellitus195 (84.4)11,594 (74.4)0.0002
      Cardiovascular disease204 (88.3)11,694 (75.1)<0.0001
      Cerebrovascular disease101 (43.7)3,988 (25.6)<0.0001
      Arrhythmia82 (35.5)4,205 (27.0)0.1
      Rheumatologic disease30 (13.0)1,382 (8.9)0.3
       Tx-related
      Living (vs deceased) donor56 (24.2)3,824 (24.6)0.8
      Acute rejection, history of32 (13.9)1,871 (12.0)0.7
      Missing0 (0.0)22 (0.1)
      PRA > 80%14 (7.9)824 (6.2)0.6
      Missing53 (22.9)2,285 (14.7)
       Immunosuppressant drugs
      Tacrolimus125 (54.1)9,807 (63.0)0.6
      Cyclosporine71 (30.7)2,861 (18.4)0.006
      MMF/mycophenolic acid153 (66.2)11,138 (71.5)0.2
      Azathioprine29 (12.6)551 (3.5)<0.0001
      mTOR inhibitor37 (16.0)1,536 (9.9)0.0003
      Corticosteroid187 (81.0)10,681 (68.6)0.0001
       Bisphosphonate use61 (26.4)2,111 (13.6)<0.0001
      Note: Values for categorical variables are given as count (percent); for continuous variables, as mean ± standard deviation.
      Abbreviations: MMF, mycophenolate mofetil; mTOR, mammalian target of rapamycin; PRA, panel-reactive antibodies; Tx, transplantation.
      a Obtained from a univariate conditional logistic regression model using a complete-case analysis. P value for male sex and African American race cannot be computed because these variables were hard matched.
      b Controls were matched to ±3 years.
      c Controls were matched to ±1 year.
      The association of hip fracture with prior PPI use was estimated using both unadjusted and multivariable conditional (on the matching set of hip fracture event and all matched controls; pair identification) logistic regression and was expressed as odds ratios (ORs) with corresponding 95% confidence intervals (CIs). All baseline variables listed in Table 1 except for those hard matched between cases and controls were included in the multivariable model. We specifically tested for effect modification of the PPI (any use)–hip fracture association by calcineurin inhibitor use given their competing metabolism by the cytochrome P450 3A4 enzyme (CYP3A4) in the liver.
      • Ogawa R.
      • Echizen H.
      Drug-drug interaction profiles of proton pump inhibitors.
      • Maguire M.
      • Franz T.
      • Hains D.S.
      A clinically significant interaction between tacrolimus and multiple proton pump inhibitors in a kidney transplant recipient.
      About 40% of observations had at least 1 variable missing. We assumed the data to be missing at random and performed multiple imputation to generate 40 imputed data sets.
      • Montez-Rath M.E.
      • Winkelmayer W.C.
      • Desai M.
      Addressing missing data in clinical studies of kidney diseases.
      We used a fully conditional specification approach.
      • van Buuren S.
      • Brand J.P.L.
      • Groothuis-Oudshoorn C.G.M.
      • Rubin D.B.
      Fully conditional specification in multivariate imputation.
      The imputation model included all variables plus a fixed effect for matching set to account for the fact that specific controls were matched to specific cases.
      This study was conducted using SAS software (version 9.3; SAS Institute Inc) and StataMP (version 13; Stata Corp) and approved by institutional review boards at Stanford University (IRB-17904) and Baylor College of Medicine (H-36408). The study was granted a waiver of informed consent.

      Results

      For 2007 to 2011, we identified 231 cases of hip fracture that fulfilled the stated inclusion and exclusion criteria that were then matched with 15,575 controls. The number of matched controls per case ranged between 1 and 225, with a median of 56 (interquartile range, 29-93). In terms of unmatched baseline characteristics, cases had a higher prevalence of diabetes mellitus, cardiovascular disease, cerebrovascular disease, arrhythmia, and rheumatologic disease, as well as higher steroid, mTOR inhibitor, cyclosporine, azathioprine, and bisphosphonate use (Table 1). Mean differences in age and time since transplantation between case and controls within case-control matched group were 1.7 year and 0.3 year, respectively, reflecting good matching within the prespecified bounds (3 years for age and 1 year for time since transplantation; Table S2).
      In the year prior to the index date, 65.4% of cases and 57.4% of controls filled a prescription for a PPI; 34.6% of cases and 28.9% of controls were higher PPI users, having filled PPI prescriptions covering at least 292 of the 365 days (>80%) preceding the index date (Table 2).
      Table 2Frequency and Intensity of PPI Use Prior to Index Date
      PPI UseCases (n = 231)Controls (n = 15,575)
      No use80 (34.6)6,628 (42.6)
      Lesser use71 (30.7)4,443 (28.5)
      Higher use80 (34.6)4,504 (28.9)
      Note: Lesser PPI use was defined as <80% of days covered with prescribed PPI in the year prior to index; higher use was ≥80% of days covered.
      Abbreviation: PPI, proton pump inhibitor.
      Unadjusted ORs of hip fracture associated with any, lesser, and higher PPI use compared to no use were 1.55 (95% CI, 1.18-2.05), 1.45 (95% CI, 1.04-2.02), and 1.65 (95% CI, 1.20-2.27), respectively. In adjusted analysis, hip fracture status remained associated with any and high PPI use, with ORs of 1.39 (95% CI, 1.04-1.84) and 1.41 (95% CI, 1.02-1.95), respectively (Fig 1).
      Figure thumbnail gr1
      Figure 1Multivariable association between hip fracture versus control status and prior proton pump inhibitor (PPI) use. Odds ratios and corresponding 95% confidence intervals (CIs) were estimated from case-control sets that were matched on age, sex, race, and time since transplantation using a conditional logistic regression model with adjustment for all other variables listed in . Lesser PPI use was defined as <80% of days covered with prescribed PPI in the year prior to index; higher use was ≥80% of days covered.
      A formal test for interaction between any prior calcineurin inhibitor use and any PPI use was not significant (P > 0.99).

      Discussion

      We used the large nationwide registry of US kidney transplant recipients to examine whether an association existed between use of a PPI, a commonly prescribed drug class in kidney transplant recipients, and hip fracture risk. Clearly, posttransplantation bone disease is complex and multifactorial. However, in PPI therapy, we have identified a potentially avoidable risk factor. Histamine-2 (H2) receptor antagonists such as famotidine offer a viable therapeutic alternative to PPI therapy for many patients and have not been linked to increased fracture risk.
      • Vestergaard P.
      • Rejnmark L.
      • Mosekilde L.
      Proton pump inhibitors, histamine H2 receptor antagonists, and other antacid medications and the risk of fracture.
      • Kwok C.S.
      • Yeong J.K.
      • Loke Y.K.
      Meta-analysis: risk of fractures with acid-suppressing medication.
      A recent study from Denmark reported a significant association between PPI use and increased risk for any fracture in the dialysis and general populations, but found no significant association in the kidney transplantation population (hazard ratio, 1.02; 95% CI, 0.78-1.33).
      • Hansen D.
      • Olesen J.B.
      • Gislason G.H.
      • Abrahamsen B.
      • Hommel K.
      Risk of fracture in adults on renal replacement therapy: a Danish national cohort study.
      However, it appears that only 10% of the 265 fractures were hip fractures. Several other reports have suggested an association between PPI use and fracture risk in the general population.
      • Khalili H.
      • Huang E.S.
      • Jacobson B.C.
      • Camargo C.A.
      • Feskanich D.
      • Chan A.T.
      Use of proton pump inhibitors and risk of hip fracture in relation to dietary and lifestyle factors: a prospective cohort study.
      • Yang Y.X.
      • Lewis J.D.
      • Epstein S.
      • Metz D.C.
      Long-term proton pump inhibitor therapy and risk of hip fracture.
      • Kwok C.S.
      • Yeong J.K.
      • Loke Y.K.
      Meta-analysis: risk of fractures with acid-suppressing medication.
      • Freedberg D.E.
      • Haynes K.
      • Denburg M.R.
      • et al.
      Use of proton pump inhibitors is associated with fractures in young adults: a population-based study.
      Moreover, hip fracture risk appears to increase in a dose-dependent manner with PPI dose and duration of PPI use,
      • Yang Y.X.
      • Lewis J.D.
      • Epstein S.
      • Metz D.C.
      Long-term proton pump inhibitor therapy and risk of hip fracture.
      a criterion that strengthens the argument that the association may be causal.
      • Hill A.B.
      The environment and disease: association or causation?.
      However, the biological mechanisms underlying the PPI–hip fracture association are not well understood. Proposed explanations include the following:
      • 1.
        Treatment with PPIs adversely affects bone health by reducing intestinal calcium absorption. An acidic environment promotes the dissociation of poorly soluble calcium salts (such as calcium carbonate) into water-soluble calcium ions.
        • Goss S.L.
        • Lemons K.A.
        • Kerstetter J.E.
        • Bogner R.H.
        Determination of calcium salt solubility with changes in pH and P(CO(2)), simulating varying gastrointestinal environments.
        The importance of an acidic gastric environment for calcium homeostasis is supported by the findings that patients with achlorhydria have reduced calcium absorption compared with healthy individuals
        • Recker R.R.
        Calcium absorption and achlorhydria.
        and that patients who have undergone gastrectomy have associated increases in parathyroid hormone levels, bone loss, and fracture.
        • Baek K.H.
        • Jeon H.M.
        • Lee S.S.
        • et al.
        Short-term changes in bone and mineral metabolism following gastrectomy in gastric cancer patients.
        • Zittel T.T.
        • Zeeb B.
        • Maier G.W.
        • et al.
        High prevalence of bone disorders after gastrectomy.
        • Mellström D.
        • Johansson C.
        • Johnell O.
        • et al.
        Osteoporosis, metabolic aberrations, and increased risk for vertebral fractures after partial gastrectomy.
        However, results of studies in humans on the effect of PPI therapy on calcium absorption are conflicting. In one study, gastrointestinal absorption of radiolabeled calcium decreased following 1 week of omeprazole (vs placebo) treatment,
        • O'Connell M.B.
        • Madden D.M.
        • Murray A.M.
        • Heaney R.P.
        • Kerzner L.J.
        Effects of proton pump inhibitors on calcium carbonate absorption in women: a randomized crossover trial.
        and in a small group of kidney transplant recipients, omeprazole treatment was associated with reduced urinary calcium excretion.
        • Graziani G.
        • Como G.
        • Badalamenti S.
        • et al.
        Effect of gastric acid secretion on intestinal phosphate and calcium absorption in normal subjects.
        Conversely, a 1-month course of omeprazole treatment was not associated with any change in calcium absorption in 21 healthy postmenopausal women.
        • Hansen K.E.
        • Jones A.N.
        • Lindstrom M.J.
        • et al.
        Do proton pump inhibitors decrease calcium absorption?.
        Furthermore, H2-blocker therapy is not associated with increased fracture risk despite also lowering (albeit likely to a lesser degree) gastric pH, arguing against PPI-associated hypochlorhydria as the primary mechanism behind PPI-associated fracture risk.
        • Vestergaard P.
        • Rejnmark L.
        • Mosekilde L.
        Proton pump inhibitors, histamine H2 receptor antagonists, and other antacid medications and the risk of fracture.
        • Armbrecht U.
        • Abucar A.
        • Hameeteman W.
        • Schneider A.
        • Stockbrügger R.W.
        Treatment of reflux oesophagitis of moderate and severe grade with ranitidine or pantoprazole–comparison of 24-hour intragastric and oesophageal pH.
      • 2.
        In vivo data suggest that PPI treatment may directly impair both osteoblast and osteoclast cell function, resulting in abnormal bone turnover.
        • Costa-Rodrigues J.
        • Reis S.
        • Teixeira S.
        • Lopes S.
        • Fernandes M.H.
        Dose-dependent inhibitory effects of proton pump inhibitors on human osteoclastic and osteoblastic cell activity.
        Indeed, PPIs inhibit the osteoclast vacuolar-type hydrogen adenosine triphosphatase pump in a manner similar to the gastric hydrogen/potassium adenosine triphosphatase pump.
        • Mizunashi K.
        • Furukawa Y.
        • Katano K.
        • Abe K.
        Effect of omeprazole, an inhibitor of H+,K(+)-ATPase, on bone resorption in humans.
        In humans, this is supported by the finding that omeprazole (vs H2-blocker) treatment is associated with decreased excretion of the bone reabsorption marker hydroxyproline.
        • Mizunashi K.
        • Furukawa Y.
        • Katano K.
        • Abe K.
        Effect of omeprazole, an inhibitor of H+,K(+)-ATPase, on bone resorption in humans.
      • 3.
        In chickens, PPI-associated secondary hypergastrinemia has been shown to promote parathyroid gland hypertrophy and hyperplasia.
        • Gagnemo-Persson R.
        • Håkanson R.
        • Sundler F.
        • Persson P.
        Growth of the parathyroid glands in omeprazole-treated chickens.
        This raises the intriguing possibility that PPI-treatment may prolong posttransplantation hyperparathyroidism through gastrin-mediated parathyroid gland stimulation.
      • 4.
        Vitamin B12 deficiency is seen more commonly among PPI users.
        • Lam J.R.
        • Schneider J.L.
        • Zhao W.
        • Corley D.A.
        Proton pump inhibitor and histamine 2 receptor antagonist use and vitamin B12 deficiency.
        Vitamin B12 is a cofactor in homocysteine metabolism and its deficiency can lead to hyperhomocysteinemia. A high homocysteine level can affect collagen crosslinks and interfere with bone remodeling
        • Herrmann M.
        • Widmann T.
        • Herrmann W.
        Homocysteine–a newly recognised risk factor for osteoporosis.
        and has been shown to be associated with fractures.
        • van Meurs J.B.
        • Dhonukshe-Rutten R.A.
        • Pluijm S.M.
        • et al.
        Homocysteine levels and the risk of osteoporotic fracture.
        • Yang J.
        • Hu X.
        • Zhang Q.
        • Cao H.
        • Wang J.
        • Liu B.
        Homocysteine level and risk of fracture: a meta-analysis and systematic review.
      • 5.
        Treatment with PPIs has also been shown to be associated with an increase in fall risk, perhaps as a result of vitamin B12 deficiency–associated weakness and gait instability.
        • Lewis J.R.
        • Barre D.
        • Zhu K.
        • et al.
        Long-term proton pump inhibitor therapy and falls and fractures in elderly women: a prospective cohort study.
      In addition, the calcineurin inhibitors tacrolimus and cyclosporine and PPIs are all metabolized by CYP3A4 in the liver, and increases in cyclosporine and tacrolimus concentrations associated with PPI coadministration are well recognized.
      • Ogawa R.
      • Echizen H.
      Drug-drug interaction profiles of proton pump inhibitors.
      • Maguire M.
      • Franz T.
      • Hains D.S.
      A clinically significant interaction between tacrolimus and multiple proton pump inhibitors in a kidney transplant recipient.
      Although purely speculative, it is possible that PPI levels may also be increased in the setting of concurrent calcineurin inhibitor treatment, potentially exposing transplant recipients to higher than expected concentrations of PPIs. However, a formal test for an interaction of the PPI−hip fracture association by calcineurin inhibitor use was not statistically significant.
      Reported hip fracture incidence in kidney transplant recipients varies according to study population.
      • Ball A.M.
      • Gillen D.L.
      • Sherrard D.
      • et al.
      Risk of hip fracture among dialysis and renal transplant recipients.
      Hip fracture incidence is highest in the early years posttransplantation and diminishes thereafter. Our group has previously reported an unadjusted incidence rate of 3.8 hip fractures/1,000 person-years (60,740 kidney transplants; median follow-up, 3 years) in a US kidney transplant cohort.
      • Sukumaran Nair S.
      • Lenihan C.R.
      • Montez-Rath M.E.
      • Lowenberg D.W.
      • Chertow G.M.
      • Winkelmayer W.C.
      Temporal trends in the incidence, treatment and outcomes of hip fracture after first kidney transplantation in the United States.
      A study from the United Kingdom noted a lower unadjusted event rate for hip fracture of 1.54/1,000-patient years.
      • Ferro C.J.
      • Arnold J.
      • Bagnall D.
      • Ray D.
      • Sharif A.
      Fracture risk and mortality post-kidney transplantation.
      Adult Canadian kidney transplant recipients (transplantations performed 1994-2009) were found to have an incidence rate of 1.9 hip fractures/1,000 person-years, with 3- and 10-year cumulative hip fracture incidences of 0.4% and 1.7%, respectively.
      • Naylor K.L.
      • Zou G.
      • Leslie W.D.
      • et al.
      Risk factors for fracture in adult kidney transplant recipients.
      Data about therapies for prevention of fracture in kidney transplant recipients are limited when compared with those in the general population. Steroid withdrawal is associated with a reduction in fracture risk, and the incidence of early posttransplantation hip fracture has decreased in recent years, corresponding in time to the increased use of steroid-free immunosuppression.
      • Sukumaran Nair S.
      • Lenihan C.R.
      • Montez-Rath M.E.
      • Lowenberg D.W.
      • Chertow G.M.
      • Winkelmayer W.C.
      Temporal trends in the incidence, treatment and outcomes of hip fracture after first kidney transplantation in the United States.
      • Nikkel L.E.
      • Mohan S.
      • Zhang A.
      • et al.
      Reduced fracture risk with early corticosteroid withdrawal after kidney transplant.
      Calcium and calcitriol, bisphosphonate therapy, and the receptor activator of nuclear factor kappa-B ligand (RANKL) inhibitor denosumab have all been shown to reduce measures of posttransplantation bone loss.
      • Josephson M.A.
      • Schumm L.P.
      • Chiu M.Y.
      • Marshall C.
      • Thistlethwaite J.R.
      • Sprague S.M.
      Calcium and calcitriol prophylaxis attenuates posttransplant bone loss.
      • Conley E.
      • Muth B.
      • Samaniego M.
      • et al.
      Bisphosphonates and bone fractures in long-term kidney transplant recipients.
      • Bonani M.
      • Frey D.
      • Brockmann J.
      • et al.
      Effect of twice-yearly denosumab on prevention of bone mineral density loss in de novo kidney transplant recipients: a randomized controlled trial.
      However, few interventional studies in the kidney transplant population have shown an associated reduction in fracture risk. Furthermore, although posttransplantation hyperparathyroidism is associated with increased fracture risk, short-term treatment of posttransplantation hyperparathyroidism with cinacalcet (vs placebo) is not associated with beneficial changes in posttransplantation bone mineral density or bone turnover markers.
      • Perrin P.
      • Caillard S.
      • Javier R.M.
      • et al.
      Persistent hyperparathyroidism is a major risk factor for fractures in the five years after kidney transplantation.
      • Evenepoel P.
      • Cooper K.
      • Holdaas H.
      • et al.
      A randomized study evaluating cinacalcet to treat hypercalcemia in renal transplant recipients with persistent hyperparathyroidism.
      Our study has a number of strengths. We used a large and contemporary cohort of kidney transplant recipients. By using Medicare Parts A, B, and D claims data, we were able to garner important baseline comorbid conditions and prescription claims data, which allowed us to include clinically relevant patient data in our adjusted model. We used a design that controlled for some known risk factors through matching while adjusting for others in the model. This way we were able to maintain model parsimony and avoid model overfitting in a setting of relatively few outcomes.
      Our study also has certain limitations. As with any retrospective study, residual confounding through either incorrectly ascertained or unavailable confounders remains a possibility. For instance, we lacked data pertinent to bone health, including serum parathyroid hormone concentration and bone mineral density. Another potential source of bias in our study is confounding by indication. We specifically included bisphosphonate use in our analysis because we were concerned about the clinical scenario in which a patient with low bone mineral density is started on a bisphosphonate and at the same time or afterward prescribed a PPI because of bisphosphonate-related upper gastrointestinal adverse effects. Additionally, some of the studies that found an association between fracture risk and PPI use were published during the period of our study
      • Yang Y.X.
      • Lewis J.D.
      • Epstein S.
      • Metz D.C.
      Long-term proton pump inhibitor therapy and risk of hip fracture.
      • Vestergaard P.
      • Rejnmark L.
      • Mosekilde L.
      Proton pump inhibitors, histamine H2 receptor antagonists, and other antacid medications and the risk of fracture.
      ; it is possible that physicians who were aware of those studies may have avoided prescribing PPIs in patients at perceived high risk for fracture. However, if anything, this phenomenon would tend to bias our results toward the null.
      We did not determine the dose and type of PPI therapy and lacked data for the original indication and start date for PPI treatment. We also may have missed some PPI users because over-the-counter PPIs have been available in the United States since 2003. However, restriction to Medicare Part D beneficiaries receiving a low-income subsidy minimized this concern because low-income subsidy status confers (almost) first dollar coverage and deincentivizes purchasing drugs over the counter. Although low-income subsidy status alleviated this concern, it raises potential issues of generalizability to more affluent patients or to patients not insured through the Medicare program. Furthermore, we lacked data for: (1) the use of alternative peptic ulcer prophylaxis/anti–gastroesophageal reflux agents; (2) the use of calcium, vitamin D, and other bone health−related treatments; and (3) smoking status.
      Traditionally, PPIs have been regarded as very safe medications. The combination of the high prevalence of upper gastrointestinal symptoms and the perception of PPIs as a benign therapy has undoubtedly contributed to their widespread use in both the general and kidney transplantation populations.

      2013 Atlas of CKD & ESRD. http://www.usrds.org/2013/pdf/v2_ch7_13.pdf. Accessed June 20, 2016.

      • Ekberg H.
      • Kyllönen L.
      • Madsen S.
      • Grave G.
      • Solbu D.
      • Holdaas H.
      Increased prevalence of gastrointestinal symptoms associated with impaired quality of life in renal transplant recipients.
      However, both economic and safety concerns have led to increased focus on their overuse.
      • Forgacs I.
      • Loganayagam A.
      Overprescribing proton pump inhibitors.
      Potential adverse associations are not limited to fracture risk. PPI use has also been linked to the development of infection caused by Clostridium difficile,
      • Dial S.
      • Delaney J.A.
      • Barkun A.N.
      • Suissa S.
      Use of gastric acid-suppressive agents and the risk of community-acquired Clostridium difficile-associated disease.
      community-acquired pneumonia,
      • Lambert A.A.
      • Lam J.O.
      • Paik J.J.
      • Ugarte-Gil C.
      • Drummond M.B.
      • Crowell T.A.
      Risk of community-acquired pneumonia with outpatient proton-pump inhibitor therapy: a systematic review and meta-analysis.
      and chronic kidney disease.
      • Lazarus B.
      • Chen Y.
      • Wilson F.P.
      • et al.
      Proton pump inhibitor use and the risk of chronic kidney disease.
      In summary, we found that PPI use was associated with increased odds of hip fracture in kidney transplant recipients. Our findings argue for a more judicious approach to prescription of PPIs in this population.

      Acknowledgements

      This work was conducted under a data use agreement between Dr Winkelmayer and the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). An NIDDK officer reviewed the manuscript for research compliance and approved of its submission for publication. Data reported herein were supplied by the USRDS. Interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as official policy or interpretation of the US government.
      Support: Dr Lenihan receives grant support through a Mentored Clinical and Population Research Award from the American Heart Association (Western State Affiliate) and a Norman S. Coplon Extramural Grant for Clinical Applied Research from Satellite Healthcare. Dr Sukumaran Nair was funded by a fellowship grant from Sanofi-Aventis. The Stanford Fellowship Program in Nephrology is supported by grant T32 DK007357.
      Financial Disclosure: The authors declare that they have no other relevant financial interests.
      Contributions: Research area and study design: CRL, MEM-R, WCW; data acquisition: MEM-R, WCW: data analysis/interpretation: CRL, SSN, CV, VR, MEM-R, WCW; statistical analysis: MEM-R; supervision or mentorship: WCW. Each author contributed important intellectual content during manuscript drafting or revision and accepts accountability for the overall work by ensuring that questions pertaining to the accuracy or integrity of any portion of the work are appropriately investigated and resolved. WCW takes responsibility that this study has been reported honestly, accurately, and transparently; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned have been explained.
      Peer Review: Evaluated by 4 external peer reviewers, a statistician, and an Acting Editor-in-Chief.

      Supplementary Material

      References

        • Hannan E.L.
        • Magaziner J.
        • Wang J.J.
        • et al.
        Mortality and locomotion 6 months after hospitalization for hip fracture: risk factors and risk-adjusted hospital outcomes.
        JAMA. 2001; 285: 2736-2742
        • Magaziner J.
        • Hawkes W.
        • Hebel J.R.
        • et al.
        Recovery from hip fracture in eight areas of function.
        J Gerontol A Biol Sci Med Sci. 2000; 55: M498-M507
        • Mossey J.M.
        • Mutran E.
        • Knott K.
        • Craik R.
        Determinants of recovery 12 months after hip fracture: the importance of psychosocial factors.
        Am J Public Health. 1989; 79: 279-286
        • Alem A.M.
        • Sherrard D.J.
        • Gillen D.L.
        • et al.
        Increased risk of hip fracture among patients with end-stage renal disease.
        Kidney Int. 2000; 58: 396-399
        • Ball A.M.
        • Gillen D.L.
        • Sherrard D.
        • et al.
        Risk of hip fracture among dialysis and renal transplant recipients.
        JAMA. 2002; 288: 3014-3018
        • Kalantar-Zadeh K.
        • Molnar M.Z.
        • Kovesdy C.P.
        • Mucsi I.
        • Bunnapradist S.
        Management of mineral and bone disorder after kidney transplantation.
        Curr Opin Nephrol Hypertens. 2012; 21: 389-403
      1. 2013 Atlas of CKD & ESRD. http://www.usrds.org/2013/pdf/v2_ch7_13.pdf. Accessed June 20, 2016.

        • Khalili H.
        • Huang E.S.
        • Jacobson B.C.
        • Camargo C.A.
        • Feskanich D.
        • Chan A.T.
        Use of proton pump inhibitors and risk of hip fracture in relation to dietary and lifestyle factors: a prospective cohort study.
        BMJ. 2012; 344: e372
        • Yang Y.X.
        • Lewis J.D.
        • Epstein S.
        • Metz D.C.
        Long-term proton pump inhibitor therapy and risk of hip fracture.
        JAMA. 2006; 296: 2947-2953
        • Winkelmayer W.C.
        • Glynn R.J.
        • Levin R.
        • Mittleman M.A.
        • Pliskin J.S.
        • Avorn J.
        Late nephrologist referral and access to renal transplantation.
        Transplantation. 2002; 73: 1918-1923
        • Naylor K.L.
        • Li A.H.
        • Lam N.N.
        • Hodsman A.B.
        • Jamal S.A.
        • Garg A.X.
        Fracture risk in kidney transplant recipients: a systematic review.
        Transplantation. 2013; 95: 1461-1470
        • Farmer M.E.
        • White L.R.
        • Brody J.A.
        • Bailey K.R.
        Race and sex differences in hip fracture incidence.
        Am J Public Health. 1984; 74: 1374-1380
        • Sennerby U.
        • Melhus H.
        • Gedeborg R.
        • et al.
        Cardiovascular diseases and risk of hip fracture.
        JAMA. 2009; 302: 1666-1673
        • Lipscombe L.L.
        • Jamal S.A.
        • Booth G.L.
        • Hawker G.A.
        The risk of hip fractures in older individuals with diabetes: a population-based study.
        Diabetes Care. 2007; 30: 835-841
        • van Staa T.P.
        • Geusens P.
        • Bijlsma J.W.
        • Leufkens H.G.
        • Cooper C.
        Clinical assessment of the long-term risk of fracture in patients with rheumatoid arthritis.
        Arthritis Rheum. 2006; 54: 3104-3112
        • Ogawa R.
        • Echizen H.
        Drug-drug interaction profiles of proton pump inhibitors.
        Clin Pharmacokinet. 2010; 49: 509-533
        • Maguire M.
        • Franz T.
        • Hains D.S.
        A clinically significant interaction between tacrolimus and multiple proton pump inhibitors in a kidney transplant recipient.
        Pediatr Transplant. 2012; 16: E217-E220
        • Montez-Rath M.E.
        • Winkelmayer W.C.
        • Desai M.
        Addressing missing data in clinical studies of kidney diseases.
        Clin J Am Soc Nephrol. 2014; 9: 1328-1335
        • van Buuren S.
        • Brand J.P.L.
        • Groothuis-Oudshoorn C.G.M.
        • Rubin D.B.
        Fully conditional specification in multivariate imputation.
        J Stat Comput Simulation. 2006; 76: 1049-1064
        • Vestergaard P.
        • Rejnmark L.
        • Mosekilde L.
        Proton pump inhibitors, histamine H2 receptor antagonists, and other antacid medications and the risk of fracture.
        Calcif Tissue Int. 2006; 79: 76-83
        • Kwok C.S.
        • Yeong J.K.
        • Loke Y.K.
        Meta-analysis: risk of fractures with acid-suppressing medication.
        Bone. 2011; 48: 768-776
        • Hansen D.
        • Olesen J.B.
        • Gislason G.H.
        • Abrahamsen B.
        • Hommel K.
        Risk of fracture in adults on renal replacement therapy: a Danish national cohort study.
        Nephrol Dial Transplant. 2016; 31: 1654-1662
        • Freedberg D.E.
        • Haynes K.
        • Denburg M.R.
        • et al.
        Use of proton pump inhibitors is associated with fractures in young adults: a population-based study.
        Osteoporos Int. 2015; 26: 2501-2507
        • Hill A.B.
        The environment and disease: association or causation?.
        Proc R Soc Med. 1965; 58: 295-300
        • Goss S.L.
        • Lemons K.A.
        • Kerstetter J.E.
        • Bogner R.H.
        Determination of calcium salt solubility with changes in pH and P(CO(2)), simulating varying gastrointestinal environments.
        J Pharm Pharmacol. 2007; 59: 1485-1492
        • Recker R.R.
        Calcium absorption and achlorhydria.
        N Engl J Med. 1985; 313: 70-73
        • Baek K.H.
        • Jeon H.M.
        • Lee S.S.
        • et al.
        Short-term changes in bone and mineral metabolism following gastrectomy in gastric cancer patients.
        Bone. 2008; 42: 61-67
        • Zittel T.T.
        • Zeeb B.
        • Maier G.W.
        • et al.
        High prevalence of bone disorders after gastrectomy.
        Am J Surg. 1997; 174: 431-438
        • Mellström D.
        • Johansson C.
        • Johnell O.
        • et al.
        Osteoporosis, metabolic aberrations, and increased risk for vertebral fractures after partial gastrectomy.
        Calcif Tissue Int. 1993; 53: 370-377
        • O'Connell M.B.
        • Madden D.M.
        • Murray A.M.
        • Heaney R.P.
        • Kerzner L.J.
        Effects of proton pump inhibitors on calcium carbonate absorption in women: a randomized crossover trial.
        Am J Med. 2005; 118: 778-781
        • Graziani G.
        • Como G.
        • Badalamenti S.
        • et al.
        Effect of gastric acid secretion on intestinal phosphate and calcium absorption in normal subjects.
        Nephrol Dial Transplant. 1995; 10: 1376-1380
        • Hansen K.E.
        • Jones A.N.
        • Lindstrom M.J.
        • et al.
        Do proton pump inhibitors decrease calcium absorption?.
        J Bone Miner Res. 2010; 25: 2786-2795
        • Armbrecht U.
        • Abucar A.
        • Hameeteman W.
        • Schneider A.
        • Stockbrügger R.W.
        Treatment of reflux oesophagitis of moderate and severe grade with ranitidine or pantoprazole–comparison of 24-hour intragastric and oesophageal pH.
        Aliment Pharmacol Ther. 1997; 11: 959-965
        • Costa-Rodrigues J.
        • Reis S.
        • Teixeira S.
        • Lopes S.
        • Fernandes M.H.
        Dose-dependent inhibitory effects of proton pump inhibitors on human osteoclastic and osteoblastic cell activity.
        FEBS J. 2013; 280: 5052-5064
        • Mizunashi K.
        • Furukawa Y.
        • Katano K.
        • Abe K.
        Effect of omeprazole, an inhibitor of H+,K(+)-ATPase, on bone resorption in humans.
        Calcif Tissue Int. 1993; 53: 21-25
        • Gagnemo-Persson R.
        • Håkanson R.
        • Sundler F.
        • Persson P.
        Growth of the parathyroid glands in omeprazole-treated chickens.
        Scand J Gastroenterol. 1994; 29: 493-497
        • Lam J.R.
        • Schneider J.L.
        • Zhao W.
        • Corley D.A.
        Proton pump inhibitor and histamine 2 receptor antagonist use and vitamin B12 deficiency.
        JAMA. 2013; 310: 2435-2442
        • Herrmann M.
        • Widmann T.
        • Herrmann W.
        Homocysteine–a newly recognised risk factor for osteoporosis.
        Clin Chem Lab Med. 2005; 43: 1111-1117
        • van Meurs J.B.
        • Dhonukshe-Rutten R.A.
        • Pluijm S.M.
        • et al.
        Homocysteine levels and the risk of osteoporotic fracture.
        N Engl J Med. 2004; 350: 2033-2041
        • Yang J.
        • Hu X.
        • Zhang Q.
        • Cao H.
        • Wang J.
        • Liu B.
        Homocysteine level and risk of fracture: a meta-analysis and systematic review.
        Bone. 2012; 51: 376-382
        • Lewis J.R.
        • Barre D.
        • Zhu K.
        • et al.
        Long-term proton pump inhibitor therapy and falls and fractures in elderly women: a prospective cohort study.
        J Bone Miner Res. 2014; 29: 2489-2497
        • Sukumaran Nair S.
        • Lenihan C.R.
        • Montez-Rath M.E.
        • Lowenberg D.W.
        • Chertow G.M.
        • Winkelmayer W.C.
        Temporal trends in the incidence, treatment and outcomes of hip fracture after first kidney transplantation in the United States.
        Am J Transplant. 2014; 14: 943-951
        • Ferro C.J.
        • Arnold J.
        • Bagnall D.
        • Ray D.
        • Sharif A.
        Fracture risk and mortality post-kidney transplantation.
        Clin Transplant. 2015; 29: 1004-1012
        • Naylor K.L.
        • Zou G.
        • Leslie W.D.
        • et al.
        Risk factors for fracture in adult kidney transplant recipients.
        World J Transplant. 2016; 6: 370-379
        • Nikkel L.E.
        • Mohan S.
        • Zhang A.
        • et al.
        Reduced fracture risk with early corticosteroid withdrawal after kidney transplant.
        Am J Transplant. 2012; 12: 649-659
        • Josephson M.A.
        • Schumm L.P.
        • Chiu M.Y.
        • Marshall C.
        • Thistlethwaite J.R.
        • Sprague S.M.
        Calcium and calcitriol prophylaxis attenuates posttransplant bone loss.
        Transplantation. 2004; 78: 1233-1236
        • Conley E.
        • Muth B.
        • Samaniego M.
        • et al.
        Bisphosphonates and bone fractures in long-term kidney transplant recipients.
        Transplantation. 2008; 86: 231-237
        • Bonani M.
        • Frey D.
        • Brockmann J.
        • et al.
        Effect of twice-yearly denosumab on prevention of bone mineral density loss in de novo kidney transplant recipients: a randomized controlled trial.
        Am J Transplant. 2016; 16: 1882-1891
        • Perrin P.
        • Caillard S.
        • Javier R.M.
        • et al.
        Persistent hyperparathyroidism is a major risk factor for fractures in the five years after kidney transplantation.
        Am J Transplant. 2013; 13: 2653-2663
        • Evenepoel P.
        • Cooper K.
        • Holdaas H.
        • et al.
        A randomized study evaluating cinacalcet to treat hypercalcemia in renal transplant recipients with persistent hyperparathyroidism.
        Am J Transplant. 2014; 14: 2545-2555
        • Ekberg H.
        • Kyllönen L.
        • Madsen S.
        • Grave G.
        • Solbu D.
        • Holdaas H.
        Increased prevalence of gastrointestinal symptoms associated with impaired quality of life in renal transplant recipients.
        Transplantation. 2007; 83: 282-289
        • Forgacs I.
        • Loganayagam A.
        Overprescribing proton pump inhibitors.
        BMJ. 2008; 336: 2-3
        • Dial S.
        • Delaney J.A.
        • Barkun A.N.
        • Suissa S.
        Use of gastric acid-suppressive agents and the risk of community-acquired Clostridium difficile-associated disease.
        JAMA. 2005; 294: 2989-2995
        • Lambert A.A.
        • Lam J.O.
        • Paik J.J.
        • Ugarte-Gil C.
        • Drummond M.B.
        • Crowell T.A.
        Risk of community-acquired pneumonia with outpatient proton-pump inhibitor therapy: a systematic review and meta-analysis.
        PLoS One. 2015; 10: e0128004
        • Lazarus B.
        • Chen Y.
        • Wilson F.P.
        • et al.
        Proton pump inhibitor use and the risk of chronic kidney disease.
        JAMA Intern Med. 2016; 176: 238-246