Outpatient Management of Delayed Graft Function Is Associated With Reduced Length of Stay Without an Increase in Adverse Events

Transcription

1 American Journal of Transplantation 2016; 16: Wiley Periodicals Inc. Brief Communication Copyright 2016 The American Society of Transplantation and the American Society of Transplant Surgeons doi: /ajt Outpatient Management of Delayed Graft Function Is Associated With Reduced Length of Stay Without an Increase in Adverse Events B. L. Muth 1, *, B. C. Astor 1,2, J. Turk 1, M. Mohamed 1, S. Parajuli 1, D. B. Kaufman 3, D. A. Mandelbrot 1 and A. Djamali 1,3 1 Division of Nephrology, Department of Medicine, University of Wisconsin School of Medicine and Public Health, Madison, WI 2 Department of Population Health Sciences, University of Wisconsin School of Medicine and Public Health, Madison, WI 3 Division of Transplantation, Department of Surgery, University of Wisconsin School of Medicine and Public Health, University of Wisconsin Hospital and Clinics, Madison, WI *Corresponding author: Brenda Muth, blm@medicine.wisc.edu Delayed graft function () is a common and costly complication of kidney transplantation. In July 2011, we established a multidisciplinary clinic managed by nurse practitioners to facilitate early discharge and intensive management of in the outpatient setting. We compared length of stay, 30- day readmission, acute rejection, and patient/graft survival in 697 consecutive deceased donor kidney transplantations performed between July 2009 and July Patients were divided into three groups: no (n = 487), before implementation of the clinic (n = 118), and clinic (n = 92). Baseline characteristics including age, gender, panel reactive antibody, retransplantation rates, HLA mismatches, induction, and maintenance immunosuppression were not significantly different between pre- and post- clinic groups. Length of stay was significantly longer in pre- clinic ( vs days, p < 1). Thirty-day readmission (21% vs. 16%), graft loss (7% vs. 20%), and patient death (2% vs. 11%) did not differ significantly between pre- and post- clinic. Patients in the clinic were less likely to develop acute rejection (21% vs. 40%, p = 6). Outpatient management of in a specialized clinic is associated with substantially shorter hospitalization and lower incidence of acute rejection without significant difference in 30-day readmission or patient and graft survival. Abbreviations: APP, advanced practice provider; ATG, antithymocyte globulin; CIT, cold ischemia time; DCD, donor after circulatory death; DD, deceased donor kidney transplantation;, delayed graft function; ECD, expanded criteria donor; egfr, estimated glomerular filtration rate; ESRD, end-stage renal disease; LOS, length of stay; PRA, panel reactive antibody; UPC, urine protein:creatinine ratio Received 26 July 2015, revised 20 November 2015 and accepted for publication 14 December 2015 Introduction Delayed graft function (), generally defined as the need for dialysis within the first week after transplantation, is a common occurrence after deceased donor kidney transplantation (DD). The incidence of ranges from 20% to 40% depending on the cause of donor death and donor quality (1,2) and has remained relatively constant over time. This rate is influenced by better immunologic management of recipients as well as a greater acceptance of both older donors and donors with a higher terminal serum creatinine level (2). Patients with have a higher risk of rejection and therefore require close monitoring of graft function and immunosuppression (3 6). is also associated with a higher risk of graft loss and death with a functioning graft (5 8). Patients with are often medically complex and have a high likelihood for readmission within the first 30 days after transplantation (8). Options for managing these patients have traditionally been limited to (i) prolonged hospitalization until they demonstrate allograft recovery or (ii) discharge to their home dialysis unit with periodic follow-up in the outpatient transplant clinic. Once discharged to the home dialysis unit, these patients may be overdialyzed as they begin to recover, and their volume status may be treated inappropriately. Opportunities for close immunosuppression management, monitoring of graft function, and timely biopsy may also be missed. Transplant nephrology physicians and advanced practice providers (APPs), however, are uniquely positioned to manage these patients in the outpatient setting. We created a multidisciplinary clinic in July 2011 for outpatient management of patients. Experienced nurse practitioners in the field of transplantation were selected to manage the clinic. We 1604

2 Outpatient Management of hypothesized that early discharge and outpatient management of patients with in a specialized clinic would be associated with reduced length of stay (LOS) without increasing readmission or compromising clinical outcomes. Methods Study population We performed a retrospective review of data from 697 consecutive patients who underwent DD between July 2009 and July 2014 at the University of Wisconsin. We compared the outcomes of patients without, those with not managed in the clinic (pre- clinic), and those with who were managed in the clinic (post- clinic). Primary end points were 30-day readmission, acute rejection, kidney allograft loss, and patient death. Recipient characteristics were collected from the Wisconsin Allograft Recipient Database, maintained by the Division of Transplantation and Department of Medicine in the University of Wisconsin School of Medicine and Public Health. The study was approved by the University of Wisconsin Health Sciences Institutional Review Board. Variables and definitions was defined as dialysis in the first week after transplantation. Recipient data included age, gender, ethnicity, body mass index (BMI), pretransplant dialysis, cause of end-stage renal disease (ESRD), donor after circulatory death (DCD), expanded criteria donor (ECD), cold ischemia time (CIT), panel reactive antibody (PRA), prior transplants, induction agent, HLA mismatch, and discharge tacrolimus level. Outcomes of interest included LOS, 30-day readmission, cause of readmission, graft loss, death, and serum creatinine, estimated glomerular filtration rate (egfr) by Modification in Diet for Renal Disease Study equation, and urine protein:creatinine ratio (UPC) at 1 year posttransplantation. All biopsies were indication biopsies; no protocol biopsies were performed. There was no difference in indication for dialysis between pre- clinic patients and post- clinic patients. Immunosuppression All patients received induction with either T cell depletion with antithymocyte globulin (ATG: Thymoglobulin â ) or with interleukin-2 blockade (basiliximab) and were maintained on triple immunosuppressive therapy with prednisone, mycophenolic acid, and tacrolimus as described previously (9,10). Tacrolimus is usually started on postoperative day 1 and titrated to a target trough concentration of 8 10 ng/ml. Delayed graft function clinic (Figure 1) Patients with are seen in consultation early postoperatively by the transplant nephrology team, usually by the transplant APPs. Patients are discharged from the University of Wisconsin Hospital and Clinics to their home or, if they live >40 miles away, to a nearby hotel (Figure 1). They must have a support person stay with them during this time. They are scheduled for a morning clinic appointment with the APP within 1 3 days of discharge, depending on estimated dialysis needs. The patient is given a log for recording daily weight, blood pressure, heart rate, 24-h urine output, and glucose. Any patient receiving insulin therapy is also given the contact information of the diabetes management service. Before each clinic visit, blood is drawn in the central hospital laboratory. Laboratory values measured three times per week include a complete blood count with differential, basic metabolic panel, calcium, b 2 -microglobulin, trough tacrolimus level, urinalysis, and UPC. Phosphorus is monitored weekly. The patient is then seen by the transplant APP, who reviews the patient s recent history, medications, vital signs, urine output, and glucose; performs a physical examination; adjusts immunosuppression, antihypertensive drugs, and diuretics; and manages anemia, pain, nausea, and other pertinent health problems. If indicated, the APP orders dialysis and the patient reports to the inpatient unit for dialysis as scheduled. The transplant APP also consults with the transplant surgeon if wound or surgical complications are identified. The diabetes service APPs adjust insu- Hospital Discharge Discharge to hotel with support person Meal vouchers Scheduled for clinic visit within 1 to 3 days after discharge Pre Clinic Visit Labs (CBC, BMP, calcium, ß 2 -microglobulin, urinalysis, UPC) Weight, vital signs, medication review, review of systems Clinic Visit Physical exam, and medical and immunosuppressive management with APP Meet with Pharmacy, Social Worker, Transplant Coordinator and Diabetes service as needed Post Clinic Visit Transplant clinic check out, and review of any treatment changes Schedule next clinic visit Dialysis Schedule for Dialysis in the inpatient dialysis unit later that same day if indicated Biopsy Schedule for transplant kidney biopsy within 7 to 14 days after transplant if no improvement in graft function CBC, complete blood count; BMP, basic metabolic panel; UPC, urine protein to creatinine ratio; APP, advanced practice provider Figure 1: clinic flow., delayed graft function; CBC, complete blood count; BMP, basic metabolic panel; UPC, urine protein: creatinine ratio; APP, advanced practice provider. American Journal of Transplantation 2016; 16:

3 Muth et al lin either by contacting the patient directly via telephone or by personally seeing the patient in clinic. A social worker meets with the patient at the first clinic visit and is available for subsequent visits as needed. The pharmacist reviews medications with the patient, ensures his or her medication box is set up appropriately, and reinforces medication teaching that occurred during the transplant hospitalization. Patients are welcome to attend education classes that are available on the inpatient transplant service. The transplant coordinator sees the patient at the end of the first clinic visit to reinforce any changes in therapy that were made and is available to the patient for subsequent clinic visits. Patients are scheduled for biopsy in the clinic between day 7 and day 14 if no evidence of improving graft function; biopsies are performed by the transplant nephrologist in the same transplant clinic. At the end of their clinic visit, the patient and their support person are given a voucher for a meal in the cafeteria. Clinic July Deceased Donor Kidney transplants (DD) July 2009 N=210 DD N=697 Pre Clinic N= 118 Post Clinic N= 92 No N=487 Statistical analysis Categorical variables were compared across groups by v 2 tests, and continuous variables were compared by Student t-tests. Multivariate logistic regression models were developed to assess the independent associations between baseline characteristics and incidence of. Patient and kidney survival and time to allograft rejection were analyzed using Kaplan Meier survival analysis and Cox proportional hazards regression models. STATA/MP 12.1 ( was used for all analyses. Results Demographics and baseline characteristics A total of 697 patients underwent DD between July 2009 and July Of these, 210 (30.1%) developed (Figure 2). One hundred eighteen patients with were seen before establishment of the clinic, and July 2014 Figure 2: Study population description. DD, deceased donor kidney transplantation;, delayed graft function. 92 were seen after establishment of the clinic (Table 1). Patients with were less likely to be female (31.0% vs. 41.8%, p = 9), receive a preemptive transplant (4.3% vs. 18.9%, p < 1), and more likely to have higher BMI ( vs kg/m 2, p = 1) and to receive a kidney from a DCD (56% vs. 19%, p < 1) or an ECD (26% vs. 17%, p = 0.04). There was no difference in age, ethnicity, cause of kidney disease, CIT, PRA, prior transplantation, or use of ATG induction. Pre- and post- clinic patients were similar in terms of age, gender, ethnicity, BMI, pretrans- Table 1: Baseline characteristics No Pre- clinic Post- clinic p* No Age (mean SD) Female (%) 204 (42) 41 (35) 25 (27) 0.2 Nonwhite (%) 122 (25) 36 (31) 27 (29) 0.8 BMI (mean SD) Pretransplantation dialysis (%) 395 (81) 111 (94) 90 (98) 0.2 Cause of ESRD Diabetes 114 (23) 29 (25) 21 (23) 0.5 Hypertension 65 (13) 22 (19) 13 (14) Polycystic kidney disease 66 (14) 16 (14) 8 (9) Glomerulonephritis 126 (26) 21 (18) 23 (25) Other 116 (24) 30 (25) 27 (29) DCD (%) 92 (19) 62 (52) 57 (62) 0.17 ECD (%) 70 (14) 27 (23) 16 (17) 0.32 CIT (mean SD h) Panel reactive antibody (mean SD) Retransplantation (%) 106 (22) 26 (22) 23 (25) 0.6 Antithymocyte globulin induction (%) 136 (28) 47 (40) 32 (35) 0.3 Average HLA mismatch (mean SD) Length of stay, days (SD) <1 Discharge TAC level (ng/ml) BMI, body mass index; DCD, donor after circulatory death; ECD, extended criteria donor; CIT, cold ischemia time; ESRD, end-stage renal disease; TAC, tacrolimus; LOS, length of stay. *p value comparing pre- and post- clinic management strategies American Journal of Transplantation 2016; 16:

4 Outpatient Management of Pre Clinic Post Clinic LOS (days) LOS, length of stay;, delayed graft function Figure 3: Hospital LOS for patients in the pre- versus post- clinic. LOS, length of stay;, delayed graft function. plant dialysis months, cause of ESRD, PRA, retransplant rates, DCD, ECD, CIT, ATG induction, and mean HLA mismatch. Discharge trough tacrolimus levels were significantly lower in the post- clinic group ( vs ng/ml, p = 1). Mean LOS was significantly shorter for post- patients ( vs days, p < 1). The majority of post- clinic patients were discharged on day 5 or 6, with three patients discharged as early as day 3, whereas 21% of pre- clinic patients had an LOS of 14 days or longer (Figure 3). Readmissions, acute rejection, and outcomes Patients with had a higher 30-day readmission rate compared with those without (18% vs. 12%; Table 2: Causes of 30-day readmission No Pre- clinic Post- clinic p* Surgical complications Infection Gastrointestinal Fluid and electrolyte Rejection Genitourinary Cardiovascular Endocrine Other *p value comparing pre and post- clinic management outcomes. p = 0.02). However, the incidence of readmission did not differ significantly between pre- clinic and post- clinic patients (16% vs. 21%, p = 0.39, Tables 2 and 3). The causes for readmission were similar for pre- clinic and post- clinic patients (Table 2). Regression analyses demonstrated that patients on dialysis before transplantation had a lower risk of readmission after (Table 4). No other significant risk factors for readmission were otherwise identified. was associated with a higher rate of acute rejection (Figure 4A). However, patients in the clinic had lower acute rejection rates compared with pre- clinic patients (p = 6; Figure 4B and Table 3). Similarly, although was associated with a higher rate of graft loss (p = 0.02; Figure 5A), there was no difference in graft survival between pre- clinic and post- patients (Figure 5B). There was no difference in patient survival in versus no or in pre- versus post- clinic patients (Figures 6A and B). The 1-year serum creatinine was similar between and no- groups, mg/dl and mg/dl, respectively (p = 0.12), as well as between pre- and post-, mg/dl and mg/dl respectively (p = 0.07). UPC was similar in compared with no-, vs (p = 0.38) and pre- and post- clinic, compared with (p = 0.1). Only egfr was better in no ( ml/min) compared with ( ml/min; p < 1). The egfr was no different between pre- and post-, and , respectively (p = 0.9). American Journal of Transplantation 2016; 16:

5 Muth et al Table 3: Outcomes No Pre- clinic Post- clinic p* n Length of stay, days (SD) <1 30-Day readmission (%) 57 (12) 19 (16) 19 (21) 0.39 Acute rejection (%) 77 (16) 47 (40) 19 (21) 6 Graft failure (%) 48 (10) 24 (20) 6 (7) 0.47 Patient mortality (%) 30 (6) 13 (11) 2 (2) 0.98 *p value comparing pre and post- clinic management outcomes. Table 4: Readmission risks in patients with Discussion Odds ratio 95% CI p Age Female Nonwhite BMI Pretransplantation dialysis PRA Antithymocyte globulin Discharge tacrolimus level HLA mismatch BMI, body mass index;, delayed graft function; PRA, panel reactive antibody. In this observational study, we found that outpatient management of patients with in a specialized clinic was associated with substantially shorter hospitalization and lower incidence of acute rejection without a significant difference in 30-day readmission or patient and graft survival. We service a large geographic area and many patients cannot easily travel from home to the transplant center 3 times per week; therefore, the hotel and clinic have been invaluable in providing close posttransplantation follow-up. Patients requiring posttransplantation dialysis on discharge in the pre- clinic era were sent back to their local dialysis unit, precluding close monitoring of immunosuppression and graft function. Because acute kidney injury recovery usually begins with incremental changes in kidney function that may not be recognized in a timely fashion, it is possible that patients were dialyzed longer than necessary in the pre- clinic era. is also frightening for patients as they worry about their continuing need for dialysis. Close follow-up in the clinic reassures them during their recovery. is associated with increased LOS and greater cost (11). Reducing the LOS is beneficial in reducing the costs of hospitalization, increasing the availability of that inpatient bed for another admission, and reducing the risk of patients developing a hospital-acquired infection. We A Incidence of Acute Rejection in vs. no B Incidence of Acute Rejection in Pre vs. Post- Clinic P= <01 P= 6 Pre- Clinic Acute Rejection, proportion No Post- Clinic Figure 4: Kaplan Meier of acute rejection in patients with and without (A) and pre- and post- clinic (B)., delayed graft function American Journal of Transplantation 2016; 16:

6 A Graft failure incidence in vs. no Outpatient Management of B Graft failure incidence in Pre vs. Post- Clinic P= 0.02 P= Graft Failure, proportion No Post- Clinic Pre- Clinic Figure 5: Kaplan Meier of graft survival rejection in patients with and without (A) and pre- and post- clinic (B)., delayed graft function. A Incidence of death in vs. no B Incidence of death in Pre vs. Post- Clinic 0.06 P= P= Patient Death, proportion No Post- Clinic Pre- Clinic Figure 6: Kaplan Meier of patient death rejection in patients with and without (A) and pre- and post- clinic (B)., delayed graft function. demonstrated that it is possible to safely discharge medically complex patients who received a high-risk organ resulting in to an outpatient setting. The mean LOS was 4.8 days shorter in our post- clinic group than in the pre- clinic group. Heilman et al also found that outpatient management of was associated with shorter LOS (12) The method of outpatient management, however, is not described in detail. Farney et al found a mean LOS of 6.4 days in patients with, which was similar to that of their non- patients (13). In this American Journal of Transplantation 2016; 16:

7 Muth et al study, however, resolved in two-thirds of patients before discharge, and their management after discharge was not described. To discharge patients, the appropriate infrastructure needs to be in place. Because recovery of acute kidney injury in the form of is a dynamic process, medication changes and close immunosuppression management can occur with each multidisciplinary clinic visit. McAdams-DeMarco et al found a 31% early readmission rate posttransplantation using Medicare claims; 27.4% of DD transplants with early readmission had (14). Others have also found an association between and readmission (8). The length of the original hospitalization has also been associated with readmission (8,1,14). We did not observe a greater incidence of readmissions among patients treated in our clinic, despite being discharged an average of 4 5 days earlier. Additionally, the causes of readmissions were similar for the pre- clinic and post- clinic patients. Our post- clinic patients had a lower incidence of acute rejection, despite similar ATG use. The potential explanations for this difference are speculative. Post- clinic patients underwent indication biopsy between postoperative days 7 and 14 if there was no evidence of improvement in graft function. The threshold for biopsy, and thus detection of rejection, however, may have differed for inpatients compared with those seen in the outpatient clinic. can mask rejection, and acute rejection is likely more nephrotoxic compared with the vasoconstrictive effects of tacrolimus. Post- clinic patients had a lower tacrolimus level on discharge compared with pre- clinic patients. This was likely due to the longer LOS of the pre- clinic patients and their having more time to achieve a therapeutic level. We have previously found that discharge tacrolimus level 8 ng/ml compared with levels <8 ng/ml did not affect (15). According to Kidney Disease Improving Global Outcomes guidelines, immediate calcineurin inhibitor use has not been shown to delay kidney recovery, and achieving therapeutic levels is effective in preventing acute rejection (16). patients had significantly more graft loss than did no- patients; however, we found no difference in graft survival in the pre- clinic and post- clinic patients. has been shown to be a potent risk factor for kidney graft loss in the first year and beyond. Quiroga et al found kidney graft survival was affected by, with the highest risk of graft loss within the first 3 months, though this risk persisted beyond 1 year (17). Yarlagadda et al also found an increased risk of kidney graft loss in patients with, with the highest risk in the first year and persistence beyond the first year (7). There are emerging data, however, that is not uniformly associated with poor graft survival. Two recent studies found that associated with donor acute kidney injury, specifically, was not associated with worse graft function or graft survival (13,18). Although is associated with patient death in the literature (8,19,20), we found no difference in patient survival between and no- and between pre- and post- clinic patients. The retrospective study design and comparison to historical controls have inherent limitations, precluding conclusions regarding causation. The shorter LOS among patients in the clinic may be due, in part, to growing experience in managing contributing to earlier discharge. Likewise, the lower risk of acute rejection observed among patients in the clinic may be due to other changes to treatment or monitoring, in addition to the outpatient clinic. In summary, patients with are medically complex, tend to have a longer hospitalization, and are at risk for early readmission. Thirty percent of our DD kidney transplant patients had. Outpatient management of these patients in a specialized multidisciplinary clinic was associated with reduced LOS without any greater readmission rate. Close monitoring of patients in the outpatient setting, close manipulation of immunosuppression, and timely biopsies for screening for rejection may allow successful discharge of patients with. Interventions are still needed to prevent and improve long-term graft survival in these patients. Author Contribution Statement Dr. Muth was responsible for concept, design, data collection, analysis, manuscript preparation, and editing. Dr. Astor was responsible for design, data collection, analysis, manuscript preparation, and editing. Dr. Turk was responsible for data collection, analysis, and editing. Dr. Mohamed was responsible for analysis and editing. Dr. Parajuli was responsible for analysis and editing. Dr. Kaufman was responsible for analysis and editing. Dr. Mandelbrot was responsible for analysis and editing. Dr. Djamali was responsible for design, analysis, manuscript preparation, and editing. Disclosure The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation. References 1. Matas AS, Smith JM, Skeans MA, Thompson B, Gustafson SK, Schnitzler MA. OPTN/SRTR 2012 annual data report: Kidney. Am J Transplant 2014; 14: Irish WD, Ilsely JN, Schnitzler MA, Feng S, Brennan DC. A risk prediction model for delayed graft function in the current era of 1610 American Journal of Transplantation 2016; 16:

8 Outpatient Management of deceased donor renal transplantation. Am J Transplant 2010; 10: Perico N, Cattaneo D, Sayegh MH, Remuzzi G. Delayed graft function in kidney transplantation. Lancet 2004; 364: Brennan DC, Daller JA, Lake KD, Cibrik D, Del Castillo D. Thymoglobulin Induction Study Group. Rabbit antithymocyte globulin versus basiliximab in renal transplantation. N Engl J Med 2006; 355: Qureshi F, Rabb H, Kasiske BL. Silent acute rejection during prolonged delayed graft function reduces kidney allograft survival. Transplantation 2002; 74: Patel S, Duhart B, Krauss A, et al. Risk factors and consequences of delayed graft function in deceased donor renal transplant patients receiving antithymocyte globulin induction. Transplantation 2008; 86: Yarlagadda SG, Coca SG, Formica RN, Poggio ED, Parikh CR. Association between delayed graft function and allograft and patient survival: A systematic review and meta analysis. Nephrol Dial Transplant 2009; 24: Harhay M, Lin E, Pai A, et al. Early rehospitalization after kidney transplantation: Assessing preventability and prognosis. Am J Transplant 2013; 13: Niederhaus SV, Muth B, Lorentzen DF, et al. Luminex-based desensitization protocols: The University of Wisconsin initial experience. Transplantation 2011; 92: Djamali A, Muth B, Ellis T, et al. Increased C4d in post-reperfusion biopsies and increased donor specific antibodies at one-week post transplant are risk factors for acute rejection in mild to moderately sensitized kidney transplant recipients. Kidney Int 2013; 83: Saidi RF, Elias N, Kawai T, et al. Outcome of kidney transplantation using expanded criteria donors and donation after cardiac death kidneys: Realities and costs. Am J Transplant 2007; 7: Heilman RL, Smith ML, Kurian SM, et al. Transplanting kidneys from deceased donors with severe acute kidney injury. Am J Transplant 2015; 15: Farney AC, Rogers J, Orlando G, et al. Evolving experience using kidneys from deceased donors with terminal acute kidney injury. J Am Coll Surg 2013; 216: McAdams-DeMarco MA, Grams ME, Hall EC, Coresh J, Segev DL. Early hospital readmission after kidney transplantation: Patient and center-level associations. Am J Transplant 2012; 12: Richards KR, Hager D, Muth B, Astor B, Kaufman D, Djamali A. Tacrolimus trough level at discharge predicts acute rejection in moderately sensitized renal transplant patients. Transplantation 2014; 97: Kasiske B, Zeier M, Chapman J, et al. KDIGO clinical practice guideline for the care of kidney transplant recipients: A summary. Kidney Int 2010; 77: Quiroga I, McShane P, Koo D, et al. Major effects of delayed graft function and cold ischemia time on renal allograft survival. Nephrol Dial Transplant 2006; 21: Hall IE, Schroppel B, Doshi MD, et al. Associations of deceased donor kidney injury with kidney discard and function after transplantation. Am J Transplant 2015; 15: Narayanan R, Cardella C, Cattran D, et al. Delayed graft function and the risk of death with graft function in living donor kidney transplant recipients. Am J Kidney Dis 2010; 56: Tapiawala S, Tinckam K, Cardella C, et al. Delayed graft function and the risk for death with a functioning graft. J Am Soc Nephrol 2010; 21: American Journal of Transplantation 2016; 16: