Delayed graft function (DGF) is a common occurrence in cadaveric renal transplantation that has been linked to increased rates of acute rejection and reduced graft survival (1). The primary mechanism is ischemia-reperfusion (IR) injury; however, other factors such as alloimmunity and brain death triggered tissue inflammation have also been implicated (2). Although there have been a succession of agents that can effectively ameliorate IR injury in rodent models, none has yet proven effective in clinical transplantation.
The bioflavonoids curcumin and quercetin are naturally occurring polyphenolic compounds with several documented effects that might be beneficial in early posttransplant injury. We have shown that quercetin and curcumin reduce IR injury in the rat (3) and reduce both IR injury and rejection synergistically with mycophenolate mofetil (MMF) (4). Finally, in a phase I open label study, administration of curcumin and quercetin to renal transplant recipients was associated with no side effects, improvement in early renal function and a surprising decrease in drug induced tremor (5).
We therefore wished to study the effect of these bioflavonoids on IR injury, DGF, acute rejection and drug toxicities in the early posttransplant period in a randomized, placebo controlled study, in patients treated concomitantly with MMF.
MATERIALS AND METHODS
Between September 2002 and August 2004, 43 dialysis dependent primary recipients of cadaveric kidneys were enrolled into this randomized placebo controlled study after appropriate informed consent for this IRB approved protocol. During this period, 45 cadaveric transplants were performed. As in our open label study (5), we used the bioflavonoid preparation Oxy-Q (Farr Labs, Santa Monica, CA) which contains 480 mg of curcumin and 20 mg of quercetin, a ratio similar to what we found most beneficial in our rodent experiments (3). Patients were randomized in a blinded fashion to receive either the Oxy-Q or placebo according to the following schedule: control: one placebo capsule twice a day; low dose: one Oxy-Q in the morning and one placebo in the evening; high dose: one Oxy-Q capsule twice a day. Treatment was started within 24 hr of surgery and continued for 1 month. Immunosuppression included daclizumab (1 mg/kg on day 0, 7, and every 2 weeks for five total doses), tacrolimus (adjusted to trough levels), mycophenolate mofetil (1 g twice daily) and steroids on a tapering schedule. High immunologic risk patients and those who developed DGF received thymoglobulin (1–1.5 mg/k per day) instead of daclizumab. Cause of donor death was not recorded.
DGF was defined as dialysis need in the first week, slow function (SGF) as failure of creatinine (Cr) to drop within the first 48 hr and/or Cr>2.5 mg/dl by day 10 and early function (EF) as the rest. Standard laboratory work was collected on all patients including serum creatinine and tacrolimus trough levels. Urine was collected from the Foley bag or as a midstream urine on days 2, 14 and 30. Patients had a protocol biopsy at 1 month and all presumed acute rejection episodes in the first 6 months were biopsy proven. Patients were asked specifically about noticeable tremor and also examined for tremor in the clinic on days 14 and 30.
HO-1 was measured using the Stress XPress Human HO-1 ELISA kit from Stressgen (Vancouver, Canada). The HO-1 was extracted from the pellet obtained after the centrifugation of 3 ml of urine. The extraction and immunoassay were performed following the manufacturer’s instructions. Lipoxin A4 and 15-epi-lipoxin A4 was also measured by ELISA. Prior to performing the immunoassay, 5 ml of centrifuged recipient urine was acidified with 1N HCl and applied to a preconditioned C18 column (JTBaker, Phillipsburg, NJ). The column was then washed with water and petroleum ether. The Lipoxin A4 and 15-epi-lipoxin A4 were eluted with methyl formate. After evaporation of the methyl formate the samples were resuspended in 250 μl of Extraction Buffer (supplied with the kit). To measure the concentrations of lipoxin A4 and 15-epi-lipoxin A4 we used enzyme immunoassay kits from Neogen Corporation (Lexington, KY), 50 μl of extracted samples was assayed in duplicate according to the manufacturer’s instructions.
Category variables were compared by chi squared and continuous variables by the Kruskal-Wallis test. Other possible covariates for confounders were sought using the variance-weighted least square regression model. Level of significance was taken at P<0.05. Sample size calculation was based upon primary endpoints of early function and day 30 creatinine with an estimated power of 80% to detect a 20% difference, with a target of 40 patients, given the standard deviations calculated in our open label study.
As seen in Table 1, the three treatment groups were well matched by age, cold ischemia time, timing of pulsatile perfusion, HLA mismatches and proportion of highly sensitized (PRA>80%) patients. The only difference was the control group having a higher proportion of men (71%). There were four withdrawals: one by patient choice and three for urine leak (data prior to withdrawal was included). One urine leak was in the low-dose group and two were in the high-dose group, although one of these was due to a technical stent problem and no ureteral ischemia was seen at the time of surgical repair. All withdrawals were between postoperative day 2 and 10, therefore data on early function was on an intent-to-treat basis and later mean creatinine values were per protocol.
In terms of early functioning of the grafts, the control group had 2/15 patients with DGF compared with none in either treatment group. We then compared EF kidneys against those with DGF or SGF, since SGF has been shown to have a deleterious impact on graft survival and increased risk of acute rejection similar to that seen with DGF (6). EF was seen in 6 of 14 (43%) of controls, 10 of 14 (71%) of low dose and 13 of 14 (93%) of the high dose group (P=0.013). As seen in Figure 1, serum creatinine was lower at every time point in patients treated with the bioflavonoids. Median serum Cr was significantly lower at 2 days (control 7.65 mg/dl, range 4.7–11.4, low dose 5.85, 1.4–9.3, high dose 4.15, 1.6–5.7; P=0.0002) and 30 days (control 1.6 mg/dl, 1.2–3.0, low dose 1.65, 1.0–2.3, high dose 1.2, 0.9–2.2; P=0.026). Incidence of acute rejection within 6 months (including protocol biopsy) was 2 of 14 control (14.3%), 2 of 14 low dose 14.3% and 0 high dose. Both rejections in the control group were Banff IIb, while the two rejections in the low dose group were Ia (as well as being subclinical) and Ib. Tacrolimus associated tremor was detected on physical examination on day 30 in 2 high dose patients (13%) vs. 18 of 40 (45%) of the rest. It was mentioned as a problem by patients in 7% of the high dose group (1 patient) vs. 10.3% of the rest.
Because there were more men in the control group, we performed a variance weighted least square regression to insure that the lower creatinines in the bioflavonoid groups was not due to having lower creatinine in females. For baseline characteristics, sex was not a cofounder between the groups (P=0.642). For serum creatinine, use of Oxy-Q was significantly associated with decrease serum creatinine at 2 days (both high and low dose vs. placebo, P=0.0001) and at 30 days (high dose vs. placebo, P=0.022) and this effect was independent of sex, although women were also independently associated with lower creatinine levels on 30-days as compared to men.
Activity of HO-1 in the urine was measured in the early postoperative period. As seen in Figure 2, there was a stepwise dose response progression of increasing HO-1 activity from control to low dose to high dose groups at all days measured (P=0.046), with pairwise significance in the control vs. high dose comparison (P=0.04). Similarly, as seen in Figure 3, there was lowest HO-1 activity in the 2 patients with DGF, intermediate activity in the patients with SGF and highest activity in the EF group (P=0.006), with pairwise significance for EF vs. SGF (P=0.05) and EF vs. DGF (P=0.02). Total antioxidant capacity, lipoxin A4 and 15-epi lipoxin A4 were also measured, however results were statistically and clinically equivalent in all treatment and outcome groups (data not shown).
The medication was well tolerated with minimal side effects. One patient in the high dose group had a transient unexplained elevation of liver function tests. In the control group, one patient had a new onset of a seizure disorder and another patient had a prolonged ileus.
The deleterious effects of DGF (7) and SGF (6) on short and long term graft function, especially when associated with acute rejection, are well described. Multiple therapeutic interventions are successful in reversing acute ischemic injury in rodent models but that success has not yet translated into clinically effective therapies (8, 9). The failure of therapies targeted to a specific pathway in the IR cascade likely stems from the multifactorial nature of DGF including both antigen-dependent and -independent mechanisms as well as the fact that much of the injury leading to DGF may begin at the onset of brain death prior to organ procurement (10).
The bioflavonoids quercetin and curcumin have multiple effects that should theoretically be of benefit in IR and immune mediated injury. They are antioxidants that inhibit xanthine oxide (11) and scavenge free radicals (12). They reduce pro-inflammatory cytokines through blockade of NF-kB (13) and induction of the hemoxygenase-1 (HO-1) enzyme (14). They are renoprotective in several models of rodent ischemic and toxic injury (15). We have shown that quercetin and curcumin prevent renal injury in rodent models of IR (3) and ureteral obstruction (16). Furthermore, these bioflavonoids were synergistic with MMF in reducing ischemic injury as well as acute rejection (17). Alloimmune responses may be inhibited by bioflavonoids through blockade of cell mediated cytotoxicity (14) and T cell activation (13). These findings led to an open label phase I study of quercetin and curcumin in renal transplant recipients (5). We found that it was well tolerated, did not effect cyclosporine or tacrolimus levels, had no deleterious effects in patients with good graft function and showed improvement in graft function in those with DGF or chronic allograft nephropathy. Surprisingly, we also saw that patients with drug induced tremor reported a reduction or cessation when they were taking the bioflavonoids.
In the present randomized placebo controlled study, we again found that the bioflavonoids were well tolerated and had beneficial effects in the early posttransplant period. Because our region puts all cadaveric kidneys on a pulsatile perfusion pump, our incidence of dialysis dependent DGF is low. We do, however, have long cold ischemia times. Nevertheless, the only 2 instances of DGF in our study were in the control group, with zero DGF with either dose of bioflavonoids. Even in patients who don’t require posttransplant dialysis, SGF has been shown to be deleterious to outcomes (6). In this study, DGF and SGF were least likely in the high dose bioflavonoid group. In addition, the high dose bioflavonoid group had the lowest serum creatinine values, the least neurotoxicity and an acute rejection rate at 6 months (including subclinical rejection) of 0%. This beneficial effect may have come from any of the previously mentioned known mechanisms of bioflavonoids or from other undiscovered pathways.
One plausible explanation for the beneficial effects of these bioflavonoids is the induction of HO-1, an inducible enzyme that produces carbon monoxide. There has been great recent interest in HO-1 induction in organ transplantation (18) for its ability to reduce IR damage and alloimmunity. HO-1 deficient rodent kidneys have higher levels of proinflammatory Th1 cytokines (19) and kidney transplants from human donors with a genetic phenotype predisposing to increased HO-1 activity have lower serum creatinines at 1 year (20). In our study there was a stepwise increase in urinary HO-1 activity with increasing bioflavonoid dose. Quercetin has been shown to increase HO-1 production in rat smooth muscle cells (21). Furthermore, curcumin induces HO-1 mRNA in human renal proximal tubule cells (22) and that this induction depends upon the Nrf2 transcription factor (23). These results are completely observational and it is possible that the increased urinary HO-1 activity seen in this study did not reflect intra-renal changes and may have been due to factors other than the ingested bioflavonoids. The mechanism of this effect deserves further study.
Neurotoxicity associated with calcineurin inhibitors is a common and often annoying problem for patients that does not always resolve with dose reduction. We had previously observed that patients with calcineurin inhibitor induced tremor had reduction or resolution while taking bioflavonoids and recurrence once they stopped. Again in this study, observation of hand tremor was lowest in the high dose bioflavonoid group. The mechanism for this is not clear, however in one animal model, administration of curcumin reduced lead induced neurotoxicity (24).
In conclusion, the use of Oxy-Q, a bioflavonoid preparation containing quercetin and curcumin, improved early renal function in our patient population. This effect was likely mediated by induction of HO-1 activity. Future studies should examine whether the benefits are also seen in cadaveric kidneys preserved by cold storage only and whether pretreatment of the donor and/or recipient would give equivalent or superior results.
1. Shoskes DA, Shahed AR, Kim S. Delayed graft function. Influence on outcome and strategies for prevention. Urol Clin North Am
2001; 28: 721.
2. Pratschke J, Wilhelm MJ, Kusaka M, et al. Brain death and its influence on donor organ quality and outcome after transplantation. Transplantation
1999; 67: 343.
3. Shoskes DA. Effect of bioflavonoids quercetin and curcumin on ischemic renal injury: a new class of renoprotective agents. Transplantation
1998; 66: 147.
4. Shoskes DA, Jones EA, Shahed A. Synergy of mycophenolate mofetil and bioflavonoids in prevention of immune and ischemic injury. Transplant Proc
2000; 32: 798.
5. Shoskes DA, Thomas M, Pobgee R, et al. Phase I study of oral bioflavonoids in cadaveric renal transplant recipients: effects on delayed graft function and calcineurin inhibitor toxicities. Transplant Proc
2003; 35: 841.
6. Humar A, Johnson EM, Payne WD, et al. Effect of initial slow graft function on renal allograft rejection and survival. Clin Transplant
1997; 11: 623.
7. Shoskes DA, Cecka JM. Deleterious effects of delayed graft function in cadaveric renal transplant recipients independent of acute rejection. Transplantation
1998; 66: 1697.
8. Noel C, Hazzan M, Coppin MC, et al. A randomized controlled trial of pentoxifylline for the prevention of delayed graft function in cadaveric kidney
graft. Clin Transplant
1997; 11: 169.
9. Salmela K, Wramner L, Ekberg H, et al. A randomized multicenter trial of the anti-ICAM-1 monoclonal antibody (enlimomab) for the prevention of acute rejection and delayed onset of graft function in cadaveric renal transplantation: a report of the European Anti-ICAM-1 Renal Transplant Study Group. Transplantation
1999; 67: 729.
10. Nijboer WN, Schuurs TA, van der Hoeven JA, et al. Effect of brain death on gene expression and tissue activation in human donor kidneys. Transplantation
2004; 78: 978.
11. Chang WS, Lee YJ, Lu FJ, Chiang HC. Inhibitory effects of flavonoids on xanthine oxidase. Anticancer Res
1993; 13: 2165.
12. Huk I, Brovkovych V, Nanobash Vili J, et al. Bioflavonoid quercetin scavenges superoxide and increases nitric oxide concentration in ischaemia-reperfusion injury: an experimental study. Br J Surg
1998; 85: 1080.
13. Ranjan D, Chen C, Johnston TD, et al. Curcumin inhibits mitogen stimulated lymphocyte proliferation, NFkappaB activation, and IL-2 signaling. J Surg Res
2004; 121: 171.
14. Gao X, Kuo J, Jiang H, et al. Immunomodulatory activity of curcumin: suppression of lymphocyte proliferation, development of cell-mediated cytotoxicity, and cytokine production in vitro. Biochem Pharmacol
2004; 68: 51.
15. Singh D, Chander V, Chopra K. Quercetin, a bioflavonoid, attenuates ferric nitrilotriacetate-induced oxidative renal injury in rats. Drug Chem Toxicol
2004; 27: 145.
16. Jones EA, Shahed A, Shoskes DA. Modulation of apoptotic and inflammatory genes by bioflavonoids and angiotensin II inhibition in ureteral obstruction. Urology
2000; 56: 346.
17. Jones EA, Shoskes DA. The effect of mycophenolate mofetil and polyphenolic bioflavonoids on renal ischemia reperfusion injury and repair. J Urol
2000; 163: 999.
18. Bach FH. Heme oxygenase-1 as a protective gene. Wien Klin Wochenschr
2002; 114 Suppl 4: 1.
19. Kapturczak MH, Wasserfall C, Brusko T, et al. Heme oxygenase-1 modulates early inflammatory responses: evidence from the heme oxygenase-1-deficient mouse. Am J Pathol
2004; 165: 1045.
20. Exner M, Bohmig GA, Schillinger M, et al. Donor heme oxygenase-1 genotype is associated with renal allograft function. Transplantation
2004; 77: 538.
21. Lin HC, Cheng TH, Chen YC, Juan SH. Mechanism of heme oxygenase-1 gene induction by quercetin in rat aortic smooth muscle cells. Pharmacology
2004; 71: 107.
22. Hill-Kapturczak N, Thamilselvan V, Liu F, et al. Mechanism of heme oxygenase-1 gene induction by curcumin in human renal proximal tubule cells. Am J Physiol Renal Physiol
2001; 281: F851.
23. Balogun E, Hoque M, Gong P, et al. Curcumin activates the heme oxygenase-1 gene via regulation of Nrf2 and the antioxidant responsive element. Biochem J
2003; 371: 887.
24. Shukla PK, Khanna VK, Khan MY, Srimal RC. Protective effect of curcumin against lead neurotoxicity in rat. Hum Exp Toxicol
2003; 22: 653.