Impact of Food and Herbal Medication on Calcineurin Inhibitor Dose in Renal Transplant Patients: A Cross-Sectional Study

Abstract

The incidence and severity of interactions of herbal products with calcineurin inhibitor (CNI) metabolism in renal transplant recipients have not been systematically investigated. These patients have a high rate of herbal product consumption, including products interfering with CNI metabolism. The study aimed at identifying an impact of herbs and foods on CNI metabolism in a cohort of renal transplant recipients by conducting dietary interviews (1) in patients with very low and high CNI maintenance dose requirements and (2) by retrospective analysis of unexplained marked deviations from CNI baseline trough levels. Of 73 renal transplant recipients, 59 were treated with a CNI-based immunosuppressive regimen. Seven patients with an exceptionally high or low CNI dose were interviewed. Five of these seven patients had not consumed any plant product with known influence on CNI metabolism. In one patient chicory-coffee and bitter chocolate had been suspected as contributing to high CNI dose requirement, but the dose could not be lowered after discontinuation of these foods. Participating nephrologists reported three as yet unexplained temporary deviations from baseline CNI trough levels, of which two could be linked to newly started consumption of high volumes of herbal teas and the other to St. John's wort. Consumption of herbal products within the study cohort had no detectable impact on maintenance doses of CNI. However, herbal products, and specifically teas when consumed by the liter, could be linked to temporary strong deviations from CNI trough levels. The study demonstrates that as yet unnoticed herbal interactions with CNI can be detected by detailed dietary analysis, but that the overall impact on maintenance doses of CNI appears to be low.

Introduction

A high proportion of renal transplant recipients consume complementary and alternative medication and drink herbal teas as documented in recent cross-sectional studies in Bavaria and Switzerland.^1,2 Many of the consumed products could potentially interfere with the metabolism of immunosuppressants and endanger transplant function.³ Specifically, calcineurin inhibitors (CNIs) like cyclosporin A (CyA) and tacrolimus (Tac) with their narrow therapeutic ranges are vulnerable to being affected by such interactions. Clinical reports had repeatedly documented interactions of the herbal antidepressant St. John's wort (SJW) with CNI, but the associated clinical risks of other herbs to cause interactions are not known. It is specifically unresolved whether herbs that interact with elements of CNI metabolism like cytochrome-P 450 (CYP) enzymes and drug-transporting proteins in vitro need to be principally considered as modifiers of CNI doses in patients.

Hardly any nephrologist recalls having encountered interactions of foods and herbal drugs with immunosuppressants apart from those with SJW, but many of them have dealt with marked temporary deviations of CNI blood levels from target ranges and with exceptionally high or low CNI dose requirements that they had been unable to explain.

It was hypothesized that herbal products or foods may account for these characteristics of CNI treatment. The study aimed at detecting this by detailed dietary interview of patients with very low or high body weight-corrected CNI maintenance doses and as yet unexplained temporary deviations from baseline CNI trough levels.

Subjects and Methods

Patients

Clinical data of renal transplant recipients of five outpatient renal departments with associated dialysis units (in Neu-Ulm, Mindelheim, Kaufbeuren, Kempten, and Lindau/Lindenberg) in the largely rural region of Bavarian-Swabia in southern Germany were assessed. Together, these centers treat approximately 590 dialysis patients and 130 transplant patients. None of the centers employed renal dieticians for counseling of their patients.

Questionnaires and telephone interview of patients

A questionnaire on current immunosuppressive regimens, immunosuppressive drug monitoring, and selected demographic data of patients (sex, age, and time on transplantation) were sent to the renal centers in December 2008. It was requested that data be reported of consecutive transplant patients seen within 3 months (until March 2009) for routine controls in the outpatient clinic. Drug monitoring results and immunosuppressant dose of the last three visits were recorded. Informed consent was obtained from patients by a letter inviting them to participate and informing them about the intention of the study. After analysis of the returned questionnaires, the participating nephrologists were informed about those patients with exceptionally high or low CNI doses. Following their informed consent (obtained by their nephrologists), the respective patients were interviewed by the first author for their current consumption of foods and herbal drugs with a potential influence on CNI metabolism by using a detailed structured questionnaire. This questionnaire covered all known herbal modifiers of CNI metabolism as recently reviewed by the first author.⁴

Nephrologists reporting unexplained deviations of CNI dosage

Participating nephrologists were asked to report events of unexpected deviations of CNI blood levels from the target range within the last 2 years that they could not explain (e.g., by attributing them to the influence of pharmacotherapy). Criteria to further analyze such events for herbal impact were a sudden drop or rise of CNI blood levels (change by at least 50%) from stable baselines or failure to remain within the therapeutic range despite at least two adjustments of CNI dose. Events fulfilling these criteria were investigated by the first author by telephone interview of the respective patients about their dietary habits at the time of the event. This interview was again strongly based on the above-mentioned questionnaire covering products of documented influence on CNI metabolism.

Statistical analysis

Descriptive statistics were conducted using the software package StatsDirect version 2.6.5 (StatsDirect Ltd., Altrincham, United Kingdom).

Results

Immunosuppressant regimens of 85 transplant patients in current outpatient surveillance were received, representing about 65% of living transplant patients at the participating centers. Twelve patients lacking current drug monitoring data had to be excluded from the analysis, leaving 73 complete data sets to be analyzed.

Demographic data

Patients had a median age of 56 years (range, 18–78 years), their sex distribution was 48 males versus 25 females, and all were white apart from one African patient. The median time after transplantation was 25 months, and none of them had been transplanted less than 6 months before the survey.

Pattern of immunosuppressive drug regimens

Patients had been treated with variable combinations of immunosuppressive drugs (Table 1). CNI-based regimens were administered in 59 patients (40 CyA, 19 Tac). From these patients, drug trough levels (C0) and CNI dose of three visits within the last year were available for calculation of CNI dose requirements. One CyA-treated patient had to be excluded because his very low trough levels were presumably due to incompliance (information by his nephrologist).

Table 1.

Use of Immunosuppressive Drugs in Transplant Patients (n = 73)

Drug(s)	Number of patients
Cyclosporin	4
Cyclosporin + MMF	26
Cyclosporin + MMF + steroids	8
Cylosporin + azathioprin	1
Cyclosporin + everolimus + steroids	1
Tacrolimus + MMF	14
Tacrolimus + MMF + steroids	3
Tacrolimus + rapamycin	1
Tacrolimus + steroids	1
Rapamycin + MMF	5
Rapamycin + steroids	1
Everolimus + MMF	1
Everolimus + MMF + steroids	1
MMF + steroids	5
No immunosuppressive drugs	1

MMF, mycophenolate mofetil.

Dose of CNI-treated patients

Individual maintenance doses of CNI-treated patients and their relation to CNI trough levels were estimated by two approaches. Mean trough levels (C0) of CyA or Tac and the concurrently administered mean dose of the respective drugs were calculated. From these data weight-adjusted CNI dose (in mg/kg) and dose-adjusted trough level (C0/dose; L^–1) were calculated (Table 2), including means, SDs, and 95% confidence intervals. As the parameters were distributed in a non-Gaussian fashion, the 95% confidence interval was not appropriate for selecting patients with exceptional weight-adjusted doses and dose-adjusted trough levels. Therefore, the thresholds for exceptional dose requirements to be further analyzed were arbitrarily set as follows: for CyA, weight-adjusted doses of <1.5 and >4 mg/kg/day and dose-adjusted trough levels of <0.25 and >1 L^–1; for Tac, weight- adjusted doses of <0.02 and >0.07 mg/kg/day and dose-adjusted trough levels of <0.8 and >3 L^–1.

Table 2.

Calcineurin Inhibitor Dose, Drug Monitoring Results, and Individual Dose Requirements

	Cyclosporin A (n = 39)	Tacrolimus (n = 19)
Dose (mg/day)	168 ± 10.7	4.2 ± 0.93
95% confidence interval	± 21.8	± 1.97
Trough level (μg/L)	92 ± 4.3	5.4 ± 0.5
95% confidence interval	± 8.7	± 1.05
Weight-adjusted dose (mg/kg/day)	2.3 ± 0.18	0.06 ± 0.02
95% confidence interval	± 0.36	± 0.04
Dose adjusted trough level (C0/dose; L^–1)	0.6 ± 0.05	1.75 ± 0.22
95% confidence interval	± 0.09	± 0.46

C0, drug trough level.

Nine patients with exceptional CNI doses were approached to be interviewed about their dietary habits. Two of these patients declined to be interviewed. Of the remaining seven patients (four taking CyA, three taking Tac), five denied having consumed any herbal product in question, and their exceptional doses could not be attributed to dietary influences.

Two patients reported unusual dietary habits of potential influence on their CNI dose:

The patient (a 40-year-old woman) with the highest individual CyA dose of this survey (340 mg/day, 5.7 mg/kg) reported a habitual daily consumption of 1 L of a coffee substitute (prepared from the root of chicory [Cichorium intybus]) and of 50 g of bitter chocolate. Both products have not been reported as modifiers of CNI metabolism. However, the patient, who would have preferred to consume a lower CNI dose, readily agreed to abandon both foods for a while. After 4 months, however, the CyA dose necessary to maintain therapeutic levels had not decreased.

The second patient (a 38-year-old man), an immigrant from Sri Lanka who adhered to his traditional diet rich in spices, needed exceptionally low CyA doses of 60–90 mg/day (1.3 mg/kg). His CyA trough levels had repeatedly exceeded the therapeutic range, and a transplant biopsy had revealed CNI toxicity-related chronic allograft nephropathy. The patient had, unfortunately, not been available for an interview on his dietary habits despite repeated efforts.

Unexplained deviations from CNI trough level baselines

Participating nephrologists remembered three incidents of marked deviations from stable baseline CNI trough levels that could be investigated for a potential impact of herbal products. Based on this retrospective information the yearly incidence of herb/drug interactions in this cohort is roughly estimated as 1%:

In one patient (a 53-year-old man) a sudden increase of CyA trough levels into the toxic range occurred concurrently with the consumption of high volumes of a blend of herbal teas from the Himalayas. His nephrologist had advised discontinuing the tea, and the CyA levels soon dropped into the therapeutic range. The composition of the tea could not be elicited.

Following an initially uncomplicated triple immunosuppression with CyA in a 31-year-old woman, CyA trough levels decreased below the therapeutic range despite repeated dose adjustments. Her nephrologist suspected consumption of SJW, which proved right. The drug had been prescribed by the patient's general practitioner, who forgot to inform the nephrologist. The problem resolved with discontinuation of SJW.

The complication in the third patient (a 49-year-old man) occurred during hospitalization and is particularly well documented. The patient had received a second renal transplant and was treated with triple immunosuppression (steroids, mycophenolate mofetil, Tac). Until Day 24 post-transplant Tac trough levels had leveled off at 8–9 ng/mL with a dose of 4 mg/day (2 × 2 mg). Then, Tac trough levels rose unexpectedly into the toxic range (16–19 ng/mL), and serum creatinine and potassium (6.5 mmol/L) increased while the urinary output decreased. The patient had not taken any additional drugs, but a detailed dietary interview disclosed that he had started drinking 2–3 L/day camomile tea prior to the complication. Tac had to be discontinued for several days until the trough levels had normalized, and the camomile tea was replaced by other drinks. After this change in diet, the further Tac dosing was uneventful. A year later, however, a similar episode happened, when the patient was hospitalized for diagnosis and treatment of fever. Before his hospital admission Tac trough levels had been stable at 6 ng/mL, but then the levels rose to 16 ng/mL within 2 days, again accompanied by disturbed transplant function (hyperkalemia of 6.4 mmol/L, oliguria, and rise of serum creatinine from 1.3 mg/dL to 2.1 mg/dL). This time the patient had consumed large quantities of herbal “house-tea” provided by the hospital staff. Drinking large volumes of this tea had been recommended to him to fight fever-induced hypovolemia. Replacement of the tea by other drinks (mainly water) soon allowed the resumption of the patient's usual Tac dose.

Discussion

CNIs are widely used in solid organ transplantation. Drug monitoring by trough level measurements is advised to ensure sufficient immunosuppressive activity and to circumvent drug toxicity. Inadequately high as well as strongly fluctuating trough levels of CNI contribute to premature graft failures due to chronic transplant nephropathy.^5,6 Keeping CNI levels steadily within therapeutic ranges can be compromised by concomitantly administered drugs and phytochemicals in foods or herbal medication interfering with CNI metabolism by CYP enzymes and drug-transporting proteins.

It is currently unresolved whether the many plant products that interfere with CYP enzymes in vitro trigger interactions with CNI in patients when they consume these products in customary quantities. So far, relevant interactions have only been observed in conjunction with SJW and certain blends of herbal tea.^7,8 Grapefruit juice, however, a well-documented inhibitor of CYP3A4 and enhancer of CNI bioavailability,⁹ has not been reported as a cause of complications in transplantation. Thus, it appears that the value of in vitro research is limited to predict clinical interactions. Clinical observations or studies are warranted to clarify which plant products are capable of causing interactions with CNI metabolism in transplant recipients.³

The present investigation is a first attempt to answer these questions using a twofold approach. It was performed in a population of renal transplant recipients with a previously documented 50% rate of herbal product consumption.¹ By calculating the CNI maintenance dose within this population from daily doses and trough levels, patients with an exceptionally low or high dose needed to maintain therapeutic CNI trough levels were identified.

The CNI dose in our cohort was variable as is known for these drugs in the renal transplant setting.⁵ Because different target trough levels had been accepted by the responsible nephrologists, dose-adjusted trough levels are probably more appropriate for comparing dose requirements than weight-adjusted doses. In those patients with exceptional doses available for dietary interview, an influence of foods and herbs could not be retraced. The case of a patient from Sri Lanka remained unresolved, and it can only be speculated that his traditional cuisine rich in spices that are known in vitro modifiers of CYP enzymes may have contributed to his low CNI dose requirements.

It is concluded that a dietary impact on the low or high extremes of CNI maintenance doses is either absent in this cohort or not detectable by the chosen approach. The decision to search for a dietary influence among those patients with extreme CNI doses had a practical reason. It was supposed that its discovery would have direct relevance for affected patients. A low but still efficient CNI dose could, for example, rely so strongly on a powerful dietary inhibitor of CYP3A4 that this product should not be withdrawn from the diet without concurrent drug monitoring and increase of CNI dose.

Is the variability of maintenance CNI doses in this cohort altogether attributable to genetic differences of CNI-metabolism?¹⁰ Probably, although many questions still remain in this respect. The fact that the oral bioavailability of CNI in African-Americans is 20–50% lower than in whites or non–African-Americans is usually explained by an inter-ethnic variability of CYP3A4 activity.¹¹ However, studies could not convincingly attribute this difference to polymorphisms of CYP genes.^11
–13 It appears that nongenetic as well as different still unexplored genetic factors also account for these inter-ethnic differences.¹⁴ To our best knowledge, an attempt to clarify a potential role of inter-ethnic dietary differences in this respect has not been made so far.

With the second approach of this study three cases of as yet unexplained temporary changes in CNI bioavailability were detected and analyzed for a potential role of herbal products. Thus, in one patient SJW was identified as a trigger of a serious interaction with CyA levels dropping into the ineffective range. The case confirms this well-known risk related to SJW. In another patient, camomile tea and at a later occasion a tea of unknown composition caused a sudden rise of Tac levels from previously stable baselines. Supposedly, the high volumes of consumed teas rather than the quality of phytochemicals alone had been crucial to provoke these changes in Tac bioavailability. Even phytochemicals of moderate influence on CYP enzymes and transporters such as those in camomile¹⁵ could affect CNI metabolism strongly when excessively ingested and absorbed. Transplant recipients should probably not comply with the recommended high fluid intake by drinking herbal teas by the liter.

Conclusions

This study is a first attempt to fill the information hiatus between in vitro documented interference of herbs with elements of CNI metabolism, on the one hand, and the reported prevalence of herbal product consumption by transplant patients, on the other hand, with clinical data on the impact of herbs on CNI dose requirements and bioavailability fluctuations. Although there was no discernible influence on CNI maintenance doses, several as yet unexplained changes in CNI bioavailability could be attributed to the consumption of herbs. We suggest broader, more sophisticated studies to better understand this still unexplored reality of clinical transplantation.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Nowack

, Ballé

, Birnkammer

, Koch

, Sessler

, Birck

. Complementary and alternative medication consumed by renal patients in southern Germany. J Ren Nutr, 2009; 19:211–219.

Hess

, De Geest

, Halter

, Dickenmann

, Denhaerynck

. Prevalence and correlates of selected alternative and complementary medicine in adult renal transplant patients. Clin Transplant, 2009; 23:56–62.

Nowack

, Andrassy

, Fischereder

, Unger

. Effects of dietary factors on drug transport and metabolism—and its impact on dosage guidelines in renal transplant patients. Clin Pharmacol Ther, 2009; 85:439–443.

Nowack

. Cytochrome P 450 enzyme- and transport-protein mediated herb-drug interactions in renal transplant patients: grapefruit juice, St. John's wort—and beyond! Nephrology, 2008; 13:337–347.

Stoves

, Newstead

. Variability of cyclosporine exposure and its relevance to chronic allograft nephropathy: a case-control study. Transplantation, 2002; 74:1794–1797.

, Johnston

. Variable cyclosporine exposure: a risk factor for chronic allograft nephropathy and graft loss? Transplant Proc, 2004; 36:1321–1326.

Breidenbach

, Kliem

, Burg

, Radermacher

, Hoffmann

, Klempnauer

. Profound drop of cyclosporin A whole blood trough levels caused by St. John's wort (Hypericum perforatum). Transplantation, 2000; 69:2229–2230.

Nowack

, Nowak

. Herbal teas interfere with cyclosporin levels in renal transplant patients. Nephrol Dial Transplant, 2005; 20:2554–2556.

Edwards

, Fitzsimmons

, Schuetz

et al. 6′,7′-Dihydroxybergamottin in grapefruit juice and Seville orange juice: effects on cyclosporine disposition, enterocyte CYP3A4, and P-glycoprotein. Clin Pharmacol Ther, 1999; 65:237–244.

10.

Lamba

, Lin

, Schuetz

, Thummel

. Genetic contribution to variable human CYP3A-mediated metabolism. Adv Drug Deliv Rev, 2002; 54:1271–1294.

11.

Wandel

, Witte

, Hall

, Stein

, Wood

, Wilkinson

. CYP3A activity in African American and European American men: population differences and functional effect of the CYP3A4*1B5′-promoter region polymorphism. Clin Pharmacol Ther, 2000; 68:82–91.

12.

Dirks

, Huth

, Yates

, Meibohm

. Pharmacokinetics of immunosuppressants: a perspective on ethnic differences. Int J Clin Pharmacol Ther, 2004; 42:701–718.

13.

Crettol

, Venetz

, Fontana

, Aubert

, Pascual

, Eap

. CYP3A7, CYP3A5, CYP3A4, and ABCB1 genetic polymorphisms, cyclosporine concentration, and dose requirement in transplant recipients. Ther Drug Monit, 2008; 30:689–699.

14.

Cattaneo

, Baldelli

, Perico

. Pharmacogenetics of immunosuppressants: progress, pitfalls and promises. Am J Transplant, 2008; 8:1374–1383.

15.

Budzinski

, Foster

, Vandenhoek

, Arnason

. An in vitro evaluation of human cytochrome P450 3A4 inhibition by selected commercial herbal extracts and tinctures. Phytomedicine, 2000; 7:273–282.