For personal use only. Reproduce with permission from The Lancet Publishing Group. ARTICLES THE LANCET • Vol 357 • February 24, 2001 587 Summary Background Renal dysfunction is a major complication of long-term immunosuppressive therapy with calcineurin inhibitors (CNI) in liver-transplant recipients. We undertook a randomised study to assess the safety and efficacy of CNI withdrawal and replacement by mycophenolate mofetil. Methods 28 people who had had renal dysfunction attributable to suspected CNI toxicity after liver transplantation participated in the study. We replaced CNI with mycophenolate mofetil in a stepwise pattern in half the group (study patients); the other half (controls) stayed on CNI immunosuppression. Renal function, blood pressure, uric acid, and blood lipids were measured before and 6 months after study entry. Side-effects of medication and graft function were recorded throughout the study. Findings At the end of the study, mean (SD) serum creatinine had fallen by 44·4 (48·7) mol/L in study patients compared with 3·1 (14·3) mol/L in controls; a mean difference of 41·3 mol/L (95% CI 12·4–70·2). Moreover, systolic and diastolic blood pressure, and serum uric acid decreased significantly in the study group but not in the control group (mean [95% CI] between group differences 10·8 mm Hg [3·0–18·6], 5·0 mm Hg [0·9–9·2], and 83·1 mol/L [12·7–153·6], respectively). There were no changes in cholesterol or triglyceride concentrations in either group. Side- effects were reported by eight of the study patients. Three reversible episodes of acute graft rejection occurred in study patients during mycophenolate mofetil monotherapy, whereas none occurred in the control group. Interpretation Substitution of CNI by mycophenolate mofetil can improve renal function, blood pressure, and uric acid concentration of liver-transplant patients, but there is an increased rejection risk with mycophenolate mofetil monotherapy. Lancet 2001; 357: 587–91 See Commentary page 571 Introduction Calcineurin inhibitors (CNI) were one of the most important advances in transplantation medicine. The introduction of ciclosporin in the early 1980s improved early graft-survival rates and helped to establish organ transplantation as routine. Most liver-transplant patients receive one of the two calcineurin inhibitors ciclosporin or tacrolimus as basic immunosuppressive treatment. Both drugs are T-cell- specific immunosuppressants but have many non- immunological side-effects such as renal dysfunction, 1,2 arterial hypertension, hyperlipidaemia, and hyper- uricaemia, which with long-term treatment are major causes of morbidity. 3 Some workers 4–6 have suggested that replacement of CNI by azathioprine as soon as 3 months after kidney transplantation may have beneficial long-term effects. 2 Mycophenolate mofetil is an immunosuppressant specific for T cells and B cells; its non-immunological side- effects 7 are different from those of CNI. Sollinger 8 suggested that this non-nephrotoxic drug is a better immunosuppressant than azathioprine in renal-transplant patients. In recent non-randomised trials, investigators have shown that replacement of CNI by mycophenolate mofetil in patients receiving liver 9–11 or heart 12 transplants can improve renal function. In our randomised study we have analysed the consequences of substitution of CNI by mycophenolate mofetil in liver-transplant recipients. Methods Patients and randomisation We included 30 adult liver-transplant patients who had had transplantation at least 6 months previously, had stable graft function, and had renal dysfunction as a probable side-effect of ciclosporin (n=22) or tacrolimus (8) immunosuppression. Patients with a history of severe (ie, steroid-resistant or repeated) rejection episodes, other renal diseases, or serious gastrointestinal problems were excluded. The study was approved by an internal ethical review board and all patients gave informed consent. We defined renal dysfunction as serum creatinine of 125 mol/L or higher (normal range 60–120) measured on at least two successive occasions more than 1 month apart. Stable graft function was defined as stable aspartate aminotransferase and alanine aminotransferase concentra- tions (<two-fold the upper limit over the previous 3 months) and normal bilirubin. We considered all liver- transplant patients who visited an outpatients’ clinic between July, 1998, and October, 1999, and who fulfilled the criteria for inclusion. After informed consent was obtained, participants were randomised with sealed envelopes. Study design The primary endpoint was a decrease in serum creatinine after 6 months. The study had 80% power to detect a mean decrease in creatinine of 10%, with p=0·05. Replacement of calcineurin inhibitors with mycophenolate mofetil in liver-transplant patients with renal dysfunction: a randomised controlled study Hans J Schlitt, Annette Barkmann, Klaus H W Böker, Hartmut H J Schmidt, Nikos Emmanouilidis, Jens Rosenau, Matthias J Bahr, Günter Tusch, Michael P Manns, Björn Nashan, Jürgen Klempnauer Klinik für Viszeral und Transplantationschirurgie (Prof H J Schlitt MD, A Barkmann MD, N Emmanouilidis MD, Prof B Nashan MD, Prof J Klempnauer MD), Abteilung Gastroenterologie und Hepatologie (K H W Böker MD, J Rosenau MD, M J Bahr MD, Prof M P Manns MD), and Abteilung Biometrie (G Tusch PhD), Medizinische Hochschule Hannover, D-30623 Hannover, Germany Correspondence to: Prof Hans J Schlitt (e-mail: schlitt.hans@mh-hannover.de)