Ciclosporin

Hypertension

Compared with azathioprine, hypertension has been considered one of the main long-term risks in patients taking ciclosporin, with major concerns about the post-transplantation increase in cardiovascular morbidity and mortality. However, there are many susceptibility factors for cardiovascular disease in transplant patients [ ], and it is difficult to take into account their complex interplay. Ciclosporin-associated hypertension appears to be dose-related, and higher whole-blood ciclosporin concentrations were found during the preceding months in patients who had thromboembolic complications compared with patients who did not [ ].

De novo or aggravated hypertension is very common in patients taking ciclosporin, with the highest incidence in cases of heart transplant (71–100%) and the lowest incidence in bone marrow transplant recipients (33–60%) [ ]. In addition, 30–45% of patients with psoriasis, rheumatoid arthritis, or uveitis had hypertension, suggesting that ciclosporin is a significant cause of hypertension in organ transplantation. Ciclosporin-associated hypertension can cause acute vascular injury, with microangiopathic hemolysis, encephalopathy, seizures, and intracranial hemorrhage.

In 3365 adults with a functioning graft after the first year, the prevalence of hypertension increased progressively and significantly during follow-up [ ]. The presence of arterial hypertension at 1 year was significantly associated with recipient sex (male), donor age (under 60 years), immunosuppressive therapy (ciclosporin), serum creatinine, and year of transplantation. Arterial hypertension was not associated with graft survival or cardiovascular mortality. The prevalence and severity of hypertension was significantly lower in patients treated with tacrolimus than with ciclosporin.

In a patient undergoing hemodialysis cardiothoracic bioimpedance showed evidence of raised systemic vascular resistance with normal blood volume [ ]. She stopped taking ciclosporin and her blood pressure fell. Ciclosporin was restarted in a dosage of 25 mg/day and her systemic vascular resistance was 1275 dyne/sec5. When the dosage was increased to 100 mg/day her blood pressure rose to 200/80 mmHg and her systemic vascular resistance to 1874 dyne/sec5, with normal cardiac output and a low-normal intrathoracic fluid volume.

Calcineurin inhibitors potentially contribute to the risk of cardiovascular events through the development of new-onset diabetes mellitus, hypertension, and hyperlipidemia. Trials have consistently shown a higher incidence of new-onset diabetes mellitus with tacrolimus, which has been borne out in large-scale registry analyses. However, the risk of hypertension is about 5% higher with ciclosporin than tacrolimus, as is the risk of hyperlipidemia [ ].

The incidence, clinical features, consequences, and management of ciclosporin-induced hypertension have been reviewed [ ]. The prevalence was 29–54% in non - transplant patients and 65–100% in heart and liver transplant patients also taking glucocorticoids. Disturbed circadian rhythm with a loss of nocturnal blood pressure fall was the main characteristic, and patients therefore had higher risks of left ventricular hypertrophy, cerebrovascular damage, microalbuminuria, and other target organ damage.

The pathophysiology of ciclosporin-induced hypertension is complex and not yet fully elucidated. Increased systemic vascular resistance subsequent to altered vascular endothelium function, renal vasoconstriction with reduced glomerular filtration and sodium-water retention, and/or increased activity of the sympathetic nervous system were suggested, while only a minor role or none was attributed to the renin–angiotensin system [ ]. However, hypertension often occurs before changes in renal function or sodium balance can be demonstrated, and ciclosporin nephrotoxicity alone does not explain ciclosporin-associated hypertension [ , ].

It has also been suggested that the increase in blood pressure induced by calcineurin inhibitors is mediated by excitation of the sympathetic nervous system. In 24 renal transplant patients who were randomly assigned to either withdrawal or continuation of ciclosporin, mean arterial pressure fell significantly during withdrawal but not during continued therapy [ ]. Muscle sympathetic nerve activity and plasma noradrenaline concentrations did not change in either group, and graft function remained stable.

The effects of antihypertensive agents have been evaluated in patients taking ciclosporin. Collectively, dihydropyridine calcium channel blockers that do not affect ciclosporin blood concentrations substantially or at all (felodipine, isradipine, and nifedipine) are usually considered to be the drugs of choice. However, the risk of gingival hyperplasia with nifedipine, which ciclosporin also causes, should be borne in mind. Combination therapy with angiotensin-converting enzyme inhibitors or beta-blockers, or the use of other calcium channel blockers (verapamil or diltiazem) should also be considered, but careful monitoring of ciclosporin blood concentrations is recommended with the latter because they inhibit ciclosporin metabolism.

Ischemic heart disease

Major cardiovascular risk factors have been analysed and the risk of coronary artery disease estimated in a comparative 6-month study of microemulsified ciclosporin (n = 271) versus tacrolimus (n = 286) concomitant with azathioprine and glucocorticoids [ ]. The primary endpoints were the incidence of and time to acute rejection. Blood pressure, serum cholesterol, HDL cholesterol, triglycerides, and blood glucose were measured at baseline and at months 1, 3, and 6. The 10-year risk of coronary heart disease was estimated according to the Framingham risk algorithm. Tacrolimus lowered serum cholesterol and mean arterial blood pressure, but in a higher summary measure of blood glucose than ciclosporin. Serum triglycerides were not different between tacrolimus and ciclosporin. The mean 10-year coronary artery disease risk estimate was significantly lowered in men who took tacrolimus, but was unchanged in women.

Ischemic heart disease after renal transplantation has been reported to be three to four times higher than in the general population. The records of recipients of cadaveric kidneys between January 1985 and November 1999 have been reviewed in a multicenter study in Spain [ ]. All received ciclosporin and steroids with or without azathioprine as initial therapy. There were 163 with ischemic heart disease and 362 control patients without ischemic heart disease. Of the patients with ischemic heart disease after transplantation, 38% had the cardiac event during the first 12 months; 29 had previously been known to have ischemic heart disease, but all were asymptomatic at the time of transplantation. After transplantation 21 revascularization procedures were performed, eight during the first year and 13 thereafter. At the time the study was performed, 28% of the patients had lost the graft and 17% (86/525) had died; 34% (55/163) from the ischemic heart disease group and 8.6% (31/362) from the control group. Death was related to ischemic heart disease in 66% of the patients with ischemic heart disease. Donor age was higher in the patients with ischemic heart disease than in the controls, as were hypercholesterolemia and hypertriglyceridemia after transplantation. Graft function was similar in the two groups, measured by serum creatinine. Multivariate analysis showed that age at transplantation increased the risk of ischemic heart disease by 5% per year. Men had a more than two-fold higher risk of ischemic heart disease than women. Body weight was associated with a 2% increased risk of ischemic heart disease. Other independent predictors were a previous history of cardiovascular disease and hypercholesterolemia before transplantation.

Raynaud’s disease

A possible role of ciclosporin in the exacerbation or development of Raynaud’s disease has been suggested on one occasion; such an effect could be linked to endothelial damage or changes in platelet function [ ].

Erythromelalgia

Erythromelalgia is a symptom complex of painful inflammatory vasodilatation of the extremities, often regarded as the inverse of Raynaud′s phenomenon. It is usually idiopathic or due to thrombocythemia, and rarely caused by calcium channel blockers.

• A 37 year old man took ciclosporin 150 mg/day for psoriasis and after 4 weeks developed marked erythema, edema, and tenderness over the fingers and toes [ ]. His symptoms increased with warmth and were partly relieved by cold compresses. His full blood count, serum biochemistry, urine analysis, and collagen profile were normal. Ciclosporin was withdrawn and the lesions regressed within 1 week but recurred when ciclosporin was restarted.

Thrombotic microangiopathy

The association between the risk of thrombotic microangiopathy and the use of the combinations ciclosporin + mycophenolate mofetil, ciclosporin + sirolimus, tacrolimus + mycophenolate mofetil, and tacrolimus + sirolimus has been studied in 368 kidney or kidney-pancreas transplant recipients [ ]. Biopsy-proven thrombotic microangiopathy was detected in 13 patients in the absence of vascular rejection. The incidence of thrombotic microangiopathy was highest with ciclosporin + sirolimus (21%). The relative risk of thrombotic microangiopathy was 16 (95% CI = 4.3, 61) for ciclosporin + sirolimus compared with tacrolimus + mycophenolate mofetil. Ciclosporin + sirolimus was the only regimen that had concomitant pro-necrotic and anti-angiogenic effects on arterial endothelial cells. This suggests that ciclosporin + sirolimus causes thrombotic microangiopathy through dual effects on endothelial cell death and repair.

Capillary leak syndrome

A capillary leak syndrome with subsequent pulmonary edema has also been reported after intravenous ciclosporin [ ].

Phlebitis

Infusion phlebitis has been attributed to intravenous ciclosporin [ ].

• A 28-year-old man with ulcerative colitis had acute recurrent infusion phlebitis during administration of intravenous ciclosporin following intravenous hydrocortisone. The intravenous catheter and its site needed to be replaced repeatedly during treatment, which eventually led to complete remission of the ulcerative colitis. After 8 months, he was still in remission, with no permanent signs of damage to the phlebitic veins.

Incidence

Neurological adverse effects of ciclosporin have been reported in up to 39% of all transplant patients. Most are mild. The most frequent is a fine tremor, the mechanism of which is not known. From many case reports or studies in transplant patients, the pattern of ciclosporin neurotoxicity ranges from common and mild to moderate symptoms, such as headaches, tremors, paresthesia, restlessness, mood changes, sleep disturbances, confusion, agitation, and visual hallucinations, to rare but severe or life-threatening disorders, including acute psychotic episodes, cerebellar disorders, cortical blindness (permanent in one report), spasticity or paralysis of the limbs, catatonia, speech disorders or mutism, chorea, seizures, leukoencephalopathy, and coma [ ].

A 19% incidence of central nervous system toxicity with ciclosporin has been reported in pediatric renal transplantation patients; the symptoms included seizures, drowsiness, confusion, hallucinations, visual disturbances, and mental changes [ ].

Neurological symptoms were observed in 12–25% of liver-transplant patients and in 29% of bone marrow transplant patients, but severe neurotoxicity occurred only in about 1% [ , , ].They usually appeared within the first month of treatment, but were sometimes delayed [ ]. Particular attention should be paid to prompt recognition of severe neurotoxicity, because abnormalities of the white matter can occur. Patients usually improved rapidly after temporary ciclosporin withdrawal or dosage reduction, and tacrolimus has sometimes been used successfully instead [ ]. However, recurrence of seizures and persistent electroencephalographic abnormalities were found in 46 and 70% of pediatric transplant patients respectively who had had ciclosporin acute encephalopathy and seizure syndrome and who were followed-up for 49 months [ ].

Mechanisms

Although the role of many other factors should be considered when neurological symptoms occur in transplant recipients, isolated reports of neurotoxicity in non-transplanted patients are in keeping with a causal role of ciclosporin. There are many susceptibility factors in ciclosporin neurotoxicity. Blood ciclosporin concentrations are sometimes raised, but severe neurological symptoms have been observed in some patients with concentrations in the usual target range [ ]. Other possible susceptibility factors for ciclosporin neurotoxicity include hypocholesterolemia, hypomagnesemia, aluminium overload, concomitant high-dose glucocorticoid therapy, hypertension, and concomitant microangiopathic hemolytic anemia [ , ]. Acute graft-versus-host disease or HLA-mismatched and unrelated donor transplants were also potential susceptibility factors in recipients of bone marrow transplants [ ].

Ciclosporin-induced vasculopathy, with endothelial injury and derangement of the blood–brain barrier, is the postulated mechanism of neurological damage. Transient cerebral perfusion abnormalities, demonstrable in SPECT scans of the brain, have been suggested as a reliable indicator of ciclosporin neurotoxicity [ ]. Clinical symptoms as well as CT and/or MRI scans were very similar to those observed in hypertensive encephalopathy, with predominant and reversible white-matter occipital lesions [ ]. There was complete neurological recovery in most patients after blood pressure was normalized, and deaths due to intracranial hemorrhage are reported only exceptionally

There has been debate about whether ciclosporin crosses the blood–brain barrier and enters the cerebrospinal fluid. Ciclosporin could not be identified in cerebrospinal fluid from 14 patients with liver transplants who had various neurological complications [ ]. Ciclosporin metabolites were measurable in the cerebrospinal fluid in only four patients, who had evidence of acute renal insufficiency, cholestasis, and raised blood concentrations of ciclosporin metabolites but identical ciclosporin parent drug blood concentrations compared with 10 patients with undetectable concentrations of ciclosporin metabolites in the cerebrospinal fluid. Ciclosporin metabolites enter the cerebrospinal fluid, and direct neurotoxicity is therefore possible in at least some patients with renal or hepatic dysfunction.

Endogenous ligands for ciclosporin and tacrolimus, known as immunophilins, are found in very high concentrations in the basal ganglia, and ciclosporin can alter dopamine phosphorylation in the medium-sized neurons in the striatum. Changes in basal ganglia glucose metabolism have been studied in a patient with severe ciclosporin-related tremor [ ].

• A 37-year-old man received ciclosporin after bone marrow transplantation for chronic myelogenous leukemia. Soon afterwards he developed a severe tremor, which persisted despite dosage reduction. A brain MRI scan was normal. After 22 months he developed a personality change. A high resolution PET scan showed symmetrical increases in 18 F-deoxyglucose uptake in both caudate and putamen.

These results confirm that ciclosporin can modulate dopaminergic transmission in the striatum, presumably by inhibition of calcineurin.

• A 16-year-old girl with end-stage renal insufficiency underwent successful renal transplantation and was given ciclosporin on day 1 [ ]. On day 10 she complained of tinnitus and tremor and had a right facial nerve palsy. An MRI scan showed areas of increased signal in the white matter of the periventricular region. The dose of ciclosporin was reduced, since no other cause could be determined. Her tremor and tinnitus resolved, but the facial nerve palsy persisted. She was given tacrolimus, but the tremor and tinnitus recurred. She was then given mycophenolate mofetil and prednisone, and the tremor and tinnitus disappeared, although the facial nerve palsy persisted. The MRI scan 3 months later was normal.

The serum magnesium concentration was below the reference range in this case, which may have favored the development of neurotoxicity.

Ciclosporin can cause pseudotumor cerebri [ ].

• A 15-year-old girl with hypodiploid acute lymphoblastic leukemia took methotrexate and ciclosporin 1.5 mg/kg after bone marrow transplantation and 30 days later developed a continuous frontal headache with bilateral papilledema. Her blood ciclosporin concentration was 85 μg/l. A CT scan showed mild prominence of the lateral and third ventricles with no intracranial masses or obstructing lesions. The CSF was normal. Ciclosporin was withdrawn and replaced by tacrolimus. After a few days her headache resolved and fundoscopy showed less disk edema.

• A 12-year-old boy with myelodysplastic syndrome took methotrexate and ciclosporin 1.5 mg/kg after stem cell transplantation and the blood ciclosporin concentration was kept at 150–200 μg/l. After 45 days he developed persistent unexplained nausea and vomiting with no headache, unresponsive to antiemetics, and bilateral papilledema. A CT scan showed no evidence of intracranial masses and the CSF was normal. He was given acetazolamide and ciclosporin was replaced by mycophenolate mofetil. His symptoms resolved after a short course of methylprednisolone and the papilledema improved.

In 20–40% of liver transplant recipients taking calcineurin inhibitors central nervous system toxicity presents a wide spectrum of mild to severe neurological and psychiatric disorders, for which intravenous lipid may be beneficial. Of 45 Japanese patients who received living-donor transplants 10 developed neurological complications, such as changes in mental status, confusion, seizures, altered level of consciousness, neuralgia, headache, and tremor; three were taking ciclosporin and seven tacrolimus [ ]. If the blood–brain barrier is impaired, unbound calcineurin inhibitors might be able to enter the brain more readily. Based on this hypothesis, five patients (two of whom were taking ciclosporin and three tacrolimus) received an intravenous lipid supplement of 20% soybean oil 0.1–0.2 ml/kg/hour for 2–3 days, after which, and without a change in the dosage of the calcineurin inhibitor, their neurological symptoms improved remarkably.

Susceptibility factors

A retrospective study identified a significantly higher incidence of central nervous system symptoms in patients with Behçet’s disease [ ]. Headache, fever, paralysis, ataxia, dysarthria, or disturbed consciousness occurred in 12 of 47 ciclosporin-treated patients compared with nine of 270 patients not treated or taking other drugs. CT and/or MRI scans were abnormal in all 12. As the clinical findings were very similar to the neurological effects of Behçet’s disease, it was suggested that ciclosporin can promote the development of neurological complications in this population.

Ciclosporin neurotoxicity is particularly frequent in liver and bone marrow transplant patients, who usually recover after temporary dosage reduction or withdrawal. However, a fatal outcome has been reported [ ].

• A 54-year-old man was given ciclosporin and methotrexate after allogeneic bone marrow transplantation. He noted blurred vision during several days and became confused 11 weeks after transplantation. Generalized tonic-clonic seizures occurred the day after and he was given phenytoin and antibiotics. His neurological condition deteriorated during the next 5 days, despite ciclosporin withdrawal, and he died from respiratory failure. Postmortem examination showed white matter edema and astrocyte injury without demyelination.

After liver transplantation, 13 of 142 recipients ciclosporin and methylprednisolone (9.2%) had neurological symptoms, including five with central pontine myelinolysis and eight with cerebral hemorrhages or infarcts [ ]. Factors that were associated with central pontine myelinolysis were hyponatremia, a rapid rise in serum sodium concentration, a postoperative increase in plasma osmolality, duration of operation, and high ciclosporin concentrations.

Clinical reports

Adverse nervous system effects of ciclosporin sometimes cause seizures.

• A 10-year-old boy with beta-thalassemia underwent autologous stem cell transplantation followed by ciclosporin 1 mg/kg/day [ ]. The trough ciclosporin concentration on day 14 was 392 ng/ml and the patient subsequently had a generalized tonic-clonic seizure. Ciclosporin was withdrawn, phenytoin was started, and the patient recovered. Six weeks later, a control brain MRI scan showed nearly complete resolution of multifocal high signal abnormalities on T2-weighted sequences in the cortex and subcortical white matter.

• A 58-year-old man with severe chronic obstructive pulmonary disease underwent bilateral lung transplantation with allografts from a cytomegalovirus seropositive donor; the recipient was seronegative [ ]. Immunosuppression was maintained with ciclosporin, prednisone, and mycophenolate mofetil. The dose of ciclosporin was adjusted to achieve blood concentrations of 350–400 ng/ml. On postoperative day 2 he had a generalized tonic–clonic seizure, when the ciclosporin concentration was 52 ng/ml; he responded to lorazepam and valproic acid. On day 10 he was given intravenous ganciclovir for prophylaxis against cytomegalovirus. By day 12 he was drowsy and unable to move his limbs. He had bilateral ptosis with intact extraocular muscles. His motor strength in the arms and legs was 0/5, with absent deep tendon reflexes. There was a five-fold increase in CSF protein with no white cells or red cells; bacterial and viral cultures of the CSF, blood, and bronchoalveolar fluid were negative, including PCR for cytomegalovirus. Electromyography and nerve-conduction studies showed a length-dependent, symmetrical, sensorimotor polyneuropathy with axonal features but no demyelination. An MRI scan of the cervical spinal cord was normal. Plasma exchange with albumin (30 ml/kg) every other day for a total of 7 sessions was undertaken and ciclosporin was replaced with tacrolimus. After the first plasmapheresis his strength improved to 2/5 in the arms and 3/5 in the legs. Over the next several weeks he continued to have steady neurological improvement.

• A 16-year-old man with chronic autoimmune hepatitis received a cadaveric orthotopic liver transplant and immunosuppression with prednisone, ciclosporin, and mycophenolate mofetil. Postoperatively he developed had generalized tonic–clonic seizures. His blood pressure was normal and electroencephalography was inconclusive. The cerebrospinal fluid was clear, with 25 cells, a protein concentration of 2 g/l, and a glucose concentration of 3.3 mmol/l. Cultures were negative. The blood ciclosporin concentration was 405 ng/ml. An MRI scan showed changes compatible with a reversible posterior leukoencephalopathy. Ciclosporin was replaced with tacrolimus and he recovered.

Three patients underwent hemopoietic stem cell transplantation for thalassemia.

Prophylaxis of graft-versus-host disease included ciclosporin, prednisolone, and methotrexate. All developed status epilepticus with cortical signal alterations mainly in the occipital regions on T2-weighted MRI. The neurological symptoms and radiographic findings completely disappeared without withdrawing ciclosporin. Children with thalassemia receiving busulfan or cyclophosphamide as a preparative regimen seem to be especially susceptible to ciclosporin-associated seizures [ ].

In one series of 367 adult and pediatric liver recipients there were 17 cases of new-onset seizures [ ]. However, information on the immunosuppressants they were taking was scanty: three were reported as having ciclosporin toxicity and three as having tacrolimus (FK506) toxicity. The causes were neurotoxicity due to immunosuppressive therapy (n = 6), cerebrovascular disease (n = 4), severe metabolic derangements by sepsis or rejection (n = 3), hyperglycemia (n = 1), brain edema due to fulminant hepatic failure (n = 1), and unknown (n = 2). Seizures recurred in 15 patients, including nine on the same day. The incidence of death or persistent vegetative state in those with seizures was almost 10 times higher than in those without (53% versus 5.7%). The prognosis in patients with seizures due to cerebrovascular disease and severe metabolic derangement due to sepsis or rejection was poorer than in patients with seizures caused by drug neurotoxicity. The eight surviving patients were free of seizures for a follow-up period of 43 (16–58) months.

Reversible cortical blindness is a rare manifestation of ciclosporin toxicity and occurred in a lung transplant recipient [ ].

Tremor has been studied in patients with severe liver disease without hepatic encephalopathy and after liver transplantation, comparing ciclosporin (n = 29) and tacrolimus (n = 6) [ ]. Compared with controls the patients had significant postural hand tremor before and after liver transplantation. The mean tremor amplitude increased after liver transplantation during treatment with both ciclosporin and tacrolimus. At higher ciclosporin plasma concentrations there was a reduction in the dominant tremor frequency with weight load and an increase in tremor amplitude, suggesting enhanced physiological or toxic tremor.

Severe headache is an uncommon adverse effect of ciclosporin [ ].

• A 66-year-old woman with no history of headache or adverse drug reactions developed membranous glomerulonephritis and autoimmune non-central neutropenia. She was given ciclosporin microemulsion 2.5 mg/kg/day. Whole blood ciclosporin concentrations were within the target range. After 5 days she reported a severe, disabling, holocranial headache. A CT scan was normal and other causes of headache were ruled out. The headache subsided when ciclosporin was withdrawn.

Pseudotumor cerebri caused bilateral disk edema with cerebrospinal hypertension in a child treated with ciclosporin [ ].

• An 11-year-old boy with recurrent tubulointerstitial nephritis associated with uveitis (TINU syndrome) was treated with ciclosporin to induce sustained remission. Ciclosporin was introduced as a steroid-sparing drug because of extreme obesity (BMI 32 kg/m ² ). Although he did not complain of any symptoms, eye inspection after 7 months showed bilateral disk edema with retinal bleeding and he developed cerebrospinal hypertension. Pseudotumor cerebri was diagnosed by measuring the intracranial pressure (31 cm H ₂ O) and normal CT and MRI scans. Ciclosporin was withdrawn, and treatment with mycophenolate mofetil led to resolution within 12 weeks.

• A 67-year-old woman taking ciclosporin developed rapidly progressive sensorimotor changes a few months after kidney transplantation [ ]. The symptoms improved after ciclosporin withdrawal. Other causes were ruled out.

After liver transplantation, a patient developed a chronic inflammatory demyelinating polyradiculoneuropathy while taking ciclosporin + prednisolone [ ]. Treatment with intravenous immunoglobulin significantly improved the neuropathy.

Ciclosporin can cause leukoencephalopathy in organ transplant recipients [ ].

• A 43-year-old man with a history of chronic renal insufficiency of unknown cause received a cadaveric kidney transplant and immunosuppression with antithymocyte-globulin, mycophenolate mofetil, ciclosporin, and prednisone. Four months later, he developed pneumonia and received fluoroquinolones for 14 days. His blood ciclosporin concentration was 659 ng/ml. 10 days after discharge, he had a headache, meningism, and a bilateral sixth nerve palsy. The cerebrospinal fluid was clear with 270 cells, including 60% polymorphonuclear leukocytes and 40% mononuclear cells and a glucose concentration of 3 mmol/l. There was no evidence of infection, but ceftriaxone, vancomycin, ampicillin, and ganciclovir were given empirically. An MRI scan showed many hypersignals in the subcortical white matter, particularly in the broadcast crown and semioval centers. Ciclosporin was withdrawn, with complete clinical improvement.

• A 50-year-old man with end-stage renal disease due to mesangiocapillary glomerulonephritis received a cadaveric kidney transplant and immunosuppression with azathioprine and prednisone. Because of gouty arthritis azathioprine was replaced by ciclosporin before treatment with allopurinol. Two weeks later he developed confusion, irritability, and personality changes. The blood ciclosporin concentration was 547 ng/ml. An MRI scan showed changes compatible with a leukoencephalopathy. He recovered quickly when mycophenolate mofetil replaced ciclosporin.

In kidney transplant recipients and children with kidney diseases, there were 20 cases (13 boys and 7 girls, aged 2–18 years) of the posterior reversible encephalopathy syndrome [ ]. Of 177 kidney transplant recipients, six of 127 patients who were given ciclosporin and four of 50 patients who were given tacrolimus developed the syndrome after transplantation. Seven had idiopathic nephrotic syndrome (all treated with ciclosporin), two had acute poststreptococcal glomerulonephritis, and one had diffuse mesangial sclerosis. The symptoms varied from headache to status epilepticus. All the patients had hypertension and had cerebellar lesions on CT or MRI scans. With the exception of one patient with a delayed diagnosis and a patient with developmental delay before transplantation, all recovered clinically and radiologically within 10 weeks, with optimal control of hypertension and seizures, as well as withdrawal of the calcineurin-inhibitors.

In three children with posterior reversible encephalopathy associated with tacrolimus (n = 2) or ciclosporin (n = 1), electroencephalography showed continuous focal rhythmic activity, which normalized after the clinical effects had resolved [ ].

Most studies have focused on the central nervous system adverse effects of ciclosporin, and there have been few reports of peripheral neuropathy. In two patients, ciclosporin was suggested as a possible cause of an entrapment neuropathy, and surgery was required in both [ ]. However, the report did not provide sufficient evidence to assess the causal relation fully. Another patient developed a symmetric polyneuropathy with flaccid paraplegia while her ciclosporin serum concentrations were about twice normal [ ]. Electromyography showed features of axonal degeneration in the peripheral nerves and neurological symptoms improved on ciclosporin dosage reduction.

Migraine associated with ciclosporin is sometimes resistant to classical treatment and the consequences can be even more severe.

• Three young adult renal transplant patients, including two with a previous history of moderate migraine, had severe attacks of unilateral throbbing migraine associated with vomiting during ciclosporin treatment [ ]. In two patients, vomiting was severe enough to reduce compliance with the immunosuppressive regimen, and both subsequently lost their grafts. The same sequence of events was again observed after retransplantation.

Substitution by tacrolimus may be beneficial in such cases.

Severe ciclosporin neurotoxicity has mostly been reported in transplant patients, but should also be considered in non-transplanted patients.

• An 87-year-old patient with resistant nodular prurigo was successfully treated with ciclosporin (3 mg/kg/day) and prednisone (10 mg/day) [ ]. Bilateral numbness and distal limb weakness developed after 18 months. Clinical examination, electromyography, and nerve conduction studies confirmed a diffuse axonal neuropathy which rapidly progressed over the next 2 months. Ciclosporin alone was withdrawn and complete remission was observed within 3 months.

Unfortunately, ciclosporin blood concentrations and renal function at the time of diagnosis were not reported.

Very severe or fatal neurotoxicity has been reported in isolated patients only.

• Based on postmortem findings in a 32-year-old woman who died with an acute encephalopathy [ ] and another report of two patients investigated with transcranial Doppler ultrasound and MRI for symptoms of ciclosporin neurotoxicity [ ], vascular changes with vasospasm and dissection of the vascular intima strongly suggest that vasculopathy is a possible mechanism of ciclosporin-induced encephalopathy.

Prolonged confusion is a recognized complication of ciclosporin, and can be due to non-convulsive status epilepticus [ ]. Three patients who developed neurotoxicity following treatment with ciclosporin manifested with generalized tonic-clonic seizures and dysarthria. The plasma ciclosporin concentration in these patients increased as the neurological signs appeared, and the signs resolved quickly after dosage reduction [ ]. Tonic-clonic seizures have been reported in a child taking ciclosporin [ ].

• A 13-year-old boy with severe Crohn’s disease developed hematochezia and required blood transfusion. He was given ciclosporin on day 22 because of persistent rectal bleeding and diarrhea, despite high-dose intravenous glucocorticoids. After 6 days he developed multiple episodes of generalized tonic-clonic seizures, with MRI findings typical but not pathognomonic of ciclosporin: prominent meningeal enhancement, bifrontal, bitemporal, biparietal, and bioccipital cortical and subcortical white matter high-signal changes, and swelling of the gyri, which obliterated the sulci.

This case illustrates that severe ciclosporin neurotoxicity can develop in patients with predisposing factors, such as hypomagnesemia, hypocholesterolemia, hypertension, and glucocorticoid therapy.

Ocular administration

Ciclosporin eye-drops have been used after keratoplasty, in high-risk cases, to prevent graft rejection and to treat severe vernal conjunctivitis, keratoconjunctivitis sicca, and various immune-related corneal disorders. Despite its severe adverse effects after systemic use, topical ciclosporin can generally be used without serious adverse reactions [ , ].

Ciclosporin oil-in-water emulsion has been used in the local treatment of moderate to severe dry eye disease. Chronic dry eye disease results from inflammation mediated by cytokines and receptors for autoimmune antibodies in the lacrimal glands. It affects the lacrimal gland acini and ducts, leading to abnormalities in the tear film, and ultimately disrupting the homeostasis of the ocular surface. Topical ciclosporin reduces the cell-mediated inflammatory response associated with inflammatory ocular surface diseases.

In two large, randomized controlled trials in 977 patients, the adverse effects associated with ciclosporin ophthalmic emulsion for the treatment of dry eye disease were minimal and consisted mostly of mild ocular burning and stinging [ ]. However, topical application of ciclosporin eye-drops was the suspected cause of severe visual loss with bilateral white corneal deposits in a 45-year-old patient with dry eye syndrome caused by graft-versus-host disease [ ]. Infrared spectroscopy and X-ray analysis suggested that the deposits contained ciclosporin. A reduction in tear clearance and compromised epithelial barrier function caused by the concomitant use of oxybuprocaine may have precipitated this adverse effect.

The efficacy, safety, tolerability, and optimal dose of ciclosporin eye-drops have been studied in a randomized, double-masked, vehicle-controlled multicenter trial in 162 patients with keratoconjunctivitis sicca with or without Sjögren’s disease and refractory to conventional treatment [ ]. Ciclosporin ophthalmic emulsion 0.05, 0.1, 0.2, or 0.4%, or the vehicle alone was instilled twice daily into both eyes for 12 weeks, followed by a 4-week observation period. There was no clear dose–response relation; ciclosporin 0.1% emulsion produced the most consistent improvement in objective and subjective end-points and ciclosporin 0.05% gave the most consistent improvement in symptoms. The vehicle also performed well, perhaps because of its long residence time on the ocular surface. There were no significant adverse effects, no microbial overgrowth, and no residence time of the vehicle emulsion on the ocular surface. All treatments were well tolerated and the highest ciclosporin blood concentration detected was 0.16 ng/ml.

To study the efficacy and safety of ciclosporin 0.05 and 0.1% ophthalmic emulsions and their vehicle in patients with moderate to severe dry eye disease, two identical multicenter, randomized, double-masked, vehicle-controlled trials have been performed in 877 patients for 6 months [ ]. More than 76% completed the course. Ciclosporin 0.05 or 0.1% eye-drops gave significantly greater improvement than the vehicle in two objective signs of dry eye disease (corneal staining and Schirmer values). Ciclosporin 0.05% also gave significantly greater improvement in three subjective measures (blurred vision, need for concomitant artificial tears, and the physician’s evaluation of global response to treatment). There was no dose–response effect and there were no topical or systemic adverse findings.

Corneal deposition of ciclosporin can occur [ ].

• A 45-year-old woman with dry eye syndrome caused by graft-versus-host disease after bone marrow transplantation for acute leukemia was given systemic ciclosporin and topical 0.1% sodium hyaluronate, 0.3% ofloxacin, 0.1% fluorometholone, and isotonic saline. She was also given 0.4% oxybuprocaine for the relief of severe ocular pain. The bilateral corneal epithelial defects persisted even after the application of punctal plugs, and 2% ciclosporin in olive oil was added as eye-drops three times a day bilaterally. Five days later she complained of severe visual loss in association with bilateral corneal opacities, which covered the pupil and the punctal plugs bilaterally. As she did not agree to keratectomy, infrared spectroscopy and X-ray analysis were conducted on the deposits on the plugs. The spectroscopic pattern and X-ray analysis showed that the deposits had the properties of ciclosporin.

As the corneal deposits did not abate after withdrawal of the ciclosporin eye-drops, the systemic ciclosporin as well as its topical use may have contributed to the deposits. One should be aware that precipitation of ciclosporin on a compromised cornea can lead to severe visual impairment.

Diabetes mellitus

Diabetes mellitus after transplantation is recognized as an important adverse effect of immunosuppressants, and has been extensively reviewed [ ]. In one study the use of ciclosporin in immunosuppressive regimens was not associated with diabetes mellitus after transplantation (10–20%) [ ].

However, other studies have shown an increased risk. Tacrolimus + sirolimus, tacrolimus + mycophenolate mofetil, and ciclosporin + sirolimus have been compared in recipients of their first kidney transplant [ ]. One-year patient and graft survival did not differ. Ciclosporin + sirolimus was associated with a higher incidence of post-transplant diabetes mellitus, more antihyperlipidemic drug therapy, increased serum creatinine concentrations, reduced creatinine clearance, and more frequent protocol discontinuation.

The effect of long-term ciclosporin on glucose metabolism was analysed in heart transplant recipients who developed post-transplant hyperglycemia, 102 with impaired glycemic control and 20 with clinical diabetes [ ]. There was a significant negative correlation between ciclosporin concentration and insulin in both groups, a significant negative correlation between ciclosporin concentration and proinsulin, C-peptide blood concentration in those with impaired glycemic control and a significant positive correlation between ciclosporin and glucose blood concentration in both groups.

The incidence of diabetes mellitus has been investigated in a prospective multicenter study at 24 months after kidney transplantation in 1276 patients taking tacrolimus, mycophenolate mofetil, and steroids and in 507 patients taking ciclosporin, mycophenolate mofetil, and steroids [ ]. Significantly more of the patients who were taking tacrolimus (74/161, 46%) needed insulin than those who were taking ciclosporin (14/50, 28%). By multivariate Cox regression analysis, age over 60 years, a BMI over 30 kg/m, and immunosuppression with tacrolimus were associated with diabetes mellitus after transplantation.

Hyperlipidemia

Ciclosporin is potentially more toxic in patients with altered LDL concentrations or a low total serum cholesterol [ ]. Ciclosporin therapy itself significantly raises plasma lipoprotein concentrations by increasing the total serum cholesterol; this is due to an increase in LDL cholesterol, demonstrated in a prospective, double-blind, randomized, placebo-controlled trial in 36 men with amyotrophic lateral sclerosis [ ]. In 22 patients there were significant increases in mean serum triglycerides and cholesterol 2 weeks after they started to take low-dose ciclosporin [ ]. Hypertriglyceridemia developed in seven patients taking ciclosporin 2.0–7.5 mg/kg/day for psoriasis during the first month of therapy; the values were greater than the upper limit in age- and sex-matched controls [ ].

The pathology of hyperlipidemia after transplantation is multifactorial, but it is clearly dose-dependently related to immunosuppressive therapy [ ]. This results in cardiovascular disease, which is one of the most common causes of morbidity and mortality in long-term survivors of organ transplantation [ ]. Hyperlipidemia can also cause renal atheroma, resulting in graft rejection. The possible impact of ciclosporin on lipids includes an increase in total cholesterol, LDL cholesterol, and apolipoprotein B concentrations, and a reduction in HDL cholesterol [ ]. The influence of ciclosporin on lipoprotein(a) concentrations has been debated [ , ]. Post-transplant hyperlipidemia is multifactorial and can be affected by impaired renal function, diuretics and beta-blockers, increased age, and female sex. A combination of lipid-lowering drugs and optimization of immunosuppressive regimens compatible with long-term allograft survival is probably required to reduce post-transplantation hyperlipidemia.

Whereas azathioprine is considered to play no role, glucocorticoid use correlates positively with increased serum cholesterol concentrations. It is uncertain whether these lipid changes reflect primarily an effect of ciclosporin alone or an additive/synergistic effect of the drug plus glucocorticoids. Ciclosporin has been considered as a possible independent susceptibility factor by several investigators, but others were unable to find an association between hyperlipidemia and ciclosporin [ , ]. There was indirect evidence for a causal role of ciclosporin in several studies; hyperlipidemia developed in non-transplant patients taking ciclosporin alone; there was a transient reduction in hyperlipidemia after ciclosporin withdrawal; there was a significant correlation between ciclosporin blood concentrations and lipid abnormalities; and there was a higher incidence of lipid abnormalities in patients taking ciclosporin alone compared with patients taking azathioprine and prednisolone [ , ]. Other studies have provided striking evidence that hyperlipidemia is more frequent in patients taking ciclosporin than in those taking tacrolimus, with more patients classified as having high cholesterol concentrations in the ciclosporin group or a significant fall in total cholesterol or LDL cholesterol in patients switched from ciclosporin to tacrolimus [ , ]. Although the glucocorticoid-sparing effect of tacrolimus may account for these differences, the concept that the glucocorticoid dose is a confounding factor has been disputed [ ]. Whether these differences translate to a higher risk of cardiovascular complications in patients taking ciclosporin has not been carefully assessed. The treatment of hyperlipidemia in transplant patients may represent a major dilemma, because of several drug interactions, with an increased risk of myopathy and rhabdomyolysis after the combined use of ciclosporin and several lipid-lowering drugs.

Of 295 patients after renal transplantation, 76 were given tacrolimus and 126 ciclosporin [ ]. Lipid concentrations were similar in the two groups at day 0. However, 12 months later, total cholesterol and LDL cholesterol were significantly higher in those who were taking ciclosporin; triglyceride concentrations were similar in the two groups.

Hyperuricemia

Significant hyperuricemia has been observed in as many as 80% of patients taking ciclosporin [ ]. In one series, hyperuricemia occurred in 72% of male and 82% of female patients taking ciclosporin after cardiac transplantation; there was also an increased incidence of gouty arthritis in these patients [ ]. Episodes of gout developed mostly in men taking diuretics, but the incidence was lower than in the hyperuricemic population. In renal transplant patients, the incidence of gout was 5–24% and tophi sometimes developed rapidly after the onset of gout [ ]. The potential mechanisms of hyperuricemia include reduced renal function and impaired tubular secretion of acid uric, with hypertension and diuretics as confounding factors [ , ].

In 32 children (median age 14 years), all of whom received triple immunosuppressive therapy (ciclosporin or tacrolimus, azathioprine or mycophenolate mofetil, and prednisone), 47% had hyperuricemia, with uric-acid concentrations above the age-related reference range; 55% were taking ciclosporin, 30% tacrolimus [ ]. Plasma uric acid concentrations did not differ among patients taking ciclosporin or tacrolimus. There was only one case of gout, 8 years after renal transplantation in a 15-year-old boy who also taking furosemide.

Presentation

Acute renal impairment: Acute ciclosporin-induced nephrotoxicity, causing reduced renal function, develops within the first month, and includes a dose-related rise in serum creatinine concentrations and hyperkalemia. Fatal acute tubular necrosis has also been noted after very high intravenous doses [ ]. Although it is clinically often difficult to differentiate from acute allograft rejection in renal transplant patients, the alteration in renal function promptly resolves on ciclosporin withdrawal or dosage reduction, and initial acute renal insufficiency is not clearly associated with the development of subsequent chronic renal dysfunction [ ]. Several conditions, such as pre-existing hypovolemia, concomitant diuretic treatment, or renal artery stenosis, are susceptibility factors. Hypothyroidism was thought to be involved in one patient [ ], and the transplanted kidney itself rather than interindividual differences between recipients was thought to play a role [ ].

Hemolytic–uremic syndrome and thrombotic microangiopathy: Hemolytic–uremic syndrome, with histological findings of thrombotic microangiopathy and possible evolution to graft loss or death, is another instance of very severe acute nephrotoxicity [ ]. It usually occurs at between the second and fourth weeks after transplant, with associated fever, thrombocytopenia, erythrocyte fragmentation, neurotoxicity, and renal impairment. Uncommon clinical features have been reported.

• In two women, hemolytic–uremic syndrome was apparently revealed by an episode of severe acute depression [ ].

• In another patient, a single injection of ciclosporin may have induced the development of fibrin thrombi seen in the perioperative graft biopsy [ ]. Later on, she was confirmed to have clinical and biological features of hemolytic–uremic syndrome, which reversed after ciclosporin withdrawal.

Hemolytic–uremic syndrome has also been reported during ciclosporin treatment for Behçet’s disease [ ]. Both early and delayed hemolytic–uremic syndrome can occur after transplantation, and its actual incidence may have been underestimated. Thrombotic microangiopathy with clinical features of hemolytic–uremic syndrome was found in 3.5–5% of renal transplant patients [ , ]. Graft loss was mostly found in patients who developed hemolytic–uremic syndrome early after transplantation, and the clinical distinction from acute rejection can be very difficult. Hemolytic–uremic syndrome does not recur after initial withdrawal and further ciclosporin reintroduction, once renal function has normalized, or even despite ciclosporin maintenance with dosage reduction [ ]. In some cases, the patient was successfully switched to tacrolimus, and only one case of hemolytic–uremic syndrome with recurrence on ciclosporin rechallenge has been reported [ ]. In contrast, both ciclosporin and tacrolimus were significant susceptibility factors for recurrence of hemolytic–uremic syndrome in patients who had undergone renal transplantation for end-stage renal disease [ ].

The factors that contribute to the development of thrombotic microangiopathy have been retrospectively investigated in 50 of 188 patients with kidney or kidney + pancreas allografts who underwent graft biopsies and in 19 control patients who had never had renal graft dysfunction or a biopsy [ ]. There were definite histological features of thrombotic microangiopathy 4 days to 6 years after transplantation in 26 patients, of whom 24 were taking ciclosporin and two were taking tacrolimus, showing that this complication can occur at any time after transplantation. Eight patients had graft loss, but only two had associated systemic evidence of microangiopathy, that is thrombocytopenia and intravascular hemolysis, suggesting that thrombotic microangiopathy should be considered in any patients with renal graft dysfunction, even if there are no suggestive systemic symptoms. Although the more frequent use of the microemulsion form of ciclosporin (Neoral) in patients with confirmed thrombotic microangiopathy than in controls suggested a possible role of this formulation, this issue remains to be further investigated, because the number of evaluable patients was small. None of the other investigated variables (age, sex, race, living-related or cadaveric donor status, the degree of HLA mismatch, the type of allograft, or the incidence of urinary tract infections after transplantation) was significantly associated with the occurrence of thrombotic microangiopathy compared with patients without thrombotic microangiopathy. Finally, the most successful strategy was a switch from ciclosporin to tacrolimus, which resulted in normalization of graft function in 81% of these patients.

Chronic renal insufficiency: Chronic renal impairment, as first reported in cardiac transplant patients [ ], is of major concern, because of possible irreversible renal dysfunction. A considerable amount of work has subsequently accumulated on the development of progressive renal dysfunction in patients receiving long-term ciclosporin for organ transplantation or chronic inflammatory disease, and there have been several comprehensive reviews [ ]. About one-third of all patients have increased serum creatinine concentrations and reduced glomerular filtration rate during ciclosporin maintenance therapy. The histopathological features of chronic nephropathy consist mostly of non-specific tubular atrophy and interstitial fibrosis [ ]; arteriolar lesions are considered very suggestive of ciclosporin nephrotoxicity. The prevalence of renal damage due to ciclosporin has fallen considerably since the use of lower doses. Arteriolopathy sometimes improves after reducing or withdrawing ciclosporin. Morphological features in patients with autoimmune diseases are non-specific, and include a wide range of lesions, mostly characterized by tubulointerstitial changes and arteriolopathy. There is no significant correlation between histological findings and ciclosporin dose. Severe histological lesions can be identified in some patients with normal renal function [ ]; the severity of tubulointerstitial lesions has been deemed to be a better index than the glomerular filtration rate for predicting the occurrence of chronic nephropathy, but in one study there was no correlation between histological renal findings and various measures of renal function [ ].

Because the possibility of irreversible renal dysfunction is a major problem in ciclosporin maintenance in both transplant and non-transplant patients, this issue continues to receive attention. Even though many studies have been performed, the long-term prognosis is a source of conflicting opinions, and the initial assumption that long-term use of ciclosporin will sooner or later cause irreversible chronic nephropathy is hotly debated. It is still unclear to what extent long-term ciclosporin contributes to progressive renal insufficiency and whether chronic ciclosporin nephropathy is irreversible or improves after dosage reduction. A retrospective analysis of more than 12 000 renal transplant patients showed that long-term maintenance with a glucocorticoid-free ciclosporin regimen (ciclosporin alone or with azathioprine) significantly increased renal graft and patient survival, compared with patients taking other immunosuppressive regimens [ ]. This allowed the use of higher doses of ciclosporin without increasing the frequency of nephrotoxicity. In contrast, several investigators have considered a change to a ciclosporin-free regimen in 40% of patients, because of progressive renal deterioration with histological signs of nephrotoxicity [ ]. The incidence of end-stage renal insufficiency requiring dialysis or renal transplantation ranges from 1% in renal transplant patients to 3–6% in heart-transplant patients [ ]. Nevertheless, several investigators have shown that despite an initial reduction in renal function, serum creatinine concentrations stabilized with no strong evidence of progressive nephropathy after several years of surveillance in various organ transplant patients [ , , ].

The potential long-term consequences of ciclosporin nephrotoxicity constitute a major disadvantage in non-transplant patients. Renal function was assessed 7 years after the end of a 1-year ciclosporin treatment period in 36 young patients from a randomized, placebo-controlled trial of ciclosporin in diabetes mellitus, 19 taking ciclosporin, and 17 taking placebo [ ]. Blood pressure did not differ between the groups. Compared with baseline values, urinary albumin excretion rate was significantly higher and estimated glomerular filtration rate significantly lower with ciclosporin. The results in the placebo group showed no change or increases. In addition, there was progression to micro- or macro-albuminuria in four patients taking ciclosporin, and two of five patients who underwent renal biopsy had arteriolar hyalinosis. It is not known whether these changes will translate to an increased risk of nephropathy, but they suggest that ciclosporin might enhance it. Of 91 consecutive patients with renal transplants with a minimum graft survival of 1 year who were followed for 7–8 years, 65% had stable renal function despite ciclosporin serum concentrations of 200–250 ng/ml [ ]. In addition, none of the 26 patients with worsening renal function had features of ciclosporin nephrotoxicity on renal biopsy.

Frequency

In a meta-analysis of 18 trials involving ciclosporin doses below 10 mg/kg/day for at least 2 months in autoimmune diseases, the weighted percentage increase in serum creatinine concentrations was 17% in ciclosporin-treated patients and 1.7% in controls [ ]. The corrected risk difference for an increase of more than 50% of pretreatment serum creatinine concentrations between the two groups was 21% (95% CI = 12, 30) [ ]. This meta-analysis did not fully consider the long-term outcome, but clinical and histological evidence of sustained or progressive ciclosporin nephropathy in this population continues to accumulate. Unfortunately, some of the findings are discordant [ , ].

In a retrospective study of 106 patients following renal transplantation who had been treated with ciclosporin, 85% were hypertensive compared with 54% of patients taking azathioprine [ ]. Renal function was significantly better in hypertensive patients treated with nifedipine than with other antihypertensive medication (beta-blockers and vasodilators), and it was similar to that of normotensive patients treated with ciclosporin.

Pathophysiology

The pathogenesis of chronic ciclosporin nephrotoxicity is not fully understood [ ]. Intrarenal afferent arteriolar vasoconstriction may play an important part, particularly in acute nephrotoxicity [ ], in which a marked reduction in renal blood flow associated with an increase in renal vascular resistance, probably due to postglomerular vasoconstriction, has been demonstrated [ , ]. The supposed mechanism is primarily an imbalance between several regulatory mechanisms of renal vasodilatation and vasoconstriction, leading to increased renal vasoconstriction, and the explanations that have been proposed include activation of the renin–angiotensin system, prostaglandin inhibition, and sympathetic nervous system activation [ ]. However, it is unclear whether a continuous increase in renal vascular resistance can account for chronic renal dysfunction in patients taking long-term ciclosporin. There are many possible mediators of renal vasoconstriction, for example nitric oxide, the renin–angiotensin and kallikrein–kinin systems, endothelin-1 release, and stimulation of sympathetic nervous system activity. A major effect of ciclosporin is to promote calcium accumulation in the mitochondrial matrix, which in turn reduces ATP synthesis [ ]. The main morphological abnormality that has been demonstrated in the kidneys of patients taking long-term ciclosporin is interstitial fibrosis. Vascular lesions, predominantly arteriolar, with arterial intimal fibrosis have been noted in renal biopsies from patients with chronic ciclosporin nephrotoxicity [ ].

Renal morphology has been studied in 17 patients who received ciclosporin for sight-threatening uveitis. Most had not received other potentially nephrotoxic drugs. Variable interstitial fibrosis, frequently associated with tubular atrophy, was noted in all 17. The extent of the pathological changes did not correlate with the age, treatment duration, or average cumulative dose [ ]. Ciclosporin nephrotoxicity mimics the histological features of acute allograft rejection and tubular necrosis. It is important to be able to distinguish clinically between ciclosporin toxicity on the one hand (necessitating a reduction in dose) and rejection (requiring an increase in dose) on the other.

The long-term effects of ciclosporin on renal function in 11 liver transplant recipients were evaluated over a follow-up period of 6–26 months [ ]. Immediately postoperatively, glomerular filtration rate (GFR) and effective renal plasma flow (ERPF) fell by 60%, subsequently settling at 45–60% of normal. There were additional toxic effects on renal tubular function. Histopathological findings were mild to moderate; notably, arterial and arteriolar nephrosclerosis. Renal function improved as the dose of ciclosporin was reduced, despite continued administration of the drug. This suggests a persistent, potentially reversible, functional component to chronic ciclosporin nephrotoxicity.

Treatment with calcineurin inhibitors is often associated with chronic renal insufficiency and conversion to mycophenolate may improve renal function [ ]. In a single-center study 49 patients who underwent liver transplantation between 1986 and 2002 were given a combination of calcineurin inhibitors and glucocorticoids; azathioprine was used as adjuvant therapy in 71%, but was withdrawn before the introduction of mycophenolate, starting at 500 mg bd during the first week, followed by an increase every week up to 1 g bd. The daily dose of calcineurin inhibitor was reduced by 10–20% stepwise every 4 weeks after the optimal dose of mycophenolate was reached; the calcineurin inhibitors were withdrawn if possible within 6 months. Mean creatinine clearance was 43 ml/minute when mycophenolate was introduced and in 45 patients it was less than 60 ml/minute. Of 39 patients who responded to a reduced dose of calcineurin inhibitor, 18 recovered completely and 21 had partial recovery of renal function; 10 were non-responders; 14 patients in whom the calcineurin inhibitor was completely withdrawn at 1 year after mycophenolate introduction tended to have a higher increase in creatinine clearance (22 ml/minute). Patients with alcoholic cirrhosis had the worst deterioration in renal function and a smaller improvement after dosage reduction. In one patient, mycophenolate was withdrawn because of severe diarrhea. In 32 patients other adverse effects were easily controlled with symptomatic treatment or a reduction in dosage.

In 40 patients with renal dysfunction (creatinine clearance 40–80 ml/minute) who randomized to continue calcineurin inhibitor therapy or switch to sirolimus renal function improved in the latter within 3 months (75 versus 56 ml/minute) [ ]. At 12 months this difference persisted (72 versus 58 ml/minute) but was not significant. Two patients developed steroid-sensitive rejection, one in each arm. Other adverse effects were mild, and included mouth sores (25%), hyperlipidemia requiring treatment (15%), and pruritus (5%).

The biochemical basis of nephrotoxicity due to calcineurin inhibitors and their interaction with inhibitors of mTOR, such as tacrolimus is still poorly understood. However, there is evidence that nephrotoxicity is caused by drug-induced mitochondrial dysfunction and that inhibitors of mTOR enhance the negative effects of calcineurin inhibitors on cell energy metabolism [ ].

Although the pathogenesis of ciclosporin nephrotoxicity is not completely defined, there is evidence that suggests a role of reactive oxygen species. In numerous in vivo and in vitro experiments ciclosporin caused renal insufficiency and increased the synthesis of reactive oxygen species, thromboxane, and lipid peroxidation products in the kidney. Furthermore, it modified the expression and activity of several renal enzymes (cyclo-oxygenase, superoxide dismutase, catalase, and glutathione-peroxidase). Antioxidant nutrients (for example vitamins E and C) can neutralize some of the effects that ciclosporin produces in the kidney [ ]. Thus, vitamin E inhibited the synthesis of reactive oxygen species and thromboxane and the lipid peroxidation process induced by ciclosporin. Antioxidants can also improve renal function and histological damage produced by ciclosporin. Although there are few data in humans taking ciclosporin, it is possibility that antioxidants also neutralize ciclosporin nephrotoxicity and LDL oxidation. Thus, antioxidant nutrients could have a therapeutic role in transplant patients taking ciclosporin.

The respective roles of organ preservation and ciclosporin in the pathogenesis of post-transplant renal damage have been studied in an in vitro model that simulates the hypothermic kidney preserved before surgery in Collins’ solution and exposed after transplantation to ciclosporin [ ]. The results showed that preservation sensitizes the kidney to ciclosporin injury, which is consistent with clinical experience [ ]. If the preserved kidney cells were given a period of repair before administration of ciclosporin, further injury did not happen. In animal experiments, prolonged cold preservation causes progressive deterioration in the renal cortical microcirculation; concentration of ciclosporin in the renal cortex of hypoperfused kidneys markedly potentiates the vascular damage caused by cold preservation [ ].

Ciclosporin-induced renal hypoperfusion was detected by quantitative cine-loop color Doppler imaging after kidney transplantation in 22 patients [ ]. They were taking ciclosporin + mycophenolate mofetil + prednisolone (n = 7), tacrolimus + mycophenolate mofetil + prednisolone (n = 7), or ciclosporin + a calcium channel blocker (n = 8). The mean effect occurred 1.1 hours after ciclosporin dosing and was prevented by calcium channel blockers. Main renal artery velocities, resistive index, and small vessel perfusion were unchanged, suggesting that medium-sized arteries mediated vasoconstriction. In contrast, tacrolimus did not alter renal vascularity.

Ciclosporin-induced nephrotoxicity has been studied in prospective protocol kidney biopsies (n = 888) from 99 patients taken regularly for 10 years after kidney transplantation [ ]. The most sensitive histological marker of ciclosporin-induced nephrotoxicity was arteriolar hyalinosis. Structural nephrotoxicity occurred in two phases, with different clinical and histological characteristics. The acute phase occurred with a median onset of 6 months after kidney transplantation, was usually reversible, and was associated with functional nephrotoxicity, high ciclosporin blood concentrations, and mild arteriolar hyalinosis. The chronic phase persisted over several biopsies and occurred at a median onset of 3 years. It was associated with lower ciclosporin doses and trough concentrations, was largely irreversible, and was accompanied by severe arteriolar hyalinosis and progressive glomerulosclerosis. These pathological changes exacerbated chronic allograft nephropathy and strategies to ameliorate or avoid nephrotoxicity are therefore urgently needed.

Chronic interstitial fibrosis is an adverse prognostic feature of chronic allograft nephropathy. It has been analysed in protocol kidney biopsies (n = 959) obtained regularly for 10 years after kidney transplantation [ ]. There was substantial interstitial fibrosis within 1 year after kidney transplantation, with maximum intensity within the first 3 months as a result of early ischemia–reperfusion injury and acute, subacute, or persistent interstitial inflammation. Ciclosporin increased the risk of interstitial fibrosis compared with tacrolimus, and mycophenolate mofetil was protective compared with azathioprine.

Large-scale studies with long periods of follow-up have emphasized the major role of graft arteriopathy (chronic graft rejection) rather than chronic ciclosporin nephrotoxicity as the primary cause of graft failure [ , , ]. Severe ciclosporin nephropathy was the cause of renal transplant failure in less than 1% of patients.

Susceptibility factors and prediction

Many factors have been postulated as being relevant to ciclosporin nephrotoxicity. Whereas in several studies initial high doses of ciclosporin increased the risk of chronic nephrotoxicity [ , ], others suggested that patients maintained on relatively high ciclosporin concentrations had no more chance than others of developing toxic nephropathy [ ]. Neither the daily dose nor the duration of ciclosporin treatment reasonably predicts the risk of chronic renal insufficiency. Chronic renal dysfunction can be observed, despite the maintenance of ciclosporin blood concentrations below 400 ng/ml. However, age, sustained hypertension, hypertriglyceridemia, low HDL cholesterol concentrations, and recurrent episodes of severe acute nephrotoxicity increase susceptibility to chronic ciclosporin nephrotoxicity [ , ].

From a prospective study in 36 heart transplant patients with stable renal function for at least 6 months after transplantation, it was suggested that high urinary retinol-binding protein concentrations may indicate tubulointerstitial damage and therefore detect patients who are at risk of ciclosporin nephrotoxicity [ ]. At the start of the study, 13 patients had high urinary retinol-binding protein concentrations and 23 had normal concentrations. After 5 years of follow-up, five of the 13 patients developed end-stage renal insufficiency requiring dialysis, whereas none of the 23 other patients had terminal renal insufficiency. Although these data await confirmation, the authors suggested that ciclosporin dosage reduction should be considered in patients with high urinary retinol-binding protein concentrations, in order to limit renal damage.

Ciclosporin can cause tubulointerstitial lesions, the pathogenesis of which is unclear. In 37 patients, the duration of ciclosporin treatment and of heavy proteinuria were independent risk factors for ciclosporin-induced tubulointerstitial disease [ ].

P glycoprotein contributes substantially to ciclosporin nephrotoxicity. The TT genotype at the ABCB1 3435C → T polymorphism is associated with reduced expression of P glycoprotein in renal tissue and has been implicated as a susceptibility factor for ciclosporin nephrotoxicity. Ciclosporin nephrotoxicity in 18 of 97 patients completely recovered after switching to a calcineurin inhibitor-free regimen [ ]. The P glycoprotein low expressor genotype 3435TT in the kidney donors, but not in the recipients, was over-represented in the cases of ciclosporin nephrotoxicity. Ciclosporin dosage, trough concentrations, and the concentration per dose ratio were not different between the groups. In a multivariate model that included several other non-genetic covariates, only the donor′s ABCB1 3435TT genotype was strongly associated with ciclosporin nephrotoxicity.

Comparative studies

The effect of microemulsion ciclosporin and tacrolimus on the development of renal allograft fibrosis has been evaluated in a randomized trial in 102 kidney recipients randomized to either microemulsion ciclosporin 15 mg/kg/day (trough concentration range 200–300 ng/ml) or tacrolimus 0.2 mg/kg/day (trough concentration range 8–15 ng/ml) in conjunction with glucocorticoids, or at a lower dose (ciclosporin 7 mg/kg/day and tacrolimus 0.1 mg/kg/day), with the addition of azathioprine for recipients of non-heart-beating kidneys [ ]. There was a significant increase in allograft interstitial fibrosis in the patients who used microemulsion ciclosporin compared with tacrolimus. There was no significant difference in the demographic characteristics between the patients or in the incidence of acute rejection (ciclosporin 36% versus tacrolimus 35%) or glucocorticoid-resistant rejection (both 10%). The incidence of insulin resistance was higher with tacrolimus, but this was not significant. Ciclosporin was associated with significant increases in total cholesterol and low-density lipoprotein concentrations, which persisted throughout the study.

In 44 patients randomized 3 months after kidney transplantation to ciclosporin + sirolimus + prednisolone (group I) or to sirolimus + prednisolone (group II), baseline graft biopsy showed a higher degree of renal damage in group II [ ]. At 1 year after transplantation, chronic allograft nephropathy was diagnosed in 55% of the patients, of whom 64% were in group I and 36% in group II. Lesions of chronic allograft nephropathy were scored as moderate to severe in 90% of group I patients but only 32% of group II patients. There was a vascular score greater than or equal to 2 in 90% of group I patients and in 38% of group II patients. Group I patients had significantly worse kidney graft function (serum creatinine 177 versus 115 μmol/l). The authors suggested that early withdrawal of ciclosporin is a safe option that allows significant reduction of chronic histological damage, particularly vascular injury, to cadaveric kidney allografts.

Kidney recipients (n = 31) with biopsy-confirmed chronic allograft nephropathy were prospectively randomized to receive a 40% ciclosporin dosage reduction either with sirolimus 2 mg/day (n = 16) or without (control, n = 15) [ ]. Sirolimus did not improve functional, molecular, or histological outcomes in patients with chronic allograft nephropathy after ciclosporin dosage reduction.