The clinical management of severe ulcerative colitis

Transcription

1 GASTROENTEROLOGY 2003;125: Randomized, Double-Blind Comparison of 4 mg/kg Versus 2 mg/kg Intravenous Cyclosporine in Severe Ulcerative Colitis GERT VAN ASSCHE,* GEERT D HAENS,* MAJA NOMAN,* SÉVERINE VERMEIRE,* MARTIN HIELE,* KATRIEN ASNONG,* JORIS ARTS,* ANDRE D HOORE, FREDDY PENNINCKX, and PAUL RUTGEERTS* *Department of Gastroenterology and Abdominal Surgery, University of Leuven, Leuven, Belgium Background & Aims: Cyclosporine A is highly effective in severe attacks of ulcerative colitis (UC) but is associated with important adverse effects that are mainly dose dependent. Our single center, randomized, double-blind, controlled trial aimed to evaluate the additional clinical benefit of 4 mg/kg over 2 mg/kg IV cyclosporine in the acute treatment of severe UC. Methods: Primary end point was the proportion of patients with a clinical response. Secondary end points included time to response, colectomy rate, and adverse effects. Results: Seventy-three patients were included. Day-8 response rates were 84.2% (32 of 38, 4 mg/kg) and 85.7% (32 of 35, 2 mg/kg) after a median of 4 days in both groups. Short-term colectomy rates were 13.1% (4 mg/kg) and 8.6% (2 mg/kg). Mean cyclosporine blood levels were in the 2-mg/kg group and ng/ml in the 4-mg/kg group. Active smoking was inversely correlated with clinical response (odds ratio, 0.06), but concomitant azathioprine or steroids were not predictive. A trend toward a higher incidence of hypertension was observed in the 4-mg/kg group (23.7% vs. 8.6%, 2 mg/kg, P < 0.08). Conclusions: High-dose IV cyclosporine has no additional clinical benefit over low dose in the treatment of severe UC. Although we did not observe differences in adverse effects on the short term, the use of 2 mg/kg IV cyclosporine should provide an improved toxicity profile for medical treatment of severe UC. The clinical management of severe ulcerative colitis (UC) remains a major challenge in that 15% of UC patients will experience a severe flare-up in the course of their disease, usually requiring hospitalization. These patients face the risk of life-threatening complications such as toxic megacolon, perforation, and major colonic bleeds. 1 The standard treatment for patients with acute severe colitis in most centers consists of intravenous corticosteroids. This treatment has been successful in approximately 60% of patients, whereas nonresponding patients were inevitably colectomized. 2,3 In 1990, Lichtiger and Present reported on the efficacy of intravenous cyclosporine in the treatment of severe UC. 4 This observation has been confirmed by several uncontrolled trials. 5 8 In the first, and only, placebo-controlled trial, 4 mg/kg of IV cyclosporine, given to patients not responding to intravenous (IV) glucocorticosteroids, induced a substantial clinical improvement in 9 of 11 patients, whereas none of the 9 placebo-treated patients improved despite continued IV steroid therapy. 9 Moreover, a recent, double-blind, controlled study from our group has demonstrated that cyclosporine as single therapy is at least as effective as corticosteroids in the treatment of severe UC flare-ups. 10 Cyclosporine is an immunosuppressive macrolide that inhibits the production of interleukin 2 by activated T lymphocytes through a calcineurin-dependent pathway. In addition, it down-regulates the synthesis of other inflammatory cytokines. 11 The proposed role of cyclosporine in the management of acute severe colitis has been to induce clinical remission as an intravenous therapy and further on as an oral agent to bridge the gap needed for the full efficacy of azathioprine or 6-mercapropurine. 5 Indeed, because of the unfavorable toxicity profile associated with long-term cyclosporine therapy, azathioprine is usually started when patients are commenced on oral cyclosporine and discharged from the hospital. The IV starting dose of 4 mg/kg was used in the placebo-controlled trial by Lichtiger et al. 9 and consecutively in most of the uncontrolled trials. However, in solid organ transplantation and in other immunity-related disorders, lower IV cyclosporine doses have been successfully used. For most of the adverse effects associated with the use of IV cyclosporine such as hypertension, impaired renal function, and neurotoxicity, a dosedependent relationship has been demonstrated Because of the toxicity profile associated with the use of cyclosporine, which includes mortality in a significant Abbreviation used in this paper: UC, ulcerative colitis by the American Gastroenterological Association /03/$30.00 doi: /s (03)

2 1026 VAN ASSCHE ET AL. GASTROENTEROLOGY Vol. 125, No. 4 minority of patients, the benefit of avoiding colectomy should always be balanced against the risk of inducing profound immunosuppression. The aim of the present study, therefore, was to investigate whether a high-dose IV cyclosporine induction regimen of 4 mg/kg was superior to a lower dose regimen of 2 mg/kg at alleviating signs and symptoms of acute severe colitis and at avoiding colectomy. In addition, we wanted to compare the toxicity profile associated with the 2 dosing strategies. Materials and Methods Patient Selection and Eligibility We designed a single center, randomized, controlled trial in patients admitted for severe UC to the Leuven University Hospital. Male and female patients between 18 and 70 years of age with an attack of severe UC as defined by a score of 10 or more in the Lichtiger clinical activity index (CAI; Lichtiger-modified Truelove and Witts criteria) were eligible. In previous studies, patients with a CAI of 10 were shown to have severe disease because they needed hospitalization and IV treatment. 4,9 Flexible sigmoidoscopy with biopsies was performed on the day of admission and repeated at the end of the IV cyclosporine treatment period. Plain abdominal x-ray was done to exclude toxic megacolon or perforation. A stool culture was obtained including ova/parasites and a specific determination of Clostridium difficile toxin. If a microbial or parasitic enteric pathogen was found, patients were not eligible. Other criteria for exclusion included renal insufficiency with a serum creatinine of more than 2 mg/dl, elevation of liver enzymes or bilirubin ( 2 times upper limit of normal), serum cholesterol below 150 mg/dl, uncontrolled hypertension, active viral or bacterial infections, and pregnancy. Randomization and Drug Administration Randomization was performed on patient inclusion at the central pharmacy of the hospital. Patients assigned to the high-dose group were started on a continuous 24-hour infusion of Sandimmune (cyclosporine-a; Novartis, Basel, Switzerland) at an initial dose of 4 mg/kg. Patients in the low-dose group started at a dose of 2 mg/kg IV. From day 1 through day 8, patients were treated with continuous cyclosporine infusions. Blood levels were controlled by an independent physician (M.H.), who was not involved in the care of the patients. Dose changes were directly communicated to the central pharmacy for the next IV infusion bag to obtain blood levels between 250 and 350 ng/ml in the 4-mg/kg-dose group and between 150 and 250 ng/ml in the 2-mg/kg-dose group. All other physicians and the patients were blinded for treatment assignment. They were not blinded for serum creatinine levels or blood pressure measurements. Blood cyclosporine levels were determined using a monoclonal antibody-based fluorescence polarization immunoassay technique (Abbott laboratories, Abbott Park, IL). Using this assay at our institution, blood levels between 150 and 350 ng/ml are considered therapeutic. During the first week of treatment, abdominal x-ray and blood analysis, including inflammatory parameters, magnesium, and glucose levels, and renal and liver function tests were regularly performed, and clinical scores with physical examination were obtained daily. At day 8, all responding patients (see end points) were switched to 8 mg/kg oral cyclosporine (Neoral Sandimmune; Novartis), and fasted blood levels were maintained between 150 and 300 ng/ml in both groups for 3 months. Nonresponding patients were offered to enter an open-phase treatment arm with 4 mg/kg IV cyclosporine for a maximum of 8 additional days. Restorative proctocolectomy was performed at any time during the blinded or unblinded phase of the study when considered clinically necessary by treating physicians and surgeons. Prophylaxis with sulfamethoxazol/trimethoprim 800/160 for the prevention of Pneumocystis pneumonia was started on day 8 and continued until the end of Neoral (Novartis) therapy. The study protocol was approved by the ethical committee of the Leuven University Hospital, and all patients gave written informed consent prior to any study-related procedure. Concomitant Medications Intravenous corticosteroids were allowed if given prior to enrollment at a stable dose for at least 5 days without clinical response and were kept stable until day 8 of the trial. Patients on oral corticosteroids were eligible if they had been started at least 14 days from inclusion without clinical benefit. Oral corticosteroids were discontinued on day 1, and patients were converted to IV steroids. At day 8, patients conversion to oral steroids was again performed, and steroids were tapered by 5 mg of prednisolone (or equivalent) per week. Azathioprine or 6-mercaptopurine was allowed if they had been started at least 3 months prior to inclusion and the dose had not been changed in the 4 weeks before admission. In those patients, doses were kept stable throughout the study. In all other patients, azathioprine mg/kg (Imuran; Glaxo-Smith Kline, Middlesex, United Kingdom) was initiated at day 8 and continued with regular monitoring for toxicity. Oral mesalamine or sulphasalazine was maintained at stable doses, and rectal mesalamine was also maintained at identical doses for the first 8 days, provided the patient was able to retain the enema. Patients receiving antibiotics at inclusion were continued on the antibiotics if judged clinically necessary, and, during the study, institution of antibiotics was only allowed for intercurrent infections. Endoscopic Analysis Flexible sigmoidoscopy was performed at inclusion and at day 8 in all patients. A consultant gastroenterologist, blinded to treatment assignment performed all procedures, and introduction of the endoscope was stopped at the first sign of discomfort from the patient. Endoscopy scores were assessed using the Mayo scoring system. 16 Grade 0 is normal or healed mucosa; grade 1 is mild inflammation with fading of the vascular pattern and contact

3 October 2003 IV CYCLOSPORINE IN SEVERE ULCERATIVE COLITIS 1027 Table 1. Patient Characteristics Group Table 2. Efficacy End Points of the Study Group 4 mg/kg 2 mg/kg N (male/female) 38 (21/17) 35 (21/14) Age (yr) Concomitant steroids 55.2% (21) 60.0% (22) Concomitant azathioprine 21.0% (8) 25.7% (9) Active smokers 10.5% (4) 11.4% (4) Disease extension (% pancolitis) 42% 48% Median CAI at D0 13 (10 17) 11 (10 16) Mean CRP SD at D NOTE. All values were not significantly different between treatment groups. bleeding; grade 2, moderate inflammation, is characterized by loss of vascular pattern, erosions, and contact bleeding; grade 3, severe disease, is marked by ulcers and spontaneous bleeding. Study End Points and Data Analysis The proportion of patients with a clinical response in each group was the primary end point. Clinical response was defined as a score of less than 10 at day 8 with a drop of 3 as compared with baseline. 2 Tests with Yates correction were used to analyze proportions. Sample size estimates showed that, with a sample size of 35 patients in each group, a 30% difference in the proportion of clinical responders could be demonstrated with 80% power ( -error, 0.05), based on the assumption that 82% of patients would respond to 4 mg/kg 6 and 50% to 2 mg/kg IV cyclosporine. Secondary end points included colectomy rates, median change in clinical activity index, median time to response, incidence of hypertension, and mean increase in serum creatinine. All patients were analyzed on an intention-to-treat basis. For quantitative data, statistical analysis was performed using 1-way analysis of variance for multiple comparisons, followed by a 2-tailed, paired t test for parametric, or Wilcoxon Rank sum test for nonparametric observations. Statistical significance was accepted at a P value Mulivariate analysis with stepwise logistic regression was performed to test for parameters influencing clinical response. Evaluated parameters were concomitant steroids and immunosuppressant, age, smoking, disease extension, and IV cyclosporine dose. Results Patient Characteristics and Clinical Response Between August 1996 and April 2002, 73 patients, all white, were included in this study: thirty-eight received 4 mg/kg daily and 35 received 2 mg/kg of cyclosporine daily. As indicated in Table 1, the mean age, sex ratio, smoking ratio, proportion of patients on steroids or azathioprine, disease extent, median CAI, and 4 mg/kg 2 mg/kg Response rate 84.2% (32/38) 85.7% (30/35) Median time to response 4 (1 7) days 4 (1 8) days Colectomy rate (within 14 days) 13.1% (5/38) 8.6% (3/35) Median drop CAI (day 8 vs. day 0) 7 ( 1 to 12) 6 (0to11) Mean drop in CRP (day 8) (mg/ml) Median endoscopy score D0 2 (1 3) 2 (1 3) D8 2 (1 3) 2 (1 3) NOTE. All values were not significantly different between treatment groups. mean CRP values at inclusion were not different between the groups. The disease extent was based on a total colonoscopy in all patients at diagnosis or in their past history because, at inclusion, only a sigmoidoscopy or left-sided colonoscopy was performed. One patient out of 38 in the 4-mg/kg group vs. 3 out of 35 in the 2-mg/kg group reported smoking cessation within 6 months from entry in the study. The primary end point clinical response was reached by 32 of 38 (84%) in the 4-mg/kg group and by 30 of 35 (85%) in the 2-mg/kg dose group (Table 2). One patient in the 4-mg/kg group had an anaphylactic reaction immediately after starting the first infusion and was withdrawn from the study. This patient was treated with oral cyclosporine and did well. The median change in CAI was 7 (95% CI, ) in the 4-mg/kg group and 6 (95% CI, ) in the 2-mg/kg group. The median time to response was 4 days in both groups (4 mg/kg: range, 1 7 days; 95% CI, ; 2 mg/kg: range, 1 8 days; 95% CI, ; Figure 1). Figure 1. Cumulative response to IV cycloporine in the 2 treatment groups. The percentage of responding patients is expressed at daily intervals from D0 to D8.

4 1028 VAN ASSCHE ET AL. GASTROENTEROLOGY Vol. 125, No. 4 Figure 2. Difference in mean cyclosporine doses and blood concentrations between the 2 treatment groups. Data are expressed as means SD. Blood levels were strictly controlled to vary between 150 and 250 ng/ml for the 2-mg/kg group and 250 and 350 for the 4-mg/kg group. Short-term (14 days) colectomy rates were similar (Table 2). The mean change in c-reactive protein level at D8 as compared with baseline was similar in the 2 groups (4 mg/kg, mg/l; 2 mg/kg, mg/l). Active smoking, mean cylosporine dose, age, location of disease (left-sided vs. pancolitis), and concomitant steroid and azathioprine therapy were evaluated for their predicting value toward response. In multivariate analysis, only active smoking was inversely correlated with clinical response (OR, 0.06; 95% CI, ). Differences in Cyclosporine Doses and Blood Levels For the cyclosporine infusions at day 0 and day 1, doses were strictly adjusted to 4- or 2-mg/kg body weight. Afterwards, adjustments were made based on blood levels using predefined criteria. The mean daily doses administered to patients over 8 days were (2-mg/kg group) and mg/kg (4-mg/kg group), respectively (P ; Figure 2). The area under the curve for the doses during 8 days was (2 mg/kg) vs mg/kg/day (4 mg/kg) (P Figure 3. Novel cases of diastolic hypertension in the 2 treatment groups. Diastolic hypertension was defined as a diastolic blood pressure of 90 mm Hg for at least 2 days from D1 to D ). Mean cyclosporine blood levels throughout 8 days of treatment were ng/ml in the 2-mg/kg group and ng/ml in the 4-mg/kg group (P ) (Figure 2). The area under the curve for the cyclosporine levels was ng/ml/day (2 mg/ kg) and ng/ml/day (4 mg/kg), respectively (P ). Table 3 shows the differences for blood levels and cyclosporine doses at different time points from D0 to D8. Adverse Effects Spontaneously reported adverse effects by the patients and findings on daily clinical evaluation were filed in the clinical records. Neurologic adverse effects (tremor or paresthesia) were found in 7.9% (3 of 38, 4 mg/kg) and in 5.7% (2 of 35, 2 mg/kg) of patients, respectively. Seizures did not occur. Novel cases of hypertension were noted in 9 of 38 (23.7%) patients in the high-dose groups vs. 3 of 35 (8.6%) in the low-dose group (P 0.08, 2, Figure 3). Hypertension was defined as a diastolic blood pressure of more than 90 mm Hg or a Table 3. Differences in Cyclosporine Dose, Blood Levels, and Serum Creatinine at Different Time Points Treatment group (mg/kg) D0 D2 D4 D6 D8 Mean cyclosporine level (ng/ml) a b a b Mean cyclosporine dose (mg/kg) a a a c Mean serum creatinine (mg/dl) NOTE. Values are given as mean SD. At day 0 and day 1, cyclosporine doses were calculated per kilogram body weight. From day 2 until day 8, blood levels were strictly controlled to vary between 150 and 250 ng/ml for the 2-mg/kg and 250 and 350 for the 4-mg/kg group. The cyclosporine dose for day 8 was calculated on day 7. a P b P c P 0.05 (Student t test, 2- vs. 4-mg/kg group).

5 October 2003 IV CYCLOSPORINE IN SEVERE ULCERATIVE COLITIS 1029 Table 4. Adverse Events of Patients systolic blood pressure of more than 140 mm Hg on 2 or more days from day 1 through day 8. An increase in serum creatinine of 10% or more was observed in 6 of 35 (17.1%) in the 2-mg/kg group and 7 of 38 (18.4%) in the 4-mg/kg group, respectively. None of the patients had an increase of 30% or more. The mean change in serum creatinine from day 8 as compared with day 1 was not different in both groups (4 mg/kg, ; 2 mg/kg, ). In the 2 patients who had moderate increases in serum creatinine at day 0, no changes in creatinine levels were found. Other adverse effects included fever, headache, and diabetes mellitus. The prevalence of these adverse effects was not different between both groups (Table 4). Endoscopic Evolution Sigmoidoscopic evaluation was performed in all patients on day 0 and day 8 (end of IV treatment). No significant changes in endoscopic scores were observed. The median score was 2 in both groups at day 0 and at day 8 (4 mg/kg: day 0, ; day 8, ; 2 mg/kg: day 0, ; day 8, ). Discussion Group 4 mg/kg 2 mg/kg Hypertension 9/38 3/35 Increase serum creatinine ( 10%) 7/38 6/35 Tremor/paresthesia 3/38 2/35 Fever 3/38 1/35 Diabetes mellitus 1/38 0/35 The results of this controlled trial show that there is no additional clinical benefit of a high-dose over a low-dose IV cyclosporine induction regimen in severe acute colitis. This was true for all clinical end points: the proportion of patients responding, the median time to response, and the short-term colectomy rate. In contrast, there was a clear difference in cyclosporine doses and blood levels between the 2 groups. The value of cyclosporine in the acute management of severe colitis is accepted in major referral centers, but concerns about toxicity have prevented its use in many communitybased hospitals. 5 The ideal management of these patients should ensure clinical benefit on the long term and an acceptable toxicity profile. Usually, azathioprine or 6-mercaptopurine is initiated when the acute colitis attack has been controlled because long-term cyclosporine with combined immunosuppression therapy is considered too toxic in a patient population with an otherwise almost normal life expectancy. 6 Because most of the adverse effects associated with the use of cyclosporine are dose dependent, we believe that this drug should always be administered at the lowest effective dose as is recommended for all accepted drug therapies. The clinical response obtained in the present study with a response rate of around 85% in both groups is comparable with what was observed in previous controlled trials using 4 mg/kg cyclosporine. Also, the other end points that we used in this study, such as short-term colectomy rates and the median time to response, are equivalent to what was previously observed. Therefore, it is highly unlikely that the absence of a clinical benefit from a high-dose treatment regimen vs. a low-dose regimen is due to a poor clinical outcome in the 4-mg/kg group. Our findings in a randomized, controlled trial, therefore, confirm the uncontrolled, open-label experience on the high efficacy of 2 mg/kg IV cyclosporine in severe UC. 17 Although the median time to response was 4 days, some patients responded as late as day 8 and, therefore, the decision for colectomy can be performed after 1 week of treatment if the patient s general condition does not deteriorate. Also, we did not observe significant changes in endoscopic scores after 8 days of IV cyclosporine despite clinical response in the majority of patients. In a previous study with 4 mg/kg IV cyclosporine, we observed a change in endoscopic disease severity but only at week 4 and not at earlier time points. 10 Therefore, we would not advocate a follow-up endoscopy at day 8 in clinical practice. The patients in our high-dose group were initially treated with 4 mg/kg cyclosporine. However, applying strict blood cyclosporine monitoring, these doses were not maintained throughout the study. Nevertheless, there was a significant difference in blood cyclosporine levels and in mean cyclosporine doses between the 2 treatment groups. In the placebo-controlled trial reported by Lichtiger et al., 9 higher blood levels were reached. The therapeutic levels for the assay used in that study were clearly higher than in our study ( vs ng/ml). We are also unable to correlate the mean cyclosporine levels in the high-dose group (2.65 ng/ml) with the data from previous controlled experience because the mean dose received by the patients was not reported in the Lichtiger et al. trial. To optimize patient safety, the physicians evaluating clinical scores were not blinded for CRP values, which are known to correlate with clinical response in severe UC. 5 However, knowledge of the CRP values was not unblinding toward

6 1030 VAN ASSCHE ET AL. GASTROENTEROLOGY Vol. 125, No. 4 group assignment because the mean change in CRP levels was similar in both groups. For the cyclosporine-associated adverse effects, there was no difference between the low- and the high-dose group. Nevertheless, we observed a clear trend toward an increased number of novel cases of diastolic hypertension in the high-dose group. The lack of statistical significance was probably caused by a type II error. Renal function as evaluated with serum creatinine levels was not affected in either group. Using inulin clearance, a more sensitive test for glomerular filtration, we have previously demonstrated that patients treated with 4 mg/kg IV cyclosporine had a significant decrease in renal function despite unchanged serum creatinine levels. We observed a relatively low number of other cyclosporineassociated adverse effects such as paresthesias, tremor, or seizures in our patients. This may be explained by strict control of cyclosporine, magnesium, and cholesterol levels. Also, minor adverse effects were noted as spontaneously reported by the patients, which could account for the low prevalence of nausea, headache, and paresthesias. The case of anaphylaxis that occurred in 1 patient (4- mg/kg group) was most likely due to an additive in the IV Sandimmune formulation because the patient was subsequently treated with oral cyclosporine without any further reactions. We evaluated different parameters with possible bearings on the clinical response to cyclosporine. Of these, only active smoking was associated with a poor outcome. We interpret this finding to indicate that, in patients with severe colitis and active smoking, the disease becomes refractory to all medical treatments. However, only 8 of 73 patients were active smokers at inclusion, and, therefore, this result should be treated with caution. More importantly, preexisting steroid or azathioprine therapy was not predictive of clinical response. This further confirms our previous study indicating that cyclosporine monotherapy is equally effective as IV steroids in the treatment of severe UC. 10 However, we did not interrupt IV steroids because there are no controlled data to support such action. Further studies are needed to evaluate whether decreasing the dose to 1 mg/kg of cylcosporine IV preserves efficacy. Also, it would be of interest to know whether, in responding patients, continuation of cyclosporine beyond day 8 is necessary for a sustained response. All these strategies may improve the toxicity profile of cyclosporine. We did not use oral cyclosporine initially because most patients with severe colitis feel sick and may not be able to comply with taking Neoral (Novartis) capsules. Furthermore, controlled studies assessing the efficacy of Neoral (Novartis) as initial therapy were not available when we initiated this study. In summary, we conclude that there is no additional efficacy of 4 mg/kg IV cyclosporine over 2 mg/kg in the acute management of severe UC attacks. Because most of the cyclosporine-associated adverse effects are dose dependent and the long-term benefit of avoiding colectomy should be balanced against the adverse effects of medical therapy, we propose to clinicians to use this lower dose. References 1. Edwards FC, Truelove SC. The course and prognosis of ulcerative colitis. Part 1: short-term prognosis. Gut 1963;4: Truelove SC, Lee EG, Willoughby CP, Ketllewell MG. Further experience in the treatment of severe attacks of ulcerative colitis. Lancet 1978;2: Jamerot G, Rolny P, Sandberg-Gertzen H. Intensive intravenous treatment of ulcerative colitis. Gastroenterology 1985;89: Lichtiger S, Present DH. Preliminary report: cyclosporine in the treatment of severe ulcerative colitis. Lancet 1990;336: Sandborn WJ. A critical review of cyclosporin therapy in inflammatory bowel disease. Inflamm Bowel Dis 1995;1: Cohen RD, Stein R, Hanauer SB. Intravenous cyclosporine in ulcerative colitis: a five-year experience. Am J Gastroenterol 1999;94: Stack WA, Long RG, Hawkey CJ: Short- and long-term outcome of patients treated with CSA for severe acute ulcerative colitis. Aliment Pharmacol Ther 1998; Carbonnel F, Boruchowitz A, Duclos B, Soule JC, Lerebours E, Lemann M, Belaiche J, Colombel JF, Cosnes J, Gendre JP. Intravenous cyclosporine in attacks of ulcerative colitis: short-term and long-term response. Dig Dis Sci 1996; Lichtiger S, Present DH, Kornbluth A, Gelernt I, Bauer J, Galler G, Michelassi F, Hanauer S. Cyclosporine in severe ulcerative colitis refractory to steroid therapy. N Engl J Med 1994;330: D Haens G, Lemmens L, Geboes K, Vandeputte L, Van Acker F, Mortelmans L, Peeters M, Vermeire S, Penninckx F, Nevens F, Hiele M, Rutgeerts P. Inravenous cyclosporine versus intravenous corticosteroids as single therapy for severe attacks of ulcerative colitis. Gastroenterology 2001;120: Gerber DA, Bonham CA, Thomson AW. Immunosuppressive agents: recent developments in molecular action and clinical application. Transplant Proc 1998;30: Feutren G, Mihatsch MJ. Risk factors for cyclosporine-induced nephropathy in patients with auto-immune diseases. International kidney biopsy registry of cyclosporine for autoimmune diseases. N Engl J Med 1992;326: Wijdicks EF, Wiesner RH, Krom RA. Neurotoxicity in liver transplant recipients with cyclosporine immunosuppression. Neurology 1995;45: de Mattos AM, Olyaei AJ, Bennett WM. Nephrotoxicity of immunosuppressive drugs: long-term consequences and challenges for the future. Am J Kidney Dis 2000;35: Porter GA, Bennett WM, Sheps SG. Cyclosporine-associated hypertension. National High Blood Pressure Education Program. Arch Intern Med 1990;150: Schroeder KW, Tremaine WJ, Ilstrup DM. Coated oral 5-aminosalicylic acid therapy for mildly to moderately active ulcerative colitis. N Engl J Med 1987;317: Actis GC, Ottobrelli A, Pera A, Barletti C, Ponti V, Pinna-Pintor M, Verne G. Continuously infused cyclosporine at low dose is sufficient to avoid emergency colectomy in acute attacks of ulcerative

7 October 2003 IV CYCLOSPORINE IN SEVERE ULCERATIVE COLITIS 1031 colitis without the need for high-dose steroids. J Clin Gastroenterol 1993;17: Received May 2, Accepted June 26, Address requests for reprints to: Gert A. Van Assche, Department of Gastroenterology, University of Leuven, 49, Herestraat, Leuven 3000, Belgium. gert.vanassche@uz.kuleuven.ac.be; fax: (32) Gert Van Assche, Severine Vermeire, Geert D Haens, and Paul Rutgeerts have been instrumental in the design of the study, trial management, data analysis, and writing the paper. Maja Noman had a major contribution in the clinical ambulatory follow-up of the patients in the trial. Martin Hiele followed cyclosporine levels and adjusted drug doses of patients in the trial and provided statistical advice. Katrien Asnong was the study coordinator and had a major share in data analysis. Joris Arts analyzed safety data and followed patients clinically during the trial. Andre D Hoore and Freddy Penninckx performed the surgical interventions in patients failing the trial and substantially contributed in evaluating patients for colectomy. Chagas of Chagas Disease Copyright holder unknown. Photo obtained from the National Library of Medicine Website ( Carlos Justiniano Ribeiro Chagas ( ) born in Oliveira, Minas Gerais, Brazil, was the eldest son of a coffee planter. His mother, widowed when Carlos was only 4 years old, hoped for her son to become a mining engineer, but the entreaties of a physician-uncle persuaded the boy to study medicine with a view to combat endemic diseases that hindered the progress of his native country. At the Faculty of Medicine in Rio de Janeiro, his M.D. thesis was devoted to the hematological aspects of malaria. In 1906, after a brief stint of private practice, he joined his friend Oswaldo Cruz ( ) at a newly established institute dedicated to eradication of debilitating infectious diseases. It was there that Chagas first became aware of the barber bug (genus Triatoma), so called because of its propensity to bite the victim s face. In the hindgut of the bug, Chagas identified a new species of trypanosome that he named cruzi as a tribute to his mentor. Two years of intense effort by Chagas resulted in fulfilling Koch s postulates, whereby the existence and cause of American trypanosomiasis became known in its acute form. Later, it was found that the infection, in some of those afflicted, impaired the ganglia of the enteric myenteric plexus, resulting in an alimentary mega syndrome. That Carlos Chagas, a relatively obscure doctor working under primitive conditions in an undeveloped country, could identify a disease that threatened millions, provides an inspiring chapter in the history of medicine. Contributed by WILLIAM S. HAUBRICH, M.D. Scripps Clinic and Reseach Foundation, La Jolla, California