Gut 1993; 34:1081-1085 1081 Bone marrow toxicity caused by azathioprine in inflammatory bowel disease: 27 years of experience Gut: first published as 10.1136/gut.34.8.1081 on 1 August 1993. Downloaded from

W R Connell, M A Kamm, J K Ritchie, J E Lennard-Jones

Abstract toid arthritis, and inflammatory bowel disease. Myelosuppression is an important and The commonest sudden side effects include potentially lethal complication of azathioprine nausea, fever, pancreatitis, and bone marrow treatment. The blood count has been reviewed suppression."7 One large study of 1349 patients in all patients treated with azathioprine for with a variety of immunological diseases not inflammatory bowel disease over 27 years in associated with transplant showed a slightly one hospital. Altogether 739 patients (422 with increased risk of neoplasia, but the extent to Crohn's disease, 284 with ulcerative colitis, which the drug alone was responsible was and 33 with indeterminate colitis) were treated uncertain. 18 with 2 mg/kg/day azathioprine for a median of The most common but potentially serious side 12-5 months (range 0.5-132) between 1964 and effect is bone marrow suppression. The effect of 1991. Full blood counts were performed azathioprine on bone marrow is largely dose- monthly for the duration of treatment. In 37 dependent. A conventional immunosuppressive patients (5%) who developed bone marrow dose ofless than 2 5 mg/k/day causes predictable toxicity, the drug was withdrawn or the dose macrocytosis, a raised mean corpuscular haemo- reduced. Thirty two of these patients were globin concentration, and mild leucopenia.'920 asymptomatic and five developed symptoms. Rarely, sudden, severe, and unexpected myelo- Leucopenia (white blood count less than suppression, which is possibly idiosyncratic, has 3Oxl10s/1 ) occurred in 28 (3.8%) patients, in also been reported when low to moderate doses nine of whom it was severe (white blood count (<2 mg/kg/day) of the drug have been used.A22 <2 Ox 109/l). Ofthese nine patients, three were This myelosuppression consists of severe leuco- pancytopenic: two died from sepsis and the penia, thrombocytopenia, and sometimes other had pneumonia but recovered. A further pancytopenia. two patients with severe leucopenia developed In a retrospective study of 396 patients with http://gut.bmj.com/ a mild upper respiratory infection only. inflammatory bowel disease who were treated Thrombocytopenia (platelet count <100 000 x with 6-mercaptopurine, 2% developed clinically 106/1) in 15 patients was associated with severe leucopenia (white blood cell count leucopenia in six and developed in isolation in <2 5 x 109/1). 7 Our experience with azathioprine a further nine (total 2%). Isolated thrombo- in a similar group of patients extends over 27 cytopenia was never clinically severe. Myelo- years. All patients treated with this drug had toxicity from azathioprine developed at any their peripheral blood count monitored each on September 26, 2021 by guest. Protected copyright. time during drug treatment (range 2 weeks-li month, providing a record of all abnormal years after starting the drug) and occurred counts and clinically adverse effects related to either suddenly or over several months. Bone marrow suppression. We present the results of marrow suppression as a result of azathioprine this experience in this report. treatment is uncommon when a moderate dose is used, but is potentially severe. Leucopenia is the commonest and most important Methods haematological complication. Regular Between 1964 and 1991 760 patients at St Mark's monitoring of the full blood count is recom- hospital were treated with azathioprine at a mended during treatment. standard dose of 2 mg/kg. Each subject was (Gut 1993; 34: 1081-1085) reviewed regularly as an outpatient during treat- ment and full blood counts were obtained monthly. Some patients who had been treated Azathioprine is well established as an effective with azathioprine for several years had blood drug in the treatment of inflammatory bowel tests every two months. disease. Although its usefulness was not shown A record was maintained for every patient in four controlled studies of patients with active treated with azathioprine, detailing any haema- Crohn's disease or ulcerative colitis,A a further tological or other side effects as well as other six controlled trials examining this drug or its reasons for stopping or changing the dose of the metabolite, 6-mercaptopurine, in chronic active drug. St Mark's Hospital, London Crohn's disease and ulcerative colitis have shown The normal white blood count range at this W R Connell steroid sparing effects in all subjects and a clear hospital is 4-1 lx 109/1. In this study, leucopenia M A Kamm therapeutic advantage in most.s 10 Additional is defined as moderate (white blood count 2 0- J K Ritchie reports of uncontrolled studies support 3 or severe 0x Thrombo- J E Lennard-Jones this 0x 109/l) (<2 109/1). impression. 11-16 to a count Correspondence to: cytopenia refers platelet less than Dr M A Kamm, St Mark's Concerns about the safety of azathioprine and 1 00OOOx 106/1. Hospital, City Road, London EC1V 2PS 6-mercaptopurine, remain, however. Consider- For the purposes of this study, 21 patients Accepted for publication able clinical experience of azathioprine has now were have been excluded from the analysis - five 21 December 1992 been obtained in organ transplantation, rheuma- because of poor drug compliance or inadequate 1082 Connell, Kamm, Ritchie, Lennard-ones

TABLE I Clinical details of739 patients treated with azathioprine 132 . E 1203, Crohn's Ulcerative Indeterninate 84 ' Gut: first published as 10.1136/gut.34.8.1081 on 1 August 1993. Downloaded from disease colitis colitis Total S 0 Co 72 - No of patients 422 284 33 739 E 0 Male 204 152 14 370 a) 60 ' Female 218 132 19 369 48 0 Median duration of 0 treatment (mth) 14-0 9 0 16-0 12-0 0) 36 0-01-183 -c 0 Range(mth) 0-01-183 0-1-171 0-5-99 24 Total no of patients - CD months of treatment 12 867 5757 839 19 463 -J 12 a _ 0- I 9 Severe Thrombocytopenia leucopenia TABLE II Patients with Crohn's disease (CD) and ulcerative colitis (UC) who developed bone Moderate marrow suppression while taking azathioprine leucopenia Leucopenia* and The number ofmonths' treatment with azathioprine before the Leucopenia thrombocytopeniat Thrombocytopenia onset ofsevere (white blood count <2 Ox 10911) or moderate CD UC CD UC CD UC (white blood count 2-0-3-Ox 109/l) leucopenia and thrombocytopenia (platelet count <100 OOOx ICt/I). Medians No of patients 16 6 4 2 6 3 are indicated Male 5 2 4 2 3 3 Female 11 4 - - 3 - Mean age at onset ofmyelotoxicity (y) 36 33 34 18 45 38 count was lower than the counts for other white Leucopenia: white blood count <3-Ox 10'/1; thrombocytopenia - platelet count <100 000X 106/1. cell lines in all of these patients. Three of them had severe leucopenia as part of pancytopenia. follow up, 10 in whom treatment was monitored Another patient with severe leucopenia and two at other centres, five who were treated with with moderate leucopenia also had associated azathioprine for conditions other than inflam- mild thrombocytopenia (platelet count matory bowel disease, and one who was <100 OOOx 106/1). Isolated mild and asympto- inadvertently given an excessive dose ofthe drug matic thrombocytopenia (60 000-100 000x (6 mg/kg). Blood counts of the remaining 739 106/1) occurred in a further nine patients. A total subjects who form the basis of this report were of 15 patients (2-0%) experienced some degree of examined to identify all haematological reactions thrombocytopenia. and case notes were reviewed for those with abnormalities.

Clinical indications for azathioprine treatment SYMPTOMS http://gut.bmj.com/ included long term steroid use, failure to control Five patients (0-7%), all with severe leucopenia, symptoms with convential treatment, and the had symptoms from myelotoxicity. Profound presence of internal or external fistulae. When pancytopenia occurred in three, resulting in two required, patients continued to receive other deaths from sepsis and one serious lung infection conventional anti-inflammatory agents includ- which resolved eventually. Two others with ing corticosteroids, sulphasalazine (or 5-acetyl- severe leuopenia developed mild upper respira-

salicylic acid derivatives), or metronidazole. No tory infections which responded to oral anti- on September 26, 2021 by guest. Protected copyright. patient received other immunosuppressive biotics and withdrawal of azathioprine. The drugs. clinical details of one of the patients who died have been described previously.9 Results Altogether 422 patients had Crohn's disease, 284 TIMING OF BONE MARROW TOXICITY IN RELATION ulcerative colitis, and 33 indeterminate colitis. TO LENGTH OF TREATMENT The median duration of treatment with azathio- Although bone marrow suppression occurred at prine was 12-0 (range 0-01-184) months. The any time during treatment, it generally developed total of patient years of treatment was 1622. The early (Fig 1). The nine patients with severe clinical details of these patients are given in leucopenia, including the three with pancyto- Table I. One hundred and forty patients received penia, developed myelotoxicity after 0-5-132 the drug for only one month or less because of months of treatment (median 9 months). This acute side effects or elective cessation ofthe drug complication occurred within three months in in the early years ofits use; none ofthese patients four patients and within a year in a further two. stopped taking the drug because of myelosup- Myelotoxicity in the 19 patients with moderate pression. Patients who developed myelosuppres- leucopenia developed after 1-76 (median 12-5) sion had no clinical evidence ofrenal, hepatic, or months oftreatment. Isolated thrombocytopenia splenic dysfunction beforehand. developed after 2-51 months of drug treatment. Bone marrow toxicity, a white blood count <3-0x 109/1, a platelet count <1 00000x 106/1, or pancytopenia caused by azathioprine occurred in RAPIDITY OF ONSET OF BONE MARROW 37 patients (5-0%). Their details are given in SUPPRESSION Table II. The drug was stopped in all except two Severe leucopenia occurred abruptly in six of the patients in whom it was continued at a reduced nine patients; blood counts one month previously dose. Leucopenia occurred in 28 patients (3-8%). had been normal in these cases. The other three In nine this was severe (white blood count patients developed severe leucopenia gradually <2-0x 109/l) and in 19 it was moderate (white over more than two months; initial reductions in blood count 2-0-3-0x 109/1). The neutrophil the white blood count were overlooked and the Azathioprine and myelotoxicity in IBD 1083

drug dose was changed only when severe share similar clinical and side effects. The dose of leucopenia occurred. 6-mercaptopurine required to produce

Four of the five symptomatic patients, equivalent therapeutic and toxic effects as Gut: first published as 10.1136/gut.34.8.1081 on 1 August 1993. Downloaded from includingthethreewithpancytopenia, developed azathioprine is about 55% that of azathioprine.23 severe leucopenia suddenly; their blood counts These drugs possess both cytotoxic and one month earlier had been normal. A gradual immunosuppressive properties. Cytotoxicity is and progressive fall in the white blood count believed to result from the incorporation of occurred in the fifth, symptomatic patient who thiopurine metabolites into cellularnucleic acids, presented with a mild respiratory infection while the immunosuppressive effects are associated with severe leucopenia. secondary to inhibition of de novo purine ribo- All but three ofthe 19 patients who developed nucleotide synthesis and interconversion.24 moderate leucopenia had had normal white Mild leucopenia, defined as a white cell count blood cell counts one month earlier. Two ofthese between 3 0 and 4 0x 109/l, is the commonest 16 patients had had a slight fall in their platelet haematological side effect of azathioprine. This count, to just below 100 000x 106/1, one month effect is thought to be dose-dependent and before their white blood cell count became relates to a shift towards less mature forms and a abnormal. Myelosuppression was progressive in decreased myeloid-erythroid ratio.'9 The the three other patients in whom azathioprine principal effect of this drug on white blood cells was continued despite the presence of moderate is granulocytopenia, which occurs shortly after leucopenia. starting azathioprine. Accompanying bone Isolated thrombocytopenia occurred suddenly marrow changes on granulopoiesis are not seen in six of nine patients. Their platelet and other until the peripheral blood count falls below blood counts had been well within the normal 2 0x 109/1.25 More profound bone marrow sup- range previously. In the other three patients pression occurs with higher doses ofazathioprine there was a gradual decline in the platelet count or can develop unexpectedly as an idiosyncratic over two or more months. reaction. In these cases, agranulocytosis or pancytopenia results from disturbed marrow cellularity and impaired cell line maturation.25 RECOVERY FROM MYELOTOXICITY The risk of leucopenia in inflammatory bowel Apart from the two patients who died, the full disease as a result of azathioprine or 6- blood counts returned to normal within one mercaptopurine treatment is reported to be 2 to month in all patients who had myelotoxicity as a 5%, though few studies define the term result of azathioprine treatment, including the leucopenia clearly. 12 13 15-17 26 In the National

patient with pancytopenia and severe Crohn's Disease Cooperative Study, 15% of http://gut.bmj.com/ pneumonia. patients treated with 2-5 mg/kg developed a white blood cell count less than 4 0x l09/1. In the same study, treatment with 1-0 mg/kg azathio- SUBSEQUENT TREATMENT WITH AZATHIOPRINE prine was associated with a 2% incidence of Azathioprine was reintroduced to 19 patients leucopenia.27 In another study, 2% of 396 who had developed myelotoxicity. Two of the patients treated with 6-mercaptopurine in a dose

nine patients with severe leucopenia resumed the of 1-5 mg/kg/day developed a white blood cell on September 26, 2021 by guest. Protected copyright. drug at a lower dose; one had an uncomplicated count ofless than 2-5x 109.'7 In a meta-analysis of course over several years, the other enjoyed good 542 patients treated with azathioprine for health for a further 10 years until the drug was rheumatoid arthritis, 14 patients (2 6%) stopped electively. Nine years later, however, developed a white blood count <2 5 x 109/1 and a this 74 year old woman has developed possible further 34 patients (6 3%) developed a count of early myelodysplastic syndrome. between 2-5 and 3-5x 109/1.26 Thirteen of 19 patients with moderate Other effects of azathioprine are seen in the leucopenia resumed azathioprine. Three ofthese peripheral blood. Macrocytosis is common; it were treated again at the same dose and all occurs in approximately 70% of renal transplant developed moderate leucopenia again. The other recipients.2829 Reversible erythroid hypoplasia 10 patients were restarted at a lower dose and two has been reported in one patient given 6- of them developed recurrent mild leucopenia. mercaptopurine for Crohn's disease and four The drug was stopped in those with recurrent renal transplant recipients treated with azathio- leucopenia. prine.'73 In keepting with our own findings, A further patient with combined mild previous studies have shown that thrombocyto- leucopenia and thrombocytopenia developed penia occurs less commonly than leucopenia and these side effects again when azathioprine was is not usually ofclinical severity. 7 2631 reintroduced at an identical dose. Three patients To our knowledge five detailed case reports of with isolated thrombocytopenia were restarted pancytopenia related to the use of azathioprine on azathioprine at a reduced dose: none of these for immunosuppression have been published. developed any subsequent abnormality of the Four patients with connective tissue disease and full blood count. one with ulcerative colitis developed reversible pancytopenia after 0-5-15 months of therapy with 75-100 mg/day azathioprine.2"2'32 Further Discussion reports of pancytopenia have been reported to Azathioprine is a synthetic analogue of the the drug's manufacturers.'9 There has been only naturally occurring purine, adenosine. It is con- one reported death related to bone marow verted to 6-mercaptopurine and since their toxicity from immunosuppressive azathioprine subsequent metabolism is identical, both drugs treatment.33 1084 Connell, Kamm, Ritchie, Lennard-Jones

The reasons for individual variation in bone serious bone marrow toxicity is unpredictable. marrow susceptibility to azathioprine deserve Complications affecting the bone marrow had a

attention. Bone marrow toxicity seems to be very variable time of onset after starting treat- Gut: first published as 10.1136/gut.34.8.1081 on 1 August 1993. Downloaded from related to the excessive intracellular concentra- ment in the other 35 patients with myelosuppres- tion of the cytotoxic active metabolities 6- sion. Whilst reduced thiopurine methyltrans- thioguanine nucleotides.2' These substances may ferase activities may be responsible for accumulate if the activity ofthe catalytic enzyme azathioprine-induced myelotoxicity in some thiopurine methyltransferase is deficient. patients, the late onset of this complication in An inverse relationship between the red cell others suggests that additional factors must also concentration of 6-thioguanine nucleotides and be involved. the neutrophil count has been reported in Both leucopenia and thrombocytopenia may children given azathioprine for leukemia.34 recur with subsequent courses of azathioprine, Furthermore, five patients who developed acute particularly if the same dose is used; it is less myelosuppression after taking azathioprine at a likely to develop ifdose reductions are made. dose of 1 0-2-5 mg/day showed very low It seems prudent to stop the drug if moderate thiopurine methyltransferase activities and suppression of the white blood cell or platelet abnormally high 6-thioguanine nucleotide con- count occurs. We feel that azathioprine should centrations compared with 16 patients who were never be used without blood tests; intervention also receiving azathioprine but did not have only when symptoms occur may be too late. It myelosuppression.2 could be argued that regular monitoring of the Further cases of patients with reduced peripheral blood count did not influence the thiopurine methyltransferase activity and course of myelosuppression in our patients. In leucopenia secondary to azathioprine or 6- six ofthe nine patients with severe leucopenia the mercaptopurine provide supporting evidence onset was sudden and unpredictable. This that low thiopurine methyltransferase activity includes the two patients who died and the may be a major risk factor for azathioprine patient with pneumonia. However, the develop- induced myelosuppression.3'36 ment of severe leucopenia could have been The activity of thiopurine methyltransferase averted in the other three patients with more seems to be controlled by a common genetic careful attention to the serial blood counts. polymorphism. One in 300 individuals in the Routine blood testing throughout the duration population has very low or no enzyme activity of azathioprine treatment requires close and and 11% have intermediate activity.37 committed coordination between specialist, Homozygotes and heterozygotes for this enzyme general practitioner, laboratory, and patient. An may have different thiopurine methyltransferase azathioprine card containing details of dose, activity and therefore different susceptibility to possible side effects, and blood testing require- http://gut.bmj.com/ azathioprine toxicity. The information on ments should be given to patients taking this thiopurine methyltransferase and susceptibility drug. The manufacturers of azathioprine to myelotoxicity from azathioprine is relatively recommend weekly blood testing for the first few recent, and post dates most of the data collected months of treatment and three monthly or less in this series. Testing for thiopurine methyl- thereafter. We believe that in inflammatory transferase activity is not routinely available at bowel disease where immunosuppressive doses present, though it is possible to measure serum of azathioprine are used, two weekly blood tests on September 26, 2021 by guest. Protected copyright. 6-thioguanine nucleotide concentrations. in the first three months oftherapy are sufficient. Our experience with azathioprine in inflam- Since the onset of myelosuppression is un- matory bowel disease confirms that the risk of predictable, and often abrupt, we advocate more myelotoxicity is low but that this has potentially frequent testing of the blood count than three serious clinical sequelae. Leucopenia is the monthly for the remainder oftreatment. commonest and most serious toxic effect of Dr Connell's research is sponsored by a John Nott Travelling azathioprine on the bone marrow. All five Scholarship and a Scholarship from the Association of Common- patients who developed symptoms related to wealth Universities. bone marrow suppression were severely leuco- penic. None of the patients with moderate 1 Rhodes J, Bainton D, Beck P, Campbell H. Controlled trial of azathioprine in Crohn's disease. Lancet 1971; 2: 1273-76. leucopenia had symptoms. In other words, 2 Klein M, Binder HJ, Mitchell M, Aaronson R, Spiro H. warning signs in the form of symptoms did not Treatment ofCrohn's disease with azathioprine: a controlled evaluation. Gastroenterology 1974; 66: 916-22. occur until myelosuppression was advanced. 3 Summers RW, Switz DM, Sessions Jr JT et al. The national Of the 37 patients with myelosuppression, 28 cooperative Crohn's disease study: results ofdrug treatment. Gastroenterology 1979; 77: 847-69. developed this complication within one month of 4 Jewell DP, Truelove SC. Azathioprine in ulcerative colitis: a normal full blood count and the drug was final report on controlled therapeutic trial. BMJ 1974; 4: 627-30. stopped or its dose reduced. Twenty four of 5 Willoughby JMT, Kumar PJ, Beckett J, Dawson AM. these patients were asymptomatic and it is not Controlled trial of azathioprine in Crohn's disease. Lancet 1971; 2: 944-6. known how many would have gone on to develop 6 Kirk AP, Lennard-Jones JE. Controlled trial of azathioprine more severe myelosuppression had the drug in ulcerative colitis. BMJr 1982; 284: 1291-2. 7 Rosenberg JL, Levin B, Wall AJ, Kirsner JR. A controlled been continued at the same dose. Myelotoxicity trial of azathioprine in Crohn's disease.AmJ- Dig Dis 1975; was not recognised initially in the other nine 8 Rosenberg JL, Wall AJ, Levin B, Binder HJ, Kirsner JB. A patients and azathioprine was continued at the controlled trial ofazathioprine in the management of chronic same dose; a progressive deterioration in the ulcerative colitis. Gasn'oenrerology 1975; 69: 96-9. 9 O'Donoghue DP, Dawson AM, Powell-Tuck J, Bown RL, blood count occurred over the next month until Lennard-Jones JE. Double-blind withdrawal trial of the drug was stopped. azathioprine as maintenance treatment for Crohn's disease. Lancet 1978; 2: 955-7. The deaths oftwo patients, after three and 132 10 Present DH, Korelitz BI, Wisch N, Glass JL, Sachar DB, months of azathioprine treatment, show that Pasternack BS. Treatment of Crohn's disease with 6- Azathioprine and myelotoxicity in IBD 1085

mercaptopurine. A long-term, randomised double-blind 25 Kolle G. Knockenmarksschadigung bei immunsuppressiver study. NEnglJ3Med 1980; 302: 981-7. therapie der juvenilen rheumatoiden arthritis und des Still- 11 Theodor E, Niv Y, Bat L. Imuran in the treatment of syndroms mit azathioprin. Dtsch Med Wochenschr 1969; 94:

ulcerative colitis. AmJ Gastroenterol 1981; 76: 262-6. 2268-73. Gut: first published as 10.1136/gut.34.8.1081 on 1 August 1993. Downloaded from 12 Nyman M, Hansson I, Eriksson S. Long-term immuno- 26 Whisnant JK, Pelkey J. Rheumatoid arthritis: treatment with suppressive treatment in Crohn's disease. Scand J Gastro- azathioprine [Imuran (R)]. Clinical side-effects and labora- enterol 1985; 20: 1197-203. tory abnormalities. Ann Rheum Dis 1982; 41 (suppl): 44-47. 13 Goldstein F. Immunosuppressant treatment of inflammatory 27 SingletonJW, LawDH, KelleyML, Mekhjian HS, Sturdevant bowel disease. Pharmacological and clinical aspects. J Clin RAL. National cooperative Crohn's disease study: adverse Gastroenterol 1987; 9(6): 654-8. reactions to study drugs. Gastroenterology 1979; 77: 870-82. 14 Present DH, Chapman ML, Rubin PH. Efficacy of 6- 28 Wickramasinghe SN, Dodsworth H, Rault RMJ, Hulme B. mercaptopurine in refractory ulcerative colitis. Gastro- Observation on the incidence and cause of macrocytosis in enterology 1988; 94 (2): A359. patients on azathioprine therapy following renal transplanta- 15 Lobo AJ, Foster PN, Burke DA, Johnston D, Axon ATR. tion. Transplantation 1974; 18: 443-6. The role of azathioprine in the management of ulcerative 29 McGrath BP, Ibels LS, Raik E, Hargave M, Mahony JF, colitis. Dis Colon Rectum 1990; 33: 334-7. Stewart JH. Erythroid toxicity of azathioprine. Q Jf Med 16 O'Brien JJ, Bayless TM, Bayless JA. Use of azathioprine of 6- 1975; 173: 57-63. mercaptopurine in the treatment ofCrohn's disease. Gastro- 30 Old CW, Flannery EP, Grogan TM, Stone WH, San Antonio enterology 1991; 101: 39-46. RP. Azathioprine-induced pure red blood cell aplasia. 17 Present DH, Meltzer SJ, Krumholz MP, Wolke A, Korelitz JAMA 1978; 240: 552-4. BI. 6-Mercaptopurine in the management of inflammatory 31 Bunch TW, O'Duffy DM. Disease modifying drugs for bowel disease: short- and long-term toxicity. Ann Intern Med progressive rheumatoid arthritis. Series on pharmacology in 1989; 111:641-9. practice. Mayo Clin Proc 1980; 55: 161-79. 18 Kinlen LJ. Incidence of cancer in rheumatoid arthritis and 32 Bacon BR, Treuhaft WH, Goodman AM. Azathioprine- other disorders after immunosuppressive treatment. Am J induced pancytopaenia. Occurrence in two patients with Med 1985; 78(1A): 44-49. connective-tissue diseases. Arch IntMed 1981; 141: 223-6. 19 Lawson DH, Lovatt GE, Gurton CS, Hennings RC. Adverse 33 Avery Jones F, Lennard-Jones JE, Hinton JM, Reeves WG. effects ofazathioprine. Advances in DrugReactions andAcute Dangers of immunosuppressive drugs in ulcerative colitis. Poisoning Revue 1984; 3: 161-71. BMJ 1966; 1: 1418. 20 Jeurissen MEC, Boerbooms AMTh, Van de Putte LBA. 34 Lennard L, Rees, CA, Lilleyman JS, Maddocks JL. Child- Pancytopaenia related to azathioprine in rheumatoid hood leukemia: a relationship between intracellular 6- arthritis. Ann Rheum Dis 1988; 47: 503-5. mercaptopurine metabolites and neutropenia. Br J Clin 21 Burke DA, Dixon MF, Axon ATR. Ulcerative colitis: pro- Pharmacol 1983; 16: 359-63. longed remission following azathioprine induced pancyto- 35 Evans WE, Homer M, Chu YQ, Kalwinsky D, Roberts WM, paenia. J Clin Gastroenterol 1989; 11(3): 327-30. Altered mercaptopurine metabolism, toxic effects, and 22 Lennard L, Van Loon JA, Weinshilboum RM. Pharmaco- dosage requirement in a thiopurine methyltransferase- genetics of acute azathioprine toxicity: relationship to deficient child with acute lymphocytic leukemia. J Paediatr thiopurine methyltransferase genetic polymorphism. Clin 1991; 119:985-9. Pharmacol Ther 1989; 46: 149-54. 36 Chochair PR, Duley JA, Simmonds HA, Cameron JS. The 23 Goldstein F. Maintenance treatment for Crohn's disease: has importance of thiopurine methyltransferase activity for the the time arrived? AmJ Gastroenterol 1992; 87: 551-6. use of azathioprine in transplant recipients. Transplantation 24 Van Scoik KG, Johnson CA, Porter WR. The pharmacology 1992; 53: 1051-6. and metabolism of the thiopurine drugs 6-mercaptopurine 37 Van Loon J, Weinshilboum RM. Thiopurine methyltrans- and azathioprine. Drug Metabolism Reviews 1985; 16(1&2): ferase biochemical genetics human lymphocyte activity. 157-74. Biochem Genet 1982; 20: 637-58. http://gut.bmj.com/ on September 26, 2021 by guest. Protected copyright.