Zaigham Abbas  ( Department of Medicine, Aga Khan University Hospital, Stadium Road, Karachi. )
Imtiaz Malik  ( Department of Medicine, Aga Khan University Hospital, Stadium Road, Karachi. )
Ata Khan  ( Department of Medicine, Aga Khan University Hospital, Stadium Road, Karachi. )

Chloramphenicol is a widely used antibiotic. It is also well known to cause some potentially lethal com­plications such as aplastic anemia, acute leukemia and paroxysmal nocturnal haemoglobinuria Al though development of aplastic anemia is relatively more fre­quent, evidence in favour of acute leukemia caused by chloramphenicol is mostly circumstantial. Very rarely, aplastic anemia caused by chloramphenicol may degenerate into acute leukemia. We describe a case who developed aplastic anemia following chl oramphenicol treatment. Bone marrow done at that time was severely aplastic with no evidence of leukemia. Her bone marrow improved completely with resolution of pancytopenia. Six months later she was again found to have pan­cytopenia. Bone marrow examination this time con­firmed her to be suffering from acute mycloid leukemia. This sequential development of aplastic anemia and acute leukemia carries some interesting pathogenetic implications.

A twenty year old lady was admitted to the Aga Khan University Hospital with complaints of fever and general­ized weakness. Two months back she had developed high grade fever. Her widal test for salmonella typhi was positive at a titre of 1:320. She was treated with chloramphenicol. Her fever subsided but two weeks later she developed increasing weakness and pallor. She was taken to a local clinic, found to be severely pancytopenic and transferredto this hospital. On examination she was found to be very pale. There was no lymphadenopathy. Liver was slightly enlarged and splenic tip was barely palpable. Her haemoglobin was 8g/dl, WBC 700/cmm and platelets 84,000/cmm. Bone marrow examination revealed a hypocellular marrow (Figure 1).

She was found to have F. coli bacteremia and treated with antibiotics. Over the next few days, her blood counts improved. Repeat bone marrow examination was nor­mal. She was discharged with haemobglobin of 10.9g/dl, WBC 5100/cmm with sixty percent neutrophils and a normal platelet count. The patient remained symptom-free for the next six months. She then started having weakness, anorexia and body aches. She was readmitted to this hospital with fever. Blood examination revealed haemoglobin 3.8g/dl, WBC 4700/cmm with seventy-five percent blasts and platelet count of 10,000/cmm. Bone marrow aspirate showed hypercellular marrow heavily infiltrated with blasts (Figure 2).

The blasts were PAS negative, sudan black positive and non-specific esterase positive. She was categorized as acute myelomonocytic leukemia (M4). She received induction therapy with three days of daunorubicin and seven days of cytosine-arabinoside. She responded well to this treatment and a marrow examination done three weeks later confirmed clinical complete remission of acute leukemia.

Chloramphenicol is an important antibiotic that has been widely used for a IonS time. It has been very effectively used for the treatment of typhoid fever and for several other infectious disorders of childhood. Its usage apparently is more common in the third world countries. In Pakistan, it is widely used in intravenous, oral and topical forms. Chloramphenicol is well known to cause some potentially lethal complications such as aplastic anemia, acute leukemia and paroxysmal nocturnal hemoglobinuria^1-4. It can cause bone marrow suppres­sion by two mechanisms. More commonly, it can cause a dose dependent, usually reversible, bone marrow sup­pression that initially affects eryrthroid cell line and later involves other cellular elements. Less commonly, chloramphenicol can cause an idiosyncratic, non dose-dependent, usually irreversible, most often fatal, bone marrow suppression⁵. The actual risk of developing fatal asplastic anemia after being treated with chloram­phenicol is about 1 in 20,000 to 1 in 30,000⁶. This is 13 times higher incidence than the risk of aplastic anemia in the general population. It has been suggested that chloramphenicol causes dose-dependent marrow failure by affecting prolifera­tion and maturation of differentiated cells. It is heralded by increase in serum iron, decrease in reticulocytes and vacuolization of the normoblasts⁷. Another explanation is that chloramphenicol and some of its intestinal metabolites interact with granulocyte-macrophage colony stimulating factor (GM-CSF) and the cells produc­ing it. GM-CSF is an important factor as it stimulates the growth and differentiation of human hematopoietic progenitor cells by activating transcription of specific genes⁹. This inhibitory effect of chloramphenicol can be reversed by the use of recombinant human GM-CSF⁸. In the aplastic anaemia of idiosyncratic type, chloram­phenicol affects the cellular DNA and renders the cells incapable of differentiation, In some cases a genetic predisposition of the stem cells to this inhibitory effect of chloramphenicol on the nucleic acid synthesisniay exist¹⁰. This effect has been observed even after topical uses of chloramphenicol¹¹. Chloramphenicol has also been incriminated to cause acute non- lymphoid leukemia. This effect is similar to that of benzene, ionizing radiation and alkylating agents^12-15. The acute leukemia may or may not be preceded byaplastic anemia¹⁶. Marrows cells have shown unusual chromosomal changes including vacuolization and a large telocentric marker chromosome. Similar changes have been observed in lymphoid cells cultured in the presence of chloramphenicol¹⁷. A population based case-control study carried out in Shanghai sug­gested a significant dose response relationship between the drug use and the risk ofacute leukemias¹⁸. Most of this evidence, however, is still circumstantial. Our patient provides a clear association ofchloram­phenicol induced aplastic anemia and subsequent development of acute leukemia. Initial recovery from aplastic anemia in this case is more suggestive of dose-dependent effect rather than an idiosyncratic reac­tion. Chloramphenicol may have caused DNA damage while inducing aplasia, subsequently degenerating into acute leukemia. However, it is more plausible that patient may have already sub-clinical acute leukemia at the time of initial presentation and that may have made the stem cells more susceptible to the inhibitory effect of chloram­phenicol. Dose- dependent type of aplasia and short time interval between aplastic anemia and subsequent acute leukemia are in favour of this hypothesis. In conclusion we have described a patient who developed chloramphenicol induced aplastic anemia, who subsequently developed acute leukemia. We hypothesize that sub-clinical leukemia may have already existed and could have increased the tendency to develop chloramphenicol induced bone marrow sup­pression.

1. Chuansumerit, A., Hathirat, P. and Isarangkura, P.Acquired aplasticanemia in children: A review of 100 patients Southeast Asian J.Trop. Med. Public Health, 1990;21:313-20.
2. West, B.C., Devault, GA., clement, J.C., and Williams, 1DM. Aplastic anemiaassociated with parenteral chloramphenicol: Review of 10 cases, Rev, infect. Din., 1988;10:1048-51.
3. Braucr, M.J. and Dameshek, W. Hypoplaatic anemia and mycioblastic leukemia following chloramphenicol therapy. N. Engl.J. Med., 1967;277:1003-5.
4. Ouagliana, 3M., Cartwright, G.E. and Wintrobe, MM. Paroxysmal nocturnal hacmogiobinuria following drug-induced aplastic anemia. Ann. Intern. Med., 1965;61:1045-52.
5. Malkin, D., Koren, G. and Saunders, E.F. Drug induced aplastic anemia; pathogenesia and clinical aspects. Am.). Pediatr., Hcmatol. Oncol., 1990;12:402-10.
6. Wallerstein, R.O., condit, P.K., K,spcr, etc. Brown, .1W. and Morrison, FR. Statewide study of chloramphenicol therapy of fatal aplastic anemia. JAMA., 1969;208:2045-50.
7. Rosenback, L.M., Caviles, A.P. and Mitus, W.J., Chloramphenicol toxicity; reversible vacuolization of crythroid cells. N.Engl.J.Med., 1960;263:724-8.
8. Jimenez., U., Jimenez., J.G., Daghistani, D. and Yunis, A.A. Interaction of chloram­phenicol and mctabolites with colony etimulating factors; possible role in chloram­phenicol-induced boric marrow injury. Am.J.Med, Sci., 1990;300:350-3.
9. Adunyah, S.E., Unlap, TM., Wagner, F. and Kraft, AS. Regulation of c-jun expression and AP-1 enhancer activity by granulocyte- nsacrophage colony stimulating factor. J. Biol.Chem., 1991;266:5670-5.
10. Smith, A.L., and Wcber, A. Pharmacology of chloramphenicol., Pediatr. din. NorthAm., 1983;30:209-36.
11. Abrams, SM., Dcgnan, T.J., and vineiguerra, V. Marrow aplasia following topical application of chloraamphenicol eye ointment. Arch. Intern. Med., 1980;140:576-77.
12. Adamson, R.H., and Sieber, SM. chemically induced leukemias in humans. Environ. Health Perspect., 1981;39:93- 103.
13. Gcary, cci. Lymphoma and leukemia due to druge. Br.J., Hosp.Mcd., 1980;24:538-47.
14. The cooperative Study Group for Aplastic and RcfractoLy Anemias: Toxic etiology of aplastic anemia: Sex trsnsm. Dis., 1984;11 (Suppl 4):343-6.
15. Tsongas, TA. Occupational factors in the epidemiology ofebemicalty induced lymphoid and hcmopoietic cancers, in toxicology of the blood and bone marrow. Edited by R.D., Iron, New York, Raven Press, 1985, pp. 149-77.
16. Frsumeni, J.F., Bone marrow depression induced by chloramphenicol or phenyl­butazone, leukemia and other sequelae. JAMA> 1967;201:828-34.
17. Cohen, Hi, and Haung, A.T.E. A marker chromosome abnormality; occurrence in chloramphenicol-associated acute leukemia. Arch, intern. Med., 1973;132:440-3.
18. Shu, X.O., Gao, Y.T., Linet, MS., Brinton, LA., Gao, R.N., Jin, F.and Fraumeni, J.F., Jr., Chlorsmphenicol useand childhood leukemia in Shanghai. Lancet, 1987; 1987;2:934­