May 1997, Volume 47, Issue 5

Original Article

Bone Marrow Changes in Human Malaria: A Retrospective Study

Abdus Salam Khan Gandapur  ( Department of Basic Medical Sciences, Gomal University, Dera Ismail Khan. )
Salman A. Malik  ( Department of Biology, Quaid-e-Azam University, Islamabad. )
Fazal Raziq  ( Department of Pathologyt, Postgraduate Medical Institute, Lady Reading Hospital, Peshawar. )


The bone marrow reports of 1966 patients admitted to a provincial teaching hospital between January, 1992 to April, 1995 were retrospectively analyzed . Twenty-six (1.3%) bone marrows showed the presence of malarial parasites. Sixteen (62%) patients had Plasmodium falciparum, 9 (34%) Vivax malaria and one (4%) mixed infection. All these patients gave a history of prolonged illness and had low parasite counts, Plasmodium vivax malaria was not associated with any significant pathology in the bone marrow, except iron deficiency anaemia, The bone marrows with Plasmodium falciparum malaria showed myeloid hyperplasia, erythroid hyperplasia, megaloblastosis and hypoplasia in different proportions. No evidence of dyserythropoiesis was found in this series. The possible mechanisms producing these changes and the factors responsible for the discrepancy in bone marrow findings in different geographical areas are discussed (JPMA 47:137, 1997).


The haematological changes seen in patients with malaria, include anaemia, neutropenia, reactive lymphocytosis, monocvtosis, eosinopenia, neutrophil leukocytosis and thrombocytopenia. Of these abnormalities, the most important is anaemia as this is often associated with considerable morbidity particUlarly in areas with sustained year round infection1. The morphological abnormalities affecting the bone marrow have also been reported in Gambian children with severn anaemia and falcipanim malaria. Large numbers of erythmcytes are destroyed by parasite replication, but other important elements include depression of eiythrocytc production by the bone marrow and phagocytosis of intact erythrocytes both of which are exacerbated by tumor necrosis factor (TNIF). The above complication is more likely tobe produced by chronic TNF production thanby acutebuists of TNF release due to repeated or untreated infection. The purpose of this retrospective study was comparative assessment of bone marrow changes in malaria seen in Pakistan where the disease is endemic wih those seen in other areas.

Patients and Methods

A retrospective analysis of all bone marrow reports in the Lady Reading Hospital Peshawar, was performed by reviewing patient’s records. One thousand nine hundred and sixty-six cases were recorded during the period from Januaiy, 1992 to April, 1995. The red cell indices, including haemoglobin, red bloodcell count, totalwhite blood cell count and platelet count were measured by conventional manual methods. Thick and thin blood films were made and stained with Giemsa stain. The plasmodium falciparum, asexual forms, were examined and counted against 200 WBC ma thick blood film and the absolute parasite count was calculated from the Observed WBC count as defined by WHO2.The indication for bone marrow examination was the diagnosis of various haematological disorders, like iron deficiency anaemia, thrombocytopenia, leukaemia and pyrexia of unknown origin. The bone marrows were aspiratedfrom posterior supenor iliac crests. The smears were stained with May Grunwald and Giemsa stain and in 11(44%) patients staining for iron was done. The degree of megaloblastic change in the erythroid series was recorded and scored as +, ++, and +++. The bone marrow cellularity was reported ashypercellular, hypocellular and normocellular. The erythropoiesis was expressed as normoblastic, hyperplastic and megaloblastic. The bone marrow slides were reviewed by two competent haematologists. The statistical analysis of age, parasite count, haemoglobin and white blood cell count were determined by the Statgraphic program (Version 5.0).


Of 1966 bone marrow reports, 26 (1.3%) showed the presence of malarial parasites. The total duration of illness ranged from 7 days to 6 months. Sixteen (62%) patients had Plasmodium falciparum malaria, 9(34%) were suffering from Vivax malaria and one (4%) patient had mixed infection. In nine cases ofPlasmodiumvivax malaria, 4(44%) patients had absent iron stores with no sideroblasts. One (11%) patient had few megaloblasts, while 3 (33%) patients had normal bone marrow. In one patient with Vivax malaria marrow showed 20% plasma cells. In sixteen cases with Plasmodium falcipamm malaria 3 (19%) bone marrows were nonnal, 3 (19%) showed myeloid hyperplasia, 6 (38%) megaloblastic ezythropoiesis, 2 (12.5%) had bone marrow hypoplasia, and the remaining 2 (12.5%) showed eiythroid hyperplasia. One (6%) case with mixed infection had 30-40% megaloblasts in the bone marrow.
Erythropoiesis in patients with falciparum malaria
All these patients with falciparum malaria had a low parasite Count and decreased haemoglobin. The patients were divided into two groups i.e., mild anaemia (Hb >8 G%) group 1 and severe anaemia (Hb <8G%) group 2. There were 10 patients in group 1 and 6 in group 2. Two patients in group 1 and 4 patients in group 2 had megaloblastic changes in the bone marrow. The bone marrow cellularity was expressed as normocellular, hypocellular and hypercellular. Two (20%) patients in group 1 and 3 (50%) patients in group 2 had hypercellular marrow, while 2 (12%) patients, one 1mm each group had hypocellular marrow (Table I).

The mean and SD values of age, duration of disease, haemoglobin level, platelet count and parasite count are shown in Table II.

The bone marrow intracellular iron was assessed only in nine patients. There was no intracellular iron in two (22%) patients scanty in 2(22%), nonnal in2 (22%) and increased in 3 (34%).


The present study reports the bone marrow changes in malaria in North West Frontier Province of Pakistan, where malaria is endemic and to compare it with other studies done in other parts of the World. Vivax malaria was not associated with significant pathology except mild megaloblastic erythropoiesis in one (11%) case. The single (4%) patientwith mixed infection had hypercellular marrow and megaloblastic erythropoiesis.
Out of 26 bone marrows recorded during 39 months period, 16 (6 1%) showed the presence of plasmodium falciparum. In 50% the bone marrow was normocellular, in 31% hypercellular and in 12% hypocellular. All the patients with hypercellular marrow belonged to severely anaemic group 2. The bone marrows of 20 Gambian children showed varying degrees of hypercellulanty which was more marked in severely anaemic patients. The hypercellularity was due to erythroid hypeiplasia in patients with chronic malaria3. In our patients, the hypercellularity in severely anaemic patients, was both due to ezythroid myeloid hyperplasia.
In another study of seven Gambian children suffering from falciparum malaria, bone marrow had severe eiythroid hyperplasia and dyseiythropoiesis. One (14%) each had megaloblastic and macronormoblastic and 5 (7 1%) normoblastic eiythropoiesis4. Three (18%) patients in this series had erythroid and myeloid hyperplasia. Fleming5 examined the bone majrows of34 pregnant patients suffering from falciparum malaria in Ndolla (Zambia) and reported erythroid hyperplasia in 26(76%), myeloid hyperplasia in 25 (74%), excess eosinophil precursors in 3 (9%), increased lymphocytes in 3 (9%) and normal plasma cells in 11(38%) cases. He reported definite megaloblastic erythropoiesis and myelopoiesis in 23 (68%) bone marrows. He observed dysery thropoiesis in only one (3%) bone marrow. Anaemia associated with malaria and megaloblastic etythropoiesis in his series was charactensed by younger age of the patients, greater severity of anaemia, macrocytosis, eiythroid hyperplasia and neutrophilleucocytosis.
The findings inthis study are similar to those reported by Fleming5 except the presence of dyseiythropoiesis in one case, whichwas not found in any patient inour series, although he studied pregnant patients with falciparum malaria. In our patients with megaloblastic changes both red cell folate and serum vitamin B12 level were normal. Malarial parasite produces glycosyl phosphatidyl inositol (GPO, a novel class of glycolipid toxins, which substitute for the endogenous inositolglycan based signal transduction pathways of the host. GPI stimulates high levels of tumor necrosis factor (TNF)and interleukin 1 produced by macrophages. These mediators are responsible for the production of these morphological abnonnalities and anaemia6. Elshoura7 reported several ultra structural changes in non- parasitized erythroblasts in the bone marrow and erythrocytes in the peripheral blood of 28 Saudi patients who were anaemic and suffering from acute falciparum malaria. The erthroblasts for the first time showed conspicuous surface knobs which were previously described only forthe parasitized erythrocytes. Ultrastructural changes of these cells were suggested to be due to dyserythropoiesis, which were in turn attributed to an imbalance in metabolism as they were being overproduced in response to infection. Numerous haemoglobin like particles were being liberated through the ezythrocyte plasma membrane indicating severe haemolysis which is considered to be one of the major factors producing anaemia during malarial infection. Jootar and coworkers8 cultured the bone marrows of 21 Thai adults infected with Plasmodium falciparum for CFU and BFU using AB serum, autologus serum (parasitaemia) and autologus serum (post parasitaemia). Fifteen patients had haematological complications. The autologus sera during parasitaemia suppressed the growth of CFU-E and BFU-E, both during and after parasitaemia They postulated that two possible mechanisms for this suppression are the reduction of erythropoietin or the increased TNF production during malarial infection.
This retrospective study shows that bone marrow involvement by the Plasmodium falcipamm does occur in Pakistan where malaria is endemic. The changes in the bone marrow particularly in the red blood cells are not so marked as reported by workers from other countries3-8. Why Plasmodium falciparum produces dyseiythropoiesis in some infected individuals but not in others, as in our situation is poorly understood. This may be due to the polymorphism of Plasmodiumfalciparumisolates in inducing different levels of TNF which may differ by 100 folds as demonstrated by Allan and coworkers12. Again TNF production may also be determined by the variation in the propensity of the host to produce TNF, the variation in the propensity of the host to produce TNF, the population dynamics of the parasites within the host and the acquisition Of antitoxic antibodies and other immune adaptations of the host.


1. Gilles, H.M. and Warrel, D.A. In Bruce-Chwatts Ed essential malariology.Lon­don. Edward Arnold, 1993, PP. 53-54.
2. World Health Organization Severe and complicated malaria. Trans. R. Soc. Trop. Med. Hyg., 1986;80(suppl):3-50.
3. Abdalla, SR., Wickramasinge, S.N. and Weatherall, D.J. The deoxyundine suppression test in severe anaemia following Plasmodium falciparum malaria. Trans. R. Soc. Trop. Med. Nyg., 1984;78:60-63.
4. Abdalla. S.H. Iron and folate status in Gambian children with malaria. Ann. Trop. Med. Parasitol., 1990;10:265-72
5. Fleming, A.F. The aetiology ofsevere anaemia in pregnancy in Ndolla (Zambia). Ann. Trop. Med. Parasitol., 1989;83:37-49.
6. Schofield, L., Vivas,L., Hackett, F. et al. Signal transduction in host cells by a glycosylphosphadityl inositol toxin of malaria parasite. Ann, Trop. Med. Parasitol., 1993;87:61 7-26.
7. Elshora, S.M. Falciparum malaria in naturally infected human patients. Ultrastructural alterations of non-parasitized red blood cells during anaemia. Appi. Parasitol., 1993;34:173-79.
8. Jootar, S., Chaisiri-Poomkere, W., Pholvicha, P. et al. Suppression of erythroid progenitor cells during malarial infection in Thai adults caused by serum inhibitors. Clin. Lab. Haematol., 1993;15:87-92.
9. Allen, R. J.,Beattie, P., Bat, C. et al. Strain variations in tumor necrosis factor induction by parasite from children with acute falciparum malaria. Infect. Immun., 1995;63:1173-75.

Journal of the Pakistan Medical Association has agreed to receive and publish manuscripts in accordance with the principles of the following committees: