Plasmodium in the bone marrow: case series from a hospital in Pakistan, 2007–2015

Acute malaria is a life-threatening, multisystem disease caused by protozoan of the Genus Plasmodium that infect hundreds of millions of people annually, with several hundred thousand not surviving [1]. These parasites occur as different forms in various tissues, including liver, blood, spleen, and bone marrow [2]. Plasmodium vivax, in particular, exhibits specific molecular traits that would appear to accord with strong tropisms for haemopoietic tissues [3‐5]. Merozoites of that species require transferrin receptor 1 (CD71) to invade erythrocytes [6], a molecule that vanishes from the surface of that cell during early reticulocyte development [5]. Furthermore, synergistic interaction of Plasmodium vivax reticulocyte binding protein 2b (PvRBP2b) and P. vivax Duffy-binding protein (PvDBP) with reticulocyte-tropic transferrin receptor 1 (TfR/CD71) and Duffy receptor antigen for chemokines (DARC), respectively, provide compelling evidence of erythroblast and reticulocyte infection in haemopoietic tissue niches including bone marrow [4]. The parasite would thus appear to have almost no opportunity to reproduce in peripheral blood but, would have almost limitless access to cells bearing that receptor in the extravascular spaces of haemopoietic tissues. Indeed, recent work with P. vivax in splenectomized rhesus macaques demonstrated those spaces of bone marrow as the dominant compartment of that infection [7]. Plasmodium falciparum has also been shown to interact with bone marrow; employing it as a site of maturation and for evasion of the host immune response [8‐13]. For initiation of active P. falciparum merozoites entry, the erythrocyte binding-like (EBL) and reticulocyte binding-like (Rh) protein families are responsible for binding to specific erythrocyte receptors [14]. In fact, higher prevalence of immature P. falciparum immature gametocytes in bone marrow than in peripheral blood has been demonstrated using quantitative polymerase chain reaction [15]. The anaemia that often develops with acute malaria involves bone marrow suppression and ineffective erythropoiesis [16‐19].

Samples of bone marrow are obtained by an invasive procedure and collected when clinically indicated by suspicion of specific illnesses, or at autopsy. The presence of malarial parasites in the bone marrow was first reported in 1894 in malarial fevers occurring during summer and autumn [20]. Later, bone marrow aspiration was described as an accessory and useful tool in the diagnosis of malaria especially in cases where strong suspicion of malaria could not be confirmed on routine peripheral smears [9‐11]. Peripheral smear examination under light microscopy remains the gold standard of malaria diagnosis and consistently negative findings in a patient are considered definitive. Point-of-care rapid diagnostic tests (RDTs) on peripheral blood are also widely used. However, malaria in endemic zones is often dominated by sub-patent and typically asymptomatic infections [21]. The finding of Plasmodium in samples of bone marrow in endemic settings may thus occur with negative findings from peripheral blood [22‐24].

Currently, bone marrow trephine biopsy and aspiration are routinely done for the workup of various haematological, infiltrative and storage abnormalities [25, 26]. In general, plasmodial infection of bone marrow is usually an incidental finding to those investigations. It is not clear if undiagnosed malaria explained the illnesses prompting bone marrow examination, or if it was clinically silent and of no direct relationship to the presenting illness. The findings on bone marrow of patients with malaria range from normocellular, normoblastic erythropoiesis to hypercellular, dyserythropoietic erythropoiesis [27‐29]. The current study adds to that scant body of evidence by reviewing the finding plasmodial infection incidental to bone marrow examination at teaching hospital in Karachi, Pakistan.

A retrospective record review of all malaria parasite-positive bone marrow specimens examined at Aga Khan University Hospital (AKUH), a large tertiary care hospital in Karachi, Pakistan, from the years 2007 to 2015 was conducted. Endemic malarial transmission occurs seasonally in and around Karachi, and across much of Pakistan [30].

Bone marrow aspiration was either done at Aga Khan University, Main Campus or submitted for review at one of the collection points across Pakistan. A complete blood cell count and peripheral blood smear was also routinely done at the time of bone marrow aspiration. In all cases, bone marrow aspirate and trephine biopsy were collected from the posterior superior iliac spine. Bone marrow aspirates were collected as per International Council for Standardization in Haematology (ICSH) guidelines. Ten to 20 ml plastic syringe, without anticoagulant, was used to draw out the aspirate. The first 0.5ml were used to produce smears at the bedside while the remainder was put in a tube containing a proportional amount of di-potassium ethylenediaminetetraacetic acid (EDTA) to prevent sample coagulation. Bone marrow slides were prepared immediately after aspiration. The slides were stained with Leishman stain for microscopy.

Core biopsies of bone were taken using a Jamshidi needle. Whenever possible, a 2 cm long core was obtained in each case. Specimens were fixed and transported in 10 % buffered formalin. After decalcification in 10 % formic acid, paraffin-wax embedding was done. Sections, 4-µm‐thick, were cut by experienced histotechnologists using standard equipment and used for haematoxylin and eosin staining.

Patients included in the study were either already diagnosed with malaria and their bone marrow was examined for work-up of another condition, or the finding of malarial pigment/parasites was incidentally discovered in the bone marrow sample.

A peripheral blood smear was considered positive when any plasmodial life cycle stage was identified. In addition, bone marrow aspirate was considered positive for malarial parasite when any stage of plasmodial life cycle or haemozoin was identified.

Following exemption from the ethics review committee (3672-Pat-ERC-15), anonymized data was collected from past records and analysed using Statistical Package for the Social Sciences (IBM SPSS) version 19. The data reviewed for each patient included: (a) patient demographics, (b) relevant, significant findings on history and physical examination, (c) a complete blood count and blood peripheral film of a blood sample taken at the time of bone marrow sampling and (d) final report of bone marrow aspirate and/or trephine biopsy done on bone marrow samples.

Frequencies and percentages were calculated for categorical variables and mean and standard deviation for continuous variables. For statistical analysis, Fisher’s exact test was used for categorical variables whereas Mann-Witney U was employed for continuous data.