The setting of the study: The study was conducted at the Department of Medicine and Department of Pathology, Institute of Post Graduate Medical Education and Research, Kolkata, West Bengal.

Duration of the study: The study was conducted over 2 years, between March 2016 to March 2018.

Type of Study: A prospective descriptive clinical study.

Sampling method: All HIV positive patients (those who satisfied the inclusion and exclusion criteria), both symptomatic and asymptomatic, reporting to antiretroviral therapy centre and admitted to a tertiary care institution of West Bengal were included within the study. HIV was diagnosed by the ELISA method as per NACO guidelines.

Sample size calculation: A total of 100 HIV patients were included in the study.

Inclusion criteria: Indoor patients from medicine wards and those attending ART clinics included in the study.

Exclusion criteria: Patients of malignancy not related to HIV disease and patients receivingchemotherapy were excluded.

Data Collection Procedure: Detailed history was taken which mainly included age, sex, place of residence, occupation, history of blood or blood product transfusions, high-risk behavior, fever, weight loss, diarrhoea, oral or genital ulcerations, bleeding diathesis or history suggestive of systemic involvement. All patients were subjected to thorough physical examination both, systemic and general with necessary investigations like CD4 count, Complete Hemogram by cell counter, USG abdomen, CSF examination, CT scan etc.

Definition of AIDS and other parameters: HIV was diagnosed by the ELISA method as per NACO guidelines. All patients were classified as those having AIDS (CD4 ≤200/μL) and Non-AIDS (CD4 > 200/μL), according to NACO criteria. Hematologic parameters were obtained by Sysmex KX-21 automated cell counter.In our study, Anemia is defined as hemoglobin level less than 10 gm/dl. Leucopenia was defined as a total count less than 4000 cells/cumm, Neutropenia was defined as absolute neutrophil count <1000 cells/cumm.

We had included 100 patients detected as HIV positive by ELISA method, reporting to ICTC clinic and admitted in medicine ward of our institution.

Table 1: Age distribution of patients.

The commonest age group involved, was 21 to 40 years (82%), similar findings were reported from other studies. [Table 1] Out of 100 HIV positive patients, 87 were male and 13 were female patients. There was a male preponderance with a male to female ratio of 6.69:1.

Pie Chart 1: Occupation of patients.

The commonest population affected was that of drivers (43%) and labourers (25%) [Pie Chart 1]

Table 2: CD4 Counts.

All cases were divided into two groups’ on-ART and Non-ART with 62 and 38 patients respectively. CD4 count could be done in 91 out of 100 patients due to technical difficulty and financial problems. In that 37 patient’s CD4 counts <200/microliter considered as AIDS according to the CDC criteria and 54 patients CD4 counts >200/microliter considered as Non- AIDS [Table 2].

Anemia (Hb % <10g/dl) was found in 95.45% cases, of this normocytic normochromic in 77.25%.

Total WBC counts 2185 ± 960, 9482 ± 4925, Granulocytes counts 974 ± 646, 4431 ± 1948, Lymphocyte counts 960 ± 505, 4340 ± 3847,Granulocyte% 42.67 ± 18.18, 54.05 ± 20.37, Platelet counts (lac/mm³) 1.84 ± 0.72, 2.14 ± 0.97,CD4 counts /microliter 166 ± 23, 385 ± 156 (Table 3), Except platelet count, these values showed statistical significance.

Table 4: Bone marrow Cellularity findings in patients on ART and non-ART.

In the peripheral blood smear- various morphological abnormalities were detected in RBC like anisopoikilocytosis, basophilic stippling, acanthocytes. Polychromasia and nucleated RBCs were also detected. Amongst the WBC series – pseudo-pelger-huet cells and activated lymphocytes were found in smear. Even, giant platelets were detected in some of the cases.Bone marrow morphology was assessed and analyzed considering features like adequacy, cellularity, myeloid to erythroid ratio, features of erythroid, myeloid and megakaryocytic series, lymphoid cells, plasma cells, macrophages, iron content, presence of abnormal cells, parasites, fungi and acid-fast bacilli.In patients on ART bone marrow were hypercellular, hypocellular, and normocellular in 09 (14.51%), 17(27.41%), and 36(58.06%) respectively. In the non-ART group, bone marrow was hypercellular, hypocellular, and normocellular in 06(15.89%), 05(13.15%), 27(71.05%) respectively. The bone marrow had been found normocellular in the majority cases of the non-ART group as compared to ART (p-value =0.9145)[Table 4].

Table 5: Bone-Marrow cellularity of AIDS and Non-AIDS patients.

In our study, we also found that most of the marrow was normocellular 85.71 % of patients in AIDS and non-AIDS, hypercellular in 5.49%, hypocellular in

Dysplasia was considered based on anemia and myelodysplasia was noted in two cell lines. Among the AIDS patients, dysplasia in above mentioned two cell lines (erythroid 16.21 % and granulocytic 18.91%) were more common than the non-AIDS patients (erythroid 7.40 % and granulocytic 3.70 %). We did not get any case of dysmegakryopoiesis [Table 6].

Bar Diagram 1: Opportunistic infections and Bone Marrow abnormalities (apart from Myelodysplasia) in HIV patients.

Out of 62 Cases, which were taking HAART, 4 patients with dysgranulopoiesis had underlying pathology of pulmonary tuberculosis, compared with 1 case belonging to a non-ART group(n=38). [Bar Diagram 1]Plasmacytosis in the bone marrow had been a pretty common feature, both in ART(4 cases) and non-ART (7 cases) groups.

Photomicrogram 3: Showing the presence of Histoplasma in Bone Marrow Aspiration Smear.

We got two interesting cases with opportunistic infections. The presence of LD bodies (one case each in ART and non-ART group) [Photomicrogram

That defective progenitor cell growth might be secondary to suppressor T cells is suggested by the observation that T-lymphocyte depletion enhances colony formation by progenitor cells in AIDS patients but not in normal individuals. The inhibition appears to be mediated by a serum antibody directed to an HIV structural protein. In addition, there is now evidence that CD34 progenitor cells from normal bone marrow and fetal hepatic hematopoietic cells can be infected with HIV in vitro although the extent of infection may be limited in asymptomatic patients. There has been an alteration of the T4:T8 ratio. The ratio is seen to be inversely proportional to HIV infectivity.[1]. As a result, in the peripheral blood picture, we found anemia, bi or pancytopenia.

Comparison of Hematological changes in our study with other studies: Patients were classified according to CD4 count in the line of NACO guidelines into two groups- CD4 count < 200/cumm into AIDS group and >200/cumm into the non-AIDS group. In our study 40.65% of patients in AIDS (37 patients), while 59.34% in non –AIDS (54 patients), as we were able to do CD4 count only in 91 patients out of 100 patients.Our study reveals that anemia was one of the most common findings in HIV positive patients. Jerry L. et al (1984)showed thrombocytopenia was initially in 3(25%) patients and subsequently developed in 2 others. Patwardhan M.S et al (2002) showed thrombocytopenia in 65 patients (13%) with average platelet count 0.92x10 (3)/ul[18].In our study thrombocytopenia on marrow showed increased megakaryocytes number secondary to increase peripheral destruction of the platelet. It is mainly due to auto-immune mediated destruction of platelet in peripheral blood.

Comparison of changes of Bone- Marrow cellularity in our study with other studies: In the ART group, bone marrow was normocellular in 36 patients and hypocellular marrow in 17 patients, which was due to patchy involvement mostly by fatty degeneration of bone marrow. Hypercellular marrow was in 9 patients because of erythroid hyperplasia. Whereas patients on Non-ART, normocellular bone marrow in 27 patients. Hypocellular marrow was found in 5 patients, hypercellular in 6 patients. The difference between the ART and Non-ART group found to be statistically insignificant (P= 0.2423). This finding is in concordance with the study done by Castellaet al,who reported normocellular bone marrow in

5(38%) in group A and 5(24%)in group B [11]. Various dyspoietic features seen in the erythroid series in our study were megaloblastoid change, bi/multinucleation, cytoplasmic vacuolation, nuclear budding, micronormoblast and ringed sideroblasts. Megaloblastoid changes seen in HIV related myelodysplasia is unrelated to serum cobalamin and folate levels, or drug therapy with zidovudine orfolate antagonists, although these drugs may accelerate it [1]. Dysplastic changes involving myeloid series, seen in our study include nuclear dysmorphism, giant metamyelocytes, cytoplasmic vacuolation,hypogranulation and pseudo-pelger-huet anomaly. Direct infection of marrow precursors by HIV may contribute to these defects, although this issue remains controversial.

Comparison of opportunistic infections and other changes of bone marrow in our study with other studies: Commonest infection was found to be pulmonary tuberculosis; among them 4 patients also on HAART (Nevirapine, zidovudine) therapy. 2 cases were having seropositivity of HBsAg and 1 case having anti-HCV. These cases were on ART therapy.Plasma cells were often strikingly increased in the marrow of HIV infected patients seen in 31-85% of patients [14]. Our patient population had plasmacytosis in 12.08 % of patients (4 cases of ART-group and 7 cases of the non-ART group). It might represent a physiological response to antigenic stimulation by viruses or other infective agents or may be secondary to dysregulated B-cell proliferation due to HIV [15]. Another interesting aspect of our study was to identify histoplasma (one case) and L-D bodies (2 cases, one case each from ART and non-ART group) in the bone marrow.Bone marrow aspirate stained by Giemsa showed oval globose yeast-like cells measuring 3–4 mm × 2–3 mm in size, suggestive of H. capsulatum in proximity to irregular clusters of epithelioid cells showing vague granuloma formation. However, no giant cell (Langhans’ or foreign body) on those aspirate slides.PAS stain revealed these yeast-like cells as bright eosinophilic structures.We have detected 2 cases of visceral leishmaniasis, where the Amastigote form of leishmania donovai was detected both extracellular and within histiocytes. Bone marrow findings of those two cases of visceral leishmaniasis are summarized below–

1. Presence of LD bodies
2. Decreased myeloid-erythroid ratio indicating relativesuppression of myelopoiesis.

3) No follow up was done to review the haematological changes on HAART with improvement in CD4 counts.

Our study emphasized the need for a thorough examination of bone marrow in each case of HIV infection. It had been noticed that one of the underlying causes of Anemia/Pancytopenia might be Mylelodysplasia. Hypocelluarity of bone marrow is a common picture in the advanced stage and comparatively high incidences of myelodysplasia have been found in AIDS patients. Opportunistic infections like Leishmaniasis and disseminated Histoplasmosis often involve the bone marrow. So, any suspicious case of such infections needs a thorough examination of bone marrow.

Prof.(Dr.) Mamata Guha Mallick (Sinha): Concepts, Manuscript review, Guidance of research work. Dr. Soumya Kanti Pramanik: Literature Search, Data acquisition, statistics, Manuscript Preparation.