Peripheral Blood and Bone Marrow Findings in Chronic Alcoholics with Special Reference to Acquired Sideroblastic Anemia

Abstract

Anemia associated with alcoholism has numerous causes, most common being megaloblastic anemia and acquired sideroblastic anemia (SA). The bone marrow aspirate (BMA) and bone marrow iron (BMIr) findings and their correlation with peripheral blood smear (PBS) have not been extensively described in literature. We aim to study the spectrum of hematological abnormalities in chronic alcoholics. Complete blood count (CBC), PBS, BMA and BMIr of 71 chronic alcoholics were studied retrospectively over a period of 3 years. The slides were reviewed by 2 pathologists. The clinical history, CBC, PBS, BMA and BMIr findings were recorded. Out of 71 patients, 68 (95.77%) had anaemia. Red cell morphology varied from normocytic-normochromic, microcytic-hypochromic, macrocytic, to dimorphic anaemia. Principal findings seen on BMA were erythroid hyperplasia and megaloblastic maturation. BMIr was available in 41 patients; iron stores were decreased in 2 (4.88%), normal in 14 (34.15%), increased in 25 (60.97%). Seven (17.07%) cases showed presence of ring sideroblasts. Chronic alcoholics show a variety of abnormalities in BMA, which closely mimic many haematological disorders. A history of alcoholism should always be taken in these circumstances. SA should be ruled out in all chronic alcoholics with anaemia not responding to vitamin B₁₂/folic acid, even with macrocytic picture on PBS.

Introduction

Patients with chronic alcoholism show a variety of hematological abnormalities [1]. Anemia associated with alcoholism has numerous causes, most common being megaloblastic anemia due to folic acid deficiency and acquired sideroblastic anemia [2]. Peripheral blood smear (PBS) findings of sideroblastic anemia (SA) can be masked by coexisting nutritional deficiencies and potentially pose as a diagnostic dilemma [2]. SA are a reversible cause of anemia and respond to a trial of pyridoxine, hence it is important to recognize this entity in patients presenting with anemia not adequately responding to haematinics [3]. As chronic alcoholics do not routinely undergo bone marrow aspiration (BMA), the BMA and bone marrow iron (BMIr) findings and their correlation with PBS has not been extensively described in literature. We aim to study the spectrum of hematological abnormalities in chronic alcoholics.

Materials and Methods

It was a retrospective observational study conducted in the Department of Pathology (Division of Hematology) of a tertiary care general hospital. Seventy-one patients clinically labeled as alcoholics, who had undergone bone marrow examination due to abnormalities in peripheral blood counts, without established diagnosis of specific hematologic diseases, malignancies, or infections, during the period 2013–2016 were included in the study. In our study, patients with history of heavy drinking for more than 5 years were labeled as alcoholics. Heavy drinking was defined as more than 8 drinks per week for women and more than 15 drinks per week for men [4]. Assessment included relevant clinical history, complete blood counts (CBC), morphological review of peripheral blood and bone marrow, and bone marrow iron stores. Hemogram: Hb, hematocrit (Hct), mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH) and mean corpuscular hemoglobin concentration (MCHC) was determined by Bechman Coulter LH 500. Anemia was graded as mild (11.0–12.9 g/dL for males, 11.0–11.9 g/dL for females), moderate (8.0–10.9 g/dL) and severe (< 8.0 g/dL) [5]. Thrombocytopenia was graded as mild (100,000–150,000/µL), moderate (50,000–100,000/µL) and severe (< 50,000/µL) [6]. PBS and BMA stained by Wright’s stain were independently reviewed by 2 pathologists. Fixed smears of BMA stained by Prussian blue were graded for hemosiderin stores from 0 to 6. In these smears, at least 100 normoblasts were inspected for intracellular iron inclusion bodies. Ring sideroblasts were defined as normoblasts with minimum 5 siderotic granules, covering at least one third of the circumference of the nucleus [7]. Statistical analysis was performed using MS EXCEL.

Results

Clinical Profile

Mean ± SD age of the patients included in our study was 40.04 ± 13.63 years, majority being in the age group of 31–50 years (67.6%). 70 of the 71 patients in the study were males. Of the 71 patients, 38 (53.52%) had history of single or more episodes of jaundice in the past. Thirty two (45.07%) patients had hepatosplenomegaly on examination, at the time of BMA. Fifty-seven (80.2%) patients had history of intake of hematinics, out of which 35 (49.30%) patients received blood transfusion in the past.

Complete Blood Counts and Peripheral Smear Morphology (Table 1 and Fig. 1)

Table 1.

Complete blood count findings in patients

Parameter	N	Incidence (%)
Hb (g/dL)
Anemia	68	95.76
Severe (< 8.0)	58	85.29
Moderate (8.0–10.9)	7	10.29
Mild (11.0–12.9 for males, 11.0–11.9 for females)	3	4.42
Platelet count/microL
Thrombocytopenia	58	81.69
Severe (< 50,000)	24	41.38
Moderate (50,000–100,000)	24	41.38
Mild (100,000–150,000)	10	17.24
TLC/mm3
Leukopenia (< 4000)	40	56.33
Leukocytosis (> 11,000)	1	1.41
MCV (fL)
Normal (83–101)	38	53.52
Increased	30	42.25
Decreased	3	4.23

Hb hemoglobin, MCV mean corpuscular volume, TLC total leucocyte count

Fig. 1.

a Wrights stain (×400): Macrocytes, macroovalocytes, and normocytes on peripheral smear, b (×1000): bone marrow aspirate showing erythroid hyperplasia with megaloblastic maturation and giant myeloid forms and c (×1000) Perl’s stain shows ring sideroblasts (marked with arrow)

Out of 71 patients, 34 (47.88%) had pancytopenia, 68 (95.77%) patients had anemia, 58 (81.69%) had thrombocytopenia, and 40 (56.33%) had leukopenia. 1 (1.41%) patient had polycythaemia, and 2 (2.82%) patients had leucocytosis.

Hb levels in our patients varied from 3 to 17.7 g/dL, mean ± SD being 6.2 ± 2.63 g/dL. Out of 68 patients, anemia was severe in 58 (85.29%), moderate in 7 (10.29%) and mild in 3 (4.42%) patients. Platelet counts varied from 4 to 480 × 10³/mm³, mean ± SD being 92.7 ± 77.6 × 10³/mm³. Thrombocytopenia was severe in 24 (41.38%), moderate in 24 (41.38%), and mild in 10 (17.24%) patients. Total leukocyte count (TLC) varied from 0.7 to 17.7 × 10³/mm³, mean ± SD being 4.42 ± 2.75 × 10³/mm³. MCV ranged from 68 to 125 fL, with a mean ± SD of 99.3 ± 12.33 fL. MCV was normal in 38 (53.52%), increased in 30 (42.25%), and decreased in 3 (4.23%) patients. Red cell morphology varied from normocytic normochromic 5 (7.04%), microcytic hypochromic 4 (5.63%), macrocytic 23 (32.39%) (Fig. 1a), to dimorphic anemia 37 (52.11%). Red cell morphology was poorly preserved in 2 (2.83%) patients. 18 (25.35%) patients showed nucleated red cells in the PBS.

Bone Marrow Aspirate and Bone Marrow Iron

Fifty-six (78.87%) of the BMA were adequate for opinion. Sixteen (28.57%) marrows were normocellular for age, 38 (67.86%) marrows were hypercellular for age, and 2 (3.57%) marrows were hypocellular for age. 42 (75.0%) of the aspirates showed evidence of erythroid hyperplasia and 3 (5.36%) aspirates showed evidence of erythroid suppression. Erythroid series showed normoblastic maturation in 18 (32.14%) patients, megaloblastic maturation in 36 (64.29%) patients (Fig. 1b), and micronormoblastic maturation in 2 (3.57%) patients, with evidence of dyserythropoiesis in 33 (58.93%) patients. Myeloid series showed giant myeloid forms in 24 (42.86%) and evidence of dysmyelopoiesis in 3 (5.36%) patients. Plasma cells were increased in 4 (7.14%) patients and histiocytes were increased in 3 (5.36%) patients. BMIr was available in 41 patients; iron stores were decreased in 2 (4.88%), normal in 14 (34.15%), and increased in 25 (60.97%) patients. Seven (17.07%) cases showed presence of ring sideroblasts (Fig. 1c).

Correlation of Bone Marrow Iron and Peripheral Smear Morphology (Fig. 2)

Fig. 2.

Correlation of bone marrow iron and peripheral smear morphology

In 25 patients with increased iron stores, red cell morphology varied from macrocytic (12), dimorphic (macrocytes and microcytes) (11), microcytic (1) and normocytic (2). In patients with normal iron stores (14), red cell morphology on PBS was dimorphic (10) or macrocytic (4). Twenty-one (53.85%) patients with increased or normal iron stores, showed presence of microcytes on PBS. In patients with decreased iron stores (2), red cell morphology was dimorphic on PBS.

Peripheral Blood and Bone Marrow Findings in Patients with Ring Sideroblasts

In these cases, 5/7 patients had pancytopenia, 2/7 had anemia and thrombocytopenia on blood counts. Anemia was severe in 5/7, and moderate in 2/7 cases. PBS showed macrocytic anemia in 3/7, and dimorphic anemia (macroytic and microcytic) in 4/7 cases.

Six out of seven BMA were hypercellular, and 1/7 was normocellular for age. Erythroid series showed hyperplasia in 5/7 cases, and megaloblastic maturation in 7/7 cases. Other findings seen were dyserythropoiesis (6/7), dysmyelopoiesis (2/7), and dysmegakaryopoiesis (1/7). BMIr was increased in 6/7, and normal in 1/7 cases.

Discussion

The adverse effects of alcoholism on haematopoiesis and iron metabolism have been well documented in literature. These include, megaloblastic anemia secondary to folic acid deficiency, iron deficiency after blood loss or inadequate intake, SA, chronic hemolytic anemia associated with cirrhosis, acute spherocytic haemolytic anemia, transient hemolysis with hyperlipemia, thrombocytopenia, granulocytopenia and vacuolated marrow precursor cells [1]. However, not many studies have correlated the BMA and BMIr findings with their PBS findings. We studied the spectrum of hematological abnormalities in chronic alcoholics.

Majority of the patients in our study were middle aged males as alcoholism is more prevalent among this gender and age group. Alcoholism is a leading cause of chronic liver disease and significant number of our patients had history of jaundice and hepatosplenomegaly. Eighty% (57/71) had history of hematinic intake, 60% (35/57) of whom had to undergo blood transfusion for the management of anemia. Hence, 50% of our clinically labelled chronic alcoholics presented to us with anemia not responding to modalities other than blood transfusion, at the time of BMA.

Diverse patterns of hematological effects were observed ranging from abnormalities in blood cell indices to bone marrow morphology. The primary coulter findings were pancytopenia (47.88%), anemia (95.77%), thrombocytopenia (81.69%), leukopenia (56.33%) and increased MCV (42%). On PBS, macrocytic (32.4%) anemia and dimorphic anemia (52%), were frequent findings.

Anemia was found in > 95% of the alcohol abusers, severe in 85.3% patients. On morphology, it was more commonly dimorphic (52%) and macrocytic (32.4%), whereas only 4 patients had microcytic hypochromic anemia. Eichner et al. had demonstrated that majority of alcoholics have abnormalities in red cell production with > 40% having megaloblastic anemia. The various causes of macrocytic anemia in alcoholics include folate deficiency due to depletion of already subnormal folic acid stores in chronic alcoholic subjects, interference with absorption of folic acid or inhibition of the conversion of folic acid to N-5-methyltetrahydrofolic acid, reticulocytosis, liver disease and macrocytosis of alcoholism [1].

More than 95% of alcoholics had normal to raised MCV. Increased MCV is a hallmark of macrocytic anemia in alcohol abusers and has long been proposed to be used as a screening test for detecting alcohol abuse. Acetaldehyde, the first metabolite of ethanol, has been demonstrated to impair cell proliferation and disturb cell growth in vitro, increasing MCV classic of macrocytosis of alcoholism. Post abstinence normalisation of MCV may take 2–4 months [1].

Isolated thrombocytopenia is commonly seen in severely alcoholic patients. It is often due to direct toxic effect of ethanol on platelet production [1]. With withdrawal of alcohol, the platelet counts return to normal in 1–3 weeks. We observed thrombocytopenia in 82.7% of alcoholics which was moderate-severe in 82.3% patients. Similar findings were reported earlier by Cowen and John et al. [8] with 81% alcoholics having severe thrombocytopenia.

The proposed mechanisms of ethanol induced hemato-toxicity include structural modifications and direct suppression of hematopoietic stem cell proliferation. This may be responsible for leukopenia occurring along with anemia and thrombocytopenia as reported by Latvala et al. [9]. We found pancytopenia in 47.88% and leukopenia in 56.33% of patients.

Seventy-five percent of the adequate BMA showed erythroid hyperplasia with megaloblastic maturation in 64% and evidence of dyserythropoiesis in 59%. BMIr is considered the gold standard for assessment of iron stores. BMIr was raised in 25 (60.97%), reduced in 2 (4.88%) and normal in 14 (34.15%) patients. This is in concordance with various studies which show that there is evidence of increased iron stores in alcoholics [1, 2]. Of the 41 cases assessed for iron stores, 7 showed ring sideroblasts.

Sideroblastic anaemias can be acquired or hereditary, acquired causes being more common. Acquired causes include drugs and toxins, myelodysplastic syndrome (MDS) and nutritional deficiencies; hereditary causes are mostly X linked, few being autosomal dominant or recessive [10]. Alcindor and Bridges [10] stated that ethanol is the most common cause of toxin induced SA and is a result of dietary deficiency and/or antagonism of pyridoxal phosphate. SA is a frequent cause of anemia in chronic alcoholics second only to megaloblastic anemia. Also, megaloblastic and SA are frequently found to coexist. Cazzola and Invernizzi [11] stated that congenital SA show microcytic hypochromic cells, basophilic stippling, and only occassionally normocytes and macrocytes on PBS. In contrast, on correlation with PBS, our study showed the presence of macrocytic anemia in 3/7, and dimorphic anemia (macroytic and microcytic) in 4/7 cases. The PBS and BMA findings in acquired SA in chronic alcoholics have not been extensively studied in the literature. In our study, the PBS findings of SA have been possibly masked by the presence of coexistent macrocytic anemia.

SA in alcoholics is a reversible cause of anemia and can be treated by administration of pyridoxine or pyridoxal phosphate [3]. Hence we wish to emphasize that a possibility of SA should be considered in chronic alcoholics with anemia not adequately responding to vitamin B12/folic acid, even with macrocytic picture on PBS.

Numerous studies have stated that iron metabolism in chronic alcoholics is dysregulated [1, 2]. Twenty-one (53.85%) of our patients with increased or normal iron stores, showed presence of microcytes on PBS pointing towards defect in utilisation of iron stores.

The limitation of this study is that no follow up of these patients is available as it is a retrospective, descriptive cross sectional study. Although BMIr is gold standard for assessment of iron overload as well as iron deficiency, but serum iron and transferrin saturation are minimally invasive investigations to assess iron status, which were not done in our study, as all patients were directly taken up for bone marrow examination. Also, serum levels of vitamin B₁₂ and folate were not available. MDS is a diagnosis of exclusion, and is a disease of elderly patients mostly. Patients in our study were middle aged males with a history of alcohol intake; however, cytogenetics was not performed in our study to exclude MDS.

To conclude, chronic alcoholics show a variety of abnormalities in BMA, which closely mimic many hematological disorders. A specific history for alcoholism should always be taken in these circumstances. Not all SA in chronic alcoholics show a microcytic hypochromic or dimorphic (microcytic and normocytic) picture on PBS. Hence, SA should be ruled out in all chronic alcoholics with anemia not responding to haematinics, regardless of the picture on PBS. Further studies should be undertaken to establish if a trial of pyridoxine/pyridoxal phosphate can be beneficial in chronic alcoholics with anemia not adequately responding to B₁₂/folic acid therapy prior to blood transfusion.

Abbreviations

PBS

Peripheral blood smear

SA

Sideroblastic anemia

BMA

Bone marrow aspiration

BMIr

Bone marrow iron

Hct

Hematocrit

MCV

Mean corpuscular volume

MCH

Mean corpuscular hemoglobin

MCHC

Mean corpuscular hemoglobin concentration

MDS

Myelodysplastic syndrome

CBC

Complete blood count

Funding

None.

Compliance with Ethical Standards

Conflict of interest

The authors declare that there are no conflicts of interest.