Original Articles      172

Megaloblastic anaemia: Prevalence and causative factors

Uma Khanduri, Archna Sharma

ABSTRACT
   Background. Megaloblastic anaemia is not uncommon in India, but data are insufficient regarding its prevalence, and causative and precipitating factors. We did a prospective study to document such data for patients of megaloblastic anaemia.
   Methods. All patients presenting to our hospital over a period of 6 months with a haemoglobin <10 g/dl and/or mean corpuscular volume >95 fL and blood film findings consistent with megaloblastosis were included in the study. Demographic data, diet, drug intake, previous blood transfusion and presenting symptoms were recorded. Clinical findings were obtained from medical records of patients. Complete blood counts, blood film examination, reticulocyte count and cobalamin and folate assays were done. Results of liver function tests and bone marrow slides were available for review.
   Results. Megaloblastic anaemia was diagnosed in 175 patients with anaemia. Assays were done on 120 patients (55 were lost to follow up) and results showed cobalamin deficiency in 78 patients (65%), combined cobalamin and folate deficiency in 20 patients (12%) and pure folate deficiency in 8 patients (6%). Fifteen per cent of patients had normal or high values of both vitamins, having received blood or haematinics before the diagnosis was established. The peak incidence of megaloblastic anaemia was in the age group of 10–30 years (48%), with female preponderance (71%). The predominant symptoms were fatigue, anorexia and gastritis, low grade fever, shortness of breath, palpitations and mild jaundice. Twenty-five per cent of patients were on acid-suppressing medication and 15% had previous transfusion for anaemia. Eighty-seven per cent of patients with cobalamin deficiency and 75% with folate deficiency were lactovegetarians. In the combined deficiency cohort, 71% were vegetarians and 29% were occasional non-vegetarians. Physical findings were pallor (85%), glossitis (29%), mild icterus (25%) and hyperpigmentation (18%).
   Abnormal haematological findings were mean corpuscular volume 77–123 fL (9 patients had iron deficiency), red cell distribution width 16%–44%, pancytopenia in 62% of patients, reticulocyte count >2% in 42% of patients and typical megaloblastic blood films in all patients. Bone marrow smears available in 22 patients showed moderate-to-severe megaloblastosis. Thirty-two per cent of patients in whom liver function tests were done showed indirect bilirubinaemia with normal enzymes.
   Conclusion. Megaloblastic anaemia was diagnosed from complete blood counts, red cell indices, blood film examination and assays of the two vitamins. Bone marrow examination was not essential for diagnosis. Cobalamin deficiency was the major cause of megaloblastosis. Aetiological factors were a diet poor in cobalamin or folate, increased requirements during the growth period and pregnancy, and the use of acid-suppressing medication. Physicians managing these patients need to be aware of the timing of blood sampling for assays, that haematinics and transfusions provide only short term benefits, and that long term follow up and diet counselling is crucial.

   Three patients were pregnant at the time of investigation and all of them had cobalamin deficiency. Two girls <1 year of age with severe anaemia had cobalamin deficiency. The nutritional status of their mothers was not known. All patients were residents of Delhi and its suburbs within a radius of 25 km from central Delhi. Eighty-seven per cent of patients with cobalamin deficiency and 75% of patients with folate deficiency were lactovegetarians. In the combined deficiency cohort, 71% were lactovegetarians. Even non-vegetarian patients ate meat only occasionally. All patients were from the middle and low income groups. A history of intake of H2 receptor blockers or proton pump inhibitors namely ranitidine and omeprazole was obtained in 30 patients. These drugs had been prescribed or bought over-the-counter for symptomatic relief of gastritis and anorexia.
   The predominant symptoms were fatigue (70%), anorexia and gastritis (60%), low grade fever (50%), cardiovascular (shortness of breath, palpitations and syncope) (30%) and yellow discoloration of eyes (20%). Paraesthesias, diarrhoea, hyperpigmentation and early graying of hair were present in <10% of patients. The duration of symptoms ranged from a few days to 3 years. Eighteen patients (15%) had received blood transfusions for anaemia, 1–3 years before the present hospital visit.
   The physical signs recorded by the attending physicians included pallor (85%), glossitis (29%), mild icterus (25%) and hyper-pigmentation of knuckles (18%). A detailed neurological evaluation was not recorded.
   The main haematological findings are shown in Table II. The MCV ranged from 77 fL to 123 fL. Nine patients whose MCVs were <95 fL and belonged to Group C were also iron-deficient. The red cell distribution width (RDW), which is an indicator of the variation in the size of red cells, ranged from 16% to 44% (normal: up to 13.5%).
   Pancytopenia was present in 74 patients (62%). Reticulocyte count was done in 74 patients and was found to be >2% in 42%.

FIG 1. Age and sex distribution of patients

FIG 2. Peripheral blood film (´1000) showing paucity of red cells, marked anisopoikilocytosis, tear drop cells. Macrocytes were recognized by comparing the size of red blood cells with that of mature small lymphocytes (L). Platelets were reduced in number. Basophilic stippling was seen (arrow).

Table II. Haematological data

* 10 patients had haemoglobin (Hb) >10 g/dl with mean corpuscular volume (MCV) >95 fL (Hb normal range 11.5–18 g/dl and MCV normal range 82–92 fL)    RDW Red cell distribution width (normal 11.5%–13.5%)    
White cell count normal range 4.5–10.5x109/L

accounted for 6%. We were unable to determine which vitamin was deficient in 17% as the patients had received haematinics and blood transfusions before sampling for the assays could be done.
   Megaloblastic anaemia is a chronic condition developing over a period of time and most patients are well compensated. There is no indication for urgent blood transfusion. Serum samples for assay of the 2 vitamins should be drawn before any form of therapy is given since assays alone can determine which vitamin is deficient.
   The majority of our patients were lactovegetarians. The average Indian vegetarian diet is deficient in cobalamin.^7,8 An earlier pilot study reported by us had shown that 40% of normal Indian subjects with normal haemograms were cobalamin-deficient.⁶ A 1973 study by WHO on the nutritional status of pregnant women in India documented iron, folate and cobalamin deficiency.⁹In obstetric practice supplementation of iron and folate is the norm. Folate supplementation alone in the presence of occult cobalamin deficiency may precipitate neurological complications.^2,10
   We attempted to identify factors that might be responsible for converting occult cobalamin deficiency into florid megaloblastic anaemia. In Caucasian and Chinese populations, megaloblastic anaemia is reported to occur in older age groups with an equal sex ratio or male preponderance.^11,12 In contrast, the peak incidence in our study was seen in the age group of 10–30 years (48% of patients) and there was a preponderance of women (71%). It is possible that the increased demand during growth spurt, puberty and child-bearing age group in a population already deficient in cobalamin precipitated the anaemia.
   Gastritis, anorexia, nausea and vomiting were present in 60% of patients. The lining epithelium of the gastrointestinal tract becomes atrophic in megaloblastosis. A vicious cycle of megaloblastosis leading to atrophy of mucosa, and subsequent malabsorption of the two vitamins, worsens megaloblastic anaemia.
   A history of intake of acid-suppressing medication (H2 receptor antagonists and proton pump inhibitors) was present in 25% of our patients. These drugs had been prescribed for gastritis by their primary physicians and were often purchased from pharmacies over-the-counter without prescriptions. These drugs may play a role in malabsorption of cobalamin.^13,14 Strict regulation in prescribing and dispensing these medications should be considered. We did not document malabsorption in these patients.
   For a laboratory diagnosis of megaloblastic anaemia, a CBC with red cell indices, examination of a well stained blood film and assay of the 2 vitamins are sufficient to make a definitive diagnosis. Pancytopenia was present in 62% of patients. Other authors have also observed that megaloblastic anaemia must be an important differential diagnosis in patients presenting with pancytopenia.^15,16 Bone marrow examination does not contribute to the diagnosis of the underlying aetiology and should be done when a diagnosis of myelodysplasia is being considered.
   Liver function tests showed a mild indirect hyperbilirubinaemia with normal enzymes in 32% of the patients tested. Estimation of serum methyl malonic acid and homocysteine, which are better indicators of cobalamin and folate deficiency at the tissue level,^17,19 were not done due to cost constraints.
   In conclusion, megaloblastic anaemia causes substantial morbidity in patients with anaemia. Data regarding the magnitude of the problem in different parts of India and the factors that might influence its incidence are lacking. Megaloblastic anaemia must be considered in the differential diagnosis of patients presenting with pyrexia of unknown origin, mild icterus or pancytopenia. Documentation of occult cobalamin deficiency in different ethnic and socioeconomic groups and in pregnant women needs to be done. The effect on neonates of cobalamin-deficient mothers should also be studied.
   A large volume of recent literature links serum levels of homocysteine and methyl malonic acid in cobalamin and folate deficiency to occlusive cardiovascular disease and neurological manifestations.^20-23 Complete evaluation for subtle neurological signs and cardiac function needs to be done in the at-risk population to assess the deficiency of these vitamins.
   Patients are being treated in the short term with haematinics and transfusions with relief of symptoms. In most instances long term follow up and diet counselling are not being done. The fortification of diet to prevent megaloblastosis needs to be taken up as a national public health issue.

St Stephen’s Hospital, Tis Hazari, Delhi 110054, India
Uma khanduri, archna sharma Department of Haematology
Correspondence to Uma Khanduri, prakash_uma@yahoo.com