Iron deficiency anaemia (IDA) is a common and complex problem in chronic inflammatory rheumatic diseases. The predominant cause of IDA is gastrointestinal blood loss, often due to drug treatment. However, asymptomatic colonic and gastric carcinoma may present with IDA and exclusion of these conditions is of prime concern.

The British Society of Gastroenterology has recently revised the guidelines for the diagnosis and management of IDA in the general population.¹ These guidelines use a combination of low haemoglobin, low mean cell volume (MCV), and low serum ferritin to diagnose IDA.

Diagnosing IDA in the presence of chronic inflammation as seen in rheumatoid arthritis, poses considerable difficulty because serum ferritin is an acute phase reactant and rises in the presence of inflammation. A further complicating factor in our population is that MCV tends to be spuriously raised as a result of disease modifying antirheumatic drug (DMARD) treatment—in particular, sulfasalazine, methotrexate, and azathioprine.

It has been suggested by Jolobe² and Broin et al³ that low mean cell haemoglobin (MCH) correlates better with low ferritin levels and hence is better than low MCV as an aid to identifying patients with IDA, though this is still not widely used in routine clinical practice.

This study aimed at investigating whether MCV or MCH in combination with serum ferritin could be used effectively in screening for IDA in our patients.

METHODS AND RESULTS

We undertook a retrospective study of our patients with chronic inflammatory rheumatic diseases who were receiving regular blood monitoring for second line treatment. Those who were anaemic (haemoglobin <115 g/l in women and <130 g/l in men, which are the lower limits of normal range of haemoglobin concentration for the local laboratory¹) on at least two occasions at least 1 month apart, were identified from “drug monitoring records”. Further, those who were “iron deficient” were identified from APEX (computerised laboratory results system) based on low serum ferritin (<20 μg/l), low MCV (<80 fl), and low MCH (<27 pg).

A total of 1231 records were examined (patient characteristics are outlined in table 1). Three hundred and four (24.7%) were found to be anaemic during a 12 month period. Sixty eight of these (22.4%) were identified as “definite IDA” (serum ferritin levels <20 μg/l, identified as the lower limit of normal range for the local laboratory). Of these 68, 44 (65%) had low MCV, but 56 (82%) had low MCH levels (p = 0.016, Fisher’s exact test). In 36 patients with “probable IDA” (serum ferritin levels 20–100 μg/l), MCV was low in 14 (39%), but MCH was low in 26 (72%) (p = 0.004).

DISCUSSION

IDA is common and often difficult to identify accurately in patients with chronic inflammatory rheumatic diseases. Bone marrow aspiration remains the preferred test for its diagnosis, but has the disadvantage of being invasive. Thus, we are limited to using serological tests of iron stores, the best validated of which is serum ferritin, which is the most powerful test of iron deficiency.⁴

It has been proposed by Goddard et al that a serum ferritin concentration of >100 μg/l excludes IDA in the presence of concurrent inflammation, malignancy, or hepatic disease.¹ Further, though serum transferrin receptor assay can help to distinguish between the anaemia of chronic disease and iron deficiency, it is no better than serum ferritin.⁵ However, there are no widely accepted guidelines for the diagnosis and management of IDA in our patient population, which represent a unique subset of patients.

Our study shows that a higher proportion of patients with both “definite IDA” and “probable IDA” had a low MCH compared with a low MCV. Low MCH correlated better with iron deficiency than low MCV.

We suggest that in chronic inflammatory arthropathies, if the haemoglobin is low, then MCH is a better marker of iron deficiency than MCV. We therefore propose that MCH in conjunction with serum ferritin is a better predictor of IDA in patients with chronic inflammatory rheumatic diseases.