Iron deficiency anaemia is the most prevalent type of anaemia worldwide, affecting over one billion people. As also one of the most treatable, it is essential for all clinicians to be able to recognise this condition. Due to its prevalence and its occurrence secondary to other medical conditions, it may also co-exist in the same patient with other causes of anaemia and this situation requires a high index of suspicion so that it can be detected and rectified. Furthermore, recognising iron deficiency is only a partial diagnosis; in most cases it is necessary to take steps to recognise the underlying cause, or at least exclude major treatable disorders, usually of the gastrointestinal (GI) tract.
There are two main aspects to dealing with a patient with suspected iron deficiency anaemia: Diagnosis and treatment. Both are commenced in primary care. Diagnosis includes recognising iron deficiency anaemia and distinguishing it from other forms of anaemia, and also involves identifying the underlying cause.
<h3 class=”subheadMIstyles”>Causes</h3>
Inadequate diet makes a major contribution to iron deficiency anaemia, although in adults in developed countries it is rarely the sole factor; most patients will have some cause of increased blood loss or, in a much smaller number, malabsorption. This review focuses mainly on adults, but iron deficiency anaemia is common in infants, especially those fed on cow’s milk rather than breast milk or fortified formula, and common also in toddlers.
In adults, menstrual blood loss is the most common predisposing factor and explains the much higher prevalence of iron deficiency anaemia in women compared to men.
Pregnancy also imposes a large negative iron balance on the woman. Menstrual periods do not need to be subjectively heavy to have a negative effect on iron status and indeed subjective estimation by women of menstrual blood loss has been shown not to correlate well with experimentally measured blood loss.
GI blood loss, usually occult, ie, not visible to the patient, is the main cause of iron deficiency in male adults and in postmenopausal women and detection of the cause is the key step in diagnosis to detect potentially treatable gastrointestinal lesions such as colon cancer.
Blood losses from other sites, eg, epistaxis, haemoptysis, haematuria, or even more rarely, haemoglobinuria, is only rarely a major contributor to iron deficiency anaemia unless the losses are extreme in severity and clinically obvious.
<h3 class=”subheadMIstyles”>Approach to patient</h3>
The approach to the patient with suspected iron deficiency anaemia, and indeed the information needed to be provided on any onward referral, follows automatically from the epidemiology of the disease. History should focus on symptoms of anaemia, which are the primary reason for instituting therapy. Poor exercise tolerance, weariness, and shortness of breath on exertion are important symptoms. The rate of development of anaemia influences the presence and severity of symptoms. For example, severe anaemia (haemoglobin <7g/dL) in the absence of symptoms usually indicates a gradual onset of the anaemia. In chronic anaemia the reduction in red cell volume will be compensated for by an increase of plasma volume. An unnecessary transfusion can easily provoke fluid overload in a patient with severe chronic anaemia. In general, blood transfusions are not indicated for iron deficiency anaemia, unless there are symptoms or organ failure, which cannot be tolerated until a response to iron occurs. If transfusion is used in iron-deficient patients without clinically evident bleeding, reassessment after a single unit of red blood cells should be done and this will often be sufficient to relieve symptoms, with iron treatment doing the rest. A patient with a severely low haematocrit may have a circulating red cell volume of less than 1L (blood volume in adult 70ml/kg: multiply by Hct to get red cell volume) and one or two units of red blood cells will represent a very significant increase in oxygen carrying capacity.
Key aspects of the history relate to the potential causes of iron deficiency. If a woman, is the patient menstruating; how frequent and heavy are her periods reported to be? Has there been recent pregnancy? If postmenopausal, when was the last menstrual period? Are there gastrointestinal symptoms especially change in bowel habit or any evidence of blood loss (haematemesis, melaena, blood per rectum)? Chronic upper GI symptoms, eg, heartburn may also be relevant though in the era of proton-pump inhibitor therapy and triple therapy for <em>Helicobacter pylori</em> infection, anaemia due to peptic ulceration in young patients is much less common than previously. Non-steroidal anti-inflammatory drug (NSAID)-induced upper GI bleeding is important to consider. Frequent blood donation can frequently lead to reduced iron stores but should not lead to overt anaemia.
Drug history in relation to current or previous iron therapy and the response thereto is essential information. In general, there is no need to delay therapy with iron if symptomatic anaemia is present and iron deficiency is evident. The induction of constipation prior to planned colonoscopy is only a major consideration if the colonoscopy is planned very soon.
Family history is particularly important in relation to first-degree family history of colorectal cancer, especially if the proband(s) were young at presentation. In a patient borderline on other criteria (eg, a premenopausal woman in her 40s) this may well make the difference in terms of requesting invasive GI investigations. Family history of recurrent nosebleeds and severe iron deficiency may reveal a diagnosis of hereditary haemorrhagic telangiectasia (HHT). A very rare syndrome of iron resistant iron deficiency anaemia also exists.
<blockquote> <div>
Inadequate diet makes a major contribution to iron deficiency anaemia, although in adults in developed countries it is rarely the sole factor; most patients will have some cause of increased blood loss or, in a much smaller number, malabsorption
</div> </blockquote> <h3 class=”subheadMIstyles”>Investigations</h3>
Full blood count is the starting point. Iron deficiency anaemia is typically microcytic, but can be normocytic, especially if mild, of recent onset, or as part of mixed pathology.
Where a microcytic hypochromic picture is present, there are essentially only four possible causes (or a combination): Iron deficiency anaemia; anaemia of chronic disease (anaemia of defective iron utilisation), though most cases are normocytic; thalassaemia or one of its variants; or the rare sideroblastic anaemias, where in fact usually only the inherited forms are microcytic – acquired sideroblastic anaemia as a type of myelodysplastic syndrome being often macrocytic.
A previously normal MCV/MCH (after the first few months of life) exclude thalassaemia. Thalassaemia and thalassaemic haemoglobinopathies do occur but are rare in those of Western European genetic heritage but more frequently seen nowadays in Ireland due to immigration, marriage, and adoption: One cannot rely on surname to indicate ethnicity – from Africa, the Middle East, and Asia. There are also differences in the red cell indices between thalassaemia minor and iron deficiency anaemia, with various discriminant factors described in the literature (a raised red cell count would be unusual in iron deficiency – unless due to polycythaemia vera – and could indicate thalassaemia minor as an alternative diagnosis); a call to your local haematologist is an alternative means of discrimination. Due to the frequency of iron deficiency, it can frequently co-exist with other conditions, including thalassaemia minor. In addition, the straightforward test for beta thalassaemia minor (raised haemoglobin A2) can be falsely negative where there is concomitant iron deficiency, so in these mild microcytic anaemias the correct sequence is first to exclude iron deficiency with a serum ferritin and if low, treat with iron until replete (ie, serum ferritin well within normal range). If the MCV and MCH normalise then thalassaemia minor is not a consideration.
Excluding iron deficiency in patients with inflammatory disorders is difficult – see discussion below re ferritin levels in these patients – and in some cases a therapeutic trial of iron therapy is needed. Sideroblastic anaemia is rare and usually associated with increased iron stores (elevated serum ferritin).
Where a classic microcytic picture is present in the right setting, it is not completely necessary to confirm the diagnosis with a serum ferritin, although one may wish to do this to avoid doubt. A serum ferritin is the primary test to diagnose low iron stores, although it does have some limitations. These limitations almost exclusively relate to ‘falsely’ (ie, not reflecting the true level of iron stores) high serum ferritins in people with concomitant conditions, primarily inflammation. A low serum ferritin never lies – the iron stores are low with virtually no exceptions. However, a normal serum ferritin can mask low iron stores in patients with inflammation. In young patients without comorbidities, this is rarely an issue, but in patients with multiple conditions, a serum ferritin below 100ug/L or even 200ug/L must not be taken as excluding iron deficiency. In patients with severe chronic kidney disease, iron deficiency is presumed to be present if transferrin saturation is less than 20 per cent and ferritin is less than 100ug/L (pre dialysis patients) or 200ug/L (haemodialysis patients). Iron treatment in this setting reduces the required erythropoietin dose. Functional iron deficiency may occur in severe CKD patients with higher ferritin levels but transferrin saturation <20 per cent.
A low serum ferritin is diagnostic of low iron stores and the patient can be treated with iron; further tests of iron status should not be done. It is worth pointing out that, due to these being 95 per cent ranges, the reference range for ferritin often extends down into a range where iron stores are quite low, with some laboratories reporting a serum ferritin as low as 11ug/L as being in the normal range. In my own practice, I never record a serum ferritin as ‘normal’ unless it is well over 100ug/L, even higher in patients with known inflammation or chronic renal disease and I always record the exact figure and interpret in the context of the full blood count and history. A ferritin of less than 30ug/L can be taken as indicative of iron deficiency anaemia in the right setting, especially a microcytic hypochromic pattern and a compatible history. A significant number of referring doctors assume that the diagnosis is not iron deficiency as the ferritin is ‘in the normal range’. This is a false dichotomy. The lower the ferritin is, in the lower part of the normal range, the more likely iron deficiency is present and the ferritin result must be interpreted in the context that the true iron status could be lower than the ferritin result indicates, but is not going to be higher. Response to therapy with rising haemoglobin after iron therapy (I usually wait at least two weeks and more typically four to pick up slower responders) confirms the diagnosis.
<h3 class=”subheadMIstyles”>Detection of underlying causes</h3>
Diagnosis is incomplete without assessment of the underlying cause, although sometimes this can be elusive. The key priority is to avoid missing treatable GI lesions. It is important not to forget coeliac disease; patients of all ages who have iron deficiency without a clear explanation should have a test for anti-tissue transglutaminase (anti-tTG) antibodies. The test is an IgA antibody and will be falsely negative in patients with selective IgA deficiency, which is rare but frequently asymptomatic and is more frequently found in patients with coeliac disease than in the normal population. We therefore do an IgA level along with anti-tTG if a non-deficient IgA level is not already documented for the patient, but there are conflicting views in the literature on this practice as it may not be cost-effective due to the rarity of selective IgA deficiency. If IgA deficiency is present, an IgG anti-endomysial antibody test can be done. Anti-gliadin antibodies are less specific and not recommended. The sensitivity of the serological tests is high for coeliac disease and unless there is a high suspicion, eg, as a result of symptoms, family history of coeliac disease, or other clinical features such as folate or fat soluble vitamin deficiencies, invasive investigations are not needed, although it is good practice to do a duodenal biopsy if an oesophagogastroduodenoscopy is done in a patient with iron deficiency.
<blockquote> <div>
Patients with GI symptoms and males or postmenopausal females should be referred for OGD and colonoscopy
</div> </blockquote>
Patients with GI symptoms and males or postmenopausal females should be referred for OGD and colonoscopy. Women over the age of 40 years who are still menstruating need careful consideration: If menorrhagia is clearly the cause and there are no gastrointestinal symptoms or family history of colorectal carcinoma, invasive investigations can be foregone. In the woman over 50 years without clear menorrhagia it may be advisable to refer for GI investigations and a first-degree family history of colorectal carcinoma would prompt me to refer at a much younger age. Where GI investigation is clearly indicated, eg, males and postmenopausal females, I would not use a faecal occult blood (FOB) test, as a negative result may just indicate intermittent blood loss and would not have a sufficient negative predictive value when iron deficiency is present. An FOB test could, if strongly positive, indicate a more major degree of blood loss, requiring urgent referral. A positive test in a borderline case for referral would make the case for invasive investigations.
<h3 class=”subheadMIstyles”>Treatment</h3>
Treatment for iron deficiency anaemia should be started in general practice with oral therapy. Though the patient may need referral to secondary care, especially for GI investigations, treatment need not be delayed. However, if there is a very early date for colonoscopy it is reasonable to wait until after the procedure. Use of a rapid release iron salt with adequate dosage is important. There are some products on the market that are little more than spring water with a moderate iron content. These are not adequate for treatment of iron deficiency. They may well contain the recommended daily allowance of iron for maintenance of normal health, but most of us do not need such supplementation. Of course such products will be better tolerated in many cases, due to the lower iron content. Ferrous sulphate, fumarate or gluconate are all suitable. It should be noted that the usual iron content in the usual size of tablet or capsule varies between these salts; for example, the full dose of ferrous gluconate amounts to six tablets daily and therefore one a day of this salt is unlikely to be an adequate dose. There are some recent studies showing a law of diminishing returns: Increasing oral iron dosage results in a reduced fractional absorption, but it is important to note that the absolute amount of iron absorbed may still increase with higher dosage. Intolerance of oral iron can be managed by reducing the dose; taking with food or changing the product. Delayed release products are not generally recommended as first choice, but may still be effective and better tolerated if the patient has GI intolerance of the rapid release iron salts. The duration of iron treatment is often underestimated. The general rule of thumb is to treat until the haemoglobin has normalised and for three months more to build up iron stores. If the patient has ongoing severe blood losses that have not been corrected, eg, angiodysplasia of the colon, or menorrhagia, the treatment duration should be extended and guided by repeat serum ferritin measurement.
Failure to respond or intolerance can be an indication for intravenous (IV) iron. We do not recommend intramuscular iron; it requires multiple doses and local pain or discolouration at injection sites is common. We generally use IV ferric carboxymaltose. A dose of 1,000mg is the usual and maximum single dose in an adult. If needed, a second dose can be given one week later. The SmPC of the product gives simplified dosing instructions, or the time honoured Ganzoni formula can be used. I usually use the latter as it is available in a number of free apps, including Medcalc, but I will generally round the dose off to 1,000mg as a single dose or 1,000mg followed a week later by 500mg. The calculation includes 500mg extra to build up iron stores, approximately equivalent to the three additional months of oral iron therapy referred to above. Serum ferritin will immediately be much higher after IV iron, frequently above the reference range and a recheck soon after the infusion is not needed and does not indicate that the dose was appropriate or excessive. After effective iron therapy, haemoglobin should rise by approximately 1g/dL in a week or two.
In severe anaemias it will take several weeks to incorporate all the iron into new red blood cells. All parenteral iron preparations can rarely be associated with severe anaphylactoid or more rarely true anaphylactic reactions. We would be reluctant to use IV iron in patients with a history of severe drug reactions or allergies or severe asthma. Resorting to IV iron in patients with no or very mild anaemia, even if they have very low iron stores, should also be avoided.
It is important also to attempt to correct underlying causes if feasible; otherwise the anaemia is likely to recur.
Dietary advice may be helpful. Animal products contain haem iron, which is better absorbed than iron from plants; the <em>Popeye</em> story regarding the power of spinach is an enduring myth, that said, green vegetables are very desirable in the diet for other reasons. Vitamin C enhances iron absorption and fruit or juice with meals may therefore help absorption, but high-dose supplements are not indicated. Tea reduces iron absorption and I advise patients with iron deficiency to avoid tea close to meals.
Menorrhagia will often improve with measures such as insertion of a progestogen-releasing intrauterine system, oral hormonal therapy or gynaecological procedures such as curettage, endometrial ablation or uterine myomectomy. The efficacy of the intrauterine system in particular has reduced greatly the need for hysterectomy to treat menorrhagia.
GI lesions should be treated medically or surgically as appropriate and coeliac disease treated with gluten free diet. In a few patients there is ongoing GI blood loss that cannot be corrected: Patients with gastric antral vascular ectasia (watermelon stomach), patients with angiodysplasia of the colon or patients with hereditary haemorrhagic telangiectasia are examples, as are some patients with inoperable tumours. These patients can be supported by regular IV iron or blood transfusion if indicated and should be regularly reviewed at a hospital department, which can deliver this support, rather than left to repeatedly attend in severely anaemic condition to the emergency department. Repeated endoscopic treatment with Argon plasma coagulation has an important role, especially in patients with GAVE and HHT, and also in some colonic lesions.
<h3 class=”subheadMIstyles”>Conclusion</h3>
The management of iron deficiency may require multidisciplinary management. Straightforward cases should be managed in primary care, both in terms of diagnosis and treatment, with referral for OGD and colonoscopy as appropriate.
Haematology advice is mainly required for those patients where the type of anaemia is not clear, however, a therapeutic trial of iron will often clear this up. Referral for consideration of IV iron is appropriate in anaemic patients who cannot tolerate or do not respond to appropriate doses of oral iron.
<div style=”background: #e8edf0; padding: 10px 15px; margin-bottom: 15px;”>
Dr Gerard Crotty, Consultant Haematologist, and Dr Ilah Nik Mahmood, Haemotology Registrar, Midland Regional Hospital, Tullamore, Co Offaly.
</div>
<h5 class=”referencesonrequestMIstyles”>References on request</h5>
