**3.2 Sideroblastic anemia**

250 New Advances in the Basic and Clinical Gastroenterology

Fig. 1. Approximate relationship of anemias according to hemoglobin and red cell size

is determined to be likely, the cause of the deficiency must be determined, and it is important for the clinician to distinguish blood loss from other causes. This list of causes varies depending on the age of the patient, the severity of the anemia and the severity of microcytosis. The chapter will outline a more thorough differential for all causes, leading towards a proper diagnosis before initiating treatment. We will emphasize that iron therapy alone is not adequate therapy; as such treatment alone will leave the cause of the anemia

Since iron deficiency anemia is almost always due to chronic blood loss, the distinction between acute and chronic blood loss must be made. While both lead to anemia, the causes as well as the signs and symptoms are profoundly different between the two. As to the cause, acute blood loss can occur externally by any route or internally into practically any space, including intracranial, intrathoracic, retroperitoneal and abdominal. Isolated acute blood loss is normocytic, with a reticulocyte count that is not elevated until several days

(MCV)

undiagnosed and untreated.

**3.1 Acute blood loss** 

The sideroblastic anemias are a heterogeneous group of congenital and acquired bone marrow disorders defined by the presence of pathologic iron deposits within the mitochondria of erythroid precursors. The anemia may be mild, moderate or severe and may be normocytic, microcytic, or macrocytic and is typically characterized by a relative reticulocytopenia (Flemming, 2009). There are several known and incompletely characterized congenital causes, but all are rare including atransferrinemia, hypotranferrinemia (Shamsian et al., 2009), and others (Iolascon et al., 2009). Acquired causes include a variety of toxic exposures, drugs, and myelodysplastic syndromes, as well as copper deficiency. Many of these, although included in a full discussion of anemia and microcytic anemia, do not need to be considered in this chapter because they only rarely present as a severe anemia that is severely microcytic. Copper deficiency, for example is extremely rare, is usually normocytic, and the major manifestations are neurologic rather than hematologic.

#### **3.3 The anemia of "chronic disease" (Case 1)**

A 10 year old boy was referred to pediatric hematology clinic for anemia occurring on a previous blood test. Two months earlier he had been hospitalized briefly for a high fever and rash. While in the hospital, a CBC showed a mild microcytic anemia (hgb 10.0 g/dL (normal 11.9-15.4 g/dL), MCV 75 fL (normal 80-95 fL). He recovered well and was subsequently discharged. He had since been well, with no subsequent fevers, and no weight loss or night sweats. He had had no increased fatigue or exercise intolerance. His family history was negative for any hematologic disorders or malignancies. He had no chronic medical conditions, took no medications, and had no known drug allergies.

In clinic, he appeared healthy, cooperative and in no distress. His skin showed no rash, bruises, jaundice, or pallor. He had no lymphadenopathy, scleral icterus, or conjunctival pallor. His lungs were clear with no wheezes, and his heart sounds were normal with no

A Case Based Approach to Severe Microcytic Anemia in Children 253

A 23 month old girl was referred to Pediatric Hematology for investigation of anemia. Thalassemia was suspected due to a family history of "Mediterranean anemia." Her father and some of his family members were known to be a carrier for thalassemia. He had been treated with high dose iron, but said that it did not help and made him ill. His family was

The patient had generally been healthy, with normal growth and development, except for a brief febrile seizure a few months earlier. Her diet included fruits, vegetables, pasta, and lentils. She took no medications. On exam, she appeared healthy and alert. She was engaging and playful. She was slightly pale but showed no signs of icterus, petechiae, or ecchymoses. She had no adenopathy or splenomegaly. Her heart rate was regular, and her lungs were clear. Screening lab work was remarkable for a mild anemia that was severely microcytic with a hemoglobin of 11 g/dL (normal 10.5-14 g/dL), MCV 60.2 fL (normal 70-90 fL) a year earlier and more recently 9.6 g/dL and 54.9 fL. A hemoglobin electrophoresis was remarkable for an elevated hgb A2 of 5.5 percent. Molecular studies had been done and showed that she

This patient is a carrier of beta thalassemia, as is clear from the history as well as lab results. The clue is the relatively mild anemia with severe microcytosis. The blood smear in thalassemia trait typically has significant target cells present, and usually a hemoglobin electrophoresis shows an elevated hemoglobin A2 or F percentage. More quantitative ways to distinguish thalassemia trait from iron deficiency include the Mentzer index (Mentzer, 1973), as well as a variety of other methods of varying degrees of sensitivity and specificity (Demir et al. ,2002; Piomelli et al., 1976). The Mentzer index is calculated as the MCV divided by the RBC; a value over 13 is somewhat predictive of iron deficiency, while a value below 13 is more typical of beta thalassemia trait. Unfortunately in a practical sense, many cases fall near the line, especially those that are not clinically obvious one way or the other. The RDW is of limited use since it is increased in iron deficiency as well as a variety of

The presence of thalassemia trait does not exclude iron deficiency, which usually can be ruled out with a ferritin level and iron studies. Without iron deficiency, regular iron supplementation should not be given; not only will iron not improve the mild anemia, but over time could lead to iron overload, particularly because the mild anemia causes increased iron absorption form the gut. Thalassemia trait itself requires no therapy or follow-up, but genetic counseling should be considered because of the risk of having children with thalassemia major. To avoid unnecessary lab evaluations and therapy, the patient and family need to have a clear understanding of these points. Iron or folic acid deficiency, pregnancy or intercurrent illness may exacerbate the anemia in patients with thalassemia trait. The specific clinical phenotype depends on the precise genetic defect in heterozygous and especially homozygous individuals. In patients with microcytosis, hypochromia and erythrocytosis but without evidence of iron deficiency or elevated hgb A2 and Hgb F levels, one of the alpha thalassemia carrier states is likely. These too are benign conditions where

proper genetic and medical counseling of the patient and family is imperative.

originally from Sicily, where he lived for 8 years. She had no siblings.

was a carrier of two separate thalassemia mutations.

hemoglobinopathies including thalassemia.

**3.5 Thalassemia trait (Case 2)** 

**3.5.1 Discussion** 

murmur. His abdomen was soft and nontender with no organomegaly, although his liver was palpable and slightly tender to palpation, but his spleen was not palpable. His extremities were well perfused with no edema. Two months after his original CBC his hemoglobin had risen to 11 g/dL (normal 12-15 g/dL), and MCV was 72.9 fL (normal 74- 82). His ferritin was 38 ng/mL (normal 10-200 ng/mL). Iron studies were normal as was his hemoglobin electrophoresis and blood smear.

### **3.3.1 Discussion**

Typically the anemia of chronic disease, also called the anemia of inflammation, is a mild normocytic or mildly microcytic anemia. Usually it is associated with chronic conditions such as IBD, and the anemia resolves with effective treatment of the underlying condition. As this case illustrates however, the anemia of "chronic" disease can be associated with inflammatory conditions of shorter duration such as chronic strep infections, recurrent otitis media, or viral infections. The RDW is often slightly elevated as a result of slight anisocytosis, reflecting ineffective iron recycling. The anemia itself is mild and asymptomatic and will not respond to iron unless there is concomitant iron deficiency. Since ferritin is an acute phase reactant, an elevated level does not imply the patient has adequate iron stores, but a low level indicates inadequate stores. Again, this type of anemia is included in a full discussion of microcytic anemia in children, but will not be discussed further in this chapter because the anemia by itself does not have the features that would or should bring it to the attention of a gastroenterologist, i.e. by the severity of the anemia or the microcytosis, although the underlying inflammatory condition itself may involve the GI tract. Distinguishing iron deficiency from the anemia of inflammation is usually not difficult from the history and CBC, but serum transferrin receptor (sTfR) levels, if available, have been shown to help distinguish them (Oliveras et al., 2000; Skikne, 1998, Vazquez et al., 2006).

### **3.4 Lead poisoning**

The association between lead poisoning and microcytic anemia has long been assumed to be a direct inhibition of lead on heme synthesis, leading to a sideroblastic effect. This mechanism is supported by the basophilic stippling present and the markedly elevated free erythrocyte protoporphyrin values in these patients. More likely, however, the lead poisoning is a secondary effect of severe iron deficiency. The mechanism is twofold, in that severe iron deficiency increases the incidence of pica (Buchanan, 1999, 2003; Eden, 1999), and increases the absorption of lead from the gut. The incidence of Pica may be higher than generally assumed (Corbett et al., 2003). The precise pathophysiology of Pica is unknown, but is probably related to CNS iron deficiency. Patients often consume laundry starch, ice, soil, or clay. Both clay and starch can bind iron in the gastrointestinal tract and exacerbate iron deficiency (Crosby, 1982; Gonzalez et al., 1982; Thomas et al., 1976). At any concentration of lead, inhibition of ferrochelatase is most marked when the iron concentration is lower (Piomelli et al., 1987), but the neurologic effects of lead may be severe in the absence of anemia in children. Thus, it is important to address both the lead poisoning and the anemia and to find and treat the cause of the bleeding rather than solely addressing the lead poisoning. There is evidence that replacing iron in patients with concomitant iron deficiency and lead poisoning helps the body eliminate lead (McGeehan, 2003; Rondó et al., 2006; Wright et al., 2003).

#### **3.5 Thalassemia trait (Case 2)**

252 New Advances in the Basic and Clinical Gastroenterology

murmur. His abdomen was soft and nontender with no organomegaly, although his liver was palpable and slightly tender to palpation, but his spleen was not palpable. His extremities were well perfused with no edema. Two months after his original CBC his hemoglobin had risen to 11 g/dL (normal 12-15 g/dL), and MCV was 72.9 fL (normal 74- 82). His ferritin was 38 ng/mL (normal 10-200 ng/mL). Iron studies were normal as was his

Typically the anemia of chronic disease, also called the anemia of inflammation, is a mild normocytic or mildly microcytic anemia. Usually it is associated with chronic conditions such as IBD, and the anemia resolves with effective treatment of the underlying condition. As this case illustrates however, the anemia of "chronic" disease can be associated with inflammatory conditions of shorter duration such as chronic strep infections, recurrent otitis media, or viral infections. The RDW is often slightly elevated as a result of slight anisocytosis, reflecting ineffective iron recycling. The anemia itself is mild and asymptomatic and will not respond to iron unless there is concomitant iron deficiency. Since ferritin is an acute phase reactant, an elevated level does not imply the patient has adequate iron stores, but a low level indicates inadequate stores. Again, this type of anemia is included in a full discussion of microcytic anemia in children, but will not be discussed further in this chapter because the anemia by itself does not have the features that would or should bring it to the attention of a gastroenterologist, i.e. by the severity of the anemia or the microcytosis, although the underlying inflammatory condition itself may involve the GI tract. Distinguishing iron deficiency from the anemia of inflammation is usually not difficult from the history and CBC, but serum transferrin receptor (sTfR) levels, if available, have been shown to help distinguish

The association between lead poisoning and microcytic anemia has long been assumed to be a direct inhibition of lead on heme synthesis, leading to a sideroblastic effect. This mechanism is supported by the basophilic stippling present and the markedly elevated free erythrocyte protoporphyrin values in these patients. More likely, however, the lead poisoning is a secondary effect of severe iron deficiency. The mechanism is twofold, in that severe iron deficiency increases the incidence of pica (Buchanan, 1999, 2003; Eden, 1999), and increases the absorption of lead from the gut. The incidence of Pica may be higher than generally assumed (Corbett et al., 2003). The precise pathophysiology of Pica is unknown, but is probably related to CNS iron deficiency. Patients often consume laundry starch, ice, soil, or clay. Both clay and starch can bind iron in the gastrointestinal tract and exacerbate iron deficiency (Crosby, 1982; Gonzalez et al., 1982; Thomas et al., 1976). At any concentration of lead, inhibition of ferrochelatase is most marked when the iron concentration is lower (Piomelli et al., 1987), but the neurologic effects of lead may be severe in the absence of anemia in children. Thus, it is important to address both the lead poisoning and the anemia and to find and treat the cause of the bleeding rather than solely addressing the lead poisoning. There is evidence that replacing iron in patients with concomitant iron deficiency and lead poisoning helps the body eliminate lead (McGeehan, 2003; Rondó et al.,

hemoglobin electrophoresis and blood smear.

them (Oliveras et al., 2000; Skikne, 1998, Vazquez et al., 2006).

**3.3.1 Discussion** 

**3.4 Lead poisoning** 

2006; Wright et al., 2003).

A 23 month old girl was referred to Pediatric Hematology for investigation of anemia. Thalassemia was suspected due to a family history of "Mediterranean anemia." Her father and some of his family members were known to be a carrier for thalassemia. He had been treated with high dose iron, but said that it did not help and made him ill. His family was originally from Sicily, where he lived for 8 years. She had no siblings.

The patient had generally been healthy, with normal growth and development, except for a brief febrile seizure a few months earlier. Her diet included fruits, vegetables, pasta, and lentils. She took no medications. On exam, she appeared healthy and alert. She was engaging and playful. She was slightly pale but showed no signs of icterus, petechiae, or ecchymoses. She had no adenopathy or splenomegaly. Her heart rate was regular, and her lungs were clear.

Screening lab work was remarkable for a mild anemia that was severely microcytic with a hemoglobin of 11 g/dL (normal 10.5-14 g/dL), MCV 60.2 fL (normal 70-90 fL) a year earlier and more recently 9.6 g/dL and 54.9 fL. A hemoglobin electrophoresis was remarkable for an elevated hgb A2 of 5.5 percent. Molecular studies had been done and showed that she was a carrier of two separate thalassemia mutations.

#### **3.5.1 Discussion**

This patient is a carrier of beta thalassemia, as is clear from the history as well as lab results. The clue is the relatively mild anemia with severe microcytosis. The blood smear in thalassemia trait typically has significant target cells present, and usually a hemoglobin electrophoresis shows an elevated hemoglobin A2 or F percentage. More quantitative ways to distinguish thalassemia trait from iron deficiency include the Mentzer index (Mentzer, 1973), as well as a variety of other methods of varying degrees of sensitivity and specificity (Demir et al. ,2002; Piomelli et al., 1976). The Mentzer index is calculated as the MCV divided by the RBC; a value over 13 is somewhat predictive of iron deficiency, while a value below 13 is more typical of beta thalassemia trait. Unfortunately in a practical sense, many cases fall near the line, especially those that are not clinically obvious one way or the other. The RDW is of limited use since it is increased in iron deficiency as well as a variety of hemoglobinopathies including thalassemia.

The presence of thalassemia trait does not exclude iron deficiency, which usually can be ruled out with a ferritin level and iron studies. Without iron deficiency, regular iron supplementation should not be given; not only will iron not improve the mild anemia, but over time could lead to iron overload, particularly because the mild anemia causes increased iron absorption form the gut. Thalassemia trait itself requires no therapy or follow-up, but genetic counseling should be considered because of the risk of having children with thalassemia major. To avoid unnecessary lab evaluations and therapy, the patient and family need to have a clear understanding of these points. Iron or folic acid deficiency, pregnancy or intercurrent illness may exacerbate the anemia in patients with thalassemia trait. The specific clinical phenotype depends on the precise genetic defect in heterozygous and especially homozygous individuals. In patients with microcytosis, hypochromia and erythrocytosis but without evidence of iron deficiency or elevated hgb A2 and Hgb F levels, one of the alpha thalassemia carrier states is likely. These too are benign conditions where proper genetic and medical counseling of the patient and family is imperative.

A Case Based Approach to Severe Microcytic Anemia in Children 255

A 63 year old male presented to his physician with complaints of fatigue and shortness of breath, progressive over several weeks. The physician noted that the patient looked pale and ordered a complete blood count. The CBC revealed a hemoglobin of 5.9 g/dL and an MCV of 59 fL. The physician correctly interpreted this as a severe microcytic anemia, and that this anemia was of relatively recent onset. The physician considered transfusions and a prescription for oral iron as therapy for the anemia. However, upon considering the etiology of the anemia with the above history, he knows that the microcytosis is not consistent with a lesion leading to acute blood loss. Chronic blood loss is the only possibility, and that blood loss MUST be gastrointestinal. He referred the man to a gastroenterologist for a thorough

For this patent, while several more benign lesions leading to chronic blood loss can be responsible, the lesion must be found, as colon cancer must be considered highly likely until proven otherwise. The sequence of events is therefore clear: The patient has developed a lesion in his GI tract that has been losing blood slowly over at least several months. Since the blood loss is relatively slow, the patient's bone marrow easily makes up for the anemia at first, but eventually stores of iron are depleted from his liver and other sites. At that point his anemia becomes more severe and begins to be microcytic, until the anemia itself becomes severe enough to cause the complaints that lead him to seek help from his physician. In retrospect the patient may or may not have noticed the typically black or tarry stools, depending on the rate of bleeding, and in fact the stool may not even consistently test positive for blood since the bleeding is often intermittent. In fact, testing the stool for blood in such a case is not necessary, as there can be no other cause of this type of anemia in this man but gastrointestinal bleeding. A negative stool test is in no way reassuring; this man must be referred to find the specific lesion. Transfusion and iron therapy do not treat the bleeding. It should even be stated that in such a case of anemia the hematologist plays no role. At any age the more severe the anemia and the microcytosis, the more likely that blood

The following pediatric cases will use the same concepts illustrated in the above case to emphasize the role of the gastroenterologic system in the etiology and therapy of severe

A 14 month old boy presented to his pediatrician for a one year well child checkup. At the visit however his parents explained to the pediatrician that he had been pale for about the last month or so, not feeding well, and had some pica like behaviors. He had recently been more fussy and was sleeping poorly. They denied any jaundice, change in bowel or urinary habits or breathing difficulties. They denied bloody or black stools. They reported poor feeding when offered a variety of foods, and until very recently his diet relied heavily on milk, and he would typically go through a gallon every other day. The family lived in an older apartment with chipping paint that the parents admitted the child sometimes peeled from the walls. His growth and development had otherwise been normal. He took no

loss rather than a nutritional cause is the responsible mechanism.

**4.1 Severe microcytic anemia in an adult (Case 3)** 

investigation.

**4.1.1 Discussion** 

microcytic anemias.

**4.2 Milk enteropathy (Case 4)** 
