A study on the effects of pica and iron-deficiency anemia on oxidative stress,antioxidant capacity and trace elements

Abstract

Pica is defined as developmentally inappropriate consumption of nonnutritive substances for at least 1 month. There are a few studies on serum trace element levels of patients with pica. The literature contains contracting data on the levels of oxidative stress and antioxidant levels in patients with iron-deficiency anemia (IDA). The effect of pica on oxidative stress and antioxidant capacity has not been investigated yet. The present study evaluated the effects of pica and IDA on oxidative stress and antioxidant capacity as well as on the levels of trace elements including serum zinc and selenium in 47 children with IDA plus pica, 22 children with IDA only and 21 nonanemic children as controls. The results demonstrated significantly lower levels of serum selenium and zinc in pica and IDA groups compared to the control group. Total oxidant levels were highest in the pica group and consistently, the lowest total antioxidant capacity was observed again in the pica group. Comparison of pica and IDA groups yielded significantly lower levels of total antioxidant levels and significantly higher oxidative stress index in the pica group. Consequently, it is thought that the detrimental effects of pica within the organism were mediated by adverse impacts on antioxidant capacity and oxidative stress. These effects should be kept in mind while managing patients with pica.

Keywords

Pica anemia selenium zinc antioxidant capacity

Introduction

Iron-deficiency anemia (IDA) and pica are important health concerns worldwide. Pica is defined as consumption of nonnutritive substances inappropriate for development and cultural practices.¹ Besides excessive salt intake, there are other inappropriate objects including ice, nutshell, starch, coffee beans, soil, hair, feces, flaking paint, lime, tomato seed, paper and cigarette butts that have been described as pica objects. The prevalence of pica is not exactly known but is comparable between the two genders. The highest prevalence is in children aged 1–6 years. Most of these children are from families with lower socioeconomic levels and often have behavioral disorders.² Pica may be seen in any geographical region, and earth eating (geophagia) is the most common form. In Turkey, geophagia is a long-known concern. Cavdar and Arcasoy reported geophagia as the most common form of pica in Turkey and noted that individuals with the disorder existed in 70% of the country.³ Pica is most prevalent in rural areas of the Central Anatolian region of Turkey. Clinical presentation of pica is changeable and is related to the specific nature of the comorbiding conditions and the type of the pica object. Patients may present with gastrointestinal complaints such as constipation or abdominal pain.⁴ There is no consensus on the etiology of pica in most cases. The most emphasized hypotheses include nutritional, physiological, cultural and pharmacological backgrounds. There is no definite treatment for pica and the patients require a multidisciplinary approach. Substitution of some essential nutrients, aversion therapy, emetics, behavioral changes, formation therapy and correctional therapy have been shown to result in the decline of pica habits even in the presence of mental retardation. Recently, dopaminergic agents have been demonstrated to be potentially beneficial in refractory cases not responding to behavior therapy.⁵

Although pica is usually considered as a harmless habit, it is an eating disorder and may cause highly serious medical problems.

Iron, a trace element that is necessary for normal development and the biological functions of human, is the second most abundant metal of the earth’s crust. Iron is widely used in the human organism; its deficiency affects all systems, while systemic symptoms and clinical manifestations become evident at further stages. Nevertheless, iron deficiency at varying degrees is observed in different societies. Iron deficiency particularly impacts on hematological parameters as well as other elements including zinc, copper and selenium that have not only multitude of biological functions but also roles in some stages of iron metabolism. Thus, variations in the trace elements of zinc, copper and selenium have also been reported in children with iron deficiency. All these findings emphasize the critical role of iron, zinc, copper and selenium for sustained health and life.

Free radicals are atoms or molecules that contain one or multiple unpaired molecules.⁶ The most critical free radicals in the biological system are those formed from oxygen. Free radicals are also released during catalytic cycles of several enzymes. Intracellular free radicals production may be significantly enhanced by some foreign toxic agents. Free radicals affect all critical compounds of the cells including lipid, protein, DNA, carbohydrate and enzymes.^7,8 All major classes of biomolecules are affected by free radicals, although the lipids are the most susceptible ones. Oxygen radicals induce lipid peroxidation by affecting polyunsaturated fatty acids. Membrane damage resulting from lipid peroxidation is irreversible.⁹ Several defense mechanisms called “antioxidant defense system” are developed by the body to prevent the formation of oxygen species and the damage associated with them. The defense systems constitute free radical scavengers and some enzymes, and the effectiveness of the defense system is primarily dependent on the enzymatic system.¹⁰ Superoxide dismutase (SOD), glutathione peroxidase (GSH-Px) and catalase are the main enzymes involved in the mechanisms of defense against the damage induced by oxygen radicals. Selenium is needed for GSH-Px in erythrocytes for it to be biologically active. Prolonged selenium deficiency reduces GSH-Px activity in all tissues of the body. It has been reported to be a key element of the GSH-Px enzyme with intake of sufficient levels being necessary in infancy for healthy growth and development due to its further antioxidant functions.¹¹

The literature contains contradicting data on the levels of oxidative stress and antioxidant levels in patients with IDA. Effects of pica on oxidative stress and antioxidant capacity have not been investigated yet. However, there are a few studies on serum trace elements in patients with pica. The objective of the present study was to investigate the oxidative stress and antioxidant levels, to evaluate the association of trace elements including serum zinc and selenium with pica and IDA and to determine the effect of pica on duodenal mucosa in children with IDA plus pica, with IDA only and in nonanemic children as the control group.

Materials and methods

A total of 47 children with IDA plus pica and 22 children with IDA without history of pica, who were admitted in the Pediatrics Clinic of the Medical Faculty of Yuzuncu Yil University, were enrolled in the study. The subjects were aged from 1 to 15 years. A total of 21 healthy children of the same age group were included as controls. Consumption of substances that are not regarded as food was considered as pica. The diagnosis of IDA was established in the presence of transferrin saturation below 15% and ferritin levels below 14 ng/l in subjects with anemia with no chronic comorbidity and no chronic infection for the past 2 weeks.¹² It was ensured that peripheral smear results were consistent with iron deficiency in each child diagnosed with IDA. Whether the patient had pica was learned from the patient or his/her family. The type, duration and extent of pica and whether other members of the family had pica were questioned. Detailed physical examinations were carried out, and height and body weight measurements were recorded.

Collection of blood samples

An informed consent was obtained from the family members of the children in the study and control groups. Venous blood samples were collected under fasting conditions in the morning. Hemoglobin (Hb), hematocrit (Htc), white blood cell, mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), red cell distribution width (RDW), serum iron, total iron-binding capacity, ferritin levels and hepatic and renal functions were investigated. For zinc, selenium, oxidative enzymes and antioxidant capacity analyses, the collected blood samples were centrifuged and kept in deep freeze at −20°C. On the day of analysis, the samples were removed from the deep freeze and were tested in the corresponding laboratories.

Serum iron and total iron-binding capacity were analyzed calorimetrically using a Roche modular autoanalyzer (Tokyo, Japan), and serum ferritin levels were measured by chemiluminescent immunometric assays (Immulite DPC, Los Angeles, CA, USA). Serum selenium and zinc levels, antioxidants and total antioxidant capacity (TAC) were investigated at the Biochemistry Laboratory of the Medical faculty of Harran University. Serum selenium levels were analyzed by a Zeeman Atomic Absorption Spectrometer (AA 2402, Australia) using graft furnace (Varian GTA-120i, Australia) as per the predefined “internal standard addition method.” Serum zinc levels were measured using Atomic Absorption Spectrometer (SpectrAA 240 FS, Varian, Springvale, Australia).^13,14,15

For patients in the study group, fecal occult blood, parasites, Helicobacter pylori, tissue transglutaminase and antiendomysium antibodies were studied. In addition, duodenal endoscopy and duodenal biopsy were performed for 22 patients positive for tissue transglutaminase and antiendomysium antibodies with clinical findings suggestive of celiac disease.

Analysis of duodenal biopsy

Endoscopic biopsy material of the duodenum was analyzed in Holland solution. Following regular tissue monitoring procedures, the tissues were immersed in paraffin; sections of 4–6 µ were obtained using microtome and were stained with hematoxylin eosin. The obtained preparations were analyzed for villous atrophy, crypt hyperplasia, erosion, Brunner’s hyperplasia, giardia, intraepithelial lymphocyte count and chronic inflammatory cell filtration into the lamina propria.

Measurement of TAC

Plasma TAC levels were determined using a novel automated colorimetric measurement method that measures the TAC of the body against free radicals developed by Erel_Fe2 ± o-dianisidine Fenton-type reaction by forming complexes with the OH radical, which results in the formation of hydrogen peroxide (H₂O₂).¹⁶ This strong reducing reactive oxygen species at low pH reacts with colorless o-dianisidine molecules to form a yellow-brown dianisidyl. Dianisidyl radicals participating in oxidation reactions increase color formation but inhibit color formation of antioxidants in the samples. This reaction is measured spectrophotometrically using an automatic analyzer andthen the results are given. The results are expressed as millimoles of Trolox equivalent per litre.¹⁷

Measurement of total oxidant level

Oxidants in the sample oxidize the ferrous ion–o-dianisidine complex to ferric ion. Glycerol accelerates the rate of reaction about three times. Ferric ions constitute a colored complex with xylenol orange in acidic media. In this fully automated colorimetric method, the color intensity that is associated with the amount of oxidants in the sample is measured spectrophotometrically. The results are expressed in terms of micromolar hydrogen peroxide equivalent per liter (µmol H₂O₂ equiv./l).¹⁸

Evaluation of oxidative stress index

The percentage ratio of total oxidant level (TOL) to TAC yields the oxidative stress index (OSI), an indicator of the degree of oxidative stress. TAC for the account (millimoles of Trolox equivalent/liter) unit, micromolar Trolox equivalent/liter and OSI calculated using the formula [TOS (mol/l)/TAC (mmol Trolox equivalent/l)] × 100.¹⁸

Statistical analysis

The commercial statistical package for the Social Sciences (SPSS) 11.0 software was used for statistical analyses and figures. Student’s t test was used to evaluate inter-group differences, and Pearson’s correlation analysis was performed to evaluate the relationships between parameters. The results were expressed as mean ± SD. Statistical significance was set at p < 0.05.

Results

Of the 90 subjects enrolled in the study in total, 21 children randomized as controls, while 47 had IDA with pica and 22 had IDA without pica. Gender profiles of the groups were 24 males/23 females in the IDA plus pica group, 11 males and 11 females in the IDA group, and 9 females and 12 males in the control group. The subjects included in the study were aged from 1 to 15 years; the mean age in the pica subjects was 7.6 ± 4.5 (1–15) years, 5.7 ± 4.2 (1–13) years in IDA subjects and 6.5 ± 4.0 (1–14) years in the control group. The groups did not differ significantly in age and gender (p > 0.05).

Paleness was the most common complaint in the pica + IDA and IDA groups (84%). The most common pica object was earth (39 patients) followed by ice,³ ash,² paper² and licking walls,¹ respectively. Parasites were identified in 10 (21%) children in the pica group and in 3 children in each of the IDA and control groups (14%). Fecal occult blood was noted in 9 (19%) children in the pica group and in 7 (32%) children in the IDA group and none in the control group. The groups did not differ significantly in terms of parasite or fecal occult blood presence (p > 0.05). H. pylori was identified in 20 (43%) children in the pica group and was positive in 4 (19%) children in each of the IDA and control groups. Frequency of H. pylori was significantly higher in the pica group than the other two groups (p < 0.05).

When we have compared the patients (pica and IDA groups) with the controls, the Hb, Htc, MCV, MCH and MCHC were diminished while the RDW levels were increased among the patients. Moreover, the serum iron, transferrin saturation and ferritin levels of both the pica and IDA groups’ were significantly lower than the controls. Particularly, the iron binding capacity of the pica group was significantly elevated in contrast to the IDA group (Table 1).

Table 1.

Intergroup comparison of serum iron, iron-binding capacity, transferrin saturation.^a

	Pica	IDA	Control	p ^b1–2
	Mean ± SD (min–max)	Mean ± SD (min–max)	Mean ± SD (min–max)	p ^b1–2
Iron (ng/ml)	12.9 ± 7.9 (4–54)	16.5 ± 9 (10–54)	43.9 ± 8.4 (30–60)	0.2
Iron-binding capacity (ng/ml)	407.2 ± 104.9 (140–618)	351 ± 73.8 (181–490)	226.3 ± 31 (180–280)	0.04
Transferrin Saturation (TFS) (%)	3.7 ± 3.1 (0.8–16.6)	4.9 ± 3 (2.9-17)	19.6 ± 3.6 (15-27.1)	0.32
Ferritin (ng/ml)	5 ± 4.5 (1–18)	7.7 ± 6.2 (1–26)	74.2 ± 14.8 (45–101)	0.45

IDA: iron-deficiency anemia.

^a TFS and ferritin values.

^b The p value shows the comparison of pica and IDA groups.

There was no significant difference concerning the trace element levels between the groups (Table 2). Serum zinc levels were significantly lower both in the pica and IDA groups compared to the controls. Despite the insignificance of the findings, serum zinc levels were relatively lower in the pica group in contrast to the IDA group (p > 0.05). Serum zinc levels decreased following the diminution of serum iron levels, but a significant correlation was not discovered between these two markers (p > 0.05). Serum selenium concentration was significantly lower in the pica and IDA groups compared to the control group (p < 0.001). Although the selenium concentration was significantly lower in the pica group comparing with IDA, the difference was not significant (p > 0.05). Although serum selenium levels decreased following the diminution of serum iron levels, no significant correlation was observed between these two parameters (p > 0.05).

Table 2.

Intergroup comparison of serum trace element levels.

	Pica	IDA	Control	p ^a	p ^b	p ^c
	Mean ± SD (min–max)	Mean ± SD (min–max)	Mean ± SD (min–max)	p ^a	p ^b	p ^c
Serum iron (ng/ml)	12.9 ± 78.9 (4–54)	16.5 ± 9 (10–54)	43.9 ± 8.4 (30–60)	0.2	0.001	0.001
Serum zinc (µg/l)	65.7 ± 13 (34.5–98.4)	66.6 ± 14 (46.5–91.1)	97.9 ± 19.2 (56–132.5)	0.97	0.001	0.001
Serum selenium (µg/l)	57.5 ± 11 2 (27.6–77.6)	62 ± 11 (40.1–76.9)	80.9 ± 13.1 (61.8–120.8)	0.33	0.001	0.001

IDA: iron-deficiency anemia.

^a The p value shows the comparison of pica and IDA groups.

^b The p value shows the comparison of pica and control groups.

^c The p value shows the comparison of IDA and control groups.

Given that children with celiac disease may present only with anemia, antiendomysium antibodies were studied in anemic subjects and was found to be positive in 4 (9%) children in the pica group and 1 (5%) child in the IDA group. Tissue transglutaminase was positive in 5 (11%) children in the pica group and 2 (9%) in the IDA group (Table 3). Endoscopy and duodenal biopsy were performed for 22 patients including those suspected to have the symptoms of celiac disease. The endoscopy and duodenal biopsy analyses revealed celiac disease in 8.5% of the pica group and 4.5% of the IDA group.

Table 3.

Celiac disease data.

	Pica (%)	IDA (%)
Tissue transglutaminase positive	5 (10.6)	2 (9)
Antiendomysium positive	4 (8.5)	1 (4.5)
Clinic suspicion^a	16 (34)	6 (27.2)
Duodenal biopsy positive	4 (8.5)	1 (4.5)
Tissue transglutaminase + antiendomysium positive	4 (8.5)	1 (4.5)
Tissue transglutaminase + antiendomysium + duodenal biopsy positive	4 (8.5)	1 (4.5)

IDA: iron-deficiency anemia.

^a Gastrointestinal symptoms, growth retardation and so on.

The children in the study group were also analyzed for TAC, TOL and OSI. TAC levels of pica, IDA and the controls were 1.3 ± 0.3, 1.6 ± 0.4 and 1.7 ± 0.3, respectively. TOL levels of pica, IDA and the controls were 20.9 ± 13.6, 16.6 ± 7.4 and 9.2 ± 2.4, respectively. The patients in the pica group had highest TOL and lowest TAC levels. When we have compared the pica group with IDA, the TAC levels were significantly lower (p < 0.05) and the OSI levels were significantly higher (p < 0.05) among the patients with pica. Additionally, the TOL levels were also higher in the pica group but the difference was not significant (p > 0.05). Concerning the differences between the pica and control groups, all three parameters were significantly different (p < 0.05), however, the same situation was observed only for the OSI levels when the IDA and control groups were compared (p < 0.05; Table 4). Serum iron, selenium, zinc, TOL, OSI and TAC were not significantly correlated (p > 0.05 for all comparisons).

Table 4.

Comparison of subjects’ TAC, TOL and OSI values.

	Pica	IDA	Control	p ^a	p ^b	p ^c
	Mean ± SD (min–max)	Mean ± SD (min–max)	Mean ± SD (min–max)	p ^a	p ^b	p ^c
TAC (micromole Trolox eq./l)	1.3 ± 0.3 (0.61–1.9)	1.6 ± 0.4 (1–2.6)	1.7 ± 0.3 (1.2–2.2)	0.018	0.001	0.6
TOL (micromole H₂O₂/l)	20.9 ± 13.6 (5.9–70.2)	16.6 ± 7.4 (6.6–35.8)	9.2 ± 2.4 (5.5–12.5)	0.29	0.001	0.08
OSI	1.6 ± 0.9 (0.4–4)	1.1 ± 0.5 (0.3–2.4)	0.6 ± 0.1 (0.3–0.7)	0.046	0.001	0.044

TAC: total antioxidant capacity; TOL: total oxidant level; OSI: oxidative stress index; H₂O₂: hydrogen peroxide; IDA: iron-deficiency anemia.

^a The p value shows the comparison of pica and IDA groups.

^b The p value shows the comparison of pica and control groups.

^c The p value shows the comparison of IDA and control groups.

Discussion

Most of the pica cases reported in Turkey involve consumption of Turkish clay and the majority of physiopathological studies are limited to Turkish clay. In the present study, the pica object was earth in the majority of cases (83%) followed by a few cases of ice, paper and ash. Iron is an element with the highest incidence of deficiency in childhood.¹² Coexistence of pica and IDA is a common problem. In the studies carried out by Berçem et al.¹⁹ and Göktaş et al.²⁰ in Turkey, coexistence of anemia and pica was reported in up to 70% of the subjects, whereas this rate was found to be 17% in another city called Elazığ by Gürgöze.²¹ In the present study, we have formed three groups; the first group included children with both pica and IDA, the second group included patients with IDA only, and the third group included healthy controls. We have investigated the potential role of pica and IDA in hematologic and biochemical parameters, trace element levels and oxidant–antioxidant capacity. Considering the presence of celiac disease, duodenal biopsy was performed for 21 children with pica and IDA, in whom celiac disease was suspected, to evaluate the effects of pica on duodenum, the main site of absorption of iron.

The most common presenting complaint of the children at the time of admission to the outpatient service was paleness (84%), which was also the prominent symptom in studies performed by Gürgöze²¹ and Berçem et al.¹⁹ (93 and 94%, respectively), consistent with the results of our findings. Fecal parasites, occult blood and H. pylori antigen were studied in all groups but the results were not significantly different. To determine whether the H. pylori infection, the association of which with IDA is well known, is one of the factors that may affect the pathogenesis of pica, the frequency of H. pylori infection was compared between the pica and IDA groups. H. pylori infection was significantly more common in the pica group (p < 0.05). We, therefore, concluded that H. pylori infection may be an important factor affecting pica pathogenesis. H. pylori incidence has been reported to be 20–80% in IDA.^22

–25 Our results were consistent with these studies. No comparison regarding H. pylori could be made for pica subjects, since there were no relevant studies in the literature except a poster presented by Boğa et al.²⁶ reporting a higher incidence of H. pylori in patients with the habit of eating ice.

The levels of iron and iron depot parameters were investigated for IDA diagnosis and they were lower in both the pica and IDA groups in contrast to the controls. Of these parameters, only iron binding capacity differed significantly between the pica and IDA groups (p < 0.05).

When we have compared the pica and IDA groups with the controls in terms of trace element levels (selenium and zinc), they were significantly diminished in both pica and IDA groups (p < 0.05) in contrast to the controls.

The relationship between pica and zinc was first described in 1960s. Danford²⁷ reported zinc deficiency in 52% of the subjects with long-term pica. In a study conducted in Turkey by Boğa et al.²⁸ evaluating the children and adults in the Central Anatolia region, the incidence of zinc deficiency was reported to be 15.3% in pica patients. The serum zinc levels were relatively lower in the pica group, but the difference was not significant (p > 0.05). This may suggest that zinc deficiency develops due to nutritional factors since the cause is the same in patients diagnosed with IDA. In addition, substitution of zinc for iron in protoporphyrin due to iron deficiency secondary to increased production of zinc porphyrin hemoglobin in IDA may be another cause for low zinc levels. One should consider that decreased zinc absorption in addition to dietary factors might have contributed to the reduced zinc levels in the patients with pica. Gonzales et al.²⁹ have reported that unabsorbed compounds may form as a result of iron and zinc substitution with Turkish clay.

In a recent study by Gürgöze,²¹ low zinc levels in the IDA group were reported. On the other hand, Singhi³⁰ described that serum zinc levels in nonanemic children with pica were lower compared to the control group. Regarding healthy children, the serum selenium levels of the controls in our study were consistent with those reported in a study by Hıncal et al.³¹ Serum selenium level in the IDA group was 62 ± 11 μg/l, in our study. The IDA group had significantly lower level of serum selenium in contrast to the controls. In a study including children diagnosed with IDA, Kalayci et al.³² determined a serum selenium level of 61.2 ± 11 μg/l, which was similar to the serum selenium levels of the IDA group in the present study. In our study, the IDA group differed significantly from the control group in terms of serum iron and selenium levels, although there was no significant correlation.

Serum selenium was 57.5 ± 11.2 μg/l in the pica group, and it was significantly lower than the controls. Although the difference was not statistically significant, the pica group had relatively lower serum selenium levels than the IDA group. The literature review about a relationship between pica and selenium did not yield a particular study. A study carried out by Boğa²⁸ in Ihlara Valley measured the selenium levels of 38 cases with geophagia and 20 individuals without geophagia, and the selenium levels were significantly lower in cases with geophagia.

Celiac disease may be presented with only IDA without any gastrointestinal symptoms.³³ Kalaycı et al.³² determined celiac disease in 6 of 135 children with IDA (4.4%). They reported pica history in 2 of these 6 children. In our study, 4 of the children were diagnosed with celiac disease from the pica group and 1 from the IDA group, corresponding to 8.5% in the pica group and 4.5% for the IDA group. Our results were consistent with those reported previously by Kalaycı et al.³² The gold standard for the diagnosis of celiac disease is the presence of classical triad in distal duodenal biopsy with intraepithelial lymphocytes accompanied by villous atrophy and crypt hyperplasia.³³

In our study, we have evaluated the effects of pica on duodenal mucosa in addition to the celiac disease on duodenal biopsies. Reference studies are very limited in number and all were performed in 1970s. Incidence of abnormalities in intestinal mucosa in pica is similar to that in iron deficiency (60%). Some investigators have argued that these changes were due to iron deficiency, while others, Arcasoy³⁴ in particular, reported that the changes in geophagia were rather secondary to chronic zinc deficiency.

Our study presents the largest series of duodenal biopsy results for the subjects with pica. In the pica group, villous atrophy was identified in 12 (60%) patients, crypt hyperplasia in 12 (60%) and erosion in 13 (65%). Brunner’s gland hyperplasia was noted in 5 (25%) patients. Mean intraepithelial lymphocyte count was 29.1 ± 11 (13–57) in the same group, 15 patients had a score above 20. We believe that the changes described above that occur in the duodenum, which is the primary site for iron absorption, may be one of the causes that lead to IDA.

Reactive oxygen species produced in the body are eliminated by enzymatic and nonenzymatic antioxidant mechanisms. Since it is difficult to study all these antioxidants in practice, a TAC measurement method has been developed recently by Erel¹⁸ TAC demonstrates both enzymatic and nonenzymatic antioxidants. There is an association between oxidative stress and anemia. Reactive oxygen species have been shown to damage erythrocyte membranes and alter their structure.^35,36 Erythrocytes, however, are equipped with multi-effect antioxidant enzyme systems including SOD, catalase, GSH-Pxs and glutathione reductase.³⁷

There are limited and contradicting data in the literature on oxidative stress and antioxidant defense capacity in individuals with iron deficiency. A study carried out by Kurtoğlu et al.³⁷ in 63 adults with iron deficiency reported increases in antioxidant enzyme levels such as malondialdehyde and decreases in antioxidant enzyme levels such as GSH-Px. The authors also reported that oxidant enzyme levels returned to normal in patients who received iron therapy for 6 weeks. Tekin et al.³⁸ determined lower GSH-Px levels in the group with iron deficiency but did not find any differences in SOD and catalase levels compared to the control group. In their study with 19 IDA children, Meral et al.³⁹ reported that iron deficiency did not lead to lipid peroxidation and changes in oxidant enzyme levels.

In our study, we have investigated TOL, TAC and OSI calculated as TOL/TAC in pica, IDA and control groups. Pica group had significantly highest TOL and significantly lowest TAC levels. The patients with pica had significantly lower levels of TAC and OSI and higher levels of TOL in contrast to the controls. However, the patients with IDA had only significant OSI levels in contrast to the controls, and the other parameters were not statistically significant (p > 0.05). Our results with the IDA group were consistent with those reported by Acharya et al.⁴⁰ and İşler et al.,⁴¹ but were contrary to the result of most of the studies discussed above.

The pica group had significantly higher OSI and lower TAC in contrast to the patients with IDA (p < 0.05). Due to the absence of any study comparing patients with pica and anemia, we were unable to compare our results with the literature. However, these results indicate that IDA alone induces some changes in oxidative stress and antioxidant capacity, but that these effects are potentiated to a significant extent in the presence of pica plus IDA.

Yetgin et al.⁴² found significant serum selenium concentrations in patients with IDA. Since GSH-Px is a selenium-dependent enzyme, the authors believed this could be attributed to selenium deficiency and have identified a significant correlation between serum iron, selenium and GSH-Px in their study. Perona et al.⁴³ obtained similar results in their study with 14 patients. In our study, no correlation was identified between serum iron, selenium, zinc levels and oxidative stress or antioxidant capacity.

Conclusion

Our study demonstrated that trace elements including selenium and zinc are significantly lower in pica and IDA groups. A decrement in serum zinc and selenium levels corresponds to reduced serum iron levels, but there is no significant correlation. Celiac may coexist in up to 8.5% of individuals with pica and 4.5% of individuals with IDA. TOL and OSI were higher in patients with pica. TOL and OSI were higher and TAC was lower in IDA, while only OSI was significantly higher in patients with pica. Pica exerts its harmful effects on the organism both by inducing morphological changes in the duodenum and with adverse effects on oxidative stress and antioxidant capacity. These effects should also be considered when managing patients with pica. However, further studies with larger patient groups are required to verify our findings.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

References

Feldman

. Pica: current perspectives. Psychosomatics 1986; 27: 519–523.

Robinson

Tolan

Golding-Beecher

Childhood pica. Some aspects of the clinical profile in Manchester, Jamaica. West Indian Med J 1990; 39: 20–26.

Cavdar

Arcasoy

Hematologic and biochemical studies of Turkish children with pica. A presumptive explanation for the syndrome of geophagia, iron deficiency anemia, hepatosplenomegaly and hypogonadism. Clin Pediatr (Phila) 1972; 11: 215–223.

Ginaldi

. Geophagia: an uncommon cause of acute abdomen. Ann Emerg Med 1988; 17: 979–981.

Sayetta

. Pica: an overview. Am Fam Physician 1986; 33: 181–185.

Del Baño

Lorente

Castillo

Phenolic diterpenes, flavones, and rosmarinic acid distribution during the development of leaves, flowers, stems, and roots of Rosmarinus officinalis. Antioxidant activity. J Agric Food Chem 2003; 51: 4247–4253.

Kim

Yang

. Antioxidative effects of Cichorium intybus root extract on LDL (low density lipoprotein) oxidation. Arch Pharm Res 2001; 24: 431–436.

Parinandi

Thompson

Schmid

. Diabetic heart and kidney exhibit increased resistance to lipid peroxidation. Biochim Biophys Acta 1990; 1047: 63–69.

Burton

. Antioxidant action of carotenoids. J Nutr 1989; 119: 109–111.

10.

Seven

İnci

Civelek

Burçak

Larenks kanserli olgularda lipid peroksidasyon ve antioksidan statü göstergelerinin dokuda incelenmesi. Türk Orl Arşivi 1998; 36: 33–36.

11.

Kuppusamy

Dharmani

Kanthimathi

Antioxidant enzyme activities of human peripheral blood mononuclear cells exposed to trace elements. Biol Trace Elem Res 2005; 106: 29–40.

12.

Oner

Bay

. Demir eksikliği anemisi. Türkiye Klinikleri J Pediatr Sci 2005; 1: 7–15.

13.

Milde

Novak

Stuzka

Serum levels of selenium, manganese, copper, and iron in colorectal cancer patients. Biol Trace Elem Res 2001; 79: 107–114.

14.

Fernandez

. Micromethod for lead determination in whole blood by atomic absorption, with use of graphite furnace. Clin Chem 1975; 21: 558–561.

15.

Manser

Haider

Lalani

Trace element studies on Karachi population, part II: normal ranges for blood copper, zinc and magnesium for children and adolescents. J Pak Med Assoc 1989; 39: 205–208.

16.

Erel

. A novel automated method to measure total antioxidant response against potent free radical reactions. Clin Biochem 2004; 37: 112–119.

17.

Cao

Prior

. Comparison of different analytical methods for assessing total antioxidant capacity of human serum. Clin Chem 1998; 44: 1309–1315.

18.

Erel

. A new automated colorimetric method for measuring total oxidant status. Clin Biochem 2005; 38: 1103–1111.

19.

Berçem

İçağasioğlu

Cevit

The prevalence of iron deficiency and iron deficiency anemia in adolescents. T Klin Pediatr 1999; 8: 15–20.

20.

Göktaş

Yıldırmak

Tanyer

Sağlıklı çocuklarda nutrisyonel demir eksikliği anemisi. Ege Pediatri Bülteni 1999; 6: 49–55.

21.

Gurgöze

. Demir eksikliği olan çocuklarda çinko, bakır ve selenyum düzeyleri. Tıpta Uzmanlık Tezi, Elazığ, 2001.

22.

Kurekci

Atay

Sarici

Is there a relationship between childhood Helicobacter pylori infection and iron deficiency anemia?

J Trop Pediatr 2005; 51: 166–169.

23.

Abasiyanik

Tunc

Salih

. Enzyme immunoassay and immunoblotting analysis of helicobacter pylori infection in Turkish asymptomatic subjects. Diagn Microbiol Infect Dis 2004; 50: 173–177.

24.

Annibale

Marignani

Monarca

Reversal of iron deficiency anemia after helicobacter pylori eradication in patients with asymptomatic gastritis. Ann Intern Med 1999; 131: 668–672.

25.

Cardenas

Mulla

Ortiz

Iron deficiency and Helicobacter pylori infection in the United States. Am J Epidemiol 2006; 163: 127–134.

26.

Boğa

Özdoğu

Asma

Pagophagia-patogenizi etkileyen faktörlerden birisi: helikobakter pilori infeksiyonu. Turk J Hematol 2004; 21: 51–52.

27.

Danford

Huber

. Pica among mentally retarded adults. Am J Ment Defic 1982; 87: 141–146.

28.

Boğa

. Pikada eser elementler, anemi, psikososyal stres ve çevre faktörleri. Tıpta uzmanlık tezi, Adana, 1990.

29.

Gonzalez

Owens

Ungaro

Clay ingestion: a rare cause of hypokalemia. Ann Intern Med 1982; 97: 65–66.

30.

Singhi

Ravishanker

Singhi

Low plasma zinc and iron in pica. Indian J Pediatr 2003; 70: 139–143.

31.

Hincal

Yetgin

Ataceri

Selenium status in Turkey. I. Serum selenium levels in infants and children in Ankara. Biol Trace Elem Res 1989; 20: 161–167.

32.

Kalayci

Kanber

Birinci

The prevalence of coeliac disease as detected by screening in children with iron deficiency anemia. Acta Paediatr 2005; 94: 678–681.

33.

Maki

Collin

. Coeliac disease. Lancet 1997; 349: 1755–1759.

34.

Arcasoy

Akar

Ors

Ultrastructural changes in the mucosa of the small intestine in patients with geophagia (Prasad’s syndrome). J Pediatr Gastroenterol Nutr 1990; 11: 279–282.

35.

Mccord

. Human disease, free radicals and the oxidant/antioxidant balance. Clin Biochem 1993; 26: 351–357.

36.

Gropper

Kerr

Barksdale

. Non-anemic iron deficiency, oral iron supplementation, and oxidative damage in college-aged females. J Nutr Biochem 2003; 14: 409–415.

37.

Kurtoglu

Ugur

Baltaci

Effect of iron supplementation on oxidative stress and antioxidant status in iron-deficiency anemia. Biol Trace Elem Res 2003; 96: 117–123.

38.

Tekin

Yavuzer

Tekin

Possible effects of antioxidant status on increased platelet aggregation in childhood iron-deficiency anemia. Pediatr Int 2001; 43: 74–77.

39.

Meral

Tuncel

Surmen

Lipid peroxidation and antioxidant status in beta-thalassemia. Pediatr Hematol Oncol 2000; 17: 687–693.

40.

Acharya

Punchard

Taylor

Red cell lipid peroxidation and antioxidant enzymes in iron deficiency. Eur J Haematol 1991; 47: 287–291.

41.

Isler

Delibas

Guclu

Superoxide dismutase and glutathione peroxidase in erythrocytes of patients with iron-deficiency anemia: effects of different treatment modalities. Croat Med J 2002; 43: 16–19.

42.

Yetgin

Hincal

Basaran

Serum selenium status in children with iron deficiency anemia. Acta Haematol 1992; 88: 185–188.

43.

Perona

Cellerino

Guidi

Erythrocytic glutathione peroxidase: its relationship to plasma selenium in man. Scand J Haematol 1977; 19: 116–120.