Abstract
Purpose:
The purpose of this study was to determine whether there are differences in choroidal thickness in children with iron deficiency anemia (IDA).
Methods:
Fifty-two patients with IDA and 54 healthy children between 3 and 16 years of age were enrolled in this study. After complete eye examinations were conducted for each participant, the choroidal thickness was measured using optical coherence tomography. Correlations between the choroidal thickness and clinical and laboratory parameters were also evaluated.
Results:
There were no statistically significant differences between the two groups in terms of visual acuity, intraocular pressure, central corneal thickness, or axial length (P > 0.05). The choroidal thicknesses at the foveal center were 303.13 ± 27.14 μm in the IDA patients and 333.67 ± 39.77 μm in the healthy control children (P < 0.001); additionally, the choroidal thicknesses at each point within the horizontal nasal and temporal quadrants were thinner in the IDA group. There were positive correlations between the choroidal thickness and hemoglobin (r = 0.337; P < 0.001), mean corpuscular volume (r = 0.305; P = 0.001), iron (r = 0.264; P = 0.006), and ferritin (r = 0.287; P = 0.003) levels; however, there were no correlations between the clinical or ocular characteristics and the choroidal thickness.
Conclusions:
The patients with IDA had significantly thinner choroidal thicknesses than those of the healthy children. Choroidal thinning in childhood may be an early sign of deterioration in the ocular blood circulation, without any risk of atherosclerosis in advanced age in the patients with IDA.
Iron deficiency anemia (IDA) is still the most common type of nutritional anemia among children worldwide.
1–3 Iron plays an important role in oxygen transport and additional roles in the central nervous system, including normal myelination, neurotransmitter synthesis, and neurometabolism.
4 Iron deficiency in children has been implicated in developmental abnormalities, ischemic stroke, venous thrombosis, breath-holding episodes, and other neurologic problems.
5
Central retinal vein occlusion, retinal hemorrhage, ischemic retinopathy, and papilledema have been reported in patients with IDA.
6,7 The exact pathophysiology of fundus lesions is still unknown, but it has been suggested, is related to hypoxia. In retinal tissue, iron is particularly important for the visual phototransduction cascade. Photoreceptor cells are persistently shedding and synthesizing their outer segments, which contain disc membranes, and these cells depend on iron-containing enzymes for the synthesis of the lipids used in producing new disc membranes.
8 The choroid exists between the retina and the sclera, and consists of many vascular beds that supply nutrients and oxygen to the retinal tissues, including the retinal pigment epithelium and photoreceptors. Consequently, any structural or functional changes in the choroidal blood flow may cause deterioration in the structure or function of the retina.
9
To the best of our knowledge, there have been no studies to date that have evaluated the choroidal thicknesses in children with IDA. Therefore, this study represents the first effort to determine whether there are differences in choroidal thickness in children with IDA.
This prospective cross-sectional study was performed in the Department of Ophthalmology at Adiyaman University Hospital in Turkey, between January and July 2015. The study was designed according to the tenets of the Declaration of Helsinki and was approved by the ethics committee of Adiyaman University. Informed written consent was obtained from all the parents of the participants prior to their admission into the study.
Fifty-two patients with IDA and 54 healthy children between 3 and 16 years of age were included in this study. The patients' IDA was diagnosed using microcytic hypochromic anemia (on the basis of a peripheral blood smear, hemoglobin <10.5 g/dL; mean corpuscular volume [MCV] <70 fL; mean corpuscular hemoglobin <23 pg; and mean corpuscular hemoglobin concentration <30 g/dL) and their iron status (serum ferritin level <12 ng/mL and serum transferrin saturation <10%). The control group was composed of healthy children who were randomly selected after a general screening at the ophthalmology outpatient clinic and who had hemoglobin levels of >12 g/dL, ferritin levels of >12 ng/mL, and transferrin saturation >10%.
Individuals with a history of diabetic retinopathy, glaucoma, keratoconus, contact lens use, ocular trauma, ocular surgery, any chronic disease or uveitis, amblyopia, strabismus, and vascular abnormalities detected upon ophthalmologic examination were excluded from the study. Patients with myopic and hypermetropic and astigmatic refractive errors of greater than ±1.0 D were also excluded. Finally, patients who did not agree to the required optical coherence tomography (OCT) examinations were excluded.
All patients underwent detailed ophthalmologic examinations, including refractive error, visual acuity, slit-lamp examination, IOP, and dilated fundus examinations. The best visual acuity values (as measured on a Snellen chart) were recorded for each patient. The values for mean visual acuity were obtained by using descriptive statistics. The axial length of the eye was measured using an optical biometer (Lenstar LS 900; Haag Streit AG, Koeniz, Switzerland).
The choroid was imaged with undilated pupils, using the choroidal mode of the spectral-domain OCT system (RTVue-100 OCT; Optovue, İnc., Fremont, CA, USA) with version 6.3 software. All images were recorded by 2 experienced ophthalmologists, and 2 sets of measurements were made and averaged. Differences between the readings of the masked physicians were determined to be within 10% of the mean, and those images with signal strength indexes >60 were included. All examinations were performed between 1:00 PM and 3:00 PM to exclude diurnal variations, and right eye measurements were recorded for statistical analyses. The measurements of both right and left eyes in each patient and in controls were obtained to assess the reproducibility of their OCT machine measurements. The choroidal thickness was measured manually as the distance from the outer retinal pigment epithelial line to the hyper-reflective line behind the large vessel layers of the scleral interface of the choroid. It was measured at 7 points: at the foveal center and within the horizontal nasal and temporal quadrants at 500-μm intervals, to a distance of 1500 μm from the foveal center (
Fig. 1). The central corneal thickness (CCT) was measured during the OCT using the pachymetry map obtained with the use of an optional attached lens (8 × 1024 A-scans); images with signal strength indexes >30 were included.
The statistical analysis was performed using the Statistical Package for the Social Sciences version 15.0 (SPSS, Inc., Chicago, IL, USA). Data were expressed as the mean ± SD, and the normality of the continuous data was assessed using the Kolmogorov-Smirnov test. The continuous variables, including age, ferritin, hemoglobin, visual acuity, axial length, IOP, and CCT, were assessed through the independent 2-sample Student's t-test. Categorical variables were compared using the chi-square test. The Pearson and point-biserial correlation analysis was used to evaluate the relationship between the variables. A P value of <0.05 was considered statistically significant.
We evaluated a total of 61 patients with IDA during the study period; however, 9 patients were excluded from the study (4 patients because of noncooperation, 3 patients because of refractive errors, and 2 patients because of image quality problems). After these exclusions, 52 patients with IDA and 54 healthy children remained as participants in this study. Of these 52 patients, 24 (46.2%) were males, and 28 (53.8%) were females. In the control group, 26 patients (48.1%) were males, and 28 (51.9%) were females. The mean ages for the study group and for the control group were 7.87 ± 2.57 and 7.52 ± 1.38 years of age, respectively, and there were no statistically significant differences between the two groups in terms of sex or age (P = 0.837 and P = 0.387, respectively).
The mean serum iron level of the children with IDA was 21.66 ± 8.44 μg/dL, and that of the control group was 93.1 ± 17.7 μg/dL (
P < 0.001). The mean serum ferritin level of the children with IDA was 7.54 ± 1.36 ng/mL, and that of the control group was 17.64 ± 1.88 ng/mL (
P < 0.001). The mean hemoglobin level of the patient group was 8.26 ± 1.62 (range, 5.4–10.2 g/dL), and that of the control group was 13.52 ± 0.81 g/dL (
P < 0.001). The refractive status of each group was similar (
P = 0.309), and there were no statistically significant differences between the groups in terms of IOP, CCT, or axial length (
P > 0.05). The demographic and ocular characteristics of the groups are shown in
Table 1, and there were no retinal pathologies in the patient group.
Measurements were performed by two investigators. There were no statistically significant differences between the measurements performed by the two investigators (
P = 0.288). The choroidal thicknesses at the foveal center were 333.67 ± 39.77 μm in the healthy control children and 303.13 ± 27.14 μm in the IDA patients (
Figs. 1,
2). The choroidal thickness was significantly thinner in the IDA group (
P < 0.001). Additionally, the choroidal thicknesses at each point within the horizontal nasal and temporal quadrants were thinner in the IDA group than in the healthy group (
Table 2). There were positive correlations between the choroidal thickness and hemoglobin (
r = 0.337;
P < 0.001), mean corpuscular volume (
r = 0.305;
P = 0.001), iron (
r = 0.264;
P = 0.006), and ferritin (
r = 0.287;
P = 0.003) levels, but there were no correlations between the clinical or ocular characteristics and the choroidal thickness (
Table 3) . The distribution graphs for the correlation between choroidal thickness and hemoglobin and ferritin levels were given in
Figure 3.
To assess the reproducibility of the participants' OCT machine measurements, the left eyes of the study group were compared with those of the control group. There was a significant reduction in the choroidal thickness of the left eyes in the children with IDA too (
Table 4).
In the present study, we demonstrate that pediatric patients with IDA had a significantly thinner choroidal thickness than that in healthy children. The choroid was thinner in all quadrants, from the nasal to the temporal regions.
Before using the OCT devices, the choroid was evaluated using indocyanine green angiography, ultrasonography, or laser flowmetry. These techniques are useful for showing abnormalities in the choroidal vessels and changes in the blood flow, but they do not provide three-dimensional anatomical information about the choroid. Recently, OCT has been shown to provide advanced imaging to evaluate the three-dimensional anatomy of the choroid layers and retinal pigment epithelium.
10,11 In studies using OCT, the choroidal thickness has been reported to be decreased in patients with diabetic retinopathy, rheumatoid arthritis, and myopia.
12–14 Xin et al.
15 reported that the choroidal thickness was thinner in patients with obstructive sleep apnea syndrome due to intermittent hypoxia. Additionally, Mathew et al.
16 reported that the decreased choroidal thickness in patients with sickle cell disease may be due to ischemia and a slower blood flow.
Iron is necessary for the structural stability and maintenance of the optic nerve and for phototransduction in the retina.
17,18 In previous studies, retinal vein occlusion, ischemic retinopathy, and ischemic optic neuropathy have been reported in patients with IDA.
6,19 In addition, Turkyilmaz et al.
20 reported that the average retinal nerve fiber layer and the retinal nerve fiber layers of the superior and inferior quadrants were significantly thinner in the IDA group than in the control group. Furthermore, Carraro et al.
7 suggested that severe anemia plays a significant role in the pathogenesis of retinal abnormalities. The choroid receives the major blood supply to the eye, especially the outer retinal structures. Based on these observations, we suggest that IDA, which causes a decrease in the iron and ferritin levels and hypoxia, can disrupt the structure of the choroid that allows the blood to flow to the eye and that this disruption may cause disorders in other ocular components.
Some blood abnormalities, including IDA, may disturb the retinochoroidal circulation in young adult patients who do not have arteriosclerosis. This disturbance causes retinopathy, including retinal vein and retinal artery occlusion.
6,21 The mechanism of the disturbance of the retinochoroidal circulation in patients with IDA remains unclear; however, iron deficiency may contribute to a hypercoagulable state by affecting blood flow patterns within the vessels because of reduced deformability and increased viscosity of microcytic red blood cells.
22 The other postulated mechanism behind hypercoagulability was reduced antioxidant defense due to IDA and reduced activity of glutathione peroxidase, which corresponds to increased platelet aggregation, reactive thrombocytosis, and elevated factor VIII levels.
23
Foulds et al.
24 revealed a higher prevalence of retinopathy in individuals with hemoglobin levels of less than 6 g/dL; however, Aisen et al.
25 suggested that anemia may be associated with retinopathy, regardless of its severity or type. Carraro et al.
7 emphasized the importance of age as a possible risk factor for fundus pathologies in anemic patients; although, in our study, the IDA patients were very young and had no retinal pathologies, and the choroidal thickness was not correlated with age. The hemoglobin levels of the patients ranged between 5.4 and 10.2 g/dL, but the choroidal thickness was reduced in all patients. Furthermore, the hemoglobin, ferritin, MCV, and iron levels were positively correlated with the choroidal thickness.
A reduction in the choroidal thickness is known to correlate with advanced age, longer axial length, increased myopia, higher IOP, and thicker CCT, but in those previous studies, the factors associated with choroidal thickness were investigated in aging populations.
26–28 In our study, no significant correlations were found between the choroidal thickness and age, CCT, IOP, or axial length. This difference may be due to the fact that those studies used subjects whose health status and ages differed from those in our study; these facts could have affected the measurements. Ruiz-Moreno et al.
29 reported that the temporal choroid was thicker than the subfoveal choroid in the pediatric population. On the other hand, Chhablani et al.
30 reported that subfoveal choroid was thicker than temporal choroid. Consistent with their study, in our study the subfoveal choroid was thicker than the temporal choroid.
The present study had several limitations; for example, our IDA group was of a relatively small size. Secondly, we do not know how long the patients had iron deficiency anemia previous to the study; therefore, we do not know how long it takes to change the choroidal thickness in anemia.
In summary, iron deficiency anemia may cause retinal vein or artery occlusion in young adult patients who do not have arteriosclerosis. Our results suggest that, choroidal thinning in childhood may be an early sign of deterioration in the ocular blood circulation, without any risk of atherosclerosis in advanced age in the patients with IDA. However, it is necessary to support our work using a larger study group.
Disclosure: A. Simsek, None; M. Tekin, None; A. Bilen, None; A.S. Karadag, None; I.H. Bucak, None; M. Turgut None