This action might not be possible to undo. Are you sure you want to continue?
Anatomical and functional impairment of the retina and optic nerve in patients with anorexia nervosa without vision loss
Marilita M Moschos,1 Fragiskos Gonidakis,2 Eleftheria Varsou,2 Ioannis Markopoulos,1 Alexandros Rouvas,1 Ioannis Ladas,1 George N Papadimitriou2
1st Department of Ophthalmology, University of Athens, Medical School, Athens, Greece 2 1st Department of Psychiatry, University of Athens, Medical School, Athens, Greece Correspondence to Dr Marilita M Moschos, 6, Ikarias Street, Ekali, 14578, Athens, Greece; firstname.lastname@example.org Accepted 18 July 2010
ABSTRACT Aim The aim of this cross-sectional study is to evaluate the macular and retinal nerve ﬁbre layer (RNFL) thickness, as well as the electrical activity of the macula in female patients suffering from anorexia nervosa (AN) without visual failure. Material and methods 13 female patients (26 eyes) suffering from AN without visual failure and 20 age and sex-matched healthy female controls (40 eyes) were studied. For the measurement of the macula thickness and the electrical activity of the macula, the optical coherence tomography (OCT) and the multifocal electroretinogram were used respectively. Results The visual acuity, as well as the visual ﬁelds, the colour vision testing and the dark adaptation test of all patients were normal. However, the mean foveal thickness was 140.04 mm (vs 150.85 in the control group, p¼0.005), and the RNFL thickness was limited to 116.42 mm in the superior area (vs 123.15 in the control group, p¼0.372) and 121.08 mm in the inferior area (vs 137.6 in the control group, p<0.001) around the optic nerve. Also, the mean P1 response density amplitude of the foveal area was 159.04 nV/deg2 (vs 292.43 in the control group, p<0.0001), and the perifoveal area was 79.04 nV/deg2 (vs 82.63 in the control group, p¼0.118). Conclusion The present study shows that in patients with AN, even without visual failure there is a decrease in macular and RNFL thickness, as well as a decrease in the electrical activity of the macula.
and perforation5 and cataract.6 7 One case with rod dysfunction8 and another with central vein occlusion9 were also reported. No other contextual papers on nutrition and retinal impact were found. In our study, we evaluated the thickness of the macula and the retinal nerve ﬁbre layer (RNFL) as well as the electrical activity of the macula with optical coherence tomography (OCT) and multifocal electroretinogram (mf-ERG), respectively, in patients with AN. To our knowledge, this is the ﬁrst time a study has been performed to evaluate the anatomical and functional damage of the macula and optic nerve in anorectic patients before presenting irreversible visual impairment.
MATERIALS AND METHODS
Consecutive patients with AN followed by the 1st Department of Psychiatry were recruited for our study. On admission to the 1st Department of Ophthalmology, none of the patients suffered from osteoporosis, cardiovascular complications or electrolytic abnormalities. Plasma vitamin A and B12 were normal, and only a subclinical deﬁciency of iron and folic acid was present in three cases. Exclusion criteria were a history of ocular surgery, ocular diseases as well as high refractive errors 66 D. Informed consent for imaging and data collection was obtained from the patients after an explanation of the nature of the study. Patients had no history of ocular disease or eye surgery, and no subjective symptoms, such as itching, redness, photophobia, tearing or low vision, were mentioned by the patients. The best-corrected visual acuity was 6/6 in all eyes, and ﬂuorescein angiography did not reveal any leakage or pigmentary lesion of the macula. The study included 66 eyes of 33 female individuals who were divided into two groups. Group A consisted of 26 eyes of 13 female patients suffering from AN. The mean age was 28.62 (SD 6.818) years. Group B consisted of 40 eyes of 20 age and sexmatched volunteers ophthalmologically normal with no ocular or systemic symptoms, who served as normal control subjects. All were female. The mean age was 28.20 (SD 8.118) years. The study was conducted in accordance with the tenets of the Declaration of Helsinki. Two anorectic females refused to participate and were not included in the study. Each individual included in our study underwent a complete ophthalmic examination in the 1st Department of Ophthalmology of Athens
1 of 6
In developed countries, anorexia nervosa (AN) is a psychiatric disorder affecting 1e3% of middleand upper-class women. It affects predominantly female adolescents, and the female/male sex ratio is approximately 10:1. In teenage girls, AN is the third most common chronic disease with an estimated mortality of 5e10%.1 According to the Diagnostic and Statistical Manual of Mental Disorders (DSMIV), the diagnostic criteria for AN are weight loss leading to a body weight 15% below normal, intense fear of gaining weight and becoming fat, distorted body self-image, amenorrhoea of at least three expected menstrual cycles and absence of other physical disorders causing weight loss.2 AN affects almost every organ system such as the skin, gastrointestinal system and hypothalamicepituitaryeovarian hormonal axis.3 4 The literature investigating the ocular affection in AN is very poor and is limited to the anterior part of the eye with the appearance of corneal ulcer
Moschos MM, Gonidakis F, Varsou E, et al. Br J Ophthalmol (2010). doi:10.1136/bjo.2009.177899
(C) 3D appearance of mf-ERG recording and mf-ERG traces of the left eye. Optical coherence tomography (OCT) OCT examination was performed with the OCT model 3000 (Stratus OCT3. The retinal mapping software was used. The bandwidth of the ampliﬁer was 10e300 Hz. (B) OCT scans around the left optic nerve. Varsou E. All eyes were scanned in a radial spoke pattern centred on the foveola with scan length of 6 mm.1136/bjo. The stimulus matrix consisted of 61 segments displayed on a CRT colour monitor driven at a frame rate of 72 Hz. For signal acquisition. and intraocular pressure measurement. colour vision testing.Clinical science University.2009. These hexagons elicit an approximately equal signal amplitude at all locations on a normal retina. the VERIS III (Visual Evoked Response Imaging System.177899 . Japan) was used. OCT scan and mf-ERG recording. et al. Each hexagon was independently alternated between black and white at a rate of 72 Hz. The pupils of the patients were dilated by means of tropicamide 0. and the eyes were optically corrected for near vision to see clearly the small ﬁxation spot in the centre of the stimulus matrix. CA. Dublin.5% and phenylephrine 5%. and the results were similar with little variation. and the stimulation technique allowed a retinal response from each stimulus. calculating the averaged retinal thickness of the central ring.4 mm around the optic disc. Carl Zeiss Meditec. fundus examination. Figure 1 (A) Optical coherence tomography (OCT) scans of the fovea of the left eye of a normal control female subject. ﬂuorescein angiography. including best-corrected visual-acuity assessment (standard Snellen chart). superior (468 to 1358 on a unit circle). USA). Normalised and aligned scans identify the outer and inner layers of the retinal nerve ﬁbre layer. OCT and mf-ERG recording was performed twice by two experienced ophthalmologists. 2 of 6 Moschos MM. The radius of the stimulus array subtended approximately 208 high and 258 wide. providing low-coherence infrared illumination that generates crosssectional images of the retina with an axial resolution of less than 10 mm. The recording of mf-ERG and OCT is not inﬂuenced by refractive errors because the subjects during the examination are fully corrected with the appropriate eye contact lens. doi:10. using an M-sequence 1023. and the ampliﬁcation was 310000. a light-emitting diode was used. Also. nasal (1368 to 2258 on a unit circle) and inferior (2268 to 3158 on a unit circle). Tomey. Gonidakis F.4)’ protocol is designed to acquire three circle scans of diameter of 3. standardised retinal photography with a wide-angle fundus camera using overlapping ﬁelds. The ‘RNFL thickness (3. Multifocal ERG For the recording of the mf-ERG. Nagoya. Br J Ophthalmol (2010). To measure RNFL thickness. Measurements of RNFL thickness from three scans were averaged to provide a mean measurement of the RNFL thickness average as well as the following retinal regions: temporal (3168 to 458 on a unit circle).
The normal ranges for these amplitudes were deﬁned by calculating the median and the 95% CIs in both eyes of 20 normal volunteers (Group B). Br J Ophthalmol (2010).2009. P1-response density amplitude and RNFL measurements and factors related to AN. The response density (amplitude per unit retinal area. The mf-ERG stimuli location and anatomical areas corresponded roughly as follows: ring 1 to the fovea.02) and RNFL of the inferior area (Z¼2.001) mf-ERG ring 2 (Z¼1. we may have a post hoc approach for the sample size needed for a parametric analysis of the data under examination (32 anorectic patients and 32 controls at least). RESULTS Thirteen anorectic female patients with a disease duration of 10. correlations between foveal thickness. ring 2 to the parafovea. These averages give a more accurate view of the relative response densities of each group. Normalised and aligned scans identify the outer and inner layers of the retinal nerve ﬁbre layer. Moschos MM.5.1136/bjo. p¼0. We have conducted a post hoc power calculation according to the GPower software application for power analysis and the simulations of Olejnik and Algina. and the duration of the data acquisition was 8 min divided into eight sessions of 60 s. doi:10. we used a non-parametric test (ManneWhitney U test) to compare the results between the two groups. The amplitude of each group was scaled to reﬂect the angular size of the stimulus hexagon. a KolmogoroveSmirnov Z test was conducted in order to detect differences in both the location and the shape of the distribution between the two groups’ Figure 2 (A) Optical coherence tomography (OCT) scans of the fovea of the right eye of an anorectic female patient. p¼0.468.7.177899 3 of 6 . min BMI ever measured during adulthood. (B) OCT scans around the right optic nerve. A generalised linear model test was used to explore further the differences between the two groups. For the same reason. Because of these results.Clinical science a bipolar contact lens was used in which the active and the reference electrodes were incorporated in the contact lens. 2). The Z test showed signiﬁcant differences in mf-ERG ring 1 (Z¼3.83 vs age of PD patients) participated in the study (ﬁgures 1. ring 4 to the near periphery and ring 5 to the central part of the middle periphery. (C) 3D appearance of mf-ERG recording and mf-ERG traces of the right eye.10 According to the above approach. The retinal response density decreases with eccentricity. It is noticeable that the results from the anorectic group show Statistical analysis Because of the small sample size. although there is no further decrease from ring 4 to ring 5. Gonidakis F. The ground electrode was attached to the ear lobe. Retinal response densities of the fovea are considerably decreased for both eyes. measurements. The fellow eye was closed. The recording procedure was repeated if there were any spurious potentials from eye blinks or if ocular movement was recorded. Box plots of the clinical measurements are presented for the anorectic group (ﬁgure 3) and the control group (ﬁgure 4). Varsou E. et al.001). such as body mass index (BMI).3. p¼0. and duration of the disorder in years were conducted using the Spearman rank test. nV/ deg2) of each local response was estimated as the dot product between the normalised response template and each local response.4 years (mean6SD) and 20 age-matched healthy female controls (p¼0. which produces the response. ring 3 to the perifovea.
0006) were different between the two groups (table 4). The ManneWhitney U test was used for the comparison of the two groups. The generalised linear model analysis showed that the inferior RNFL thickness (p¼0. RNFL temporal. 4 of 6 DISCUSSION Several physical complications are associated with AN. second. with AN patients who were not vomiting having a higher P1-response density amplitude in both the right (median¼164) and left eye (median¼185).8. compared with the AN patients who were inducing vomit (right eye median¼107. The two groups showed differences in three measurements.0001) and the P1-response density ampli- Figure 4 Box plots of the clinical measurements for the left eye of both groups. (tables 1. RNFL superior. p¼0.001).2009. retinal nerve ﬁbre layer (RNFL) inferior. z1.0001) and ring 2 (p¼0. Br J Ophthalmol (2010). The Spearman rank test was used to investigate possible correlation between ophthalmological measurements and factors related to AN (BMI. doi:10.1136/bjo. z1. 2). None of the ophthalmological measurements showed a correlation with the reported min BMI (table 5). the P1-response density amplitude of ring 1 in anorectic patients was lower than in controls (p¼0. n. ring 2. the inferior area (p¼0. RNFL temporal. and third.5.04) were different between the two groups (table 3). central macular thickness. i. the P1-response density amplitude of ring 1 in anorectic patients was lower than in controls (p¼0. Restrictive anorectic patients reduce their daily caloric intake. When the patients (eight patients) who were not inducing vomit were compared with the patients (ﬁve patients) who were inducing vomit. s. CMT. All other differences did not reach statistical signiﬁcance. First. p¼0. Similarly. the analysis of the inferior area in anorectic patients showed a lower RNFL thickness (p¼0. The duration of the illness was also correlated negatively with the RNFL of the superior area (p¼0.177899 . The results of the comparison between the two groups regarding the right eye’s measurements are presented in table 1.001). RNFL superior. ring 2.001) and the superior RNFL (p¼0. the ManneWhitney U test was used to compare the ophthalmological measurements between restrictive-type AN patients (six patients) and bingeepurge-type AN patients (seven patients). RNFL nasal. p¼0. First. while bingeepurge anorectic patients occasionally succumb to binge episodes that are followed by purging behaviours such as induced vomit and laxative and diuretics abuse.001).001) and the average RNFL (p¼0. min BMI and duration of the disorder in years). t. s. with the AN restrictive type having a higher foveal thickness (median¼142) than the AN bingeepurge type (median¼134).07).06) and left eye (z¼1. CMT. Figure 3 Box plots of the clinical measurements for the right eye of both groups. the ManneWhitney U test was used to compare the ophthalmological measurements between AN patients who reported to be under any kind of psychiatric medication (ﬁve patients.Clinical science tude of ring 1 (p¼0. central macular thickness. Gonidakis F. retinal nerve ﬁbre layer (RNFL) inferior. t. The two groups showed differences in two measurements. n.001).37.002). No difference was found between the two groups in any of the measurements. a much wider spread from the median than those in the control group. and second. z2.02). the only marginal statistical difference that was found concerned the P1-response density amplitude of ring 1 of mf-ERG in both the right (z¼1.7. the analysis of the inferior area in anorectic patients showed a lower RNFL thickness (p¼0. Varsou E. the foveal thickness measured by OCT in anorectic patients was lower than in controls (p¼0. there are two types of AN: the restrictive type and the binge/purging type.0003). the nasal area (p¼0. RNFL nasal.001) and ring 2 (p¼0.0002) and the P1-response density amplitude of ring 1 (p¼0. 16 eyes). The generalised linear model analysis showed that the foveal thickness measured by OCT (p¼0.0003). 10 eyes) and the AN patients who have never received any kind of psychiatric medication (eight patients.0071). Because of the small size of our sample. ring 1. the inferior RNFL thickness (p¼0. z2. The test showed that BMI of anorectic patients at presentation correlated negatively with RNFL of the inferior area (p¼0.021). The two groups differed only in the left eye OCT measurements (Z¼2. et al. All other differences did not reach statistical signiﬁcance. ring 1.005). According to the American Psychiatric Association Diagnostic and Statistical Manual (DSM-IV) 2 classiﬁcation for mental disorders. The results of the comparison between the two groups regarding the left eye’s measurements are presented in table 2. left eye median¼153). Many of these problems in behaviour were associated with controlling Moschos MM. i.
89 13. n[20 Upper quartile 138 184 104 78 133 135 85 Median 152 289 83 73 125 139 77 Lower quartile 141 281. possibly resulting from impaired dopaminergic neurotransmission. Berthout et al8 described a case with affection of the cones and rods of the retina resulting in a decrease in central vision and visual-ﬁeld constriction bilaterally.0013 0.25 Measurements Optical coherence tomography Multifocal electroretinogram (ring 1) Multifocal electroretinogram (ring 2) RNFL T RNFL S RNFL I RNFL N Median 134 159 76 70 124 128 72 Lower quartile 132 147 68 65 119 110 63 RNFL I.2009. et al. retinal nerve ﬁbre layer (RNFL) inferior. ring 2. However. z2.19 Standard 0.36 À0. temporal. Also.0024 0.07 0.41 0. which is much greater in anorectic patients than in controls.74 À0. A larger population study would probably provide a reason for this ﬁnding.79 p Value 0. Vision is frequently affected in AN.5 90.13 À4.14 p Value 0.001 0. RNFL nasal.63 À1. RNFL superior. RNFL T. and the ﬁndings are very interesting. ring 2. Fluorescein angiography did not reveal any lesion of the pigmentary epithelium of the macula and the perimacular area.0022 Wald 2. RNFL N. Br J Ophthalmol (2010).0001 Table 4 Variables Generalised linear model for the left eye Estimate 0.5 77 128 140.5 Upper quartile 160 302. and the affection of the retina was attributed to a lack of vitamin A.021 0. This impairment may be related to decreased appetitive function.36 14. because a similar high spread of the measurements was not observed in the OCT and mf-ERG.5 6 only one case of an anorectic patient is reported with retinal lesions. Laurence et al11 support that in AN there is impairment in visual discrimination learning.0071 0. n[20 Median 147 295 81 74 1225 138 79 Lower quartile 138. ring 1.026 Wald 3.and b-wave amplitude and absence of oscillatory potentials. where there is a reduction in dopamine in the retina and especially in the amacrine cells.0003 0.1 0. there is Table 3 Variables Generalised linear model for the right eye Estimate À0.04 0.61 0.0074 0. Table 2 Comparison between anorectic and control groups (ManneWhitney U test) for the left eye Anorectic group.0050 0.0022 0. Dopamine is an important neurotransmitter in the visual pathway. RNFL N. and most of these problems resolved once eating habits and weight loss returned to normal.1136/bjo.17 0. There is a dearth of literature investigating the ocular impact of AN.59 0.5 136 70.5 284.008484 Standard 0.0058 À0. Gonidakis F.81 À0. The scotopic electroretinogram was almost absent.5 83 ManneWhitney U test Z À2.0055 0.6 0.35 107. n[13 Control group. RNFL S.53 0.00096 0. body weight in an unhealthy manner. RNFL T. z2.5 79 127.002 0.3 À1. ring 1.5 ManneWhitney U test Z À1.55 0.177899 5 of 6 . RNFL superior. and the photopic ERG was abnormal with a decrease in a.18 À1. and its presence was documented in the mammalian retina. either as a result of food restriction or.5 Upper quartile 163 299 89. It is also interesting that these ﬁndings correlated negatively with the duration of the disease and the BMI on admission to the Department of Ophthalmology.47 0. Moschos MM. as indicated in the box plot ﬁgures.0006 0.16 À0.57 4.005 0. but they never show any correlation with the patient’s min BMI.00056 0. there is a high spread of the RNFL measures and mf-ERG responses of ring 1.5 77 67 118. temporal. Our results show that the retinal thickness of the macula is higher in restrictive-type anorectic patients than in bingeepurge type patients.86 p Value 0.69 0.0002 Intercept Optical coherence tomography Multifocal electroretinogram z1 Multifocal electroretinogram z2 Retinal nerve ﬁbre layer inferior z1.26 0.Clinical science Table 1 Comparison between anorectic and control groups (ManneWhitney U test) for the right eye Anorectic group.073 0. Indeed. Except from two studies concerning the anterior part of the eye.71 À4.0001 0.4 RNFL I.00051 0.35 10. more intriguingly.12 13 show electrophysiological changes in the retina.79 À2. related to the underlying pathophysiology of AN itself. Inzelberg et al17 suggested that it is possible that impoverished dopaminergic input to a subset of retinal ganglion cells contributes to abnormal production of glutamate and atrophy of these selected optic nerve ﬁbres and localised thinning of RNFL.04 113. RNFL nasal. which means that the anatomical impairment of the fovea is greater in the AN bingeepurge type.41 0. n[13 Measurements Optical coherence tomography Multifocal electroretinogram (ring 1) Multifocal electroretinogram (ring 2) RNFL T RNFL S RNFL I RNFL N Median 140 131 72 75 114 111 76 Lower quartile 130 96 65 67 103 105 62 Upper quartile 150 199 87 85 124 132 81 Control group.0023 0.5 74 67. doi:10.5 120 135 71.75 À0. Previous studies in Parkinson’s disease.0023 0.43 p Value 0.166 0. This is difﬁcult to explain. In our study. retinal nerve ﬁbre layer (RNFL) inferior. Varsou E. the electrical activity of the macula and the thickness of the macular area in AN without visual failure have been studied.52 À2.29 11.5 140 85. including the human retina. Intercept Optical coherence tomography Multifocal electroretinogram z1 Multifocal electroretinogram z2 Retinal nerve ﬁbre layer inferior z1.0003 0.14 15 and Palmowski et al16 in a recent study found a reduction in the amplitude of central oscillatory potentials. RNFL S.
Appetite 2003. temporal.23 À0. To the best of our knowledge. 17.97:1001e7. Behnke S. Anorexia nervosa associated with cataract (report of a case). Retinal nerve ﬁber layer thinning in Parkinson’s disease. Electrophysiological signs in dopamine deﬁciency in recently diagnosed Parkinson’s disease and a follow-up study.04 À0. 4th edition text revision (DSM-IV-TR).34 0. Patient consent Obtained.119:109e10. body mass index.39 À0.23 À0.6y 0. Ikeda H. A case of anorexia nervosa in a young woman with development of subcapsular cataract.63y 0. Vitamin E.16 RNFL N À0. 25 0. J Fr Ophtalmol 2007. Weghaupt H.18 The limited number of patients included in our study prevents deﬁnitive conclusions.79y 0.53y À0. et al. Nevertheless. Bellodi L. Whether these deﬁcits are preliminary signs. 01 0. Langan SM. 11. 12. vitamin A and essential fatty acid status of patients hospitalized for anorexia nervosa. 03 0. 13 À0. Haggendal J.05 1 0.25:241e86. 1986:666e8.16 0.39 0. Measures of effect size for comparative studies: Applications.19 0. Head GM. Decrease dopamine in the retina of patients with Parkinson disease. Washington. RNFL I.29 0. Berthout A.44:2793e7. Provenance and peer review Not commissioned. Handbook of Eating DisordersdPhysiology Pathology and Treatment Obesity.78:217e22. which will conduct to visual failure. as shown by a reduced concentration of dopamine metabolite homovenillic acid and altered growth hormone response to apomorphine stimulation. Gilbert JM. 2.41* 0.64:58e66.36 Multifocal electroretinogram z2 0. BMI min and duration of anorexia nervosa) and ophthalmological measurements Variables Optical coherence Multifocal Multifocal tomography electroretinogram z1 electroretinogram z2 RNFL T RNFL S RNFL I RNFL Av BMI 0. et al. Weiss JS.177899 . Inzelberg R. Laurence AD. Acta Physiol Scand 1963. Evidence of dopamine concerning neurons in the retina of the rabbit. In: Brownell Kd. Arancio C. Inﬂuence of dopamine deﬁciency in early Parkinson’s disease on the slow stimulation multifocal ERG.4* 0.19 À0.41:1054e60. Semin Adolesc Med 1966.23 À0.112:209e15. retinal nerve ﬁbre layer (RNFL) Av. Effect of levedopa on the human pattern electroretinogram and pattern visual evoked potentials. we presume that the substantial evidence against vitamin A deﬁciency in these cases may support the idea that other nutritional abnormalities or impaired dopaminergic neurotransmission play a more substantial role in the function of the photoreceptors. Interpretations and limitations. Harnois C.56y 0. Olejnik S.64y 0.07 RNFL Av 0. Although we cannot pinpoint the cause of the photoreceptors’ abnormality.31:2473e5. Miller DA. American Psychiatric Association. Varsou E.2:21e5. superior. 2000:65e8. Hayasaka S.47* À0. ring 1. Shaw H. 6 of 6 Moschos MM.19 RNFL T 0. 9. there are no publications concerning the relationship between the affection of photoreceptors and the lack of dopamine in AN. Anorexia nervosa: medical and physiological aspects. z2. 8. Impaired visual discrimination learning in anorexia nervosa.Clinical science Table 5 Spearman correlation matrix between factors related to anorexia nervosa (BMI. Sattler AL.23 0. Graefes Arch Clin Exp Ophthalmol 1989.35 À0.14 À0.1136/bjo.36 1 0. Vision Res 2004. Despite the limitations of the study. 7. Brambilla F. et al.26:393e408. 15. 06 À0. 18. Am J Ophthalmol 1995. Ocular manifestations and impression cytology of anorexia nervosa. Ethics approval Ethics approval was provided by the Athens University.227:421e7. Ellis CJ. RNFL S. Invest Ophthalmol Vis Sci 1990. 16. Diagnostic and Statistical Manual of Mental Disorders. Trans Ophthalmol Soc U K 1958. Farell PM.18 0. Central dopaminergic function in anorexia and bulimia nervosa. Anorexia and Bulimia.17 0.8y 1 Duration of the BMI min disorder in years À0.19 0.01 À0.19 0. Sellam M. Di Polo T. Malmfors T. Sheniker IR.25 0. New York: The Press.30:15.23 0. average. it is stated that patients with AN have an impaired dopamine function.40:85e9. et al. 5. a statistically signiﬁcant decrease in macular and RNFL thickness as well as a decrease in electrical activity in the macula and especially of the foveal area. inferior.32 1 À0. Mitchell JE. 4. Gottlob I. Gonidakis F. 14 À0.2009. et al. REFERENCES 1.33 À0.64y 1 0.64y 0. Br J Ophthalmol (2010). Competing interests None. Central retinal vein occlusion in a patient with anorexia nervosa.17 0.09 À0. Denimal F. minimum. et al.23 À0.71 0. externally peer reviewed.05 *Correlation is signiﬁcant at the 0. Vision Res 1994. 24 0. Diagnosis and early management of anorexia nervosa and bulimia.41* RNFL I 0. Nisipeanu P.19 À0. Dowson J.58y Optical coherence tomography 1 Multifocal electroretinogram z1 0. 14.001 level (two-tailed).8y 0.15 À0. Vass C. Acta Ophthalmol 1965.05 level (two-tailed).38 À0. doi:10. 38 0. 13.32 À0.79y 0. Oeil et anorexie. Contemp Educ Psychol 2000. Foxal GL. A propos d’un cas (in French).32 0. Psychoneuroendocrinology 2001. Perez MT. Algina J.38 À0. a signiﬁcant anatomical and functional impairment has been proven for the ﬁrst time in the literature. min. 3. 6. Stigmar G. Ramirez JA. RNFL N. Shibuya Y. BMI. ring 2. 57y À0. et al. 04 0. nasal. Ophthalmology 1990. ed.24 0.12 À0. z1. et al.19 RNFL S 0. or will be normalised on recovery is an important issue for further research.34:2629e38. Palmowski-Wolfe AM. DC: American Psychiatric Association. yCorrelation is signiﬁcant at the 0.63y À0. Doc Ophthalmol 2006.56y 1 0. 10. 23 0.43:787e9.05 0. Am J Clin Nutr 1985. RNFL T.
This action might not be possible to undo. Are you sure you want to continue?
We've moved you to where you read on your other device.
Get the full title to continue reading from where you left off, or restart the preview.