Acta Ophthalmologica 2014
Retinal vessel diameters decrease with macular ganglion cell layer thickness in autosomal dominant optic atrophy and in healthy subjects Cecilia R€ onnb€ack,1,2 Karen Grønskov2,3 and Michael Larsen1,2,4 1
Department of Ophthalmology, Glostrup Hospital, Glostrup, Denmark Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark 3 Applied Human Molecular Genetics, Kennedy Center, Rigshospitalet, Copenhagen, Denmark 4 National Eye Clinic, Kennedy Center, Rigshospitalet, Copenhagen, Denmark 2
ABSTRACT. Purpose: To investigate retinal trunk vessel diameters in subjects with autosomal dominant optic atrophy (ADOA) and mutation-free healthy relatives. Methods: This cross-sectional study included 52 ADOA patients with the optic atrophy 1 (OPA1) exon 28 (c.2826_2836delinsGGATGCTCCA) mutation (age 8.6–83.5 years) (best-corrected visual acuity (BCVA) 8–94 Early Treatment Diabetic Retinopathy Study (ETDRS) letters) and 55 mutation-free first-degree healthy relatives (age 8.9–68.7 years, BCVA 80–99). Analysis of fundus photographs provided integrated magnification-corrected measures of retinal vessel diameters (central retinal artery equivalent, CRAE, and central retinal vein equivalent, CRVE). Statistical analysis was corrected for age, gender, spherical equivalent refraction, axial length and mean arterial blood pressure (MABP) in a mixed model analysis. Results: Retinal arteries and veins were thinner in ADOA than in healthy controls (CRAE (mean 2 standard deviations (SD)) 153.9 41.0 lm and CRVE 236.1 42.0 lm in ADOA, CRAE 172.5 25.0 lm (p = 0.0004) and CRVE 254.2 37.6 lm (p = 0.0019) in healthy controls). MABP was comparable in the two groups (p = 0.18), and in both groups, CRAE decreased with increasing MABP (p = 0.01 and p < 0.0001, respectively). In ADOA, CRAE and CRVE decreased with age (p = 0.011 and p = 0.020, respectively) and CRAE decreased with decreasing BCVA (p = 0.011). In patients with ADOA and in healthy controls, CRAE decreased with decreasing average macular ganglion cell–inner plexiform layer (GC-IPL) thickness (p = 0.0017 and p = 0.0057, respectively). Conclusion: Narrow retinal arteries and veins were associated not only with the severity of ADOA but with ganglion cell volume in patients with ADOA and in healthy subjects. This suggests that narrow vessels are a consequence rather than the cause of inner retinal hypoplasia or atrophy, although longitudinal studies are needed to confirm this hypothesis. Key words: dominant optic atrophy – autosomal dominant optic atrophy – retinal vessel diameters – hypertension
Acta Ophthalmol. 2014: 92: 670–674 ª 2014 Acta Ophthalmologica Scandinavica Foundation. Published by John Wiley & Sons Ltd
doi: 10.1111/aos.12378
670
Introduction Autosomal dominant optic atrophy (ADOA), also called Kjer disease (OMIM 165500), is the most common inherited optic neuropathy (Kjer 1959; Kjer et al. 1996). The prevalence is around 1:50 000 in most countries where it has been investigated, but considerably higher, 1:10 000, in Denmark (Kjer et al. 1996). Affected subjects have a defect of the optic atrophy 1 (OPA1) gene on chromosome 3q28q29 that codes for a protein involved in mitochondrial membrane integrity and biogenesis (Alexander et al. 2000; Delettre et al. 2000). The phenotype is characterized by a deficit of ganglion cells, most prominently in the macula, temporal pallor of the optic disc, progressive visual field defects and tritan colour vision anomaly (Kjer et al. 1996). The onset of symptoms is generally before the age of twenty and often in the first decade of life (Kjer 1959). Visual field loss is the principal clinical manifestation of ADOA, although sensorineural hearing loss can be observed in some patients (Delettre et al. 2002; Newman 2005). Visual acuity varies from normal to legal blindness, and a high variance is seen within affected families. The disease has a high incomplete penetrance with variable expression (Votruba et al. 1998). Retinal nerve fibre layer (RNFL) thickness and the thickness of the combined ganglion cell and inner plexiform layers (GC-IPL) are affected,
Acta Ophthalmologica 2014
where recent studies have found evidence that ADOA, to some extent, is an optic hypoplasia (Barboni et al. 2010; Milea et al. 2010; Yu-Wai-Man et al. 2011). A study on a small number of patients with ADOA has found decreased blood perfusion and blood flow velocities (Granse et al. 2003). OPA1 consists of 30 exons and spans more than 100 kb (Delettre et al. 2000). Alternative splicing generates several isoforms. The main isoform is 960 amino acids long, encoded by 28 exons. Many OPA1 mutations have been described, of which c.2826_2836delinsGGATGCTCCA in exon 28 is the most common in Denmark (Thiselton et al. 2001; Almind et al. 2012). This provides a favourable opportunity to study the effects of non-OPA1 factors on ADOA in genetically homogenous cohorts. This study is a cross-sectional study on patients with c.2826_2836delinsGGATGCTCCA mutation in OPA1 and unaffected first-degree family members without the mutation. The purpose of the study was to investigate the retinal anatomy by routine ophthalmological examinations in a homogenous group of patients with ADOA. The GCIPL thickness and RNFL thickness in the same population have previously been published (Ronnback et al. 2013). In this, we wanted to further examine the retinal vessel diameters and blood pressure in the above-mentioned crosssectional study population to find out whether ADOA is a general hypoplasia or atrophy of the retina and to add information about decreased blood perfusion in patients with ADOA found by Granse et al. (2003).
Materials and Methods The study included 52 patients with a confirmed c.2826_2836delinsGGATG CTCCA mutation in OPA1 and a control group of 55 mutation-free first-degree relatives. The control group was required not to have a history of eye disease other than refractive anomaly of magnitude within 7 dioptres spherical equivalent refraction, having clear refractive media, visual acuity of 80 Early Treatment Diabetic Retinopathy Study (ETDRS) letters or better and absence of c.2826_2836delinsGG ATGCTCCA mutation, as described in Ronnback et al. (2013). For both groups, the minimal age for participation was 8 years. Sixty-seven patients
with ADOA and 57 mutation-free relatives were screened for the study. Fifteen patients and two controls were excluded mainly due to poor image quality. Patients and unaffected first-degree relatives were recruited from a national register and examined in the order they volunteered. Written informed consent was obtained from all patients and healthy subjects. For participants under the age of 18 years, informed consent was obtained by their respective guardian. The study was approved by the Medical Ethics Committee of Copenhagen County and followed the tenets of the Declaration of Helsinki. Subjects underwent an ophthalmologic examination including applanation tonometry, refractioning and determination of best-corrected visual acuity (BCVA) in ETDRS letters, slitlamp and fundoscopy examination, including intra-ocular pressure measurement (IOP), fundus photography and blood pressure measurement. Subjects were asked about their smoking habits. Pupils were dilated using tropicamide 1% and phenylephrine hydrochloride 10% before fundus photography (digital grey scale, 50 degrees, 2000–3000 pixel panchromatic sensor) in red-free illumination (FF 450plus, Zeiss, Jena, Germany). Only tropicamide 0.5% was used in subjects under 15 years. Optic disc-centred fundus photographs from right eyes were analysed using customized digital image analysis software (Taarnhoj et al. 2006) that tracks and measures the diameters of the six largest peripapillary arteries and the six largest veins (minimal length > 80 lm) within a circular zone spanning the distance from one half to one diameters from the optic disc. The diameters are summarized into a single value for arteries and one for veins representing a hypothetical central retinal artery equivalent diameter (CRAE) and a hypothetical central retinal vein equivalent diameter (CRVE) calculated using the empirical formulae developed by Knudtson et al. (2003). The relative diameters are described by the artery-to-vein diameter ratio (AVR = CRAE/CRVE). Two or more photographs were analysed per eye. Magnification (M) by the camera and ocular refractive media was estimated according to Bengtsson & Krakau (1992), where M = 1 0.017 *spherical equivalent refraction.
Blood pressure and pulse were measured using computer-controlled brachial cuff manometry. Normal blood pressure was defined as having systolic blood pressure
Retinal vessel diameters decrease with macular ganglion cell layer thickness in autosomal dominant optic atrophy and in healthy subjects Cecilia R€ onnb€ack,1,2 Karen Grønskov2,3 and Michael Larsen1,2,4 1
Department of Ophthalmology, Glostrup Hospital, Glostrup, Denmark Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark 3 Applied Human Molecular Genetics, Kennedy Center, Rigshospitalet, Copenhagen, Denmark 4 National Eye Clinic, Kennedy Center, Rigshospitalet, Copenhagen, Denmark 2
ABSTRACT. Purpose: To investigate retinal trunk vessel diameters in subjects with autosomal dominant optic atrophy (ADOA) and mutation-free healthy relatives. Methods: This cross-sectional study included 52 ADOA patients with the optic atrophy 1 (OPA1) exon 28 (c.2826_2836delinsGGATGCTCCA) mutation (age 8.6–83.5 years) (best-corrected visual acuity (BCVA) 8–94 Early Treatment Diabetic Retinopathy Study (ETDRS) letters) and 55 mutation-free first-degree healthy relatives (age 8.9–68.7 years, BCVA 80–99). Analysis of fundus photographs provided integrated magnification-corrected measures of retinal vessel diameters (central retinal artery equivalent, CRAE, and central retinal vein equivalent, CRVE). Statistical analysis was corrected for age, gender, spherical equivalent refraction, axial length and mean arterial blood pressure (MABP) in a mixed model analysis. Results: Retinal arteries and veins were thinner in ADOA than in healthy controls (CRAE (mean 2 standard deviations (SD)) 153.9 41.0 lm and CRVE 236.1 42.0 lm in ADOA, CRAE 172.5 25.0 lm (p = 0.0004) and CRVE 254.2 37.6 lm (p = 0.0019) in healthy controls). MABP was comparable in the two groups (p = 0.18), and in both groups, CRAE decreased with increasing MABP (p = 0.01 and p < 0.0001, respectively). In ADOA, CRAE and CRVE decreased with age (p = 0.011 and p = 0.020, respectively) and CRAE decreased with decreasing BCVA (p = 0.011). In patients with ADOA and in healthy controls, CRAE decreased with decreasing average macular ganglion cell–inner plexiform layer (GC-IPL) thickness (p = 0.0017 and p = 0.0057, respectively). Conclusion: Narrow retinal arteries and veins were associated not only with the severity of ADOA but with ganglion cell volume in patients with ADOA and in healthy subjects. This suggests that narrow vessels are a consequence rather than the cause of inner retinal hypoplasia or atrophy, although longitudinal studies are needed to confirm this hypothesis. Key words: dominant optic atrophy – autosomal dominant optic atrophy – retinal vessel diameters – hypertension
Acta Ophthalmol. 2014: 92: 670–674 ª 2014 Acta Ophthalmologica Scandinavica Foundation. Published by John Wiley & Sons Ltd
doi: 10.1111/aos.12378
670
Introduction Autosomal dominant optic atrophy (ADOA), also called Kjer disease (OMIM 165500), is the most common inherited optic neuropathy (Kjer 1959; Kjer et al. 1996). The prevalence is around 1:50 000 in most countries where it has been investigated, but considerably higher, 1:10 000, in Denmark (Kjer et al. 1996). Affected subjects have a defect of the optic atrophy 1 (OPA1) gene on chromosome 3q28q29 that codes for a protein involved in mitochondrial membrane integrity and biogenesis (Alexander et al. 2000; Delettre et al. 2000). The phenotype is characterized by a deficit of ganglion cells, most prominently in the macula, temporal pallor of the optic disc, progressive visual field defects and tritan colour vision anomaly (Kjer et al. 1996). The onset of symptoms is generally before the age of twenty and often in the first decade of life (Kjer 1959). Visual field loss is the principal clinical manifestation of ADOA, although sensorineural hearing loss can be observed in some patients (Delettre et al. 2002; Newman 2005). Visual acuity varies from normal to legal blindness, and a high variance is seen within affected families. The disease has a high incomplete penetrance with variable expression (Votruba et al. 1998). Retinal nerve fibre layer (RNFL) thickness and the thickness of the combined ganglion cell and inner plexiform layers (GC-IPL) are affected,
Acta Ophthalmologica 2014
where recent studies have found evidence that ADOA, to some extent, is an optic hypoplasia (Barboni et al. 2010; Milea et al. 2010; Yu-Wai-Man et al. 2011). A study on a small number of patients with ADOA has found decreased blood perfusion and blood flow velocities (Granse et al. 2003). OPA1 consists of 30 exons and spans more than 100 kb (Delettre et al. 2000). Alternative splicing generates several isoforms. The main isoform is 960 amino acids long, encoded by 28 exons. Many OPA1 mutations have been described, of which c.2826_2836delinsGGATGCTCCA in exon 28 is the most common in Denmark (Thiselton et al. 2001; Almind et al. 2012). This provides a favourable opportunity to study the effects of non-OPA1 factors on ADOA in genetically homogenous cohorts. This study is a cross-sectional study on patients with c.2826_2836delinsGGATGCTCCA mutation in OPA1 and unaffected first-degree family members without the mutation. The purpose of the study was to investigate the retinal anatomy by routine ophthalmological examinations in a homogenous group of patients with ADOA. The GCIPL thickness and RNFL thickness in the same population have previously been published (Ronnback et al. 2013). In this, we wanted to further examine the retinal vessel diameters and blood pressure in the above-mentioned crosssectional study population to find out whether ADOA is a general hypoplasia or atrophy of the retina and to add information about decreased blood perfusion in patients with ADOA found by Granse et al. (2003).
Materials and Methods The study included 52 patients with a confirmed c.2826_2836delinsGGATG CTCCA mutation in OPA1 and a control group of 55 mutation-free first-degree relatives. The control group was required not to have a history of eye disease other than refractive anomaly of magnitude within 7 dioptres spherical equivalent refraction, having clear refractive media, visual acuity of 80 Early Treatment Diabetic Retinopathy Study (ETDRS) letters or better and absence of c.2826_2836delinsGG ATGCTCCA mutation, as described in Ronnback et al. (2013). For both groups, the minimal age for participation was 8 years. Sixty-seven patients
with ADOA and 57 mutation-free relatives were screened for the study. Fifteen patients and two controls were excluded mainly due to poor image quality. Patients and unaffected first-degree relatives were recruited from a national register and examined in the order they volunteered. Written informed consent was obtained from all patients and healthy subjects. For participants under the age of 18 years, informed consent was obtained by their respective guardian. The study was approved by the Medical Ethics Committee of Copenhagen County and followed the tenets of the Declaration of Helsinki. Subjects underwent an ophthalmologic examination including applanation tonometry, refractioning and determination of best-corrected visual acuity (BCVA) in ETDRS letters, slitlamp and fundoscopy examination, including intra-ocular pressure measurement (IOP), fundus photography and blood pressure measurement. Subjects were asked about their smoking habits. Pupils were dilated using tropicamide 1% and phenylephrine hydrochloride 10% before fundus photography (digital grey scale, 50 degrees, 2000–3000 pixel panchromatic sensor) in red-free illumination (FF 450plus, Zeiss, Jena, Germany). Only tropicamide 0.5% was used in subjects under 15 years. Optic disc-centred fundus photographs from right eyes were analysed using customized digital image analysis software (Taarnhoj et al. 2006) that tracks and measures the diameters of the six largest peripapillary arteries and the six largest veins (minimal length > 80 lm) within a circular zone spanning the distance from one half to one diameters from the optic disc. The diameters are summarized into a single value for arteries and one for veins representing a hypothetical central retinal artery equivalent diameter (CRAE) and a hypothetical central retinal vein equivalent diameter (CRVE) calculated using the empirical formulae developed by Knudtson et al. (2003). The relative diameters are described by the artery-to-vein diameter ratio (AVR = CRAE/CRVE). Two or more photographs were analysed per eye. Magnification (M) by the camera and ocular refractive media was estimated according to Bengtsson & Krakau (1992), where M = 1 0.017 *spherical equivalent refraction.
Blood pressure and pulse were measured using computer-controlled brachial cuff manometry. Normal blood pressure was defined as having systolic blood pressure
