Discussion
Glaucoma is a major public health problem and the leading cause of blindness across the world and a progressive optic neuropathy characterised by both ocular morphological changes, such as optic nerve disc cupping or loss of retinal ganglion cell, and ocular functional changes, such as visual field defects that correspond with the loss of the area of the optic neuronal rim. Some of the risk factors for primary open-angle glaucoma (POAG), such as high IOP,14 advancing age,15 sex,16 family history,17 myopia,18 race,19 high systolic blood pressure, low diastolic blood pressure20 and diabetes mellitus,21 have been widely investigated and described. Considered a multifactorial disorder, several theories for the pathogenesis of glaucomatous optic neuropathy have been described as a mechanical compression,22 a vascular23 and a genetic theory.24
Specific examination of the optic disc is a very important part of the assessment in patients with glaucoma. There is a push for more objective optic disc assessment, such as the HRT, ocular coherence tomography and GDx nerve fibre analyser. HRT is an established imaging device that creates both accurate and reproducible measurements of optical nerve head parameters, including VCDR, for the diagnosis and monitoring of glaucoma.25 VCDR was classified by five CDR classes, including 0.7 (VCDR >97.5th percentile), the value that was used as glaucoma diagnosis according to the International Society of Geographical and Epidemiological Ophthalmology criteria.13 There was a higher proportion of older groups and a lower proportion of younger groups in the upper CDR class. Conversely there was a lower proportion of older groups and a higher proportion of younger groups in the lower CDR class. Overall there was a significant positive relationship between CDR class and age group.
In univariate analysis, IOP and CCT were significantly greater for men than for women in the fourth and fifth decade groups. IOP is influenced by many factors, such as CCT and corneal biomechanical properties.26 CCT especially has a significant influence on IOP measurement. It has been reported that thinner corneas result in artificially lower IOP readings and that thicker corneas cause artificially high IOP readings. This is because the thick cornea has greater resistance to corneal shape distortion against external pressure than the thinner cornea, leading to the projected IOP reading being less than the real IOP in a thin CCT.27
IOP, BMI, systolic blood pressure and diastolic blood pressure were significantly greater for men than for women in the fourth and fifth decade groups. Pasquale et al reported that BMI had a significantly positive relation with IOP after adjusting for several parameters.9 Previous reports explained that increasing intraorbital fat tissue caused a high intraorbital pressure, an increase in episcleral venous pressure and a consequent decrease in outflow facility. As a result, IOP may increase. Obesity increases blood viscosity through increasing red cell count, haemoglobin and haematocrit, as well as consequently increased outflow resistance of episcleral veins, resulting in increased IOP.28 29 Elevated blood pressure increases IOP by increasing ciliary artery pressure and ultrafiltration of the aqueous humour.30 31 In addition, some studies have documented positive associations between BMI and blood pressure in several ethnicities.8
VCDR and optic disc area were significantly greater for men than for women in the seventh decade group. Several researchers reported positive association between VCDR and optic disc area in normal and glaucomatous eyes.32–34 Because several factors are intricately related to each other, multivariate analyses were performed using linear mixed model adjusted for age, optic disc area and blood pressure.
In multivariate analysis, only BMI was a significant VCDR-related factor in men after adjusting for age, optic disc area and blood pressure. Even though BMI is a number based on weight and height, neither height nor weight was found to be a significant determinant of greater VCDR after adjusting for age, gender and disc area in a stepwise multiple regression analysis. Further studies will be needed in order to explore this relationship and the meaning of BMI. In contrast, IOP and CCT were significant VCDR-related factors in women after adjusting for age, optic disc area and blood pressure. Several authors described that CCT was significantly thinner in normal tension glaucoma (NTG) than in POAG or normal subjects.35 36 At a glance, BMI and CCT were negatively related to VCDR, while IOP was positively related to VCDR (figure 2). A large epidemiological study suggested that the significant factors of elevated VCDR were increased age, being male, higher IOP, lower diastolic blood pressure, lower BMI and previous cataract surgery.37 Previous studies suggested that higher IOP was the most significant factor of elevated VCDR. However, our data suggested that the most significant factor of elevated VCDR was IOP in women, but BMI in men. For this possible reason, a large population-based epidemiological study reported that the Japanese population seems to have a higher incidence rate of NTG.38 39 NTG is a common form of POAG and considered a mechanism shared with both conditions.40 However, IOP seems to be a relatively lower potential factor in NTG than in POAG. A low BMI may be related to NTG for men. Pasquale et al
9 reported that higher BMI among women was associated with a low risk of NTG (IOP ≤21 mm Hg). Also, a population-based study in Singapore found that men generally had greater VCDR than women and that lower BMI among both sexes was associated with greater VCDR.37 These studies hypothesised that higher oestrogen levels from fatty tissue may lead to neuroprotection. On the other hand, some researchers revealed a statistically significant correlation between elevated cerebrospinal fluid pressure (CSFP) levels and higher BMI.41 42 They concluded that a depressed CSFP caused by low BMI might be a potential risk factor for NTG. We speculated that in men BMI levels are inversely related to glaucoma because BMI is inversely related to VCDR. Our hypothesis is that CSFP levels, not oestrogen levels from body fat tissue, affect VCDR since we could not find out the relationship between VCDR and BFP in our multivariate analysis.
IOP may be a relatively low risk factor in NTG compared with POAG, though. Considering the possible mechanism of a translaminar pressure gradient and treatment difficulties in NTG, we believe there must be other factors contributing.
In conclusion, we observed a significant negative relationship between VCDR and BMI for men and VCDR and CCT for women, as well as a significant positive relationship between VCDR and IOP for women in a large population-based study. Male participants with higher BMI had elevated VCDR than those with lower BMI. Female participants with thinner CCT had elevated VCDR than those with thicker CCT. Also, female participants with higher IOP had elevated VCDR than those with lower IOP. These are important findings that may help to clarify the reason for higher proportions of NTG in Asia (including Japan) and to find novel therapeutic options for refractory glaucoma in which the visual field defect worsens despite treatment with topical and/or surgical therapy. Moreover, it may shed light on the different mechanisms leading to glaucoma in men and women.