Influence of Sex on Neuroretinal Degeneration: Six-Month Follow-Up in Rats With Chronic Glaucoma

Abstract

Purpose

To evaluate differences by sex in the neuroretina of rats with chronic glaucoma over 24 weeks of follow-up, and to assess by sex the influence on neurodegeneration of different methods of inducing ocular hypertension.

Methods

Forty-six Long–Evans rats—18 males and 28 females—with induced chronic glaucoma were analyzed. Glaucoma was achieved via 2 models: repeatedly sclerosing the episcleral veins (9 male/14 female) or by injecting poly(lactic-co-glycolic acid) microspheres measuring 20 to 10 µm (Ms20/10) into the anterior chamber (9 male/14 female). The IOP was measured weekly by tonometer; neuroretinal function was recorded by dark/light-adapted electroretinography at baseline and weeks 12 and 24; and structure was analyzed by optical coherence tomography using the retina posterior pole, retinal nerve fiber layer and ganglion cell layer protocols at baseline and weeks 8, 12, 18, and 24.

Results

Males showed statistically significant (P < 0.05) higher IOP in both chronic glaucoma models, and greater differences were found in the episcleral model at earlier stages. Males with episclerally induced glaucoma showed a statistically higher increase in retinal thickness in optical coherence tomography recordings than females and also when comparing Ms20/10 at 12 weeks. Males showed a higher percentage of retinal nerve fiber layer thickness loss in both models. Ganglion cell layer thickness loss was only detected in the Ms20/10 model. Males exhibited worse dark/light-adapted functionality in chronic glaucoma models, which worsened in the episcleral sclerosis model at 12 weeks, than females.

Conclusions

Female rats with chronic glaucoma experienced lower IOP and structural loss and better neuroretinal functionality than males. Sex and the ocular hypertension–inducing method influenced neuroretinal degeneration.

Traditionally, the eye was considered a sex-neutral organ. Recent publications, however, have evidenced male/female sex divergence. There are sex steroid receptors in nearly every part of the eye^1,2 and even the retina synthetizes neuroestrogens via aromatase enzymes.³ Differences in ocular blood flow were found between sexes,⁴ and variations in the optic nerve head were described according to phase of menstrual cycle and were correlated to hormonal levels.^5,6 Premenopausal females showed better recordings by perimetry⁷ and electroretinography (ERG).⁸ Physiological ophthalmologic changes caused by hyperestrogenism are also described in pregnancy.^9–11 Moreover, a clear tendency to gender difference is found in ophthalmologic,^12,13 neuropsychiatric, and neurodevelopmental pathologies with immune system implications^14,15 and modulation of aging-linked genes influenced by sex.^16,17

Glaucoma is the second biggest cause of blindness worldwide and will affect 111.8 million people by 2040.¹⁸ Glaucoma is considered an aging-linked pathology caused by progressive death of retinal ganglion cells (RGCs). The IOP is the major modifiable risk factor, with other well-known nonmodifiable factors being age and race.¹⁹ However, there are discrepancies regarding gender in the literature.²⁰ There is evidence of higher incidence of POAG in men (adjusted by age),¹⁸ although the higher prevalence and risk of blindness are found in elderly women, possibly owing to their increased life-expectancy.²¹ Nonetheless, ophthalmic research data on sex differences in glaucoma remain scarce.

The protective effect of estrogens on cardiovascular areas is widely known and it has been suggested that they have a similar effect on the eye, delaying the start of glaucomatous pathology.¹³ In fact, the risk of POAG increases in conditions of low estrogenic exposure, such as late menarche onset, early menopause, or polycystic ovarian syndrome^13,22 and better ophthalmologic parameters were found in studies with hormone replacement therapy, although its use is not recommended.^20,23–25 Estrogens, and especially 17β estradiol, activate endothelial nitric oxide (NO) synthase, which diminishes vascular resistance, increases blood flow in the optic nerve²⁶ and decreases the IOP. All of these effects of estrogens could explain the increase in IOP occurring after the onset of menopause,²⁷ when estrogenic hormone levels decrease abruptly. Furthermore, estrogens exert a protective effect on the extracellular matrix of the trabecular meshwork and lamina cribosa.²⁸ In contrast, estrogens act as an endogen neuroprotector owing to their chemical properties and phenolic ring,^29,30 helping to maintain RGC viability^31,32 and intraretinal synapsis.^12,33 Neuroprotection exerted by estrogens has been demonstrated by topical, intravitreal and systemic administration in animal studies,^34–36 even under conditions of induced ocular hypertension (OHT).

Sex equity in animal studies has been recently recommended on ethical grounds. Moreover, there is growing interest in increasing knowledge about the influence of sex on aging and aging-linked pathologies such as glaucoma.³⁷ Most animal glaucoma studies are performed by inducing OHT, but usually only 1 sex is used. Moreover, even when both sexes are used, differences between them are not analyzed.

To our knowledge, this study is the first to analyze differences by sex in IOP and neuroretinal structure and function under conditions of chronic glaucoma in 2 different OHT models. We demonstrate that females showed physiological protection because they suffered from lower OHT and less structural and functional neuroretinal damage than age-matched males at an early stage of chronic glaucoma. We also showed that the induced model used to increase IOP influenced sex-dependent neuroretinal degeneration. We therefore suggest our outcomes be considered in future studies into neurodegeneration and protection.

Methods

Animals

All work with animals was approved by the Ethics Committee for Animal Research (PI34/17) and was carried out in strict accordance with the Association for Research in Vision and Ophthalmology's Statement on the Use of Animals at the Biomedical Research Center of Aragon (CIBA: Centro de Investigaciones Biomédicas de Aragón). Long–Evans rats at 4 weeks of age and weighing 50 to 100 g at the beginning of the study were housed in standard cages. To keep the same animals together throughout the study, and to preserve the animals’ welfare, a maximum of either 3 males or 4 females were placed in each cage and given environmental enrichment and water and food ad libitum. They were kept under animal welfare conditions (e.g., a 12-hour dark–light cycled room, temperature of 22 °C, and relative humidity of 55%). Body weight was measured periodically to calculate anesthetic dosage.

A total of 46 animals were used: 18 males and 28 females. Chronic glaucoma models were induced using 2 OHT methods. OHT was induced by biweekly sclerosing injections of the episcleral veins (EPI model), as described by Morrison et al³⁸ in 23 rats (9 male/14 female). In the other 23 rats (9 male/14 female), OHT was induced by injecting a suspension of Ms20/10 (10% w/v) into the anterior chamber of the eye (Ms20/10 model) at baseline, doing so biweekly for the first month and then once monthly until week 20, as described by Rodrigo et al.³⁹ The Ms were made of poly(lactic-co-glycolic acid) (PLGA) according to the oil in water emulsion solvent extraction–evaporation technique previously described by Garcia-Herranz et al.⁴⁰ and were prepared by the Innovation, Therapy and Pharmaceutical Development in Ophthalmology (InnOftal) Research Group (Faculty of Pharmacy, Complutense University of Madrid, Spain). All OHT injections were performed in the right eye under surgical conditions: controlled temperature, topical tetracaine (1 mg/mL + oxibuprocaine 4 mg/mL) eye drops (Anestesico doble Colirccusi, Alcon Cusí SA, Barcelona, Spain) and intraperitoneal (60 mg/kg of ketamine + 0.25 mg/kg of dexmedetomidine) anesthetic. Afterward, the rats were left to recover in an enriched 2.5% oxygen atmosphere and were treated with antibiotic ointment (erythromycin 5 mg/g [Oftalmolosa Cusí eritromicina, Alcon Cusí SA]). The Ms20/10 were injected at baseline and at weeks 2, 4, 8, 12, 16, and 20. Episcleral injections were administered biweekly if the IOP was less than 20 mm Hg. Animal weight during the study was similar to that reported by the supplier (Janvier-labs, Le Genest-Saint-Isle, France). Glaucoma induction did not alter the morphometric weight parameter versus healthy animals. No correlation studies were carried out between body weight and retinal loss since these parameters have been previously demonstrated—by means of data standardization—to be independent.⁴¹

Ophthalmologic Clinical Evaluation

IOP measurements were recorded weekly, using a rodent-specific rebound tonometer (Tonolab, TiolatOy, Helsinki, Finland), under a sedative mixture of 3% sevoflurane gas and 1.5% oxygen for less than 3 minutes as recommended.⁴² The IOP value was obtained by averaging three consecutive measurements, taken from the average of 6 rebounds.

Optical coherence tomography (OCT) (OCT Spectralis, Heidelberg Engineering, Germany) was used to quantify the neuroretinal structure in vivo, using a contact lens adapted to the rat's cornea in order to obtain higher quality recordings. The retina posterior pole, retinal nerve fiber layer (RNFL), and ganglion cell layer (GCL) protocols were used.^{39–41,43,44} All of them measured a 3-mm-diameter area centered on the optic disc using 61 b-scans, and subsequent follow-up examinations were performed at this same location using eye-tracking software and a follow-up application. The retina posterior pole protocol analyzed the thickness from the inner limiting membrane to the retinal pigment epithelium divided into the 9 Early Treatment Diabetic Retinopathy areas,⁴⁵ which included a central 1-mm circle centered on the optic disc (no fovea exists in rats), and inner (inferior, superior, nasal, temporal) and outer (inferior, superior, nasal, temporal) rings measuring 2 and 3 mm in diameter, respectively, as well as total volume; the RNFL protocol analyzed the thickness from the inner limiting membrane to the GCL boundaries divided into 6 sectors (inferotemporal, inferonasal, superotemporal, superonasal, nasal, and temporal); and the GCL protocol analyzed from the RNFL to the inner nuclear layer boundaries within the Early Treatment Diabetic Retinopathy areas.

ERG (Roland consult RETIanimal ERG, Germany) was used to quantify neuroretinal functionality using the scotopic full-field ERG and the light-adapted photopic negative response protocols. The scotopic ERG test was performed after 12 hours of dark adaptation and the eyes were prepared to full dilatation using topical eye drops containing tetracaine 1 mg/mL + oxibuprocaine 4 mg/mL (Anestesico doble Colirccusi, Alcon Cusí SA) and mydriatics (tropicamide 10 mg/mL, phenylephrine 100 mg/mL [Alcon Cusí SA]). Eyes were lubricated with hypromellose 2% (Methocel OmniVision, Puchheim, Germany). Active electrodes were placed on the cornea, references were placed at both sides under the skin, and the ground electrode was placed near the tail. Electrode impedance with a difference of less than 2 kΩ between electrodes was accepted.

Both eyes were tested simultaneously using a Ganzfeld Q450 SC sphere with white LED flashes for stimuli, and 7 steps with increasing luminance intensity and intervals were performed (rod response: step 1: –40 dB, 0.0003 cds/m², 0.2 Hz [20 recordings averaged]; step 2: –30 dB, 0.003 cds/m², 0.125 Hz [18 recordings averaged]; step 3: –20 dB, 0.03 cds/m², 8.929 Hz [14 recordings averaged]; step 4: –20 dB, 0.03 cds/m², 0.111 Hz [15 recordings averaged]; step 5: –10 dB, 0.3 cds/m², 0.077 Hz [15 recordings averaged]; mixed rod–cone response: step 6: 0 dB, 3.0 cds/m², 0.067 Hz [12 recordings averaged]; and oscillatory potentials: step 7: 0 dB, 3.0 cds/m², 29.412 Hz [10 recordings averaged]). The photopic negative response protocol was performed after light adaptation to a blue background (470 nm, 25 cds/m²), and a red LED flash (625 nm, –10 dB, 0.30 cds/m², 1.199 Hz [20 recordings averaged]) was used as stimulus. Latency (in milliseconds) and amplitude (in microvolts) were studied in a, b and photopic negative response waves.

OCT tests were performed at baseline, 8 weeks, 12 weeks, 18 weeks, and 24 weeks; ERG tests were performed at baseline, 12 weeks, and 24 weeks, both under intraperitoneally administered anesthesia (60 mg/kg of ketamine + 0.25 mg/kg of dexmedetomidine, as mentioned above for ocular injections). A trained, masked researcher rejected or manually corrected the recordings if the algorithm had clearly failed.

Statistical Analysis

Data from this longitudinal study were recorded in an Excel database and statistical analysis was performed using SPSS software version 20.0 (SPSS Inc., Chicago, IL). To assess sample distribution, the Kolmogorov–Smirnov test was used; however, given the nonparametric distribution of most of the data, the Mann–Whitney U test was performed to evaluate the differences between both cohorts, and a paired Wilcoxon test was used to compare the changes recorded in each eye over the 24-week study period. All values were expressed as means ± standard deviations. Values of P < 0.05 (expressed as *) were considered to indicate statistical significance, and the Bonferroni correction for multiple comparisons was calculated to avoid a high false-positive rate. The level of significance for each variable was established based on Bonferroni calculations (expressed as ^#).

Results

IOP

Both OHT models showed an increase in IOP over follow-up, which occurred earlier and more abruptly in the episcleral sclerosis model. Males in both OHT models, but especially the episcleral model, showed statistically significant higher IOP measurements than females (Fig. 1 and Supplementary Table S1).

Figure 1.

IOP curve in both sexes and chronic glaucoma models over follow-up. EPIm, episcleral sclerosis model; Ms 20/10, microsphere 20/10 model; M, male; F, female. #P < 0.020, for Bonferroni correction for multiple comparisons.

Structural Neuroretinal Analysis by in VIVO OCT

Neuroretinal thickness in males and females was quantified over the study in both chronic glaucoma models (see Supplementary Tables S2 and S3). Males in the episcleral sclerosis model had greater retinal thickness (but similar RNFL and GCL) than females at the end of the study. Moreover, males showed greater fluctuations in retina posterior pole and RNFL thickness than females, which exhibited a progressive and sustained decrease in all three OCT parameters. However, these features were not observed in the Ms20/10 model, in which males had lesser thicknesses and females exhibited greater fluctuations in RNFL (Fig. 2).

Figure 2.

Structural analysis of the neuroretina using OCT in both sexes and chronic glaucoma models. EPIm, episcleral sclerosis model; Ms 20/10, microsphere 20/10 model; OCT, optical coherence tomography; RNFL, retinal nerve fiber layer; GCL, ganglion cell layer complex; average thickness in microns (μm).

The percentages of thickness loss in both sexes and both chronic glaucoma models were then analyzed and compared over the study. As can be seen in Figure 3, the percentage of RNFL thickness loss was greater in males in both OHT models. It was also greater in the GCL and retina posterior pole in the Ms20/10 model, reaching statistical significance at an early stage (week 8). Interestingly, increased retina posterior pole thickness was found in males in the episcleral sclerosis model.

Figure 3.

Percentage neuroretinal structure loss measured using OCT in both sexes and chronic glaucoma models over 24 weeks of follow-up. EPI, episcleral sclerosis model; Ms20/10, microsphere 20/10 model; RNFL, retinal nerve fiber layer; GCL, ganglion cell layer complex; M, male; F, female.

Although perceptual loss of OCT thickness in each sector was also analyzed to identify a sex-dependent neuroretinal degeneration pattern, no obvious pattern was found. Both chronic glaucoma models and sexes lost more (on average) in the inner retina posterior pole and GCL sectors and in the superior-inferior axis sectors in the RNFL. Furthermore, losses in contiguous sectors were found in females in the Ms20/10 model. The patterns of degeneration found are shown in Figure 4.

Figure 4.

Neuroretinal percentage loss in OCT sectors and loss trend averaged over 24 weeks of follow-up. R, retina; C, central, II, inner inferior; OI, outer inferior; IS, inner superior; OS, outer superior; IN, inner nasal; ON, outer nasal; IT, inner temporal; OT, outer temporal; IT, inferior temporal; IN, inferior nasal; ST, superior temporal; SN, superior nasal; N, nasal; T, temporal; EPIm, episcleral sclerosis model; Ms20/10, microsphere 20/10 model; TV, total volume; I, inferior; S, superior; N, nasal; T, temporal; >, greater loss than; ≈, similar; M, male; F, female.

To standardize the neuroretinal loss, the right eye loss rate per day and the millimeters of mercury of increased IOP each week were calculated from the average of all OCT sectors and expressed in micrometers per millimeters of mercury per day. The parameter most affected in both OHT models per millimeter of mercury increase was GCL followed by RNFL and, finally, retina posterior pole. In the Ms20/10 model, both sexes showed a similar rate of GCL loss over time; in the episcleral sclerosis model, a similar loss rate was only observed in the later stages. This feature was also observed in retina posterior pole, although males in the episcleral sclerosis model experienced an inverted loss rate in the early stages. RNFL was the parameter that exhibited widest variations between sexes and models. At the end of the study (week 24), the loss rate in the episcleral sclerosis model was similar between sexes; however, in the Ms20/10 model females showed a slightly lower loss rate than males (Fig. 5).

Figure 5.

Neuroretinal loss rate measured using OCT in both sexes and chronic glaucoma models. EPIm, episcleral sclerosis model; Ms20/10, microsphere 20/10 model; M, male; F, female.

Functional Neuroretinal Analysis by in VIVO ERG

In the episcleral sclerosis model, males exhibited statistically significant lower scotopic ERG signals at week 12; in the Ms20/10 model, no differences by sex were found. In the light-adapted photopic negative response test exploring RGC functionality the males showed a clear tendency to produce worse measurements in both chronic glaucoma models up to week 24. Lower recordings were measured in both models over time, with the episcleral sclerosis model producing the lowest values (see Fig. 6).

Figure 6.

Neuroretinal functionality using dark- and light-adapted ERG in both chronic glaucoma models and sexes. PhNR, photopic negative response; EPIm, episcleral sclerosis model; Ms20/10, microsphere 20/10 model; M, male; F, female; Phases 1 to 7 of dark-adapted ERG in the EPIm at week 12.

Discussion

Several models have been developed for the study of glaucoma. However, the results regarding neurodegeneration differ among different groups, and it has been proposed that this may be due to the use of different animals, animal strains, amounts of injected solution, injection times, and even particle sizes when using microbeads.^46,47 However, until now, no study considered sex as an influential factor in glaucomatous degeneration (Table). In fact, most studies employed only one sex and, when males and females were used together, differences between sexes were not analyzed.^48–82 Glaucoma is a neurodegenerative disease in which associations with another neurodegenerative diseases that are influenced by sex, such as Alzheimer's disease, have been found.^83,84

Table.

Sex in Rodent Glaucoma Models

Model	Sex	References
Episcleral vein sclerosis model	Male	48, 49, 50, 56, 58, 62, 63, 66, 67, 68, 69, 71, 73, 74, 76
	Female	–
	Sex not mentioned	59, 60, 64, 65, 70, 75, 77
	Both sexes	61, 72 (sex differences not analyzed in either)
Microsphere model	Male	51, 53, 57, 82
	Female	52, 54, 55, 78, 81
	Sex not mentioned	80
	Both sexes	79 (sex differences not analyzed)

Given the lack of knowledge about the influence of sex in glaucoma, and given current needs,⁸⁵ the purpose of our study was to investigate the existence of differences between males and females in 2 animal models of glaucoma (sclerosis of episcleral veins and injection of PLGA microspheres in the anterior chamber of the eye).

As stated elsewhere in this article, IOP is the main risk factor for development and worsening of glaucomatous pathology. Several glaucoma models have been developed to induce increased IOP. The episcleral sclerosis model produces very high pressures at early stages, leading to faster neuroretinal degeneration. However, the microsphere model, in which it is possible to modify injection frequency, allows for modulation of the hypertensive curve, thereby achieving a slower and more progressive IOP increase,⁸⁶ as is consistent with our results. Our results also showed that females exhibit lower IOP levels than males in both models. A possible explanation for this finding is the presence of estrogenic receptors in the trabecular meshwork and endothelial cells that regulate NO synthase, these being responsible for increased vasodilation and a lower IOP.⁴ Therefore, females may also benefit from a protective effect in OHT. However, both sexes suffered from higher IOP in the episcleral model. This finding may be due to the induction method itself, because hypersaline solution is injected into the episcleral veins and, via a retrograde pathway, reaches the trabecular meshwork and may even opacify the lens as evidence of the solution entering the eye.⁴⁷ The hyperosmotic effect could damage the endothelial cells, altering the effect of the NO synthase and eliminating, at least partially, the protective effect exerted by estrogen on the vasodilation-mediating enzyme (NO synthase) and increasing IOP in both sexes. In contrast, mineralocorticoid receptors have been found in the iris, ciliary body, endothelial cells, and even in the neuroretina and pigmentary epithelium.⁸⁷ In addition to an NaCl imbalance (as with 1.8 M NaCl solution), aldosterone is upregulated and binds to the mineralocorticoid receptors, causing endothelial dysfunction and vasculopathy.⁸⁸ However, because this damage would not occur so abruptly with the injection of microspheres, it seems that the OHT-inducing method could influence the results differently by sex.

As the increase in IOP produces neuroretinal degeneration, an OCT-based study was designed to evaluate the changes in the neuroretina over time. This technology allows consecutive re-explorations while decreasing the number of euthanized animals. In addition, several studies have shown an adequate correlation between OCT and immunocytochemistry and ERG.^89–91 The OCT recordings were performed from baseline with first re-exploration at 8 weeks of follow-up (corresponding to 12 weeks of rat life). At this age, development and growth of the retina end and the retina reaches maturity.⁹²

In a previous article, we demonstrated that the episcleral sclerosis model induced a more aggressive retinal glial reaction than the microsphere model.³⁹ However, differences between the sexes were not evaluated. This article now shows that these differences are mainly found in males. It seems that the glaucoma-inducing method could also influence sex-dependent neurodegeneration. Our results showed a significant increase in retinal thickness in males at week 12 (which decreased at week 18 and increased again at week 24) in the episcleral sclerosis model; it did not occur in females and did not occur in the Ms20/10 model. These fluctuations³⁹ may be due to greater immune infiltration⁹³ or increased glia sensitivity to hydroelectrolytic or vascular changes, especially in males. A potential transfer of hyperosmotic solution to the posterior pole could have upregulated aldosterone. It has been demonstrated that the systemic administration of aldosterone decreases RNFL thickness under ocular normotensive conditions, and that it was counteracted by the antagonist, spironolactone.⁹⁴ In this sense, Takasago et al.⁹⁵ found a significant negative correlation between the aldosterone levels in plasma and the number of RGCs, which decreased earlier on the periphery and later in central sectors. However, these authors only analyzed males in their study. The mineralocorticoid receptors of the neuroretina and pigment epithelium act in retinal/choroidal homeostasis by balancing sodium and reabsorbing the fluid,⁹⁶ but mineralocorticoid receptor overactivation could induce damage that has been linked to central serous chorioretinopathy.⁹⁷ Central serous chorioretinopathy courses with increased capillary permeability and pressure in choroidal vessels⁹⁸ that increase retinal thickness and cause edema. Moreover, this pathological entity is oddly more prevalent in males. Testosterone increases the risk of central serous chorioretinopathy with increased retinal thickness, and its improvement with finasteride has been demonstrated.⁹⁹ According to the facts mentioned elsewhere in this article, bimonthly injections with hyperosmolar solution for 6 months could overactivate mineralocorticoid receptors and increase retinal thickness, although no evident edema was detected by OCT. In contrast, it is known that glial cell edema (which is also present in glaucoma) is produced by intracellular Na overload (which could be aggravated by the hyperosmotic solution). In both models, a similar pattern of RNFL fluctuation was detected, with increases in thickness at 12 and 24 weeks after induction of the increased IOP. Increases in thickness owing to immune infiltration³⁹ or edema have been detected before cell death.¹⁰⁰ In our episcleral sclerosis model results, fluctuations in RNFL were observed in males, but it seems that premenopausal females counteracted it. In this regard, Neuman et al.¹⁰¹ showed that sex steroids such as progesterone inhibit the swelling of glial cells. The suggestion therefore is that the episcleral sclerosis model would affect males more. In contrast, in the Ms20/10 model a greater fluctuation occurred in females. This suggests that sex influences axonal damage (measured as RNFL thickness change) depending on whether the hypertensive Noxa is pretrabecular (Ms 20/10 model) or post-trabecular (episcleral sclerosis model), although other unknown factors may also exert an influence.

Furthermore, O'Steen et al.¹⁰² demonstrated in rats that, in a situation of chronic stress, retinal thickness (especially in the outer layers) decreased significantly in both sexes, although males lost more than females. The degeneration began on the periphery and later affected the inner sectors. Previous OCT studies by our group found damage at an earlier stage in the outer retinal sectors and affectation of the inner sectors over the follow-up.^39,40 In this study, the degeneration pattern showed greater loss (on average) in the inner sectors, which could be explained by greater damage or could be a consequence of the averaged result already including the inner sectors.

Estrogens have demonstrated a neuroretinal protective effect,^12,31–36 as has progesterone in attenuating microglial-driven neurodegeneration,¹⁰³ and our results showed that females experienced less neurodegeneration in all parameters in the Ms20/10 model as well as in the RNFL in the episcleral sclerosis model. O'Steen¹⁰⁴ proposed that 17β-estradiol was to protect photoreceptors from light damage and Chaychi et al.⁸ showed greater functionality in premenopausal female rats than in males. We found similar results even with OHT Noxa in the episcleral sclerosis model, as the females had greater amplitudes in the scotopic ERG. Interestingly, even when the percentage loss of GCL in the episcleral sclerosis model was similar between sexes (or even slightly higher in females), ERG, and photopic negative response functionality were better preserved in females than in males, indicating better synaptic function in females. In this regard, neurosteroids (such as estrogens) synthesized in the retina have been reported to modulate the gamma-aminobutyric acid and glutamate receptors intervening in the synaptic function,⁸⁵ and estrogens have shown an improvement in synaptic functionality in neuroretinal models. Nevertheless, our results partially contradict the results of other authors who claim that RGC dysfunction occurs earlier and progressively preceding cell death. It is worth mentioning that these studies were carried out with monkeys (sex was not mentioned), and with male rats.¹⁰⁵ In our study, the same outcomes were found in males, but not in females. Thus, there could be a lag, or maintenance of the function and/or nonidentical sequence in females.

A study in humans with POAG showed that photopic negative response was decreased in the early stages of the disease. Patients showed lower photopic negative response amplitude with the decrease in visual field sensitivity, which was correlated with the vertical cupping/disc ratio. However, no differences were found between the sexes.¹⁰⁶ Nevertheless, the results from our study showed that females exhibited greater photopic negative response amplitudes than males in both models.

In a previous study⁴¹ analyzing differences in neurodegeneration in healthy Long–Evans rats by age and sex, we found that males had higher IOP by the end of the study (7 months of age), exhibited greater neuroretinal thickness but higher structural percentage loss, and had worse dark- and light-adapted function than females. These facts served as hypotheses for this article analyzing whether sex could also influence the findings in the OHT models studied. Moreover, as mentioned elsewhere in this article, the episcleral sclerosis model and Ms 20/10 models induce differing extents of damage at different stages of the study. episcleral sclerosis model showed earlier damage than the Ms20/10 model.³⁹ This study constitutes a step forward; indeed, analyzing data by sex has been essential to detecting changes that would have otherwise gone unnoticed. If sex had not been considered, the average of both the male and female curves would have been taken, masking differences that would not be statistically significant in the structural and functional tests owing to an underestimation or overestimation of results (in males and females, respectively) that would counteract each other. Similarly, if the OHT model had not been considered separately, we would be facing the same dilemma owing to underestimation or overestimation (episcleral sclerosis model and Ms 20/10 model, respectively) of the thicknesses and electrical responses, ignoring differences existing since intermediate stages of the study. This argument would also apply to the analysis of the response to potential neuroprotective and hypotensive drugs. Therefore, we demonstrate that Noxa type and sex are important factors in glaucoma onset and progression.

Our group is aware of the main limitation of this study, which is the absence of histologic analysis, even when OCT is a reliable method, so it would be beneficial to support our results with histologic studies. Furthermore, hormone levels were not measured, and our animals did not reach senescence, which is when the prevalence of glaucoma rises, especially in women. It would therefore be advantageous to study more advanced ages in future research.

In conclusion, to our knowledge this study is the first to analyze the influence of sex in 2 different animal POAG models. Our results showed different patterns of degeneration and different loss rates between males and females in the same hypertensive model and also between different models. Premenopausal females exhibited lower IOP and less neuroretinal damage in glaucoma than age-matched males. Differences or discrepancies in the results of previous studies could be due in part to a sex bias, as well as to the OHT-inducing model used. It would be beneficial to consider both sex and the OHT-inducing model in future studies when evaluating both hypotensive and neuroprotective therapies, since our results suggest that premenopausal females maintained a protective effect in relation to OHT and the therapeutic outcomes could be overestimated or underestimated.