Noninvasive evaluation of anterior segment and tear film parameters and morphology of meibomian glands in a pediatric population with hypogonadism

The Ocular Surface xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

The Ocular Surface journal homepage: www.elsevier.com/locate/jtos

Original Research

Noninvasive evaluation of anterior segment and tear film parameters and morphology of meibomian glands in a pediatric population with hypogonadism Gamze Dereli Cana,∗, Özlem Karab a b

Department of Ophthalmology, Bursa Yuksek Ihtisas Training and Research Hospital, Turkey Department of Pediatric Endocrinology and Metabolism Clinic, Bursa Yuksek Ihtisas Training and Research Hospital, Turkey

ARTICLE INFO

ABSTRACT

Keywords: Corneal endothelial cell density Corneal thickness Dry eye Non-invasive tear film break-up time Meibomian gland

Purpose: To compare the meibomian gland (MG), non-invasive tear film break-up time (NITFBUT), anterior segment measurements between healthy children and children with hypogonadism. Methods: A total of 80 eyes of 40 children with hypogonadism and 86 eyes of 43 age- and sex-matched healthy subjects were included in the study. The mean keratometry (Km), maximum keratometry (Kmax), central (CCT), thinnest (TCT) and apical (ACT) corneal thicknesses, corneal volume (CV), anterior chamber depth (ACD), iridocorneal angle (ICA), first and average non-invasive NITFBUT, MG loss, morphology of MGs, and MG distortion grade, specular endothelial cell density (CD), coefficient of variation (CoV), and percentage of hexagonal cells (HG) were analysed. Results: The mean CCT and TCT values were approximately 20 μm lower on average in patients with hypogonadism (p < 0.05). MG loss was present 56.1% of the healthy children, the ratio increased to 81.3% in children with hypogonadism (p < 0.001). The morphology and distortion grade did not show any significant differences between groups (p > 0.05). The mean NITFBUT value were similar between groups (p > 0.05). The mean CD value did not show any significant difference between groups, however it decreased in the hormone replacement therapy (HRT) group (p = 0.005). Conclusions: MG loss is a physiological process that is prominent in the condition of sex steroid deficiency, but does not cause tear film alterations in children. Future studies investigating sex and gender effect on the ocular surface system in an age-based fashion are required to clearly communicate influences in the arenas of ocular surface research.

1. Introduction The hypothalamic-pituitary axis is the primary regulator of the endocrine system. The inputs arise from the central nervous system and peripheral endocrine organs are processed in the hypothalamus and converted to signals inducing pituitary gland to release hormones affecting various tissues and regulating whole body functions [1]. The influence of sex steroids (i.e. androgens, estrogens and progestins) on the eye has been known for almost more than two-thousand years. Differences associated with hypothalamic-pituitary-gonadal (HPG) axis and sex steroids have been identified in most of the ocular surface and anterior segment structure such as, meibomian gland, lacrimal gland, conjunctiva, cornea, and anterior chamber. However, the sex-related diversities are not due solely to the influence of sex steroids. As detailed

in Tear Film and Ocular Surface Dry Eye Work Shop II (TFOS DEWS II) report, overall hypothalamic-pituitary hormones, glucocorticoids, insulin, insulin-like growth factor-1 (IGF-1) and thyroid hormones also participate especially in the regulation of ocular surface and adnexal tissues [2]. Competent functioning at all levels of the HPG axis is essential for gonadal development and subsequent sex steroid production. Abnormalities within the HPG axis lead to hypogonadotropic hypogonadism (central hypogonadism) while primary isolated gonadal failure is characterized as hypergonadotropic hypogonadism (primary hypogonadism, peripheral hypogonadism). Both the central and peripheral hypogonadism can be attributed to a variety of congenital origins including chromosomal abnormalities, syndromes, or genetic mutations and acquired conditions such as trauma, irradiation, chemotherapy, and

∗ Corresponding author. Department of Ophthalmology, Bursa Yuksek Ihtisas Training and Research Hospital, Mimar Sinan Mah. Emniyet Cad, Polis Okulu Karşısı Yıldırım, Bursa, 16310, Turkey. E-mail address: [email protected] (G. Dereli Can).

https://doi.org/10.1016/j.jtos.2019.09.001 Received 7 March 2019; Received in revised form 14 August 2019; Accepted 5 September 2019 1542-0124/ © 2019 Elsevier Inc. All rights reserved.

Please cite this article as: Gamze Dereli Can and Özlem Kara, The Ocular Surface, https://doi.org/10.1016/j.jtos.2019.09.001

The Ocular Surface xxx (xxxx) xxx–xxx

G. Dereli Can and Ö. Kara

tumors [3]. Clinically, the most common cause of hypogonadism is sex chromosome aneuploidy as is present both in Turner Syndrome and Klinefelter Syndrome. Isolated abnormalities of sex chromosomes are also associated with primary gonadal failure [4]. Diagnosis of Turner Syndrome requires the combination of characteristic phenotypic features as well as missing or abnormal X chromosome. Although variants exist with different numbers of X chromosomes, the most common genotype of Klinefelter Syndrome is XXY [5]. Even though generally uncommon, individuals with these syndromes exhibit a spectrum of ophthalmic manifestations including ptosis, strabismus, nystagmus, red-green color deficiency, and eyelid abnormalities [6,7]. Recent advances in ophthalmic imaging have allowed researchers to perform high-quality analysis of anterior segment structures with scheimpflug camera. Sirius Topography System (CSO, Italy) combines placido disk topography with rotating Scheimpflug tomography of the anterior segment and provides information on elevation, curvature and dioptric power of both corneal surfaces, and biometric measurements of the anterior chamber [8]. Additionally, it allows digital analysis of meibomian gland (MG) and tear film break-up time (TFBUT) with different camera attachments [9]. Previous studies demonstrating changes in MG structure and tear film homeostasis have focused primarily on adult population or have been limited by the sex and systemic conditions such as menopause, andropause and hormone replacement therapy (HRT) of these adult population [10–12]. As emphasised in DEWS II report, regulation of ocular surface and adnexal tissues have both been influenced by hormonal and age-related changes [2]. In recent studies, establishing a baseline for MG morphology and analysing anterior segment structures in children have gained popularity [13–15]. Therefore, to demonstrate the details of ocular surface, tear film and anterior segment structures in a pediatric population with sex steroid deficiency exhibit an importance as an identical condition with adult population. The purpose of the current report is to compare the qualitative and quantitative features of the MGs, TFBUT, dry eye symptoms, corneal and anterior segment measurements between healthy children without any reported symptoms and children with hypogonadism by a non-invasive, patient-friendly examination method for the first time.

(TSH), free T4 (fT4), follicle stimulating hormone (FSH), luteinizing hormone (LH), estradiol (E2), testosterone (T), gonadotropinreleasing hormone (GnRH), prolactin, IGF-1, adrenocorticotropic hormone (ACTH), cortisol were all collected in hypogonadism group. Bone age is more determinative than the chronological age in showing the physiological growth potential and pubertal developmental stages. In healthy children, bone age and chronological age are similar and the difference between them is not more than 12 months [16]. Children with hypogonadism were divided into four subgroups: Central Hypogonadism (n: 24), Primary Chromosomal Gonadal Failure (n: 12) that is further divided into two subgroups; Turner Syndrome (n: 9) and Klinefelter Syndrome (n: 3), Primary İdiopathic Gonadal Failure (n: 4). Genetic analysis of patients with Turner Syndrome revealed that 8 of 9 had a 45, X0 karyotype and the remaining 1 had mosaic karyotype. Genetic analysis of patients with Klinefelter Syndrome revealed that 2 of 3 had a 47, XXY karyotype and remaining 1 had mosaic pattern. Eighty-six eyes of 43 age- and sex-matched healthy subjects that presented to ophthalmology clinic for a routine ocular examination were assigned as control group. Each children underwent full ophthalmologic examination including refraction measurement (RK-F2, Canon, Japan), best-corrected Snellen visual acuity (20 feet) (BCVA), anterior and posterior segment examination with slit-lamp biomicroscopy, intraocular pressure (IOP) measurement (CT.1P, Topcon, Japan), and red and green color discrimination by Ishihara cards. Then, all the patients were transferred to the anterior segment examination room with a standard light, temperature and humidity values during examinations. Corneal topography, anterior segment analysis, non-invasive TFBUT (NITFBUT) evaluation and meibography were performed using a noncontact, non-invasive rotating Scheimpflug camera system (Sirius Topography, CSO, Italy) with different attachments. Three measurements were made per eye; the one with the best alignment and fixation was selected for data analysis. For topographic analysis, corneal thickness and refractive map indices were evaluated with the best sphere fit for each subject. Mean keratometry (Km) values for the both front and back surfaces of the cornea, maximum keratometry (Kmax) value, central (CCT), thinnest (TCT) and apical (ACT) corneal thicknesses, corneal volume (CV), horizontal visible iris diameter (HVID), anterior chamber volume (ACV), anterior chamber depth (ACD), and irido-corneal angle (ICA) values were collected. Non-invasive TFBUT was performed by Sirius in combination with the Phoenix-tear film imaging module. This innovative test performed through videokeratoscopy, the video recording of the Placido's disk ring projection on the corneal surface. Any distortion or interruption of the disk's reflected rings showing exactly when and where the break-up has occurred. The software processes the video in real time, and also performing the measurement without any need for the user to intervene, thus avoiding the detailed manual checking of the film. The patient asked to blink two times followed by continuation of openness of the eyelids as soon as possible during imaging. A maximum of 17 s the imaging continues based on the patient demand to eye blink or distortion of the rings occurred on the corneal surface. Consequently, first and average NITFBUT time has been acquired per image. Meibography was performed by Sirius using the instrument's infrared illumination module. Once the infrared images of the MGs have been obtained using the live acquisition software, they can be manually processed inside Phoenix program. Because the number of MGs in the upper eyelid was found to be greater in adolescents than in young children and the lower lid showed similarity between age groups, the lower eyelid was selected to evaluate meiboscale to avoid impact of age independent from the effect of the hormonal status [17]. The lower eyelid was everted and 5 images were captured to verify the best one with high quality and good focus. The silhouette of MGs through infrared illumination of the everted eyelid from the conjunctival side was appeared as hypoilluminant vertical clusters (Fig. 1a). Starting near the punctum, examiners marked the meibomian gland area of interest that

2. Patients and methods This cross-sectional study was performed from July 2018 to February 2019 at the department of ophthalmology, Yuksek Ihtisas Training and Research Hospital at Bursa, Turkey, and at the pediatric endocrinology and metabolism clinic of the same hospital. The study protocol was approved by the Institutional Review Board of the hospital (2011-KAEK-25 2018/07–24) in accordance with the tenets of the declaration of Helsinki. Written informed consent to participate in this study was obtained from the children and/or parents or legal guardians of the children after explaining the nature and purpose of the study. Exclusion criteria included blepharoconjunctivitis, any systemic or ocular disease requiring daily medication including allergic, autoimmune or dystrophic diseases, a history of eye surgery, high spherical (> -5.0 D or > +3 D) or cylindrical (> ± 2.0 D) refractive errors, contact lens wear, a history of reported or diagnosed dry eye disease that would influence MG architecture or interfere with tear film production and function. Patient's clinical chart review was performed to obtain relevant information regarding the patient's systemic hormonal status and ocular history. Children with age less than 5 years and higher than 18 years, and disability to cooperate with image acquisition were also excluded. A total of 80 eyes of 40 Caucasian children with different types of hypogonadism that were consecutively referred to ophthalmology clinic for the purpose of eye examination were enrolled in this study. Chronological age, bone age, body weight, body weight standard deviation score (SDS), height, height SDS, body mass index, body mass index SDS, tanner stages, and the levels of thyroid-stimulating hormone 2

The Ocular Surface xxx (xxxx) xxx–xxx

G. Dereli Can and Ö. Kara

Fig. 1. Defining meiboscale in Sirius Topography with meibography attachment: a) silhouette of meibomian glands through infrared illumination of the everted lower eyelid from the conjunctival side was appeared as hypoilluminant vertical clusters. b) starting near the punctum, examiners marked the meibomian gland area of interest that build a trapezoidal shape. Moving each blue or yellow points allows to adjust the shape to trace the interested area. c) The meibomian glands that were not visualized (red) were indicated as “MG loss”. See the percentage of meibomian gland loss and meiboscale grade on the left upper side.

build a trapezoidal shape. Moving each blue or yellow points allows to adjust the shape to trace the interested area (Fig. 1b). The MGs that were not visualized were indicated as “MG loss” (Fig. 1c). Meiboscale on the left side of the program used to describe the MG loss and the grades for each eyelid were divided as follows: grade 0 (no loss of MGs), grade 1(the affected area was ≤25%), grade 2 (the affected area was 26%–50%), grade 3 (the affected area was 51%–75%), and grade 4 (the affected area was > 75%) [18]. Subsequently, the morphology of MGs were classified as follows: Vertical (straight glandular duct), tortuous (folding glandular duct into a defined angle), overriding (crossing glandular duct), U-shaped (connecting the ends of two glandular ducts as a “U”), and hooked (bending of the proximal part of the duct as a hook) (Fig. 2). This morphological classification was adopted from the report by Zhao et al. [19]. Furthermore, MG distortion was defined as any duct distorted > 45°, and clinically scored using a grading scale modified from Arita et al.‘s [20] report as follows: grade 0 (no distortion), grade 1 (the affected area was < 1/3), grade 2 (the affected area was 1/3 to 2/3), grade 3 (the affected area was > 2/3). Clinical specular microscopy (NSP-9900, NonconRobo, Konan, Japan) was used to view non-invasively the image of the corneal endothelial cell layer in all subjects. The specular reflex occurred at a regular, smoothed surfaced interface of two refractive indices (corneal endothelium and aqueous humor) was used to analyse the endothelial cell density (CD), coefficient of variation (CoV) in cell size, percentage of hexagonal cells (HG). Three measurements per subject in the center of the cornea were performed and at least 100 contiguous cells were

analysed by the program. Finally, each subject (or parent of the patient) was asked to complete the Ocular Surface Disease Index (OSDI) questionnaire that has been broadly using to screen for dry eye symptoms. The error that can be seen widely in statistical analysis in ophthalmological studies is the independent evaluation of the information about two eyes belonging to the same individual. If the values are highly correlated, then methods of analysis in which each eye is considered as an independent random variable are not valid. For the analysis of ocular level parameters, both eyes were used (166 eyes of 83 patients) in this study. While analysing 166 eyes Generalized Linear Mixed Models (GLMM) were used for continuous variables. For GLMM, group variable set as fixed effect and side (left, right) was set as both random effect and repeated measures. The random effect covariance type selected as scaled identity, unstructured or variance component and the best model was selected according to Akaike Corrected Information Criteria and Bayesian Information Criteria (AIC, BIC). The statistical software function was GENLINMIXED. For the post estimation satterthwaite, approximation and robust estimation were used. Correction was not required for p value because all comparisons were made only with control group. For patient level analysis, the results were presented for categorical variables as numbers and percentages, for continuous variables as mean ± standard deviation (SD). While comparing groups, Kruskal Wallis test was used with Dunn's test as a post-hoc test. The statistical level of significance for all tests was considered when the p value ≤ 0.05. Statistical analysis was performed

Fig. 2. The morphology of meibomian glands: VERTICAL; straight glandular duct, TORTUOUS; folding glandular duct into a defined angle, OVERRIDING; crossing glandular duct, U-SHAPED; connecting the ends of two glandular ducts as a “U”, and HOOKED; bending of the proximal part of the duct as a hook. 3

The Ocular Surface xxx (xxxx) xxx–xxx

G. Dereli Can and Ö. Kara

Table 1 Demographic features, body measurements, and tanner stages of whole subjects. Groups

p

Control

CH

TS

KS

PIGF

Age (years) Body Weight (kg) Body Weight SDS Height (m) Height SDS Body Mass Index (kg/m2) Body Mass Index SDS

13.7 ± 4.0 47.7 ± 14.7 −0.1 ± 0.8 1.52 ± 0.2 −0.1 ± 0.8 19.9 ± 2.9 −0.1 ± 0.9

14.8 ± 3.3 56.2 ± 20.6 −0.3 ± 2.1 1.52 ± 0.2 −1.4 ± 1.5 23.3 ± 5.4 0.4 ± 2.0

11.8 ± 3.9 35.6 ± 13.6 −1.5 ± 1.6 1.29 ± 0.2 −3.0 ± 1.8 20.3 ± 4.0 0.3 ± 1.4

16.0 ± 1.0 51.7 ± 21.8 −1.9 ± 2.4 1.61 ± 0.1 −1.6 ± 1.2 19.2 ± 5.9 −1.5 ± 2.4

16.3 ± 1.5 61.1 ± 6.2 0.4 ± 0.9 1.60 ± 0.0 −0.5 ± 1.3 23.7 ± 1.8 0.8 ± 0.7

0.175 0.020 0.082 0.008 < 0.001 0.011 0.045

Gender Female Male

29(67.4) 14(32.6)

13(54.2) 11(45.8)

9(100.0) 0(0.0)

0(0.0) 3(100.0)

3(75.0) 1(25.0)

0.018

Tanner Stages Stage 1 Stage 2 Stage 3 Stage 4 Stage 5

10(23.3) 4(9.3) 3(7.0) 4(9.3) 22(51.2)

9(37.5) 8(33.3) 1(4.2) 4(16.7) 2(8.3)

6(66.7) 2(22.2) 1(11.1) 0(0.0) 0(0.0)

3(100.0) 0(0.0) 0(0.0) 0(0.0) 0(0.0)

1(25.0) 0(0.0) 0(0.0) 2(50.0) 1(25.0)

0.003

CH: Central Hypogonadism, TS: Turner Syndrome, KS: Klinefelter Syndrome, PIGF: Primary Isolated Gonadal Failure, SDS: standard deviation score. Results are shown as Mean ± SD or n(%).

using the IBM SPSS ver. 19 package program (IBM Corp. Released 2010. IBM SPSS Statistics for Windows, Version 19.0. Armonk, NY: IBM Corp.).

stage were summarized in Supplementary Table 1 [21,22]. In general ophthalmologic examination, spherical equivalence (−0.3 ± 0.8 vs −0.5 ± 1.0; p = 0.112), BCVA (0.9 ± 0.0 vs 1.0 ± 0.0; p = 0.999), anterior and posterior segment examination and IOP measurement (16.3 ± 0.5 vs 16.9 ± 0.4; p = 0.188) were not significantly different between hypogonadism and control groups, respectively. Two of the 9 patients of Turner Syndrome showed strabismus.

3. Results 3.1. Subject demographics and medical history

3.2. Corneal topography, anterior segment parameters and specular microscopy

Demographic features, body measurements, and tanner stages of whole subjects were summarized in Table 1. There was no statistically significant difference in age of the participants between groups (p > 0.05). There were 25 (62.5%) females and 15 (37.5%) males in children with hypogonadism, and 29 (67.4%) females and 14 (32.6%) males in control group (p = 0.637). Bone age, hormonal status and HRT time of children with hypogonadism were summarized in Table 2. Hormone replacement therapy was used in 21 of the 40 (52.5%) subjects in children with hypogonadism (the number of patients treated previously or currently under treatment), estradiol (Estrofem®, NovoNordisk) in females and testosterone (Sustanon®, Merck) in males. The normal ranges of specific hormones based on age, gender, and pubertal

A comparison of the corneal topographic and anterior chamber parameters and corneal endothelial cell analysis between groups appears in Table 3. The mean CCT and TCT values were significantly lower in children with hypogonadism than those of control (p = 0.012 and 0.08, respectively). There was a significant decrease in HVID in children with hypogonadism (p = 0.007). Back Km values were steeper in children with hypogonadism than healthy controls (p = 0.038). The mean ACT, ACV, ICA, and CV values were similar between groups (p > 0.05, for all). The mean CD, CoV and HG values were similar between hypogonadism and control groups (p > 0.05, for all). However, the mean specular pachymetry value was significantly lower in hypogonadism group (p = 0.02). The pachymetry values measured by Sirius topography were significantly thicker than those measured by specular microscopy (p < 0.001). A comparison was performed with regards to HRT in patients with hypogonadism (Table 4). There were insignificant differences in the mean values of CCT (543.2 ± 36.2 μm vs 550.9 ± 28.6 μm, p = 0.360), TCT (534.3 ± 45.7 μm vs 546.9 ± 29.5 μm, p = 0.216), and ACT (574.9 ± 57.1 μm vs 599.0 ± 57.2 μm, p = 0.064) between children that were not received treatment and children under treatment. Although there was no significant difference for specular CD exist between healthy controls and hypogonadism group, children were exposed to HRT showed significantly lower mean CD values than children that were not receive any hormonal treatment (p = 0.005). The correlation analysis of whole data was also showed in Supplementary Table 2.

Table 2 Bone age, hormonal status and hormone replacement therapy time of children with hypogonadism.

Bone Age (years) TSH (μU/mL) fT4 (ng/dL) FSH (mIU/mL) LH (mIU/mL) E2 (pg/mL) Testosteron (ng/dL) Peak LH (mIU/mL) Prolactin (ng/mL) IGF-1(ng/mL) ACTH (pg/mL) Cortizol (μg/dL) HRT Time (months)

Mean ± SD

Median (Min– Max)

12.3 ± 3.6 2.7 ± 1.3 1.2 ± 0.2 32.8 ± 46.8 9.0 ± 13.7 16.6 ± 9.7 48.6 ± 50.6 1.6 ± 1.3 9.0 ± 4.3 223.5 ± 223.2 22.6 ± 13.4 14.1 ± 6.7 9.9 ± 12.0

13.0(2.0–18.0) 2.6(0.9–6.3) 1.1(0.6–1.6) 2.6(0.0–183.0) 0.2(0.0–47.0) 12.0(5.0–46.0) 31.0(0.6–194.0) 1.4(0.1–4.7) 8.1(2.7–23.0) 113.0(44.0–980.0) 20.3(5.0–79.9) 12.0(3.0–28.8) 4.0(0.0–38.0)

3.3. Tear film analysis and OSDI score

SD: Standard deviation, Min: minimum, Max: maximum, TSH: thyroid-stimulating hormone, fT4: free T4, FSH: follicle stimulating hormone, LH: luteinizing hormone, E2: estradiol, GnRH: Gonadotropin Releasing Hormone, IGF-1: insulin-like growth factor-1, ACTH: adrenocorticotropic hormone, HRT: hormone replacement therapy.

The mean first and average NITFBUT of hypogonadism group were 10.3 ± 1.0 s and 10.9 ± 0.9 s, respectively. They were insignificantly longer in healthy controls (p > 0.05, for both). All subjects completed 4

The Ocular Surface xxx (xxxx) xxx–xxx

G. Dereli Can and Ö. Kara

Table 3 Comparison of the corneal topographic and anterior chamber parameters and corneal endothelial cell analysis between groups.

BCVA IOP (mmHg) OSDI NITFBUT First (sec) NITFBUT Average (sec) CCT (μm) TCT (μm) ACT (μm) HVID (mm) ACD (mm) ACV (mm3) ICA (0) CV (mm3) Front Km (D) Back Km (D) Apikal K (D) MG Loss (%) CD (cells/mm2) CoV (%) HG (%) Specular pachymetry (μm)

Control

Hypogonadism

p

1.0 ± 0.0 16.9 ± 0.4 22.6 ± 3.0 11.3 ± 0.7 11.9 ± 0.7 565.8 ± 5.3 562.0 ± 5.3 605.7 ± 8.4 12.2 ± 0.1 3.1 ± 0.0 162.3 ± 4.8 42.3 ± 0.8 59.9 ± 0.5 43.4 ± 0.2 −6.2 ± 0.0 45.1 ± 0.3 6.7 ± 1.1 3122 ± 37.0 43.5 ± 1.2 51.4 ± 1.4 533.4 ± 5.4

0.9 ± 0.0 16.3 ± 0.5 19.2 ± 4.2 10.3 ± 1.0 10.9 ± 0.9 547.3 ± 7.3 541.1 ± 7.8 587.8 ± 11.8 11.9 ± 0.1 3.1 ± 0.0 156.1 ± 6.8 41.7 ± 1.2 58.9 ± 0.8 43.5 ± 0.4 −6.3 ± 0.0 45.3 ± 0.0 15.4 ± 1.9 3186 ± 66.0 41.4 ± 1.7 52.6 ± 1.8 514.7 ± 8.3

0.999 0.188 0.422 0.318 0.285 0.012 0.008 0.133 0.007 0.766 0.357 0.550 0.200 0.895 0.038 0.999 < 0.001 0.332 0.223 0.525 0.028

Table 5 The meiboscale, tortuosity and morphology analysis of meibomian glands between groups. Hypogonadism Meiboscale %0 14(18.7) %0-25 43(57.3) %26-50 18(24.0) %51-75 0(0) %76-100 0(0) Meibomian Gland Morphology Vertical 32(42.7) Tortuous 23(30.7) Overriding 11(14.7) Hooked 2(2.7) U-Shaped 7(9.3) Meibomian Gland Tortuosity Grade No Distortion 32(42.7) < 1/3 area 20(26.7) 1/3–2/3 area 16(21.3) > 2/3 area 7(9.3)

Control

p

36(43.9) 42(51.2) 4(4.9) 0(0) 0(0)

< 0.001

39(47.6) 23(28.0) 8(9.8) 3(3.7) 9(11.0)

0.876

40(48.8) 23(28.0) 10(12.2) 9(11.0)

0.489

Results are shown as n(%).

the OSDI questionnaire by themselves and/or the aid of parents. The average OSDI score was 19.2 ± 4.2 and 22.6 ± 3.0 for children with hypogonadism and healthy group, respectively (p = 0.422). Table 3 summarized both the mean values of tear film analysis and OSDI scores. Comparison with regard to HRT showed a significant increase in the mean OSDI scores in hypogonadotropic children receiving HRT (Table 4). The correlations between tear film parameters and other results were summarized in Supplementary Table 2.

The values are represented as Mean ± SD. BCVA: Best Corrected Visual Acuity, IOP: Intra Ocular Pressure, OSDI: Ocular Surface Disease Index, NITFBUT: noninvasive tear film break-up time, ave: average, CCT: Central Corneal Thickness, TCT: Thinnest Corneal Thickness, ACT: Apical Corneal Thickness, HVID: Horizontal Visible Iris Diameter, ACD: Anterior Chamber Depth, ACV: Anterior Chamber Volume, ICA: Irido-corneal Angle, CV: Corneal Volume, Km: Mean Keratometry, MG: Meibomian Gland, CD: Cell Density, CoV: Coefficient of Variation, HG: Hegzagonality.

3.4. Meiboscale and morphology of MGs Table 4 Comparison with regards to hormone replacement therapy in patients with hypogonadism.

The mean MG loss values were 15.4 ± 1.9% and 6.7 ± 1.1% in hypogonadism and control groups, respectively (p < 0.001). The mean meiboscale values of hypogonadism and control groups were 1.0 ± 0.6 and 0.6 ± 0.5 (p < 0.001) and the mean tortuosity scores were 0.9 ± 1.0 and 0.8 ± 1.0 (p = 0.462) in hypogonadism and control groups, respectively. In total, 61 (81.3%) eyes had MG loss or part of tarsus without MG distribution (meiboscale≥1), and 43 (57.3%) eyes had evidence of any degree of MG tortuosity (tortuosity grade≥1) in hypogonadism group. Of all healthy controls, 46 eyes (56.1%) had MG loss or part of tarsus without MG distribution (meiboscale≥1) (p < 0.001), and 42 (51.2%) subjects had evidence of any degree of MG tortuosity (tortuosity grade≥1) (p = 0.489). Fig. 2 depicts five distinct shapes of MG ducts observed in whole experimental group. There was no statistically significant difference of the distrubition of MG morphology between groups (p = 0.876). The meiboscale, tortuosity and morphology analysis of MGs between groups were summarized in Table 5. The correlations between MG parameters and other results were summarized in Supplementary Table 2.

Hormone Replacement Therapy

BCVA IOP (mmHg) OSDI NITFBUT first (sec) NITFBUT average (sec) CCT (μm) TCT (μm) ACT (μm) HVID (mm) ACD (mm) ACV (mm3) ICA (0) CV (mm3) Front Km (D) Back Km (D) Apical K (D) MG Loss (%) CD (cells/mm2) CoV (%) HG (%) Specular pachymetry (μm)

–

+

p

1.0 ± 0.1 16.0 ± 2.0 11.8 ± 7.5 10.5 ± 5.5 11.1 ± 5.1 543.2 ± 36.2 534.3 ± 45.7 574.9 ± 57.1 12.0 ± 0.5 3.1 ± 0.3 151.9 ± 34.0 41.1 ± 7.3 57.9 ± 4.3 43.1 ± 2.0 −6.2 ± 0.3 44.9 ± 2.3 18.0 ± 12.9 3323.8 ± 412.7 40.4 ± 7.5 54.4 ± 7.0 513.7 ± 48.1

1.0 ± 0.0 16.4 ± 2.3 25.7 ± 19.0 10.2 ± 5.8 10.8 ± 5.4 550.9 ± 28.6 546.9 ± 29.5 599.0 ± 57.2 12.0 ± 0.5 3.2 ± 0.2 159.6 ± 26.9 42.2 ± 7.0 59.7 ± 3.9 43.8 ± 1.7 −6.4 ± 0.3 45.7 ± 2.2 13.3 ± 9.8 3063.0 ± 279.4 42.3 ± 10.0 51.0 ± 10.7 515.7 ± 38.1

0.262 0.403 0.003 0.868 0.881 0.360 0.216 0.064 0.999 0.067 0.194 0.460 0.107 0.030 0.027 0.125 0.082 0.005 0.478 0.092 0.509

3.5. Ocular examinations based on gender When the comparison analysis was made separately on female and male genders, only the MG loss and meiboscale values and specular CD values showed statistically significant changes between hypogonadism and healthy controls in both gender. However, corneal thickness values, HVID, and back Km values showed significant differences between hypogonadism and healthy controls only in female genders. Table 6 shows comparison of ocular examinations between hypogonadism and control groups for both male and female gender.

The values are represented as Mean ± SD. BCVA: Best Corrected Visual Acuity, IOP: Intra Ocular Pressure, OSDI: Ocular Surface Disease Index, NITFBUT: Noninvasive Tear Film Break-up Time, ave: average, CCT: Central Corneal Thickness, TCT: Thinnest Corneal Thickness, ACT: Apical Corneal Thickness, HVID: Horizontal Visible Iris Diameter, ACD: Anterior Chamber Depth, ACV: Anterior Chamber Volume, ICA: Irido-corneal Angle, CV: Corneal Volume, Km: Mean Keratometry, MG: Meibomian Gland, CD: Cell Density, CoV: Coefficient of Variation, HG: Hegzagonality.

5

The Ocular Surface xxx (xxxx) xxx–xxx

G. Dereli Can and Ö. Kara

Table 6 Comparison of ocular examinations between hypogonadism and control groups in male and female gender. Male

BCVA IOP (mmHg) OSDI NITFBUT first (sec) NITFBUT ave (sec) CCT (μm) TCT (μm) ACT (μm) HVID (mm) ACD (mm) ACV (mm3) ICA (0) CV (mm3) Front Km (D) Back Km (D) Apikal K (D) MG Loss (%) CD (cells/mm2) CoV (%) HG (%) Specular pachymetry (μm)

Female

Control

Hypogonadism

p

Control

Hypogonadism

p

1.0 ± 0.0 16.5 ± 0.6 21.1 ± 4.9 13.2 ± 1.0 13.6 ± 0.9 562.3 ± 12.1 558.3 ± 12.2 602.5 ± 18.3 12.3 ± 0.1 3.2 ± 0.1 173.2 ± 10.5 44.0 ± 1.8 59.5 ± 1.2 42.8 ± 0.4 −6.1 ± 0.0 44.3 ± 0.5 4.9 ± 1.5 2980 ± 42.0 40.7 ± 2.1 55.3 ± 2.8 532.8 ± 12.5

0.99 ± 0.0 16.1 ± 0.7 25.9 ± 8.2 10.6 ± 1.5 10.9 ± 1.4 539.7 ± 14.9 531.8 ± 10.5 592.0 ± 22.5 12.2 ± 0.2 3.2 ± 0.1 165.8 ± 13.5 44.1 ± 2.1 57.1 ± 1.5 42.9 ± 0.4 −6.2 ± 0.1 44.9 ± 0.7 12.3 ± 2.8 3206 ± 78.0 37.8 ± 3.0 56.5 ± 3.7 513.7 ± 16.3

0.305 0.558 0.559 0.106 0.079 0.136 0.107 0.743 0.592 0.577 0.585 0.936 0.248 0.877 0.248 0.416 0.013 0.013 0.328 0.749 0.247

1.0 ± 0.0 17.1 ± 0.5 23.1 ± 3.7 10.5 ± 0.9 11.2 ± 0.8 568.1 ± 4.8 563.7 ± 5.2 607.8 ± 8.6 12.2 ± 0.1 3.1 ± 0.0 157.1 ± 4.9 41.7 ± 0.8 60.2 ± 0.5 43.7 ± 0.3 −6.2 ± 0.0 45.5 ± 0.3 7.5 ± 1.4 3082 ± 38.0 44.6 ± 1.4 49.7 ± 1.3 532.8 ± 5.3

0.99 ± 0.0 16.3 ± 0.7 16.8 ± 4.8 10.1 ± 1.3 10.9 ± 1.2 551.3 ± 7.5 546.9 ± 8.1 586.1 ± 13.8 11.8 ± 0.1 3.1 ± 0.0 150.0 ± 7.1 40.4 ± 1.5 59.8 ± 0.9 43.9 ± 0.5 −6.4 ± 0.0 45.6 ± 0.6 17.7 ± 2.4 3311 ± 83.0 43.4 ± 1.9 50.1 ± 1.8 515.4 ± 9.7

0.147 0.283 0.189 0.721 0.772 0.030 0.040 0.118 0.035 0.665 0.316 0.391 0.634 0.723 0.044 0.818 < 0.001 0.007 0.538 0.842 0.206

The values are represented as Mean ± SD. BCVA: Best Corrected Visual Acuity, IOP: Intra Ocular Pressure, OSDI: Ocular Surface Disease Index, NITFBUT: Noninvasive Tear Film Break-up Time, ave: average, CCT: Central Corneal Thickness, TCT: Thinnest Corneal Thickness, ACT: Apical Corneal Thickness, HVID: Horizontal Visible Iris Diameter, ACD: Anterior Chamber Depth, ACV: Anterior Chamber Volume, ICA: Irido-corneal Angle, CV: Corneal Volume, Km: Mean Keratometry, MG: Meibomian Gland, CD: Cell Density, CoV: Coefficient of Variation, HG: Hegzagonality.

treatment. Moreover, the correlation analysis indicated that a mild-tomoderate positive correlation was exist between IGF-1 values and ACT and CV values. Growth hormone (GH) is seemed to be an important endocrine stimulus for somatic growth of the organism through the production of IGF-1. Animal studies showed that GH, IGF-1, and recombinant growth factor treatment implement a stimulative impact on ocular development via the synthesis of the extracellular matrix [23]. However, there have been contradictory results for corneal thickness values in patients with pituitary GH excess or GH deficiency in the literature. Parentin et al. [24] showed greater CCT values in patients with GH deficiency and concluded that a greater CCT can represent a sign of a delayed eye growth caused by the lack of possible remodeling and stretching of collagen fibres of sclera and cornea in GH deficiency group. On the other hand, Ciresi et al. [25] demonstrated significantly greater CCT values in acromegalic patients than healthy controls. Although GH deficiency is not a feature any type of hypogonadism, the serum IGF-1 analysis showed lesser values in some central hypogonadotropic hypogonadism patients in our study. This result aided the interpretation of the literature and debate of the GH influence on corneal thickness. Another standpoint for the evaluation of corneal changes is sex steroids. A report by Keskin et al. [26] showed a significant decrease (approximately 40 μm) in mean CCT value and a linear correlation between CCT and serum E2 levels of postmenopausal women. Similarly, Giuffre et al. [27] advocated that E2 may have a role in corneal physiology and showed changes in the CCT during various phases of the menstrual cycle. On the other hand, Gokce et al. [28] reported that serum testosterone levels or androgen replacement therapy have no short-term effects on corneal structure in patients with idiopathic hypogonadotropic hypogonadism (IHH). Our results supported the previous knowledge. The comparison of CCT and TCT values between healthy and hypogonadotropic patients showed significant changes in female children, whereas such a relationship was not found in male children. Additionally, mean CCT, TCT and ACT values were insignificantly higher in hypogonadotropic children that were under treatment than hypogonadotropic children that were not received HRT. Furthermore, though statistically insignificant, difference in mean IOP values between healthy and hypogonadotropic patients was higher in female group. According as, although statistically insignificant, the relationship between lower serum E2 level, corneal

4. Discussion Overall, sex, gender and hormones play an important role in the regulation of ocular surface and adnexal tissues, and also appear to induce differences in tear film parameters and MG function between women and men. The subcommittee report of TFOS DEWS II critiqued the nature of this role to promote future research on interrelationships between sex, gender, hormones and ocular surface characteristics. They advised that such studies need to determine whether the sex difference in ocular surface parameters especially for tear film parameters lessens with more advanced age and becoming more similar among women and men. This report clearly demonstrated that ocular surface parameters may be influenced by both the body metabolism and aging [2]. Therefore, to elucidate the fundamental role of endocrine system and sex steroids on tear film parameters, MG morphology and function, and corneal and anterior segment parameters, one should evaluate these parameters in a pediatric population with hormonal changes. To date, the anterior segment evaluation of patients with specific chromosomal abnormalities such as Turner Syndrome and Klinefelter Syndrome has almost been demonstrated in the literature, however TFBUT and MG parameters have not been investigated yet. To the best of our knowledge, this is the first study evaluating the anterior segment parameters, tear film parameters, dry eye symptoms and MG morphology in a pediatric population with low E2 and testosterone levels. The current report will therefore provide a contribution to previous studies evaluating the impact of sex hormone differences on MGs in adult population and studies evaluating morphology of MGs in healthy pediatric patients. The results of this study showed that CCT and TCT values were approximately 20 μm lower on average in patients with hypogonadism than in healthy subjects; IOP values were also lower but the difference was not statistically significant. When the IGF-1 values were evaluated it has been understood that it was lower in some children with central hypogonadism than in age- and sex-matched healthy controls. Additionally, the comparison based on the HRT in hypogonadotrophic pediatric group showed that children receiving therapy had insignificantly greater CCT values than age- and sex-matched hypogonadotrophic children that were not under 6

The Ocular Surface xxx (xxxx) xxx–xxx

G. Dereli Can and Ö. Kara

thickness and IOP measurement should be considered as a potential risk factor upon accurate IOP readings. Also, corneal thickening in HRT group in our study was compatible with the previous reports in the literature. In addition to corneal thickness changes in our study, Km values for the back of the corneal surface was significantly steeper in hypogonadotropic children than healthy controls. Similarly to CCT and TCT values, the gender-based comparison showed this difference was mainly arised from the female gender that could be interpreted as an effect of E2 on the keratometric values. It has been shown that the frequency of keratoconus is higher especially in Turner Syndrome many times in the literature [29]. Scheimpflug based imaging can provide a measurement of not only the anterior corneal surface but also the posterior surface and characterize corneal architecture in three dimensions which is important for detection of forme fruste keratoconus. Our results will contribute the information on corneal biomechanical and structural changes in a pediatric age group with low sex steroid levels in which the keratometric values have an important role for visual acuity and keratoconus screening. Although the relationship between HRT and change in keratometric values for both front and back surfaces of the cornea could not interpreted from the current literature, the results of the current report prove that a consideration should be keep in mind while evaluating children with low sex steroid levels and those under HRT. Another interesting finding in our study was children who were exposed to HRT showed significantly lower mean CD values than children that were not receive any hormonal treatment in the hypogonadism group. Up to the present, the effect of HRT on corneal endothelium has only been reported by Gokce et al. [28]. They showed that androgen replacement therapy does not significantly affect CCT, IOP and corneal endothelial cell morphology and density measured by confocal microscopy at 3-month follow-up in men with IHH. Although several studies have showed the effect of hormonal status and HRT (estrogen, medroxyprogesterone acetate, and testosterone) on both corneal epithelium and thickness, to date no study has demonstrated how this therapy affects CCT [30–32]. As a matter of fact, corneal endothelial cell analysis is crucial for corneal transparency and thickness assessment, therefore it is rational to take into consideration whether HRT may affect or be a possible cytotoxic agent to corneal endothelial cells. An importance of our study comes to fore at this point as it evaluated both the corneal topographic and endothelial features simultaneously. In the current report, we have found significantly lower CD values in addition to insignificantly higher corneal thickness values. Previous studies showed that the HRT can individually affect the IOP measurement, however the mechanism is not clear [33]. Briefly, significantly lower CD, insignificantly higher CCT and IOP results obtained in the current report can be attributable to initial corneal endothelial cell toxicity as a precursor of subsequent alterations in the anterior segment. However, it has also been shown that CoV and HG, rather than CD, are the important biomarkers of corneal endothelial stress and they were found normal in our study [34]. Therefore, further data are still needed to establish the clinical significance of HRT on corneal endothelium, corneal thickness and IOP. The other significantly important data was lower HVID measurements for hypogonadotropic children. The sex-based subgroup analysis indicated that the difference between healthy and hypogonadotropic groups came principally from the female gender. It has been shown that corneal diameters might be affected by the body and facial measurements in the literature. Iyamu et al. [35] found that males had significantly wider HVID and they interpreted this data with the common fact that men are generally taller and have correspondingly larger eyes than women. Based on these data, the smaller measurements of HVID in female gender with low estrogen levels of the current report can be linked to the combined effect of gender and lower sex steroid levels. To date, the anterior segment evaluation of patients with specific chromosomal abnormalities such as Turner Syndrome and Klinefelter Syndrome has almost been demonstrated in the literature, however

TFBUT and MG parameters have not been investigated yet. The high prevalence of eye abnormalities in such syndromic hypogonadism patients may be explained by the similar genetic pathway for ocular and gonadal development [36]. As distinct from genetic theory, sex steroid levels are also important for the coordinated ocular surface system comprising the lacrimal glands, MGs, corneal and conjunctival epithelium, goblet cells and regulating by both nervous and endocrine systems. Meibomian gland dysfunction is characterized by duct obstruction and/or changes in the glandular secretion that is further concluded as gland atrophy. A prevalence study by Gupta et al. [14] showed that approximately 42% of the asymptomatic pediatric population had some evidence of MG changes, with the majority of these patients having mild MG atrophy (meiboscale = 1). Gupta et al. [14] also represented that the process of MG atrophy starts early in life course and not all patients with MG atrophy develop MG dysfunction and dry eye disease. On the other hand, Zhao et al. [19] advocated that it is difficult to determine whether MG deficiency is congenital or acquired, therefore they proposed the term “MG deficiency” instead of “MG atrophy” that implies a condition represented by shortened glands and trailing “empty” space and occurs over time. The results of the study by Zhao et al. [19] revealed that approximately 45.5% of the asymptomatic pediatric population had 20–30% MG deficiency (grade 1) and this grade was linked to the electronic device use and environmental aggravation. Similarly Wu et al. [17] emphasised “Congenital MG dropout” as a physiological phenomenon that is mainly found at the fornix of the eyelids, especially in the nasal and temporal regions and the mechanism of dropout (glandular branching disorder and complete loss from orifice to fornix) differs from the age-related MG atrophy (loss of meibocyte progenitors). Wu et al. [17] compared differences in MG morphology between children and adolescents and reported 28% patients exhibited MG loss. They also showed significant morphological changes in the upper eyelid of adolescents. Similar to previous reports, our results revealed that 56.1% of the healthy controls had MG loss or part of tarsus without MG distribution (meiboscale≥1). However, this ratio significantly increased to 81.3% in patients with hypogonadism. The analysis of NITFBUT and OSDI scores showed similar results between patients with hypogonadism and healthy controls. Already, biomicroscopic anterior segment evaluation did not show any considerable finding for dry eye disease or ocular surface deterioration. Discrepancies between the meiboscale and OSDI score in our study are not necessarily unexpected because children are not always reliable self-reporters of their symptoms [37]. However, NITFBUT should be considered as an objective non-invasive assessment for tear film and ocular surface health which was found normal in the current report. We found an obvious MG loss in children with hypogonadism but without an adequate relationship to tear film parameters. This can be attributed to lack of compounding factors, such as continued digital device use (not examined in the current report) and systemic disease that increase the risk for developing dry eye disease. On contrary to MG deficiency results, tortuosity and morphology analysis between healthy and hypogonadotropic children did not show significant differences between groups. This indicates that sex steroid may have a minor role on the qualitative features of MGs as opposed to the quantitative features of MGs such as loss or deficiency. However it is not clear whether deficiency in sex steroids affect the MGs in utero development or in the postpartum period. The weakness of this study includes the lack of investigation on daily life activities especially electronic device use and personal preferences. Relatively small sample size seems to be another limitation of the study. As indicated in the statistical analysis section, both eyes of patients were included in the study basing on the GLLM analysis. However, the subgroups of hypogonadism included relatively small sample size that prevent subgroup comparisons. Other possible debatable parameter is that the evaluation of MG characteristics is subjective. It is essential to establish an objective analysing system to evaluate the MG morphology and distortion. 7

The Ocular Surface xxx (xxxx) xxx–xxx

G. Dereli Can and Ö. Kara

5. Conclusion

atrophy in a pediatric population. Cornea 2018;37:426. [15] Mizoguchi T, Arita R, Fukuoka S, Morishige N. Morphology and function of meibomian glands and other tear film parameters in junior high school students. Cornea 2017;36:922. [16] Dahlberg PS, Mosdøl A, Ding Y, Bleka Ø, Rolseth V, Straumann GH, et al. A systematic review of the agreement between chronological age and skeletal age based on the Greulich and Pyle atlas. Eur Radiol 2018:1–13. [17] Wu Y, Li H, Tang Y, Yan X. Morphological evaluation of meibomian glands in children and adolescents using noncontact infrared meibography. J Pediatr Ophthalmol Strabismus 2017;54:78–83. [18] Pult H, Riede-Pult B. Comparison of subjective grading and objective assessment in meibography. Contact Lens Anterior Eye 2013;36:22–7. [19] Zhao Y, Chen S, Wang S, Chen Y, Li J, Fu Y, et al. The significance of meibomian gland changes in asymptomatic children. Ocul Surf 2018;16:301–5. [20] Arita R, Itoh K, Maeda S, Maeda K, Tomidokoro A, Amano S. Association of contact lens-related allergic conjunctivitis with changes in the morphology of meibomian glands. Jpn J Ophthalmol 2012;56:14–9. [21] Tetlow LJ, Clayton PE. 25 Tests and normal values in pediatric endocrinology. Charles GD Brook; 2005. p. 523. [22] Bereket A, Turan S, Omar A, Berber M, Ozen A, Akbenlioglu C, et al. Serum IGF-I and IGFBP-3 levels of Turkish children during childhood and adolescence: establishment of reference ranges with emphasis on puberty. Horm Res Paediatr 2006;65:96–105. [23] Burren CP, Berka JL, Edmondson SR, Werther GA, Batch JA. Localization of mRNAs for insulin-like growth factor-I (IGF-I), IGF-I receptor, and IGF binding proteins in rat eye. Investig Ophthalmol Vis Sci 1996;37:1459–68. [24] Parentin F, Pensiero S. Central corneal thickness in children with growth hormone deficiency. Acta Ophthalmol 2010;88:692–4. [25] Ciresi A, Morreale R, Radellini S, Cillino S, Giordano C. Corneal thickness in children with growth hormone deficiency: the effect of GH treatment. Growth Hormone IGF Res 2014;24:150–4. [26] Keskin N, Cantürk S, Aydin S, Saygili H, Özgün C. An objectıve method to determıne corneal changes durıng menopause. Clin Exp Obstet Gynecol 2009;36:176. [27] Giuffre G, Di Rosa L, Fiorino F, Bubella DM, Lodato G. Variations in central corneal thickness during the menstrual cycle in women. Cornea 2007;26:144–6. [28] Gokce G, Hurmeric V, Mumcuoglu T, Ozge G, Basaran Y, Unal HU, et al. Effects of androgen replacement therapy on cornea and tear function in men with idiopathic hypogonadotropic hypogonadism. PGM (Postgrad Med) 2015;127:376–80. [29] Macsai M, Maguen E, Nucci P. Keratoconus and Turner's syndrome. 1997. [30] Kilavuzoglu AEB, Cosar CB, Bildirici I, Cetin O, Ozbasli E. Estrogen-and progesterone-induced variation in corneal parameters according to hormonal status. Eye Contact Lens 2018;44:S179–84. [31] Affinito P, Sardo ADS, Di Carlo C, Sammartino A, Tommaselli GA, Bifulco G, et al. Effects of hormone replacement therapy on ocular function in postmenopause. Menopause 2003;10:482–7. [32] Sorrentino C, Affinito P, Mattace FR, Loffredo M, Merlino P, Loffredo A, et al. Effect of hormone replacement therapy on postmenopausal ocular function. Minerva Ginecol 1998;50:19–24. [33] Vajaranant TS, Maki PM, Pasquale LR, Lee A, Kim H, Haan MN. Effects of hormone therapy on intraocular pressure: the women's health initiative-sight exam study. Am J Ophthalmol 2016;165:115–24. [34] Bahar I, Kaiserman I, Lange AP, Slomovic A, Levinger E, Sansanayudh W, et al. The effect of mitomycin C on corneal endothelium in pterygium surgery. Am J Ophthalmol 2009;147. 447-52. e1. [35] Iyamu E, Osuobeni E. Age, gender, corneal diameter, corneal curvature and central corneal thickness in Nigerians with normal intra ocular pressure. J Optom 2012;5:87–97. [36] Ranke MB, Saenger P. Turner's syndrome. The Lancet 2001;358:309–14. [37] Han SB, Yang HK, Hyon JY, Hwang J-M. Children with dry eye type conditions may report less severe symptoms than adult patients. Graefe’s Arch Clin Exp Ophthalmol 2013;251:791–6.

As a conclusion, results lead to an idea that MG loss is a physiological process that is prominent in the condition of sex steroid deficiency, but does not cause tear film alterations. The mechanism of CCT changes and relationship between sex steroids, HRT and corneal endothelial alterations should be annotated. Still, future studies investigating sex and gender effect on the ocular surface system in an agebased fashion are required to clearly communicate influences in the arenas of ocular surface research. Disclosure statement The authors have no commercial or proprietary interest in any concept or product described in this article. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.jtos.2019.09.001. References [1] Jacobson L. Hypothalamic–pituitary–adrenocortical axis regulation. Endocrinol Metabol Clin 2005;34:271–92. [2] Sullivan DA, Rocha EM, Aragona P, Clayton JA, Ding J, Golebiowski B, et al. TFOS DEWS II sex, gender, and hormones report. Ocul Surf 2017;15:284–333. [3] Viswanathan V, Eugster EA. Etiology and treatment of hypogonadism in adolescents. Pediatr Clin 2011;58:1181–200. [4] Howard SR, Dunkel L. Management of hypogonadism from birth to adolescence. Best Pract Res Clin Endocrinol Metabol 2018;32:355–72. [5] Tyler C, Edman JC. Down syndrome, Turner syndrome, and Klinefelter syndrome: primary care throughout the life span. Prim Care Clin Off Pract 2004;31:627–48. [6] Denniston A, Butler L. Ophthalmic features of Turner's syndrome. Eye 2004;18:680. [7] Juhn AT, Nabi NU, Levin AV. Ocular anomalies in an infant with Klinefelter syndrome. Ophthalmic Genet 2012;33:232–4. [8] Savini G, Barboni P, Carbonelli M, Hoffer KJ. Repeatability of automatic measurements by a new Scheimpflug camera combined with Placido topography. J Cataract Refract Surg 2011;37:1809–16. [9] Dogan AS, Kosker M, Arslan N, Gurdal C. Interexaminer reliability of meibography: upper or lower eyelid? Eye Contact Lens 2018;44:113–7. [10] Schaumberg DA, Nichols JJ, Papas EB, Tong L, Uchino M, Nichols KK. The international workshop on meibomian gland dysfunction: report of the subcommittee on the epidemiology of, and associated risk factors for, MGD. Investig Ophthalmol Vis Sci 2011;52:1994–2005. [11] Alghamdi YA, Mercado C, McClellan AL, Batawi H, Karp CL, Galor A. The epidemiology of meibomian gland dysfunction in an elderly population. Cornea 2016;35:731. [12] Chhadva P, Goldhardt R, Galor A. Meibomian gland disease: the role of gland dysfunction in dry eye disease. Ophthalmology 2017;124:S20–6. [13] Shirakawa R, Arita R, Amano S. Meibomian gland morphology in Japanese infants, children, and adults observed using a mobile pen-shaped infrared meibography device. Am J Ophthalmol 2013;155:1099–103. e1. [14] Gupta PK, Stevens MN, Kashyap N, Priestley Y. Prevalence of meibomian gland

8