Striae Distensae

Experimenters scanned all the tissue slices with a fluorescent microscope to identify brain regions that expressed either MC3R or MC4R mRNA. Twenty-six brain regions were chosen based on findings in previous literature, functional sites of melanocortin action, and visual inspection of expression within tissue by the experimenter. The 26 brain regions were identified based on A Stereotaxic Atlas of the Golden Hamster Brain (Morin and Wood, 2001 ). Schematic diagrams from the hamster atlas were used at the level that corresponds to where the region of interest was imaged and overlayed with a color fill-in using BioRender to help illustrate the mRNA labeling within a region. The serial sections were taken at a specific location (matched across brains) within each brain region, as indicated by the atlas plates and represent a rostral-caudal distance of less than 0.5 mm. Anatomical abbreviations are as follows:

ARC	Arcuate nucleus of the hypothalamus
BIC	Nucleus of the brachium of the inferior colliculus
BNST	Bed nucleus of the stria terminalis
CPu medial, lateral, dorsal	Caudate-putamen medial, lateral, dorsal
DEn	Dorsal endopiriform nucleus
DRN	Dorsal raphe nucleus
HPC	Hippocampus (dorsal/ventral CA1, CA2)
IL	Infralimbic cortex
LHb	Lateral habenula
LS	Lateral septum
MePD	Medial amygdaloid nucleus, Posterodorsal part
MD	Mediodorsal thalamic nucleus
MS	Medial septum
MPOA	Medial preoptic area
NAc core	Nucleus accumbens core
NAc shell	Nucleus accumbens shell
OT	Olfactory tubercle
PAG	Periaqueductal gray
PrL	Prelimbic cortex
PVN	Periventricular hypothalamus
VMH	Ventromedial hypothalamus
VTA	Ventral tegmental area

All images were acquired on a Leica TCS SPE confocal microscope under the same scanning parameters. An image was collected in the left and right hemisphere from two tissue sections (in series) for all brain regions for each subject (resulting in a sample of four images for each brain region for each subject). Images were collected with a 20X/0.60 advanced correction system objective with a pixel distribution of 1,024 × 1,024 at a frequency of 8 kHz. A solid-state laser with 488 and 532 nm wavelengths and an ultra-high dynamic PMT detector was utilized to capture z-stack images with 1.5 μm spacing for a maximum of 15 steps. The pinhole size was 1 airy unit (AU), 2 frames were averaged, and optical zoom was 1.00X. 3D images produced were 550 × 550 × 14 μm.

We conducted this single-center prospective study in a tertiary ophthalmology center (Centre Hospitalier National d’Ophtalmologie des Quinze-Vingts, Paris, France) between May 2021 and July 2021. We included over two consecutive months all patients presenting to ophthalmology consultations in two specialized consultations and presenting with one of the following pathologies: keratoconus (KC group) and Ocular Surface Disease (OSD group). The exclusion criteria were as follows: age younger than 18 years, history of recent eye surgery (<3 months) and presence of a language barrier preventing reliable responses. Patients who incorrectly completed the questionnaire (e.g., missing page) were excluded retrospectively. The questionnaire was given to patients in consultation on a voluntary basis. Informed consent was obtained from the subjects after explanation of the nature of the study. The research followed the tenets of the Declaration of Helsinki.
The first part of the questionnaire included an interrogation concerning the patient’s ophthalmological, dermatological and psychiatric background and a self-evaluation of ophthalmological symptoms. The following data were collected: age, sex, working conditions, medical and surgical ophthalmological history, ophthalmological treatments, use of glasses and contact lenses. Next, we assessed the ocular symptoms experienced by the patients using an open-ended question. Symptomatology was then evaluated in terms of frequency and intensity using the SANDE (Symptom Assessment In Dry Eye) questionnaire, which estimates the frequency of occurrence of symptoms and their intensity [15 (link)]. Psychiatric history was sought: addictive disorder, personal psychiatric disorder or family history. We looked for the presence of dermatological history as well as the existence of skin pruritus and scratching.
The second part of the questionnaire dealt with eye rubbing. As for the ocular symptoms, we used the SANDE scale to evaluate the frequency and intensity of eye rubbing. We looked for daily or occasional occurrences as well as the period of day. We used the Goodman and DSM-5 diagnostic criteria for addiction to obtain a 16-item cognitive–behavioral assessment of eye rubbing [16 (link)]. We chose the limit of 5 positive responses as an indicator of addictive-like behavior. The CAGE questionnaire used to predict the existence of an alcohol use disorder was also submitted to the patient with four questions adapted to eye rubbing [17 (link)]. We considered the presence of 2 positive responses to the CAGE questionnaire as predictive of an addictive-like disorder related to eye rubbing. Finally, we asked patients if they intended to stop eye rubbing after completing the questionnaire. The primary endpoint was the assessment of Goodman criteria for patients with eye rubbing [16 (link)]. Secondary endpoints were the evaluation of CAGE score results, psychiatric and addictive comorbidities.
An ophthalmologic examination by a specialist was performed on the same day for each patient. The following data were collected for all patients: best corrected distance visual acuity (CDVA), intraocular pressure (IOP), presence of follicles, papillae or blepharitis, break-up time (BUT) and Oxford grading score (Oxf). For keratoconus, we evaluated the presence of Vogt’s striae and corneal opacities, the maximum keratometry (Kmax) and the thinnest pachymetry (measured on Pentacam, Oculus).
Statistical analysis of the data was performed with MedCalc^® software v20.218 (MedCalc Software Ltd.^®, Ostend, Belgium). Chi-square test, Student t-test, ANOVA test, and logistic regression were used. Variables were first studied in univariate analysis and then analyzed in multivariate regression. Regression analysis was performed using Spearman’s correlation coefficient. The threshold for statistical significance was set at p < 0.05 for the entire study. We separated patients into two groups: patients with (group R) and without (group NR) eye rubbing.

The diagnosis of age-related cataracts and the grading of lens nuclear hardness were determined by the same senior ophthalmologist under a slit lamp according to the Lens Opacities Classification System II criteria [16 (link)]. An experienced ophthalmologist performed all the operations with the same procedure and an Intrepid balanced tip in the CENTURION^® Vision System (Alcon Laboratories, Fort Worth, TX, USA). When the phacoemulsification was performed with the AFS, the target IOP was set at 50 mmHg. When the surgery was performed with the GFS, the bottle height was put at 90 cm. The vacuum level and aspiration flow rate were 450 mmHg and 45 cc/min, respectively, in both systems. Patients who participated in the AGSPC were grouped on the basis of the randomization results. Patients who did not agree to participate in the clinical trial, but were eligible for AGSPC, chose the surgical system according to their own desires. All patients were treated with the same prescription and viscoelastics (medical sodium hyaluronate gel, Iviz, Bausch + Lomb, Rochester, NY, USA) in the perioperative period. More details could be extracted from the protocol [15 (link)].
The prediction model was to predict the incidence of CE in patients on the first day after phacoemulsification. One ophthalmologist completed slit-lamp biomicroscopic and corneal endothelial microscopic examinations for all patients at one day postoperatively. Patients were asked about any discomfort at the same time. When the corneal stroma shows diffuse gray or milky clouding, CE is defined. However, if the opacity of the cornea is limited to the vicinity of the surgical incision and does not reach the visual axis area, the cornea is not classified as edema. Central corneal thickness (CCT) measured with a noncontact specular microscope (SP·3000P, Topcon, Tokyo, Japan) is another auxiliary indicator to assess CE. When CE is detected under a slit lamp, there is a notable increase in CCT compared to the preoperative period, which is usually greater than 50 μm. Nevertheless, there was a subset of patients who did not have obvious diffuse corneal clouding despite a significant increase in CCT. This may have correlated with the patient’s different preoperative CCT, so observation under a slit lamp was adopted as the main method of diagnosing CE. Some patients postoperatively developed Descemet’s membrane striae, but did not show a significant CCT increase and opacity in the visual axis area. Therefore, these patients were considered to have only mild corneal complications that were not defined as CE.
We reviewed previously published studies that focused on risk factors for the development of CE after phacoemulsification, and referenced potential risk factors therein [5 (link),17 (link),18 (link),19 (link)]. Combining the data collected under the AGSPC protocol, we identified 17 potential predictors to construct prediction models for CE after phacoemulsification: gender, age, hypertension (HBP), diabetes (Dia), best corrected visual acuity (BCVA), preoperative intraocular pressure (Pre IOP), nuclear hardness (NH), axial length (AL), anterior chamber depth (ACD), lens thickness (LT), CCT, ECD, fluidics, cumulative dissipated energy (CDE), ultrasound time (U/S time), estimated fluid used (EFU), and total aspiration time (TAT).