In the recent years, numerous researchers studied the problem of formation of connective tissue in the eye [1–4]. However, there is one special vitreoretinal pathology, macular fibrosis, which causes significant deterioration and even irreversible loss of visual function in working age population.
Despite the considerable progress in understanding etiological, pathological and genetic mechanisms of formation of epiretinal membrane (ERM), some questions pertaining to the development idiopathic ERM remain unanswered [5]. The main factors contributing to the development of this pathology are: impaired biomechanical processes at the vitreoretinal interface, namely, posterior vitreous detachment (PVD); micropores in the internal limiting membrane (ILM); pathological changes of the macular microvasculature [6–8]. Regardless of the pathogenetic mechanism of ERM development, migration and proliferation of various cell types play a key role in its formation and progression, the cells being glial cells (Muller retinal cells, astrocytes and microglia), hyalocytes, macrophages, retinal pigment epithelium (RPE) and retinal surface fibroblasts [9, 10].

Influenced by cytokines and growth factors, these cells transdifferentiate into a phenotype similar to myofibroblasts [11]. With aseptic inflammation in the background, myofibroblasts undergo apoptosis [11, 12]. In fibrosis, myofibroblasts activate and, when persisting for a long period of time, cause excessive deposition of collagen followed by its remodeling [13]. Evidence obtained through ophthalmoscope examination [14] and with the help of a number of current fundus pathology investigation methods allows isolating three main stages (grades) of ERM development: stage 0 — cellophane maculopathy, stage 1 — crinkled cellophane maculopathy, and stage 2 — macular puckering. Depending on the stage, clinical manifestations of the disease vary from total lack of symptoms to a significant visual function deterioration [15]. To date, there are no safe and effective methods that allow slowing down cell proliferation and progression of fibrosis in its early stages.

Globally, medical researchers and practitioners successfully subject late-stage ERM to vitreoretinal surgery, having accumulated significant related experience. However, despite the positive surgery results, numerous histological studies have shown that ILM peeling damages Muller cells and compromises retina architectonics and biomechanical strength [16, 17]. As a result, complete restoration of vision after such surgery occurs in 5–25% of cases only [18]. Moreover, this kind of treatment is prescribed in case the clinical symptoms, i.e. visual function alteration and/or deterioration, are pronounced. Up to the present, the main tactic for managing ERM patients has been dynamic observation [19].
Thus, development of an effective and safe early-stage ERM treatment technique yielding stabilization and/or improvement of vision and functional indicators is an urgent problem.
Given the urgency, such a technique was developed in the Research Center for Ophthalmology of Pirogov Russian National Research Medical University. The technique combines grid laser photocoagulation and exposure to 577 nm subthreshold micropulse laser light (RF patent №. 2634684, 02.11.2017), the two techniques that differ in their mechanism of action [20]. This study aimed to evaluate efficacy and safety of the combined laser technique designed to treat early idiopathic ERM (stages 0–1).

METHODS

Ninety-two patients aged 64.7 ± 9.6 years (mean) participated in the clinical research. The inclusion criteria were: early idiopathic epiretinal membrane (stage 0–1) with concomitant lens pathology: pseudophakia or early cataract.
All patients were randomized into three groups depending on the treatment tactics. The treatment group included 32 patients (32 eyes); they underwent laser treatment following the combined technique developed at the Research Center for Ophthalmology of Pirogov Russian National Research Medical University. The comparison group consisted of 30 patients (30 eyes); their treatment was laser coagulation, "lattice" type. The control group included 30 patients (30 eyes); they received neither laser nor conservative treatment, and natural proliferation of their fibroses was under dynamic observation.
The new technique is a combination of grid laser photocoagulation and subthreshold micropulse laser irradiation [20]. We used IRIDEX IQ 577 retinal surgery laser (IRIDEX Corporation, Mountain View; USA) that works in continuous and micropulse modes.

Grid laser photocoagulation was the first stage of the treatment. We irradiated the entire surface of ERM except for the avascular zone; the wavelength was 577 nm, power — 50 mW, pulse duration — 0.05 s, spot diameter — 100 μm, distance between coagulates — 150 μm. After two weeks, ERM was exposed to subthreshold laser micropulses (second stage of the combined laser treatment technique); the wavelength was 577 nm, pulse packet duration — 30 ms, micropulse duration — 50 μs, pulse ratio — 4.7%, spot diameter — 100 μm, power — 50 mW (tab. 1). Since we registered no significant changes after the 4th session, treatment group patients had only 3 sessions of subthreshold micropulse laser treatment, each a month apart.
All patients underwent standard ophthalmic and special examinations: multispectral with various filters (Blue, Green, Infrared Reflectance, MultiColor), spectral optical coherent tomography (SOCT) «Spectralis OCT» (Heidelberg Engineering, Inc; Germany ) and microperimetry MAIA (CenterVue; Italy).
All participants were examined before treatment and 3 months, 6 months, 1 year, 2 years, 3 years, 4 years and 5 years after treatment. Treatment and comparison group patients also underwent examination after each stage of laser treatment. Central retinal sensitivity was used as the basis for clinical assessment of safety of laser treatment.
We assessed uncorrected visual acuity (UVA), best corrected visual acuity (BCVA), central retinal thickness (CRT), central retinal sensitivity (CRS). These indicators were subjected to normal distribution; methods of parametric statistics (paired sample t-test) were applied to the data acquired to compare benchmark figures and results of treatment at different observation timepoints. Single-factor analysis of variance (ANOVA) was used to assess the significance of differences in comparison of results from more than two independent groups. The differences were considered statistically significant at p < 0.05.

RESULTS

Results of initial examination of patients from the treatment group (32 eyes): UVA — 0.45 ± 0.31, BCVA — 0.9 ± 0.13, CRS (microperimetry data) — 26.3 ± 1.65 dB, CRT (OCT data, mean) — 282.8 ± 27.1 μm.
In the comparison group (30 eyes), the pre-surgery indicators were: UVA — 0.44 ± 0.26, BCVA (mean) — 0.86 ± 0.15, CRT (mean) — 26.3 ± 1.57, CRT (mean) — 292.4 ± 62.2 μm.
Initial indicators registered in the control group (30 eyes) were as follows: UVA (mean) — 0.64 ± 0.23, BCVA — 0.87 ± 0.14, CRS (mean, microperimetry data) — 27.1 ± 1.52 dB, CRT (mean) — 301.4 ± 44.8 μm.
tab. 2 presents analysis of clinical and functional results of treatment and dynamic observation of patients with idiopathic ERM.
In the control group, mean UVA and BCVA were decreasing significantly throughout the observation period (fig. 1, fig. 2).
According to the OCT data, mean CRT was increasing significantly from the 3rd month on (fig. 3), while mean CRS, on the contrary, was significantly increasing from the 3rd month on (fig. 4). Multispectral imaging revealed continued proliferation on the retinal surface (fig. 5A, B).
In the comparison group, the mean UVA was increasing significantly up to the 3rd month and decreasing from the 6th month on (fig. 1). Within the first 3 months of observation, mean BCVA changes were insignificant, but from the 12th month on this indicator was decreasing gradually, the difference being significant (fig. 2). The analysis of mean CRT did not reveal significant differences from the benchmark data within the first 6 months, but from the 12th month on, this indicator was increasing significantly (fig. 3) and ERM was developing (fig. 6A, B). Computer microperimetry did not register significant changes of mean CRS within the first 12 months, but from the 2nd to the 5th years it was decreasing, and these changes were significant (fig. 4).
Comparing the results registered in treatment, comparison and control groups we established that only patients of the treatment group enjoyed significant improvement of visual and functional indicators (UVA, BCVA and CRS) combined with decreasing CRT and idiopathic ERM involution, these results staying stable throughout observation (fig. 7A, B). In the comparison group, the indicators improved for a short period of time (up to the 3rd month), and then began to gradually deteriorate; in the control group, such deterioration, significant, was registered at all observation timepoints (fig. 1–fig. 4).
Laser treatment did not inflict biomechanical damage to the retina in treatment and comparison groups, as confirmed by the results of computer microperimetry and SOCT.

CONCLUSIONS

Assessed against the results registered in comparison and control groups, the suggested combined laser treatment technique applied in the treatment group proved to be highly effective in maintaining/improving visual functional indicators and stabilizing/improving morphofunctional indicators throughout the entire period of observation. As for the morphological and functional structures of sensory retina, the combined laser treatment technique is a safe option, which is proved by the retinal sensitivity improvements registered at the different timepoints of the observation period.