Minimally invasive laser treatment combined with intravitreal injection of anti-vascular endothelial growth factor for diabetic macular oedema

The purpose of this study was to investigate the effect of the combination of minimally invasive laser treatment to the intravitreal injection of anti-vascular endothelial growth factor (VEGF) for diabetic macular oedema (DME). This study was retrospective longitudinal study of thirty-four eyes of 31 patients with DME. Either once or several times of intravitreal anti-VEGF injection was followed by the single minimally invasive laser within a month. The mean best corrected visual acuity (VA) and the central macular thickness (CMT) were measured before treatment, 1, 3, 6 and 12 months after the first anti-VEGF injection. The mean logMAR VA had improved from 0.52 ± 0.34 at baseline to 0.44 ± 0.32 (p = 0.003), 0.40 ± 0.34 (p = 0.006), 0.43 ± 0.33 (p = 0.063), and 0.41 ± 0.34 (p = 0.009), at 1, 3, 6, and 12 months after treatment, respectively. The mean CMT decreased significantly by 1 month and maintained over 12 months (491.1 ± 133.9 µm at baseline, 396.6 ± 116.8 µm (p = 0.001), 385.2 ± 156.2 µm (p = 0.002), 336.5 ± 86.3 µm (p = 0.000), and 354.8 ± 120.4 µm (p = 0.000) at 1, 3, 6, and 12 months, respectively). The average number of the anti-VEGF injection in 1 year was 3.6 ± 2.1 in all patients. The combined intravitreal anti-VEGF and minimally invasive laser therapy improves the VA, alleviates DME, and may decrease the required number of anti-VEGF injections.


Materials and Methods
This was a retrospective study based on a review of medical records. The study protocol followed the tenets of the Declaration of Helsinki and was approved by the ethics committee of St. Luke's International University. Informed consent for the analysis and use of treatment results was obtained from all patients at the time of the first visit.
In total, 34 eyes of 31 patients with DME-defined as clinically significant macular oedema according to the ETDRS criteria accompanied by diffuse and/or localized leakage from microaneurysms on fluorescein angiography-who have received intravitreal anti-VEGF injections and a minimally invasive laser therapy at St. Luke's International Hospital between March in 2014 and September in 2015 were involved in this study. Eyes with the pre-treatment visual acuity ≤0.1 in logMAR (≥0.8 in decimal) or dense cataract, the eyes that had underwent intraocular surgeries, such as cataract surgery and vitreous surgery, or other treatments for macular oedema, such as intravitreal anti-VEGF injection or sub-Tenon's capsule injection of triamcinolone, within the last 3 months were excluded. Among all eyes, 24 eyes of 22 patients with logMAR VA ≥ 0.3 (≤0.5 in decimal) were classified as a poor visual acuity group for a subgroup analysis.
General treatment protocol. All eyes initially received intravitreal anti-VEGF injection either once or multiple times on a monthly basis, followed by the single minimally invasive laser irradiation within a month after the latest injection. Initial series of injection before laser treatment was conducted until the macular oedema is disappeared on OCT. The further anti-VEGF treatment was determined with a Pro re nata (PRN) regimen under monthly control over 12-month follow up period after the first injection.
Retreatment criteria were based on reduction in BCVA or an increase of CMT with approximately 20% or more, or presence of sub macular fluid. Patients who did not wish the loading phase due to economic problems or patients with mild macular oedema (less than 350 µm) did not completed loading phase.
Minimally invasive laser therapy. The IQ577 (577 nm, OcuLight SLx, Iridex; Mountain View, CA, USA) and the PASCAL Streamline Yellow (xxx nm, Topcon Medical Laser Systems, Inc.; Santa Clara, CA, USA) were used as a minimally invasive micropulse laser therapy and a minimally invasive continuous wave laser endpoint management, respectively.
Minimally invasive micropulse laser therapy with IQ577. Before the treatment irradiations, test irradiations were performed at a non-oedematous area outside the vascular arcade to determine the threshold laser power for the retinal whitening, the power with which barely visible spots are observed approximately 3 s after irradiation. For that, the irradiation condition of the laser was chosen as follows; a spot diameter of 100 µm, total duration of 0.2 s, and 5% duty cycle. The irradiation power for treatment was set to 50% to 60% of the threshold value determined in test irradiations. The areas of oedema with a strong fluorescence leakage was irradiated with zero spacing. The used parameters for the titration were the power ranged from 850 to 1000 mW (t = 0.2 s, 5% duty cycle).
Direct micropulse coagulation for microaneurysms was performed using the IQ577, 15% duty cycle, 50-µm spot diameter, and 0.2-s duration. The leaking microaneurysms ≥500 µm away from the fovea defined with a fluorescein angiography was irradiated with a power which a hint of microaneurysm colour change was observed.
Minimally invasive continuous wave laser therapy with PASCAL Streamline Yellow (endpoint management). As in the micropulse laser, test irradiations were performed for the energy titration at a non-oedematous area outside the vascular arcade. The minimum power at which coagulation spots were slightly visible approximately 3 s after irradiation with an irradiation diameter of 200 µm and duration of 0.015 s was set as the threshold. The threshold ranged from approximately 100 to 130 mW under these conditions. Then, the endpoint management was set to 30%-50% and irradiation was performed using a macular grid pattern (donut pattern) with the spacing of 0.5. The areas that could not be covered by the donut pattern were treated with additional coagulation using a 2 × 2 pattern.
When a non-perfusion area was observed outside the archade vessels on fluorescein angiography, we performed targeted retinal photocoagulation with a pattern scan laser using the PASCAL Streamline Yellow (Topcon Medical Systems, Inc.).
Clinical examinations. The visual acuity (VA) and central macular thickness (CMT) were measured before treatment and 1, 3, 6, and 12 months after treatment. CMT was measured using optical coherence tomography (OCT). The decimal visual acuity was measured and converted to the logarithm of the minimum angle of resolution (logMAR) values for the analysis. www.nature.com/scientificreports www.nature.com/scientificreports/
Minimally invasive laser treatment. The IQ577 and PASCAL Streamline Yellow were used for 19 (55.9%) and 15 (44.1%) eyes, respectively. The laser models and treatments are shown in Table 2.
An increase of ≥0.3 logMAR at 12 months after the treatment was observed for seven of the 34 eyes (20.6%). The VA remained unchanged for 25 (73.5%) and deteriorated for two (5.9%) of the 34 eyes. Of the 24 eyes with poor baseline-VA, seven (29.1%) showed an increase of ≥0.3 logMAR, 15 (62.5%) showed no change, and two (8.3%) showed deterioration.

Comparison of the mean VA, CMt, and number of intravitreal anti-VeGF injections between
the IQ577 group and PASCAL group. Comparison of the treatment results between the IQ577 group and PASCAL group is shown in Table 4. There was no difference between the groups in mean VA, CMT, and number of anti-VEGF injections in a year. the CMT after combined laser and anti-VEGF therapy for patients with diabetic macular oedema: (a) VA was significantly improved at 1, 3, and 12 months after treatment (p = 0.003, p = 0.006, and p = 0.009, respectively). In addition, a significant improvement was observed in the subgroup with worse baseline VA (logMAR VA > 0.3) 1 month after treatment, which was maintained 12 months after treatment. (b) CMT was considerably reduced 1 month after treatment, and the improvement was maintained 12 months after treatment (p < 0.001). Eyes with decimal visual acuities ≤0.5 also showed significant decreases in oedema 1 month after treatment, with maintenance of this improvement 12 months after treatment (p < 0.05). (2019) 9:7585 | https://doi.org/10.1038/s41598-019-44130-5 www.nature.com/scientificreports www.nature.com/scientificreports/ Adverse effects/complications. Cerebral infarction was observed in one case (2.9%, 1/34) that received ranibizumab in the first injection. There were no ocular complications associated with laser treatment or intravitreal anti-VEGF injection. Fig. 2. The patient was a 58-year-old man with focal serous retinal detachment in the fovea and circinate exudation with numerous microaneurysms on the temporal side of the fovea. The pre-treatment corrected decimal visual acuity was 0.4 ( Fig. 2a-c). Following the reduction in oedema after three anti-VEGF injections, subthreshold micropulse laser coagulation using the IQ577 and direct coagulation were performed for the residual temporal oedema at week 14 ( Fig. 2d-f). After laser application, macular oedema improved, and the corrected visual acuity had improved to 0.5 ( Fig. 2g-i).

Discussion
In this study, we evaluated the efficacy of intravitreal anti-VEGF injection combined with minimally invasive laser treatment for the treatment of DME and assessed the 12-month outcomes. The results showed that macular oedema was significantly reduced while visual acuity was significantly improved at 1 month after treatment, and that the improvements were maintained at 12 months after treatment.
Twelve months after treatment, the mean number of anti-VEGF injections and the corresponding ETDRS letter gain were 3.6 and 5.9, respectively, in the entire cohort, and 4 and 6.9, respectively, in the poor visual acuity subgroup. The corresponding values for the anti-VEGF monotherapy in previous studies were as follows: 10.2 and 10.3, respectively, in the RESOLVE study 16 , 7.8 and 6.6, respectively, in the REVEAL study 17 , and 7.4 and 7.9, respectively, in the RESTORE study 18 . Thus, the visual gain in the current study within 12 months after minimally invasive laser with intravitreal anti-VEGF injection was slightly worse than the average gain in previous prospective studies with anti-VEGF treatment alone. However, focusing on the number of injections, very small number of anti-VEGF injections (3.6) were required in our combination study group. The number of anti-VEGF injections required for treatment is often a burden for patients, both socially and economically. Therefore, one of the advantages of combined anti-VEGF and deferred laser therapy may lie in its potential to decrease the required number of anti-VEGF injections. A study with longer follow-up time is necessary to evaluate the effect of combination therapy on long-term visual outcome.
For several decades, laser monotherapy was the standard treatment strategy for DME. However, this treatment modality is associated with the side effect of deterioration of macular function. Subthreshold micropulse laser photocoagulation was introduced to overcome the side effects of focal macular laser therapy, and its efficacy for reducing macular oedema in DME is reported to be equal to or higher than that of conventional modified-ETDRS coagulation 19 . Furthermore, macular sensitivity is better with subthreshold micropulse photocoagulation than with conventional laser therapy 10 . Anti-VEGF therapy recently gained approval for DME treatment and is   www.nature.com/scientificreports www.nature.com/scientificreports/ currently the standard treatment for this condition. Therefore, the indication of laser monotherapy is limited to focal macular oedema, which can be treated with subthreshold laser therapy alone. However, in cases of severe diffuse oedema, laser monotherapy is not recommended, considering known evidence regarding the efficacy of anti-VEGF therapy. Therefore, we decided to combine anti-VEGF and minimally invasive laser therapy for the treatment of patients with DME and investigated the efficacy of this combination treatment in the present study.
The literature shows that the combination of focal macular laser therapy and anti-VEGF injections, particularly for initial treatment, does not improve the visual acuity. The Diabetic Retinopathy Clinical Research Network (DRCR-net) documented an ETDRS letter gain of 7.2 when treatment was initiated as combination treatment (immediate laser group) and 9.8 letters when laser treatment was added after 6 months of anti-VEGF treatment (deferred laser group). Thus, the number of letters gained was larger in the deferred laser group 20 .
In contrast, Liegl et al. 3 compared pro re nata (PRN) administration after three loading doses of anti-VEGF (anti-VEGF monotherapy) with NAVILAS laser therapy after three loading doses of anti-VEGF (anti-VEGF + laser combination) and showed that 84% eyes in the monotherapy group and 36% eyes in the combination therapy group required additional injections within 12 months. Furthermore, the improvement in visual acuity and reduction in CMT were not significantly different when comparing the anti-VEGF monotherapy group to the anti-VEGF + laser combination group. In the present study, we found that anti-VEGF + laser combination therapy had the potential to reduce the required number of intravitreal anti-VEGF injections.
The discrepancies in the results of the two abovementioned studies may be related to the timing of the laser therapy. It may be possible that laser therapy under a thickened retina induces severe thermal damage, and more laser energy is required to treat severe macular oedema, when the faint whitening of the retinal colour is used as the indicator of the effective burn in the conventional laser therapy. We posit that tissue damage resulting from laser therapy may account for a decreased visual acuity. In contrast, deferred laser therapy administered after a reduction in macular oedema results in the use of lower laser energy levels, potentially leading to visual improvements.
Intravitreal anti-VEGF drugs exert their effects on macular oedema soon after administration, following which they are cleared from the system. On the other hand, minimally invasive laser treatment, including subthreshold micropulse photocoagulation, has been shown to exert its effects 3 months after irradiation 9,11,13,14 . Thus, combination therapy allows for a persistent reduction in macular oedema after the effects of the anti-VEGF drug have attenuated because of clearance.
In the present study, laser treatment was intentionally applied after three loading doses of intravitreal anti-VEGF injections in 50% of patients. In addition to the advantages of reduced laser energy and persistent Figure 2. Representative case of a 58-year-old man with diabetic macular oedema treated with combined laser and anti-vascular endothelial growth factor (anti-VEGF) therapy. The patient exhibited diabetic macular oedema with serous retinal detachment in the fovea. His corrected visual acuity was 0.4 (a-c). Following the resolution of oedema after three intravitreal anti-VEGF injections, subthreshold micropulse laser coagulation and direct coagulation were performed to treat the residual temporal oedema (blue circle) (d) at week 14 (d-f). Nineteen weeks after the first injection, after three anti-VEGF injections and a single session of laser treatment, the oedema had disappeared, and the corrected visual acuity had improved to 0.5 (g-i). RV: Right vision.
www.nature.com/scientificreports www.nature.com/scientificreports/ reduction in macular oedema, this treatment strategy has another technical advantage. In subthreshold laser treatment, the laser spot is invisible. Therefore, laser energy has to be determined using titration methods. Because titration is performed in the retina, where no oedema is present, there is always a discrepancy in the energy used at the laser site if macular oedema is present. If the laser application site is not oedematous, the energy is presumably set at the appropriate level.
Although the mechanism of subthreshold micropulse photocoagulation for macular oedema has not been elucidated, this treatment is considered to reduce macular oedema by providing heat stimulation without destroying the RPE. Thus, its mechanism of action is different from that of anti-VEGF therapy. In addition, unlike conventional laser treatment that destroys the structure of the retinal layers, this treatment modality is non-invasive and allows for additional irradiation 8,13 . The endpoint management irradiation system with the PASCAL laser automatically calculates coagulation conditions via a unique algorithm for performing grid pattern scan laser photocoagulation, which enables simple and appropriate performance of subthreshold photocoagulation. This system is considered a minimally invasive laser treatment system that does not destroy the RPE, similar to subthreshold micropulse laser photocoagulation 21,22 .
Finally, no treatment-related complications, including the enlargement of coagulation spots and emergence of scotoma after laser irradiation, occurred in the present study. This suggests that the treatment efficacy and safety are better with a combination of minimally invasive laser therapy and intravitreal anti-VEGF injections than with a combination of conventional laser therapy and intravitreal anti-VEGF injection or a with a conventional laser therapy alone.

Conclusions
The results of this study suggest that the combination of intravitreal anti-VEGF injections with minimally invasive laser therapy for DME effectively improves the visual acuity and reduces macular oedema by 1 year after treatment. In addition, it reduces the required number of intravitreal anti-VEGF injections, thus reducing the economic burden on patients and healthcare systems. It would be important to create an algorithm for the combination therapy. On the basis of this study, we recommend initial anti-VEGF therapy with a loading dose (3 or more) until macular oedema is almost resolved, followed by application of a minimally invasive laser for residual oedema. If macular oedema recurs, the initial dose (which may not require 3 loading doses) should be repeated.
However, considering the limitations of the small sample size, short follow-up period (12 months), and retrospective nonrandomized design, a long-term follow-up study with a larger sample size is essential to confirm our findings.

Data Availability
The data used to support the findings of this study are included within the article.