Subthreshold Micropulse Laser for Diabetic Macular Edema: A Review

Diabetic macular edema (DME) is one of the main causes of visual impairment in patients of working age. DME occurs in 4% of patients at all stages of diabetic retinopathy. Using a subthreshold micropulse laser is an alternative or adjuvant treatment of DME. Micropulse technology demonstrates a high safety profile by selectively targeting the retinal pigment epithelium. There are no standardized protocols for micropulse treatment, however, a 577 nm laser application over the entire macula using a 200 μm retinal spot, 200 ms pulse duration, 400 mW power, and 5% duty cycle is a cost-effective, noninvasive, and safe therapy in mild and moderate macular edemas with retinal thickness below 400 μm. Micropulse lasers, as an addition to the current gold-standard treatment for DME, i.e., anti-vascular endothelial growth factor (anti-VEGF), stabilize the anatomic and functional retinal parameters 3 months after the procedure and reduce the number of required injections per year. This paper discusses the published literature on the safety and application of subthreshold micropulse lasers in DME and compares them with intravitreal anti-VEGF or steroid therapies and conventional grid laser photocoagulation. Only English peer-reviewed articles reporting research within the years 2010–2022 were included.


Introduction
Diabetes mellitus (DM) has become a civilization disease associated with a sedentary lifestyle and the aging of the population in the contemporary world. It is estimated that DM affects around 10% of the global population [1]. The prevalence of diabetes is increasing rapidly, and the World Health Organization (WHO) has recognized diabetes as a noncommunicable disease which is causing an epidemic in the 21st century. An insufficiently controlled and long-term disease is associated with a high risk of multiorgan complications, including those involving eyes. One of the main retinal complications is diabetic macular edema (DME), which leads to gradual visual impairment, especially at working ages. DME occurs in 4% of patients diagnosed with DM, even at the early stage of diabetic retinopathy. The estimated number of adults worldwide in 2020 with clinically significant DME was 18.8 million, and this is projected to increase by half in 2045 [2].
According to the current international guidelines for the management of DME by the European Society of Retina Specialists (EURETINA), intravitreal anti-vascular endothelial growth factor (anti-VEGF) was established as first-line therapy in DME with visual impairment [3]. After publishing the results of the DRCR.net Protocol I and Protocol S studies, laser therapy was regarded as being inferior to anti-VEGF treatments [4].
The availability of anti-VEGF injections has changed the standard of care for DME patients [5]. The agents improve both the functional and anatomical parameters of the retina. Currently, different anti-VEGF agents such as ranibizumab, aflibercept, brolucizumab, faricimab, and off-label bevacizumab have become the therapy of choice in DME treatments [6]. Intravitreal injections have a high efficacy and safety profile, however, on the phyll in the neurosensory retina. Commercially available devices can deliver conventional and micropulse shots at 577 nm that simultaneously enable the combination therapy of the grid micropulse laser and the direct photocoagulation of microaneurysms [18]. The 670 nm laser is less scattered and not absorbed by hemoglobin and xanthophyll, thus, it seems to be safe for the neurosensory retina [19]. There is no consensus on which wavelength is the most favorable for the treatment for DMEs, thus, all the above-described devices have a high safety profile and are recommended for micropulse use.
Currently, the specific indications for the application of MPLT have not been established. It is considered as an alternative treatment in macular disorders such as DME (Figure 1), central serous chorioretinopathy, and macular edemas that are secondary to retinal vein occlusion [20][21][22][23]. MPLT was proven to be efficient and free from adverse events in minor and moderate macular edemas with a central retinal thickness (CRT) below 400 µm and relatively good visual acuity [24]. As an adjuvant to anti-VEGF agents, it helps to stabilize the anatomic and functional retinal parameters with a lower required number of injections. of the micropulse was performed with a 810 nm laser modality, where the wavelength deeply penetrated the retina and, thus, it was not absorbed by the macular carotenoids. The 577 nm wavelength targets oxyhemoglobin and melanin, and it is not absorbed by xanthophyll in the neurosensory retina. Commercially available devices can deliver conventional and micropulse shots at 577 nm that simultaneously enable the combination therapy of the grid micropulse laser and the direct photocoagulation of microaneurysms [18]. The 670 nm laser is less scattered and not absorbed by hemoglobin and xanthophyll, thus, it seems to be safe for the neurosensory retina [19]. There is no consensus on which wavelength is the most favorable for the treatment for DMEs, thus, all the above-described devices have a high safety profile and are recommended for micropulse use.
Currently, the specific indications for the application of MPLT have not been established. It is considered as an alternative treatment in macular disorders such as DME (Figure 1), central serous chorioretinopathy, and macular edemas that are secondary to retinal vein occlusion [20][21][22][23]. MPLT was proven to be efficient and free from adverse events in minor and moderate macular edemas with a central retinal thickness (CRT) below 400 μm and relatively good visual acuity [24]. As an adjuvant to anti-VEGF agents, it helps to stabilize the anatomic and functional retinal parameters with a lower required number of injections.

Materials and Methods
The present paper reviews all the relevant literature on DME treatments with a subthreshold micropulse laser. The PubMed database and Mendeley were used as a source of studies within the years 2010-2022. Only peer-reviewed articles published in English reporting research were included. Relevant studies were identified using the following terms in combination with Boolean operators: subthreshold laser, micropulse laser, diabetic macular edema, clinically significant macular edema, anti-VEGF, intravitreal steroid, vitrectomy, conventional photocoagulation, ETDRS photocoagulation, continuous-wave photocoagulation, combined therapy, and safety. Subsequently, a manual search of the

Materials and Methods
The present paper reviews all the relevant literature on DME treatments with a subthreshold micropulse laser. The PubMed database and Mendeley were used as a source of studies within the years 2010-2022. Only peer-reviewed articles published in English reporting research were included. Relevant studies were identified using the following terms in combination with Boolean operators: subthreshold laser, micropulse laser, diabetic macular edema, clinically significant macular edema, anti-VEGF, intravitreal steroid, vitrectomy, conventional photocoagulation, ETDRS photocoagulation, continuous-wave photocoagulation, combined therapy, and safety. Subsequently, a manual search of the reference lists in the retrieved manuscripts was performed. Studies discussing the use of a micropulse transscleral laser for the treatment of glaucoma were excluded. A total of 68 full-text articles on MPLT were assessed for eligibility and divided into four sections covering safety, efficacy, and comparisons with conventional laser and intravitreal therapies (Table 1).

Safety of MPLT
A high safety profile of MPLT was reported in the in vivo and in vitro studies (Table 2). Potential damages were assessed using mathematical models, investigated using animal and stem cell cultures, and measured in imaging tests such as infrared (IR) and redfree fundus photos, optical coherent tomography (OCT), fundus autofluorescence (FAF), microperimetry, fluoresceine angiography (FA), and indocyanine angiography (IGCA).  Ohkoshi et al. [25] detected sites of the application of the micropulse laser in scanning laser ophthalmoscopy in the retro mode. Dark spots were visible immediately after photostimulation, and they were not identified in FAF nor in the fundus photos. However, after 1 week, the alterations were no longer observed. This study implied that MPLT affects the RPE cells and can cause the localized swelling of the treated region.
Luttrull et al. [26] assessed the risk of laser-induced retinal thermal injury by comparing computer modeling of the tissue temperature after MPLT using clinical findings in imaging tests such as IR and red-free fundus photography, FAF, FFA, and OCT. According to the study, an increased risk of retinal damage was related to higher retinal irradiance, and it was found in none of the patients treated with MPLT at a 5% duty cycle.
Wells-Gray et al. [28] confirmed the structural damage after MPLT by measuring the integrity of cone photoreceptors using advanced adaptive optics imaging.
Midena et al., in their studies, pointed to the role of the influence of MPLT on the retinal biomarker levels in aqueous humor [29,30,32]. A strong correlation in protein concentration between the aqueous and vitreous humor was previously proven [37], therefore, a simpler accessible anterior chamber fluid was used for the samples. The authors measured the concentration of the biomarkers of RPE, Müller cells, and a panel of inflammatory molecules in eyes with the DME before and after the MPLT treatment and compared the values with the control groups with healthy ones. The results of their papers were consistent, and they found the effect of MPLT on the expression of aqueous humor markers to be statistically significant. The decrease in proinflammatory proteins and the VEGF level suggested that the MPLT may deactivate the retinal microglia and reduce diabetes-induced inflammation. Moreover, a significant decrease in bioindicators of Müller cell activation implied that MLPT induced positive retinal metabolic and morphology alterations.
Vujosevic et al. [27] showed that both 577 nm and 810 nm micropulse lasers in a "highdensity" pattern with 5% DC were safe and efficient in mild DMEs. No retinal damage was detected during any clinical imaging examination. They suggested that the MPLT with the lowest CD and without titration could be a repeatable and simple treatment for patients. In reference to this study, Chang et al. [31] used the same micropulse laser parameters to assess the kinetics of RPE heat-shock protein (HSP) activation. HSP is a group of proteins that are produced in response to cell exposure to stress and during tissue remodeling. This report showed that both the lasers were equally efficient, but a higher predictability and wider safety margin resulted from the use of the 810 nm one. The upregulation of the HSP 70 family was confirmed in the study led by Shiraya et al. [33] on irradiated human RPE stem-cell cultures, which suggested that MPLT could be more beneficial for light perception, photoreceptor protection, and maintenance than a conventional laser could.
In agreement with the results of HSP observation, De Cilla and colleagues [35] proved that MPLT not only reduced oxidative stress and markers of apoptosis, but it also increased autophagia in mouse retinal cells. This study proved that the oxidant-antioxidant balance shifted in favor of the antioxidant system with an increasing number of treatments and with a younger age. Moreover, no laser effect was shown in fellow untreated eyes.
Yu et al. [34] conducted a study on the tissue section of enucleated rabbits' eyes. In the experiment, the right eyes were treated using an 810 nm micropulse laser, and the left eyes were treated using a 532 nm micropulse laser with 5%, 10%, 20%, and 40% DC. The samples were analyzed for protein marker expression and morphological changes in the retinal tissues. The histologic effect and protein regulation induced by both the lasers were not distinguishable. The 5% DC therapy caused no retinal disruption or RPE damage.
No retinal damage induced by MPLT was confirmed in another animal model investigated by Hirabayashi [36]. According to the upregulation of aquaporin 3 gene expression in retinal photoreceptors, the researchers concluded that MPLT may be responsible for suppressing macular edema and intensifying drainage of retinal fluid. However, the role of aquaporin 3 remains unclear, and it needs to be confirmed in other studies.
MPLT, micropulse laser treatment; CWL, continuous-wave laser; DME, diabetic macular edema; FU, follow-up (in months); BCVA, best corrected visual acuity; CRT, central retinal thickness; CS, clinically significant; CI, center involved; RPE, retinal pigment epithelium; FAF, fundus autofluorescence; IR, infrared; FA, fluoresceine angiography; DRIL, disorganization of inner retinal layers; mfERG, multifocal electroretinography; EDI, enhanced-depth imaging. Nakamura et al. [38] proved that functional improvement after MPLT was limited to an increase in visual acuity. According to the study, the macular sensitivity within the central 10 • in microperimetry did not improve significantly, despite the increase in BCVA and the reduction in foveal thickness.
Luttrull et al. [14] observed that significant differences between pre-and postoperative CRT were observed in eyes with CRT < 300 µm, with a maximum reduction between 4 and 7 months after MLPT. The BCVA was stable with a significant improvement between 4 and 7 months of the follow-up.
According to Kwon et al. [51], the MPLT did not cause chorioretinal scars despite repeated treatments occurring and there being an increased number of micropulse shots. The study showed a similar efficacy of the micropulse and conventional lasers.
Inagaki et al. [18] compared the efficacy of 810 nm and 577 nm MPLT combined with focal microaneurysm photocoagulation. They proved that both the wavelengths are effective in reducing CRT and maintaining visual acuity. As advantages of the 577 nm wavelength, they pointed out that it required less power and enabled them to perform both the micropulse and classic therapies using the same device. Supplementary microaneurysm photocoagulation reduced the recurrence rate. Marashi et al. [59] agreed that the hybrid threshold laser of microaneurysms with subthreshold micropulse high-density laser effectively stabilized the DMEs with minimal scar formation.
Mansouri et al. [50] concluded that the retinal thickness affects the spread of the laser energy and influences the tissue response. The authors compared the efficacy of MPLT according to anatomical severity of the edema, suggesting MPLT as an effective and safe therapy in mild and moderate DMEs. In the study, all the eyes with initial CRT > 400 µm did not respond to the therapy and required rescue injections of anti-VEGF. Citirik et al. [56] also showed the relationship between the efficacy of the micropulse laser and the central retinal thickness. The study indicated that eyes which previously underwent ineffective bevacizumab treatment responded well to MPLT if the CRT was no higher than 300 µm.
Nicolò et al. [49] suggested that the micropulse laser is ineffective in eyes which previously did not respond sufficiently to focal or grid macular photocoagulation or an anti-VEGF treatment. Additionally, the authors reported a better response to the treatment of naïve patients, with a stabilization of or improvement in the BCVA and CRT parameters. Valera-Cornejo et al. [41] observed changes in BCVA only in previously untreated patients. It should be underlined that the laser procedures were performed not only over the edema, but also over the entire macula, including the foveal center and unthickened retina. In contrast, the work by Abouhussein et al. [48] led to a different conclusion, i.e., that a single session of MPLT was effective in patients with a refractory DME below 400 µm. In terms of limitations, both the studies had short follow-up and small sample sizes without randomization.
Latalska and colleagues [47] proved that the effects of the micropulse laser were more significant in a rural environment than they were in an urban environment. Moreover, they pointed out that glycated hemoglobin level ≤ 7% significantly influenced the improvement in CRT and near visual acuity.
Optical coherence tomography angiography (OCT-A) is a novel noninvasive accessory examination, which enables imaging vascular abnormalities and microaneurysms in the superficial and deep capillary plexus. It also reveals the enlargement of the foveal avascular zone (FAZ), nonperfused areas, and neovascularization [60]. The studies by Vujosevic et al. [44,45] showed the mechanism of action of a micropulse laser via a reduction in the inflammatory biomarkers detected in OCT and OCT-A. They detected a decreased number of hyper-reflective spots and microaneurysms, whereas the chorioretinal perfusion parameters were stable in response to the MPLT.
No significant changes have been observed in fixed and variable regimens of 577 nm MPLT for mild center-involved DMEs, however, Donati et al. [16] suggested that fixed parameters facilitate the treatment and reduce the number of potential errors. Frizziero et al. [43] confirmed the safety of the fixed model. Nowacka et al. [58] reported the stabilization of the macular structure through the maintenance of the bioelectrical function of cones and bipolar cells detected in mfERG.
Ueda et al. [40] proved the entropy of RPE cells as an objective indicator of the retinal healing process. They showed a positive correlation between the decrease in CRT after MPLT and entropy measurements in RPE.
According to Işık et al. [39], the response to MPLT may be related to the status of the central RPE and glycated hemoglobin level, however, further studies on a larger group are required.
A recent study by Kikushima et al. [19] compared the 577 nm with the novel 670 nm micropulse treatment. Both the wavelengths seemed to be equally effective, however, the use of the 670 nm laser resulted in less scattering and better penetration.

Comparison of Subthreshold Micropulse and Conventional Laser Treatment
Studies comparing MPLT with conventional laser therapy are presented in Table 4. No damage was identified after MPLT in OCT scans; fewer changes in outer retina after pattern scanning laser than after conventional laser. In most reports, the authors found micropulse subthreshold laser therapy to be equivalent to conventional macular photocoagulation [61][62][63][64][65][67][68][69][70][71][72]. Vujosevic et al. pointed out that MPLT is not only as effective as classic lasers are in reducing macular edema, but it is also a less aggressive therapy, as shown by the increased retinal sensitivity in the microperimetry. The positive influence on central retinal sensitivity was also confirmed in the study by Chhlablani et al. [68]. Venkatesh and colleagues [63] suggested that MPLT did not induce any functional loss detected in multifocal electroretinography, with equally good therapeutic effects. Inagaki et al. [64] investigated Japanese patients with a more pigmented retina which could predispose them to the increased absorption of laser energy and more severe retinal damage. Changes in retinal morphology at 3 months after the laser therapy were detected only after pattern scanning and a conventional grid treatment. A recently published multicenter clinical trial by Lois et al. [72] included a large number of participants (266 eyes) with mild DMEs (<400 µm). The study confirmed the clinical effectiveness, safety, and cost-effectiveness of MPLT in compared to those of a conventional laser treatment.
Lavinsky et al. [62] observed the superiority of a high-density, confluent micropulse treatment regarding the anatomical and functional outcomes after 1 year of the follow-up. In contrast, after the normal-density treatment (two burn widths apart), no improvement was seen. Correspondingly, Fazel et al. [67] measured that MPLT significantly improved the BCVA and CRT parameters in eyes with a previously untreated, mild DME. The presented study showed MPLT to be more effective than continuous-wave treatment did in the very short term (4 months). Similarly, Bougatsou et al. [69] agreed that MPLT was more efficacious than a conventional laser was in non-center-involved clinically significant macular edemas, whereas Al-Barky et al. [70] observed slightly better functional outcomes after MPLT. Othman et al. [66] compared MPLT in treatment-naïve patients with MPLT in recurrent or persistent DME 3 months after conventional macular photocoagulation. The therapy was similarly effective in both groups, however, more patients in the secondary group required rescue therapy with an intravitreal steroid.
Available data on alternative subthreshold micropulse panretinal photocoagulation (PRP) in treating severe non-proliferative diabetic retinopathy and proliferative diabetic retinopathy are limited, and without studies of higher quality according to evidence-based medicine (EBM), it should be considered as experimental [73,74].

Subthreshold Micropulse Laser Treatment and Intravitreal Therapy
Numerous studies compared MPLT with intravitreal treatment or investigated combination therapy (Table 5).    Most articles compared MLPT with bevacizumab, ranibizumab, and aflibercept. The treatment protocol for anti-VEGF monotherapy was three loading injections at a monthly interval followed by a pro re nata (PRN) scheme. The patients qualified for micropulse therapy after receiving three initial loading anti-VEGF doses and with a CRT below 400 µm. It was suggested that additional laser treatment could decrease the burden of agent injection frequency with similar functional and anatomical outcomes [75,[78][79][80]82,83,[85][86][87]. However, the study by Akhlaghi et al. [77] led to a different conclusion: adjuvant MPLT improved BCVA and CRT in eyes resistant to the bevacizumab therapy.
Inagaki et al. [75] suggested that the initial loading dose of intravitreal anti-VEGF agent, followed by a single MPLT for residual edema reduces the number of required injections and effectively improves BVCA and CRT.
Akkaya et al. [76] proved that MPLT was superior to anti-VEGF injections in patients with mild macular oedema (CRT max. 350 µm) and good visual acuity (BCVA ≤ 0.15 logMAR) due to there being less frequent visits, lower costs, and a higher safety profile. In this regard, MPLT could be considered as the early intervention and, if it is necessary, it can be continued with anti-VEGF injections.
The study by Abdelrahman et al. [81] compared patients treated with MPL or ranibizumab with a control group for multifocal electroretinography (mfERG). The functional outcome was additionally measured not only by the subjective BCVA, but also by objective mfERG readings from the macular region. Only in the ranibizumab group was there a significant improvement in electrophysiological parameters after the treatment. They proved that both MPLT and ranibizumab improved the anatomical and functional retinal parameters, with superiority over the intravitreal agent.
A recent retrospective study by Lai et al. [88] presented that aflibercept monotherapy resulted in short-term higher functional and anatomical improvement compared to that resulting from the MPLT with rescue aflibercept therapy, however, the long-term results did not show any significant differences. In contrast to other studies, MPLT was not preceded by initial anti-VEGF injections, and it was performed with focal conventional laser treatment of microaneurysms.
In general, the authors agree that adjuvant micropulse therapy reduced the number of required intravitreal injections, apart from Koushan et al. [89], who did not find an additional benefit in using a combined therapy.
Elhamid et al. [9] treated center-involved DMEs, which previously did not respond to an anti-VEGF therapy, with a combination of an Ozurdex implant and MLPT. As in other studies, they suggested that poor response after three initial monthly injections of anti-VEGF predicts reduced the persistent response for subsequent doses. An early switch to a steroid implant diminished the number of intravitreal surgeries. In this study, the frequency of recurrence was relatively lower than it was in other trials with the dexam-ethasone implant, which can be explained by the synergic effect of MPLT. In terms of the limitation, the obtained results require confirmation in larger studies with a control group. Toto et al. [90] also demonstrated the effect of MPLT in addition to a dexamethasone implant. The combined therapy reduced the frequency and the number of required injections, thus extending the treatment-free interval.
Micropulse lasers appear to be an efficient modality to decrease persistent DMEs after pars plana vitrectomy. A comparative study by Bonfiglio et al. [91] showed that MPLT performed 6 months after surgery improved the anatomical and functional parameters in vitrectomized eyes.

Conclusions
An analysis of the available results is limited due to the scarce number of large, randomized clinical trials. The reviewed studies varied in terms of the inclusion criteria, protocols, and treatment procedures. The detailed eligibility criteria for MPLT have not been defined, however, according to the presented literature, there are some therapeutic principles.
Three meta-analyses which evaluated the efficacy of MPLT versus conventional photocoagulation or intravitreal injections have been published. Chen et al. [92] compared the mean change in BCVA and CRT, according to six randomized controlled trials (RCTs), including a total of 398 eyes. MLPT resulted in better visual acuity with similar anatomical outcome. Similarly, Qiao et al. [93] compared MPLT with an mETDRS treatment in seven RCTs on 425 eyes. They found no statistical differences in BCVA and CRT after the treatments, with less retinal damage after MPLT. Wu et al. [94] performed a Bayesian analysis of 18 studies, comprising a total of 1758 patients, which assessed the effect of lasers in monotherapy or adjuvant therapies to anti-VEGF. The findings showed that ranibizumab plus conventional photocoagulation is more effective than micropulse laser monotherapy is, however, there was no significant difference in efficacy between the MPLT and bevacizumab plus conventional laser treatments, as well as between the MPLT and conventional laser monotherapies.
There are no standardized protocols for MPLT, however, according to the reviewed articles, micropulse panmacular treatment including the fovea, with a fixed regimen, seems to be a cost-effective, noninvasive, and safe therapy. Data in the analyzed articles confirmed that 577 nm laser applications using a 200 µm retinal spot, 200 ms pulse duration, 400 mW power, and 5% DC induced significant morphologic and functional improvement in the central retina and were not associated with any adverse events. Titration can prolong and complicate the procedure. The continuous-wave test burn is performed outside the posterior pole, over non-edematous retina, until a barely visible white spot is created. There is no consensus, after reaching the threshold, on how much to modify the laser power. Some authors switched the continuous wave to the micropulse mode, multiplying the threshold value by 0.5-4. Some researchers titrated the power in micropulse mode and then divided the value by 2. The proper subthreshold value is hard to determine, and medical errors can lead to overtreatment and involuntary damage of the retina. A confluent treatment using fixed 400 mW power for yellow laser with low 5% DC and high intensity was confirmed to effectively stimulate RPE cells.
None of the presented studies detected any visible signs of chorioretinal damage in the ancillary imaging tests and animal retinal sections. In contrast to harmful conventional laser treatment, MPLT additionally increased the central retinal sensitivity.
The efficacy of the micropulse laser was proven in mild DMEs with a CRT that is smaller than 400 µm due to the diffused distribution in the target tissue. In general, the treatment helps to stabilize or improve the visual acuity and decrease the macular edema. Better results are observed in a high-density protocol covering the macular region, with no spacing between the spots. Automatic pattern systems are helpful in the application of invisible laser spots. The minimal interval from the treatment to obtain a significant response and a reduction in retinal thickness is about 3 months. Therefore, it can be recommended to start the therapy with three loading doses of anti-VEGF, followed by MPLT combined with PRN injections to achieve a quick response to anti-VEGF, which is supported by the long-lasting remodeling effect of MPLT. The increased number of micropulse sessions is associated with a greater retinal response. A combined treatment requires a lower number of anti-VEGF injections, and it is not inferior to monotherapy [95,96]. MPLT is also an emerging option as a standalone treatment for noncompliant patients and for those having contraindications for other therapies.