Panretinal Photo-stimulation in Proliferative Diabetic Retinopathy

Pascal Pan Retinal Photo-Stimulation in Pre-Proliferative Diabetic Retinopathy: a Safety and Efficacy Study

Diabetic retinopathy (DR) is a leading cause of new cases of blindness in people aged 20 to 65 years worldwide.

Patients with DR may go on to develop a more severe form of the disease called Proliferative Diabetic Retinopathy (PDR), a condition in which abnormal new blood vessels may rupture and bleed inside the eye. When this advanced stage of retinopathy occurs, pan-retinal photocoagulation (laser treatment) is usually recommended.

The purpose of the investigators study is to find if treating patients using a single session of lower intensity laser (Pascal® Pan Retinal Photo-Stimulation, P-RPhS) at an earlier stage in Diabetic Retinopathy (during the severe non-proliferative diabetic retinopathy stage) when the abnormal new vessels are not developed, will prevent diabetic patients to develop PDR.

Patients included in the study will be randomized in three arms (randomization). In one arm patients will be treated with the normal parameters used in Pascal® laser, the second arm patients will be treated with a lower intensity than normal, using the Endpoint management system (a new software from the Pascal® laser which allow us to decrees the intensity of the burns (invisible burns) showing some landmarks with normal intensity so the area which has been treated can be viewed. And in the third arm the patients will be observed.

The published evidence from clinical trials in the literature has demonstrated that conventional PRP using 2000-2500 burns may be used safely and effectively in PDR patients over 2 or 3 sessions. A recent audit of 313 Pascal laser treatments performed at Manchester Royal Eye Hospital (2007-2008) has demonstrated that 1952 burns may be used safely and effectively in routine ophthalmic practice for PDR.

A recent study (PETER PAN study) which has been presented at the Association for Research in Vision and Ophthalmology (ARVO) 2011 Annual Meeting and also submitted for publication (Am J Ophthalmol) has shown that performing primary 20ms Pascal targeted retinal photocoagulation (TRP) and reduced fluence/minimally-traumatic panretinal photocoagulation (MT-PRP) in a single session using 2,500 burns can be safe and effective in treating PDR, without negative effect on visual fields. Therefore, it seems reasonable to think that a single session of 3000 burns 20ms with Endpoint Management set at 70% fluence will be safe and effective to treat patients with severe NPDR as similar or less total energy would be delivered to the retina-RPE complex.

Topcon's Endpoint Management (EM) software is based on a computational model of retinal heating and an Arrhenius damage model to determine optimal pulse duration and laser power for various levels of pulse energy. The algorithm varies the power and duration concurrently to maximize the margins between visible and sub-visible photocoagulation endpoints, providing a linear control over an inherently nonlinear process of coagulation.

When utilising EM, the user first titrates to a comfortable visible endpoint. Establishing a titration endpoint is important as it gives a consistent baseline, ensuring repeatable results between patients. With the Endpoint Management option activated, the laser output in a pattern is a percentage of this titration energy. Power and duration are both modulated to make this energy adjustment, allowing for a fine gradation of laser dosage and control over treatment endpoints. The visible titration endpoint can be references throughout the course of treatment by enabling the Landmark feature, which crates reference lesions at the titration dose at the corners of the pattern. This provides visible feedback on dosage and positioning of treated areas.

Importantly, the EM approach to laser therapy allows the physician to consistently operate in the realm of therapeutic relevance for sub-visible treatments. When no burns are visible, the biggest risk becomes lack of therapeutic effect. The Arrhenius integral-based algorithms in Endpoint Management adjust power and duration to provide the best "path" between endpoints, moving smoothly from the ophthalmoscopically-visible titration point to angiographically-only, OCT-only, and sub-visible/ therapeutic regimes. The Landmark feature allows the user to determine the local effect of the titration dose, and make adjustments as uptake varies across the treated area. The result is more predictable sub-visible laser delivery with the widest window of safe, effective treatment.

The Pascal® system with EM utilizes 577nm laser wavelength compared to the 532nm laser wavelength which was utilized in previous Pascal® laser studies. However, previous studies have not been able to demonstrate that clinical effectiveness was dependent on the laser wavelength.

In order to reduce the risk of complications after PRP whilst maintaining clinical efficacy, the optimal regime may be to utilize a technique of a lower intensity burn with reduced power while applying a larger number of treatment burns to cover the retinal area. It may also be important to "target" the laser treatment to the retinal area of "penumbra" or critically hypoxic retina based on angiographic tests.

"Ischaemic penumbra" is the critically hypoxic tissue characterised by Hypoxia-inducible factor-1 (HIF-1) upregulation and secretion of angiogenic proteins such as Vascular endothelial growth factor (VEGF). In ischemic diabetic retinopathy, there are 3 theoretical/postulated locations of such critically hypoxic inner retina:

1. Along the interface between the inner retinal area still oxygenated via retinal capillaries and the (more peripheral) area of capillary non-perfusion,
2. Alongside the medium-sized retinal arteries and hyperoxaemic retinal veins forming arterio-venous anastomoses while traversing non-perfused inner retina ("Krogh secondary tissue cylinders"), and
3. Mid-peripheral non-perfused inner retina obtaining a degree of oxygenation from the choroid via O2 penetration of the "metabolic O2 barrier" formed by the rod inner segments.

Fundus autofluorescence and Fourier-domain OCT imaging data suggested that a lower burn intensity and reduced fluence burn could achieve effective uptake within the outer retina.. confirmed effective laser uptake using the Pascal® multi-spot technique at 10-20ms pulse durations (Laser energy delivered to the tissue can be reduced by either reducing the intensity of the laser or the duration of the burn). Furthermore, the analysis of barely-visible laser titration burns showed effective thermal destruction of the inner segments of the photoreceptors which, because of their highly energetic metabolism, are the principal tissue target for laser photocoagulation.

In the treatment of PDR, the severity of visual field loss may correlate with burn density, longer pulse duration, and higher fluence. The ETDRS recommended 1 burn-widths spot-spacing, however the investigators aimed to maximize retinal laser coverage in a single sitting, and also to allow potential re-treatments to be safely placed between previous arrays of burns thus preventing overlapping laser burns and nerve fiber layer defects.

Visual acuity, visual field (VF) testing, and optical coherence tomography (OCT) are important tools for quantifying the degree of visual dysfunction. Central (24 degree) visual field measures in diabetic patients have been shown to correlate well with the stage of retinopathy according to the ETDRS scale. Pan-retinal photocoagulation has been shown to have a marked effect upon the central visual field measured with both global indices (Mean Deviation and Pattern Standard Deviation) and point-by-point measures. OCT is a sensitive instrument to diagnose and monitor DMO, and central retinal thickness (CRT) measures are used in clinical practice.

In addition, a multispectral camera has been used effectively in a recent study (Manchester Targeted Retinal Photocoagulation Study, MTRAP), with promising results produced that may help explain the clinical effects of the study laser treatment.

ETDRS data did not show that initiating scatter photocoagulation prior to the development of high-risk proliferative diabetic retinopathy in patients with type I diabetes will reduce the risk of severe visual loss. However, ETDRS analyses did indicate that for patients with type II diabetes, it is especially important to consider scatter photocoagulation at the time of the development of severe nonproliferative or early proliferative retinopathy.

Ferris also showed that patients with type II diabetes, or older patients with diabetes, are more likely to benefit from early scatter photocoagulation than patients with type I diabetes.

The recommendation of the Royal College of Ophthalmologists is to consider PRP only once the retinopathy reaches the proliferative stage.

The reasons why scatter laser PRP is not generally recommended for severe NPDR in the UK (and to prevent progression to PRD thereby) include: (i) uncertainty over desirable clinical end-points of treatment (eg. whether moderate NPDR or mild NPDR), and (ii) the likelihood of complications of PRP (such as DMO and visual field loss) that outweigh the potential benefits of early treatment. However, all recommendation up to date are based on the experience and results using argon laser as per EDTRS guidelines.

It is important to assess the benefit of Pascal single-session subthreshold scatter panretinal photocoagulation. As previously mentioned, all previous attempts of early PRP involved the use of argon laser systems and different treatment strategies: full retinal thickness burns, less number of burns and multiple treatment sessions. Laser treatment using these parameters is currently considered to be tissue ablative. It was also not possible, at the time, neither to image with as much clarity as the investigators can today the area of retinal ischaemic penumbra nor to treat the peripheral retina as far anteriorly as it can be done today. The investigators have shown that using Pascal20ms light burns allows for a tissue healing response as shown by Optical Coherence Tomography (OCT) and Fundus Autofluorescence (AF) retinal imaging.

The use of minimally-traumatic high-burn dosimetry (3000 burns) Pascal Pan Retinal Photo-Stimulation in Pre-Proliferative Diabetic Retinopathy may have implications in terms of NHS diabetic eye care. Single session treatment with Pascal retinal laser may have a significant cost saving effect for NHS departments, as the treatment time is significantly reduced as well as the number of required sessions and outpatient clinic appointments.

The Pascal system would allow larger number of patients to be safely treated per clinic session. Reducing the number of patients at risk of conversion to proliferative disease could not only have a significant cost saving effect for the NHS by reducing the frequency of hospital visits and cost of treatment but, most importantly would offer a significant overall benefit to the patient.

Currently, the Diabetic National Service Framework and English National Screening Programme for Diabetic Retinopathy have outlined targets for treating PDR at within two weeks of diagnosis in the eye clinic and from diabetic screening services. In current practice, this target may be only achievable in around 75% of patients. If clinical hypothesis is achieved, then a higher percentage would perhaps be achievable as the number of patients with proliferative retinal diabetic disease that require treatment would go down.

Novel P-RPhS Pascal treatment strategies may improve the comfort and safety of the patient's laser journey, and increase the compliance with laser treatment and follow-up.

Inclusion Criteria:

Patient-eligibility

Inclusion criteria:

1. Older than 18 years of age
2. Male or female patients with diabetes mellitus type I or type 2 who meet the WHO or ADA criteria for diabetes
3. Able to give informed consent

Study Eye eligibility

Inclusion criteria:

1. Treatment-naive S-NPDR

2. ETDRS visual acuity equivalent to 35 letters or better (Snellen equivalent 6/60 or better)

3. Any of the following:

   • Extensive (>20) intraretinal hemorrhages in each of 4 quadrants
   • Definite venous beading in 2+ quadrants
   • Prominent IRMA in 1+ quadrant

4. Mean central retinal thickness of less than or equal to 300 microns as measured by Deep Range Imaging Optical Coherence Tomography (DRI -OCT) scans

5. No intra-retinal fluid (IRF) or sub-retinal fluid (SRF) on DRI-OCT

6. Adequate pupil dilatation and clear media to perform wide-field colour, red-free imaging and fundus fluorescein angiography (WF-FFA), wide-field fundus autofluorescence imaging (WF-AF), DRI-OCT of retina and choroid, Multispectral Imaging (MSI) of chorioretinal oxygenation and P-RPhS

7. Ability to perform accurate Humphrey visual field test

Exclusion Criteria:

Patient-eligibility

Exclusion criteria:

1. History of chronic renal failure or renal transplant for diabetic nephropathy
2. Recent (last 6 months) or on-going poor glycaemic control. H1Ac greater than 10.0mg/dL
3. Creatinine greater than 1.2 mg/dL
4. HDL equal to or greater than 40 mg/dL
5. Uncontrolled hypertension. Blood pressure greater or equal to 180/110 mmHg
6. Patient is unavailable for follow-up visits
7. Pregnant women or breast-feeding females

Study Eye eligibility

Exclusion criteria:

1. Lens opacity that could influence vision and results
2. Any surgical or non-retinal laser treatment to the study eye within 2 months
3. Narrow drainage angles with raised intraocular pressure and angle closure glaucoma.
4. Planned YAG peripheral iridotomy
5. Previous retinal laser photocoagulation, intraocular drug therapy, or macular laser treatment to treatment eye in last year
6. Any previous ocular condition that may be associated with a risk of macular oedema
7. Active lid or adnexal infection
8. Planned intra-ocular surgery within one year