eMethods. Laser Treatment Variables, Assessments, Retinal Imaging
eResults. Laser Treatment and Secondary Outcomes
eTable 1. Cross Tabulation of Baseline Clinical Diabetic Retinopathy Grade of the Enrolled Participants
eTable 2. Retinal Sensitivity Results From Microperimetry for Central Macular 12° and 4°
eTable 3. Cross Tabulation of Early Treatment Diabetic Retinopathy Study Grade of Retinopathy in Right and Left Eyes for Each Subject at Baseline and Exit
Mala Subash, Oliver Comyn, Ahmed Samy, Dania Qatarneh, Serafeim Antonakis, Manjit Mehat, James Tee, Tatiana Mansour, Wen Xing, Catey Bunce, Ananth Viswanathan, Gary Rubin, Richard Weleber, Tunde Peto, Louisa Wickham, Michel Michaelides. The Effect of Multispot Laser Panretinal Photocoagulation on Retinal Sensitivity and Driving Eligibility in Patients With Diabetic Retinopathy. JAMA Ophthalmol. 2016;134(6):666–672. doi:10.1001/jamaophthalmol.2016.0629
Panretinal photocoagulation (PRP) for proliferative diabetic retinopathy (PDR) may lead to peripheral field loss that prevents driving. Anti–vascular endothelial growth factor agents are proposed as treatments for PDR that spare peripheral vision. If multispot lasers cause less visual field loss, continuing to perform PRP may be justified.
To assess the effect of bilateral multispot laser PRP on retinal sensitivity and driving visual fields in PDR.
Design, Setting, and Participants
This prospective nonrandomized interventional cohort analysis performed at a tertiary referral center included 43 laser-naive patients with PDR that required bilateral PRP. Participants were recruited from June 27, 2012, to October 14, 2013. At baseline and 6-month follow-up, patients underwent detailed static and kinetic perimetry, microperimetry, optical coherence tomography, wide-field color fundus photography, and fluorescein angiography. Quantitative change in retinal sensitivity was assessed by comparing the mean global retinal sensitivity before and after laser treatment and by comparing the modeled hill of vision by deriving a volumetric measure. Final follow-up was completed on May 21, 2014.
Multispot laser treatment was applied using standard parameters, until neovascularization regressed or complete retinal coverage was achieved.
Main Outcomes and Measures
Participants who passed the Esterman binocular visual field test for driving in the United Kingdom (at least 120° horizontal field with no significant defects within the central 20°) and full-field and macular retinal sensitivity.
Of the 43 patients (17 men; 26 women; mean [SD] age, 46.6 [13.3] years), 38 (88%) completed the study. Before treatment, 41 of 43 patients (95%) passed the Esterman visual field test for driving; after completion of laser treatment, 35 of 38 patients (92%) passed. The mean (SD) change in retinal sensitivity on static perimetry was −1.4 (3.7) (95% CI, −2.7 to −0.1) dB OD and −2.4 (2.9) (95% CI, −3.4 to −1.4) dB OS. Mean (SD) 4° macular sensitivity decreased by 3.0 (5.2) dB OD and 2.6 (5.4) dB OS.
Conclusions and Relevance
This prospective study investigating the effects of multispot laser PRP on retinal sensitivity demonstrates a high likelihood of retaining eligibility to drive based on adequate visual field. A mild loss of retinal sensitivity was detected at 6 months after completion of laser treatment. Further change to visual fields may have occurred with longer follow-up. This study provides information that might be used to counsel patients requiring PRP and informs the debate regarding the role of anti–vascular endothelial growth factor therapy in patients with PDR who might otherwise receive laser treatment.
Untreated proliferative diabetic retinopathy (PDR) can lead to visual loss from vitreous hemorrhage and tractional retinal detachment. Since 1976, when the Diabetic Retinopathy Study demonstrated a reduction in the rates of visual loss with laser treatment,1 panretinal photocoagulation (PRP) has been the standard treatment for PDR. Although PRP has been shown to be effective in reducing the risk for severe visual loss in multiple trials,2- 5 its destructive nature may cause peripheral visual field loss, with implications for the patient’s ability to drive.6- 10 Estimates for the prevalence of visual field defects sufficient to preclude driving at 6 months after laser treatment range from 11%7 to 50%,9 although the higher estimate included patients treated with the xenon arc laser. A more recent study,11 however, identified an improvement in the central 24° threshold sensitivity at 6 months after laser treatment, with no adverse effect on visual fields required for driving.
Anti–vascular endothelial growth factor agents have been proposed as a treatment for PDR that could avoid visual field loss, and a recently published randomized clinical trial12 showed that repeated ranibizumab injections resulted in noninferior visual acuity after PRP. A secondary outcome of that study was the visual field score after treatment; the ranibizumab-treated group developed significantly lower reductions in overall visual field sensitivity. Aside from that study, current data on the effects of PRP on the visual field are significantly limited by several factors, including the retrospective nature of the studies, small sample sizes, nonstandardized laser treatments, and the wide variability of findings ranging from few to most patients having marked visual field loss after PRP. These limitations, coupled with the transition from conventional laser systems to multispot laser delivery, led to our prospective study of the effects of standardized multispot laser PRP on retinal sensitivity and driving eligibility.
In the United Kingdom, the Driver and Vehicle Licensing Agency defines field of vision requirements for holding a group 1 (car or motorcycle) license as a minimum field of 120° along the horizontal meridian, extending at least 50° left and right, with no significant defect within the 20° of fixation above or below the horizontal meridian, using a Goldmann white III4e target equivalent. Testing is typically performed by binocular Esterman perimetry.13 In obtaining consent from patients for PRP, the reduction in visual field below the driving standard must be mentioned as one of the potential risks because of previously cited study results. Patients are also required to inform the Driver and Vehicle Licensing Agency when they are receiving bilateral laser therapy for PDR. In contrast, the United States has no uniform visual standard for driving. Instead, requirements are determined at a state level, with differences in the visual acuity and visual field required to hold restricted (car) and unrestricted licenses.14
Multispot lasers allow delivery of multiple burns with a single activation. The semiautomated procedure uses a brief pulse duration of 0.02 to 0.05 seconds, combined with a rapid raster scan application of multiple spots, which allows shorter treatment delivery time. By contrast, conventional photocoagulation uses a single application of laser energy per burn that is usually delivered at a 100-millisecond duration. With the same spot size and similar power levels, overall fluence of the multispot laser is reduced and may thereby limit collateral retinal damage and reduce subsequent expansion in laser spot size.15 Multispot laser delivery systems, including the Valon (Valon Lasers) and PASCAL (pattern scan laser) Photocoagulator (OptiMedica), are in routine use. Whether multispot lasers have a different effect on the visual field than single-spot lasers remains unknown. Therefore, to provide robust and reliable information for physicians and patients, we prospectively assessed the effects of PRP delivered by multispot laser on peripheral and central retinal sensitivity and visual fields required for driving.
Question Does panretinal photocoagulation delivered by multispot laser result in a change of visual field that precludes driving?
Findings This nonrandomized interventional cohort study showed that 92% of patients with proliferative diabetic retinopathy treated in this way retained sufficient visual field to pass standards for driving in the United Kingdom. Laser treatment led to a small reduction in the overall volume of the hill of vision.
Meaning Visual fields after multispot laser panretinal photocoagulation for proliferative diabetic retinopathy are likely sufficient to pass standards for driving in the United Kingdom.
This prospective interventional study was conducted at a single center. Participants were recruited from the Medical Retina Service of Moorfields Eye Hospital. Prospective ethical approval was obtained from the United Kingdom National Research Ethics Service, and the study was conducted in accordance with the tenets of the Declaration of Helsinki.16 All participants received detailed written information with an explanation of the nature and possible consequences of participation and provided written informed consent to take part in the research before inclusion.
Adults with type 1 or 2 diabetes mellitus who required bilateral PRP for PDR or severe nonproliferative diabetic retinopathy with no previous laser photocoagulation were enrolled. Potential participants were excluded in the presence of any coexistent ocular or systemic condition that may have affected their visual field, a visual acuity of less than 20/200 that may affect the accuracy of visual field testing, the presence of vitreous hemorrhage, planned intraocular surgery or other intervention anticipated in either eye during the period of treatment, or insufficient participant cooperation for adequate visual field testing.
All participants received PRP using the Valon TT multispot laser system (Valon Lasers) with standardized parameters (eMethods in the Supplement). Participants attended at baseline and at 6 months when functional and structural assessments were performed according to predefined protocols. After the initial baseline and 2-week visits when laser treatment was mandated, further visits occurred at 6-week intervals when laser treatment was performed as required. Before undergoing laser treatment, participants underwent a series of retinal functional assessments and imaging that was repeated at 6 months. These procedures were performed in the following standardized order (detailed in eMethods in the Supplement):
Binocular Esterman visual field test for driving on the Humphrey visual field analyzer (Carl Zeiss Meditec AG);
Monocular full-field static perimetry on the Octopus 900 (Haag-Streit Diagnostics) using the German adaptive threshold estimation (GATE) strategy17;
Binocular full-field static perimetry on the Octopus 900 using the GATE strategy;
Goldmann monocular kinetic testing;
Microperimetry on the Nidek MP-1 (Nidek Technologies);
Massof Activity Inventory questionnaire, an adaptive visual function questionnaire that quantifies the difficulty of performing certain tasks18;
Ultrawide-field color and fundus fluorescein angiography; and
Spectral-domain optical coherence tomography.
A quantitative assessment of change in retinal sensitivity was undertaken by comparing the mean global retinal sensitivity and comparing the entire hill of vision (HOV), an arguably more sophisticated analysis that better represents the data by using the full data set with a volumetric approach before and after PRP.19
Outcomes were assessed 6 months after enrollment and the baseline visit. The primary purpose of the study was to assess the effect of multispot laser PRP on driving eligibility and visual fields. Functional outcomes included the participants’ pass rate for the Esterman visual field test for driving, global mean retinal sensitivity from full-field static and kinetic testing, HOV assessment of monocular and binocular static full-field perimetry, microperimetric macular sensitivity of the central 4° and 12°, and self-reported visual function.
Secondary outcomes included change in best-corrected visual acuity and change in central macular thickness. The Early Treatment Diabetic Retinopathy Study (ETDRS) grade and macular ischemia were reported at baseline and 6-month follow-up. The number of cases of macular edema and the data on laser treatment, including the numbers of visits and total treatments delivered, were reported.
This nonrandomized interventional study focused primarily on providing summary statistics. We report summary measures for the baseline characteristics and at final follow-up using means (SDs) for continuous (approximate) normally distributed variables, medians (interquartile ranges) for nonnormally distributed variables, and frequencies and percentages for categorical variables. Our assessment of normality was based on the visual examination of histograms to ensure that no overt skewness or kurtosis was present. Where changes in visual field occurred, 95% CIs were calculated. Where unavailability for follow-up occurred, data were analyzed on an available-case basis. We conducted an exploratory paired t test to assess whether the observed change in the Massof Activity Inventory score was statistically significant.
Participants were recruited from June 27, 2012, through October 14, 2013. The final follow-up visit was May 21, 2014. The study group consisted of 43 participants (17 men and 26 women); mean (SD) age was 46.6 (13.3) years. The baseline characteristics of participants enrolled in the study are summarized in Table 1. None of the 43 participants recruited had intraocular pressures of greater than 21 mm Hg, were receiving glaucoma medication, or had undergone previous vitrectomy. Eight right eyes and 11 left eyes were pseudophakic at the start of the study, and no participant had visually significant cataracts or underwent cataract surgery during the study period. Diabetic retinopathy grade at the start of the study is reported in eTable 1 in the Supplement. Thirty-eight participants (88%) completed functional assessments at 6 months. Of the 5 participants who did not complete the 6-month follow-up, 2 were unavailable for follow-up despite multiple contact attempts, 2 moved abroad, and 1 was believed to be ineligible for the study inclusion criteria after completion of initial retinal sensitivity and imaging tests. Details of laser treatment are given in eResults in the Supplement.
All 43 participants underwent Esterman visual field testing for driving at baseline, of whom 41 (95%) passed in accordance with UK driving standards before PRP. At the 6-month follow-up, 3 of the 38 participants who completed the study did not pass the Esterman visual field test for driving, including the 2 participants who failed at baseline; hence, 35 (92%) passed the test. Concordance between the initial visual field grader (M.S.) and the adjudicator (A.V.) was 100%.
Forty-two participants at baseline and 37 at the 6-month follow-up underwent monocular full-field static testing. The binocular full-field static test was completed by 40 participants at baseline and 37 at the 6-month follow-up. Mean retinal sensitivity was calculated only for reliable visual field tests (reliability factor, <20). At baseline, the monocular and binocular fields of 3 participants were excluded owing to poor reliability. At the 6-month follow-up, 5 right and 4 left monocular fields and 1 binocular field were excluded. Results at baseline and the 6-month follow-up are presented in Table 2. We found a mean reduction in sensitivity of −1.4 (3.7) (95% CI, −2.7 to −0.1) dB OD and −2.4 (2.9) (95% CI, −3.4 to −1.4) dB OS for participants who completed the baseline and exit fields. The binocular reduction in sensitivity was −1.4 (3.6) (95% CI, −2.7 to −0.2) dB.
Forty participants completed Goldmann perimetry at baseline for at least 1 stimulus size, and 33 completed it at the 6-month follow-up. All participants were judged to have adequate to excellent fixation at the time of testing. Participant fatigue meant that not every stimulus size was plotted for every participant at every visit. Results for visual field area for different stimulus sizes are shown in Table 3. The greatest reduction in field area was for the I4e target, and for each stimulus size the reduction was greater in the left than the right eye.
Volumetric analysis of the HOV was conducted where possible on monocular static fields (138-point monocular) for each eye and on the summated binocular field using the aforementioned monocular field from both eyes. Analysis was also conducted on the binocular 120-point Esterman visual fields where available. Results are presented in Table 4. A 15% to 17% decrease was found in the central 30° field volume for right eyes (−2.8 [8.8]; 95% CI, −4.7 to 0.1 dB/steradian) and left eyes (−2.8 [5.2]; 95% CI, −4.6 to −1.0 dB/steradian) and in the total field volume for right eyes (−7.2 [13.0]; 95% CI, −11.7 to −2.7 dB/steradian) and for left eyes (−7.7 [11.2]; 95% CI, −11.6 to −3.8 dB/steradian). The decrease in binocular total field volume was greater (20.7%) when measured with the summated 138-point field (−13.9 [12.3]; 95% CI, −18.1 to −9.7 dB/steradian) than the 120-point Esterman visual field (10.2%) (−4.9 [10.9]; 95% CI, −9.4 to −0.5) dB/steradian.
This study demonstrates that at 6 months after bilateral PRP delivered using a multispot laser, a visual field compatible with driving in the UK is preserved in most patients. Only 1 participant in our cohort who passed the Esterman visual field test for driving at baseline did not pass after laser treatment. We performed detailed functional assessments demonstrating relative preservation of mean retinal sensitivity and minimal effect of laser on the binocular field of vision. Our cohort of participants had poorly controlled diabetes mellitus as evidenced by the mean hemoglobin A1c level and end-organ damage in the form of severe retinopathy and other diabetes mellitus–related comorbidities. Most participants with sight-threatening retinopathy were of working age, and half of the participants were drivers. We also present novel reading center–derived data on ETDRS diabetic retinopathy grading and laser coverage with multispot therapy.
Panretinal photocoagulation remains an effective treatment for PDR and severe nonproliferative diabetic retinopathy. The laser light is absorbed primarily by melanosomes within the retinal pigment epithelium, leading to coagulation of adjacent photoreceptors and retinal pigment epithelial cells that result in the formation of laser burns. Laser scar expansion may be associated with further photoreceptor loss, retinal pigment epithelial hypertrophy, and functional loss. The annual scar expansion may increase for several years after PRP, with additional damaging effects on the visual field.20 Visual field loss secondary to PRP has been evaluated in limited previous studies. Mackie et al6 investigated 100 patients retrospectively with PDR who required unilateral and bilateral PRP and found 19% of patients failed the Esterman visual field test for driving after treatment; no testing was performed before treatment for comparison. Estimates from a small number of studies reveal a large variation (12%-50%) in failure of the UK visual field test for driving after PRP.6- 11 Likely reasons for this variation include different study designs, sample sizes, photocoagulation methods (argon or xenon laser and laser treatment variables), unilateral and bilateral treatments, and visual field assessments. We have attempted to standardize these variables in our study.
During photocoagulation, the aim is to optimize the thermally induced therapeutic effect while minimizing collateral retinal damage.15,21 Laser-tissue interaction is influenced by wavelength, spot size, power, and exposure time; thus retinal damage can be reduced by changing these variables. With the reduced exposure time that multispot systems adopt, a higher power to achieve the desired therapeutic lesion is required. This reduced exposure time does not increase complications owing to the reduced laser energy per burn reaching the eye secondary to its shorter duration. Fluence is calculated as (power × time)/area and, provided the spot size remains unchanged with a burn duration of 20 milliseconds, the fluence is less than with a 100-millisecond burn, and therefore reduced diffusion of heat and subsequent collateral damage occur. A reduced exposure time with a grid pattern of application also has the benefit of being less painful with reduced treatment duration.15 With reduced thermal energy diffusion and collateral damage, photoreceptor damage, retinal pigment epithelial hypertrophy, and retinal scarring should be reduced; thus the effects on visual field loss would be predicted to be less pronounced.
Our study found no change in best-corrected visual acuity, despite a minor increase in the number of eyes with macular ischemia or clinically significant macular edema. Results of monocular and binocular static visual field testing demonstrate relative preservation of global mean retinal sensitivity after PRP, with a reduction of less than 2 dB for the monocular and binocular full-field static tests. This reduction is also demonstrated with kinetic monocular testing; however, left eyes undergoing testing with Goldmann perimetry show a greater reduction than right eyes. This greater change in left eyes for static and kinetic perimetry may reflect fatigue because the right eyes were tested first. Microperimetry reveals relative preservation of macular function after PRP for 12° and 4° areas, with a reduction in mean retinal sensitivity of 3.0 (5.2) dB OD and 2.6 (5.4) dB OS at the central 4° area (eTable 2 in the Supplement).
We also undertook more robust assessment of retinal sensitivity data by HOV modeling of the entire full-field static perimetry data set, which allowed comparison of the magnitude and extent of change before and after PRP by using this volumetric approach. This detailed analysis reveals relatively minor loss of retinal function after PRP (10%-20% reduction) when looking at monocular, summated monocular, and binocular static grid testing, with the greatest reduction noted with summated monocular fields. The recent Diabetic Retinopathy Clinical Research Network study12 undertook only static perimetry using the 30-2 and 60-4 tests of the Humphrey field analyzer, and greater than 20% loss to follow-up for visual field testing occurred in both groups at 2 years.
We found that by using predefined standard clinical definitions of laser efficacy (see eMethods in the Supplement), 31 of 38 participants (82%) had clinical evidence of a response to PRP. The ETDRS retinopathy grading scale does not distinguish between proliferating and regressing new vessels, meaning that the Moorfields Eye Hospital Reading Centre identified a significant number of participants who still had a PDR ETDRS grade after grading of wide-field color fundus photography at the 6-month follow-up (eTable 3 in the Supplement). Where the baseline image was graded as nonproliferative retinopathy, grade 60 (PRP scars present) was not used, to allow more sensitive grading of nonproliferative features. Reading Centre ETDRS grading is not used in clinical practice to gauge the effect of PRP. Laser coverage in the retinal periphery was more extensive than in a 2 × 2 disc area nasal to the optic disc. This finding may explain the relative preservation of visual field in this study, an area that is often relatively spared by many retinal specialists for this very reason.
A limitation of our study was the number of participants who failed to attend follow-up appointments to receive adequate PRP in a timely manner. This study had a dedicated study coordinator and the visits were in a dedicated clinical research facility, with the timing of visits tailored to optimize convenience for each participant. However, 5 participants (12%) did not complete the study. Our study presents results at 6 months after laser treatment. Disease progression and changes to laser scars could further alter the visual field during a longer follow-up.
The incidence and prevalence of diabetes mellitus are predicted to rise in Europe and North America during the next decade. Despite improvement in the management of diabetes mellitus, the incidence of PDR remains high.22,23 With the increasing worldwide use of multispot laser for application of PRP, ophthalmologists must be able to advise patients accurately on whether PRP for diabetic retinopathy may jeopardize their entitlement to a driving license, particularly because patients may receive a diagnosis earlier and hence live longer with the effects of treatment. This study identified no loss of driving eligibility on the basis of visual field criteria in almost all patients, although the confidence in this result is limited by 38 of the 43 enrolled participants being available for follow-up evaluation. Taking this into account, other measures of central and peripheral retinal sensitivity were also affected to a relatively limited extent.
Corresponding Author: Michel Michaelides, MD, FRCOphth, National Institute for Health Research Biomedical Research Centre at Moorfields Eye Hospital National Health Service Foundation Trust and University College London Institute of Ophthalmology, 11-43 Bath St, London EC1V 9EL, England (email@example.com).
Submitted for Publication: December 2, 2015; final revision received February 9, 2016; accepted February 15, 2016.
Published Online: April 14, 2016. doi:10.1001/jamaophthalmol.2016.0629.
Author Contributions: Drs Subash and Comyn contributed equally to this article and are joint first authors. Drs Wickham and Michaelides are joint senior authors. Dr Michaelides had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Subash, Comyn, Samy, Viswanathan, Weleber, Wickham, Michaelides.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Subash, Comyn, Samy, Antonakis, Mehat, Tee, Bunce.
Critical revision of the manuscript for important intellectual content: Comyn, Samy, Qatarneh, Mansour, Xing, Viswanathan, Rubin, Weleber, Peto, Wickham, Michaelides.
Statistical analysis: Subash, Comyn, Bunce, Viswanathan.
Obtained funding: Subash, Wickham, Michaelides.
Administrative, technical, or material support: Subash, Samy, Antonakis, Mehat, Tee, Mansour, Viswanathan, Rubin, Weleber, Peto, Wickham, Michaelides.
Study supervision: Subash, Samy, Viswanathan, Weleber, Peto, Wickham Michaelides.
Conflict of Interest Disclosures: All authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Weleber reports holding US patent 8 657 446 for Visual Field and Modeling Analysis (VFMA) software, the intellectual property rights for which are owned by the Oregon Health and Science University (OHSU), and serving on a scientific advisory board for the Foundation Fighting Blindness, a relationship that has been reviewed and managed by OHSU. No other disclosures were reported.
Funding/Support: This study was supported by Foundation Fighting Blindness, grant 099173/Z/12/Z from the Wellcome Trust, the National Institute for Health Research Biomedical Research Centre at Moorfields Eye Hospital National Health Service Foundation Trust and University College London Institute of Ophthalmology, the Special Trustees of Moorfields Eye Hospital, the Insulin Dependent Diabetes Trust, Valon Lasers, and Optos plc.
Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Previous Presentation: This report was presented in part at the 2014 Annual Meeting of the Association for Research in Vision and Ophthalmology; May 6, 2014; Orlando, Florida.
Additional Contributions: Robin Hamilton, DM, FRCOphth, and Phil Hykin, MD, FRCOphth, Moorfields Eye Hospital, provided constructive advice during protocol development. They received no compensation for this role.