PMCCPMCCPMCC

Search tips
Search criteria 

Advanced

 
Logo of bmjclinevidLink to Publisher's site
 
BMJ Clin Evid. 2007; 2007: 0702.
Published online 2007 November 23.
PMCID: PMC2943811

Diabetic retinopathy

Abstract

Introduction

Diabetic retinopathy is the most common cause of blindness in the UK, with older people and those with worse diabetic control, hypertension, and hyperlipidaemia being most at risk. Diabetic retinopathy can cause microaneurysms, haemorrhages, exudates, changes to blood vessels, and retinal thickening.

Methods and outcomes

We conducted a systematic review and aimed to answer the following clinical questions: What are the effects of treatments in people with diabetic retinopathy? What are the effects of treatments for vitreous haemorrhage? We searched: Medline, Embase, The Cochrane Library and other important databases up to November March 2007 (BMJ Clinical Evidence reviews are updated periodically, please check our website for the most up-to-date version of this review). We included harms alerts from relevant organisations such as the US Food and Drug Administration (FDA) and the UK Medicines and Healthcare products Regulatory Agency (MHRA).

Results

We found 29 systematic reviews, RCTs, or observational studies that met our inclusion criteria. We performed a GRADE evaluation of the quality of evidence for interventions.

Conclusions

In this systematic review we present information relating to the effectiveness and safety of the following interventions: peripheral retinal laser photocoagulation, focal and grid laser photocoagulation for maculopathy, corticosteroids for macular oedema, and vitrectomy for vitreous haemorrhage.

Key Points

Diabetic retinopathy is the most common cause of blindness in the UK, with older people and those with worse diabetic control, hypertension, and hyperlipidaemia most at risk.

  • Diabetic retinopathy can cause microaneurysms, haemorrhages, exudates, changes to blood vessels, and retinal thickening.

Peripheral retinal laser photocoagulation reduces the risk of severe visual loss compared with no treatment in people with preproliferative (moderate/severe non-proliferative) retinopathy and maculopathy.

  • We don't know if any one type of laser treatment is superior.
  • We don't know whether peripheral laser photocoagulation is beneficial in people with background or preproliferative (non-proliferative) retinopathy without maculopathy.

The benefits of laser photocoagulation are more notable in people with proliferate retinopathy than in those with maculopathy.

  • Focal macular laser photocoagulation reduces the risk of moderate visual loss in eyes with clinically significant macular oedema plus mild to moderate preproliferative (moderate/severe non-proliferative) diabetic retinopathy, compared with no treatment.
  • Grid photocoagulation to zones of retinal thickening may improve visual acuity in eyes with diffuse maculopathy.
  • Photocoagulation is unlikely to be beneficial in eyes with maculopathy but without clinically significant macular oedema.

Intravitreal triamcinolone acetonide improves visual acuity and reduces macular thickness in eyes with macular oedema refractory to previous macular laser photocoagulation, but repeated injections are needed to maintain benefit.

  • Secondary ocular hypertension and progression of cataract are common complications with intravitreal triamcinolone; infectious endophthalmitis is rare.

Vitrectomy can reduce visual loss if performed early in people with vitreous haemorrhage, especially if they have severe proliferative retinopathy.

  • We don't know whether vitrectomy is beneficial in people with vitreous haemorrhage plus maculopathy.

About this condition

Definition

Diabetic retinopathy is characterised by varying degrees of microaneurysms, haemorrhages, exudates (hard exudates), venous changes, new vessel formation, and retinal thickening. It can involve the peripheral retina, the macula, or both. The range of severity of retinopathy includes background (mild non-proliferative), preproliferative (moderate/severe non-proliferative), proliferative and advanced retinopathy. Involvement of the macula can be focal, diffuse, ischaemic, or mixed (see table 1 ).

Table 1
Equivalent UK and US terminology, where different.

Incidence/ Prevalence

Diabetic eye disease is the most common cause of blindness in the UK, responsible for 12% of registrable blindness in people aged 16-64 years.

Aetiology/ Risk factors

Risk factors include age, duration and control of diabetes, raised blood pressure, and hyperlipidaemia.

Prognosis

Natural history studies from the 1960s found that at least half of people with proliferative diabetic retinopathy progressed to Snellen visual acuity of less than 6/60 (20/200) within 3-5 years. After 4 years' follow-up, the rate of progression to less than 6/60 (20/200) visual acuity in the better eye was 1.5% in people with type 1 diabetes, 2.7% in people with non-insulin-dependent type 2 diabetes, and 3.2% in people with insulin-dependent type 2 diabetes.

Aims of intervention

To prevent visual disability, partial sight, and blindness; to improve quality of life, with minimum adverse effects.

Outcomes

Visual acuity (measured using a Snellen chart, unless otherwise stated). Incidence of visual disability (visual acuity 6/24 [20/80] or worse in the better eye), partial sight registration (visual acuity 6/60 [20/200] or worse in the better eye), and registrable blindness (visual acuity 3/60 [10/200] or worse in the better eye). Much of the published data used eyes, rather than people, as the unit of analysis. Clinically important loss of vision is often defined as loss of two or more Snellen lines of acuity (vision measured on standard Snellen chart) roughly equivalent to doubling of the visual angle (visual angle is the angle subtended at the eye of the smallest letter visible by that eye) — a measure used extensively in research.

Methods

BMJ Clinical Evidence search and appraisal March 2007. The following databases were used to identify studies for this systematic review: Medline 1966 to March 2007, Embase 1980 to March 2007, and The Cochrane Database of Systematic Reviews and Cochrane Central Register of Controlled Clinical Trials 2007, Issue 1. Additional searches were carried out using these websites: NHS Centre for Reviews and Dissemination (CRD) — for Database of Abstracts of Reviews of Effects (DARE) and Health Technology Assessment (HTA), Turning Research into Practice (TRIP), and National Institute for Health and Clinical Excellence (NICE). We also searched for retractions of studies included in the Review. Abstracts of the studies retrieved from the initial search were assessed by an information specialist. Selected studies were then sent to the author for additional assessment, using predetermined criteria to evaluate relevant studies. Study design criteria for inclusion in this review were: published systematic reviews and RCTs in any language, containing more than 20 individuals, of whom more than 80% were followed up. There was no minimum length of follow-up required to include studies. Open studies were included. Additional papers were identified from manual searches performed by a previous contributor up to November 2004. In addition, we use a regular surveillance protocol to capture harms alerts from organisations such as the US Food and Drug Administration (FDA) and the UK Medicines and Healthcare products Regulatory Agency (MHRA), which are added to the reviews as required. We have performed a GRADE evaluation of the quality of evidence for interventions included in this review (see table ).

Table
GRADE evaluation of interventions for diabetic retinopathy

Glossary

Advanced retinopathy
Retinopathy characterised by traction retinal detachment, vitreous haemorrhage obscuring fundus details, or both.
Background retinopathy
(mild non-proliferative) Characterised by microaneurysms, small haemorrhages, and exudates (hard exudates).
Diffuse exudative maculopathy
Characterised by thickened oedematous retina at the fovea, often with cystic changes.
Focal exudative maculopathy
Characterised by exudates (hard exudates) within one disc diameter of the centre of the fovea or circinate rings of exudates (hard exudates) within the macula.
High-quality evidence
Further research is very unlikely to change our confidence in the estimate of effect.
Ischaemic maculopathy
Characterised by zones of capillary non-perfusion visible only on fluorescein angiography but often inferred from presence of deep blot haemorrhages within the fovea.
Low-quality evidence
Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Moderate-quality evidence
Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Preproliferative retinopathy
Mild, moderate, or severe (moderate or severe non-proliferative) depending on number/location of lesions; characterised by cotton wool spots, deep round haemorrhages, venous beading, loops and reduplication, and intraretinal microvascular anomalies.
Proliferative retinopathy
Characterised by new vessels at the disc or elsewhere.
Snellen visual acuity
The Snellen chart usually includes letters, numbers, or pictures printed in lines of decreasing size, which are read or identified from a fixed distance; distance visual acuity is usually measured from a distance of 6 m (20 feet). The Snellen visual acuity is written as a fraction: 6/18 means that from 6 m away the best line that can be read is a line that could normally be read from a distance of 18 m away.
Very low-quality evidence
Any estimate of effect is very uncertain.
Vitrectomy
The vitreous is the normally clear gelatinous material that fills most of the inside of the eye. The vitreous can be affected by bleeding, inflammatory cells, debris, or scar tissue. Vitrectomy involves removal of the abnormal vitreous material.

Notes

Disclaimer

The information contained in this publication is intended for medical professionals. Categories presented in Clinical Evidence indicate a judgement about the strength of the evidence available to our contributors prior to publication and the relevant importance of benefit and harms. We rely on our contributors to confirm the accuracy of the information presented and to adhere to describe accepted practices. Readers should be aware that professionals in the field may have different opinions. Because of this and regular advances in medical research we strongly recommend that readers' independently verify specified treatments and drugs including manufacturers' guidance. Also, the categories do not indicate whether a particular treatment is generally appropriate or whether it is suitable for a particular individual. Ultimately it is the readers' responsibility to make their own professional judgements, so to appropriately advise and treat their patients.To the fullest extent permitted by law, BMJ Publishing Group Limited and its editors are not responsible for any losses, injury or damage caused to any person or property (including under contract, by negligence, products liability or otherwise) whether they be direct or indirect, special, incidental or consequential, resulting from the application of the information in this publication.

Notes

Control of hypertension to prevent diabetic retinopathy (see review on hypertension in diabetes).

Surgical treatments for age related cataract in people with diabetic retinopathy (see review on cataract).

Contributor Information

Efstratios Mendrinos, Department of Ophthalmology, Vitreo-retinal Unit, Geneva University Hospitals, Geneva, Switzerland.

Alexandros Stangos, Department of Ophthalmology, Vitreo-retinal Unit, Geneva University Hospitals, Geneva, Switzerland.

Please enter your position here Constantin Pournaras, Department of Ophthalmology, Vitreo-retinal Unit, Geneva University Hospitals, Geneva, Switzerland.

References

1. Evans J, Rooney C, Ashwood F, et al. Blindness and partial sight in England and Wales: April 1990–March 1991. Health Trends 1996;28:5–12.
2. Ebeling P, Koivisto VA. Occurrence and interrelationships of complications in insulin-dependent diabetes in Finland. Acta Diabetol 1997;34:33–38. [PubMed]
3. Beetham WP. Visual prognosis of proliferating diabetic retinopathy. Br J Ophthalmol 1963;47:611–619. [PMC free article] [PubMed]
4. Caird FI, Burditt AF, Draper GJ. Diabetic retinopathy: a further study of prognosis for vision. Diabetes 1968;17:121–123. [PubMed]
5. Deckert T, Simonsen SE, Poulsen JE. Prognosis of proliferative retinopathy in juvenile diabetes. Diabetes 1967;10:728–733. [PubMed]
6. Klein R, Klein BEK, Moss SE. The Wisconsin epidemiologic study of diabetic retinopathy: an update. Aust NZ J Ophthalmol 1990;18:19–22. [PubMed]
7. Patz A, Schatz H, Berkow JW, et al. Macular edema — an overlooked complication of diabetic retinopathy. Trans Am Acad Ophthalmol Otol 1973;77:34–42. [PubMed]
8. Hercules BL, Gayed II, Lucas SB, et al. Peripheral retinal ablation in the treatment of proliferative diabetic retinopathy: a three-year interim report of a randomised, controlled study using the argon laser. Br J Ophthalmol 1977;61:555–563. [PMC free article] [PubMed]
9. Diabetic Retinopathy Study Research Group. DRS group 8: photocoagulation treatment of proliferative diabetic retinopathy. Ophthalmology 1981;88:583–600. [PubMed]
10. Early Treatment Diabetic Retinopathy Study Research Group. Pars plana vitrectomy in the Early Treatment Diabetic Retinopathy Study. ETDRS report number 17. Ophthalmology 1992;99:1351–1357. [PubMed]
11. Early Treatment Diabetic Retinopathy Study Research Group. Early photocoagulation for diabetic retinopathy: ETDRS report number 9. Ophthalmology 1991;98:766–785. [PubMed]
12. Ferris F. Early photocoagulation in patients with either type I or type II diabetes. Trans Am Ophthalmol Soc 1996;94:505–537. [PMC free article] [PubMed]
13. Bandello F, Brancato R, Menchini U, et al. Light panretinal photocoagulation (LPRP) versus classic panretinal photocoagulation (CPRP) in proliferative diabetic retinopathy. Semin Ophthalmol 2001;16:12–18. [PubMed]
14. Bandello F, Brancato R, Lattanzio R, et al. Double-frequency Nd:YAG laser vs argon-green laser in the treatment of proliferative diabetic retinopathy: randomized study with long-term follow-up. Lasers Surg Med 1996;19:173–176. [PubMed]
15. The Krypton Argon Regression Neovascularization Study Research Group. Randomized comparison of krypton versus argon scatter photocoagulation for diabetic disc neovascularization: the krypton argon regression neovascularization study report number 1. Ophthalmology 1993;100:1655–1664. [PubMed]
16. Doft BH, Metz DJ, Kelsey SF. Augmentation laser for proliferative diabetic retinopathy that fails to respond to initial panretinal photocoagulation. Ophthalmology 1992;99:1728-1734. [PubMed]
17. Pearson AR, Tanner V, Keightey SJ, et al. What effect does laser photocoagulation have on driving visual fields in diabetics? Eye 1998;12:64–68. [PubMed]
18. Theodossiadis GP. Central visual field changes after panretinal photocoagulation in proliferative diabetic retinopathy. Ophthalmologica 1990;201:71–78. [PubMed]
19. Mackie SW, Walsh G. Contrast and glare sensitivity in diabetic patients with and without pan-retinal photocoagulation. Ophthalmic Physiol Opt 1998;18:173–181. [PubMed]
20. Khosla PK, Rao V, Tewari HK, et al. Contrast sensitivity in diabetic retinopathy after panretinal photocoagulation. Ophthalmic Surg 1994;25:516–520. [PubMed]
21. Birch J, Hamilton AM. Xenon arc and argon laser photocoagulation in the treatment of diabetic disc neovascularization. Part 2: effect on colour vision. Trans Ophthalmol Soc UK 1981;101:93–99. [PubMed]
22. Krill AE, Archer DB, Newell FW, et al. Photocoagulation in diabetic retinopathy. Am J Ophthalmol 1971;72:299–321. [PubMed]
23. Doft BH. Single versus multiple treatment sessions of argon laser panretinal photocoagulation for proliferative diabetic retinopathy. Ophthalmology 1982;89:772–779. [PubMed]
24. Seiberth V, Schatanek S, Alexandridis E. Panretinal photocoagulation in diabetic retinopathy: argon versus dye laser coagulation. Graefes Arch Clin Exp Ophthalmol 1993;231:318–322. [PubMed]
25. Cordeiro MF, Stanford MR, Phillips PM, et al. Relationship of diabetic microvascular complications to outcome in panretinal photocoagulation treatment of proliferative diabetic retinopathy. Eye 1997;11:531–536. [PubMed]
26. Duffy SW, Rohan TE, Altman DG. A method for combining matched and unmatched binary data: application to randomized, controlled trials of photocoagulation in the treatment of diabetic retinopathy. Am J Epidemiol 1989;130:371–378. [PubMed]
27. Blankenship GW. Diabetic macular edema and argon laser photocoagulation: a prospective randomized study. Ophthalmology 1979;86:69–75. [PubMed]
28. Buckley S. Field loss after pan retinal photocoagulation with diode and argon lasers. Doc Ophthalmol 1992;82:317–322. [PubMed]
29. Early Treatment Diabetic Retinopathy Study Research Group. Photocoagulation for diabetic macular edema. Arch Ophthalmol 1985;103:1796–1806. [PubMed]
30. Early Treatment Diabetic Retinopathy Study Research Group. Focal photocoagulation treatment of diabetic macular edema: relationship of treatment effect to fluorescein angiographic and other retinal characteristics at baseline: ETDRS report number 19. Arch Ophthalmol 1995;113:1144–1155. [PubMed]
31. Bandello F, Polito A, Del BorrelloM, et al. "Light" versus "classic" laser treatment for clinically significant diabetic macular oedema. B J Ophthalmol 2005;89:864–870. [PMC free article] [PubMed]
32. Ciavarella P, Moretti G, Falsini B, et al. The pattern electroretinogram (PERG) after laser treatment of the peripheral or central retina. Curr Eye Res 1997;16:111–115. [PubMed]
33. Olk RJ. Modified grid argon (blue–green) laser photocoagulation for diffuse diabetic macular edema. Ophthalmology 1986;93:938–950. [PubMed]
34. Akduman L, Olk RJ. Diode laser (810 nm) versus argon green (514 nm) modified grid photocoagulation for diffuse diabetic macular edema. Ophthalmology 1997;104:1433–1441. [PubMed]
35. Gillies MC, Sutter FK, Simpson JM, et al. Intravitreal triamcinolone for refractory diabetic macular edema: two-year results of a double-masked, placebo-controlled, randomized clinical trial. Ophthalmology 2006;113:1533–1538. [PubMed]
36. Jonas JB, Kamppeter BA, Harder B, et al. Intravitreal triamcinolone acetonide for diabetic macular edema: A prospective, randomized study. J Ocul Pharmacol Ther 2006;22:200–207. [PubMed]
37. Lam DS, Chan CK, Mohamed S, et al. A prospective randomised trial of different doses of intravitreal triamcinolone for diabetic macular oedema. Br J Ophthalmol 2007;91:199–203. [PMC free article] [PubMed]
38. Audren F, Lecleire-Collet A, Erginay A, et al. Intravitreal triamcinolone acetonide for diffuse diabetic macular edema: phase 2 trial comparing 4 mg vs 2 mg. Am J Ophthalmol 2006;142:794–799. [PubMed]
39. Entezari M, Ahmadieh H, Dehghan MH, et al. Posterior sub-tenon triamcinolone for refractory diabetic macular edema: a randomized clinical trial. Eur J Ophthalmol 2005;15:746–750. [PubMed]
40. Jonas JB. Intravitreal triamcinolone acetonide: a change in a paradigm. Ophthalmic Res 2006;38:218–245. [PubMed]
41. Early vitrectomy for severe vitreous hemorrhage in diabetic retinopathy. Two-year results of a randomized trial. Diabetic Retinopathy Vitrectomy Study report 2. The Diabetic Retinopathy Vitrectomy Study Research Group. Arch Ophthalmol 1985;103:1644–1652. [PubMed]
42. Diabetic Retinopathy Vitrectomy Study Group. Early vitrectomy for severe vitreous hemorrhage in diabetic retinopathy. Four-year results of a randomized trial. Diabetic Retinopathy Vitrectomy Study report 5. Arch Ophthalmol 1990;108:958–964. [PubMed]
43. The Diabetic Retinopathy Vitrectomy Study Research Group. Early vitrectomy for severe proliferative diabetic retinopathy in eyes with useful vision. Results of a randomized trial. Diabetic Retinopathy Vitrectomy Study report 3. Ophthalmology 1988;95:1307–1320. [PubMed]
44. Le Mer Y, Korobelnik JF, Morel C, et al. TPA-assisted vitrectomy for proliferative diabetic retinopathy: results of a double-masked, multicenter trial. Retina 1999;19:378–382. [PubMed]
45. Sima P, Zoran T. Long-term results of vitreous surgery for proliferative diabetic retinopathy. Doc Ophthalmol 1994;87:223–232. [PubMed]
2007; 2007: 0702.
Published online 2007 November 23.

Peripheral retinal laser photocoagulation in people with preproliferative (moderate/severe non-proliferative) retinopathy and maculopathy

Summary

VISUAL ACUITY Compared with no treatment: Peripheral retinal photocoagulation is more effective at 26 months to 5 years than no treatment at reducing the risk of severe visual loss in people with preproliferative retinopathy and diabetic maculopathy ( high-quality evidence ).

Benefits

Peripheral retinal laser photocoagulation versus no treatment:

We found no systematic review, but found one RCT in people with preproliferative (moderate/severe non-proliferative) diabetic retinopathy, most of whom had diabetic maculopathy, which found that peripheral photocoagulation significantly reduced the risk of visual deterioration at 5 years compared with no treatment (see table 2 ).

Table 2
RCTs of peripheral photocoagulation versus no treatment in people with diabetic retinopathy.

Harms

See harms of Peripheral retinal laser photocoagulation in people with proliferative retinopathy.

Comment

See comment on Peripheral retinal laser photocoagulation in people with proliferative retinopathy.

Substantive changes

No new evidence

2007; 2007: 0702.
Published online 2007 November 23.

Peripheral retinal laser photocoagulation in people with proliferative retinopathy

Summary

VISUAL ACUITY Compared with no treatment: Peripheral retinal laser photocoagulation reduces risk of severe visual loss at 3–5 years in people with proliferative retinopathy compared with no treatment ( high-quality evidence ). Low-intensity compared with classic peripheral laser photocoagulation: Low-intensity peripheral laser photocoagulation is no more effective at 22 months than classic peripheral laser photocoagulation at improving visual acuity in people with high-risk proliferative diabetic retinopathy ( moderate-quality evidence ). Supplemental compared with no supplemental peripheral laser photocoagulation: Supplemental peripheral laser photocoagulation may be no more effective at 1 year than no supplemental peripheral laser photocoagulation at improving visual acuity in people who have not responded to initial photocoagulation ( very low-quality evidence ). RATES OF REGRESSION Low-intensity compared with classic peipheral laser photocoagulation: Low-intensity peripheral laser photocoagulation is no more effective at 22 months than classic peripheral laser photocoagulation at reducing the risk of regression in people with high-risk proliferative diabetic retinopathy (moderate-quality evidence). Different types of laser compared with each other: Different types of lasers are no more efffective at 3–29 months than each other at reducing the rate of regression of new blood vessels in people with proliferative diabetic retinopathy (moderate-quality evidence). REDUCTION OF RISK FACTORS Supplemental compared with no supplemental peripheral laser photocoagulation: Supplemental peripheral laser photocoagulation may not reduce risk factors of retinopathy at 1 year in people who have not responded to initial photocoagulation compared with no supplemental peripheral laser photocoagulation (very low-quality evidence).

Benefits

Peripheral retinal laser photocoagulation versus no treatment:

We found no systematic review, but found three RCTs published in five publications (see table 2 ), which recruited people with different grades of diabetic retinopathy, and compared different regimens of peripheral retinal photocoagulation versus no treatment or versus deferred treatment. One of the RCTs recruited only people with proliferative diabetic retinopathy and found that peripheral photocoagulation significantly reduced the risk of blindness after 2 or 3 years compared with no treatment (see table 2 ).Two large RCTs (reported in 4 publications) recruited people with either preproliferative (moderate/severe non-proliferative) or proliferative retinopathy. Both found that early photocoagulation decreased the risk of severe visual loss at 5 years compared with no early photocoagulation, but in one of the RCTs the rate of severe visual loss was low and the effect was not significant (see table 2 ). Subgroup analysis in one of these RCTs found that the benefit was significant in people with type 2 diabetes and with severe preproliferative (severe non-proliferative) or early proliferative retinopathy without high-risk characteristics (data presented graphically).

Low-intensity versus classic peripheral laser photocoagulation:

We found no systematic reviews, but found one RCT (50 people [65 eyes] with high-risk proliferative diabetic retinopathy.) The RCT compared low-intensity peripheral laser photocoagulation (mean power 235 mW with a very light biomicroscopic effect on the retina for each laser spot) versus classic peripheral laser photocoagulation (mean power 420 mW with a white–yellow biomicroscopic effect on the retina for each laser spot). It found no significant difference in visual acuity between groups at around 22 months (final mean logMAR visual acuity [mean follow-up]: 0.18 [22.4 months] with low-intensity treatment v 0.27 [21.6 months] with classic treatment; P = 0.231). It also found no significant difference between groups in the proportion of people with regression of high-risk proliferative diabetic retinopathy at the end of the follow-up (31/34 [91%] eyes with low-intensity treatment v 30/31 [97%] eyes with classic treatment; P = 0.615).

Different types of laser:

We found no systematic review, but found two RCTs. The first RCT (33 people, 42 eyes with proliferative diabetic retinopathy) compared argon laser versus double frequency YAG laser, and found similar rates of regression of new vessels after mean follow-up of 29 months (number of eyes improved after laser treatment: 20/21 [95.2%] with argon laser v 20/21 [95.2%] with YAG laser; P value not reported). The second RCT (multicentre, 696 people, 907 eyes with proliferative diabetic retinopathy with new vessels on the disc) found no significant difference in rates of regression of new vessels at 3 months between krypton red and argon laser (proportion with regression of neovascularisation to less than a third of disc area: 41.8% with krypton laser v 41.8% with argon laser; P = 0.92).

Supplemental versus no supplemental peripheral laser photocoagulation in non-responders to initial photocoagulation:

We found no systematic review but found one poorly-designed RCT (sealed-envelope randomisation; 35 people with visual acuity of 20/200 or better, and three or more retinopathy risk factors: vitreous or preretinal haemorrhage, new vessels, location of new vessels on the optic disc, or moderate or severe extent of new vessels; see comment below).The RCT compared supplemental photocoagulation (minimum of 500 additional burns of argon green laser filling in between previous laser burns with efforts not to re-treat previous spots) versus no supplemental photocoagulation in people who had not responded to initial peripheral retinal photocoagulation (response defined as regression to two or fewer retinopathy risk factors at 3 weeks after initial treatment). It found similar visual acuity measures between groups at 1 year (proportion of people with 20/50 or better visual acuity: 50% with supplemental treatment v 53% without supplemental treatment; proportion of people with 20/200 or better visual acuity: 75% with supplemental treatment v 74% without supplemental treatment; P values and significance not reported). It found no significant difference between groups in the mean reduction of retinopathy risk factors at 1 year (mean change in number of retinopathy risk factors: –1.12 with supplemental photocoagulation v –1.05 with no supplemental photocoagulation; P value not reported; reported as not significant). In addition, for all study participants, the persistence of three or more retinopathy risk factors 1 year after treatment was associated with a poorer visual result (proportion of people with traction retinal detachment of more than four disc ares in size, vitreous haemorrhage blocking a quarter or more of the retinal area, or visual acuity worse than 20/400: 35% in people with three or more risk factors at 1 year v 6% in people with fewer than three risk factors at one year; significance not reported).

Harms

Adverse effects of photocoagulation include loss of visual field and visual acuity, increased glare, reduced contrast and colour sensitivity, temporary choroidal effusion, exudative retinal detachment, retinal haemorrhage or partial obstruction of retinal vessels caused by inadvertent photocoagulation of middle- and large-sized retinal vessels, anterior uveitis, worsening macular oedema, and pain during treatment. Most studies were too small to provide accurate estimates of the frequency of these adverse effects, and they may overestimate the risks, as they used old treatment protocols.

Peripheral retinal laser photocoagulation versus no treatment:

In one RCT, using an argon treatment protocol that has since been modified in current practice, constriction of visual field to within 45° of fixation occurred in 5% of eyes, constriction within 30° in 0%, and loss of vision by two or more Snellen lines in 3%.

Fractionation:

One RCT found that adverse effects (including exudative retinal detachment, choroidal detachment, and angle closure) were reduced if photocoagulation was administered in multiple sessions spaced over time rather than in a single session.

Low-intensity versus classic peripheral laser photocoagulation:

The RCT found significantly lower rates of vitreous haemorrhage, and of appearance or worsening of clinically significant macular oedema (CSME) with low-intensity peripheral laser photocoagulation compared with classic treatment (vitreous haemorrhage: 0/34 [0%] eyes with low-intensity treatment v 6/31 [19%] eyes with classic treatment; P = 0.009; CSME: 1/34 [3%] eyes with low-intensity treatment v 7/31 [23%] eyes with classic treatment; P = 0.023). Low-intensity treatment was also associated with lower rates of other complications, although differences between groups were not significant (choroidal detachment: 0/34 [0%] eyes with low-intensity treatment v 3/31 [10%] eyes with classic treatment; P = 0.103; troublesome pain: 1/34 [3%] eyes with low-intensity treatment v 4/31 [13%] eyes with classic treatment; P = 0.184; neurotropic keratopathy: 0/34 [0%] eyes with low-intensity treatment v 2/31 [6%] eyes with classic treatment; P = 0.224). Low-intensity treatment required significantly fewer treatment sessions (mean total number of sessions: 7.4 with low-intensity treatment v 9.9 with classic treatment; P less than 0.001).

Different types of laser:

We found no clear evidence of different rates of complications with different lasers. Dye laser and orange laser (600 nm) may be more painful than argon for peripheral retinal photocoagulation.

Supplemental versus no supplemental peripheral laser photocoagulation in non-responders to initial photocoagulation:

The RCT found a similar proportion of complications between groups at 1 year (vitreous haemorrhage: 7 eyes with supplemental treatment v 7 eyes without; rubeosis: 1 person with supplemental treatment v 1 person without; retinal detachment: 2 people with supplemental treatment v 2 people without; need for additional surgery: 2 eyes with supplemental treatment v 2 eyes without; P values and significance not reported).

Comment

We also found one meta-analysis comparing photocoagulation versus no treatment for diabetic retinopathy, but its results are difficult to interpret because it was not based on a published systematic review, it did not include the largest RCT, and it included one RCT of macular photocoagulation.

Supplemental versus no supplemental peripheral laser photocoagulation in non-responders to initial photocoagulation:

The RCT had a number of design limitations: additional photocoagulation treatment was not standardised; visual acuity examination was not obtained by refracting the person at the time it was measured; and the number of people included in the study was too small to allow the statistical power necessary to detect small but clinically important differences between groups. Because of these limitations, the authors of the RCT did not advocate abandoning the current clinical practice of additional photocoagulation in eyes that fail to respond to initial treatment.

Clinical guide:

We found no good evidence of a harmful or beneficial effect on visual acuity with additional peripheral retinal photocoagulation. Limited prospective observational data suggest that peripheral retinal photocoagulation should be repeated until there is evidence of regression. We found no evidence that theoretical advantages with certain lasers are reflected in significant improvements in clinical outcomes. Studies of visual field loss do not consider field loss before laser photocoagulation; one study found significant field loss in people with diabetes before laser compared with people without diabetes (P less than 0.01).

Substantive changes

Peripheral retinal laser photocoagulation in people with proliferative retinopathy One RCT added;categorisation unchanged (Beneficial).

2007; 2007: 0702.
Published online 2007 November 23.

Peripheral retinal laser photocoagulation in people with background or preproliferative (non-proliferative) retinopathy without maculopathy

Summary

We found no direct results about the effects of peripheral retinal photocoagulation in people with background or preproliferative retinopathy without maculopathy.

Benefits

Peripheral retinal laser photocoagulation versus no treatment:

We found no RCTs of photocoagulation in people with preproliferative (moderate/severe non-proliferative) retinopathy who have not yet developed maculopathy.

Harms

We found no RCTs. See also harms of peripheral retinal laser photocoagulation in people with proliferative retinopathy.

Comment

See comment on peripheral retinal laser photocoagulation in people with proliferative retinopathy.

Substantive changes

No new evidence

2007; 2007: 0702.
Published online 2007 November 23.

Photocoagulation in people with clinically significant macular oedema

Summary

VISUAL ACUITY Compared with no treatment: Focal macular laser photocoagulation reduces visual loss at 2–3 years in people with macular oedema plus mild to moderate non-proliferative diabetic retinopathy compared with no treatment ( high-quality evidence ). Low-intensity compared with classic macular photocoagulation: Low-intensity macular photocoagulation may not improve visual outcomes at 12 months compared with classic macular photocoagulation ( very low-quality evidence ). NOTE Focal macular laser photocoagulation seems to be beneficial if eyes have better vision and clinically significant macular oedema, especially if the centre of the macula is involved or imminently threatened.

Benefits

Macular photocoagulation versus no treatment:

We found no systematic reviews.

Focal macular laser photocoagulation treatment versus no treatment:

We found two RCTs comparing focal macular argon laser photocoagulation versus no treatment in eyes with macular oedema plus mild to moderate diabetic retinopathy. The first RCT (39 people with symmetrical macular oedema and background [mild non-proliferative] diabetic retinopathy), found no significant difference between photocoagulation and no treatment in the incidence of visual deterioration after 2 years, but the study may have lacked power to detect a clinically important difference (visual deterioration of completing eyes: 7/30 [23%] eyes with laser v 13/30 [43%] eyes with no treatment; RR 0.54, 95% CI 0.25 to 1.16).The second and much larger RCT (2244 eyes with macular oedema plus mild to moderate non-proliferative diabetic retinopathy) compared focal laser treatment using an argon laser (754 eyes) versus no treatment (1490 eyes). It found that, after 3 years, laser photocoagulation significantly reduced moderate visual loss compared with no treatment (RR 0.50, 95% CI 0.47 to 0.53; NNT 8 eyes, 95% CI 7 eyes to 12 eyes). Subgroup analysis found that the benefits increased further in eyes with clinically significant macular oedema, particularly in people in whom the centre of the macula was involved or imminently threatened.

Low-intensity versus classic macular photocoagulation:

We found no systematic review, but found one RCT. The RCT (randomisation method not reported, 24 people [29 eyes] with mild to moderate non-proliferative diabetic retinopathy and clinically significant macular oedema) compared low-intensity (median power 50 mW with barely visible burns at the level of the retinal pigment epithelium) versus classic (median power 140 mW) Nd:YAG 532 nm (frequency doubled) green laser photocoagulation. Visual acuity outcomes were assessed using mean scores for the Early Treatment Diabetic Retinopathy Study Research Group (ETDRS) visual acuity chart read at 4 metres and the Pelli–Robson chart read at 1 metre. The RCT found no significant difference between groups in outcomes at 12 months — see comment below (proportion of people with reduction-elimination of oedema: 5/15 [33%] with low intensity v 10/14 [71%] with classic; P = 0.09; proportion of people with an increase or 5 letters or more in visual acuity: 5/15 [33%] with low intensity v 5/14 [36%] with classic; P = 1.00; proportion of people with a decrease or 5 letters or more in visual acuity: 2/15 [13%] with low intensity v 2/14 [14%] with classic; P = 1.00; number of letters change in contrast sensitivity: −0.27 with low intensity v −1.21 with classic; P = 0.56; mean deviation in the central 10 degrees visual field: −0.03 with low intensity v −0.04 with classic; P = 0.99; proportion of people with increased foveal thickness: 5/16 [33%] with low intensity v 3/14 [21%] with classic; P = 0.79).

Different types of laser treatments versus each other:

We found no systematic review or RCTs.

Harms

Uncontrolled studies reported that loss of contrast sensitivity and visual acuity occurred after direct application of the laser to the centre of the fovea. We found no accurate estimates of the frequency of adverse effects.

Focal laser photocoagulation treatment:

The largest RCT found no significant difference in the frequency of immediate visual loss, visual field or color vision scores in treated compared with untreated people (reported as non-significant, no further data reported).

Low-intensity versus classic macular photocoagulation:

The RCT found no significant difference between groups in the proportion of people who needed to be re-treated at 3 or 6 months (proportion of people re-treated, at 3 months: 3/15 [20%] with low intensity v 1/14 [7%] with classic; P = 0.60; at 6 months: 3/15 [20%] with low intensity v 4/14 [29%] with classic; P = 0.68).

Different types of laser treatments versus each other:

We found no systematic reviews or RCTs.

Comment

One prospective observational study reported a 40% reduction in macular function measured using the pattern electroretinogram in people undergoing focal argon paramacular treatment.Other complications include laser damage to the centre of the fovea, and induction of choroidal neovascularisation, but we found no reliable data on frequency.

Low-intensity versus classic macular photocoagulation:

The RCT suggested that light photocoagulation for clinically significant macular oedema may be as effective as classic laser treatment. However, it is possible that significant differences between treatments may have been detected if a larger sample size was used, or if follow-up was continued beyond 1 year. With results at 1 year, the greater gain in mean number of letters, larger decrease in mean foveal retinal thickness, and greater proportion of eyes with reduction of clinically significant macular oedema in biomicroscopy in the classic group (though not reaching statistical significance) may suggest a trend toward a larger beneficial effect compared with low-intensity treatment.

Clinical guide:

The benefits of laser photocoagulation are less notable in people with maculopathy than in those with proliferative retinopathy. However, focal laser macular photocoagulation significantly reduces moderate visual loss, and is recommended in eyes with clinically significant macular oedema and mild or moderate non-proliferative diabetic retinopathy, particularly when the center of the macula is involved or imminently threatened. We found no evidence that theoretical advantages of certain types of laser result in significant improvements in clinical outcomes.

Substantive changes

Photocoagulation in people with clinically significant macular oedema One RCT added; categorisation unchanged (Beneficial).

2007; 2007: 0702.
Published online 2007 November 23.

Photocoagulation in people with maculopathy but without clinically significant macular oedema

Summary

VISUAL ACUITY Compared with no treatment: Focal laser macular photocoagulation does not reduce visual loss at 3 years compared with no treatment in people with maculopathy and without clinically significant macular oedema ( moderate-quality evidence ). NOTE There is consensus that, prior to development of clinically significant macular oedema, the risk of visual loss is low, and there is no evidence of additional benefit from early focal laser macular photocoagulation.

Benefits

We found no systematic review but found one RCT (2244 eyes with macular oedema plus mild to moderate non-proliferative diabetic retinopathy comparing focal laser treatment using an argon laser (754 eyes) versus no treatment (1490 eyes). The RCT performed a subgroup analysis of 822 eyes with maculopathy but without clinically significant macular oedema (227 eyes assigned to immediate focal laser treatment v 545 eyes assigned to no treatment). The subgroup analysis found that, after 3 years, laser photocoagulation did not significantly reduce moderate visual loss compared with no treatment (reported as non-significant, absolute results presented graphically). Clinically significant macular edema developed in both groups over 3 years with no significant difference between groups (13/81 eyes [16%] with photocoagulation 43/173 eyes [10%] with no treatment; reported as non-significant, P value not reported; see comment below).

Harms

The RCT did not assess adverse effects separately in the subgroup of people with maculopathy but without clinically significant macular oedema. Overall, the RCT found no significant difference in the frequency of immediate visual loss, visual field or color vision scores in treated compared with untreated people (reported as non-significant, no further data reported).

Comment

In the RCT, eyes with maculopathy without clinically significant macular oedema had low rates of visual loss, especially during the first year of follow-up in both the treatment and no treatment group. Clinically significant macular oedema developed in both groups during follow-up; eyes assigned to treatment received additional focal macular photocoagulation and eyes assigned to no-treatment did not receive photocoagulation. The larger, though not significant, differences in visual loss between the two groups at 3 years can be explained, at least in part, by the demonstrated beneficial effect of focal macular photocoagulation in the eyes which developed clinically significant macular oedema during follow up.

Clinical guide

: The presence or absence of clinically significant macular oedema is the most important factor to consider when deciding to treat people with maculopathy and mild to moderate non-proliferative diabetic retinopathy. Prior to development of clinically significant macular oedema, the risk of visual loss is very low, and there is no evidence of additional benefit from early focal laser macular photocoagulation. It is recommended that people with maculopathy but without clinically significant macular oedema are followed up regularly at 4−6 month intervals. Once clinically significant macular oedema develops, focal laser macular photocoagulation can be applied.

Substantive changes

Photocoagulation in people with maculopathy but without clinically significant macular oedema Subgroup analysis from one large RCT added, which found that laser photocoagulation did not significantly reduce moderate visual loss compared with no treatment in people with maculopathy but without clinically significant macular oedema.Categorisation changed to Unlikely to be beneficial.

2007; 2007: 0702.
Published online 2007 November 23.

Grid photocoagulation to zones of retinal thickening in people with diffuse maculopathy

Summary

VISUAL ACUITY Compared with no treatment: Grid photocoagulation to zones of retinal thickening may be more effective at 12 months than no treatment at improving visual acuity, and at 24 months at reducing the risk of moderate visual loss by 50–70%, in eyes with diffuse maculopathy (2 or more disc areas of retinal thickening and involving the center of the macula) with or without cystoid macular oedema ( low-quality evidence ). Different types of grid lasers compared with each other: Argon green may be no more effective at 1 year than diode laser at improving vision or at reducing macular oedema (low-quality evidence).

Benefits

Grid laser to zones of retinal thickening versus no treatment:

We found no systematic review. We found one RCT (160 eyes with diffuse maculopathy with or without cystoid macular oedema), which found that grid laser photocoagulation significantly reduced loss of visual acuity compared with no treatment at 12 months (RR 0.84; NNT 4 eyes, 95% CI 3 eyes to 9 eyes) and at 24 months (RR 0.78, 95% CI 0.60 to 0.96; NNT 3 eyes, 95% CI 2 eyes to 7 eyes). Photocoagulation reduced the risk of moderate visual loss (defined as a doubling of the visual angle, equivalent to loss of about two Snellen lines) by 50–70%. The RCT had some loss to follow-up. The 12-month analysis was conducted on 149 people and the 24-month analysis on 79 people.

Different types of grid laser versus each other:

We found one RCT (randomisation method not reported; 91 people [171 eyes] with diffuse maculopathy) comparing argon green (514 nm) versus diode laser (810 nm) modified grid photocoagulation. It found no significant difference between groups in visual improvement of 3 lines or more (at 1 year: 8/86 [9%] with argon green v 9/85 [11%] with diode laser; P = 0.147; at 2 years: 6/42 [14%] with argon green v 6/39 [15%] with diode laser; P = 1.00), visual loss of 3 lines or more (at 1 year: 7/86 [8%] with argon green v 13/85 [15%] with diode laser; P = 0.147; at 2 years: 5/42 [12%] with argon green v 4/39 [10%] with diode laser; P = 1.00), reduction-elimination of macular oedema (at 1 year: 79/86 [92%] with argon green v 71/85 [84%] with diode laser; P = 0.098; at 2 years: 40/42 [95%] with argon green v 36/39 [92%] with diode laser; P = 0.666).

Harms

Grid laser to zones of retinal thickening versus no treatment:

In the RCT, paracentral grid-like scotomas or haze were visible to most people treated with grid photocoagulation, but the data were insufficient to estimate the frequency of this effect.

Different types of grid laser versus each other:

The RCT comparing argon green (514 nm) versus diode laser (810 nm) modified grid photocoagulation found no significant difference between groups in number of supplemental treatments per eye (at 1 year: 1.61 with argon green v 1.94 with diode laser; P = 0.733; at 2 years: 1.76 with argon green v 2.01 with diode laser; P = 0.67). Complications included non-clearing vitreous haemorrhage or traction retinal detachment requiring vitrectomy (2/86 [2%] with argon v 3/85 [4%] with diode; significance not reported) and increased macular oedema (2/86 [2%] with argon v 2/85 [2%] with diode; significance not reported). Cataract extraction and intraocular lens implantation occurred in 8/86 (9%) of eyes in the argon group and 6/85 (7%) of eyes in the diode group (significance not reported). However, the proportion of phakic or pseudophakic eyes at baseline was 86/86 (100%) in the argon group and 79/85 (93%) in the diode group.

Comment

RCTs are needed to compare efficacy and harm of focal and grid laser protocols.

Substantive changes

No new evidence

2007; 2007: 0702.
Published online 2007 November 23.

Corticosteroids (intravitreal)

Summary

VISUAL ACUITY Compared with placebo: Intravitreal triamcinolone acetonide may be more effective at 6 months to 2 years than placebo at improving visual acuity and reducing macular thickness in people with refractory diabetic macular oedema, but may be associated with a significant risk of adverse effects, and requires repeated injections to maintain its effect ( low-quality evidence ). Different doses compared with each other: We don't know if higher doses of intravitreal triamcinolone acetonide improve visual acuity or reduce macular thickness at 24 weeks to 6 months compared with lower doses ( moderate-quality evidence ). Triamcinolone acetonide posterior sub-tenon injection compared with placebo subconjunctival injection: Triamcinolone acetonide posterior sub-tenon injections may not reduce macular thickness or visual acuity loss compared with placebo subconjunctival injections (low-quality evidence). ADVERSE EFFECTS Intravitreal triamcinolone acetonide is associated with increased intraocular pressure and cataract formation or progression.

Benefits

Intravitreal triamcinolone acetonide versus intravitreal placebo:

We found no systematic reviews but found two RCTs. The first RCT (43 people [69 eyes] with diffuse or focal diabetic macular oedema involving the central fovea, at 3 months or more after at least one session of laser treatment with best corrected visual acuity in the affected eye or eyes of 20/30 or worse) compared triamcinolone acetonide (4 mg intravitreally, no more than once every 6 months) versus placebo. It found that, at 2 years, triamcinolone acetonide significantly improved visual acuity and reduced foveal thickness outcomes compared with placebo (visual acuity gain of 5 or more ETDRS letters: 19/34 [56%] eyes with intravitreal triamcinolone acetonide v 9/35 [26%] eyes with placebo; P = 0.006; see comment below; mean change from baseline in number of ETDRS letters: +3.1 with intravitreal triamcinolone acetonide v −2.9 with placebo; mean difference: 5.7, 95% CI 1.4 letters to 9.9 letters; reduction in central macular thickness: 125 μm with intravitreal triamcinolone acetonide v 71 μm with placebo; P = 0.009; mean difference 59 μm, 95% CI 15 μm to 104 μm; see comment below). It also found that a significantly smaller proportion of eyes in the intravitreal triamcinolone acetonide group than in the placebo group required further macular laser treatment (1/34 [8%] eyes with intravitreal triamcinolone acetonide v 16/35 [46%] eyes with placebo; P = 0.0001).The second RCT (2:1 allocation; randomisation method not reported; 38 people [40 eyes] with diffuse diabetic macular oedema, involving the central fovea for at least three months' duration, and refractory to previous laser photocoagulation) compared triamcinolone acetonide (about 20 mg intravitreally) versus placebo. It found that, at 6 months, triamcinolone acetonide significantly increased mean visual acuity, and the proportion of people with a visual acuity improvement of 2 lines or more (mean change in best visual acuity: 3.4 lines with treatment v 0.9 lines with placebo; P less than 0.001; visual acuity increased by 2 lines or more: 22/28 [76%] with treatment v 4/12 [33%] with placebo; P = 0.01), but not the proportion of people with an increase of 3 lines or more in visual acuity (15/28 [54%] with treatment v 3/12 [25%] with placebo; P = 0.17) compared with placebo (see comment below).

Different doses of intravitreal triamcinolone acetonide:

We found no systematic review, but found two RCTs comparing different doses of a single intravitreal triamcinolone acetonide injection for diabetic macular oedema. The first RCT (63 people [63 eyes] with clinically significant macular oedema, with or without previous macular laser photocoagulation) compared 4 mg, 6 mg, and 8 mg of intravitreal triamcinolone acetonide single dose. The 8 mg dose was significantly more effective than 4 mg at improving mean best corrected visual acuity (BCVA) at 6 months (mean BCVA [number of ETDRS letters] improvement: 3.1 with 4 mg v 9.9 with 8 mg; P = 0.047; data for 6 mg group not reported). The mean reduction in foveal thickness was higher in the 8 mg group compared with the 6 mg and 4 mg groups, but the difference failed to reach significance (29% with 4 mg v 42% with 6 mg v 61% with 8 mg; P = 0.06). It reported that BCVA was improved in all three groups at 6 months, and that visual improvement in the 4 mg group began to deteriorate after 4 weeks (maximum visual improvement with 4 mg: logMar BCVA 0.66 [+7.4 ETDRS letters] at 4 weeks) whereas, in the 8 mg group, visual acuity continued to improve until 26 weeks (maximum visual improvement with 8 mg: logMar BCVA of 59 [+9.9 ETDRS letters] at 26 weeks).The second RCT (32 people [32 eyes] with diffuse diabetic macular oedema involving the center of the macula, unresponsive to previous maximal laser photocoagulation) compared 2 mg versus 4 mg intravitreal triamcinolone acetonide single injection. It found no significant difference between the two doses in reduction of macular thickness from baseline at 4, 12, or 24 weeks ([baseline: 522.9 μm with 2 mg v 564.5 μm with 4 mg] standardised change in macular thickness, at 4 weeks: 69% with 2 mg v 72% with 4 mg; P = 0.49; at 12 weeks: 73.2% with 2 mg v 73.2% with 4 mg; P = 0.66; at 24 weeks: 35% with 2 mg v 29% with 4 mg; P = 0.81 ), or in the median time to recurrence of macular oedema (16 weeks in 9/16 [56%] eyes with 2 mg v 20 weeks in 11/16 [69%] eyes with 4 mg; P = 0.11). The RCT also found no significant difference between groups in visual acuity at 24 weeks (mean gain of ETDRS letters: 5.3 letters with 2 mg v 5.5 letters with 4 mg; P = 0.62; see comment below).

Triamcinolone acetonide posterior sub-tenon injection versus placebo subconjunctival injection:

We found no systematic review but found one RCT (randomisation method not reported; 38 people [64 eyes] with visual acuity of 20/200 or less, macular ischaemia or diffuse macular oedema unresponsive to previous laser photocoagulation performed at least 3 months earlier). The RCT compared triamcinolone acetonide 40 mg posterior sub-tenon injection versus placebo subconjunctival injection, repeated after 2 months. It found no significant difference between groups, at 4 months after injection, in central macular thickness (mean central macular thickness at 4 months [change from baseline]: 377 [−15] μm with triamcinolone acetonide v 357 [−31] μm with placebo; P = 0.627), or best corrected visual acuity scores (mean best corrected visual acuity [change from baseline]: 0.71 [−0.04] logMAR with triamcinolone acetonide v 0.88 [−0.05] logMAR with placebo; P value not reported; reported as not significant; see comment below).

Harms

Triamcinolone acetonide versus placebo intravitreal injection:

The first RCT found that the mean number of injections was significantly higher in the treatment group (2.6 injections in the treatment group v 1.8 injections on the placebo group; P = 0.006), and found that intravitreal triamcinolone acetonide injection significantly increased adverse effects compared with placebo (intraocular pressure increase of 5 mm Hg or higher: 23/34 [68%] eyes with intravitreal triamcinolone acetonide v 3/30 [10%] eyes with placebo; P less than 0.0001; glaucoma medication required: 15/34 [44%] eyes with intravitreal triamcinolone acetonide v 1/30 [3%] eyes with placebo; P = 0.0002; cataract progression of 2 or more grades: 12/28 [43%] eyes with intravitreal triamcinolone acetonide v 3/21 [14%] eyes with placebo; P = 0.03; cataract surgery performed: 15/28 [54%] eyes with intravitreal triamcinolone acetonide v 0/21 [0%] eyes with placebo; P less than 0.0001). In the intravitreal triamcinolone acetonide group 2/34 (6%) eyes required trabeculectomy compared with 0/30 (0%) eyes in the placebo group (P value and significance not reported). One person in the intravitreal triamcinolone acetonide group (1/34 [3%]) had infectious endophthalmitis.The second RCT, comparing intravitreal injection of 20 mg of triamcinolone acetonide with placebo, found that, during the 6 month follow-up, 67% of the eyes treated with triamcinolone acetonide developed a maximal intraocular pressure greater than 21 mm Hg, that was managed successfully by topical antiglaucomatous medication.

Different doses of intravitreal triamcinolone acetonide:

The first RCT found no significant difference in rates of increased cataract formation or progression between 4 mg, 6 mg or 8 mg doses (phakic eyes with increase of cataract grade: 6/19 [32%] with 4 mg v 7/16 [44%] with 6 mg v 2/12 [17%] with 8 mg; P = 0.53). It found no significant difference between groups in increased ocular pressure (proportion of eyes with intraocular pressure greater than 21 mm Hg: 9/23 [39%] with 4 mg v 6/20 [30%] eyes with 6 mg v 11/20 [55%] eyes with 8 mg; P = 0.27). It found that a higher proportion of people in the 8 mg group compared with the 6 mg and 4 mg groups remained on anti-glaucoma medication at 6 months, which reached borderline significance (4/23 [17%] with 4 mg v 3/20 [15%] with 6 mg v 9/20 [45%] with 8 mg; P = 0.05).The second RCT found no significant difference in intraocular pressure at 24 weeks between 2 mg and 4 mg (mean intraocular pressure at 24 weeks [at baseline]: 16.2 mm Hg [17.6 mm Hg] with 2 mg v 15.1 mm Hg [17.4 mm Hg] with 4 mg; P = 0.45). It reported no appreciable difference between doses in mean lens opacity outcomes (absolute data and significance not reported) or in the proportion of eyes with intraocular pressure increase of at least 10 mm Hg (5/16 [31%] with 2 mg v 4/16 [25%] with 4 mg, P value and significance not reported). It reported two serious adverse ocular events: a central retinal vein occlusion in one person with increased intraocular pressure, and a vitreous haemorrhage associated with intravitreal triamcinolone acetonide injection (dose groups not reported).

Triamcinolone acetonide posterior sub-tenon injection versus placebo subconjunctival injection:

The RCT found that, in the triamcinolone acetonide posterior sub-tenon injection group, ocular hypertension over 21 mm Hg occurred in 2/32 [6%] of eyes, and was controlled with topical medication, and chemosis occurred in 2/32 [6%] of eyes. It found no significant difference between groups in cataract progression at 4 months, which the authors of the RCT attributed to having only a short-term follow-up.

Comment

Clinical guide:

Clinicians should be aware that intravitreal triamcinolone acetonide is an off-label treatment not licensed for use in eyes, and people should be informed accordingly. One non-systematic review on harms of intravitreal triamcinolone acetonide injections reported that associated complications include: secondary ocular hypertension; medically uncontrollable high intraocular pressure requiring anti-glaucoma surgery; posterior subcapsular cataract and nuclear cataract; postoperative infectious endophthalmitis; non-infectious endophthalmitis perhaps caused by a reaction to the solvent agent; and pseudoendophthalmitis with triamcinolone acetonide crystals in the anterior chamber.The RCT comparing intravitreal triamcinolone acetonide 4 mg versus placebo showed a beneficial effect of treatment at 2 years in refractory diabetic macular oedema, but was associated with a significant risk of adverse effects, and required repeated injections to maintain its effect. In the placebo group, 26% of eyes gained at least 5 letters of vision over 2 years. The RCT authors reported that, although 8/9 (89%) of these improvements in vision occurred in people who had both eyes included in the study and thus received active treatment in the fellow eye, it seems unlikely, although not impossible, that the triamcinolone injection into one eye might have affected the fellow eye treated with placebo. A more likely explanation is that participation in the study may have stimulated people to increase their efforts in controlling the systemic features of their diabetes. This also may have contributed to the lower than expected requirement for re-treatment — 48% of eyes in the intravitreal triamcinolone group requiring only 1−2 injections over the 2 years of the study. These findings emphasise the importance of the best possible control of a person's systemic risk factors. A single intravitreal triamcinolone injection, even with a dose as high as 20 mg, has only a temporary effect. There is no evidence that lower doses of intravitreal triamcinolone acetonide than the standard 4 mg dose have a more beneficial effect or fewer adverse effects, although, in one RCT, a higher dose of 8 mg was associated with better and more sustained visual outcomes. The lack of visual improvement after two posterior sub-tenon injections of triamcinolone acetonide 40 mg may in part be due to the severity of maculopathy in the eyes studied (eyes with macular ischaemia and visual acuity of 20/200 or less were included), and therefore no visual improvement would be expected in these cases. However, there was almost no effect of triamcinolone on central macular thickness, which reflects that the sub-tenon route of administration of triamcinolone may not penetrate sufficiently through the sclera to the macula to produce an anatomical effect on central macular thickness. Future RCTs are warranted, with larger sample sizes and longer follow-up, to evaluate the optimal dosage of intravitreal triamcinolone acetonide, the necessity and possibility of reinjection, and other routes of administration for the treatment of diabetic macular oedema refractory to laser photocoagulation.

Substantive changes

Corticosteroids (intravitreal) New option with five RCTs; categorised as Beneficial.

2007; 2007: 0702.
Published online 2007 November 23.

Vitrectomy in people with vitreous haemorrhage and proliferative retinopathy

Summary

VISUAL ACUITY Early compared with deferred vitrectomy: Early vitrectomy is more effective at 2–4 years than deferred vitrectomy at improving visual acuity in people with recent severe vitreous haemorrhage — especially in people with type I diabetes with diabetes of less than 20 years' duration, and in people with severe proliferative retinopathy — with benefits increasing proportionally with degree of severity of disease ( moderate-quality evidence ).

Benefits

Early versus deferred vitrectomy:

We found no systematic review, but found one RCT (reported in three publications; total 594 people [616 eyes]; 616/616 eyes with proliferative retinopathy and recent severe vitreous haemorrhage, [defined as haemorrhage of no more than 6 months' duration reducing visual acuity to the interval between 5/200 to light perception at two follow-up visits at least 1 month apart]; 370/616 eyes with visual acuity of 10/200 or better and extensive, active proliferative retinopathy [defined as one of the following: severe new vessels and severe fibrous proliferations or severe new vessels and red vitreous haemorrhage or moderate new vessels, severe fibrous proliferations and red vitreous haemorrhage]). The RCT compared early vitrectomy (performed 1 to 6 months after the onset of haemorrhage) versus deferral of vitrectomy for 1 year (performed if persistent vitreous haemorrhage was present and visual acuity was 5/200 or worse after 12 months, or if retinal detachment involved the centre of the macula detected by ophthalmoscopic or ultrasonographic examination at any time during follow-up).The RCT found that people in the early-treatment group were significantly more likely to have visual acuity of at least 10/20 than those in the deferred-treatment group at 2 years' follow-up (24% with early vitrectomy v 15% with deferred vitrectomy; P = 0.01; absolute data presented graphically; see comment below),and at 3 years (28% with early vitrectomy v 18% with deferred vitrectomy; P = 0.016; absolute data presented graphically), but it found no significant difference between groups at 4 years (29% with early vitrectomy v 23% with deferred vitrectomy; P = 0.224; absolute data presented graphically). The study also found that, in eyes with extensive, active proliferative retinopathy and visual acuity of 10/200 or better at baseline, the proportion of eyes with final visual acuity of 10/20 or better at 4 years was higher after early vitrectomy compared with conventional treatment consisting of observation, photocoagulation, or vitrectomy deferred until traction macular detachment or 6 months of non-clearing vitreous haemorrhage (final visual acuity of 10/20 or better at 4 years: 44% with early vitrectomy v 28% with conventional treatment). The advantage of early vitrectomy was most apparent in eyes with the most severe degree of new vessel development (see comment below).Preplanned subgroup analysis in people with type 1 diabetes (defined as receiving insulin and diagnosed before the age of 20 years) found that early treatment significantly reduced visual loss compared with deferred treatment (eyes with visual acuity at least 10/20: 36% with early v 12% with deferred treatment; P = 0.0001; absolute data not reported). However, there was no such advantage in people with type 2 diabetes (defined as aged 40 years or older when diagnosed, regardless of insulin use, or not using insulin when they entered the RCT) in whom early treatment did not significantly reduce visual loss compared to deferred treatment (eyes with visual acuity of at least 10/20: 16% with early vitrectomy v 18% with deferred vitrectomy; P = 0.007; absolute data not reported). Subgroup analysis found that deferral of vitrectomy appeared to be particularly disadvantageous in people with type 1 diabetes of less than 20 years' duration (eyes with visual acuity of at least 10/20: 34% with early vitrectomy v 2% with deferred vitrectomy; P value and absolute data not reported not reported). In the subgroup of people with type 1 diabetes of more than 20 years' duration, there was a small but significant benefit in visual acuity with early compared with deferred vitrectomy (eyes with visual acuity of at least 10/20: 37% with early vitrectomy v 22% with deferred vitrectomy; P = 0.007; absolute data not reported).At 4 years, when all eyes were assessed, the proportion of eyes with visual acuity of at least 10/20 was higher in the early-vitrectomy group than in the deferred-vitrectomy group, but this difference was not significant. However, the advantage of early vitrectomy persisted at 4 years' follow-up visit for people with type 1 diabetes, particularly in those with less than 20 years' duration. Subgroup analyses also found greater benefit in people with more severe levels of proliferative retinopathy (visual acuity of at least 10/20: 43% with early treatment and 41% with deferred treatment in eyes with least-severe new vessels v 36% with early treatment and 11% with deferred treatment in eyes with very severe new vessels).

Harms

The RCT reported that adverse effects at up to 4 years included enucleation, phthisis or alcohol injection (53/308 [17%] eyes with early vitrectomy v 44/308 [14%] eyes with deferred vitrectomy; P value and significance not reported), sympathetic uveitis or endophthalmitis (3/308 [1.0%] eyes with early vitrectomy v 1/308 [0.3%] eyes with deferred vitrectomy; P value and significance not reported), corneal oedema or epithelial abnormality (45/308 [17%] eyes with early vitrectomy v 32/308 [10%] eyes with deferred vitrectomy; P value and significance not reported), neovascular glaucoma (70/308 [23%] eyes with early vitrectomy v 50/308 [16%] eyes with deferred vitrectomy; P value and significance not reported), and retinal detachment (50/308 [16%] eyes with early vitrectomy v 89/308 [29%] eyes with deferred vitrectomy; P value and significance not reported).An additional RCT found that the use of preoperative intravitreal tissue plasminogen activator failed to reduce the rate of complications in 56 people undergoing vitrectomy for the complications of proliferative diabetic retinopathy.

Comment

Clinical guide:

Early vitrectomy appeared to be clearly advantageous only in people with type 1 diabetes, and this advantage was greater when the duration of diabetes was less than 20 years, and in people with more severe levels of proliferative retinopathy. These studies were performed before advances in vitreoretinal surgery — such as the introduction of endoscopic laser photocoagulation — and modern instrumentation, which have yielded more favourable results in people with diabetes type 1 and 2 with vitreous haemorrhage and/or proliferative retinopathy.

Substantive changes

No new evidence

2007; 2007: 0702.
Published online 2007 November 23.

Vitrectomy in people with vitreous haemorrhage and maculopathy

Summary

We found no direct information about the effects of vitrectomy in people with both vitreous haemorrhage and maculopathy.

Benefits

Vitrectomy versus no treatment:

We found no systematic review or RCTs.

Harms

In one RCT assessing harms, the use of preoperative intravitreal tissue plasminogen activator failed to reduce the rate of complications in 56 people undergoing vitrectomy for the complications of proliferative diabetic retinopathy.

Comment

One retrospective study of 260 eyes treated with vitrectomy reported neovascular glaucoma in 6%, retinal detachment in 8%, and cataract in 27%. Glaucoma was more likely in people with associated preoperative retinal detachment.

Substantive changes

No new evidence


Articles from BMJ Clinical Evidence are provided here courtesy of BMJ Publishing Group