STUDY SUBJECTS

Major inclusion criteria:

• Adult patients with type 1 or 2 diabetes mellitus with severe NDPR in both eyes or PDR in one eye
• VA >20/100 in each eye

RANDOMIZATION SCHEME AND INTERVENTIONS

Both eyes of patients were randomized to either:

(a) laser photocoagulation (argon or xenon) completed in 1 or 2 sessions. Argon was 800-1600 burns, 500 microns in size, 0.1 second duration with direct treatment of NVD and NVE. Xenon technique was similar, but burns were smaller, longer duration and only direct treatment of NVE

(b) observation

The initial minimum follow up was planned for 5 years, however after an average of 15 months of follow up, there was a 60% reduction in blindness in the laser cohort and thus the protocol was changed. Initially untreated eyes that developed high risk characteristics (HRCs, moderate or severe NVD, NVD with VH or pre-retinal heme, or NVE ½ DD in size with VH and pre-retinal heme) were now considered for photocoagulation, The protocol was changed again and allowed eyes to be treated if the DRS investigator considered to be in the patient’s best interest.

RESULTS (before protocol change)

Rate of SVL

• As severity of PDR increased, rate of SVL increased in controls
• In eyes with severe NPDR, the rate of SVL was 3.6% in controls and 3.0% in treated eyes at 2 years
• In eyes with the highest risk PDR, the rate of SVL was 36.9% in controls and 20.1% in treated eyes at 2 years
• Photocoagulation reduced 2 year risk of blindness by about 60%

RESULTS (after protocol change – by 4 years, 35% of the original controls had been treated and eyes were resulted based on the original randomization cohort regardless of future treatment)

Rate of SVL

• Control eyes at 2 years: 3.2% in severe NPDR, 7.0% in eyes with PDR without HRC, 26.2% in eyes with PDR and HRC
• Control eyes at 4 years: 12.8% in severe NPDR, 20.9% in eyes with PDR without HRC, 44.0% in eyes with PDR and HRC
• Treated eyes at 2 years: 2.8% in severe NPDR, 3.2% in eyes with PDR without HRC, 10.9% in eyes with PDR and HRC
• Treated eyes at 4 years: 4.3% in severe NPDR, 7.4% in eyes with PDR without HRC, 20.4% in eyes with PDR and HRC

Other risk factors for SVL

• NVD, VA at randomization visit, hemorrhages or Mas, elevation of the NVD, arteriolar abnormalities, history of sudden VA loss, venous caliber abnormalities in fellow eye, vitreous or preretinal hemorrhage, proteinuria, and perivenous exudates
• NVE not selected as factor indicative of VA loss
• An analysis using those 10 factors plus the presence of macular edema and diastolic blood pressure found that eyes in the highest decile risk score had a SVL of 56.3% versus 3.4% in the lowest decile

DME and PRP

• 2 year risk of VA loss in eyes without baseline DME and PDR without HRC: 19.0% in controls, 16.6% in argon treated eyes, 23.2% in xenon treated eyes
• 2 year risk of VA loss in eyes with baseline DME and PDR without HRC: 45.6% in controls, 36.4% in argon treated eyes, 35.5% in xenon treated eyes
• 2 year risk of VA loss in eyes without baseline DME and PDR with HRC: 51.1% in controls, 24.9% in argon treated eyes, 33.3% in xenon treated eyes
• 2 year risk of VA loss in eyes with baseline DME and PDR with HRC: 65.5% in controls, 46.3% in argon treated eyes, 43.6% in xenon treated eyes

Adverse events

• Visual field loss was nearly 3x as common in xenon treated eyes compared to argon treated eyes

CONCLUSIONS

• Eyes with PDR and HRC: risk of SVL at 2 years without treatment is 25-35% and with treatment there is a 50-60% reduction in SVL. 11% of treated eyes saw 1 line of VA loss and 3% saw >2 lines of VA loss and thus eyes with PDR and HRC are recommended for prompt treatment.
• Eyes with PDR without HRC: risk of SVL at 2 years without treatment is 5-10% with similar rates of SVL reduction and VA loss with treatment.
• Eyes with severe NPDR: risk of SVL is 3-4% at 2 years without treatment.
• Laser photocoagulation is eyes with PDR and HRCs have significant reductions in SVL, however in eyes without HRCs and NPDR, the reduction in SVL is smaller and thus risks of treatment become more important. Argon laser resulted in fewer harmful effects than xenon laser.
• Extensive NVI is an urgent indicator for prompt scatter photocoagulation whether or not HRCs are present or not.
• In eyes with DME, it may be desirable to treat the DME with focal or grid laser first if scatter treatment is not urgent