STAR Trial: Surgery Not Superior to Pneumatic Displacement for Submacular Hemorrhage

Published on: 

The 6-month RCT investigating submacular hemorrhage secondary to nAMD found that both treatment modalities showed clinical improvement of VA, but surgery did not provide a superior visual gain.

Surgical pars plana vitrectomy (PPV) did not lead to superior visual gain compared to pneumatic displacement for submacular hemorrhage secondary to neovascular age-related macular degeneration (nAMD), according to new research.1

Both strategies lead to clinical improvement of visual acuity without safety concerns for submacular hemorrhage over 6 months; however, the team led by Pierre-Henry Gabrielle, MD, PhD, University Hospital, suggested the trial design and data could not be used to establish equivalence between treatments, necessitating a larger clinical trial designed for equivalence.

“Submacular hemorrhage should be treated promptly, and the treatment strategy should be selected based on the patient’s comorbidities, treatment, and submacular hemorrhage characteristics, and the ophthalmologist’s expertise and experience,” investigators wrote.

The prospective, randomized controlled STAR trial is the largest trial of its type in the anti-VEGF era, evaluating the safety and efficacy of submacular hemorrhage management with either surgical PPV or pneumatic displacement plus tissue plasminogen activator (TPA). Gabrielle and colleagues hypothesized that surgical treatment would lead to superior vision outcomes in the form of a 2-line improvement in vision compared to pneumatic displacement, with an acceptable safety profile.

A total of 90 patients enrolled from 13 hospitals in France between April 2016 - October 2019, with the patient’s final visit in May 2020. All visual outcomes and safety were assessed monthly up to Month 6. Eligible patients were aged ≥50 years and presented with visual loss due to a recent submacular hemorrhage secondary to nAMD (≤14 days). The presence of blood needed to predominantly overlay the retinal pigment epithelium (RPE) with a diameter greater than 2 optic disk diameters on retinal photographs.

At the inclusion visit, investigators randomized eligible patients in a 1:1 ratio to receive either surgery (PPV, subretinal TPA [max 0.5 ml/50 µg] and 20% Sulfahexafluoride tamponade) or pneumatic displacement (0.05 ml intravitreal TPA [50 µg] and 0.3 ml intravitreal pure sulfahexafluoride), followed by intravitreal ranibizumab 0.5mg at the end of the procedure. Participants received intravitreal ranibizumab administered at month 1 and 2 as the loading phase, with additional injections at Months 3, 4, 5, and 6 administered according to the presence of hemorrhage and/or sign of active choroidal neovascular lesion.

For the superiority trial, the primary efficacy endpoint was mean best-corrected visual acuity (BCVA) changes from baseline to Month 3, with secondary endpoints including changes at Month 6, Visual Function Questionnaire (VFQ-25) composite score value at Month 3 and 6, number of anti-VEGF injections, and complications during the 6-month follow-up. A total of 90 patients were randomized and 78 patients (86.7%) completed the 3-month efficacy endpoint visit, with an overall mean age of 83 years and 66.3% female.

The analysis showed the mean change in VA improved up to Month 1 and then stabilized to Month 6 in both groups, with a mean VA change from baseline to Month 3 of 16.8 letters (95% CI, 8.7 - 24.9) in the surgery group and 16.4 letters (95% CI, 7.1 - 25.7) in the pneumatic displacement group. Investigators indicated there was no significant difference between treatment groups in VA change to Month 3 (adjusted β, 1.9; 95% CI, -11.3 to 14.9; P = .787).

Additionally, there were no significant differences observed between groups for secondary endpoints. At month 6, the mean VA letter score improvement was 17.2 letters (95% CI, 9.1 - 25.4) in the surgery group and 15.4 letters (95% CI, 5.7 - 25.1) in the pneumatic displacement group, with no difference between groups. Investigators noted there was no meaningful difference in the VFQ-25 composite score value at month 3 and month 6 between groups.

Data showed the average number of optional anti-VEGF injections from Month 3 to Month 6, administered in a pro re nata regimen was similar between the surgery and pneumatic displacement groups (3.2 injections vs. 2.8 injections). Both treatment modalities reported good safety profiles, with data showing around a quarter of patients reported having at least 1 treatment-related adverse ocular event.

Gabrielle and colleagues noted that the VA improvement in each treatment group helped highlight the importance of prompt treatment of submacular hemorrhage, due to poor VA outcomes associated with a lack of treatment.

“Results from STAR will help address the comparative evidence gap, which has thus far prevented the establishment of a clear management consensus for submacular hemorrhage,” investigators wrote.


  1. Gabrielle PH, Delyfer MN, Glacet-Bernard A, et al. STAR: a randomized controlled trial for submacular hemorrhage secondary to age-related macular degeneration [published online ahead of print, 2023 Apr 21]. Ophthalmology. 2023;S0161-6420(23)00280-4. doi:10.1016/j.ophtha.2023.04.014