Outcomes similar for heavy, standard silicone oil

She presented the interim results from a prospective, multicentre, randomized trial.

"These interim findings, which were based on data from less than one-third of the planned enrollment, were the basis for halting the study early," said Dr Joussen, professor and chair, Department of Ophthalmology, Charité University Medicine, Berlin, Germany. "For clinicians, they suggest that the choice of which silicone oil to use as an endotamponade should be an individualized decision for each patient."

"Since regular silicone oil does not contact the lower retina, there can be residual fluid beneath the silicone oil bubble in the lower retina that allows for the accumulation of mediators able to promote inflammation and PVR," Dr Joussen added. "It was theorized that because heavy silicone oil would act mechanically to displace this fluid, it could thereby prevent the development of PVR in the lower retina."

Conducting the study

The HSO Study enrolled patients with inferiorly and posteriorly located PVR grade C-A6, P12, or inferior retinal detachment with giant retinal tear in the inferior hemisphere. After surgical retinal reattachment, patients were randomly assigned to one of two endotamponade groups and were followed for 12 months.

Primary study endpoints included visual acuity improvement from baseline to month 12 and anatomic success, defined as retinal attachment at month 12 and without a retina-affecting re-operation.

Secondary endpoints included improvement in quality of life, number of retina-affecting re-operations (excluding endotamponade removal), and complete retinal attachment prior to endotamponade removal.

The study was an international project, conducted at a number of centres throughout Europe, two sites in the US, and one in China. The study began in 2003. Recruitment was slow, however. Four years after study initiation, only 94 patients had been treated (46 heavy silicone oil, 48 standard silicone oil). At the time the interim analysis was performed, about one-third of patients were lost to follow-up.

For the interim analysis, data were analysed both for patients evaluable at 12 months and using an intent-to-treat population in which patients lost to follow-up were treated as having anatomic failure. The analysis also used last observation carried forward for final visual acuity.

Data were stratified by surgeon, but because the bulk of the patients were treated by just a few surgeons, patients operated on by surgeons contributing fewer than 10 patients were grouped together, Dr Joussen reported.

The two study groups were similar at baseline in their demographic and clinical characteristics. The primary efficacy analyses of evaluable patients showed anatomic success rates of 46.4% (13 of 28 patients) in the HSO group and 61.3% (19 of 31 patients) in the standard silicone oil group with no significant difference between groups. Mean logMAR visual acuity was 1.27 in the HSO group and 1.24 in eyes treated with standard silicone oil, and not significantly different between groups.

The data on retinal redetachment under the tamponade represented the most interesting finding from the study, Dr Joussen revealed.

The results indicated HSO did not prevent PVR. Retinal redetachment was more common in the HSO group than in eyes treated with standard silicone oil, 39% versus 27%, although the difference between groups was not statistically significant.

As expected, all of the redetachments in the standard silicone oil group were in the lower quadrants, between 4 and 9 o'clock. However, in a number of the patients treated with HSO, the retinal redetachment occurred in the upper quadrant.

Consistent finding

"This finding is consistent with our previous experience using HSO to prevent PVR following autologous translocation of the choroid and retinal pigment epithelium for age-related macular degeneration," Dr Joussen confirmed.

The latter analysis [Br. J. Ophthalmol. 2009; 93:128–129] showed that PVR developed in 12 (24%) of 50 eyes treated with HSO, all of which were in the upper quadrants of the eye and none was at the transplant excision site. Of 19 eyes treated with regular silicone oil, four (21%) developed PVR, of which three were at the transplant excision site.

In the HSO Study, other secondary endpoint analyses also revealed no statistically significant treatment-related differences. Fewer patients needed a retina-affecting reoperation in the standard silicone oil group compared with the HSO group (19 versus 27, respectively), but the rate was not significantly different compared with the HSO group. Rates of retinectomy, retinotomy and further buckle surgery were also similar in the two groups.

No differences in corneal changes or IOP during follow-up occurred between the groups, and the number of revision surgeries was also similar. A higher proportion of patients who originally received HSO, however, were crossed over to use of standard silicone oil in their second surgery compared with the crossover rate in the second surgery among patients who originally received standard silicone oil.