Retinal Displacement After Pneumatic Versus Vitrectomy for Retinal Detachment (ALIGN)

Rhegmatogenous retinal detachment (RRD) is an acute, sight threatening condition, with an incidence of approximately 10 per 100,000 people.

Without surgical intervention by a vitreoretinal surgeon, retinal detachment almost invariably results in permanent sight loss in the affected eye. There is an increased risk of delayed visual rehabilitation the longer the wait for surgery. Both of the treatments under investigation are widely used and accepted by vitreoretinal surgeons.

Interventions for retinal detachment:

1. Pneumatic retinopexy (PnR) has been employed to repair retinal detachments since the late 1980s and is a minor surgical intervention, carried out in a treatment room. The initial success rate (i.e. the proportion of patients in whom the retina becomes attached after one treatment) is quoted as approximately 70%. PnR is carried out under topical or local anaesthetic (a freezing injection under the conjunctiva, the superficial skin on the eye). The procedure involves injection of a small gas bubble into the eyeball via a fine needle. This step takes a maximum of 15 minutes. Two gases can be injected into the eye: perfluoropropane (C3F8), which lasts 6 weeks, and sulfahexafluoride (SF6), which lasts about 2 weeks. Both are non-toxic, equally effective, have been validated for this use, and are widely used amongst retina surgeons in the world. After injection of the gas bubble, the patient is required to maintain a strict 'head posture' (for example, head tilt to left) for up to 10 days. The purpose of this 'head posture' is to align the gas bubble (which floats within the eye) to the retinal tear. The buoyant force of the gas bubble, as well as its surface tension, act to reattach the detached retina over several days. The gas bubble spontaneously dissipates after 2-6 weeks, depending on the gas selected. Additionally, laser treatment or cryotherapy is carried out either before or 1-2 days after injection of the gas bubble, to secure the retinal tear. Both laser and cryotherapy are widely accepted methods of securing the tears in the retina and both are considered equally safe and effective. In patients where the retina does not reattach with PnR alone, vitrectomy surgery (PPV) or repeat PnR is needed (see below). However, the minority of patients who go on to need repeat treatment encounter similar final anatomical success rates and will experience the same gain in vision as those patients who underwent PPV in the first place.
2. Vitrectomy surgery (PPV) involves 'keyhole' surgery to the eyeball, via three tiny (23/25 gauge) incisions to the sclera. This procedure is carried out in the operating room, under regional anaesthetic plus sedation. During PPV, the vitreous gel is removed from the eye using a fine metal instrument called a 'vitrector'. A large gas bubble (same gases as mentioned for PnR) is injected (to reattach the retina, as in PnR), and laser or cryotherapy is applied around the retinal tear to secure it (as in PnR). After treatment, the patient will be required to maintain a 'head posture' (face down for 24 hours). The gas bubble reabsorbs after 2-6 weeks, depending on the gas selected. The surgery takes 1-1.5 hours. The success rate (i.e. the proportion of patients in whom the retina becomes attached after one treatment) is reported as being as high as 90% in the scientific literature.

Both of the treatments may be associated with complications such as bleeding, infection, increased intraocular pressure or cataract. The risk of a sight threatening complication such as a severe intraocular infection or hemorrhage is less than 1:1000 (for both procedures). The risk of cataract development (clouding of the lens, requiring cataract extraction surgery) is less than 10% for PnR and at least 70% for PPV.

Distortion and retinal displacement after retinal detachment repair:

Image distortions such as metamorphopsia and micropsia are common complaints after surgery for retinal detachment. In 2010 there was the first study demonstrating hyper-fluorescent lines, adjacent to the retinal blood vessels in Fundus auto-fluorescence imaging (FAF) of the retina after retinal detachment (RD) repair surgery. The authors proposed a theory in which these lines which are called also Retinal Vessel Printing (RVP) correspond to the area where the retinal blood vessels were located before the retinal detachment. According to this theory the RVP in FAF imaging is due to increased metabolic activity of RPE cells. Prior to surgery these RPE cells were obscured to light rays by retinal blood vessels while after surgery, due to displacement of the retina, these RPE cells became exposed to the light which leads to increase in the cells metabolic activity. This increase in metabolism is thought to be the cause for the hyper fluorescence seen in FAF imaging. Displacement of the retina after RD repair surgery can serve as anatomy basis of vision distortion. Moreover, these reference lines allow us to quantify the displacement of the retina after retinal detachment surgeries. By doing this, we can compare retinal displacement of different retinal detachment repair surgeries and may reduce post operation visual distortion.

Since the first report, several other studies looked into retinal displacement after RD repair, epiretinal membrane and macular hole. Other studies have shown that retinal displacement ratio is higher in patients with intravitreal gas compare to patients with silicon oil (71.4% vs. 22.2%). There was also a way of quantifying the rotational displacement of the retina. The authors showed that there is more than a simple rotation and probably also a temporal stretch of the retina. Recently there was a publication of the biggest study so far of 125 patients after pars plana vitrectomy (PPV) with 35.2% of patient showed signs of retinal displacement.

Recently the investigators showed in PIVOT trial that patients after pneumatic retinopexy has less vertical distortion than patients after PPV. To the best of our knowledge, no study so far looked into retinal displacement after Pneumatic Retinopexy. Moreover, wide field FAF was not used in previous studies. The investigators think there is a reason to believe that Pneumatic Retinopexy will cause less retinal displacement than PPV. Thus, the investigators propose a prospective cohort study which will compare retinal displacement of patients after RD repair by PPV versus Pneumatic Retinopexy.

The aim of this study is to compare retinal displacement and visual distortion of primary retinal detachment repair following pneumatic retinopexy (PnR) versus pars plana vitrectomy (PPV).

The primary study hypothesis is that pneumatic retinopexy will cause less retinal displacement and less visual distortion at the first 12 months for patients with primary retinal detachment.

Interventions Participants will undergo either: PnR + laser/cryotherapy or PPV + laser/cryotherapy depending on the treating physician's recommendation, regardless their participation in the study.

For patients undergoing PPV, the use of adjunctive surgical techniques such as placement of a scleral buckle, use of silicone oil, or combined cataract extraction are at the discretion of the treating surgeon. All participants undergoing the vitrectomy arm, regardless of the additional steps done during the procedure, will be considered as one group for data analysis.

In the event of primary intervention failure (i.e. failure of retinal re-attachment following primary intervention), the decision to proceed with secondary intervention, and the nature of such intervention, will rest with the treating physician in conjunction with the patient. Secondary intervention may involve any surgical procedure, as deemed clinically appropriate.

Note: Additional laser retinopexy, cryotherapy, gas injection or head positioning are not considered a failure.

Sample size:

A sample size calculation was carried out in relation to the primary outcome, using the following assumptions: minimal clinically important difference in risk of retinal displacement=20% (15% in PnR and 35% in PPV), power 80% and alpha= 0.05. We also increased the sample size per group to account for patients with primary failure, lost follow-up and ungradable images due to media opacity, yielding a total sample size of 204 patients (n=102 per group).

Data management:

Initial data collection (clinical examination findings, visual acuity, questionnaire data) will take place in a paper format. Subsequently, this data will be transferred to a digital database (Microsoft Excel). Paper data will be stored in a locked filing cabinet in the principal investigator's office and away from the study data, and will be destroyed once digital data entry has taken place. The digital spreadsheet will be held on a password protected computer in a locked room, and an encrypted memory stick. At recruitment, each patient's name and date of birth will be obtained to facilitate onward administration of follow-up appointments and safety monitoring, and stored on a face sheet (master linking log). The face sheets will be stored in a locked filing cabinet, away from the study data. Each patient will be allocated a unique study identification number, which will be used to label all paper and digital data pertaining to that patient. The face sheets (master linking log) and all paper/electronic data will be destroyed once publication takes place. The de-identified study data will be destroyed five years after publication has taken place.

Consent:

Written, informed consent will be obtained from each participant. On no occasion should consent be obtained by the treating physician or study investigator. During working hours: The study will be introduced to the patient by the examining physician. Interested patients are directed to the Research Technician who will obtain informed consent.

Data Analysis:

Continuous data: Data will be checked for normality. Normal data will be compared using a non-paired t-test. Non-normal data will be compared using non parametric tests. Categorical data: Chi squared test.

Coefficients with 95% confidence intervals will be reported. A p-value of 0.05 will be considered for statistical significance. Data will be analyzed using SPSS (SPSS Inc., Chicago, IL). Per protocol analysis will be used.