The fellow eye in retinal detachment: findings from the Scottish Retinal Detachment Study
- D Mitry1,2,
- J Singh3,
- D Yorston4,
- M A Rehman Siddiqui4,
- A L Murphy5,
- A F Wright6,
- B W Fleck3,
- H Campbell1,
- D G Charteris2
- 1Department of Public Health Sciences, Centre for Population Health Sciences, University of Edinburgh, Edinburgh, UK
- 2Moorfields Eye Hospital, City Road, London, UK
- 3Princess Alexandra Eye Pavilion, Chambers Street, Edinburgh, UK
- 4Gartnavel General Hospital, Glasgow, UK
- 5Aberdeen Royal Infirmary, Aberdeen, UK
- 6MRC Human Genetics Unit, Institute of Genetics and Molecular Medicine, Crewe Road, Edinburgh, UK
- Correspondence to Dr Danny Mitry, Department of Public Health Sciences, Centre for Population Health Sciences, University of Edinburgh, Edinburgh, UK;
- Accepted 7 February 2011
- Published Online First 3 March 2011
Aim To characterise the predisposing pathology and clinical features in the fellow eyes of patients recruited as part of the Scottish Retinal Detachment Study.
Methods The Scottish Retinal Detachment Study was a 2-year prospectively recruited population-based epidemiology study that sought to recruit all incident cases of primary rhegmatogenous retinal detachment (RRD) in Scotland.
Results A total of 1202 incident cases of primary RRD were recruited in Scotland, over a 2-year period and in 94% (1130 cases) detailed data on the clinical features of fellow eyes with RRD were available. Full-thickness retinal breaks were found in 8.4% (95/1130) of fellow eyes on presentation. Lattice degeneration was present in 14.5% (164/1130) of fellow eyes. Thirteen per cent (148/1130) of affected fellow eyes had a best corrected visual acuity of 6/18 or worse with previous RRD, the second most common cause of poor vision. Overall, 7.3% (88/1202) of cases had RRD in both eyes; 60% of cases with consecutive bilateral RRD presented before the macula were affected.
Conclusions Rhegmatogenous pathology in the fellow eye represents an important threat to vision. Fellow-eye detachments are more common in pseudophakic individuals and those with a more myopic refractive error. Fellow-eye RRD has a greater likelihood of prompt presentation.
Rhegmatogenous retinal detachment
Rhegmatogenous retinal detachment (RRD) is a sight-threatening condition commonly associated with bilateral predisposing pathology. Fellow eyes of RRD cases have been demonstrated in previous studies to have a higher risk of RRD due to increased rates of predisposing retinal degeneration and retinal breaks.1–4 Vision-threatening rhegmatogenous pathology in the fellow eyes of patients with RRD is frequent; however, large-scale epidemiological research on RRD has in general poorly characterised the predisposing pathology in fellow eyes of individuals with primary RRD.5–7 We herein present the predisposing pathology and clinical characteristics of the fellow eye in over 1000 incident cases of primary RRD recruited in the Scottish Retinal Detachment Study.
The Scottish Retinal Detachment Study8 was a prospectively recruited population-based study, the methodology of which has been described elsewhere.9 Between 1 November 2007 and 31 October 2009, a comprehensive system was established where each primary RRD presenting to one of the six vitreoretinal surgical centres in Scotland was approached for study inclusion. The study was approved by the Multi-centre Research and Ethics Committee, Scotland (MREC-06/MRE00/19).
Each case had to be a resident in Scotland and had to have a primary RRD, defined as an area of subretinal fluid greater than two disc diameters with a full-thickness retinal break identified preoperatively or intraoperatively.9 The diagnosis of RRD was made by a consultant vitreoretinal surgeon after biomicroscopic examination. A posterior vitreous detachment (PVD) was defined as the presence of a Weiss ring on biomicroscopy. Cases that met these criteria but did not undergo an operative repair for social or other medical reasons were also eligible. Traumatic RRD was classified as ‘blunt’ if there was a clear reported history of eye trauma and was classified as ‘other’ if there was a head trauma that may have been causally associated with RRD. The following cases were excluded: previous posterior segment intraocular surgery, previous penetrating injury in the presenting eye, previous RRD in the presenting eye and all other types of retinal detachment (exudative, tractional and combined).
At recruitment, all cases were classified primarily by the type of retinal break causing the detachment and also by the presence or absence of a PVD. Horse-shoe tear (HST) and giant retinal tear (GRT) RRDs were classified as PVD-associated RRDs as were round-hole (RH) RRDs with a partial or complete PVD observed on biomicroscopy. RH RRDs with no observed PVD as well as retinal dialysis and retinoschisis were classified as non-PVD RRDs. Continuous data were analysed using a two-sample t test, and differences in categorical data were analysed using the Z test of proportionality or Fisher's exact test for small samples.
A total of 1202 incident cases of primary RRD were recruited in Scotland, over the 2-year study period. In 94% (1130), there were detailed data on the clinical features of fellow eyes with RRD. All cases were classified by the presence or absence of a PVD and by the type of retinal break causing the detachment. In total, PVD was present in 87.6% (990/1130) of cases, with single or multiple HST RRDs accounting for 98.5% (975/990) of this group and GRT RRDs accounting for 1.5% (15/990). PVD was absent in 12.4% (140/1130) of cases and these comprised round-hole RRD in 40% (56/140), retinal dialysis in 47.8% (67/140) and retinoschisis RRD in 12.1% (17/140).
Fellow-eye retinal breaks
Full-thickness retinal breaks were found in 8.4% (95/1130) of fellow eyes on presentation. Of the cases with PVD, 8.5% (85/990) had fellow-eye breaks. In 89.4% (76/85), the fellow-eye break was a PVD-associated tear and in 10.6% (9/85) the fellow-eye break was an RH without a PVD. In cases without PVD, fellow-eye breaks were present in 7.1% (10/140). In RH RRD (N=7), all fellow-eye breaks were round holes; in dialysis cases (N=3), one fellow-eye break was a HST and two fellow eyes also had retinal dialysis.
Overall, lattice degeneration was present in 18.7% (211/1130) of detached eyes and in 14.5% (164/1130) of fellow eyes. Of detached eyes with lattice degeneration (N=211), lattice degeneration was present in 61.6% of fellow eyes (N=130). The fellow eyes of PVD-associated RRD exhibited lattice degeneration in 13.6% (135/990) which was significantly less than that in the fellow eyes of PVD-absent RRD (20.7%–29/140; Z=1.97; p=0.049) (table 1). The highest proportion of fellow-eye lattice degeneration was seen in the fellow eyes of RH RRD, where 39.2% (22/56) had lattice degeneration. Peripheral retinal degeneration other than lattice degeneration (pavingstone/cobblestone degeneration, peripheral chorioretinal scarring) was present in 10.7% (122/1130) of detached eyes and in 6.8% (77/1130) of fellow eyes. The fellow eyes of detached eyes affected with peripheral retinal degeneration exhibited other peripheral degeneration in 59% (79/122). The fellow eyes of PVD-associated RRD exhibited peripheral retinal degeneration other than lattice degeneration in 6.3% (63/990)—similar to the 10% (14/140) occurrence observed in PVD-absent RRD.
Fellow-eye best corrected visual acuity
The best corrected visual acuity (BCVA) by types of retinal breaks in the fellow eye of all cases is shown in table 2. Overall, 79.2% (895/1130) had a BCVA of 6/9 or better, however, 13.1% (148/1130) of affected fellow eyes had a BCVA of 6/18 or worse. In PVD-associated RRD, the most common causes of BCVA of 6/18 or worse in the fellow eye were 40% (52/130) cataract, 21.5% (28/130) RRD, 18.4% (24/130) age-related macular degeneration, 9.2% (12/130) amblyopia, 5.3% (7/130) injury and 3% (4/130) primary open-angle glaucoma. In PVD-absent RRD, the most common causes of BCVA of 6/18 or worse in the fellow eye were 44.4% (8/18) cataract, 22.2% (4/18) injury, 16.7% (3/18) RRD and 5.5% (1/18) amblyopia.
Previous retinal prophylaxis
Overall, 5.8% (66/1130) of cases had previous retinal prophylaxis and subsequently had an RRD in the treated eye. Prophylaxis was most commonly laser or cryotherapy to retinal breaks only. Previously treated peripheral retinal degeneration was uncommon. The proportions of RRD cases that had previous prophylaxis by break type were HST 5.8% (57/975), GRT 6.7% (1/15), RH 7.1% (4/56), dialysis 3% (2/67) and retinoschisis 11.8% (2/17).
Consecutive bilateral RD
Overall, 7.3% (88/1202) of cases had RRD in both the eyes. The majority (70/88) had consecutive fellow-eye RRD which was either prior to (N=63) or within (N=7) the 2-year study period. The median time between detachment in each eye was 3.8 years (interquartile range (IQR) 1.8–6.4 years) The consecutive bilateral RRD group were similar in age and gender to the unilateral group: (56.3 years (12.7) vs 57.9 years (15); p=0.3); bilateral group M:F=2:1; unilateral group M:F=1.8:1 (p=0.1). The symptoms at presentation were also similar between the bilateral and unilateral groups: floaters (58% vs 53%), flashes (24% vs 23%) and peripheral shadow (55% vs 70%); however, significantly fewer patients with bilateral RRD reported loss of central vision (24.3% vs 42.5%). This was also reflected in the proportion of macula ‘on’ detachments, where the consecutive bilateral group had macula ‘on’ RRD in 62.9% (44/70) compared to only 44.2% (461/1042) in the unilateral group (p<0.0001). Similarly, in the extent of RRD, the average clock hours of detachment in the bilateral group was significantly less than that of the unilateral group (see table 3).
Of the seven cases presenting sequentially within the study period, four cases were caused by HST in both the eyes, one case had an RH RRD in one eye and an HST RRD in the second, one case had bilateral RH RRD and one case had bilateral retinal dialysis. Of the 63 cases with consecutive bilateral RRD (with the first eye presenting outside the study period), 52.3% (33/63) presented with PVD-associated RRD (32 cases with single or multiple break HST and one case of GRT RRD) and 47.6% (30/63) presented with non-PVD-associated RRD (29 cases of RH RRD and one case of retinal dialysis). The distribution of retinal breaks in the second affected eye of the consecutive bilateral RRD group was similar to the overall unilateral RRD distribution, with 33.9% (53/156) of breaks in the superotemporal quadrant, (47/156) 30.1% in the inferotemporal quadrant, 23% (36/156) in the superonasal quadrant and 12.8% (20/156) in the inferonasal quadrant. A much higher proportion of the bilateral RRD group was pseudophakic at presentation (table 3). The proportion of lattice degeneration (22.8%, 16/70) in this consecutive bilateral RRD group was similar to the general RRD population (18.7%); however, this group was significantly more myopic and had a longer axial length compared to the unilateral RRD group (figure 1).
Simultaneous bilateral RRD
Eighteen cases (1.5%) presented with simultaneous bilateral RRD. Of these, three had bilateral HST-associated RRD, four had bilateral RH RRD, five cases had RH RRD in one eye and HST RRD in the fellow eye, five cases had bilateral retinal dialysis and one case had a bilateral GRT RRD. Three of the five retinal dialysis cases reported significant ocular trauma within 6 months prior to presentation. The proportion of lattice degeneration (23.8%) was not different from that of the general RRD population. One case was pseudophakic at presentation. All cases presented with unilateral symptoms. There was no difference in myopia or axial length measurements between the 18 simultaneous bilateral cases and the general RRD population. Bilateral cases were significantly younger (unilateral RRD mean age (SD)=57.9(15); simultaneous bilateral RRD mean age (SD)=45.4(18.6); p=0.014).
The series presented here represents over 95% of all operated cases of RRD in Scotland, over a 2-year study period, providing an extensive clinical and epidemiological resource.8 For analysis, we have stratified RRD by break morphology and vitreous attachment. This resulted in two principal groups: those secondary to PVD and tractional retinal tears and those without PVD, where atrophic (eg, round) holes or retinal dialyses were causative factor.
RRD in the fellow eye represents an important threat to vision. In our series of 1130 incident cases of RRD, 8.4% of fellow eyes had full-thickness retinal breaks at presentation, warranting laser or cryotherapy treatment. There was no difference in the proportion of fellow-eye breaks in PVD-associated RRD and non-PVD RRD (8.5% vs 7.1%). This is similar to the previous reports where over an average of 5-year follow-up, fellow-eye retinal tears were observed in 12% (30/248) of cases.1 6 10–12 Lattice degeneration was present in 11% of fellow eyes, which was slightly below the range reported previously (18–23%).6 10 11 In affected eyes with lattice (or other) degenerations, a high rate of degeneration in fellow eyes was observed (59–61%), supporting previous observations that indicate a bilaterality of lattice degeneration of between 60 and 90%.2 11 Lattice degeneration was significantly more frequent in the fellow eyes of PVD-absent RRD than in those of PVD-associated RRD (20.7% vs 13.6%).
The rate of bilateral RRD has a wide reported range, affecting between 2 and 11% in previous population-based studies and has been noted in up to 30% in large clinical series.5 13–15 Approximately 2% of our patients presented with simultaneous bilateral RRD and a further 5.3% of patients had a previous RRD in the contralateral eye with a median time interval of 3.8 years between detachments. Retinal dialyses in younger patients accounted for approximately 30% of simultaneous bilateral cases, with trauma being implicated in 55% of dialysis cases. There were significant differences between consecutive bilateral RRD and unilateral RRD series. Cases of consecutive bilateral RRD presented sooner, with a higher rate of macula on RRD and a smaller extent of detachment, compared to unilateral RRD. Pseudophakia at presentation was twice as frequent in this group compared to the unilateral RRD group. In addition, the patients in the consecutive bilateral RRD group were on average two dioptres more myopic and had on average an axial length 1.6 mm longer than unilateral RRD cases, suggesting that genetically determined predispositions to RRD may result in a higher incidence of bilateral RRD.
Laser photocoagulation and cryotherapy are commonly used to treat tractional retinal tears;16 however, up to 10% of previously treated tears nevertheless go on to develop RRD.6 17 In this series, 5.8% of RRD cases had had retinal prophylaxis to a previously identified retinal break or an area of peripheral retinal degeneration and subsequently detached in the treated eye.
Thirteen per cent (148/1130) of affected fellow eyes had a BCVA of 6/18 or worse. RRD in the fellow eye accounted for 20.9% (31/148) of the reduced visual acuity, highlighting the importance of rhegmatogenous pathology as a cause of visual loss in the other eye. It is also notable that approximately 3% of fellow eyes had a visual acuity less than 6/60, highlighting the potential for blindness secondary to RRD.
In conclusion, the fellow eye of RRD patients has an important risk for rhegmatogenous pathology. Up to 8% of fellow eyes have full-thickness retinal breaks on presentation. In our study, 7.3% of patients had bilateral RRD—1.5% presenting with simultaneous bilateral RRD. Fellow-eye detachments are more common in pseudophakic individuals and in individuals with a more myopic refractive error. Fellow-eye RRD has a greater likelihood of prompt presentation with a smaller area of detachment and an intact macula compared to unilateral RRD, suggesting enhanced patient awareness of symptoms. Despite early identification and prophylactic treatment of retinal breaks or peripheral retinal degeneration, approximately 6% of treated cases developed RRD in the treated eye. One in 10 fellow eyes has a BCVA of 6/18 or less, and in 20% of patients this is due to previous RRD.
We thank all the members of the Scottish RD Study group for their valuable contribution to this study: Princess Alexandra Eye Pavilion, Edinburgh: H. Bennett. Gartnavel General Hospital, Glasgow: A. Cox, H. Hammer, J. Murdoch, S. Saidkasimova. Ayr Hospital, Ayr: Z. Koshy, K. Madill, A. Singh. Ninewells Hospital, Dundee: G. Cormack, J. Ellis, P. Baines. Aberdeen Royal Infirmary: H.Atta, N.Lois, JV. Forrester, M.S. Mustafa. Raigmore Hospital, Inverness: S. Hewick I. Whyte.
Funding This study was supported by a major ophthalmology grant from the Royal College of Surgeons, Edinburgh; the Royal Blind School, Edinburgh/Scottish War Blinded; the W.H. Ross Foundation for the Prevention of Blindness and the Chief Scientist Office, Scotland (CZB/4/705). The authors acknowledge (a proportion of their) financial support from the Department of Health through the award from the National Institute for Health Research to Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology for a Specialist Biomedical Research Centre for Ophthalmology. The views expressed in this publication are those of the authors and not necessarily those of the Department of Health.
Competing interests None.
Patient consent Obtained.
Ethics approval This study was conducted with the approval of the Multi-Centre Research and Ethics Committee, Scotland (MREC-06/MRE00/19).
Provenance and peer review Not commissioned; externally peer reviewed.