- Email: [email protected]
Demarcation laser photocoagulation of selected macula-sparing rhegmatogenous retinal detachments
Demarcation Laser Photocoagulation of Selected Macula-sparing Rhegmatogenous Retinal Detachments Tamara R. Vrabec, MD,1 Caroline R. Baumal, MD2 Objective: To report a series of macula-sparing rhegmatogenous retinal detachments (MSRRDs) treated with demarcation laser photocoagulation (DLP). Design: Retrospective, noncomparative case series. Participants: Thirty-one patients (34 eyes) with primary or recurrent MSRRDs without associated visual field loss, necrotizing retinitis, or proliferative vitreoretinopathy (PVR), managed with DLP from November 1992 through May 1999. Intervention: Demarcation laser photocoagulation consisting of a triple row of confluent laser burns. Main Outcome Measures: Best corrected postoperative visual acuity and MSRRD progression or recurrence. Results: Thirty-four primary and recurrent MSRRDs were treated by DLP, which consisted of a triple row of confluent laser burns. Macula-sparing rhegmatogenous retinal detachments were located in all quadrants and affected 10% to 45% of the retina. Findings associated with MSRRDs included lattice degeneration (12 eyes), vitreous hemorrhage (4 eyes), and demarcation line (9 eyes). Symptoms (photopsias or floaters) were associated with 14 MSRRDs. Eight eyes were myopic and 11 were pseudophakic. Thirty-two MSRRDs were shallow, two were dome shaped, and all were smooth without corrugations. Follow-up ranged from 1.5 to 80 months (mean, 15.8 months; median, 17 months). Thirty-three of 34 detachments remained stable after DLP. Three flattened spontaneously. One eye was managed with scleral buckle 6 weeks after DLP. Progression was attributed to incomplete laser treatment. Best corrected postoperative visual acuity was the same or improved in all but one eye, in which a cataract developed. Conclusions: Demarcation laser photocoagulation is an effective method to manage acute or chronic, primary or recurrent MSRRDs without associated PVR that are shallow and smooth without corrugations. Demarcation laser photocoagulation is an alternative to both observation and surgical repair for these select MSRRDs. Ophthalmology 2000;107:1063–1067 © 2000 by the American Academy of Ophthalmology. Scleral buckle surgery and pneumatic retinopexy1,2 remain the standards for repair of uncomplicated rhegmatogenous retinal detachments (RRDs). The temporary balloon buckle3 and primary vitrectomy are other surgical alternatives. Macula-sparing cytomegalovirus retinitis-related RRDs have been successfully managed with demarcation laser photocoagulation (DLP).4,5 This series reports results of demarcation laser photocoagulation in eyes with primary and recurrent macula-sparing rhegmatogenous retinal detachments (MSRRDs) without associated visual field loss and uncomplicated by necrotizing retinitis or proliferative vitreoretinopathy.
Originally received: August 17, 1999. Accepted: March 1, 2000. Manuscript no. 99563. 1 Wills Eye Hospital, Thomas Jefferson University, Philadelphia, Pennsylvania. 2 New England Eye Center, Tufts University, Boston, Massachusetts. Presented in part at the Pan American Association of Ophthalmology meeting, Orlando, Florida, October 1999. Correspondence to Tamara R. Vrabec, MD, 3244 Midvale Avenue, Philadelphia, PA 19129; E-mail: [email protected] © 2000 by the American Academy of Ophthalmology Published by Elsevier Science Inc.
Patients and Methods We reviewed the clinical records of all patients we treated with DLP between November 1992 and May 1999. A total of 39 patients (forty-three eyes) were identified. Patients to whom DLP was recommended were those who had MSRRDs without associated visual field loss or proliferative vitreoretinopathy (PVR). Seven eyes with MSRRD resulting from cytomegalovirus-related retinal detachment, previously reported,4 were excluded. One eye was excluded because follow-up was less than 3 months. One MSRRD resulting from retinoschisis with an outer wall hole was also excluded. Thirty-one patients (34 eyes) who underwent DLP for MSRRD were included. Data obtained included age and gender of each patient; history of retinal detachment repair; presence or absence of symptoms (photopsias, floaters); number and type of retinal tears; size (% retina involved as estimated by the treating retina specialist [TRV or CRB]); location and appearance (shallow versus bullous) of retinal detachments; associated findings including refractive error, demarcation line, lattice degeneration, vitreous hemorrhage, and lens status; best corrected pre- and postlaser visual acuity; progression beyond laser barrier; and length of follow-up. The presence or absence of posterior vitreous detachment (PVD) at the time of DLP was not specifically noted in patient records. One patient who reported visual field loss was managed with DLP because she ISSN 0161-6420/00/$–see front matter PII S0161-6420(00)00091-9
1063
Ophthalmology Volume 107, Number 6, June 2000
Figure 1. Large temporal rhegmatogenous retinal detachments caused by one horseshoe tear and two atrophic holes. Detached retina (to the left of the laser barrier) remains stable 4 months after demarcation laser photocoagulation. The fellow eye had a smaller detachment stabilized by demarcation laser photocoagulation.
declined scleral buckling surgery and was not a candidate for pneumatic retinopexy. Perimetry was not performed on any patients. Informed consent was obtained from all patients after a discussion of the risks and benefits of surgery (pneumatic retinopexy and scleral buckling), observation alone, and DLP, including the possibility of progression that would necessitate additional procedures if DLP failed to stabilize the detachment. Laser was applied in three rows of confluent 200-m grey-white burns that were placed posterior to the entire detachment and which extended to the ora serrata surrounding the entire detachment. Indirect laser delivery system facilitated peripheral treatment in some cases. After DLP, patients’ behavior was not limited in any way to prevent progression. Eyes in which subretinal fluid extended so close to the optic disc or fovea that three rows of confluent laser could not be applied without damaging these structures and eyes with vitreous hemorrhage dense enough to prevent complete laser treatment were not considered acceptable candidates for DLP.
Results Twenty males and 11 females (34 eyes) between the ages of 12 and 84 years underwent DLP for MSRRD. Twenty-eight MSRRDs were primary detachments and six were limited recurrent detachments after previous surgical repair with scleral buckle (n ⫽ 2), pneumatic retinopexy (n ⫽ 1), or PPV with (n ⫽ 2) or without (n ⫽ 1) silicone oil injection. Before laser demarcation, best corrected vision ranged from 20/20 to 20/400 in patients with no previous surgery and from 20/30 to counting fingers in patients who had previous retinal detachment repair. Six patients had bilateral retinal detachments. In three, bilateral detachments were treated with DLP (Fig 1). Three others had retinal detachments in fellow eyes that had been repaired previously with scleral buckle surgery. The extent of MSRRDs ranged from 10% to 45% of the retina (median, 20%). Eight MSRRDs affected at least 30% of the retina and also extended posterior to the equator. One extended across two quadrants, but was peripheral so that the total extent of detachment was less than 30%. The locations of MSRRDs were superior (n ⫽ 14), inferior (n ⫽ 13), and temporal (n ⫽ 7). Thirty-two MSRRDs were shallow, and two were dome shaped. None were highly elevated or corrugated (bullous). Three patients
1064
had more than one tear. Type of tears included dialysis (n ⫽ 1), horseshoe tear (n ⫽ 18), and round hole (n ⫽ 20). Fourteen MSRRDs had associated symptoms including photopsia and floaters. One eye (see below) had visual field loss. The duration of symptoms was less that 1 month in 11 eyes, and more than 1 month in 3 eyes. Associated ocular conditions included lattice degeneration in 12 eyes, limited vitreous hemorrhage in 4 eyes, and incomplete pigmented demarcation lines in 9 eyes. In three eyes, a depigmented line without subretinal fibrosis was noted in the attached retina immediately posterior to the subretinal fluid. Eight eyes were known to be myopic. Refractive status could not be determined from the charts in 18 cases. Eleven eyes were pseudophakic (1 iris-fixated lens, 10 posterior chamber lenses), and one was aphakic. In most cases, initial follow-up visits were 1 week, 3 weeks, 6 weeks, 3 months, and 6 months after the laser procedure. Follow-up ranged from 1.5 to 80 months (mean, 15.8 months; median, 17 months). The only MSRRD with less than 3 months of follow-up included in the series was the one that progressed 6 weeks after DLP. Three detachments flattened spontaneously after laser. In both primary and recurrent detachments, post-treatment best corrected visual acuity was either unchanged or improved by 1 to 4 lines (median, 1 line) in all but one eye. A cataract developed in this eye. Specifically, in eyes that had previous surgery, visual acuity after DLP was improved in two patients, unchanged in four patients, and worse in one because of cataract. In eyes with no previous surgery, visual acuity after DPL was improved in 10 and unchanged in 17. Best corrected postoperative vision ranged from 20/20 to 20/300 in eyes with primary detachments and from 20/30 to counting fingers in those who had previous surgery. Final visual acuity was 20/300 in one eye with primary detachment as a result of a disciform macular scar that had been present before DLP. One retinal detachment progressed 6 weeks after laser treatment. This eye was not believed to be an acceptable candidate for DLP because subretinal fluid was so close to the fovea that only 1 to 2 lines of laser could be delivered and because the patient was aware of visual field loss. Demarcation laser photocoagulation was performed because the patient initially declined scleral buckle surgery and was not a candidate for pneumatic retinopexy. When subretinal fluid progressed to threaten the fovea, the patient consented to scleral buckle repair. Vision improved from 20/40 before surgery to 20/30 after surgery and remained stable 17 months after scleral buckle surgery.
Discussion Vitreoretinal surgery is indispensable in the management of macula-off RRDs. However, optimal management strategy for certain MSRRDs including those that have been named limited, subclinical, and asymptomatic is ambiguous. Paucity of natural history data and a plethora of definitions and reasons for treating these detachments have made it difficult to glean a clear-cut treatment approach from the literature. This series is not the first to describe the use of laser as a treatment for retinal detachment. In 1968, Okun and Cibis6 recommended photocoagulation for “limited” retinal detachments. Limited detachments were defined by size. Criteria included width at least two times but not greater than five times the diameter of the largest break, and less than 2 clock hours in size with the posterior extent not beyond the equator. Breaks included tears, holes, and dialyses. No detachment was complicated by vitreous hemorrhage. The authors did not state whether detachments were
Vrabec and Baumal 䡠 DLP of MSRRD acute or chronic, or whether symptoms or demarcation lines were present. Laser demarcation of subretinal fluid plus photocoagulation facilitated by scleral depression of the areas around the retinal break and between the break and the demarcation line caused spontaneous flattening with chorioretinal scarring 2 to 3 weeks after treatment in 42 (88%) of 48 eyes. New breaks contributed to redetachment in three eyes. Laser failed to stabilize the detachment in three other eyes. Several authors have recommended surgical repair of MSRRDs. Schepens7 recommended prophylactic treatment to selected “subclinical” detachments he defined by the absence of symptomatic visual field loss. Because 50% of fellow eyes had symptomatic RRDs, he deduced that the subclincal RRD was a precursor to a symptomatic RRD. Thirty-eight (35%) of 110 RRDs were treated with external diathermy, in most cases without drainage of subretinal fluid. Eyes selected for treatment included those with RRD in the contralateral eye; superior or temporal RRDs; large, multiple, and equatorial tears; vitreous hemorrhage; and photopsia. Complete reattachment of the retina followed surgical obliteration of retinal breaks with diathermy in 97%. The clinical course of untreated eyes was not discussed. Davis8 also recommended surgical repair of subclinical detachments defined by subretinal fluid at least one disc diameter beyond the break but not more than two disc diameters posterior to the equator. Data from two groups, symptomatic and asymptomatic were analyzed. Six of seven subclinical RRDs with symptomatic (horseshoe) breaks were repaired immediately with surgery. The other progressed to “clinical” detachment (defined as involving one or two quadrants) without macular involvement 3.5 weeks after initial diagnosis and was repaired surgically. Because only one symptomatic subclinical RRD was observed, it is not possible to draw conclusions about progression in this subset. In the subset with asymptomatic breaks followed more than 6 months, 30% (6/20) progressed to “clinical” detachment. It is important to note that only one of these clinical detachments (4.5%) affected the macula. However, Davis concluded that prophylactic surgical treatment was “probably wise” for “subclinical” RRDs. Benson and associates9 recommended scleral buckling for RRDs with demarcation lines. Twenty of 66 eyes (30%) reviewed retrospectively were asymptomatic. All retinas were reattached surgically. Reattachment rate was 98.5%, but neither visual outcome nor complications in the asymptomatic maculasparing RRDs are reported. No asymptomatic cases were observed. It is uncertain whether they would have remained stable without treatment. Other authors have recommended observation for subclinical or asymptomatic RRDs. Byer,10 who defined subclinical RRDs by the same size criteria as Davis, observed only 3 of 18 asymptomatic detachments progressed slightly during a follow-up period ranging from 1 to 18 years. All detachments remained subclinical; none required surgery. In a subsequent study of eyes with lattice degeneration, he noted 2 of 10 subclinical RRDs progressed only slightly during a follow-up interval of 8 months to 23 years.11 No macular detachments developed in any. One eye was treated with a segmental buckle. Byer concluded in both studies
that because subclinical RRDs remain localized or progress only very slowly, surgical intervention should be discouraged. “Asymptomatic” detachments, discussed by Brod and associates,12 were in most cases larger than “subclinical” RRDs. None had associated photopsia or visual field loss. Demarcation lines were present in 74%, and 61% of the demarcation lines were complete. Of 31 RRDs, only 2 (6%) progressed, one 2.25 years and the other 3.3 years after initial diagnosis. Both were repaired with scleral buckle surgery. The authors recommended observation for asymptomatic RRDs. Some MSRRDs, also know as limited, subclinical, or asymptomatic, may progress to macular detachment. The detachments in this series included ones which were larger than those previously defined as subclinical, ones with horseshoe tears, acute onset, symptoms of vitreoretinal traction, vitreous hemorrhage, superior location, and detachments in fellow eyes, features that have been considered risk factors for progression. One of 23 detachments in this series that had one or more of these risk factors progressed. This series demonstrates that surgery may be safely avoided in many of these high-risk MSRRDs. It could be argued that observation alone may have been adequate in the MSRRDs without apparent risk factors. The literature suggests that many MSRRDs, especially those that are asymptomatic, small, shallow, with demarcation lines or caused by round holes, will remain stable without progression. However, previous series have documented that even low-risk detachments may progress.8,10 –12 None of the 20 asymptomatic detachments in our series progressed to detachment in comparison with the up to 16.6% rate of progression in previous series.8,10,12 It appears that DLP is a reasonable alternative to observation alone in these low-risk detachments. Laser photocoagulation rapidly enhances retinal adhesion in vitro and in vivo to 140% of normal in 24 hours, and twice normal between 3 days and 4 weeks.13 In this series, we found DLP an effective treatment for MSRRD. Only 1 of 34 detachments (2.9%) progressed beyond the laser barrier, and progression was attributable to incomplete demarcation in the macula. For laser to be effective, the importance of adequate and correctly placed treatment should be emphasized. Three confluent rows of laser photocoagulation burns in healthy retina must completely surround the entire detachment and extend to the ora serrata. Neither the extent nor the location of the detachment need be an exclusion criteria; however, detachments in the posterior pole should be treated only if three confluent rows of laser can be placed in attached retina between the subretinal fluid and the fovea or optic disc without damaging these structures. Insufficiently confluent or incomplete laser demarcation is likely to predispose to breakthrough of subretinal fluid. The feature shared among MSRRDs stabilized with DLP in this series was a shallow, relatively tense appearance without associated corrugations. Each MSRRD resembled a low, tense blister rather than a billowing parachute (Fig 2). Okun and Cibis6 and Schepens7 also noted RRDs that responded well to laser or to external diathermy were shallow. In considering why the chorioretinal adhesion created by DLP may prevent progression of shallow MSRRDs, and
1065
Ophthalmology Volume 107, Number 6, June 2000
Figure 2. Inferior macula-sparing rhegmatogenous retinal detachment treated with a triple row of confluent demarcation laser photocoagulation is stable 9 months after treatment. Note the shallow, smooth, noncorrugated appearance of this detachment.
in fact why some demarcation lines do as well, one must reconsider the factors important in the pathogenesis of RRD. Classically, vitreoretinal traction is thought to be most important. Certainly, it contributes to the development of horseshoe tears. However, although a retinal break is a necessary precursor of RRD, all torn retinas do not detach. Only 35% of symptomatic horseshoe tears and few if any of asymptomatic retinal breaks ever progress to RRD.8 In addition to vitreoretinal traction and the consistency of the vitreous in the region of the retinal break (liquid versus gel), the physiologic forces that maintain retinal adhesion may play a significant role in determining the likelihood, extent, rate of progression, and clinical features of RRD after a retinal break has developed. Several important physiologic factors are known to contribute to retina–RPE adhesion, including adhesion molecules in the interphotoreceptor matrix, specifically cone matrix sheath glycoconjugates, active fluid transport from the subretinal space by the RPE, the ionic environment of the interphotoreceptor matrix, osmotic pressure gradients, and tissue oxygenation.14 –18 Conceivably, individual genetic differences in these factors may render the retina–RPE adhesion more or less susceptible to detachment in the presence of a retinal break. Rhegmatogenous retinal detachments with poor adhesion may peel off easily from the underlying RPE and appear bullous. In contrast, RRDs with stronger adhesion may appear more shallow and tense because vitreous fluid meets more resistance as it enters the subretinal space. These more tightly adherent detachments may have a lesser tendency to progress, and a complete demarcation line, whether physiologic or iatrogenic, may be enough to stabilize them. There are several potential advantages of demarcation laser over both surgery and observation in properly selected eyes with MSRRDs. Scleral buckling may have associated morbidity, including buckle-induced refractive error, diplopia, or vision-threatening infection or hemorrhage. Complications reported with pneumatic retinopexy include hemorrhage, infection, increased intraocular pressure, and new retinal breaks, among others.2 No complications developed as a result of DLP in this series. This is especially relevant for asymptomatic MSRRDs. Other MSRRDs particularly
1066
suited to DLP include those not amenable to pneumatic retinopexy because of inferior or widely separated breaks, those in patients unwilling or unable to undergo additional surgery, and those with pre-existing central visual loss as a result of macular degeneration or other maculopathy who are unaware of peripheral visual field loss. In summary, DLP consisting of a triple row of confluent laser photocoagulation appears to be an effective treatment in primary or recurrent, acute or chronic, symptomatic and asymptomatic MSRRDs without associated visual field loss or PVR and characterized by shallow subretinal fluid and a smooth, noncorrugated surface. Successful stabilization depends on both proper selection (shallow, tense, noncorrugated RRD) and adequate laser treatment (triple row of confluent laser spots extending around the entire detachment from ora to ora). Close follow-up to ensure there is no evidence of progression after DLP is also advisable. We recognize that the number of detachments we studied is small and that we did not randomize patients to observation versus treatment to assure that DLP indeed has a therapeutic effect, that is, that it reduces progression to macular detachment. A prospective, randomized, controlled trial would be required to study the usefulness of demarcation laser photocoagulation further as an alternative to both observation and surgical retinal repair in patients with shallow, primary, and recurrent macula-sparing rhegmatogenous retinal detachments without associated visual field loss. However, the low morbidity of this treatment suggests DLP with careful observation for progression, at which time prompt surgical repair could be performed, may be a reasonable approach in lieu of such an investigation.
References 1. Schepens CL, Okamura ID, Brockhurst RJ. The scleral buckling procedures. Arch Ophthalmol 1957;58:797– 811. 2. Tornambe PE. Pneumatic retinopexy: the evolution of case selection and surgical technique. A twelve year study of 302 eyes. Trans Am Ophthalmol Soc 1997;95:551–78. 3. Kreissig I, Failer J, Lincoff H, Ferrar F. Results of temporary balloon buckle in the treatment of 500 retinal detachments and a comparison with pneumatic retinopexy. Am J Ophthalmol 1989;107:381–9. 4. Vrabec TR. Laser photocoagulation repair of macula-sparing cytomegalovirus-related retinal detachment. Ophthalmology 1997;104: 2062–7. 5. Davis JL, Hummer J, Feuer WJ. Laser photocoagulation for retinal detachments and retinal tears in cytomegalovirus retinitis [see comments]. Ophthalmology 1997;104:2053– 60; discussion 2060 –1. Comment in: Ophthalmology 1998;105: 1353–5. 6. Okun E, Cibis PA. Photocoagulation in “limited” retinal detachment and breaks without detachment. In: McPherson A, ed. New and Controversial Aspects of Retinal Detachment. New York: Harper and Row, 1968; 164 –72. 7. Schepens CL. Subclinical retinal detachments. Arch Ophthalmol 1952;47:593– 606. 8. Davis MD. Natural history of retinal breaks without detachment. Arch Ophthalmol 1974;92:183–94. 9. Benson WE, Nantawan P, Morse PH. Characteristics and
Vrabec and Baumal 䡠 DLP of MSRRD
10. 11. 12. 13. 14.
prognosis of retinal detachments with demarcation lines. Am J Ophthalmol 1977;84:641– 4. Byer NE. The natural history of asymptomatic retinal breaks. Ophthalmology 1982;89:1033–9. Byer NE. Long-term natural history of lattice degeneration of the retina. Ophthalmology 1989;96:1396 – 401; discussion 1401–2. Brod RD, Flynn HW Jr, Lightman DA. Asymptomatic rhegmatogenous retinal detachments. Arch Ophthalmol 1995;113: 1030 –2. Yoon YH, Marmor MF. Rapid enhancement of retinal adhesion by laser photocoagulation. Ophthalmology 1988;95: 1385– 8. Hageman GS, Marmor MF, Yao XY, Johnson LV. The inter-
15. 16. 17. 18.
photoreceptor matrix mediates primate retinal adhesion. Arch Ophthalmol 1995;113:655– 60. Kita M, Marmor MF. Effects on retinal adhesive force in vivo of metabolically active agents in the subretinal space. Invest Ophthalmol Vis Sci 1992;33:1883–7. Kita M, Negi A, Marmor MF. Lowering calcium concentration in the subretinal space in vivo loosens retinal adhesion. Invest Ophthalmol Vis Sci 1992;33:23–9. Negi A, Marmor MF. Effects of subretinal and systemic osmolality on the rate of subretinal fluid resorption. Invest Ophthalmol Vis Sci 1984;25:616 –20. Marmor MF, Yao XY. The metabolic dependency of retinal adhesion in rabbit and primate. Arch Ophthalmol 1995;113: 232– 8.
1067
Originally received: August 17, 1999. Accepted: March 1, 2000. Manuscript no. 99563. 1 Wills Eye Hospital, Thomas Jefferson University, Philadelphia, Pennsylvania. 2 New England Eye Center, Tufts University, Boston, Massachusetts. Presented in part at the Pan American Association of Ophthalmology meeting, Orlando, Florida, October 1999. Correspondence to Tamara R. Vrabec, MD, 3244 Midvale Avenue, Philadelphia, PA 19129; E-mail: [email protected] © 2000 by the American Academy of Ophthalmology Published by Elsevier Science Inc.
Patients and Methods We reviewed the clinical records of all patients we treated with DLP between November 1992 and May 1999. A total of 39 patients (forty-three eyes) were identified. Patients to whom DLP was recommended were those who had MSRRDs without associated visual field loss or proliferative vitreoretinopathy (PVR). Seven eyes with MSRRD resulting from cytomegalovirus-related retinal detachment, previously reported,4 were excluded. One eye was excluded because follow-up was less than 3 months. One MSRRD resulting from retinoschisis with an outer wall hole was also excluded. Thirty-one patients (34 eyes) who underwent DLP for MSRRD were included. Data obtained included age and gender of each patient; history of retinal detachment repair; presence or absence of symptoms (photopsias, floaters); number and type of retinal tears; size (% retina involved as estimated by the treating retina specialist [TRV or CRB]); location and appearance (shallow versus bullous) of retinal detachments; associated findings including refractive error, demarcation line, lattice degeneration, vitreous hemorrhage, and lens status; best corrected pre- and postlaser visual acuity; progression beyond laser barrier; and length of follow-up. The presence or absence of posterior vitreous detachment (PVD) at the time of DLP was not specifically noted in patient records. One patient who reported visual field loss was managed with DLP because she ISSN 0161-6420/00/$–see front matter PII S0161-6420(00)00091-9
1063
Ophthalmology Volume 107, Number 6, June 2000
Figure 1. Large temporal rhegmatogenous retinal detachments caused by one horseshoe tear and two atrophic holes. Detached retina (to the left of the laser barrier) remains stable 4 months after demarcation laser photocoagulation. The fellow eye had a smaller detachment stabilized by demarcation laser photocoagulation.
declined scleral buckling surgery and was not a candidate for pneumatic retinopexy. Perimetry was not performed on any patients. Informed consent was obtained from all patients after a discussion of the risks and benefits of surgery (pneumatic retinopexy and scleral buckling), observation alone, and DLP, including the possibility of progression that would necessitate additional procedures if DLP failed to stabilize the detachment. Laser was applied in three rows of confluent 200-m grey-white burns that were placed posterior to the entire detachment and which extended to the ora serrata surrounding the entire detachment. Indirect laser delivery system facilitated peripheral treatment in some cases. After DLP, patients’ behavior was not limited in any way to prevent progression. Eyes in which subretinal fluid extended so close to the optic disc or fovea that three rows of confluent laser could not be applied without damaging these structures and eyes with vitreous hemorrhage dense enough to prevent complete laser treatment were not considered acceptable candidates for DLP.
Results Twenty males and 11 females (34 eyes) between the ages of 12 and 84 years underwent DLP for MSRRD. Twenty-eight MSRRDs were primary detachments and six were limited recurrent detachments after previous surgical repair with scleral buckle (n ⫽ 2), pneumatic retinopexy (n ⫽ 1), or PPV with (n ⫽ 2) or without (n ⫽ 1) silicone oil injection. Before laser demarcation, best corrected vision ranged from 20/20 to 20/400 in patients with no previous surgery and from 20/30 to counting fingers in patients who had previous retinal detachment repair. Six patients had bilateral retinal detachments. In three, bilateral detachments were treated with DLP (Fig 1). Three others had retinal detachments in fellow eyes that had been repaired previously with scleral buckle surgery. The extent of MSRRDs ranged from 10% to 45% of the retina (median, 20%). Eight MSRRDs affected at least 30% of the retina and also extended posterior to the equator. One extended across two quadrants, but was peripheral so that the total extent of detachment was less than 30%. The locations of MSRRDs were superior (n ⫽ 14), inferior (n ⫽ 13), and temporal (n ⫽ 7). Thirty-two MSRRDs were shallow, and two were dome shaped. None were highly elevated or corrugated (bullous). Three patients
1064
had more than one tear. Type of tears included dialysis (n ⫽ 1), horseshoe tear (n ⫽ 18), and round hole (n ⫽ 20). Fourteen MSRRDs had associated symptoms including photopsia and floaters. One eye (see below) had visual field loss. The duration of symptoms was less that 1 month in 11 eyes, and more than 1 month in 3 eyes. Associated ocular conditions included lattice degeneration in 12 eyes, limited vitreous hemorrhage in 4 eyes, and incomplete pigmented demarcation lines in 9 eyes. In three eyes, a depigmented line without subretinal fibrosis was noted in the attached retina immediately posterior to the subretinal fluid. Eight eyes were known to be myopic. Refractive status could not be determined from the charts in 18 cases. Eleven eyes were pseudophakic (1 iris-fixated lens, 10 posterior chamber lenses), and one was aphakic. In most cases, initial follow-up visits were 1 week, 3 weeks, 6 weeks, 3 months, and 6 months after the laser procedure. Follow-up ranged from 1.5 to 80 months (mean, 15.8 months; median, 17 months). The only MSRRD with less than 3 months of follow-up included in the series was the one that progressed 6 weeks after DLP. Three detachments flattened spontaneously after laser. In both primary and recurrent detachments, post-treatment best corrected visual acuity was either unchanged or improved by 1 to 4 lines (median, 1 line) in all but one eye. A cataract developed in this eye. Specifically, in eyes that had previous surgery, visual acuity after DLP was improved in two patients, unchanged in four patients, and worse in one because of cataract. In eyes with no previous surgery, visual acuity after DPL was improved in 10 and unchanged in 17. Best corrected postoperative vision ranged from 20/20 to 20/300 in eyes with primary detachments and from 20/30 to counting fingers in those who had previous surgery. Final visual acuity was 20/300 in one eye with primary detachment as a result of a disciform macular scar that had been present before DLP. One retinal detachment progressed 6 weeks after laser treatment. This eye was not believed to be an acceptable candidate for DLP because subretinal fluid was so close to the fovea that only 1 to 2 lines of laser could be delivered and because the patient was aware of visual field loss. Demarcation laser photocoagulation was performed because the patient initially declined scleral buckle surgery and was not a candidate for pneumatic retinopexy. When subretinal fluid progressed to threaten the fovea, the patient consented to scleral buckle repair. Vision improved from 20/40 before surgery to 20/30 after surgery and remained stable 17 months after scleral buckle surgery.
Discussion Vitreoretinal surgery is indispensable in the management of macula-off RRDs. However, optimal management strategy for certain MSRRDs including those that have been named limited, subclinical, and asymptomatic is ambiguous. Paucity of natural history data and a plethora of definitions and reasons for treating these detachments have made it difficult to glean a clear-cut treatment approach from the literature. This series is not the first to describe the use of laser as a treatment for retinal detachment. In 1968, Okun and Cibis6 recommended photocoagulation for “limited” retinal detachments. Limited detachments were defined by size. Criteria included width at least two times but not greater than five times the diameter of the largest break, and less than 2 clock hours in size with the posterior extent not beyond the equator. Breaks included tears, holes, and dialyses. No detachment was complicated by vitreous hemorrhage. The authors did not state whether detachments were
Vrabec and Baumal 䡠 DLP of MSRRD acute or chronic, or whether symptoms or demarcation lines were present. Laser demarcation of subretinal fluid plus photocoagulation facilitated by scleral depression of the areas around the retinal break and between the break and the demarcation line caused spontaneous flattening with chorioretinal scarring 2 to 3 weeks after treatment in 42 (88%) of 48 eyes. New breaks contributed to redetachment in three eyes. Laser failed to stabilize the detachment in three other eyes. Several authors have recommended surgical repair of MSRRDs. Schepens7 recommended prophylactic treatment to selected “subclinical” detachments he defined by the absence of symptomatic visual field loss. Because 50% of fellow eyes had symptomatic RRDs, he deduced that the subclincal RRD was a precursor to a symptomatic RRD. Thirty-eight (35%) of 110 RRDs were treated with external diathermy, in most cases without drainage of subretinal fluid. Eyes selected for treatment included those with RRD in the contralateral eye; superior or temporal RRDs; large, multiple, and equatorial tears; vitreous hemorrhage; and photopsia. Complete reattachment of the retina followed surgical obliteration of retinal breaks with diathermy in 97%. The clinical course of untreated eyes was not discussed. Davis8 also recommended surgical repair of subclinical detachments defined by subretinal fluid at least one disc diameter beyond the break but not more than two disc diameters posterior to the equator. Data from two groups, symptomatic and asymptomatic were analyzed. Six of seven subclinical RRDs with symptomatic (horseshoe) breaks were repaired immediately with surgery. The other progressed to “clinical” detachment (defined as involving one or two quadrants) without macular involvement 3.5 weeks after initial diagnosis and was repaired surgically. Because only one symptomatic subclinical RRD was observed, it is not possible to draw conclusions about progression in this subset. In the subset with asymptomatic breaks followed more than 6 months, 30% (6/20) progressed to “clinical” detachment. It is important to note that only one of these clinical detachments (4.5%) affected the macula. However, Davis concluded that prophylactic surgical treatment was “probably wise” for “subclinical” RRDs. Benson and associates9 recommended scleral buckling for RRDs with demarcation lines. Twenty of 66 eyes (30%) reviewed retrospectively were asymptomatic. All retinas were reattached surgically. Reattachment rate was 98.5%, but neither visual outcome nor complications in the asymptomatic maculasparing RRDs are reported. No asymptomatic cases were observed. It is uncertain whether they would have remained stable without treatment. Other authors have recommended observation for subclinical or asymptomatic RRDs. Byer,10 who defined subclinical RRDs by the same size criteria as Davis, observed only 3 of 18 asymptomatic detachments progressed slightly during a follow-up period ranging from 1 to 18 years. All detachments remained subclinical; none required surgery. In a subsequent study of eyes with lattice degeneration, he noted 2 of 10 subclinical RRDs progressed only slightly during a follow-up interval of 8 months to 23 years.11 No macular detachments developed in any. One eye was treated with a segmental buckle. Byer concluded in both studies
that because subclinical RRDs remain localized or progress only very slowly, surgical intervention should be discouraged. “Asymptomatic” detachments, discussed by Brod and associates,12 were in most cases larger than “subclinical” RRDs. None had associated photopsia or visual field loss. Demarcation lines were present in 74%, and 61% of the demarcation lines were complete. Of 31 RRDs, only 2 (6%) progressed, one 2.25 years and the other 3.3 years after initial diagnosis. Both were repaired with scleral buckle surgery. The authors recommended observation for asymptomatic RRDs. Some MSRRDs, also know as limited, subclinical, or asymptomatic, may progress to macular detachment. The detachments in this series included ones which were larger than those previously defined as subclinical, ones with horseshoe tears, acute onset, symptoms of vitreoretinal traction, vitreous hemorrhage, superior location, and detachments in fellow eyes, features that have been considered risk factors for progression. One of 23 detachments in this series that had one or more of these risk factors progressed. This series demonstrates that surgery may be safely avoided in many of these high-risk MSRRDs. It could be argued that observation alone may have been adequate in the MSRRDs without apparent risk factors. The literature suggests that many MSRRDs, especially those that are asymptomatic, small, shallow, with demarcation lines or caused by round holes, will remain stable without progression. However, previous series have documented that even low-risk detachments may progress.8,10 –12 None of the 20 asymptomatic detachments in our series progressed to detachment in comparison with the up to 16.6% rate of progression in previous series.8,10,12 It appears that DLP is a reasonable alternative to observation alone in these low-risk detachments. Laser photocoagulation rapidly enhances retinal adhesion in vitro and in vivo to 140% of normal in 24 hours, and twice normal between 3 days and 4 weeks.13 In this series, we found DLP an effective treatment for MSRRD. Only 1 of 34 detachments (2.9%) progressed beyond the laser barrier, and progression was attributable to incomplete demarcation in the macula. For laser to be effective, the importance of adequate and correctly placed treatment should be emphasized. Three confluent rows of laser photocoagulation burns in healthy retina must completely surround the entire detachment and extend to the ora serrata. Neither the extent nor the location of the detachment need be an exclusion criteria; however, detachments in the posterior pole should be treated only if three confluent rows of laser can be placed in attached retina between the subretinal fluid and the fovea or optic disc without damaging these structures. Insufficiently confluent or incomplete laser demarcation is likely to predispose to breakthrough of subretinal fluid. The feature shared among MSRRDs stabilized with DLP in this series was a shallow, relatively tense appearance without associated corrugations. Each MSRRD resembled a low, tense blister rather than a billowing parachute (Fig 2). Okun and Cibis6 and Schepens7 also noted RRDs that responded well to laser or to external diathermy were shallow. In considering why the chorioretinal adhesion created by DLP may prevent progression of shallow MSRRDs, and
1065
Ophthalmology Volume 107, Number 6, June 2000
Figure 2. Inferior macula-sparing rhegmatogenous retinal detachment treated with a triple row of confluent demarcation laser photocoagulation is stable 9 months after treatment. Note the shallow, smooth, noncorrugated appearance of this detachment.
in fact why some demarcation lines do as well, one must reconsider the factors important in the pathogenesis of RRD. Classically, vitreoretinal traction is thought to be most important. Certainly, it contributes to the development of horseshoe tears. However, although a retinal break is a necessary precursor of RRD, all torn retinas do not detach. Only 35% of symptomatic horseshoe tears and few if any of asymptomatic retinal breaks ever progress to RRD.8 In addition to vitreoretinal traction and the consistency of the vitreous in the region of the retinal break (liquid versus gel), the physiologic forces that maintain retinal adhesion may play a significant role in determining the likelihood, extent, rate of progression, and clinical features of RRD after a retinal break has developed. Several important physiologic factors are known to contribute to retina–RPE adhesion, including adhesion molecules in the interphotoreceptor matrix, specifically cone matrix sheath glycoconjugates, active fluid transport from the subretinal space by the RPE, the ionic environment of the interphotoreceptor matrix, osmotic pressure gradients, and tissue oxygenation.14 –18 Conceivably, individual genetic differences in these factors may render the retina–RPE adhesion more or less susceptible to detachment in the presence of a retinal break. Rhegmatogenous retinal detachments with poor adhesion may peel off easily from the underlying RPE and appear bullous. In contrast, RRDs with stronger adhesion may appear more shallow and tense because vitreous fluid meets more resistance as it enters the subretinal space. These more tightly adherent detachments may have a lesser tendency to progress, and a complete demarcation line, whether physiologic or iatrogenic, may be enough to stabilize them. There are several potential advantages of demarcation laser over both surgery and observation in properly selected eyes with MSRRDs. Scleral buckling may have associated morbidity, including buckle-induced refractive error, diplopia, or vision-threatening infection or hemorrhage. Complications reported with pneumatic retinopexy include hemorrhage, infection, increased intraocular pressure, and new retinal breaks, among others.2 No complications developed as a result of DLP in this series. This is especially relevant for asymptomatic MSRRDs. Other MSRRDs particularly
1066
suited to DLP include those not amenable to pneumatic retinopexy because of inferior or widely separated breaks, those in patients unwilling or unable to undergo additional surgery, and those with pre-existing central visual loss as a result of macular degeneration or other maculopathy who are unaware of peripheral visual field loss. In summary, DLP consisting of a triple row of confluent laser photocoagulation appears to be an effective treatment in primary or recurrent, acute or chronic, symptomatic and asymptomatic MSRRDs without associated visual field loss or PVR and characterized by shallow subretinal fluid and a smooth, noncorrugated surface. Successful stabilization depends on both proper selection (shallow, tense, noncorrugated RRD) and adequate laser treatment (triple row of confluent laser spots extending around the entire detachment from ora to ora). Close follow-up to ensure there is no evidence of progression after DLP is also advisable. We recognize that the number of detachments we studied is small and that we did not randomize patients to observation versus treatment to assure that DLP indeed has a therapeutic effect, that is, that it reduces progression to macular detachment. A prospective, randomized, controlled trial would be required to study the usefulness of demarcation laser photocoagulation further as an alternative to both observation and surgical retinal repair in patients with shallow, primary, and recurrent macula-sparing rhegmatogenous retinal detachments without associated visual field loss. However, the low morbidity of this treatment suggests DLP with careful observation for progression, at which time prompt surgical repair could be performed, may be a reasonable approach in lieu of such an investigation.
References 1. Schepens CL, Okamura ID, Brockhurst RJ. The scleral buckling procedures. Arch Ophthalmol 1957;58:797– 811. 2. Tornambe PE. Pneumatic retinopexy: the evolution of case selection and surgical technique. A twelve year study of 302 eyes. Trans Am Ophthalmol Soc 1997;95:551–78. 3. Kreissig I, Failer J, Lincoff H, Ferrar F. Results of temporary balloon buckle in the treatment of 500 retinal detachments and a comparison with pneumatic retinopexy. Am J Ophthalmol 1989;107:381–9. 4. Vrabec TR. Laser photocoagulation repair of macula-sparing cytomegalovirus-related retinal detachment. Ophthalmology 1997;104: 2062–7. 5. Davis JL, Hummer J, Feuer WJ. Laser photocoagulation for retinal detachments and retinal tears in cytomegalovirus retinitis [see comments]. Ophthalmology 1997;104:2053– 60; discussion 2060 –1. Comment in: Ophthalmology 1998;105: 1353–5. 6. Okun E, Cibis PA. Photocoagulation in “limited” retinal detachment and breaks without detachment. In: McPherson A, ed. New and Controversial Aspects of Retinal Detachment. New York: Harper and Row, 1968; 164 –72. 7. Schepens CL. Subclinical retinal detachments. Arch Ophthalmol 1952;47:593– 606. 8. Davis MD. Natural history of retinal breaks without detachment. Arch Ophthalmol 1974;92:183–94. 9. Benson WE, Nantawan P, Morse PH. Characteristics and
Vrabec and Baumal 䡠 DLP of MSRRD
10. 11. 12. 13. 14.
prognosis of retinal detachments with demarcation lines. Am J Ophthalmol 1977;84:641– 4. Byer NE. The natural history of asymptomatic retinal breaks. Ophthalmology 1982;89:1033–9. Byer NE. Long-term natural history of lattice degeneration of the retina. Ophthalmology 1989;96:1396 – 401; discussion 1401–2. Brod RD, Flynn HW Jr, Lightman DA. Asymptomatic rhegmatogenous retinal detachments. Arch Ophthalmol 1995;113: 1030 –2. Yoon YH, Marmor MF. Rapid enhancement of retinal adhesion by laser photocoagulation. Ophthalmology 1988;95: 1385– 8. Hageman GS, Marmor MF, Yao XY, Johnson LV. The inter-
15. 16. 17. 18.
photoreceptor matrix mediates primate retinal adhesion. Arch Ophthalmol 1995;113:655– 60. Kita M, Marmor MF. Effects on retinal adhesive force in vivo of metabolically active agents in the subretinal space. Invest Ophthalmol Vis Sci 1992;33:1883–7. Kita M, Negi A, Marmor MF. Lowering calcium concentration in the subretinal space in vivo loosens retinal adhesion. Invest Ophthalmol Vis Sci 1992;33:23–9. Negi A, Marmor MF. Effects of subretinal and systemic osmolality on the rate of subretinal fluid resorption. Invest Ophthalmol Vis Sci 1984;25:616 –20. Marmor MF, Yao XY. The metabolic dependency of retinal adhesion in rabbit and primate. Arch Ophthalmol 1995;113: 232– 8.
1067