Retinal Breaks: Clinical Course and Outcomes After Retinopexy

Patients are at risk for new retinal breaks or even retinal detachment in spite of retinopexy.


A 57-year-old physician with moderate myopia and a history of rhegmatogenous retinal detachment (RRD) in his left eye presented with flashes and floaters in his right eye and was found to have a single horseshoe retinal tear. He underwent laser retinopexy, but 1 month later he developed recurrent symptoms and was found to have a second retinal tear, also treated with laser. One week later, a third tear developed (Figure 1); this too was lasered. Two weeks later he developed a fourth tear that was lasered. Two years thereafter, he noted new floaters and was found to have mild vitreous hemorrhage and extension of the previous third tear, but there was no subretinal fluid and the tear remained within the laser barricade. He continues to be followed regularly, and 5 years later, his VA remains 20/20 in the right eye.

Figure 1. Sequential horseshoe retinal tears treated with laser retinopexy. Composite color fundus photograph from a middle-aged male with a history of RRD in the fellow eye who suffered 3 sequential, symptomatic horseshoe retinal tears over a 2-month period. Shown is an old horseshoe tear at 11:30 with laser scars, another tear at 10:30 with recent laser burns, and a fresh tear without laser at 9 o’clock. See text for more information on the patient’s clinical course.

This patient’s scenario is not uncommon and raises several questions for the retina specialist. When do I see the patient next? After retinopexy, what should we counsel patients regarding the risk of subsequent retinal breaks, need for additional procedures, or progression to RRD?

Jonathan F. Russell, MD, PhD, is a resident at Bascom Palmer Eye Institute, University of Miami, Miller School of Medicine in Miami, Florida. William E. Smiddy, MD, and Harry W. Flynn Jr., MD, are vitreoretinal specialists with Bascom Palmer Eye Institute. The authors report no related disclosures. Reach Dr. Flynn at

Editor’s note: This article is featured in a journal club episode of “Straight From the Cutter’s Mouth: A Retina Podcast.” Listen here: "Episode 73: Retinal Physician November 2017 Issue Discussion, PPV for Uveitis/Retinal Tears/Future of Retina."


Retinal breaks are full-thickness discontinuities in the neurosensory retina. They can be classified into tears, such as horseshoe tears and giant retinal tears, or holes, such as operculated holes and atrophic holes. Horseshoe tears classically appear as U-shaped retinal “flaps” (but the shape may vary widely) and are clearly caused by vitreous traction. These types of breaks are often found at, or anterior to, the equator (Figure 1). Giant retinal tears are defined as encompassing 3 or more clock hours, are often idiopathic, and are less commonly due to trauma, high myopia, or hereditary vitreoretinopathy such as Stickler syndrome. Tractional or atrophic macular holes are not discussed here. Operculated holes are generally round retinal breaks with overlying avulsed retina, which usually indicates release of traction from the retinal surface. Atrophic holes are not associated with vitreous traction and are more common in myopes or in areas of lattice retinal degeneration. Lattice retinal degeneration is a thinning of neurosensory retina that can also be associated with retinal tears at the posterior or lateral margin, where the vitreous is more firmly attached, particularly when an acute posterior vitreous detachment (PVD) occurs.

Retinal breaks can be asymptomatic but may occur with PVD, as heralded by symptoms of floaters, photopsias, or a visual field deficit (if there is associated RD). Prompt clinical examination aimed at detecting a tear before ensuing retinal detachment is important. Shafer sign, which is characterized as large gold-brown particles seen in the anterior vitreous, is about 90% predictive of a retinal break in phakic patients.1,2 A Weiss ring is peripapillary glial tissue suspended in the vitreous cortex, which indicates that a PVD has occurred. In eyes with acute, symptomatic PVDs, there is a 4% to 11% risk of a retinal break.3-5 However, nearly half of symptomatic PVDs with an associated retinal break lack a clearly visible Weiss ring.6 The presence of vitreous hemorrhage is 50% to 70% predictive of a retinal break.3,4 If there is no identifiable break on initial examination, the finding of hemorrhage at presentation connotes a roughly 20% risk of a break within 6 weeks.5 Even in the absence of Shafer sign, a Weiss ring, or hemorrhage, a subtle retinal break may be present. Therefore, any patient presenting with symptoms should undergo thorough indirect ophthalmoscopy with scleral indentation as needed. When media opacities such as dense cataract or vitreous hemorrhage preclude complete ophthalmoscopy to the ora serrata, B-scan ultrasonography is an acceptable alternative.

Risk factors associated with retinal breaks include myopia, aphakia, pseudophakia, inflammatory retinitis, lattice retinal degeneration, cystic retinal tufts, hereditary vitreoretinopathies, blunt trauma, and penetrating trauma. A common association of PVD, and by extension a retinal tear, is aging; vitreous syneresis (liquefaction) and synchysis (shrinkage) leads to PVD in 65% of eyes.7 With PVD liquefied vitreous migrates posterior to the vitreous gel, which then may exert traction on areas of abnormal vitreoretinal adhesion. If the tractional adhesive forces exceed retinal cohesive forces, a retinal break may ensue. Liquid within the vitreous cavity can then enter the subretinal space, gradually progressing from a cuff of subretinal fluid (SRF) to RRD.


Untreated, symptomatic horseshoe tears have been reported to lead to RRD in 30% to 50% of cases.8,9 Symptomatic horseshoe tear features that are particularly high risk for progression to RRD include breaks of acute onset, breaks located superior to the horizontal meridian, large breaks, and tears with either a cuff of SRF or subclinical RD, which is defined as extension of SRF at least 1 disc diameter away from the break but no more than 2 disc diameters posterior to the equator.

In contrast, operculated breaks imply release of focal vitreoretinal traction, as evidenced by the edges of the tear lying flat, and therefore the risk of progression to RRD is very low. Only 2 cases of symptomatic operculated breaks progressing to RRD have been recorded.8,9 In both cases, vitreoretinal traction persisted on a retinal vessel adjacent to the break. Similarly, asymptomatic retinal breaks carry a very low risk of progression to RRD.10 Even superior horseshoe tears in young, phakic patients, if they are asymptomatic, rarely lead to RRD.11,12

Lattice retinal degeneration is present in 6% to 8% of the general population but is present in 30% of phakic RRDs.13 Classically, retinal tears occur at the lateral or posterior margin of lattice areas. Young myopes with lattice-associated atrophic holes can have localized, slowly progressive, and often asymptomatic RRDs.14 This is uncommon, however, as there is a less than 1% risk of RRD in an eye with lattice if an RRD has not occurred in the fellow eye.13 If an RRD has occurred in the fellow eye, lattice carries a 2% to 5% risk of RRD.15


It is generally agreed that there is sufficient evidence to justify prompt treatment of acute-onset, symptomatic horseshoe tears as well as giant retinal tears. For other retinal breaks, there is insufficient evidence for treatment, and the surgeon must balance risks, benefits, and alternatives in the context of the patient’s clinical examination. Symptomatic operculated breaks are usually observed unless there is evidence of persistent vitreoretinal traction adjacent to the break. Likewise, all asymptomatic retinal breaks — horseshoe tears, operculated breaks, atrophic holes, and lattice retinal degeneration — can usually be observed. Exceptions may include asymptomatic breaks in fellow eyes of patients with a history of retinal detachment, or asymptomatic horseshoe tears that are either associated with lattice retinal degeneration or found in eyes with impending cataract surgery. Clinical judgment must be used as to the patient’s individual risk factors for progression to RRD.

Apart from giant retinal tears, which are usually treated with surgery, the treatment of choice for symptomatic retinal tears without associated RRD is retinopexy. Retinopexy with either laser photocoagulation or cryotherapy induces chorioretinal adhesions that counter vitreous traction and prevent liquefied vitreous from entering the subretinal space to cause RRD. Most surgeons perform laser retinopexy consisting of 2 or more near confluent rows of moderately white burns (Figure 2). A single row is probably less effective.16 The laser power is titrated to yield retinal whitening, which subsequently leads to pigmentation (Figure 3). Practically speaking, without substantial whitening (and later pigmentation) it can be difficult to be sure that laser completely surrounds the break. Laser burns are particularly difficult to visualize in a blonde fundus. Excessively heavy burns, on the other hand, may cause focal hemorrhage, excessive inflammation, and cystoid macular edema and carry the theoretical risks of inducing proliferative vitreoretinopathy (PVR) and epiretinal membrane (ERM) formation.17 The pigmentation associated with laser scars begins to form about 4 days after laser application.18

Figure 2. Fresh laser retinopexy of a symptomatic horseshoe retinal tear. A 360-degree laser barricade with more than 3 rows of moderately white burns was performed to induce chorioretinal adhesion around the retinal break.

Figure 3. Laser retinopexy scars around a retinal tear. Pigmentation begins to form about 4 days after retinopexy. This photo was taken 1 month after laser application.

Retinopexy laser is applied to form a 360-degree barricade around the retinal break (Figure 4). It is also effective to laser demarcate anterior breaks in a U-shaped fashion extending to the ora serrata to ensure an anterior margin of barricade. Cryotherapy can be useful for far anterior breaks or if visibility is difficult. It may also be useful in cases of unclear media not amenable to application with red laser. Cryotherapy can cause pain and conjunctival damage and may increase dispersion of retinal pigment epithelium cells, which may lead to PVR and ERM formation.19

Figure 4. Six-month follow-up of a symptomatic retinal tear treated with retinopexy. Note the treatment of the anterior margin of the tear.


Retinopexy may fail to prevent RRD for multiple reasons. The first is inadequate treatment to all margins, especially the anterior margins of the break, where the traction may be most dynamic.16 Second, there may have been multiple retinal breaks, but a secondary break was missed on initial examination. Third, new retinal breaks may develop elsewhere after retinopexy. A small percentage of eyes progress to RRD despite adequate treatment.

The rates of new break formation are summarized in Table 1. New retinal breaks can be single or multiple and are frequently clustered within a few clock hours of the original break.20 Studies from the 1960s and 1970s reported that 2.7% to 8.0% of treated eyes developed a new break and roughly 5% progressed to RRD. In contrast, more recent studies have reported a higher rate of new break formation. In one 1991 series of 171 eyes, 14% of treated eyes developed a new break. In that study, 22% of eyes treated for retinal breaks required additional treatment: 5% had inadequate closure of the original break, 9% developed new breaks without RRD, 5% developed new breaks with RRD, and in 4% a RRD developed from the original, presumably adequately treated break.21 About half of the retreatments took place within 1 month, but nearly a quarter were after 6 months.

Table 1: Outcomes After Retinopexy for Retinal Tears
301 5.5 6.0 1 Robertson and Norton, 197316
177 7.3 4.5 2 Combs and Welch, 198223
74 7.3 2.7 NR Straatsma et al, 196524
231 NR 5.0 NR Chignell and Shilling, 197325
701 8.0 4.7 NR Kanski and Daniel, 197526
231 0 0 0 Morse and Scheie, 197427
83 10.0 NR NR Goldberg and Boyer, 198020
171 14.0 9.0 5 Smiddy et al, 199121
155 12.2 3.0 NR Sharma et al, 200422
2,124 7.3 4.6 1.6
Adapted from Smiddy et al.21 RRD, rhegmatogenous retinal detachment; ERM, epiretinal membrane; NR, not reported. In calculating total percentages, a study was omitted if it did not report the respective outcome measures.

Another retrospective study reported that 12% of treated eyes developed new breaks, but only 3% of treated eyes went on to develop RRD.22 About two-thirds of eyes needing retreatment presented within 6 months, and 21% of the retreatments took place more than 1 year after the initial treatment.22 All retreated eyes had developed recurrent visual symptoms.22 Notably, all of the studies to date are limited by their single-center, retrospective nature with relatively short follow-up periods. In summarizing the literature, it is appropriate to counsel patients after retinopexy that the risk of new retinal breaks requiring additional laser is approximately 1 in 10 and the risk of RRD requiring surgery is approximately 1 in 20 (Table 1).

As for other post-treatment issues, one large series observed ERM formation in 5% of eyes after retinopexy.21 Nearly half of these retinopexies had involved at least some cryotherapy, and yet less than 1% of the subsequent ERMs were clinically significant enough to merit surgical intervention. Most contemporary retinopexies are performed with laser alone, and most retina specialists think the rate of ERM formation is lower despite OCT facilitating detection. Because studies to date have had short follow-up periods, it is unclear whether eyes can develop ERMs many years after retinopexy. In fact, because most retinal tears are treated with retinopexy, it is unknown whether ERMs form due to retinopexy or as a result of simply having a retinal break, because breaks may lead to the release of retinal pigment epithelial and glial cells which could later form an ERM. Finally, additional adverse events after laser retinopexy, such as burns to anterior segment structures or the macula, are rare.


Patients can be counseled that retinopexy for symptomatic retinal tears is effective and far safer than not treating, but regular follow-up examinations should be performed to ensure adequate treatment of the initial tear and detection of any subsequent breaks or progression to RRD. Patients with retinal tears should be counseled that persistence of flashes and floaters is expected but that they usually fade with time. Any new photopsias, floaters, or visual field deficits should prompt return, because they may reflect a new retinal break or progression to RRD. RP


  1. Brod RD, Lightman DA, Packer AJ, Saras HP. Correlation between vitreous pigment granules and retinal breaks in eyes with acute posterior vitreous detachment. Ophthalmology. 1991;98(9):1366-1369.
  2. Tanner V, Harle D, Tan J, et al. Acute posterior vitreous detachment: the predictive value of vitreous pigment and symptomatology. Br J Ophthalmol. 2000;84(11):1264-1268.
  3. Jaffe NS. Complications of acute posterior vitreous detachment. Arch Ophthalmol. 1968;79(5):568-571.
  4. Byer NE. Natural history of posterior vitreous detachment with early management as the premier line of defense against retinal detachment. Ophthalmology. 1994;101(9):1503-1513.
  5. Van Overdam KA, Bettink-Remeijer MW, Klaver CC, et al. Symptoms and findings predictive for the development of new retinal breaks. Arch Ophthalmol. 2005;123(4): 479-484.
  6. Goh YW, Ehrlich R, Stewart J, Polkinghorne P. The incidence of retinal breaks in the presenting and fellow eyes in patients with acute symptomatic posterior vitreous detachment and their associated risk factors. Asia Pac J Ophthalmol (Phila). 2015;4(1):5-8.
  7. Foos RY. Posterior vitreous detachment. Trans Am Acad Ophthalmol Otolaryngol. 1972;76(2):480-497.
  8. Colyear BH Jr, Pischel DK. Clinical tears in the retina without detachment. Am J Ophthalmol. 1956:41(5):773-792.
  9. Davis MD. Natural history of retinal breaks without detachment. Arch Ophthalmol. 1974;92(3)183-194.
  10. Posterior vitreous detachment, retinal breaks, and lattice degeneration PPP - 2014. American Academy of Ophthalmology: 2014. Available at .
  11. Byer NE. What happens to untreated asymptomatic retinal breaks, and are they affected by posterior vitreous detachment? Ophthalmology. 1998;105(6):1045-1049.
  12. Byer NE. The natural history of asymptomatic retinal breaks. Ophthalmology. 1982:89(9):1033-1039.
  13. Byer NE. Long-term natural history of lattice degeneration of the retina. Ophthalmology. 1989;96:1396-1401.
  14. Winslow RL, Tasman W. Juvenile rhegmatogenous retinal detachment. Ophthalmology. 1978;85(6):607-618.
  15. Folk JC, Arrindell EL, Klugman MR. The fellow eye of patients with phakic lattice retinal detachment. Ophthalmology. 1989;96(1):72-79.
  16. Robertson DM, Norton EW. Long-term follow-up of treated retinal breaks. Am J Ophthalmol. 1973;75(3):395-404.
  17. Morales PH, Hernandez E, Rolon A, Cousins SW. Laser photocoagulation increases PVR severity. Invest Ophthalmol Vis Sci. 1997;38(4):S672.
  18. Kain HL. Chorioretinal adhesion after argon laser photocoagulation. Arch Ophthalmol. 1984;102(4):612-615.
  19. Glaser BM, Vidaurri-Leal J, Michels RG, Campochiaro PA. Cryotherapy during surgery for giant retinal tears and intravitreal dispersion of viable retinal pigment epithelial cells. Ophthalmology. 1993;100(4):466-470.
  20. Goldberg RE, Boyer DS. Sequential retinal breaks following a spontaneous initial retinal break. Ophthalmology. 1980;88(1):10-12.
  21. Smiddy WE, Flynn HW Jr, Nicholson DH, et al. Results and complications in treated retinal breaks. Am J Ophthalmol. 1991;112(6):623-631.
  22. Sharma MC, Regillo CD, Shuler MF, et al. Determination of the incidence and clinical characteristics of subsequent retinal tears following treatment of the acute posterior vitreous detachment-related initial retinal tears. Am J Ophthalmol. 2004;138(2):280-284.
  23. Combs JL, Welch RB. Retinal breaks without detachment: natural history, management and long term follow-up. Trans Am Ophthalmol Soc. 80:64-97.
  24. Straatsma BR, Allen RA, Christensen RE. The prophylaxis of retinal detachment. Trans Pac Coast Otoophthalmol Soc Annu Meet. 1965;46:211-236.
  25. Chignell AH, Shilling J. Prophylaxis of retinal detachment. Br J Ophthalmol. 1973;57(5):291-298.
  26. Kanski JJ, Daniel R. Prophylaxis of retinal detachment. Am J Ophthalmol. 1975;79(2):197-205.
  27. Morse PH, Scheie HG. Prophylactic cryoretinopexy of retinal breaks. Arch Ophthalmol. 1974;92(3):204-207.