Ophthalmic Surgery, Lasers and Imaging Retina

Clinical Science 

OCT Assistance in Identifying Retinal Breaks in Symptomatic Posterior Vitreous Detachments

Waseem H. Ansari, MD; Barton L. Blackorby, MD; Gaurav K. Shah, MD; Kevin J. Blinder, MD; Sabin Dang, MD

Abstract

BACKGROUND AND OBJECTIVE:

Posterior vitreous detachment (PVD) is a separation of the posterior hyaloid from the retina that manifests as photopsias and floaters. Optical coherence tomography (OCT) has demonstrated posterior vitreous opacities (PVOs) that may correlate with Shaffer's sign, which may correlate with retinal breaks.

PATIENTS AND METHODS:

Patients with symptomatic PVDs were retrospectively reviewed at a single institution by a single provider. Masked qualitative review of SD-OCTs by a single reviewer determined presence of PVOs.

RESULTS:

Among 78 patients, PVOs were found in 32 of the patients (41%), and 19 (59%) had retinal breaks. In those without PVOs, six (13%) had a break. Sensitivity and specificity were 76.0% and 75.5%, respectively. Removing patients with vitreous hemorrhages, sensitivity, and specificity of PVOs was 82.4% and 86.4%, respectively.

CONCLUSION:

In symptomatic PVDs, PVOs on OCT correlated with the presence of a retinal break, especially in the absence of a vitreous hemorrhage.

[Ophthalmic Surg Lasers Imaging Retina. 2020;51:628–632.]

Abstract

BACKGROUND AND OBJECTIVE:

Posterior vitreous detachment (PVD) is a separation of the posterior hyaloid from the retina that manifests as photopsias and floaters. Optical coherence tomography (OCT) has demonstrated posterior vitreous opacities (PVOs) that may correlate with Shaffer's sign, which may correlate with retinal breaks.

PATIENTS AND METHODS:

Patients with symptomatic PVDs were retrospectively reviewed at a single institution by a single provider. Masked qualitative review of SD-OCTs by a single reviewer determined presence of PVOs.

RESULTS:

Among 78 patients, PVOs were found in 32 of the patients (41%), and 19 (59%) had retinal breaks. In those without PVOs, six (13%) had a break. Sensitivity and specificity were 76.0% and 75.5%, respectively. Removing patients with vitreous hemorrhages, sensitivity, and specificity of PVOs was 82.4% and 86.4%, respectively.

CONCLUSION:

In symptomatic PVDs, PVOs on OCT correlated with the presence of a retinal break, especially in the absence of a vitreous hemorrhage.

[Ophthalmic Surg Lasers Imaging Retina. 2020;51:628–632.]

Introduction

Posterior vitreous detachment (PVD) is a slow separation of the posterior hyaloid from the internal surface of the retina. This process occurs throughout the eye and eventually culminates in the acute separation from the optic nerve. The sudden separation usually manifests as symptomatic flashes and floaters. Diagnosis is typically made when the examiner identifies a Weiss ring, a ring of dense vitreous, anterior to the posterior retina. Remaining traction on the peripheral retina can result in the formation of retinal breaks in the form of tears or holes or vitreous hemorrhages.1 These breaks must be located and treated to prevent the formation of a sight-threating retinal detachment. In those presenting with flashes and floaters, retinal breaks can be found in up to 25% of patients.2 Currently, standard of care dictates referral to a retinal specialist to perform a scleral depressed exam to rule out peripheral tears. Shafer's sign or tobacco dust, the presence of pigment or photo-receptors in the vitreous, have been shown to raise the likelihood that a retinal break is present.3

Optical coherence tomography (OCT) has been shown to demonstrate posterior vitreous opacities (PVOs) that may correlate with Shaffer's sign or tobacco dust. A previous study attempted to correlate this with the likelihood of finding a retinal break on exam.4 Our study performed a similar analysis to determine the sensitivity and specify of PVOs correlating to presence of retinal breaks in patients presenting with symptoms of flashes or floaters. These results can further assist in the diagnosis of acute retinal breaks in PVOs in symptomatic patients.

Patients and Methods

Data Collection

This retrospective study was performed at The Retina Institute of St. Louis, St. Louis, MO, after receiving approval from the St. Luke's Hospital institutional review board, Chesterfield, MO. All study-related procedures were performed in accordance with good clinical practice (International Conference on Harmonization of Technical Requirements of Pharmaceuticals for Human Use [ICH] E6), applicable U.S. Food and Drug Administration regulations, and the Health Insurance Portability and Accountability Act. A retrospective chart review was performed for patients who presented with a chief complaint of flashes and/or floaters and were newly diagnosed with a posterior vitreous detachment at single institution from January 2017 through July 2018 by a single physician.

Study Participants

The inclusion criteria were limited to patients who were older than 18 years of age, had complaints of flashes and/or floaters, received a spectral-domain OCT (SD-OCT) during their visit, and had diagnosis of new PVD as determined by SDOCT or exam. Patients were excluded if any condition (such as corneal opacities, dense cataracts) precluded obtaining an SD-OCT with a signal strength lower than 7/10. Additionally, any history of previous vitrectomy or retinal break was considered exclusion criteria.

Study Variables and Analysis

Study variables collected from the medical records included demographics such as age and gender, best-corrected visual acuity (VA), intraocular pressure (IOP), lens status, symptom complaints, and symptom duration. Blinded qualitative review of SD-OCT was performed by two different readers to determine presence of posterior vitreous densities in the overall cohort and defined subgroups. A single scan through the fovea and nerve was used for analysis in each eye. Scans were considered positive if they contained at least three opacities in the preretinal vitreous (Figure 1). Variables from the fundus examination included presence of vitreous hemorrhage, posterior vitreous detachment, and presence of break. Analysis included determining sensitivity, specificity, and positive and negative predictive values of the presence of posterior vitreous densities on OCT correlation with finding of a retinal break by determining the values for each reader and then averaging the two. Additional analysis of between additional variables was performed using standard t-tests.

(A) Optical coherence tomography (OCT) demonstrating posterior vitreous opacities (PVOs) (red) with retinal tear and no vitreous hemorrhage, representing true positive. (B) OCT without PVOs or tear, representing true negative. (C) OCT with PVOs (green and yellow) and no tear with vitreous hemorrhage, representing false positive.

Figure 1.

(A) Optical coherence tomography (OCT) demonstrating posterior vitreous opacities (PVOs) (red) with retinal tear and no vitreous hemorrhage, representing true positive. (B) OCT without PVOs or tear, representing true negative. (C) OCT with PVOs (green and yellow) and no tear with vitreous hemorrhage, representing false positive.

Results

In total, 76 eyes among 76 patients met criteria to be involved in the study. Average age was 63.3 years (range: 41 to 91 years) with a 29:47 male-to-female ratio. Overall, presenting logarithmic VA was found to be 0.148, or approximately 20/25, and presenting IOP was 16.5. Approximately 80.2% of patients were phakic at the time of presentation, and patients had an average symptom duration of 10.7 days. Approximately 50 patients (66%) complained of both floaters and flashes, and the remaining 26 (34%) complained of floaters only. Patients who presented with a retinal break were more likely to be phakic (95.6%; P ≤ .001) than in those who did not present with a retinal break (77.5%). Age, gender, presenting VA, IOP, or type or duration of symptoms were not statistically significant between those with a retinal break and those without or those with PVOs and those without (Table 1).

Demographic Summary of Study Patients and Comparisons of Those Found to Have a Break or PVOs Versus Those Without

Table 1:

Demographic Summary of Study Patients and Comparisons of Those Found to Have a Break or PVOs Versus Those Without

Interobserver agreement was calculated using the Pearson correlation coefficient and was found to be 0.77 (P < .001). Overall, 23 patients (30.2%) were found to have a retinal break, and PVOs were found in 28 patients (36.8%). Of those with PVOs, 16 (57.1%) of those patients had a retinal break and in those without PVOs, only seven (14.5%) had a break (Table 2). Sensitivity and specificity of this finding were found to be 65.2% and 75.4%, respectively. The positive predictive value and negative respective value were 53.5% and 83.3%, respectively. A total of 18 patients (23.6%) presented with a vitreous hemorrhage. When eliminating all patients with a vitreous hemorrhage, the sensitivity and specificity of PVOs were found to be 63.8% and 93.7%, respectively (Table 2). The positive and negative predictive value increased to 82.3% and 85.2%, respectively.

Presence or Absence of PVO With and Without Presence of Break in All Patients and a Subset of Patients Presenting Without a Vitreous Hemorrhage

Table 2:

Presence or Absence of PVO With and Without Presence of Break in All Patients and a Subset of Patients Presenting Without a Vitreous Hemorrhage

Discussion

Unilateral flashes and floaters can be explained by traction on the retina from an acute vitreous separation that can result in retinal breaks, which can eventually lead to irreversible vision loss from a retinal detachment. Thus, a referral to a retinal specialist is warranted and identification and treatment of all retinal breaks is essential to improve patient outcomes. Numerous clinical signs have been developed to aid the examiner by preventing a missed diagnosis. Shaffer's sign, or the presence of photoreceptors, or tobacco dust, the presence of pigment in the vitreous, have both been shown to raise the likelihood that a retinal break is present. Additionally, a vitreous hemorrhage has been shown to increase the likelihood a tear is present, as well.5,6

Currently, no guidelines exist to address the utility of imaging in the setting of acute PVD. OCT was first introduced in 1991 and as the technology evolved, its clinical utility has expanded to accompany a wide variety of pathologies.7 Recently, wide-angle OCT has been used to identify whether a PVD has occurred.8 Additionally, a recent study found that the presence of vitreous opacities on OCT may correlate to the presence of a retinal break in patients presenting with flashes and floaters.4 They determined that severe vitreous opacities were associated with a positive likelihood ratio of 9.78 (95% confidence interval, 3.02–31.63). Another study published in 2015 looked at a cohort of 40 patients and found that the presence of “falling ash” on OCT correlated with a sensitivity of 77.8% and a positive predictive value of 87.5%.9

Our study determined the specificity and sensitivity of these opacities. The threshold for a positive scan was set to the presence of greater than three opacities to rule out random artifacts based on prior clinical experience with OCT imaging on patients presenting with retinal breaks. Overall, the presence of posterior vitreous opacities leads to sensitivity and specificity of 65.2% and 75.4%, respectively, similar to the previously discussed study. A vitreous hemorrhage can artificially cause posterior vitreous opacities; therefore, a subset analysis was performed in patients who presented without a vitreous hemorrhage based on clinical exam. In this subset, the sensitivity and specificity of PVOs increased to 63.8% and 93.7% respectively. Therefore, an examination in the absence of vitreous hemorrhage and vitreous opacities on OCT has a low likelihood of identifying a retinal break. We found that the presence of a vitreous hemorrhage significantly correlated with presence of PVOs. This supports that idea that in those patients without a clinical vitreous hemorrhage PVOs may represent a subclinical vitreous hemorrhage.

This study contains the largest cohort analyzed thus far in the literature and provides strong evidence that PVOs are correlated with the presence of a retinal tear. Unlike previous studies, separating out the cofounder of vitreous hemorrhage greatly strengthens the diagnostic value. Limitations include the size of the patient database and reading only a single raster line. Future studies looking at automated identification of opacities on multiple raster scans using deep learning algorithms may further reduce the time needed to evaluate the OCTs and increase the accuracy of identification.

In patients presenting with symptoms of flashes and floaters, the presence of PVOs on OCT was found to correlate with the presence of a retinal break. Notably, in the absence of a clinical vitreous hemorrhage and PVOs, there was a 94% chance that no retinal break or tear was present. Therefore, an OCT can assist the examiner by providing useful information by increasing or decreasing the likelihood a retinal break has occurred with a symptomatic PVD.

References

  1. Abdolrahimzadeh S, Piraino DC, Scavella V, et al. Spectral domain optical coherence tomography and B-scan ultrasonography in the evaluation of retinal tears in acute, incomplete posterior vitreous detachment. BMC Ophthalmol. 2016;16(1):60 doi:10.1186/s12886-016-0242-0 [CrossRef] PMID:27215604
  2. Dayan MR, Jayamanne DG, Andrews RM, Griffiths PG. Flashes and floaters as predictors of vitreoretinal pathology: is follow-up necessary for posterior vitreous detachment?Eye (Lond). 1996;10(Pt 4):456–458. doi:10.1038/eye.1996.100 [CrossRef] PMID:8944097
  3. Tanner V, Harle D, Tan J, Foote B, Williamson TH, Chignell AH. Acute posterior vitreous detachment: the predictive value of vitreous pigment and symptomatology. Br J Ophthalmol. 2000;84(11):1264–1268. doi:10.1136/bjo.84.11.1264 [CrossRef] PMID:11049952
  4. Oh JH, Oh J, Roh HC. Vitreous Hyper-Reflective Dots in Optical Coherence Tomography and Retinal Tear in Patients with Acute Posterior Vitreous Detachment. Curr Eye Res. 2017;42(8):1179–1184. doi:10.1080/02713683.2017.1289226 [CrossRef] PMID:28358220
  5. Gishti O, van den Nieuwenhof R, Verhoekx J, van Overdam K. Symptoms related to posterior vitreous detachment and the risk of developing retinal tears: a systematic review. Acta Ophthalmol. 2019;97(4):347–352. doi:10.1111/aos.14012 [CrossRef] PMID:30632695
  6. Karahan E, Karti O, Er D, et al. Risk factors for multiple retinal tears in patients with acute posterior vitreous detachment. Int Ophthalmol. 2018;38(1):257–263. PMID:28160191
  7. Fujimoto J, Swanson E. The Development, Commercialization, and Impact of Optical Coherence Tomography. Invest Ophthalmol Vis Sci. 2016;57(9):OCT1–OCT13. doi:10.1167/iovs.16-19963 [CrossRef] PMID:27409459
  8. Tsukahara M, Mori K, Gehlbach PL, Mori K. Posterior Vitreous Detachment as Observed by Wide-Angle OCT Imaging. Ophthalmology. 2018;125(9):1372–1383. doi:10.1016/j.ophtha.2018.02.039 [CrossRef] PMID:29631900
  9. Rayess N, Rahimy E, Vander JF, et al. Spectral-Domain Optical Coherence Tomography in Acute Posterior Vitreous Detachment. Ophthalmology. 2015;122(9):1946–1947. doi:10.1016/j.ophtha.2015.03.030 [CrossRef] PMID:25935788

Demographic Summary of Study Patients and Comparisons of Those Found to Have a Break or PVOs Versus Those Without

All Patients (n = 78)With Break (n = 25)Without Break (n = 53)P ValueWith PVOs (n = 32)Without PVOs (n = 46)P Value
Age (Years)63.160.764.4.0664.162.2.31
Gender (M:F)30:4814:1116:37.1814:1816:30.10
% Phakic85.89671.2.01*81.877.8.23
Presenting Visual Acuity (logMAR)0.1440.1710.137.330.1410.157.60
Intraocular Pressure16.417.216.3.4616.516.4.93
Symptoms (Floaters: Floaters and Flashes)26:527:1819:34.459:2317:29.42
Duration of Symptoms (Days)10.38.310.8.226.512.2.06
Presence of Vitreous Hemorrhage21%22%20%.3745%4%< .001*

Presence or Absence of PVO With and Without Presence of Break in All Patients and a Subset of Patients Presenting Without a Vitreous Hemorrhage

All Patients
BreakNo BreakPPV59.4%
PVOs1913NPV87.0%
No PVOs640SENS76.0%
SPEC75.5%

Patients Without Vitreous Hemorrhages
BreakNo BreakPPV70.0%
PVOs146NPV92.7%
No PVOs338SENS82.4%
SPEC86.4%
Authors

From The Retina Institute of St. Louis, St. Louis, Missouri (WHA, BLB, GKS, KJB, SB); and Washington University School of Medicine, St. Louis, Missouri (KJB).

Presented in poster form at The American Society of Retina Specialists Annual Meeting in 2019 in Vancouver, BC, Canada.

Supported by Retina Research and Development Foundation, St. Louis, MO. The sponsor of funding organization had no role in the design or conduct of this research.

Drs. Ansari and Blackorby report no relevant financial disclosures. Dr. Shah has received personal fees from Regeneron, Allergan, DORC, Novartis, and OMIC outside the submitted work. Dr. Blinder has received personal fees from Regeneron, Allergan, Bausch + Lomb, and Novartis outside the submitted work. Dr. Dang has received personal fees from Regeneron outside the submitted work.

Address correspondence to Sabin Dang, MD, 2201 S. Brentwood Blvd., Brentwood, MO 63144; email: sabin@sabindangmd.com.

Received: April 12, 2020
Accepted: September 03, 2020

10.3928/23258160-20201104-04

Sign up to receive

Journal E-contents