Real-Time In Vivo Assessment of Retinal Reattachment in Humans Using Swept-Source Optical Coherence Tomography


To assess the in vivo physiology of retinal reattachment in humans using swept-source optical coherence tomography (SS-OCT) in real time.


Prospective case series.


Fifteen consecutive patients with fovea-involving rhegmatogenous retinal detachment were undergoing pneumatic retinopexy. SS-OCT was performed at presentation and frequent intervals immediately after pneumatic retinopexy. The primary outcome was longitudinal assessment of early postoperative SS-OCT to establish stages of reattachment.


Most patients (93.3%, 14/15) achieved successful reattachment at the median follow-up duration of 13 weeks (interquartile range 7.5-18.0). Reattachment occurred in 5 specific stages: 1) redistribution of fluid and approach of the neurosensory retina toward the retinal pigment epithelium occurred in 100% (15/15); 2) reduction in cystoid macular edema and improvement of outer retinal corrugations was achieved in 100% (15/15); 3) initial contact of the neurosensory retina to the retinal pigment epithelium occurred completely in 66.7% (10/15); 4) deturgescence of the inner and outer segments of the photoreceptors occurred in 66.7% (10/15); and 5) recovery of photoreceptor integrity occurred in 3 specific substages: 5A) external limiting membrane recovery (10/15, 66.6%); 5B) ellipsoid zone recovery (9/15, 60%); and 5C) interdigitation zone/foveal bulge recovery (3/15, 20%). Twenty percent (3/15) had delayed progression through stage 2, characterized by formation of outer retinal folds. Similarly, 33.3% (5/15) developed residual subfoveal fluid blebs (delayed progression to stage 3).


This study characterizes the in vivo physiology of retinal reattachment in humans using high-resolution SS-OCT that occurs in 5 specific stages. Delayed progression through certain stages was characterized by postoperative anatomic abnormalities. Am J Ophthalmol 2021;221:•••–•••. © 2021 Elsevier Inc. All rights reserved.

Rhegmatogenous retinal detachment (RRD) is an acute sight-threatening emergency. There have been numerous publications on the pathophysiology of RRD and the effectiveness of various techniques for surgical repair. Despite substantial improvements in surgical techniques over the past century focused on increasing the single-operation reattachment rate, there is limited understanding regarding the comparative anatomic integrity of retinal reattachment with various techniques. The refinement of multimodal imaging has enhanced our ability to assess the integrity of postoperative anatomic outcomes. Many authors have evaluated a variety of optical coherence tomography (OCT) biomarkers after RRD repair and have assessed their association with functional outcomes. For example, Kobayashi and associates assessed photoreceptor microstructure using OCT and determined its association with postoperative visual acuity.

It has been demonstrated that pneumatic retinopexy (PnR) has a greater chance of achieving a high-integrity retinal attachment (HIRA) with less retinal displacement compared with pars plana vitrectomy (PPV). Furthermore, PnR may be associated with better postoperative microstructural outer retinal integrity on OCT when compared with PPV. These findings have led us to question how the retina is reattached with various techniques and if specific variations in surgical technique may lead to improvements in the integrity of anatomic attachment. There are also several postoperative anatomic complications after retinal detachment repair including persistent subretinal fluid, outer retinal folds, cystoid macular edema, and epiretinal membrane formation that may impact functional outcomes. However, a detailed understanding of the pathophysiology of these complications is still evolving.

In 1968, Machemer , and Kroll and Machemer , conducted histologic and electron microscopy studies in owl monkeys to assess changes that occur after surgically induced RRD and after retinal reattachment. Similarly, Lewis and associates and Fisher and associates used an experimental animal model of RRD to characterize cellular changes that occur after retinal detachment of varying durations and after reattachment.

There have been no human in vivo studies assessing the physiology of retinal reattachment. This is related to difficulty in assessing patients immediately postoperatively during the rapid reattachment that occurs in the first few hours after procedures such as PPV and scleral buckle. In PPV, the presence of a large gas bubble prevents detailed assessment of the retinal microstructure with OCT. With scleral buckle, significant postoperative swelling, ocular surface abnormalities, and patient discomfort limit proper early evaluation. The minimally invasive PnR, which uses a small expansile gas bubble that does not obstruct the visual axis and the availability of high resolution swept-source OCT (SS-OCT) provides a unique opportunity to assess the in vivo physiology of human retinal reattachment in a prospective longitudinal fashion in the first several hours and days after the intervention in real time.


Study Design

This was a prospective consecutive case series of patients with fovea-involving RRD undergoing PnR at St. Michael’s Hospital/Unity Health Toronto, Toronto, Ontario, Canada between July 1, 2020 and September 30, 2020. This study was approved by the Research Ethics Board at St. Michael’s Hospital/Unity Health Toronto in Toronto and adhered to the Declaration of Helsinki.


Consecutive patients with a single or multiple retinal break(s) within 3 clock hours in detached retina above the 8- and 4-o’clock meridians with any number, location, and size of retinal breaks or lattice degeneration in the attached retina, proliferative vitreoretinopathy less than or equal to grade B, and who were undergoing treatment with PnR were eligible for the study. Exclusion criteria included a history of vitrectomy or scleral buckle in the affected eye, media opacity, and preexisting macular or retinal pathology.


All patients underwent PnR as described in the PIVOT randomized trial. Patients were given strict positioning instructions to place the apex of the bubble at the responsible retinal break(s), with or without the use of an initial steamroller maneuver with face-down positioning.

Data Collection and Imaging

All patients had clinical examination and underwent imaging with the PLEX Elite 9000 SS-OCT using high-definition horizontal 1 line spotlight scan (× 100) and a 12- × 12-mm macular cube (Carl Zeiss, Dublin, California, USA). Imaging was performed at presentation and every 2 hours for the first 6 hours after gas injection, at days 1, 2, and 5, and at weeks 1, 2, 4, and 6 after PnR. Additional imaging was also performed at 3 months in most patients. Imaging of the detached retina was performed with the tracker off, with the aim of capturing a foveal scan. After retinal attachment, images were captured using FastTrac that mitigates eye motion and operator related artifacts to allow the same point to be compared over time.


The primary outcome was the longitudinal assessment of the early postoperative SS-OCT changes after PnR for RRD to assess the in vivo human physiology of retinal reattachment in real time and to determine the stages of reattachment. All data are reported descriptively using medians and interquartile range (IQRs).


Fifteen eyes of 15 consecutive patients with fovea involving RRD were enrolled. The demographic details and baseline clinical characteristics are shown in Table 1 . The median age at presentation was 62.0 years (IQR 55.0-72.5). The median number of days since loss of central vision was 14.0 days (IQR 7.5-27.0) and the median baseline logarithm of minimal angle of resolution (logMAR) visual acuity was 2.0 (Snellen equivalent: count fingers at 0.61 meters [IQR 1.5-2.7]).


Baseline Demographic and Clinical Characteristics

Patient No. Age at Presentation (y) Gender No. of Days Since Loss of Central Vision Lens Status Laterality Clock Hours of Detachment Height of Detachment (Microns) Baseline logMAR Visual Acuity
1 75 F 1 Phakic OS 7 2897 1.3
2 58 M 2 Phakic OS 8 903 1.3
3 76 F 19 Pseudophakic OD 9 1006 1.3
4 69 F 21 Phakic OD 6 1062 1.3
5 23 F 14 Phakic OS 4 81 0.3
6 62 M 2 Pseudophakic OS 9 1463 3.0
7 52 M 14 Phakic OS 12 895 1.3
8 52 M 0 Pseudophakic OD 8 32 0.4
9 71 M 11 Phakic OS 8 2110 1.4
10 57 M 10 Pseudophakic OD 9 592 0.7
11 56 M 1 Phakic OS 4 942 0.7
12 54 M 14 Phakic OS 3 1283 2.0
13 77 M 15 Phakic OD 8 1385 3.0
14 73 M 40 Phakic OD 4 1136 2.0
15 72 M 27 Phakic OD 8 911 1.0

F = female; logMAR = logarithm of minimal angle of resolution; M = male; OD = right eye; OS = left eye.

One third (33.3%, 5/15) of patients presented within 3 days of central vision loss. The median clock hours of detachment was 8 (IQR 4.5-8.7). The median height of the retinal detachment at the fovea was 1006 μm (IQR 899-1334 μm). Almost all (93.3%, 14/15) patients had a successful retinal attachment with PnR at the median follow-up duration of 13.0 weeks (IQR 7.5-18.0 weeks). Median final logMAR visual acuity was 0.4 (Snellen equivalent 20/70 [IQR 0.1-0.6]).

All patients achieved retinal reattachment visible in real time on SS-OCT (Video 1; available online at Reattachment of every point on each patient’s retina occurred in 5 specific stages ( Figure 1 and Data in Brief Figures 1-4 ). Although the stages of reattachment could be applied to any point on the retina, the emphasis for this study was on the fovea. Stage 1, defined as a redistribution of fluid and approach of the neurosensory retina towards the retinal pigment epithelium (RPE), occurred in 100% (15/15) of patients. Stage 1 had to demonstrate a reduction in the height of the RRD compared with baseline. Stage 2 is characterized by a reduction in cystoid macular edema (CME) and improvement of outer retinal corrugations which was also achieved in 100% (15/15) of patients. Stage 3 is defined by the initial contact of the neurosensory retina to the RPE which occurred completely in 66.7% (10/15) and incompletely in 33.3% (5/15) of patients. Stage 4 was defined as deturgescence of the inner and outer segments of the photoreceptors which occurred completely in 66.7% (10/15) of patients. Stage 5 was characterized by a recovery of photoreceptor integrity which occurred in 3 substages (5A, external limiting membrane [ELM; 10/15, 66.7%]; 5B, ellipsoid zone [EZ; 9/15, 60%]; and 5C, interdigitation zone/foveal bulge [3/15, 20%; Table 2 ]).

Jul 10, 2021 | Posted by in OPHTHALMOLOGY | Comments Off on Real-Time In Vivo Assessment of Retinal Reattachment in Humans Using Swept-Source Optical Coherence Tomography

Full access? Get Clinical Tree

Get Clinical Tree app for offline access