Islets persisting on OCT macular thickness map after intravitreal injections unmask telangiectatic capillaries (TelCaps) in diabetic macular edema.

Save

Purpose

Recurrence of macular edema after intravitreal injections (IVI) may result from persistent focal microvascular abnormalities such as telangiectatic capillaries (TelCaps), which can be difficult to detect within a thickened retina. This study proposes a simple screening strategy to identify TelCaps using optical coherence tomography (OCT) thickness maps following anti-VEGF treated for diabetic macular edema (DME).

Methods

This is an ancillary study from the TalaDME double-blind randomized controlled trial evaluating the benefit of adjunctive focal laser photocoagulation to anti-VEGF therapy in eyes with DME and TelCaps. OCT imaging was performed monthly and indocyanine green angiography (ICGA) quarterly over 12 months. Twenty-eight fellow eyes that had recovered from center-involved DME (CI-DME) associated with TelCaps were analyzed.

TelCaps were identified as late-phase hyperfluorescent lesions on ICGA. The topography of perifoveal retinal thickening on OCT and its spatial correlation with TelCaps were assessed.

Results

Following IVI induced resolution of central DME (defined as central macular thickness < 315 µm), OCT mapping revealed 52 perifoveal thickening spots, herein referred to as islets. TelCaps were located within 100 µm of the barycenter of 47/52 (90%) islets, observed in 27/28 (98%) eyes. During follow-up, 25 (89%) eyes showed islets enlargement, with 20 (71%) progressing to involve the fovea.

All cases of central DME recurrence occurred in eyes with post-IVI islets, and recurrent edema was topographically continuous with these islets. No recurrence was observed in eyes with documented TelCaps occlusion.

Conclusion

Following anti VEGF treatment for CI-DME associated with TelCaps, OCT thickness mapping can reveal islets of retinal thickening that frequently colocalize with residual TelCaps, potentially heralding recurrence of central edema. A systematic search for these islets may facilitate the diagnosis of TelCaps.

Introduction

Diabetic macular edema (DME) is a leading cause of vision loss in the working-age population worldwide [1]. The introduction of intravitreal injections (IVI) of anti-vascular endothelial growth factor (anti-VEGF) agents and corticosteroids has significantly improved the management of DME [2]. However, due to the high treatment burden and costs associated with repeated IVI, alternative or adjunctive strategies deserve consideration.

Cost-effectiveness analyses have shown that initial laser therapy followed by as-needed IVI can reduce treatment costs by up to 50% compared to initial IVI alone [3, 4]. As such, laser photocoagulation remains a relevant therapeutic option [5–7]. In real-world practice, 37% to 50% of eyes of patients with DME receive laser therapy in addition to IVI, according to data from the Diabetic Retinopathy Clinical Research Network (DRCRnet) [8].

While DRCRnet guidelines still support selective use of focal/grid laser in some cases of DME [9], current European guidelines tend to discourage laser use [10].

The standard approach to laser therapy for DME is based on the modified Early Treatment Diabetic Retinopathy Study (ETDRS) protocol. This includes focal photocoagulation of leaking microaneurysms and grid laser targeting the retinal pigmentary epithelium (RPE) in areas of thickening or nonperfusion, suggesting that treating both microvascular lesions and the RPE may enhance anatomical outcomes compared to RPE treatment alone [11, 12]. However, focal treatment of microaneurysms alone has also demonstrated comparable and functional results [13].

Recent research has drawn attention to large retinal capillary aneurysms ([14] telangiectactic capillaries (TelCaps) — as significant contributors to persistent or recurrent macular edema. TelCaps have been identified in up to 60% of eyes with chronic DME [15] and appear to be viable targets for focal laser therapy [16–19]. Accurate detection of TelCaps is therefore of clinical interest.

While fluorescein angiography often fails to detect TelCaps, late-phase indocyanine-green angiography (ICGA) has demonstrated superior sensitivity [20]. Alternatively, some studies have proposed using infrared reflectance or optical coherence tomography (OCT) to guide focal laser treatment [13, 21]. However, identifying TelCaps on OCT may be challenging when the retina remains thickened.

In this study, we propose a novel approach to identify TelCaps by analyzing OCT thickness maps after intravitreal therapy in patients with DME. This method may provide a practical, non-invasive strategy to guide targeted laser treatment in selected cases.

Material and methods

Study design and participants

This study is a post-hoc retrospective analysis of prospectively collected data from the fellow eyes of patients enrolled in the TalaDME (Targeted Laser for Diabetic Macular Edema) double-blind randomized controlled trial (EudraCT 2018-A00800-55 and NCT03751501), conducted across ten French university hospitals. The TalaDME trial investigates the potential benefit of adjunctive focal laser therapy in patients with DME and TelCaps [21], and received ethics committee approval. All procedures adhered to the tenets of the Declaration of Helsinki.

Throughout the 12-month follow-up, best corrected visual acuity (VA) and OCT were assessed monthly in both eyes, while ICGA was performed every three months. ICGA and spectral-domain OCT were acquired using the combined OCT-SLO Spectralis system (Heidelberg Engineering, Heidelberg, Germany). OCT settings included a macular cube scan (30° × 25°), 60µm spacing; and high resolution ART 5 mode.

ICGA images were obtained with a 30° field of view.

At each visit, VA, OCT central macular thickness (CMT), and the presence of TelCaps (as detected by ICGA) were recorded in both eyes. TelCaps were defined as: “isolated lesions” > 150 µm in diameter, or “clustered dot-like lesions” (at least three lesions, each ≥ 120 µm in diameter on OCT) grouped within a circle no larger than 1000µm in diameter, exhibiting sustained staining on ICGA images beyond 10 to 20 min post-injection [22].

The present ancillary study therefore included the fellow eyes of TalaDME participants who met the following inclusion criteria: i) baseline CI-DME associated with TelCaps, ii) achievement of CMT ≤ 315 µm after initial anti-VEGF IVI, and iii) minimum follow-up duration of 12 months. Eyes with coexisting macular diseases, including epiretinal membrane, vitreomacular traction, age-related macular degeneration, or retinal vascular occlusion, were excluded. Eyes were managed using a pro re nata (PRN) regimen until month 12.

Decisions regarding retreatment with IVI and/or focal laser were made at the discretion of the treating investigators. Images were independently graded by 2 graders (BD and DCF).

After IVI-induced resolution of central edema, pericentral thickening spots —referred to here as islets— were defined as retinal areas thicker than 350 µm, appearing orange-red to white on the Spectralis thickness map [23], with barycenters located outside the central 1000-µm ETDRS circle, with or without surrounding hard exudates. To evaluate the topographical relationship between residual thickening and TelCaps, OCT thickness maps were superimposed onto the late-phase ICGA images. During follow-up, DME recurrences were defined as ≥ 20% increase in thickness in the central and/or paracentral ETDRS subfields [24].

The spatial distribution of recurrent thickening was correlated with the location of previously identified islets and TelCaps.

Statistical analysis

Statistical analyses were performed using Stata Statistical Software, Release 18 (StataCorp LLC, College Station, TX). Quantitative variables were expressed as mean ± standard deviation (SD) or median with interquartile range [IQR], as appropriate. The Wilcoxon matched-paired test was used to compare changes in CMT, TelCaps count, and islets count over time.

All p-values were two-sided, and a significance threshold of p < 0.05 was applied.

Results

The flowchart of patient inclusion is presented in Fig. 1. Among the 148 diabetic patients enrolled in the prospective longitudinal TalaDME trial, 28 (20%) had a fellow eye that met the inclusion criteria for the ancillary analysis.

Table 1 summarizes the main baseline characteristics of the included patients and eyes, as well as the number of IV) and laser treatments administered over the 12-month period. All eyes received intravitreal injections (IVI), with a mean of 5.5 ± 3 anti-VEGF injections over 12 months. Seven of 28 eyes (25%) also received targeted laser treatment.Fig.

1Flowchart. IVI: intravitreal injection of anti-vascular endothelial growth factor; CI-DME: center-involved diabetic macular edema; CMT: central macular thicknessTable 1Characteristics of patients and eyesTotal number of eyes (one per patient), n28Age, years, mean ± SD62.8 ± 8.1Females, n (%)13 (46)Diabetes type 2, n (%)28 (100)Diabetes duration, years, mean ± SD16.9 ± 11.0HbA1C %, mean ± SD7.7 ± 0.8Hypertension, n (%)26 (87)DR severity scale * Mild NPDR2 Moderate NPDR4 Severe NPDR4 Panretinal photocoagulation for PDR18mean baseline ETDRS score58.2 ± 12.4mean baseline CMT µm ± SD453 ± 155mean number of TelCaps per eye ± SD2.6 ± 1.2Presence of circinate hard exudates, n (%)24 (86)Exudates reaching the inner ETDRS circle, n (%)12 (42)Number of eyes treated with anti-VEGF injections over 12 months, n (%)28 (100)Number of anti VEGF injections over 12 months, mean ± SD5.5 ± 3Number of eyes treated with targeted laser, n (%)7 (25)*ETDRS -Simplified American Academy of Ophthalmology Diabetic Retinopathy Severity scale (AAO 2003)ETDRS early treatment diabetic retinopathy study, DR diabetic retinopathy, VA visual acuity, CMT central macular thickness, NPDR non-proliferative diabetic retinopathy, PDR proliferative diabetic retinopathy

At baseline, the mean CMT was 453 ± 155 µm, and the mean number of TelCaps per eye was 2.6 ± 1.2. Circinate hard exudates were observed in 24 (86%) of cases.

At month 3, following the IVI administered at baseline and normalization of CMT to ≤ 315 µm (mean CMT 273 ± 44 µm), a total of 52 islets of pericentral thickening were identified across the 28 eyes (mean: 1.8 ± 1.0 islets per eye). At this timepoint, the mean number of TelCaps per eye was 2.1 ± 1.4 (p = 0.02 compared to baseline) (Fig. 2).

TelCaps were identified on ICGA within 100 µm of the barycenter of 47 out of 52 islets (90%), in 27 out of 28 eyes (98%). In the remaining five islets, no focal late ICGA staining was observed (Fig. 3).

This indicates that in approximately 10% of cases, islets were not centered on ICG-stained TelCaps, but microaneurysms may be visible. Conversely, in 10 out of 28 eyes (35%), ICG-stained TelCaps were visible on ICGA despite normal macular thickness in the surrounding retina (Fig. 2).Fig.

2Representative illustration of post-IVI islets. All images are superimposition of optical coherence tomography images and late indocyanine green angiogram images. A: Two islets with TelCaps located at the summit of each.

B. 2 islets located superior to the macula, as well as one temporal crescent-shaped islet formed by the coalescence of several mini-islets, each centered by TelCaps. There is presence of TelCaps stained with ICG, not associated with retinal thickening (arrows)Fig.

3Example of islets with and without ICG staining. A: Superimposed optical coherence tomography (OCT) thickness map and late-phase indocyanine green angiography (ICGA) images showing a large temporal islet containing multiple, prominently ICG-stained TelCaps, and a smaller inferofoveal islet. B: ICGA images alone show large TelCaps within the temporal islet (arrows), but no detectable ICG-stained TelCaps within the smaller inferofoveal islet (*).

C. OCT B-scan showing clustered TelCaps adjacent to a larger TelCap (arrow). Exudates are present at the margin of the small inferofoveal islet (*), but no TelCaps are observed within this islet

Islets enlargement was observed in 25 eyes (89%). Recurrence of edema within the central 1000 µm ETDRS circle was documented in 20 eyes (71%), and was systematically preceded by the enlargement and coalescence of islets. A peripheral recurrence (outside the central 1000-µm area) was noted in 5 eyes (18%).

In all eyes with recurrent edema, TelCaps remained visible at the barycenter of the islets (Figs. 4 and 5). In contrast, no recurrence occurred in 3 eyes (11%), where all islets and their associated TelCaps had resolved.Fig.

4Macular edema recurrence via enlargement of an islet with clustered TelCaps. A: Baseline: center-involved diabetic macular edema (CI-DME) associated with TelCaps and circinate exudates. B: Partial regression of the edema at month 3 after three monthly intravitreal injections of anti VEGF, revealing an islet with TelCaps located at the barycenter of the retinal thickening.

C: After additional intravitreal therapy (at month 6), the central macular thickness has decreased but there is a residual islet that colocalizes with the TelCaps stained on the late ICGA images. D: Recurrence of edema at month 12, six months after IVI discontinuation; the pre-existing islet centered by TelCaps is larger and its summit co-localizes with the stained spots seen initially on late ICGA imagesFig. 5Fluctuations of macular edema via enlargement of an islet with a TelCap.

A: Single islet visible at month 3 after intravitreal injections, with an isolated TelCap located at the barycenter of the retinal thickening. B: After additional IVI, the thickness of the islet has decreased, but there is a residual thickening that colocalizes with the TelCaps stained on the late ICGA images. C: After IVI discontinuation, the islet centered by the TelCap is larger, with recurrence of edema not reaching the center.

D: After resuming intravitreal therapy, the thickness of the islet has decreased, but there is a residual thickening that colocalizes with the TelCaps

At month 12, the total number of islets had decreased to 39 (p = 0.01), and the mean number of islets per eye was reduced to 1.4 ± 1.1 (p < 0.01). Mean CMT had significantly decreased to 315 ± 103 µm (p < 0.01) compared to baseline. The mean number of TelCaps per eye was 2.3 ± 1.6, with no significant change from baseline (p = 0.55) (Table 2).Table 2Evolution of central macular thickness, number of islets and TelCaps over timeN eyes = 28BaselineMonth 3PMonth 12PCMT µm, mean ± SD453 ± 155273 ± 440.01315 ± 1030.15 *Number of TelCaps per eye, mean ± SD2.6 ± 1.22.1 ± 1.40.022.3 ± 1.60.55 *number of islets/eye mean ± SDN/A1.8 ± 1-1.4 ± 1.1 < 0.01Total number of isletsN/A52-39 < 0.01*Wilcoxon matched pairs test between baseline and month 12

Among the 13 resolved islets, 11 were initially associated with TelCaps, and 2 with microaneurysms. All 13 became undetectable on late-phase ICGA during follow-up. Importantly, the 11 TelCaps-associated islets had all received successful laser therapy targeted at the TelCaps (Fig.

6), supporting the effectiveness of this intervention.Fig. 6Example of resolving islets after IVI and laser. A: Baseline: asymmetric, polylobulated, center-involved diabetic macular edema with three islets associated with clustered TelCaps visible on the late ICGA images.

B: After intravitreal injections and a targeted focal laser session, the islet and TelCaps at the inferior location are no longer visible. The superior islet containing clustered TelCaps is still visible, indicating that the laser was not sufficiently powered to occlude the lesions. Of note, the juxta-foveal cluster of TelCaps is not visible, although laser treatment was not delivered at this site.

C: After a second laser session, the superior islet and associated TelCaps have resolved. Retinal thickness is normal throughout the macular region. One juxta-foveal TelCaps is still visible, but not associated with retinal thickening

Discussion

In the present study, we analyzed the spatial and temporal characteristics of residual retinal thickening following anti VEGF IVI for CI-DME associated with TelCaps. After normalization of CMT, OCT thickness maps frequently revealed perifoveal islets of retinal thickening that colocalized with TelCaps. Theses islets often preceded recurrence of central edema, suggesting they represent sources of fluid leakage.

Previous studies have highlighted that DME may persist or recur despite IVI, with up to 30% of eyes showing residual edema after a loading dose at 6 months [8]. In particular, larger TelCaps have been associated with poorer anatomical response to IVI [23]. We hypothesized that the distribution of TelCaps, as revealed by late-phase ICGA staining, would overlap with the location of post- IVI islets of thickening.

This hypothesis was confirmed by the strong colocalization observed in our cohort, reinforcing the idea that these focal lesions are less responsive to anti-VEGF therapy than diffuse leakage areas [23, 25]. Moreover, our longitudinal data showed that recurrence of central edema was consistently preceded by enlargement and coalescence of perifoveal islets. This temporal sequence supports the hypothesis that these islets are active sites of leakage and play a central role in the re-accumulation of fluid in the macula.

The distinction between focal and diffuse DME has long been recognized as clinically relevant [26]. Focal leakage is typically caused by discrete vascular abnormalities, such as microaneurysms, while diffuse leakage results from widespread blood-retina barrier disruption without identifiable focal sources [27]. Focal DME has been reported to respond less favorably to anti-VEGF treatment than diffuse forms, and may thus require adjunctive focal laser photocoagulation [25].

Notably, although IVI reduce the number and size of microvascular abnormalities, some— particularly the largest lesions—tend to persist [23, 28], highlighting the need to refine the indications and targeting of laser therapy.

Late-phase ICGA has been shown to be a sensitive tool for identifying focal lesions associated with blood-retinal barrier disruption [29]. Importantly, TelCaps often go undetected on fluorescein angiography or OCT angiography [30], underscoring the added value of ICGA in guiding treatment decisions. Our results further support the idea that the late ICG staining reflects the severity of leakage, providing a functional correlate to structural abnormalities.

Histopathological studies suggest that TelCaps are distinct from typical microaneurysms, with features such as thickened walls and reduced endothelial cell density [31, 32]. These abnormalities may explain their poor response to anti-VEGF agents, which aim to retore tight-junction integrity. Interestingly, in our cohort, some small TelCaps resolved completely after IVI alone, consistent with previous findings indicating that closure may occur in smaller-diameter lesions [23, 28].

However, when TelCaps were specifically targeted with ICGA-guided laser, no recurrence of adjacent edema was observed, confirming the therapeutic potential of focal photocoagulation [18, 33, 34]. However, other studies have reported no added benefit of navigated angiography-guided macular laser [35].

This study has several limitations, including the relatively small sample size and the predominance of focal-type DME in the population analyzed. Especially, no control group of eyes with DME but no TelCaps and treated with anti VEGF was analyzed. Consequently, generalizability to broader DME populations, including those with mixed or diffuse patterns, should be approached with caution, and authors may only conclude that there is an association between the presence of TelCaps and the presence of islets, without pre- suppose that a causality link exist between them.

Nonetheless, given the high prevalence of TelCaps —up to 60% in chronic DME cases [15]— our findings remain clinically relevant and may inform personalized treatment strategies. In addition, given the retrospective and uncontrolled design of this ancillary cohort, we could not evaluate the success rate of laser therapy for TelCaps occlusion. Results from the forthcoming full prospective TalaDME cohort are expected to help clarify this issue.

In conclusion, following anti-VEGF therapy for CI-DME associated with TelCaps, OCT thickness maps frequently reveal perifoveal islets of residual retinal thickening that most often colocalize with TelCaps. These islets appear to represent areas of poor local response to IVI and may precede recurrence of central edema. Their identification via OCT mapping offers a simple, non-invasive method to screen for persistent focal leakage, potentially guiding further investigations using ICGA and enabling targeted laser photocoagulation.

Integrating this strategy into routine follow-up could support more personalized treatment approaches and improve long-term outcomes in patients with focal or mixed forms of DME.

Related articles

PubMed

IgD-λ POEMS syndrome presenting with edema, pulmonary hypertension, and renal impairment: A case report.

Rationale: Polyneuropathy, organomegaly, endocrinopathy, monoclonal plasma-cell disorder, and skin changes (POEMS) syndrome is a rare multisystem syndrome associated with clonal plasma-cell disorders. The IgD-λ subtype is uncommon, and a low-level monoclonal protein may escape detection by routine electrophoresis and immunofixation electrophoresis (IFE), delaying diagnosis. Patient concerns: A 38-year-old woman was admitted in April 2025 with intermittent bilateral lower-extremity edema for >...

PubMed

RETRACTION: Tristetraprolin Attenuates Brain Edema in a Rat Model of Cerebral Hemorrhage.

RETRACTION: P. Li, J. Zhang, X. Li, and H. Gao, “Tristetraprolin Attenuates Brain Edema in a Rat Model of Cerebral Hemorrhage,” Brain and Behavior 9, no. 3 (2019): e01187, https://doi.org/10.1002/brb3.1187. The above article, published online on 06 February 2019 in Wiley Online Library (wileyonlinelibrary.com), has been retracted by agreement between the journal Editor‐in‐Chief, Amanda Bischoff‐Grethe; and Wiley Periodicals, LLC. The retraction has been agreed upon following concerns raised by...