Ultra-widefield (UWF) imaging provides extended visualisation of the posterior segment beyond the traditional 30–50° field of view. It captures a broad view of the posterior pole and periphery in a single, quick, non-mydriatic acquisition. While its role in diabetic retinopathy and peripheral retinal disorders is well established, its utility in glaucoma is increasingly being recognised.
Ultra-widefield imaging has been found to have 93.9% glaucoma classification accuracy [1] and a very high level of agreement with traditional stereo fundus photos when assessed by a glaucoma specialist [2].
In population-based data, grading of UWF photographs showed high intra- and inter-observer reproducibility for vertical cup-to-disc ratio (VCDR) and good agreement with colour digital stereoscopy (CDS), supporting its use when stereoscopy or slit lamp biomicroscopy are impractical (tele-ophthalmology, screening and high-throughput clinics) [2].
Building on this, large-scale machine learning studies have demonstrated that UWF can support automated glaucomatous optic neuropathy (GON) detection. Using 22,972 UWF images across four institutions, a deep-learning system achieved AUC[SV1.1]s of 0.93–0.99 with sensitivities of ~97–98% and specificities of ~94–98%, performing comparable to experienced ophthalmologists; most false negatives arose in highly myopic discs and most false positives from other fundus lesions, useful caveats for clinical deployment [3].
Comparative work further situates UWF among contemporary imaging options. In a retrospective cross-sectional study (777 eyes), deep-learning classifiers built on UWF fundus photographs and true-colour confocal scanning. They both showed strong diagnostic performance for glaucoma and no significant difference versus optical coherence tomography (OCT) parameter–based methods (retinal nerve fibre layer, ganglion cell complex, etc.) detected. This highlights UWF’s potential as a frontline or adjunctive test in multimodal algorithms. Notably, UWF’s specificity can be lower than some alternatives, underscoring the importance of confirmatory testing and clinical context [4].
While most glaucoma-focused UWF literature centres on optic nerve/peripapillary assessment and screening paradigms, UWF can also document postoperative complications, including hypotony-related serous choroidal detachments that extends into the periphery and may be underestimated on standard views. Case reports have illustrated UWF’s ability to delineate 360° choroidal after filtration surgery, emphasising its value for recognition and follow-up [5].
We report a case of a patient with primary open-angle glaucoma (POAG) who underwent uncomplicated Perserflo® microshunt implantation. They subsequently developed hypotony-related choroidal folds. Ultra-widefield imaging (California, Optos plc, Dunfermline, UK) was used to support the detection, documentation and monitoring the postoperative changes, providing a clear record of their progression and eventual resolution.
Case report
An 89-year-old patient with POAG had well-controlled intraocular pressure (IOP) for years on bimatoprost/timolol fixed combination therapy, but then, due to intolerance, she was switched to latanoprost/timolol two years prior to their presentation. Over the preceding 12 months, progressive memory impairment led to poor adherence, with subsequent visual field deterioration despite IOP measurements in the acceptable range in the clinic – this is also a reminder that glaucoma is much more than IOP level, but a comprehensive disease that needs a holistic approach. Central corneal thickness was below average (RE 516µm, LE 505µm).
Surgical intervention
The Preserflo® microshunt with Mitomycin-C (MMC) was implanted in their left eye. Postoperatively prednisolone acetate (Pred Forte 1%, Allergan, AbbVie Inc) every two hours was administered while awake.
Figure 1: Week 1.
Postoperative course
At week one postoperatively, visual acuity (VA) measured 6/18 with an IOP of 4mmHg. On UWF imaging (Figure 1), glaucomatous optic disc changes were evident, along with macular folds consistent with hypotony maculopathy and a serous choroidal effusion, most pronounced nasally and superiorly with lesser temporal involvement. Pred Forte was reduced to QID and atropine 1% BD was initiated.
Figure 2: Week 4.
By week four, VA had improved to 6/12 with IOP 7mmHg. However, UWF demonstrated progression of the effusion into kissing choroidal (Figure 2), despite the apparent rise in IOP; atropine was stopped and Pred Forte tapered to once daily.
Figure 3: Week 10.
At week 10, VA was 6/9 with IOP 9mmHg. Ultra-widefield imaging at this stage continued to show kissing choroidal, although with minimal signs of fluid regression (Figure 3). Systemic prednisolone (20mg OD for four weeks) was commenced, together with a proton pump inhibitor (lansoprazole) for gastric protection and a calcium carbonate/vitamin D3 supplement for bone support, while topical steroids were discontinued.
Figure 4: Week 11.
By week 11, following a re-referral from the community optometrist, VA was 6/7.5 with IOP reduced again to 5mmHg; UWF imaging demonstrated only minimal improvement (Figure 4), and the management plan remained unchanged.
At week 16, VA was 6/9 with IOP 8mmHg. As significant choroidal detachment persisted, a viscoelastic injection was scheduled. In addition, Pred Forte was restarted once daily.
Figure 5: Week 19.
Three weeks later, UWF showed minimal improvement compared with earlier visits (Figure 5). Viscoelastic injection into the anterior chamber was performed, raising IOP transiently.
Figure 6: Week 20.
One week after the injection, VA was 6/12 with IOP 12mmHg. UWF demonstrated marked improvement with an almost clear posterior pole (Figure 6), and Pred Forte was increased to QID.
Figure 7: Week 25.
Five weeks later, VA remained 6/12 with IOP 8mmHg, and UWF showed near-total resolution of the choroidals (Figure 7), allowing Pred Forte to be stopped.
Finally, at week 64, VA had improved to 6/6 with IOP 12mmHg, and UWF confirmed that the choroidal effusions had fully resolved with restoration of normal posterior pole architecture.
Discussion
Hypotony and choroidal effusion are well-recognised complications following glaucoma filtration surgery, particularly when adjunctive MMC is used. Mitomycin-C enhances filtration success by reducing subconjunctival fibrosis, but it may predispose to over-filtration and prolonged hypotony. This can result in hypotony maculopathy, characterised by macular folds and reduced visual acuity, and in more severe cases, extensive choroidal detachments. While clinical diagnosis of hypotony and choroidal effusion is often made on slit lamp examination and fundoscopy, the full extent of pathology can sometimes be underestimated, particularly when peripheral involvement is present.
In recent years, multimodal imaging has been increasingly employed to support the detection of and document these postoperative complications. Standard colour fundus photography and OCT can demonstrate macular folds and serous detachments involving the posterior pole. However, they are limited by their narrow field of view and may fail to capture the widespread nature of choroidal effusions, especially when they extend into the mid-periphery or periphery.
In our case, UWF imaging contributed to diagnosis, follow-up and patient counselling throughout the postoperative period. Imaging was applied at each postoperative stage to document the evolution of hypotony-related choroidal changes. It demonstrated the progression to kissing choroidal, the persistence of detachments despite systemic and topical therapy, and the gradual resolution following viscoelastic injection. This sequential documentation correlated with the patient’s clinical course, including improvements in IOP and visual function.
More broadly, the role of UWF in glaucoma has so far centred on optic nerve and peripapillary evaluation, screening programmes and as an adjunct in multimodal imaging. This case highlights an additional application, its usefulness in recognising and monitoring postoperative complications that may be underestimated with standard techniques.
References
1. Haleem MS, Han L, Hemert J van, et al. Regional image features model for automatic classification between normal and glaucoma in fundus and scanning laser ophthalmoscopy (SLO) images. J Med Syst 2016;40(6):132.
2. Quinn NB, Azuara-Blanco A, Graham K, et al. Can ultra-wide field retinal imaging replace colour digital stereoscopy for glaucoma detection? Ophthalmic Epidemiol 2018;25(1):63–9.
3. Li Z, Guo C, Lin D, et al. Deep learning for automated glaucomatous optic neuropathy detection from ultra-widefield fundus images. BJO 2021;105(11):1548–54.
4. Shin Y, Cho H, Shin YU, et al. Comparison between deep-learning-based ultra-wide-field fundus imaging and true-colour confocal scanning for diagnosing glaucoma. J Clin Med 2022;11(11):3168.
5. Senthilkumar V, Mishra C. Ultra-widefield image of choroidal detachment after combined glaucoma filtration surgery. Indian J Ophthalmol 2020;68(8):1669.
Declaration of competing interests: None declared.
Acknowledgments: The authors would like to thank the ophthalmic imaging team at York Hospital for their support in capturing the ultra-widefield images used in this case report.
