The current contribution of SiO endotamponade following PPV to the development of glaucoma long-term is limited. The findings in this study suggest long-term silicone oil situ may alter the morphology of the AC.

Prior to the use of OCT for anterior segment imaging, ultrasound biomicroscopy (UBM) was often the imaging modality of choice for collecting high-definition images of the anterior segment [12]. UBM allows for the generation of quality images of intraocular structures through the use of 50–100 MHz high frequency soundwaves [13]. In 2006, Margio et al. [14] compared the AC morphometric parameters of eyes undergoing PPV pre-operatively and eight weeks post-operatively, finding no significant differences in parameters including ACD and angle opening distance. Comparatively, another study comparing pre-operative parameters to the immediate post-operative period noted ACD was only reduced in patients having undergone gas tamponade and there was no correlation between ACD measurements and IOP elevation [15].

Li et al. [16] compared ACD’s of PPV patients using A-scan and found no mean ACD difference in eyes that underwent PPV for VH in the three month post-operative period. Interestingly, they noted eyes with ERMs had a deeper ACD pre-operatively compared to their fellow eye, this phenomenon was observed to persist one week post procedure but resolved at subsequent follow up. This suggested ERM induced ACD deepening was reversible with successful vitrectomy.

Another study found no difference in AC volume and iridocorneal angle following either PPV alone or PPV with SiO injection [17]. At the one week follow up they did however note an increase in the ACD of the SiO injected group and a decrease in the ACD of the PPV without SiO group. Furthermore, the SiO group had a reduction in CCTs during the one month post-operative follow up period. Ünsal et al.’s [18] UBM evaluation of found PPV patients with SiO had reductions in trabecular meshwork-ciliary process distance and iris-ciliary process distance. Similar results were also observed for their gas tamponade cohort, with an additional statistically significant reduction in ACD. In this study the authors concluded gas tamponade patients had a greater alteration to AC morphology, suggesting this attributes to the greater incidence of IOP rise in this specific sup-group.

Ghomi et al. [19] found although no significant change in AC parameters following undergoing vitrectomy, there was an increase in anterior–posterior lens size. They attributed this to an increase in lens opacity secondary to PPV. Similarly, a study of phakic patients with mac-off retinal detachments had an increase in axial length, ACD and lens thickness [20]. A more recent study reported patients undergoing PPV with indented vitreous base shaving were noted to have an increase in ACD and irideocorneal angle over a 3 month post-operative period [21].

Literature using AS-OCT for the assessment of anterior morphology is far more limited. One study analysed 245 eyes prospectively noting a reduction in both ACD in the 12-month post-operative period for both patients undergoing PPV and SB. Further to this, SB patients were noted to also have a mean increase in axial length [22]. In contrast, Khodabande et al. [23] found no difference in ACD, ACd, irideocorneal angle size in patients who underwent PPV without tamponade. Worth noting, axial length changes have been noted post vitrectomy in patients with pre-operative hypotony or extreme myopia, however it remains unclear whether this translates to changes within the anterior segment [24].

The influence of PPV and long-term SiO is not entirely established, with variable morphological changes reported in literature thus far. Therefore, the precise aetiologic mechanism by which it contributes to the long-term development of glaucoma remains a topical issue. In 1988, Burk et al. [25] suggested intravitreal SiO may play a role in the development of late onset glaucoma through its presence in the AC angles. They undertook histopathology testing of six enucleated eyes, noting the presence of: fibrovascular membranes, inflammatory cells or cellular debris in both AC angles and the trabecular meshwork. Gao et al. [26] also suggested a link between SiO in the AC and IOP elevation.

Following this, one study analysed the angles of patients with SiO in situ using gonioscopy, noting at least 50% of eyes had emulsified oil present in angles regardless of SiO density injected [27]. This study found oil was predominately noted in the 1–4 O’clock areas of the superior trabecular meshwork. Subsequent to this, an evaluation of trabecular meshworks via electron microscopy detected fibrosis/thickening of trabecular spaces and narrowing of intertrabecular spaces in SiO patients [28]. Microscopic analysis also demonstrated the presence of cellular debris within the meshwork, however there were no SiO bubbles or macrophages suggestive of inflammation noted. In contrast, ultrasound biomicroscopy has demonstrated the presence of silicone oil bubbles in patients’ trabecular meshworks, peripheral synechiae were also noted [29].

In context of the above, one of the main hypotheses at present suggests SiO is thought to aggregate at the trabecular meshwork and form a physical barrier for aqueous outflow [30]. This is thought to occur over time as a result of SiO bubbles (being injected for tamponade following vitrectomy) migrating into the AC. Much more recently, a case report identified the presence of SiO in Schlemm’s canal on AS-OCT, again suggesting the possibility of SiO being involved with the development of glaucoma [31].

There is also a possibility the presence of SiO in the AC may elicit an inflammatory response, which in turn alters the structure and impedes the function of aqueous drainage pathways. Liu et al. [32] study comparing patients with and without glaucoma following SiO injection demonstrated glaucoma group had significantly higher levels levels of IL-17, IL-6 and TNF-α in their aqueous humour. Surprisingly, although the glaucoma group had a greater proportion of patients with SiO in the AC, slit lamp examination demonstrated the control group had a greater proportion of patients with emulsion overall (both results were not statistically significant). Semeraro et al. [33] compared the inflammatory activity of patients receiving standard SiO to heavy oil, they reported higher prostaglandin E2 and interleukin-1α in the heavy SiO group.

A limitation of this study is the sample size of long-term SiO patients, there were only 13 operated and 8 control eyes within this study. This is due to long-term silicone oil rarely being left in situ for PPV patients and thus a very limited proportion of clinic patients being eligible to participate in the study. In future, to validate the results of this study, a larger multicentre trial is likely required. Furthermore, future studies would benefit from ensuring all participants having pre-operative AS-OCT imaging to allow for a pre-operative comparison, although it is often difficult to predict which patients may require long-term oil indefinitely.

Another limitation of this study is the image collecting limitations of the AS-OCT. This imaging modality can be operator dependent and often given subjects had low vision-there were challenges with fixation on specified targets whilst collecting images. Furthermore, whilst the software within the Zeiss HD-OCT software is accurate, it still requires human intervention to set parameters for some measurements, thereby introducing a possibility of human error. To mitigate this, all measurements were undertaken by a single investigator.