In this study, the authors report a retrospective case series of endoscopic orbital decompression surgery for proptosis caused by TED. The results of endoscopic decompression are comparable to those reported in the literature.

The post-operative results in our study demonstrate significant improvement in proptosis within the early post-operative period (typically measured within 1–3 months after surgery). Importantly, we also assessed the long-term stability of these results over a mean duration of 1.75 years using repeated Hertel measurements.

The authors report an average decompression of 2.35 mm from medial wall-only surgery, 3.18 mm from surgery involving both medial and lateral walls, and 4.7 mm from triple-wall surgery. Those results are comparable to 2.6–6 mm improvement with endoscopic approaches and 3.2–6.5 mm with external approaches reported in the literature today. The average reduction of proptosis in the 3-wall decompression group was 7.6 mm [1, 4,5,6,7,8,9,10,11,12,13,14,15,16,17, 19,20,21,22,23,24,25,26,27,28,29,30].

The authors’ results showed good, statistically significant differences in all three methods (medial alone, inferomedial, and lateral combined with inferomedial). There was a difference in the proptosis improvement (2.35, 3.18, and 4.70 mm, respectively), although this was not statistically significant. This is likely due to the low power of the small sample size.

Results vary across the literature regarding visual acuity, probably due to the heterogeneity of patients and surgical indications [20, 31]. The visual acuity in this study remained stable in all cases. This is due to good visual acuity pre-operation in most patients (0.01 ± 0.21LogMAR) combined with a small study group. It is also known, from recent publications, that visual acuity has a better chance of improving in the active phase of the disease and in patients with compressive optic neuropathy than in those with long standing proptosis [22, 32].

Regarding common complications, endoscopic decompression has been associated with diplopia occurring at a rate of 15–45% [6, 9, 11, 15, 20, 22, 25, 28, 31]. No long-term complications were reported in this series, probably because the study group was small.

In our study, we noted a wider range of proptosis reduction in the medial wall-only decompression group compared to the multi-wall groups. We hypothesise that this variability may be due to patient-specific anatomical factors. For instance, the degree of ethmoid sinus pneumatisation can vary significantly, directly impacting the volume available for medial orbital decompression. Additionally, individual differences in orbital fat volume and distribution, particularly adjacent to the medial wall, may impact the effectiveness of decompression. Tissue compliance and the extent of fibrosis resulting from prior inflammation may also affect the amount of orbital content that can prolapse into the ethmoid space. Further radiological or volumetric studies correlating sinus anatomy and orbital tissue characteristics with surgical outcomes may help clarify these observations.

An interesting finding in our study was the slight deterioration in Hertel measurements over time, with a mean increase of 0.58 mm occurring approximately 1.75 years after surgery. Although this change is small, it raises questions regarding long-term surgical stability. We propose several possible explanations for this finding:

First, subclinical progression of TED may occur even during the quiescent phase, potentially leading to minor changes in orbital tissue volume or stiffness. Second, age-related remodelling of orbital fat and connective tissue over time may alter the positioning of the globe, independent of disease activity. Third, tissue elasticity and mechanical rebound may contribute to a mild re-expansion or settling of orbital contents following decompression. Finally, it is essential to acknowledge that Hertel exophthalmometer, although widely used, is susceptible to inter- and intra-observer variability, particularly in long-term follow-up, where measurements may not be taken under identical conditions.

This study has some limitations. Since clinical data were collected retrospectively, they lack unity and may be insufficiently documented. Secondly, the number of patients in the current study may be insufficient to yield significant results for some questions. Only 56% of the study patients underwent a third Hertel measurement, so the deterioration data was limited. It may be necessary to conduct larger multicenter prospective comparative studies to obtain sufficient data and more conclusive results. In addition, this study represents a retrospective series of endoscopic decompressions and, therefore, cannot be used to compare the outcome with that of external approaches. Smoking is a well-known risk factor for TED severity and poorer outcomes with both medical and surgical treatments. Although we recorded smoking status in our cohort, we did not perform a formal comparison of surgical results between smokers and non-smokers. Additionally, the retrospective nature of our data limited the consistency and detail of smoking history documentation, such as pack years and cessation status. Future studies should explore the impact of smoking on surgical outcomes in TED, ideally with prospective data collection and standardised definitions of smoking exposure. Furthermore, although data on pre-operative medical therapy were available for most patients in our study, the variability in treatment type and timing, along with our limited sample size, prevented a formal analysis of its impact on post-operative outcomes. Future prospective studies are needed to determine whether certain pre-operative therapies influence surgical results, particularly regarding inflammation control, tissue compliance, and healing response.