Introduction

Retinal detachment is a complex and progressive emergency condition that, if left untreated, can result in vision loss. In terms of the mechanism of development of retinal detachments, there are several types, the most common being rhegmatogenous detachment. In this case, fluid from the vitreous body enters the subretinal space through retinal tears, causing the neurosensory retina to separate from the retinal pigment epithelium.

The success of retinal detachment surgery is typically assessed by the reattachment of the retina and postoperative best corrected visual acuity (BCVA). However, despite favourable anatomical outcomes and high BCVA, patients often report experiencing distorted vision following surgery.

Metamorphopsia (MM) is a frequent postoperative symptom following pars plana vitrectomy and is often accompanied by aniseikonia [1]. This condition causes distorted vision, with straight lines appearing curved or disjointed to the patient. While the Amsler grid is traditionally used for the qualitative assessment of MM, the severity of the distortion can be measured quantitatively and more precisely using M-CHARTS [2].

To date, there is no universally accepted explanation for MM following vitrectomy, including successful retinal detachment surgery, thus the pathogenesis of MM is still not fully understood.

In our study, we aimed to determine the prevalence of MM and to quantify its occurrence in patients undergoing vitrectomy for rhegmatogenous retinal detachment. In addition, we aimed to identify correlations between MM and perioperative data from patients' records, as well as OCT findings. By analysing these associations, our goal was to identify potential risk factors for postoperative MM and to provide recommendations for optimising surgical techniques.

Materials and methods

In our retrospective case series study, we enrolled patients who had undergone vitrectomy for rhegmatogenous retinal detachment, performed by two vitreoretinal surgeons at the Department of Ophthalmology of the University of Szeged.

The exclusion criteria for our study were as follows:

• other preoperative macular diseases (such as a macular hole, epiretinal membrane, macular degeneration, and diabetic maculopathy);

• a history of previous macular/retinal surgery;

• a period of more than 6 months between the surgery and the examination;

• an uncorrectable irregularity or opacity of the refractive media; and

• incomplete patient documentation.

At follow-up visits, a Snellen visual acuity test and MM assessment using M-CHARTS (Inami & Co., Ltd., Tokyo, Japan) were routinely performed, and optical coherence tomography (OCT) images were captured (ATE Topcon 800, Tokyo, Japan). Retrospectively, B-scans were analysed to assess the regularity of the foveal depression, central subfield thickness (CSFT), the existence of epiretinal membrane (ERM) and subretinal fluid (SRF), and the integrity of the ellipsoid zone (EZ)/external limiting membrane (ELM). We supplemented our findings with the perioperative and follow-up data recorded in the patients' electronic medical records.

A statistical analysis was conducted using Student's t-test to assess the association between MM and the collected perioperative data. The statistical analysis was performed in Microsoft Excel (manufactured by Microsoft Corporation).

Results

In our study, we analysed data from 23 patients who met the inclusion criteria. The patients' preoperative data are shown in Table 1, and their postoperative data in Table 2. The average interval of 13.3 days between the onset of symptoms and the seeking of medical help is relatively long, which may have negatively affected surgical outcomes in the case of our patients.

The follow-up period ranged widely, between 1 and 6 months, because several patients attended the regular follow-up sessions at either 3–4 weeks or 8–10 weeks after surgery not at our department but at other district hospitals, only attending our clinic at a later date.

The patients' BCVA was assessed preoperatively, with an average measurement of 0.4. Following surgery, there was a notable improvement, with the mean BCVA increasing to 0.6 (Fig. 1). This improvement was statistically significant (P = 0.003).

View Full Size

The mean BCVA was 0.4 preoperatively and 0.6 postoperatively

Citation: Developments in Health Sciences 2024; 10.1556/2066.2024.00064

• Download Figure
• Download figure as PowerPoint slide

Nineteen of the 23 patients in our study (82.6%) reported experiencing distorted vision after pars plana vitrectomy (Table 2). Figure 2 illustrates how M-CHARTS are used as an assessment tool.

View Full Size

In this case, the patient's VM score is 0.4 and the HM score is 0.5 [Nowomiejska K, Oleszczuk A, Brzozowska A, Grzybowski A, Ksiazek K, Maciejewski R, Ksiazek P, Juenemann A, Rejdak R. M-charts as a tool for quantifying metamorphopsia in age-related macular degeneration treated with the bevacizumab injections. BMC Ophthalmol. 2013; 13:13. doi: 10.1186/1471-2415-13-13].

Citation: Developments in Health Sciences 2024; 10.1556/2066.2024.00064

• Download Figure
• Download figure as PowerPoint slide

We assessed both vertical MM (VM) and horizontal MM (HM), the mean VM being recorded as 0.4 and the mean HM as 0.6. However, the difference between the two was not statistically significant (P = 0.141). In the absence of a significant difference between VM and HM, we computed the average scores for both VM and HM for each patient in order to identify correlations between the MM scores and the other variables considered in our analysis.

It is well established that MM is more prevalent and severe in cases of retinal detachment involving the macula [3, 4], a trend that our study also confirmed. Among our patients, 52.2% presented with macula-on retinal detachment, while 47.8% presented with macula-off retinal detachment. The mean MM score for the cases of macula-on detachment was 0.3, compared to 0.6 for macula-off cases. However, the difference between the two groups was not statistically significant (P = 0.136).

We observed a negative correlation between preoperative BCVA and postoperative MM, with a Pearson's r of −0.27. This indicates that patients with poorer preoperative vision tended to experience more pronounced postoperative MM.

Following pars plana vitrectomy, OCT images are routinely captured during follow-up appointments to assess and analyse retinal structure. As part of our study, we evaluated the CSFT of the vitrectomised eyes, while the non-vitrectomised eyes served as controls. Our aim was to determine whether these factors influenced the development of MM, as observed in conditions such as retinal vein occlusion and macular oedema, where the severity of MM correlates with an increase in CSFT and the presence of inner retinal cysts [5]. The mean CSFT of the vitrectomised eyes in our study was 267.4 μm, compared to 253.8 μm in the control eyes. However, the difference between the two groups was not significant (P = 0.348), and no correlation was observed between MM and CSFT.

We further analysed the regularity, symmetry, and shape of the fovea in the 23 patients included in our study. Following pars plana vitrectomy, 11 patients (47.8%) exhibited a regular foveal depression, while 12 patients (52.2%) had an asymmetric or flat fovea. We compared the MM scores of the patients between these two groups, with both groups showing an MM score of 0.4 (SEM: ±0.1). The results indicated no significant difference between the two groups (P = 0.992), suggesting that the regularity of the foveal depression is not associated with the severity of MM.

In our study, only 2 of the 23 patients had SRF (8.7%). However, these patients exhibited the most prominent distorted vision, with MM scores averaging 1 and 0.9. Only 5 patients (21.7%) exhibited no abnormalities in the EZ, while 17 patients (73.9%) showed some degree of disintegration, discontinuity, or hyperreflectivity. In one case, extensive macular oedema prevented the assessment of the EZ. Among the patients with no EZ abnormalities, the mean MM was 0.2 (SEM: ±0.1), while among those with abnormal EZ it was 0.5 (SEM: ±0.1) (see Fig. 3). Although the difference between the two groups was not statistically significant (P = 0.051), the observed trend suggests that a significant difference may emerge with an increase in sample size.

View Full Size

The mean MM score for regular EZ is 0.2, while the mean MM score for disrupted or irregular EZ is 0.5

Citation: Developments in Health Sciences 2024; 10.1556/2066.2024.00064

• Download Figure
• Download figure as PowerPoint slide

With respect to the OCT images, 5 patients (21.7%) exhibited no detectable disintegration or hyperreflectivity in the external limiting membrane (ELM), while 17 patients (73.9%) did show such abnormalities. In the case of 1 patient, the ELM was not detectable due to macular oedema. The patients with a regular ELM had an average MM score of 0.2 (SEM: ±0.1), whereas those with a disrupted or hyperreflective ELM had an average MM score of 0.5 (SEM: ±0.1) (Fig. 4). Although the difference between the two groups was not statistically significant (P = 0.072), there was a noticeable trend indicating a potential association between disrupted ELM and MM. This suggests that significance may be reached with an increase in sample size.

View Full Size

The mean MM score for regular ELM is 0.2, while the mean MM score for disrupted or hyperreflective ELM is 0.5

Citation: Developments in Health Sciences 2024; 10.1556/2066.2024.00064

• Download Figure
• Download figure as PowerPoint slide

Discussion

Although we did not find a significant difference between VM and HM, and despite the limited data on whether VM or HM is more prominent after rhegmatogenous retinal detachment surgery, a 2017 study on idiopathic macular hole surgery found VM to be significantly more prominent than HM [6].

We observed a negative correlation between preoperative BCVA and postoperative MM. Several factors may contribute to this finding. Firstly, poorer preoperative BCVA suggests a higher likelihood of macular involvement in the retinal detachment, as discussed above. As noted, macula-off retinal detachment typically results in more prominent postoperative MM. Additionally, larger retinal detachments are associated with greater loss of visual acuity [3], which in turn may lead to more pronounced MM.

A study conducted in 2015 involving 45 patients with primary macular detachment found that 88.6% of patients exhibited MM, as determined by an Amsler grid test, an average of 12 months after vitrectomy. Patients with MM had significantly worse postoperative BCVA compared to those without MM (P = 0.02) [4].

The presence of SRF in the macula is widely recognised as a significant factor contributing to the development of MM following successful retinal detachment surgery, as reported in various studies [4, 7]. It is common for a certain amount of SRF to remain between the retinal tear and the posterior pole after vitrectomy, despite the performance of fluid–air exchange, and it can persist postoperatively, especially in the case of damaged or atrophic retinal pigment epithelium [7].

A study carried out in 2016 aimed to assess the prevalence of MM and predictive factors following successful rhegmatogenous retinal detachment surgery. The study examined 380 eyes and measured MM using an Amsler grid. Among those eyes in which SRF was observed on OCT scans (26 eyes, 6.84%), 20 were found to have MM. Persistent postoperative SRF was identified as an independent predictor of MM (P = 0.01) [7].

Previous studies have demonstrated that the presence of persistent SRF in the macula for up to 6 months is associated with poor functional recovery following macula-off retinal detachment surgery. Prolonged separation of the macular photoreceptors and the retinal pigment epithelium due to persistent SRF may compromise visual recovery after pars plana vitrectomy [7].

In addition to SRF, disintegration of the EZ is a crucial factor in the development of MM [4]. Patients who experience MM are more likely to exhibit EZ disintegration compared to those without MM [7]. The disruption of photoreceptors in the macula is believed to play a pivotal role in MM development following retinal detachment, as detachment-induced microcirculatory disturbances can lead to distorted vision [4, 8, 9].

Previous studies using spectral domain OCT have shown that epiretinal membranes can cause EZ disruption in some cases, with persistent MM observed even after the removal of the epiretinal membranes [10].

Zhou et al. identified a disrupted external limiting membrane (ELM) as an independent predictor of MM, indicating its ability to predict the presence of MM regardless of other factors and variables [7]. The presence of MM was assessed using the Amsler grid during follow-up visits, with a mean follow-up interval of 35.11 ± 19.6 months after surgery. The overall prevalence of MM was 46.58% (177/380), which increased to 56.69% (144/254) in cases involving macula-off detachment. The authors identified younger age, preoperative macula-off status, disrupted ELM, and postoperative SRF as independent predictors of postoperative MM. However, no statistical significance was found between gender, surgical approach, type of tamponade, and primary or recurrent retinal detachment [7]. In our study, no correlation was observed between age and MM, with a Pearson's r value of 0.045.

Conclusions

The pathomechanism underlying MM following successful rhegmatogenous retinal detachment surgery remains elusive. However, based on available data, persistent SRF, a disrupted ELM, and EZ disintegration are factors that are likely to be associated with postoperative MM after pars plana vitrectomy. Our study may be the first not only to investigate the presence of MM following rhegmatogenous retinal detachment surgery but also to assess it quantitatively using M-CHARTS. Statistical analysis using Student's t-test revealed no significant correlation between MM and the factors mentioned above. However, a notable trend was observed between MM and disrupted EZ and ELM (P = 0.051 and 0.072 respectively). Given that the results of the t-test are strongly influenced by sample size, it is highly probable that increasing the number of cases would reinforce this observed pattern. Consequently, disrupted EZ and ELM may emerge as significant contributors to the prevalence and severity of MM. In our study, we observed a relatively high incidence (82.6%) of MM, comparable to that reported in macula-off cases [4]. This can partly be attributed to the fact that nearly half the patients in the study presented with macula-off retinal detachment, a condition known to increase both the severity and prevalence of MM [3, 4]. At the same time, it makes the mechanisms behind MM development in macula-on cases even more obscure. Furthermore, the greater extent of retinal detachment, averaged across two quadrants, probably also contributed to the elevated prevalence of MM. This association is supported by previous research indicating that the larger the retinal detachment, the higher the likelihood of developing MM [3]. Finally, the likelihood of the higher sensitivity of M-CHARTS in detecting MM may also have contributed to the unfavourable result in our study. We observed a negative correlation between preoperative visual acuity and postoperative MM, indicating that poorer preoperative visual acuity was associated with higher postoperative MM. This underscores the importance of advising patients to seek medical attention promptly upon noticing any loss of vision. Early intervention increases the likelihood of successful surgery and can reduce the severity of postoperative symptoms.

Authors' contribution

All the authors contribute to this article.

R.B.: Writing – Original Draft, Writing - Review & Editing, Investigation, Visualization.

T.Á-L.: Formal analysis, Writing - Review & Editing.

A.H-K.: Conceptualization, Methodology, Writing - Review & Editing, Supervision.

Ethical approval

Our study was conducted in accordance with the Declaration of Helsinki and in conformity with all applicable local and international standards. Ethical approval was granted by the Committee of Science and Research Ethics, approval number BM/4950/2024.

Conflicts of interest/funding

The authors declare no conflicts of interest. No financial support was received for this study.