OCT angiography findings in type 2 macular telangiectasia: A meta-analysis

Abstract

This meta-analysis evaluates optical coherence tomography angiography (OCTA) results in macular telangiectasia (MacTel) type 2, a rare retinal condition marked by vascular abnormalities. We analyzed 26 studies for OCTA findings in MacTel type 2. We examined parameters like best corrected visual acuity (BCVA), foveal avascular zone (FAZ), and superficial and deep capillary plexus densities. The pooled mean BCVA was 0.37 log of minimum angle of resolution (20/50). FAZ enlargement was observed, with a pooled mean FAZ size of 0.39 mm². Foveal deep vascular density (FDVD) and parafoveal superficial vascular density (PSVD) were significantly reduced (P = 0.021 and P < 0.001, respectively). Parafoveal deep vascular density (PDVD) and foveal superficial vascular density (FSVD) did not decrease significantly (P = 0.067 and 0.114, respectively). MacTel type 2 has unique OCTA characteristics, including reduced FDVD, PSVD, and enlarged FAZ. However, FSVD and PDVD did not statistically decrease in patients. Future studies using more parameters are required to expand our understanding.

Macular telangiectasia (MacTel) type 2 diagnosis is troublesome due to its subtle vascular alterations.[1] Optical coherence tomography angiography (OCTA) has gained attention in MacTel diagnosis.[1] It is hypothesized that MacTel type 2 is associated with changes in OCTA features, such as foveal avascular zone (FAZ) enlargement and vascular density reduction.[2] Therefore, this meta-analysis investigates whether MacTel type 2 is associated with OCTA feature alterations. A systematic search across PubMed, Scopus, and Google Scholar was done based on Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, and studies that met the inclusion criteria (26 studies) were included. Future studies are needed to investigate more of MacTel’s OCTA features.

Methods

This study used the PRISMA protocol as a guideline.

Search strategy and eligibility criteria

The PubMed electronic database, Scopus, and Google Scholar were searched for pertinent publications published by January 31, 2023. The searched Medical Subject Heading terms and text keywords were “MACTEL2 OR macular telangiectasia2” AND “OCT-A OR OCT angiography OR Optical coherence tomography angiography OR Deep capillary plexus OR Superficial capillary plexus OR DCP OR SCP.” Studies were included if their subjects had untreated MacTel type 2, had undergone OCTA, were written in English, and were published before January 31, 2023. Case reports, studies without full-text access, studies with insufficient OCTA data for analysis, studies that included previously treated MacTel patients, and review studies were not included. The authors thoroughly searched the relevant literature and chose studies that fit the criteria after examining their full text. Fig. 1 illustrates the selection process. The searches were done independently by two researchers (VZ and SH).

Figure 1.

Flow diagram of study selection

Data extraction and outcomes

Selected studies underwent data extraction after a full-text review. (1) Year of publication, (2) country, (3) sample size, (4) study design, (5) number of MacTel participants, (6) number of control participants, (7) mean age and gender, (8) rate of hypertension and diabetes among participants, (9) type of MacTel disease, (10) rate of the presence of right-angled vessel, (10) OCTA parameters analyzed, and (11) best corrected visual acuity (BCVA) of the patients and controls were extracted.

Effect size

The mean difference was used as the effect size to compare the OCTA parameters between the control group and the MacTel patients. In addition, the mean of OCTA parameters of the included papers was used in the meta-analysis to obtain a pooled mean estimation.

Quality assessment and risk of bias

Mixed Methods Appraisal Tool (MMAT) was used to evaluate the publications’ quality.[3] If the response to three or more of the five MMAT questions was “No” or “Cannot tell,” the quality was deemed low. The studies covered in this meta-analysis are all of acceptable quality.

Statistical analysis

STATA 16 was used for this study’s analyses. Heterogeneity was assessed using the I² statistic. If I² exceeded 50%, heterogeneity was considered remarkable, and the random-effect model was employed for the meta-analysis. In addition, sensitivity analysis was conducted, and the inferences did not affect the studies.

Results

Fig. 1 illustrates the study selection process. Initially, 449 studies were identified from databases. After removing duplicates and reviewing the abstracts and full texts, 42 reports remained. Of these, 16 reports were excluded due to insufficient data, leaving 26 included papers.

This study used the data from various OCTA machines manufactured by Carl Zeiss Meditec Inc., Optovue, Inc., Topcon, and Heidelberg Engineering. Previous research showed that data of the different OCTA machines are not significantly different and are comparable.[4]

Of the 547 participants in our study, 343 (63%) were female. Four hundred ninety-nine study subjects (307 females, 192 males) were in the MacTel group. Furthermore, the total mean age of the participants was 60.00 (±9.2) years. In the included papers, diabetes was only assessed in 278 patients, with 122 (44%) having it. In addition, the presence of hypertension was assessed in just 152 patients, with 78 (51%) of them having it.

Of the 2094 examined eyes, 1821 were affected by MacTel type 2 disease (86%). One thousand and six eyes with MacTel type 2 were analyzed to determine whether they were proliferative or non-proliferative. Of these, 378 (37%) were classified as proliferative. In addition, 651 MacTel eyes were evaluated for right-angled vessels, and 203 (32%) had this sign.

Functional evaluation

Twenty-six studies were included in the analysis [Fig. 1]. The best and worst reported BCVA were 0.19 log of minimum angle of resolution (LogMAR; 20/32)[5,6] and 0.74 LogMAR (20/110),[7] respectively. The pooled BCVA was 0.37 LogMAR (20/50) (95% confidence interval [CI]: 0.32–0.42) [Fig. 2].

Figure 2.

Mean BCVA. The squares show the mean BCVA for the included patients, and the diamond represents the pooled BCVA (0.37 LogMAR) and corresponding CI (95% CI: 0.32–0.42) BCVA = best corrected visual acuity, CI = confidence interval, LogMAR = log of minimum angle of resolution

OCTA parameters

FAZ area measurements

The mean of FAZ ranged from 0.35 to 0.49 mm². The pooled mean FAZ was 0.39 mm^{2 (}95% CI: 0.34–0.43).[8,9,10]

Central macular thickness

The pooled mean central macular thickness (CMT) was 223.41 µm (95% CI: 163.21–283.61).[7,8,11] The highest and lowest CMT values were 288.4 and 150.28 µm, respectively.

Foveal superficial vascular density

The highest and lowest foveal superficial vascular density (FSVD) values were 13.27% and 51%, respectively.[10,12] The pooled mean FSVD was 25.66% (95% CI: 15.04–36.28) [Fig. 3a]. Fig. 3b compares FSVD of the MacTel patients with that of the control group. The mean FSVD in patients was 2.42% lower, which was statistically insignificant (P = 0.114).

Figure 3.

(a) Mean FSVD. The squares show the amount of FSVD difference between healthy subjects and patients, and the diamond represents the pooled estimation and related 95% CI. (b) Comparison of FSVD between patients and the control group. (c) Mean FDVD. The squares show the amount of difference between healthy subjects and patients, and the diamond illustrates the pooled estimation and related 95% CI. (d) Comparison of FDVD between patients and the control group. The squares show the amount of difference between healthy subjects and patients, and the diamond illustrates the pooled estimation and related 95% CI. CI = confidence interval, FDVD = foveal deep vascular density, FSVD = foveal superficial vascular density

Foveal deep vascular density

The highest and lowest foveal deep vascular density (FDVD) values were 56.4% and 22.58%, respectively.[12,13] The pooled FDVD in patients was 32.16% (95% CI: 22.06%–42.26%) [Fig. 3c]. Fig. 3d compares FDVD between patients and the control group. The FDVD was 5.40% lower in MacTel patients, which was statistically significant (P = 0.021).

Parafoveal superficial vascular density

Parafoveal superficial vascular density (PSVD) ranged from 47.06% to 53.54%.[14,15] The pooled mean PSVD was 49.28% (95% CI: 46.34–52.22).[9,10,14,15] Analysis revealed that the pooled mean of temporal PSVD was 49.46% (95% CI: 46.98–51.95). The pooled mean of nasal PSVD was 50.78% (95% CI: 47.68–53.87). A significant difference was observed in the mean of PSVD between patients and controls (P < 0.001).[14,15,16,17] PSVD was lower in patients by 2.72% [Fig. 4a]. In addition, the mean PSVD was significantly lower in patients in the nasal section by 2.05% (P = 0.006).[14,15,17] The mean of the temporal PSVD was lower by 3.41% in patients, which was statistically significant (P = 0.026).[14,15,17]

Figure 4.

(a) Comparison of parafoveal superficial vascular density between patients and the control group. (b) Comparison of parafoveal deep vascular density between patients and the control group. The squares show the amount of difference between healthy subjects and patients, and the diamond illustrates the pooled estimation and related 95% CI. CI = confidence interval

Parafoveal deep vascular density

The highest and the lowest parafoveal deep vascular density (PDVD) values were 60.38% and 52.8%, respectively.[14,15] The pooled PDVD was 56.46% (95% CI: 53.15–59.77).[9,10,14,15] The mean PDVD values of the nasal and temporal sections were calculated separately. The pooled mean PDVD in the temporal section was 56.19% (95% CI: 53.53–58.86).[9,10,14,15] In addition, the analysis demonstrated that the pooled mean in the nasal section was 57.04 (95% CI: 54.26–59.82).[9,10,14,15] Fig. 4b compares PDVD of the patients and controls. The mean of PDVD was about 0.8% lower in patients, which was not statistically significant (P = 0.067).

Discussion

OCTA is a noninvasive modality that evaluates the retinal and choroidal circulation.[18] It visualizes the two retinal plexuses, which is impossible with conventional methods like Fundus Autofluorescence (FAF).[13] This meta-analysis aims to evaluate the OCTA findings of MacTel type 2 patients.

We identified 26 studies evaluating the OCTA findings in MacTel type 2 patients. Loss of vision is one of the main symptoms of MacTel type 2, due to the gradual degeneration of the neurosensory layer of the retina.[15,19] Based on the pooled analysis, we showed that the mean BCVA of the patients was 0.37 LogMAR (20/50).

Vascular changes are common features in MacTel type 2.[20,21] These eyes are associated with microvascular changes such as FAZ enlargement and distortion that correlate with poorer vision.[9] We found that the pooled mean FAZ size was 0.39 mm².

Superficial and deep vascular plexus alterations, including decreased vascular density,[14] capillary loss, and telangiectatic vessel formation,[22] are identified in MacTel type 2. However, our analysis showed that the pooled mean of FSVD among patients was 25.665%, which is not markedly reduced. The pooled mean of FDVD was 32.16%, which is significantly lower than that of healthy subjects.

We also analyzed the parafoveal vessel density and found that the pooled mean PSVD (49.28%) was dramatically lower than in normal eyes. The pooled mean values of temporal (49.46%) and nasal (50.78%) PSVD were also significantly decreased. However, the pooled PDVD (56.46%) was not markedly reduced. In addition, the pooled mean PDVD of the temporal (56.19%) and nasal (57.04%) sections did not decrease.

Different theories have been proposed on the pathophysiology of these vascular alterations.[5,19] Muller cells are crucial in the angiogenesis of the retina.[19] Dysfunction and death of the Muller cells in MacTel type 2 cause an imbalance of proangiogenic and antiangiogenic factors, leading to vascular alterations.[2,5,19]

MacTel type 2 might also lead to retinal atrophy and decrease in CMT (from the internal limiting membrane to the retinal pigment epithelium).[7] Our analysis showed that the pooled mean CMT was 223.41 µm.

To our knowledge, this paper represents the first meta-analysis regarding the OCTA findings of MacTel type 2. In this meta-analysis, only four OCTA parameters and one functional parameter (BCVA) were meta-analyzed. However, other parameters could not be evaluated due to insufficient data. The disease’s rarity and data’s heterogeneity also made analysis troublesome. Moreover, we could not include studies published after January 2023 due to the long meta-analysis process. Thus, future meta-analyses with the newest articles included are anticipated.

Conclusion

In conclusion, this study shows that MacTel type 2 is associated with changes in the retinal vascular plexuses. Our analysis demonstrates that MacTel type 2 eyes have lower FDVD and PSVD. However, we found no relationship among FSVD, PDVD, and MacTel type 2. Identifying the OCTA features of MacTel type 2 would help clinicians make an early diagnosis and prevent the disease’s progression. Thus, more studies on the other OCTA features of MacTel type 2 are expected.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.