Introduction

Macular telangiectasia type 2 (MacTel type 2) is a rare, bilateral retinal disorder characterized by progressive capillary telangiectasia and neurodegenerative changes within the macula, affecting 0.0045 to 0.1% of the population1,2. Pathologically, the disease involves Müller cell atrophy, retinal pigment epithelial dysfunction, and gradual photoreceptor degeneration, which together lead to central vision loss3,4,5. Clinically, MacTel type 2 has traditionally been staged using the Gass & Blodi classification system, which categorizes disease progression from subtle parafoveal greying (Stage 1) to advanced subretinal neovascularization (SRNV) (Stage 5)6,7 (Table 1). The presence of SRNV, further defined by Yannuzzi et al. as the proliferative stage2, marks a critical transition associated with poorer visual prognosis.

Table 1 Gass&Blodi classification of MacTel type 2*.
Full size table

Although no curative therapy currently exists, several treatment approaches have been explored to slow disease progression or manage complications. A wide variety of interventions, including intravitreal anti-vascular endothelial growth factor (anti-VEGF) injection, focal laser photocoagulation, photodynamic therapy (PDT), and indocyanine-green-mediated photothrombosis have been reported to yield benefit in selected patients8,9,10,11. More recently, neuroprotective strategies, particularly ciliary neurotrophic factor (CNTF), have shown promising results in preserving photoreceptor structure, with phase 3 clinical trials confirming a significant reduction in ellipsoid zone (EZ) loss12.

Despite increasing recognition of MacTel type 2, several unresolved issues remain to be addressed. (1) Early diagnosis remains challenging, as patients often present with subtle or mild symptoms. In its initial stages, MacTel type 2 frequently shares overlapping features with age-related macular degeneration, such as pigmentary changes and parafoveal telangiectasia, making clinical differentiation difficult13. This highlights the need for more systematic characterization of multimodal imaging findings to improve early detection accuracy. (2) Treatment strategies remain heterogeneous and lack consensus. While anti-VEGF therapy has proven effective in proliferative MacTel type 2 by suppressing neovascular activity and improving vision, its benefit in non-proliferative disease is still debated14,15. Reported rates of visual improvement after intravitreal anti-VEGF injection vary widely, from 12.5% to 80%16,17. (3) The underlying pathogenesis of MacTel type 2 remains elusive18. Although systemic factors such as hypertension, diabetes mellitus, and female sex have been frequently reported19,20, their causal associations have yet to be confirmed in large, well-designed multicenter studies. (4) While severe vision loss is relatively uncommon, complications including lamellar or full-thickness macular hole (MH) formation, foveal detachment, and SRNV can lead to significant visual decline and reduced quality of life. However, the true incidence and prognostic impact of these complications remain uncertain due to limited and inconsistent data21,22,23.

Given the ongoing controversies and clinical challenges surrounding MacTel type 2, a comprehensive and quantitative analysis of all the currently available evidence is urgently needed. To date, only one meta-analysis has specifically addressed MacTel type 2, focusing on optical coherence tomography angiography (OCTA) findings24. Yet, a comprehensive, large-scale synthesis encompassing its clinical characteristics, imaging profiles, and therapeutic responses remains unavailable. To fill these gaps, we conducted the present meta-analysis to give a full-scale description of MacTel type 2 and to evaluate the efficacy of various interventions, with the aim of providing an evidence-based reference for ophthalmologists in clinical practice.

Methods

This study was carried out in compliance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) statement25, and the completed PRISMA checklist is provided in Supplementary Table 1. The study design aligns with previous studies on rare conditions26,27,28,29,30, combining all available epidemiological and treatment evidence.

Search strategy and study identification

A systematic literature search was conducted in PubMed, Embase, and Ovid from inception to March 2024 for relevant studies. The search was restricted to English-language publications involving human subjects. The following keywords were used: “MacTel”, “macular telangiectasia”, and “macular telangiectasis”. Because applying a restriction to “type 2” may have led to the omission of eligible studies, no subtype limitation was imposed during the search. A comprehensive search strategy was first developed for PubMed and then adapted for the other databases. The specific search criteria employed in PubMed included the following query: ((Macular Telangiectasi*) OR (Mac Tel[Title/Abstract])) AND (humans[Filter]) AND (english[Filter]). To ensure completeness, the reference lists of relevant reviews and included studies were additionally screened manually for potential eligible records.

Inclusion criteria and exclusion criteria

Inclusion criteria：(1) clinical studies (interventional or observational), including prospective cohorts, retrospective cohorts, case-control studies, and case series involving patients with MacTel type 2; (2) full-text available. Exclusion criteria：(1) studies with a sample size less than 3; (2) literature reviews, editorial, conference abstract, or technical notes; (3) studies on animal or cadaver subjects; (4) non-English or duplicated publications.

Data extraction and assessment of methodological quality

EndNote 21 software was used for reference management. After removing duplicates, the title and abstract of every search result were examined by two investigators independently (X.-y.Z. and M.-q.H.). If relevant, the full texts would be read for data collection. The following data were extracted: the first author, publication year, study design, cohort size, baseline characteristics of patients, baseline best-corrected visual acuity (BCVA), follow-up period, prevalence of clinical and imaging features at presentation (examined by slit lamp, color fundus photograph (CFP), optical coherence tomography (OCT), fluorescein fundus angiography (FFA), and optical coherence tomography angiography (OCTA)). Information on intervention strategies, clinical outcomes, and disease progression during follow-up was also collected. The primary outcome was defined as the proportion of eyes demonstrating improvement in VA. Secondary outcomes included the proportion of eyes with reduction in central macular thickness (CMT), progression to the proliferative stage, MH closure, and regression of neovascularization. For multiple studies derived from the same patient cohort, data were extracted only once to avoid duplication. If the concerned data was unavailable, we would contact the corresponding author for more information. A third reviewer (Y.-x.C.) intervened when encountering any discrepancies in eligibility and data collection. The methodological quality of non-randomized studies was assessed using the Newcastle–Ottawa Scale (NOS), with a total score greater than 5 indicating high quality31. For randomized controlled trials, the Cochrane Risk of Bias 2 (RoB 2) tool was applied32. Two investigators (X.-y.Z. and M.-q.H.) independently performed the assessments, with disagreements resolved by discussion or consultation with a third reviewer (Y.-x.C.). The certainty of evidence was evaluated using the GRADE (Grading of Recommendations, Assessment, Development, and Evaluation) framework33. Summaries of findings and evidence profiles were generated with the GRADEpro Guideline Development Tool (https://methods.cochrane.org/gradeing/gradepro-gdt)34.

Statistical methods

Statistical analyses were conducted using R version 3.6.3 (R Foundation for Statistical Computing, Vienna, Austria). Freeman–Tukey variant of arcsine square transformation was performed to determine pooling proportions with a 95% confidence interval (CI) and to build forest plots. The heterogeneity was assessed by chi-squared test and I2 statistics. A fixed-effect model would be adopted if the heterogeneity was low (p > 0.1, I2 < 50%). Otherwise, sensitivity analysis and subgroup analysis were conducted to identify the source of significant heterogeneity (p < 0.1, I2 > 50%). When the heterogeneity could not be eliminated, random-effect model would be used. Subgroup analysis was conducted between MacTel type 2 patients with proliferative stage, non-proliferative stage, and MH complication in order to assess the clinical characteristics and prognosis of different treatments.

P < 0.05 was considered as statistically significant. Publication bias was estimated by funnel plots of the Egger test.

Results

Study characteristics and quality assessment

A total of 1783 potentially relevant records were initially identified. After removing the duplicates and screening of titles and abstracts, 297 full-text articles were assessed for eligibility. Among these, 205 studies comprising 11,045 patients were eventually included in this meta-analysis. The detailed screening procedure is depicted in Fig. 1, and the main characteristics of the included studies are presented in Supplementary Table 2. The quality of our outcomes was evaluated using the GRADE approach and summarized in Supplementary Table 3.

Fig. 1
figure 1

PRISMA flowchart of selected studies.

Full size image

Baseline characteristics

MacTel type 2 was female-predominant, with a pooled male ratio of 38% (95% CI [37 ~ 40%]). Most patients were Caucasian (68%, 95% CI [47 ~ 87%]), followed by Asian individuals (31%, 95% CI [13 ~ 51%]). The majority of cases were diagnosed between 50 and 59 years (36%, 95% CI [32 ~ 39%]) and 60–69 years of age (39%, 95% CI [36 ~ 41%]). Systemic comorbidities were common among Mactel type 2 patients, with hypertension was present in 41% (95% CI [36 ~ 45%]), diabetes mellitus in 35% (95% CI [30 ~ 40%]), and dyslipidemia in 33% (95% CI [21 ~ 47%]).Regarding visual status, over half of the patients (55%, 95% CI [49 ~ 62%]) had a BCVA of 20/40 or better, whereas 19% (95% CI [15 ~ 24%]) exhibited severe visual impairment (BCVA ≤ 20/100).The detailed results of the baseline characteristics of MacTel type 2 patients are summarized in Table 2.

Table 2 The pooling results of the population distribution and baseline characteristics of MacTel type 2 patients.
Full size table

Symptoms and signs

Across the included studies, mild visual impairment and metamorphopsia were reported in 54% (95% CI [40% ~ 68%]) and 40% (95% CI [12% ~ 73%]) of patients with MacTel type 2, respectively. Based on the Gass & Blodi classification, Stage 3 was the most frequently observed stage(33%, 95% CI [25 ~ 41%]) followed by Stage 2 (26%, 95% CI [20 ~ 33%]), Stage 4 (23%, 95% CI [18 ~ 30%]), Stage 1 (10%, 95% CI [6 ~ 16%]), and Stage 5 (7%, 95% CI [5 ~ 10%]). Proliferative disease was identified in 12% (95% CI [10 ~ 15%]) of patients. Regarding multimodal imaging features, the most prevalent findings on CFP, FFA, OCT, and OCTA were loss of retinal transparency (76%, 95% CI [65% ~ 85%]), retinal vascular leakage (90%, 95% CI [78% ~ 99%]), EZ loss (70%, 95% CI [57% ~ 81%]), and retinal-choroidal anastomosis (RCA; 36%, 95% CI [11% ~ 64%]), respectively. A comprehensive summary of pooled symptoms, disease staging, and multimodal imaging characteristics is provided in Table 3. Subgroup analyses comparing proliferative and non-proliferative stages are presented in Supplementary Table 4.

Table 3 The pooling results of the disease staging and multimodal imaging characteristics of MacTel type 2 patients.
Full size table

Treatment and prognosis

MacTel type 2 patients were further divided into three groups for prognosis analysis, including proliferative stage, non-proliferative stage, and patients complicated with full-thickness MH. The pooling results are summarized in Table 4.

Table 4 The pooling results of the treatment strategy and efficacy for MacTel type 2 patients with proliferative stage, non-proliferative stage and macular hole complication.
Full size table

Proliferative stage

The pooled results indicated that visual acuity (VA) was improved in 51% (95% CI [36 ~ 66%]) of the proliferative stage MacTel type 2 patients and stable in 30% (95% CI [15 ~ 44%]). A decrease in CMT was noted in 85% (95% CI [78 ~ 92%]) of the patients, and clinical regression of neovascular complex was reported in 62% (95% CI [21 ~ 96%]) of the patients. Subgroup analysis was performed regarding different treatment modalities, including anti-VEGF injection and observation. Anti-VEGF therapy demonstrated superior efficacy, with 59% (95% CI [43 ~ 75%]) of patients achieving VA improvement and 85% (95% CI [78 ~ 92%]) showing reduced CMT (Fig. 2). By contrast, observation alone resulted in only 7% (95% CI [0 ~ 31%]) of vision improvement.

Fig. 2
figure 2

Forest plots of visual acuity outcomes following anti-VEGF therapy in patients with proliferative MacTel type 2. The x-axis displays the proportion of eyes exhibiting each VA outcome, ranging from 0 (0%) to 1.0 (100%).

Full size image

Non-proliferative stage

Our pooled results showed that VA was improved in 30% (95% CI [15 ~ 47%]) of the non-proliferative stage MacTel type 2 patients and stable in 40% (95% CI [18 ~ 65%]). CMT decreased in 59% (95% CI [33 ~ 83%]) of the patients, and progression to the proliferative stage was reported in 9% (95% CI [6 ~ 13%]) of the patients. Subgroup analysis was performed regarding different interventions, including anti-VEGF injection, laser photocoagulation, and observation. Among them, anti-VEGF injection exhibited the highest efficacy to improve VA (58%, 95% CI [29% ~ 85%]), while laser photocoagulation and observation resulted in VA improvement in 0% (95% CI [0 ~ 16%]) and 31% (95% CI [15 ~ 50%]) of the patients, respectively.

Complicated with full-thickness MH

Patients complicated with full-thickness MH were treated with pars plana vitrectomy (PPV) across the included studies. In general the pooled MH closure rate was 56% (95% CI [26 ~ 84%] with VA improved in 53% 53% (95% CI [36 ~ 70%]) and remained stable in 33% (95% CI [18 ~ 50%]) of patients following PPV. Subgroup analysis revealed distinct outcomes between surgical techniques. PPV with internal limiting membrane (ILM) flap achieved a markedly higher MH closure rate of 92% (95% CI [74 ~ 100%]), with VA improvement observed in 60% (95% CI [34 ~ 84%]) of patients. In contrast, PPV with ILM peeling resulted in an MH closure rate of 44% (95% CI [27 ~ 62%]) and VA improvement in 45% (95% CI [23 ~ 66%]) of patients.

Publication bias

Publication bias was not found to be significant using the funnel plot (Supplementary Fig. 1) and Egger’s test (P = 0.119).

Discussion

Our study summarized the demographics, clinical features, multimodal imaging characteristics, management, and prognosis of MacTel type 2. Based on pooled evidence from 205 studies encompassing 11,045 patients, our analysis revealed that MacTel type 2 predominantly affects females (male ratio: 38%, 95% CI [37 ~ 40%]) in their fifth to sixth decades of life, with a relatively high prevalence of systemic comorbidities including hypertension (41%, 95% CI [36 ~ 45%]), diabetes mellitus (35%, 95% CI [30 ~ 40%]), and dyslipidemia (33%, 95% CI [21 ~ 47%]). More than half of the patients (55%, 95% CI [49 ~ 62%]) maintained relatively good vision (BCVA ≥ 20/40), while only 19% (95% CI [15 ~ 24%]) exhibited severe vision loss (BCVA ≤ 20/100). According to the Gass & Blodi staging system, Stage 3 and Stage 2 were the most frequently observed, accounting for 33% (95% CI [25 ~ 41%]) and 26% (95% CI [20 ~ 33%]) of all cases, respectively. Proliferative disease was identified in 12% (95% CI [10 ~ 15%]) of patients. The most common CFP, FFA, OCT, and OCTA features included loss of retinal transparency (76%, 95% CI [65 ~ 85%]), retinal vascular leakage (90%, 95% CI [78 ~ 99%]), EZ disruption (70%, 95% CI [57 ~ 81%]), and RCA (36%, 95% CI [11 ~ 64%]). In proliferative stage, anti-VEGF therapy demonstrated visual improvement in 59% (95% CI [43 ~ 75%]) of patients and a CMT reduction in 85% (95% CI [78 ~ 92%]). In contrast, observation alone yielded visual improvement in only 7% (95% CI [0 ~ 31%]) of patients. In the non-proliferative stage, anti-VEGF and observation led to visual improvement in 58% (95% CI [29 ~ 85%]) and 31% (95% CI [15 ~ 50%]) of patients, respectively, whereas laser photocoagulation failed to produce any visual gain (0%, 95% CI [0 ~ 16%]). Among patients complicated by full-thickness MH, PPV with ILM flap achieved a markedly higher MH closure rate (92%, 95% CI [74 ~ 100%]) and a greater proportion of patients with visual improvement (60%, 95% CI [34 ~ 84%]). In contrast, the ILM peeling technique resulted in lower closure (44%, 95% CI [27 ~ 62%]) and visual improvement rates (45%, 95% CI [23 ~ ~ 66%]).

Baseline characteristics

Our pooled results demonstrated a clear female predominance in MacTel type 2, with males accounting for 38% (95% CI [37 ~ 40%]). This gender disparity has been reported in several studies19,35, yet contrasts with earlier research suggesting that MacTel type 2 affects both sexes equally2,6. Most patients were in their fifth to sixth decades of life, and the majority were Caucasian (68%, 95% CI [47 ~ 87%]). However, the current evidence regarding age, sex, and racial distribution is primarily derived from case series rather than population-based studies, which may limit the representativeness of these findings. Hypertension was the most frequently reported systemic comorbidity, with a pooled prevalence of 41% (95% CI [36 ~ 45%]). Vascular changes associated with hypertension, such as arteriole rigidity and wall thickening, may contribute to the characteristic retinal telangiectasia observed in MacTel type 219. Other systemic conditions, including diabetes mellitus and dyslipidemia, were also commonly identified among affected patients.

Moreover, our pooled findings indicate that visual impairment in MacTel type 2 is generally mild, with more than half of the patients maintaining relatively preserved vision (BCVA ≥ 20/40), and only approximately 20% presenting with severe visual loss (BCVA ≤ 20/100). Such subtle and slowly progressive visual decline may contribute to misdiagnosis or delayed recognition of disease progression in clinical practice. Notably, despite the often mild visual symptoms, previous studies have demonstrated a significant reduction in vision-related quality of life (VRQoL) among individuals with MacTel type 2, underscoring the broader functional burden imposed by this disease beyond measurable VA alone36.

Symptoms and signs

Our pooled results showed that metamorphopsia and mild visual impairment each affected approximately half of patients with MacTel type 2, suggesting that early symptoms can be subtle and easily overlooked in clinical practice. However, this finding should be interpreted with caution given the limited number of studies reporting ocular symptoms. Regarding disease staging, our analysis indicated that Grade 3 and Grade 2 MacTel type 2 were the most prevalent, accounting for 33% (95% CI [25 ~ 41%]) and 26% (95% CI [20 ~ 33%]) of all cases, respectively. Grade 3 is typically characterized by dilated, right-angled retinal venules, while Grade 2 is marked by reduced parafoveolar retinal transparency37. These stages are likely the most frequently encountered in clinical practice. The pooled incidence of proliferative disease was 12% (95% CI [10–15%]). According to Yannuzzi’s OCT-based classification, patients with SRNV and fibrosis are categorized as having proliferative disease. It is important to note that both the Gass & Blodi and Yannuzzi classification systems have inherent limitations. The former was developed before the advent of modern imaging techniques such as OCT, whereas the latter underestimates the natural endpoint of the disease, specifically localized photoreceptor layer atrophy38,39. More recently, Chew et al. analyzed data from the MacTel Project and proposed an updated OCT-based classification system for MacTel type 240. However, as this new system has not yet been widely adopted in published studies, it was not applied in our analysis.

Regarding multimodal imaging characteristics, most patients exhibited loss of retinal transparency, telangiectatic vessels, and retinal pigment epithelium atrophy on CFP, consistent with the underlying pathology of Müller cell dysfunction and photoreceptor degeneration41,42. On FFA, retinal vascular leakage was almost universally observed, typically presenting as diffuse late-phase hyperfluorescence. Although FFA has long been regarded as the diagnostic gold standard for MacTel type 2, it is important to note that structural alterations may occur independently of angiographically visible vascular changes and probably precede them43,44. On OCT, EZ loss was among the most common findings, supporting the neurodegenerative nature of the disease. Longitudinal analyses from the MacTel Project demonstrated that EZ loss correlates strongly with functional decline and often precedes visual symptoms, establishing it as a robust structural biomarker for disease progression and clinical trials45,46. Earlier OCT abnormalities, including hyperreflectivity of the outer plexiform layer and parafoveal capillary dilatation, have also been described as precursors to EZ loss47. On OCTA, our pooled results indicated that RCA was detected in approximately 36% (95% CI [11% ~ 64%]) of cases. Their presence correlates with hyperpigmentation, right-angled venules, and reduced VA48,49. Notably, most proliferative MacTel cases do not represent typical SRNV but rather intraretinal neovascularization or RCA. With advances in OCTA, these vascular alterations can now be visualized with greater precision than with traditional FFA. Recently, a meta-analysis evaluating OCTA-derived parameters in MacTel further complemented our findings24. It showing that foveal superficial vascular density and parafoveal deep vascular density did not significantly decrease in affected patients, providing additional quantitative insight into the microvascular changes in this disease.

Treatment and prognosis

Proliferative stage

The proliferative stage represents a major cause of severe vision loss, affecting approximately 12% (95% CI [10–15%]) of patients with MacTel type 2. Our pooled results demonstrated that anti-VEGF therapy led to VA improvement in 59% (95% CI [43 ~ 75%]) of patients, while 21% (95% CI [14–29%]) experienced VA deterioration, and 85% (95% CI [78–92%]) showed a reduction in CMT. These outcomes were markedly superior to those in the observation group, in which only 7% (95% CI [0–31%]) of patients demonstrated VA improvement and 49% (95% CI [22–77%]) experienced further vision loss. These findings are consistent with prior clinical studies and meta-analyses9,50,51,52, emphasizing that active anti-VEGF intervention plays a crucial role in improving the prognosis of proliferative stage of MacTel type 2. However, one large-scale study (n = 64) recently reported significant reductions in both CMT and macular volume but failed to demonstrate a corresponding improvement in VA following anti-VEGF treatment, even after a median follow-up of nearly 1.5 years23. This discrepancy highlights the need for further research to clarify whether anti-VEGF therapy genuinely enhances visual outcomes or primarily serves to prevent further deterioration. The mechanism underlying the therapeutic effect of anti-VEGF in proliferative MacTel type 2 remains controversial. One hypothesis proposes that endothelial cell degeneration may initiate a vasogenic process independent of ischemia and inflammation53, whereas another suggests that endothelial damage occurs secondary to Müller cell dysfunction, leading to retinal hypoxia, upregulated angiogenesis, and increased VEGF secretion54,55,56.

Historically, focal laser photocoagulation and PDT were the mainstay of treatment for proliferative MacTel type 2 in the absence of more effective options. However, the landmark ANCHOR and MARINA randomized controlled trials subsequently provided robust evidence supporting the superiority of anti-VEGF agents in macular and retinal neovascular diseases, including proliferative MacTel type 257,58. Since then, PDT and laser therapy have been regarded as obsolete, and no further evidence-based recommendations have endorsed their clinical use. Consequently, few recent studies have investigated these modalities, and their therapeutic efficacy was not analyzed in our meta-analysis.

Non-proliferative stage

Our pooled results indicated that anti-VEGF therapy achieved the highesy efficacy in patients with non-proliferative MacTel type 2, with a VA improvement rate of 58% (95% CI [29% ~ 85%]). This was superior to that observed with conservative management (31%, 95% CI [15–50%]), suggesting that anti-VEGF intervention may offer better visual outcomes in the non-proliferative stage, which is consistent with several prior retrospective studies59,60. However, an equivalent number of retrospective studies and a phase II clinical trial have reported opposite findings, suggesting that anti-VEGF therapy is unlikely to provide a general or significant benefit to patients with non-proliferative disease17,61,62,63. The non-proliferative stage typically involves mild visual impairment, which primarily results from irreversible neurodegenerative and atrophic changes, accompanied by low-grade, partially reversible exudation11,64. To date, no studies have clearly elucidated the underlying mechanism by which anti-VEGF agents act in this disease stage. Moreover, some reports have suggested that anti-VEGF therapy not only lacks benefit but may even adversely influence the course of retinal neurodegeneration in MacTel type 217,65,66. Given that non-proliferative disease generally progresses slowly and causes relatively mild visual damage, aggressive therapeutic intervention at this stage remains highly controversial, and our findings should be interpreted with caution.

Nonetheless, anti-VEGF therapy has been shown to be beneficial in certain atypical non-proliferative phenotypes. For example, studies have reported that the rare MacTel type 2 with subfoveal detachment (SFD) phenotype demonstrates an excellent response to anti-VEGF therapy, with complete resolution of SFD in all cases after treatment. The authors proposed that prompt foveal reattachment following anti-VEGF injection may protect against outer retinal degeneration caused by persistent SFD, thereby delaying the onset of foveal atrophy22,67,68.

Regarding laser photocoagulation, our analysis revealed that nearly no patients experienced VA improvement (0%, 95% CI [0–16%]), whereas the majority showed VA deterioration (76%, 95% CI [18–100%]). These results are consistent with previous reports indicating that laser therapy offers no clinical benefit in MacTel type 2 and may, in fact, exacerbate vision loss by inducing retinal fibrosis and thinning, thereby aggravating the outer retinal atrophy characteristic of the disease69,70. It is also important to note that vascular abnormalities in MacTel type 2 typically lie within 500 μm of the foveal center, which has historically led clinicians to adopt laser therapy only with extreme caution and usually in advanced disease stages when vision loss is already present. This inherent selection bias may have impact on visual outcomes. Interestingly, a recent randomized controlled trial published in 2021 reported that nondamaging retinal laser therapy was safe, well tolerated, and associated with both structural and functional improvements in patients with MacTel10. This suggests that the role of laser therapy may depend strongly on the specific laser modality used and warrants further investigation in well-designed prospective studies.

Encouragingly, recent phase III multicenter randomized sham-controlled trials have demonstrated that CNTF, a neurotrophic molecule known to reduce photoreceptor cell loss in animal models of outer retinal degeneration, significantly slowed the rate of EZ loss compared with sham procedures, confirming its neuroprotective effect against retinal degeneration12. In March 2025, the U.S. FDA approved the first-ever treatment for MacTel type 2, known as NT-501, which employs a novel encapsulated cell therapy platform to deliver sustained intraocular levels of CNTF directly to the retina. However, due to the limited number of available studies, CNTF therapy was not included in our pooled analysis.

Complicated with full-thickness MH

MH represents a rare but vision-threatening complication of MacTel type 2 with no existing consensus on management strategy. In our analysis, the pooled incidence of lamellar MH and full-thickness MH was 8% (95% CI [1 ~ 22%]) and 1% (95% CI [0 ~ 3%]), respectively. PPV with either ILM flap or ILM peeling was the most commonly reported treatment for full-thickness MH. Previous studies have also described PPV alone71, PPV with ILM graft72, and autologous retinal transplantation73 for treatment of full-thickness MH, but these approaches were not included in our pooled analysis due to limited sample sizes. Our pooled results demonstrated that the ILM flap technique achieved superior anatomical and functional outcomes compared with ILM peeling, with markedly higher MH closure (92%, 95% CI [74 ~ 100%] vs. 44%, 95% CI [27 ~ 62%]) and visual improvement (60%, 95% CI [34 ~ 84%] vs. 45%, 95% CI [23 ~ 66%]). These findings align with previous evidence in idiopathic and myopic FTMH, where the ILM flap has shown higher closure rates and better postoperative vision recovery74,75,76. The superiority of the ILM flap technique in MacTel-related MH may stem from the preservation of fragile perifoveal structures. Given the temporal thinning of the ganglion cell and nerve fiber layers in MacTel 2 and the frequent coexistence of diabetes-related inner retinal thinning, extensive ILM peeling could aggravate structural damage77,78,79. In contrast, the ILM flap approach helps maintain retinal integrity and yields better functional outcomes80. Nevertheless, although our results indicate a clear advantage of ILM flap–assisted PPV, the limited reporting of surgical details and patient selection criteria across the included studies constrains the generalizability of these findings.

Strength and limitations

To the best of our knowledge, this is the first meta-analysis to comprehensively synthesize all available evidence on MacTel type 2, encompassing demographics, clinical characteristics, and treatment outcomes. As the largest and most comprehensive analysis to date, our study provides a higher level of evidence and aims to serve as a valuable reference for ophthalmologists, enhancing understanding of the disease and facilitating more effective communication with patients.

However, our study also had several limitations. Firstly, given the rarity of MacTel type 2, we adopted a single-rate meta-analysis to maximize the inclusion of all available evidence. Consequently, baseline and final VA or CMT were not pooled to provide a direct comparison of treatment efficacy across interventions. Because baseline VA was not standardized across studies, the observed rates of visual improvement may be subject to bias and should be interpreted with caution. Secondly, the wide inclusion period from inception to 2024 may have introduced variability in medical techniques, patient characteristics, treatment strategies, and follow-up durations, which may contribute to clinical and methodological heterogeneity. We also included studies reported both unilateral and bilateral eyes, despite MacTel type 2 being predominantly a bilateral condition. Thirdly, our search cut-off was March 2024, and studies published within the most recent year may not have been captured. Although we incorporated key new findings into the discussion, some degree of temporal bias is unavoidable. Finally, adverse events are a critical component when evaluating different therapeutic interventions. However, the available data on treatment-related adverse events were insufficient to support a reliable quantitative synthesis. We therefore could not analyze safety outcomes in this meta-analysis. Future studies should not only focus on visual and anatomical prognosis but also systematically evaluate and report adverse events to allow a more comprehensive assessment of treatment safety and efficacy.

Conclusions

MacTel type 2 primarily affects females in their 50 s and 60 s and is frequently accompanied by metabolic comorbidities. Most patients experience only minor visual impairment, yet the disease presents with characteristic multimodal imaging findings. Anti-VEGF therapy provides anatomical and functional benefit in proliferative disease, whereas its role in non-proliferative disease remains uncertain. Among patients with full-thickness MH, PPV with ILM flap appears to yield superior surgical outcomes than ILM peeling. Overall, this study integrates all available evidence from diverse study designs, providing the most comprehensive synthesis to date of the demographics, clinical presentation, and treatment outcomes of MacTel type 2.