FormalPara Key Summary Points

Why carry out this study?

At present, it is not clear whether polypoidal choroidal vasculopathy (PCV) is a subtype of neovascular age-related macular degeneration (nAMD) or a distinct disease within the pachychoroid disease spectrum.

We aimed to evaluate systemic and serum risk factors in PCV and nAMD to comprehensively analyze the difference in pathogenesis between these two diseases.

What was learned from this study?

Compared to the control group, patients with nAMD and PCV were more likely to have coronary artery disease and hypertension, respectively, and hyperlipidemia was significantly associated with both nAMD and PCV.

The aberrant changes in levels of serum lipids (high-density lipoprotein, low-density lipoprotein, and total cholesterol) in PCV patients, combined with the lower level of complement 3, indicate that the molecular mechanism underlying PCV development differs from that of nAMD.

Introduction

Neovascular age-related macular degeneration (nAMD), which is characterized by the development of choroidal neovascularization (CNV), is more prevalent than geographic atrophy in China and is a common cause of blindness [1, 2]; First reported in the 1990s, polypoidal choroidal vasculopathy (PCV) is described as abnormal dilations of terminals of branching vascular networks, which are often best demonstrated by indocyanine green angiography (ICGA), and it is considered to be a subtype of AMD [3,4,5]. Although PCV appears to differ from nAMD based on clinical findings, epidemiological features, and treatments, whether PCV should be defined as a distinct disease entity remains controversial [6, 7].

To date, some studies have focused on risk factors for nAMD or PCV in the Asian population. For example, in a Chinese cohort, patients who smoked were more likely to present ocular features associated with nAMD and PCV [8]. This finding was similar to that of Masato et al. [9], who reported that patients in Japan with nAMD and PCV were frequently smokers. However, the prevalence of other factors, including sex, hypertension, coronary artery disease (CAD), hyperlipidemia, diabetes mellitus (DM), and asthma, reportedly differs between nAMD and PCV [10,11,12].

Cholesterol is an important constituent of drusen, and serum lipids are considered to be a risk factor for AMD development [13,14,15]. In patients who experience symptoms for longer than 3 months, leakage from active aneurysmal elements results in considerable lipid depositions in polypoidal vascular abnormalities [16]. Thus far, there are a limited number of studies analyzing lipid levels in PCV development, and epidemiological findings of the association of serum lipid levels with PCV are inconsistent [17,18,19]. Therefore, it is necessary to obtain more evidence to elucidate the role of lipid/lipoprotein concentrations in the development of AMD or PCV. Although it is known that complement-mediated inflammation plays an important role and excessive complement system activation leads to inflammation and tissue impairment in AMD [20,21,22], few studies have described serum concentrations of complements in patients with PCV.

The results of our previous study suggest significant associations between hyperlipidemia and PCV [23]. Thus, we performed a comprehensive analysis to compare systemic risk factors and serum levels, including high-density lipoprotein (HDL), low-density lipoprotein (LDL), total cholesterol (TC), triglycerides (TG), apolipoprotein A1 (APOA1), apolipoprotein B (APOB), complement 3 (C3), and complement 4 (C4), in cases and controls to further evaluate the characteristics of nAMD and PCV and whether they are the same ocular disease.

Methods

Subjects

A total of 458 unrelated Chinese participants were enrolled in this case–control cohort from January 2018 to December 2020, and they comprised 108 patients with nAMD, 131 patients with PCV, and 219 controls. The study subjects were recruited in a consecutive manner from the Department of Ophthalmology of Peking University People’s Hospital, and they signed informed consent before participating in this research. The procedure was approved by the Ethical Committee of Peking University People’s Hospital and was performed in accordance with the tenets of the Declaration of Helsinki. All participants underwent a standard ophthalmic examination, including assessment of visual acuity, slit-lamp biomicroscopy, and dilated fundus examinations, which was performed by two ophthalmic specialists to avoid potential diagnostic bias. All patients with nAMD and PCV underwent fluorescein fundus angiography (FFA), optical coherence tomography (OCT), and ICGA with an HRA2 device (Heidelberg Engineering, Heidelberg, Germany).

The International Classification System for Age-Related Maculopathy (ARM) was applied for nAMD diagnosis [24]. Patients were defined as having typical CNV, serous retinal detachment, subretinal hemorrhage, serous/hemorrhagic retinal pigment epithelial detachment, and fibrosis, with no evidence of polypoidal lesions. The diagnosis of PCV was based on ICGA results that indicated a branching vascular network terminating in polypoidal swelling. Cases with any other macular abnormalities, such as pathological myopia, presumed ocular histoplasmosis, idiopathic CNV, or other secondary CNV, were excluded. All diagnoses and evaluations of nAMD or PCV were performed by two retina specialists, and a third specialist was consulted in cases of discrepancy. Control participants were selected if they met the following criteria: (1) age ≥ 50 years, and (2) fundus examinations revealed normal findings without any signs of PCV, nAMD, or other vitreous/retinal/choroidal pathological alterations. Control subjects with severe cataracts or other remarkable abnormalities in the anterior segment of the eyes were excluded. Data on age, sex, hypertension, hyperlipidemia, CAD, DM, and asthma were obtained through a questionnaire and interview.

Serum Analysis

Fasting serum samples were collected on the same day (January 2018 to December 2020) as the ophthalmic examinations and analyzed for HDL, LDL, TC, TG APOA1, APOB, C3, and C4 in the clinical laboratory at Peking University People’s Hospital. The authors have a terminal system to check the results of blood examinations for participants.

Statistical Analysis

All statistical analyses were performed using SPSS software (version 16; SPSS Science, Chicago, IL). Comparisons of sex and systemic risk factors (hypertension, hyperlipidemia, CAD, DM, and asthma) between the nAMD, PCV, and control groups were analyzed by chi-square tests. Data on age and serum lipid or complement levels are expressed as the mean ± standard deviation (SD), differences between groups were assessed using independent-samples t tests. We analyzed associations between the development of nAMD or PCV and systemic or serum risk factors using a binary logistic regression model to estimate odds ratios (ORs) with 95% confidence intervals (CIs). P < 0.05 was considered statistically significant.

Results

Of the participants, 131 were PCV patients (mean age: 67.67 ± 8.97 years), 108 were nAMD patients (mean age: 69.81 ± 10.02 years), and 219 were controls (mean age: 67.88 ± 9.02 years). The demographic data and prevalence of systemic diseases of the participants are shown in Table 1. Patients with nAMD tended to have CAD (P = 0.020) and hyperlipidemia (P < 0.001); patients with PCV were more likely to have hypertension (P = 0.006) and hyperlipidemia (P < 0.001). No association of DM or asthma with nAMD or PCV was found (P > 0.05).

Table 1 Systemic characteristics of patients with nAMD or PCV and the controls
Full size table

Table 2 shows the associations of nAMD and PCV with systemic risk factors. In logistic analysis, hyperlipidemia was significantly associated with nAMD (OR = 0.437, P = 0.002; OR = 4.353, P < 0.001), PCV (OR = 0.409, P < 0.001; OR = 4.464, P < 0.001), and nAMD + PCV (OR = 0.449, P < 0.001; OR = 4.217, P < 0.001). DM (OR = 0.535, P = 0.044) was significantly associated with PCV. Other systemic risk factors were not statistically significant (P > 0.05).

Table 2 Logistic regression analysis of systemic risk factors for nAMD and PCV
Full size table

The serum levels of lipids and complements in the cases and controls are presented in Table 3. In univariate analysis, patients with nAMD and those with PCV had significantly lower levels of HDL (P < 0.001; P < 0.001), APOA1 (P < 0.001; P < 0.001), APOB (P = 0.032; P = 0.001), and C3 (P < 0.001; P < 0.001) than controls. With regard to LDL, its level was significantly associated with nAMD (P < 0.001).

Table 3 Serum levels of lipids and complements in nAMD and PCV subjects
Full size table

Furthermore, in logistic analysis (Table 4), levels of HDL, LDL, TG, APOB, and C3 remained associated with nAMD (OR < 0.001, P < 0.001; OR = 0.028, P < 0.001; OR = 0.175, P < 0.001; OR = 0.922, P = 0.022; OR < 0.001, P < 0.001), PCV (OR = 0.001, P = 0.001; OR = 0.097, P = 0.003; OR = 0.410, P = 0.037; OR = 0.895, P = 0.001; OR = 0.001, P < 0.001), and nAMD + PCV (OR = 0.000, P < 0.001; OR = 0.052, P < 0.001; OR = 0.304, P < 0.001; OR = 0.880, P < 0.001; OR = 0.001, P < 0.001).

Table 4 Logistic regression analysis of serum risk factors for nAMD and PCV
Full size table

As illustrated in Table 5 and Table 6, we also analyzed differences in systemic or serum risk factors between nAMD and PCV. Compared with nAMD patients, PCV patients were prone to having higher levels of HDL (P = 0.003) and LDL (P = 0.016) along with lower levels of TG (P = 0.039) in univariate analysis, and levels of HDL and LDL also showed a positive effect in logistic analysis (OR = 3.452, P = 0.022; OR = 4.812, P = 0.004). However, there was no significant difference in systemic risk factors between patients with nAMD and patients with PCV (P > 0.05).

Table 5 Comparison of serum risk factors between nAMD and PCV subjects
Full size table
Table 6 Comparison of systemic risk factors between nAMD and PCV subjects
Full size table

Discussion

In our study, hyperlipidemia showed significant associations with both nAMD and PCV, though other factors, including age, hypertension, CAD, DM and asthma, were not associated with PCV. Although no clear associations of all systemic factors between nAMD and PCV were found, the observed differences in serum lipid levels (HDL, LDL, and TG) suggested that PCV might not be a simple type of AMD and may be a distinct disease entity.

Subretinal hemorrhage (SRH) resulting from vascular terminal dilation in PCV development is a severe complication that causes sudden vision loss, and recurrent SRH is more common in patients with than in those without hypertension [25, 26]. Therefore, hypertension might contribute to the occurrence of polypoidal dilated lesions and sudden vascular rupture. In this study, we found a significant association between hypertension and PCV (P = 0.006). Nevertheless, it is still necessary to perform a study with a larger sample size or experimental research to reveal the potential effect of hypertension on PCV.

To date, the correlation between CAD and AMD is not well defined and rather controversial. In general, the formation of lipid deposition in arterial walls during aging and atherosclerosis is suggested to be similar to the development of esterified lipid-rich deposition or APOB lipid deposition in AMD pathology, resulting in increased choroidal vascular resistance, elevated choriocapillary pressure, decreased choroidal blood flow, and accumulated subretinal deposits. Thomas et al. [27] performed a retrospective study on 3950 patients with AMD and found the existence of an association between CAD and AMD, particularly among older adult patients in the predominantly male population. In addition, the presence of stenotic lesions in the coronary arteries and the severity of stenosis correlate independently with the early stage of AMD, suggesting that patients with CAD are more likely to develop early AMD [28]. These findings are consistent with the significant association between CAD and nAMD (P = 0.020) in our study and strengthen the potential clinical significance of timely diagnosis for AMD in individuals with signs of CAD.

The association between DM and AMD or PCV seen in previous studies is inconsistent with our findings. Cheung et al. [17] did not demonstrate any associations between DM and nAMD or PCV in the Asian population based in Singapore. Conversely, Sakurada et al. [11] found that the percentage of DM in nAMD patients was significantly higher than that in PCV patients, and the severe form of diabetic retinopathy was more prevalent in DM cases with nAMD than in DM cases with PCV. Although the relationship between blood glucose level and PCV remains unclear, the controversial results of some studies might be attributed to different dietary patterns among countries and regions.

Lipid-rich drusen are the hallmark lesions of AMD and accumulate between the retinal pigment epithelium and Bruch’s membrane [29]. Atherosclerosis, as characterized by the generation of atherosclerotic plaques, is a known risk factor for CAD and DM. It is noteworthy that drusen and atherosclerotic plaques also have various components in common, including C3, APOs, and cholesterol [10, 30]. Interestingly, our results showed hyperlipidemia to be a risk factor for both PCV and nAMD (P < 0.001). Therefore, it is necessary to distinguish PCV from nAMD to identify the unique features of the former, and we analyzed levels of serum lipids in patients with nAMD and patients with PCV.

Despite other reports of serum lipid levels associated with AMD, results are conflicting; in particular, there are limited data for PCV. We found that levels of HDL, LDL, TG and APOB were significantly associated with PCV and nAMD (P < 0.05); however, others have shown that differences in serum levels of TG, TC, HDL, and LDL do not differ significantly between patients with PCV or AMD and controls [31,32,33]. Similarly, the level of APOA1, the major protein component of HDL, was not associated with nAMD or PCV in logistic analysis (P > 0.05). In contrast, APOB, the major protein component of LDL, was significantly associated with the two diseases (P < 0.05), which was not consistent with previous findings [15]. As this difference might be due to the ethnicity of the samples, larger multiethnicity studies are needed.

Previous studies have suggested the occurrence of abnormal alterations in complements in AMD pathogenesis [34,35,36]. Our results showed that a lower level of C3 was a risk factor for PCV (P < 0.001), which indicated that complement system activation plays an important role in the development of PCV. It has been reported that depletion of C3 might result from the activation of retinal autoantibodies in AMD patients [37], and further studies are warranted to explore the ocular autoimmune mechanism underlying PCV. Considering the lower levels of lipids and C3, we sought to explore the existence of a relationship between lipid levels and complement activation levels. Comparing PCV with nAMD, we found higher levels of HDL and LDL and lower levels of TG (P < 0.05), which was not completely consistent with previous results, and these risk factors suggested that the pathology of PCV is not similar to that of nAMD [11, 17].

The major limitations of the present study are the relatively small number of cases and that the association of systemic or serum risk factors with nAMD or PCV is poorly understood, so additional studies with larger or more diverse sample sizes will be important to compare risk factors, clinical features, and prognosis between nAMD and PCV.

Conclusion

AMD is a heterogeneous disorder involving the complex interplay between aging, environmental risk factors, and genetic susceptibility [38, 39]. PCV is suggested to belong to the pachychoroid spectrum of diseases, the pathogeneses of which might be associated with vortex veins or running patterns of choroidal vessels [40, 41]. Despite the overlap of clinical or pathophysiological features, alterations in the levels of plasma markers appear to differ between nAMD and PCV [42,43,44,45]. Our findings indicate that the development of nAMD or PCV may correlate with systemic risk factors, especially hyperlipidemia. Serum lipid levels and the complement system are probably involved in both nAMD and PCV. Thus, further studies are required to differentiate between their etiologies, and such findings will be useful for applying appropriate strategies in clinical practice or scientific research in the future.