NAFLD is considered as a multisystemic disease with co-existence of risk factors for type 2 diabetes mellitus (T2DM) and CVD. This intricate relationship is explicable in the broader context of MetS. [21] Therefore, a group of experts have propounded the modified term of metabolic (dysfunction)-associated fatty liver disease (MAFLD) in order to highlight the metabolic aspect in this disease. [22] Nevertheless, the association between choroidal changes and NAFLD AND MetS components remains controversial. Prevailing studies have implied a negative association between body mass index (BMI) and CT [16, 23, 24] but some have reported a positive one. [25, 26] Hypertension is asserted to contribute to a thinner choroid. [27, 28] Additionally, Wong et al. [29] reported positive correlation between hypercholesteremia and CT while Lei et al. [30] found hypercholesteremia to be inversely associated with CT in proliferative diabetic retinopathy patients.
Literature has focused merely on diabetic retinopathy (DR) as ophthalmic manifestation of NAFLD. [31] Studies have been controversial on investigating the association of DR and NAFLD. A cross-sectional study showed that in fact the incidence of DR is lower in NAFLD patients compared to healthy subjects in Iranian population. [32] Meta-analyses performed on this subject demonstrated a lack of correlation between NAFLD and DR overall but believed ethnical differences to be determining in this connection. [33, 34] Song et al. believed that a putative cause for lower incidence of DR in Asian NAFLD patients is that insulin secretion is relatively low in this population compared to Westerners. Of note, only one study has confirmed retinal vascular changes as a predictor of NAFLD in obese patients in the absence of diabetes and hypertension which implies the intrinsic role of NAFLD in initiating retinal changes. [35]
A link between NAFLD pathophysiology and CT can be justified on a molecular basis. Asymmetric dimethyl arginine (ADMA), an endogenous inhibitor of nitric oxide synthase (NOs), plays a critical role in endothelial dysfunction in NAFLD patients. [7]Thus increased level of ADMA, leads to reduced synthesis of nitric oxide (NO) and disturbed vascular homeostasis. There is substantial evidence on NAFLD patients exhibiting higher level of ADMA. [36, 37] Dağ et al. reported that elevated ADMA level correlates well with choroidal thinning in proliferative diabetic retinopathy. [38]
The imperative hallmark of development and progression of NAFLD to NASH is inflammation. Accordingly, increased concentrations of inflammatory mediators including C-reactive protein (CRP), interleukin-1β (IL-1β), interleukin-6 (IL-6), tumor necrosis factor-α (TNF-α) and intercellular adhesion molecule-1 (ICAM-1) is associated with higher risk of NAFLD. [39] In a study by Fang et al. [40] CRP level was proven to be adversely associated with CT. Another study demonstrated that TNF-α is related to a thinner choroid while IL-17 is pertinent to a thicker choroid. [41] This study also suggested that CT is inversely associated with homocysteine, albumin, aspartate transaminase (AST), alanine transaminase (ALT) and high-density lipoprotein (HDL) all of which are applicable as biomarkers of NAFLD and hepatic damages. [42, 43] Moreover, Romero-Ibarguengoitia et al. [35] indicated that non-diabetic thin NAFLD patients with retinal vascular changes (RVC) have higher levels of IL-2, IL-6, TNF-α, and interferon-γ and CRP level independently contributes to higher risk of NAFLD, obesity and RVC.
To the best of our knowledge, choroidal microvasculature changes in NAFLD were studied in only one published research to date. [44] This study demonstrated that NAFLD confers a decrease in choroid vascular index (CVI) but does not affect mean subfoveal choroidal thickness (SFCT) which is consistent with our study.
In this study we were unable to find any significant difference between NAFLD and control groups in terms of CFT or CT in neither of distances. The only significant difference was the higher left eye nasal 1000 CT (µm) in NAFLD compared to control group which seems clinically inconclusive. We speculate that a proper difference might be seen in rather severe forms of NAFLD and CT might not apply as an early detection marker in NAFLD. Of note, Strobbe et al. [45] reported that hepatitis C virus (HCV) positive subjects with liver fibrosis had thicker choroids in comparison with those lacking fibrosis. Probably, choroidal manifestations could be discerned after the onset of hepatic injuries rather than a bland steatosis. Our study has some limitations. First, the sample size is relatively low. Second, we weren’t able to calculate CVI due to its cumbersome calculation. Third, due to the cross-sectional setting of study, we could not investigate causality relationships.