Macula, choroid and retinal nerve fiber layer optical coherence tomography biomarkers in liver dysfunction

Introduction

Primary biliary cirrhosis (PBC) is an choronic progressive autoimmune liver disease. Patients with PBC are frequently female (90%) and the age at diagnosis is usually more than 40. Chronic cholestasis and circulating anti-mitochondrial antibodies (AMA) are present in PBC. In particular, the M2 distribution of AMA is important in confirmation of the diagnosis in patients with PBC. Non-suppurative destructive cholangitis and interlobular bile duct destruction is detected on liver biopsy. Alkaline phosphatase (ALP) levels are 1.5 times above the upper limit of normal for 24 weeks. PBC is often coexists with other autoimmune diseases. The most common association is Sjögren’s syndrome (31%) [1, 2].

The liver is a very important metabolic center. Ocular findings can be detected at any age from birth due to blockage of certain metabolic pathways or accumulation of toxic substances in liver diseases. Conjunctiva, cornea, lens, retina and optic nerve are affected in liver diseases [3].

Changes in ganglion cell layer (GCL++), retinal nerve fiber layer (RNFL), macular and choroidal thickness detected by optical coherence tomography (OCT) are helpful in the early diagnosis of ocular diseases. GCL and RNFL are the first parameters to deteriorate in optic nerve pathologies. Similarly, macular and choroidal analyzes using OCT in retinal diseases are helpful in early diagnosis [4]. We aimed to determine whether the retina and optic nerve are affected in liver dysfunctions in our study.

Materials and methods

Ankara Numune Education and Research Hospital Ethics Committee approved the protocol for the study (E-18-2440). This study was performed in accordance with the Declaration of Helsinki. Written informed consent was obtained from all patients.

We included 52 patients with PBC (PBC group) and 61 healthy individuals (Control group) in our study. Blood Alanine aminotransferase (ALT), aspartate aminotransferase (AST), gamma-glutamyltransferase (GGT), alkaline phosphatase (ALP), total bilirubin, albumin, prothrombin time results as the International Normalized Ratio (INR), neutrophil (NEUT), lymphocyte (LYMP), neutrophil lymphocyte rate (NLR), platelet (PLT), mean platelet volume (MPV) levels were analyzed in both groups.

The blood samples were taken after overnight fasting. Samples are centrifuged at 3,000 rpm and stored at −20 °C until analysis time. Biochemical variables including aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP) and gamma glutamyltranspeptidase (GGT), Albumin and Protein levels were analyzed by photometric method on the Siemens ADVIA 1800 device (Siemens Healthcare, Erlangen, Germany). A complete blood count was performed by ADVIA 2120 system (Siemens, USA). The INR is calculated by taking the ratio between the prothrombin time of patients and the prothrombin time of normal individuals. Prothrombin time measurements are determined by Sysmex CS 2500 (Siemens Healthcare, Beersel, Belgium) instrument and optical reaction method.

All patients underwent ophthalmic examination. RNFL, Inferior RNFL (IRNFL), Superior RNFL (SRNFL), GCL++ (Ganglion cell layer + RNFL + Inner plexiform layer), central macular thickness (CMT) and subfoveal choroidal thickness (CT) measurements were done by swept source optical coherence tomography (DRI Triton swept source optical coherence tomography; Topcon, Tokyo, Japan). The swept source optical coherence tomography uses a adjustable laser as the light source to provide a centered wavelength of 1,050 nm, achieves a scanning speed of 100,000 A-scans per second, and provides axial and transverse resolution of 8  and 20 μm, respectively in tissue. We used 3D(H) Macula+5 Line Cross 12 × 9 mm mode and Peripapilar 3D Disc 6 × 6 mm mode. After these measurements, the thickness of the retinal layers could be obtained.

The choroidal vascular index (CVI) was obtained by using image binarization technique. Image binarization technique was performed by using image processing and analysis in Java (Image J; version 1.47; http://imagej.nih.gov/ij/). CVI was calculated by dividing luminal area by the total subfoveal circumscribed choroidal area. CVI was calculated as the luminal area and stromal area ratio (LA/SA). CVI was determined in vertical and horizontal sections [5]. Choroidal images obtained from macular sections on OCT were used to detect CVI.

Patients with glaucoma, optic neuropathy, uveitis, retinopathy, spherical equivalent values greater than ±4.0 diopters, ocular trauma or previous intraocular surgery were excluded from the study. Patients with neurological and systemic vascular disease were also excluded.

PBC patients were diagnosed with liver function tests, AMA M2 positivity and liver ultrasonography findings. Patients with Vitamin A deficiency were not included in our study.

Physicians working in ophthalmology and gastroenterology clinics without known systemic or ocular pathology were included in the control group.

SPSS 26.0 (IBM Corporation, Armonk, New York, United States) program was used in the analysis of the variables. Kolmogorov-Smirnov test was used for normalitiy distibution. Mann Whitney U test was used for two group analysis. It was considered statistically significant when p-value was less than 0.05. Neural Network analysis was used to find the variable with the highest significance associated with the groups.

Results

There were 52 patients in the PBC group, 50 female (96.2%) and 2 (3.8%) male patients. 57 females (93.44%) and 4 (6.55%) males were included to the Control group (p>0.001).

GGT (p=0.001), ALP (p<0.001), albumin (p<0.001), NEUT (p=0.046) and MPV (p<0.001) values were higher in the PBC group. INR (p=0.017) and PLT (p=0.027) values were lower in the PBC group. CMT (p=0.015), CT (p=0.002), and GCL++ (p<0.001) values were found to be significantly lower in the PBC group. The results and comparison of both groups are shown in Table 1.

Looking at the variables with the highest significance associated with the groups, 39.3% of GLC++ was also significant after ALP, albumin, NEUT. The order of importance of the variables with the highest significance associated with the groups is given in Table 2 and Figure 1. GCL++ correlated very weak with ALT (r=0.199, p=0.045) and AST (r=0.196, p=0.048) levels. RNFL correlated very poorly with ALT (r=0.222, p=0.024) and AST (r=0.210, p=0.034) levels.

Figure 1:

Neural network analysis for indicating the significance of variables [1–18].

Discussion

In the current study, we evaluated thinning in GCL++, CMT and CT measurements in PBC group. RNFL, SRNFL, IRNFL, HCVI and VCVI measurements were not different between the groups. Serum GGT is widely used as an index of liver dysfunction. Serum GGT levels increased in obstructive liver disease like PBC. This condition leads to increased free radical production and the glutathione depletion. Increased serum GGT levels are associated with coronary heart disease and high mortality risk. Increased oxidative stress in the body could be expected to affect all ocular structures [6]. In this study, GGT values were higher in PBC group.

Signaling roles for extracellular ATP, ADP, and adenosine have been founded in the eye. ALP is an important enzyme working in this system. ALP has been shown to be present in the vitreous, retina pigment epithelium, choriocapillaris, ganglion cells and optic nerve head. High levels of ALP were found especially in non-clearing vitreous hemorrhage and neovascular membranes in eyes with diabetic retinopathy. Increased ALP levels in the choriocapillaris may have an effect on ocular blood flow [7]. ALP levels were higher in PBC group. The thinning in CMT, CT and GCL++ that we detected in the PBC patients may be due to increased GGT and ALP levels. However, in our study, we found that GGT levels in 11 patients and ALP levels in 10 patients were within normal laboratory values. There was no patient with normal ALP and GGT levels together. There were 10 patients in the control group, each of whom was outside the normal laboratory limits of ALP or GGT.

Carotenoids act as optical filters in the macula, protecting cone cells from photochemical damage. Carotenoids are also powerful antioxidants. Getting enough carotenoids with the diet helps to prevent retinal diseases such as senile macular degeneration. Major carotenoids detected in ocular tissues are lutein, zeaxanthin, lycopene and their oxidative products. Zeaxanthin and lutein are present in many ocular tissues such as retina, retina pigment epithelium, choroid, iris, lens. In addition, there are high amounts of lutein and zeaxanthin in the plasma and liver. The liver plays a role in the storage and metabolism of carotenoids [8]. Carotenoid metabolism also may be responsible for the thinning of the CMT, CT and GCL++ layers, which we found in our study.

The liver has important effects on the metabolism of vitamin A, which has an important place in visual function. The storage of vitamin A in tissues is provided by the support of hepatocyte-associated hepatic stellate cells. Hepatocyte-associated hepatic stellate cells may respond to stimuli of low vitamin A in tissues throughout the body. These cells have the ability to both take up vitamin A for storage and release vitamin A into lymphatic capillaries and blood vessels. It is also hypothesized that the eye can send signals without the need for hepatocytes to release vitamin A. Vitamin A deficiency is detected in 33.5% in patients with PBC. Night blindness due to vitamin A deficiency has been reported in patients with PBC [9], [10], [11]. Therefore, patients with vitamin A deficiency were not included in our study.

Platelet-derived growth factor (PDGF) is an important mitogen for connective tissue cells and especially for the retina. PDGF has been detected in human PLT. PDGF consists of dimeric proteins that play an important role in the development of neural, glial and vascular cells. PDGF-BB activates PDGFR-β in the rod photoreceptors, which then suppresses light-induced photoreceptor damage. PDGF-BB has also neuroprotective effects. Decreased PDGF levels and related pathologies may occur in patients with reduced platelets [12], [13], [14].

It has been reported that MPV increases depending on the histopathological severity of PBC disease. Decreased platelet count is frequently detected in PBC patients. Threshold thrombocyte count causing complications is ≤132.5 × 109/L. NLR may be associated with fibrosis in the liver. Most importantly, INR and PLT are known to independently affect prognosis in patients with PBC [15], [16], [17]. There was increased MPV, NLR levels and decreased PLT counts in our PBC patients. In addition, INR levels were lower in PBC group.

Metabolic syndrome is a total disorder of the metabolism of patients. Thinning of RNFL and macular thickness is observed in patients with metabolic syndrome. Body mass index and diastolic blood pressure cause a decrease in RNFL thickness. Triglyceride and high density lipoprotein levels could affect macular thickness. The OCT parameters we measure are affected by many factors, especially the metabolic status of individuals [18]. The inadequacy of our study is that we did not include body mass index and blood pressure.

The macula is primarily affected in liver dysfunctions such as PBC. This pathology is revealed as thinning in CMT, CT and GCL++ in OCT analyses. According to our literature review, our study is the first to evaluate the reflection of liver dysfunction in OCT analysis.