Retinal and Choroidal Thickness Changes in Populations with Helicobacter pylori Infection by Swept-Source Optical Coherence Tomography

Here, we compared macular retinal and choroidal thickness in populations with and without Helicobacter pylori infection using a swept-source optical coherence tomography (SS-OCT) device, which is one of the recent milestones in the development of retinal and choroidal visualization.

Around half of the world's population is infected with Helicobacter pylori (H. pylori), which is closely related to several ocular diseases. The study aims to evaluate the retinal and choroidal thickness changes in subjects with H. pylori infection by swept-source optical coherence tomography (SS-OCT). The ophthalmic examination and ¹³C-urea breath test (¹³C-UBT) were performed on all subjects participating in the cross-sectional study. The participants were divided into H. pylori (+) and H. pylori (−) groups depending on the ¹³C-UBT results. This study covered 2574 right eyes from 2574 subjects with H. pylori infection and 2574 right eyes from 2574 age- and sex-matched individuals without H. pylori infection. Out of the nine sectors of the early treatment diabetic retinopathy study (ETDRS) grid, the maximum retinal thickness was in the inner superior sector, while the minimum was in the center sector. The maximum choroidal thickness was in the inner superior sector, while the minimum was in the outer nasal sector. The choroid of each area of the ETDRS subfield in the H. pylori (+) group was significantly thicker than that in the H. pylori (−) group, but retinal thickness did not show any difference between the two groups. Increased choroidal thickness may be an early indicator of H. pylori-associated retinal or choroidal diseases.

Helicobacter pylori (H. pylori) is a gram-negative bacterium that settles on the gastric epithelium surface and infects approximately 50% of the world's population^1,2. The incidence of H. pylori infection varies from region to region, and infections are more common in societies with low socioeconomic status³. H. pylori infections are typically acquired during childhood, and a significant proportion of individuals infected with H. pylori do not exhibit any symptoms⁴. If left untreated, H. pylori can colonize throughout an individual's lifetime.

The main manifestations of H. pylori infection are chronic gastritis, peptic ulcer, gastric cancer, and MALT lymphoma^2,5,6,7. Currently, the role of H. pylori has been shown in the pathogenesis of extra gastric pathologies, such as atherosclerosis, idiopathic thrombocytopenic purpura, urticaria, obesity, and iron-deficiency anemia^8,9,10,11,12. Some studies have also demonstrated that H. pylori is associated with several eye diseases, including central serous chorioretinopathy¹³, blepharitis¹⁴, glaucoma¹⁵, and uveitis¹⁶.

As the tissue with the highest blood flow, the choroid provides oxygen and nutrition to the retina while removing metabolic waste^17,18,19. Changes in retinal and choroidal thickness may occur due to systemic diseases¹⁸ and physiological conditions²⁰ and may also contribute to ocular pathologies such as central serous chorioretinopathy²¹, diabetic retinopathy²², age-related macular degeneration²³, uveitis²⁴, glaucoma²⁵, and myopia-related chorioretinal atrophy²⁶. Thus, accurate measurements of retinal and choroidal thickness are of great significance for monitoring the onset and progression of the diseases. Swept-source optical coherence tomography (SS-OCT), a recent advancement in retinal and choroidal visualization, offers a noninvasive means of obtaining more precise images of the retina and choroid.

Currently, there have been limited studies examining the relationship between H. pylori infection and choroidal thickness, and the data in the literature are controversial^19,27,28. Some authors have reported that choroidal thickness is related to H. pylori infection¹⁹, whereas others have not^27,28. The present study aims to evaluate the associations between H. pylori infection and the thickness of the retina and choroid.

This cross-sectional study was conducted at Huashan Hospital of Fudan University between September 2018 and April 2019. It was approved by the Institutional Review Board of Huashan Hospital, affiliated with Fudan University (No. KY2016-274). Informed consent was obtained from all participants prior to the examinations.

1. Inclusion criteria

1. Collect all subjects' general conditions, including age, sex, and a history of systemic diseases (thyroid diseases, diabetes mellitus, hypertension, dyslipidemia). Inclusion criteria are as follows:
   1. Include participants aged 18-70 years without a history of systemic diseases noted in step 1.1.
2. Collect all subjects' ocular conditions, including best-corrected visual acuity (BCVA), a history of eye diseases (retinal and choroidal diseases), and a history of ocular trauma or surgery. Inclusion criteria are as follows:
   1. Include participants with BCVA ≥ 20/25 Snellen.
   2. Include participants without a history of eye diseases.
   3. Include participants without a history of ocular trauma or surgery.

2. ¹³C-urea breath test

NOTE: Among the several noninvasive tests for the diagnosis of H. pylori infection, including stool antigen testing and serological testing, ¹³C-urea breath test (¹³C-UBT) is considered to be the gold standard with high sensitivity and specificity^1,29.

1. After an overnight fast, ask the participant to blow as indicated in the test kit. Collect a baseline breath sample.
2. Have the subject take 75 mg of the ¹³C-urea substrate in a citric acid base at the ¹³C-UBT meal. Ask the participant again to blow 30 min after urea ingestion. Collect the breath sample.
3. NOTE: This test is based on the urease produced by H. pylori present in the stomach, liberating carbon dioxide (CO₂) from urea. The uptake of labeled urea leads to the generation of labeled CO₂, which can subsequently be quantified in exhaled breath.
4. Calculate the result expressed as the difference value between the two breath samples. Consider the test positive for H. pylori infection when the delta over baseline value was >3.5%.
   NOTE: False-negative results can occur in subjects taking proton pump inhibitors (PPIs) or antibiotics that interfere with the sensitivity of ¹³C-UBT³⁰. To avoid a false-negative result, PPIs and antibiotics should be stopped for at least 3 months before ¹³C-UBT.
5. Based on the ¹³C-UBT results, divide the participants into two groups: The H. pylori (+) group refers to those with H. pylori infection, and the H. pylori (−) group refers to those without H. pylori infection.

3. Swept-source optical coherence tomography imaging

1. Open up the Power button of the SS-OCT device. Click the Radial Dia.6.0 mm Macula Overlap 4 button.
2. Capture OCT images for every eye by scanning 100,000 A-scans/s between 8 a.m. and 10 a.m.
   NOTE: The retinal and choroidal thickness were defined, respectively, referring to previous articles (Figure 1)^31,32.
3. Generate thickness images of SS-OCT automatically based on the standard early treatment diabetic retinopathy study (ETDRS) subfield (Figure 2).
4. Manually check each segmented layer line for every image to avoid automatic measurement errors^31,32.
5. Exclude the unclear OCT images due to unstable fixation or media opacity.

4. Statistical analysis

1. Click on the statistical analysis software. Import the data into the software.
   NOTE: SPSS was used for statistical analysis. Continuous data was summarized using the mean ± standard deviation, while categorical data was presented as frequency (percentage).
2. Compare continuous variables via the t-test, whereas categorical variables by the chi-square test. Define statistical significance as a p-value less than 0.05.
   NOTE: The analysis included only the right eyes, which were considered as a separate study unit.

A total of 2574 right eyes of 2574 subjects who tested positive for H. pylori and 2574 right eyes of 2574 individuals who tested negative for H. pylori were evaluated. Table 1 lists the baseline characteristics of the participants. The male/female ratio was 1334/1240 in both the H. pylori (+) group and the H. pylori (−) group. The mean age was 42.68 ± 10.76 years (range, 18-70 years) in both groups.

Out of the nine sectors of the ETDRS grid, the maximum retinal thickness was in the inner superior sector, while the minimum was in the center sector. The retinal thickness of each sector of the ETDRS grid showed no significant difference between the H. pylori (+) and H. pylori (−) groups (Table 2).

Out of the nine sectors of the ETDRS grid, the maximum choroidal thickness was in the inner superior sector, while the minimum was in the outer nasal sector. The choroid in each sector of the ETDRS grid in the H. pylori (+) group was significantly thicker than that in the H. pylori (−) group (Table 3).

Figure 1: Representative images of retinal and choroidal thickness. (A) Retinal thickness: between two green lines. (B) Choroidal thickness: between two green lines. Please click here to view a larger version of this figure.

Figure 2: Representative images of the retinal and choroidal thickness in nine sectors of the ETDRS grid. (A) Retinal thickness. (B) Choroidal thickness. Please click here to view a larger version of this figure.

Table 1: Demographic characteristics

Table 2: Retinal thickness

Table 3: Choroidal thickness

H. pylori infection has been related to many extra gastric disorders, such as neurological, dermatological, cardiovascular, and ocular diseases³³. However, the pathogenesis of H. pylori-associated ocular diseases remains elusive. We used SS-OCT to measure macular retinal and choroidal thickness in subjects with and without H. pylori infection. The results showed that the choroidal thickness was significantly increased in subjects with H. pylori infection when compared to the choroidal thickness of age- and sex-matched control subjects, whereas no difference was observed in retinal thickness between the two groups. This may suggest a correlation between H. pylori infection and choroidal thickness. Choroid may serve as a link connecting H. pylori infection and eye diseases.

There have been limited studies on the relationship between H. pylori infection and choroidal thickness. Can et al.¹⁹ compared the choroidal thickness of 25 subjects with H. pylori infection and 25 subjects without H. pylori infection in Turkey using enhanced depth imaging (EDI)-OCT instead of SS-OCT and identified H. pylori infection as a significant factor for increased choroidal thickness. Similar to this research, in our study, we observed a higher choroidal thickness in subjects with H. pylori infection. In contrast, two other studies conducted in Turkey evaluated the effect of H. pylori infection on choroidal thickness with EDI-OCT, and researchers discovered no significant difference in choroidal thickness between the subjects with and without H. pylori infection^27,28. EDI-OCT may have several inherent limitations that can result in these conflicting observations. Firstly, it may not accurately detect the chorioscleral interface (CSI) in some eyes with thicker choroids due to its low penetration through the retinal pigment epithelium (RPE). In comparison with EDI-OCT, SS-OCT can detect the CSI in 100% of eyes³⁴. Secondly, in most studies using EDI-OCT, the choroid is manually measured at only one or a few points, which can be influenced by focal thickening or thinning of the choroid, as the CSI may have an irregular shape in some eyes^35,36. Thirdly, manual measurement by different subjects can lead to variations in choroidal thickness. SS-OCT has the capacity to surmount these constraints³⁷. In this investigation, the retinal and choroidal thickness was acquired utilizing SS-OCT and subsequently averaged in accordance with the ETDRS grid through automated means, thereby ensuring confirmed reliability.

A second possible factor for the varying results can be ethnic differences among the study populations. The East Asian variant of the CagA protein exhibits greater carcinogenicity and virulence in comparison to its Western counterpart³⁸. As a major virulence determinant, H. pylori cag PAI is associated with gastric mucosal inflammatory cell infiltration and interleukin (IL)-8 secretion³⁸. Possibly, this explains why H. pylori-positive and H. pylori-negative participants in our study had significantly different choroidal thicknesses.

This study showed that the choroid in the H. pylori (+) group was thicker than that in the H. pylori (−) group. In the study of White et al.³⁹, H. pylori was shown to interact with pattern recognition receptors expressed by gastric epithelial cells and thus activated inflammation. Proinflammatory cytokines, including tumor necrosis factor α, interferon γ, IL-8, IL-1β, IL-6, IL-12, CCL2-5, CCL20, and CXCL1-3, are upregulated in the infected gastric mucosa of H. pylori-infected patients. These cytokines lead to the recruitment of inflammatory cells, such as neutrophils and macrophages, and increase endothelial permeability. Moreover, virulence factors, such as lipopolysaccharide, immunoglobulin-G antibody, and cross-reactivity of anti-CagA antibodies in response to autoimmunity, may lead to endothelial dysfunction^9,40. Increased choroidal endothelial permeability caused by inflammatory cytokines and endothelial dysfunction by direct virulence attack leads to the thickening of the choroid.

The blood-retina-barrier formed by the retinal capillary endothelium and RPE through well-developed tight junction proteins can prevent harmful substances from entering the eye and maintain the physiological environment for a functional retina, which may explain why the retinal thickness is not affected by H. pylori infection.

There are several limitations to the present study. First, the research is a cross-sectional study. It is not possible to observe changes in the retina and choroid during the disease or after treatment. We are currently conducting a longitudinal study with an extended cohort. Second, participants with subclinical diseases that may affect the retina and choroid were included despite all subjects being assessed in detail. Third, the findings of this study may not be generalizable to other populations, as the participants enrolled in this research are all from China.

In summary, the choroid was thicker in subjects with H. pylori infection compared with normal individuals. Increased choroidal thickness may be an early indicator of H. pylori-associated retinal or choroidal diseases.

None of the authors has a financial or proprietary interest in any material or method mentioned.

This study was funded by grants from the National Natural Science Foundation of China (No. 81900879) and the Science and Technology Commission of Shanghai Municipality (No. 20Y11910800).