Bulbar Conjunctival Thickness Measurements With Optical Coherence Tomography in Healthy Chinese Subjects

Abstract

Purpose.: To conduct in vivo measurements of the bulbar conjunctiva epithelial, stromal, and total thicknesses in healthy Chinese subjects using optical coherence tomography (OCT) and to compare the conjunctival thickness differences between sex and age.

Methods.: From January to July 2012, healthy Chinese subjects were recruited to undergo a Cirrus HD-OCT examination using the Anterior Segment 5 Line Raster scanning protocol. Ophthalmoscopy and slit-lamp examinations were performed on all eyes, along with the determination of visual acuity. The bulbar conjunctiva epithelial thickness, stromal thickness, and full thickness were measured on cross-sectional conjunctiva images. The average values of each parameter were compared between males and females and among eight age groups (<20, 21–30, 31–40, 41–50, 51–60, 61–70, 71–80, and >80 years).

Results.: A total of 711 eyes in 711 subjects (average age 46.5 years) were included. The average conjunctiva epithelial, stromal, and full thickness measurements were 42.4 ± 7.4 μm, 197.7 ± 32.5 μm, and 240.1 ± 29.8 μm, respectively. No significant difference was found for any of the three parameters between sexes. The conjunctiva epithelial thickness values decreased significantly after age 20 and were then maintained at relatively low levels, followed by a sharp increase at 60 years of age. The conjunctiva stromal thickness decreased throughout the entire lifetime, and the full thickness decreased from the teenage to the elderly periods.

Conclusions.: None of the Chinese conjunctiva epithelial, stromal, and full thickness was sex-related, but all of them were age-related.

Introduction

The assessment of conjunctival thickness, including the epithelial and stromal thicknesses, can be an adjunct to a clinical examination and helpful in the treatment and follow-up of various conjunctival diseases, such as conjunctival inflammation and conjunctivochalasis. ¹ As an example, thinning of the conjunctiva after anti-inflammatory medication may be viewed as an alleviation of edematous conjunctiva. Early, precise monitoring of the changes in conjunctival thickness is possible only if a database is first established in healthy subjects. To the best of our knowledge, no data have been published describing the conjunctival thickness, including the conjunctiva epithelial and stromal thicknesses, in a large, healthy population base. 

Biometric studies rely on technique. Traditional in vivo observations of the gross tissue appearance of the conjunctiva are derived from a slit-lamp biomicroscope. In vivo confocal microscopy enables a magnified view of the conjunctiva cells. However, both of these techniques cannot provide cross-sectional thickness measurements of the conjunctiva layers. The rapid development of noninvasive optical coherence tomography (OCT) has enabled the visualization of human conjunctival layers with high-resolution, cross-sectional tomographic tissue images that facilitate the measurement of the conjunctival thickness. Until now, only a few studies with small sample sizes have addressed conjunctival thickness with OCT measurements in healthy subjects. In 2008, Feng and Simpson used a modified retinal OCT (Stratus OCT, model II; Carl Zeiss Meditec, Inc., Dublin, CA) to visualize and measure the thickness of the human conjunctiva epithelial layer in 13 healthy subjects. ² In 2011, Francoz and associates used spectral-domain OCT (SD-OCT) to measure limbal and bulbar conjunctiva epithelial thickness in 55 eyes of 28 healthy subjects. ³ In our previous study, we found that the OCT (Cirrus OCT; Carl Zeiss Meditec, Inc.) observations were consistent with the histological and clinical data. ⁴ For example, in our preoperative bulbar conjunctival OCT images of the glaucoma patients eyes, below the superior, continuous, narrow, low reflective conjunctival epithelial layer, there was a relatively wider high reflective layer that was separated from the underlying progressively lower reflective layer with a narrow demarcation. With nylon 6-0 nonabsorbent threads placed beneath the conjunctival stroma during a trabeculectomy, the narrow demarcation was confirmed as the boundary between the conjunctival stroma and Tenon's capsule in the postoperative OCT images. Therefore, the middle high reflective layer of approximately uniform thickness corresponds to the bulbar conjunctival stroma. We also measured the conjunctiva epithelial, stromal, and full thicknesses in 18 healthy Chinese subjects, further proving that conjunctival thickness measurements obtained using OCT are immune to variability in measurements that occur over time both within and between sessions. ⁴ Using the same method, we have measured the conjunctival thickness in 711 eyes of 711 healthy Chinese subjects using OCT (Carl Zeiss Meditec, Inc.) and compared the differences according to age and sex. 

Methods

Between January 1 and July 31, 2012, an epidemiological survey for ocular surface diseases was conducted in the Guangyuan community, located in the central part of Shanghai city. Healthy subjects were randomly recruited from local residents in the epidemiological study; their average annual income was at the median level for Shanghai residents. The inclusion criteria were the following: Chinese race; no systemic diseases (e.g., hypertension and diabetes mellitus); no previous ocular surgery; no ocular surface disease; no history of trauma or glaucoma; no lid abnormalities; no abnormal ocular surface findings on the slit-lamp and funduscopic examinations; no psychological disorders; and willingness to follow the interviews and examinations according to the study protocol. The participants were not current contact lens wearers or users of any eyedrops up to 2 hours before the OCT imaging. 

The size (n) of each group could be calculated using the following sample size calculation equation for multiple group analysis of variance n = ψ ² (∑s_i ²/k)/(∑(x_i –X)²/(k–1)). ⁵ We set the study group numbers to k = 8, α = 0.05, and β = 0.1. Based on our former work, the average bulbar conjunctiva full-thickness from a population sample of 18 normal Chinese was 238.8 μm, standard deviation was 51.1 μm. ⁴ We made assumptions that the s_i (standard deviation) of the eight study groups was equal and the difference in the average thickness between two adjacent groups was equal. We selected the effect size (the difference between two adjacent groups divided by the standard deviation) at a medium level of 1/3, resulting in n ≈ 90. 

This study was performed at the Department of Ophthalmology, Putuo Hospital, affiliated with Shanghai Traditional Medicine University. After case collection, all participants underwent visual acuity measurements and screening for ocular diseases by ophthalmoscopy and slit-lamp examination. Because of the high correlation between both eyes, we did not use both eyes of the same subject but selected the right eye for convenience. 

We followed the measurement protocol as described in our previous report, ⁴ using a commercial HD-OCT device (Cirrus HD-OCT 4000; Carl Zeiss Meditec, Inc.) for the in vivo measurements of bulbar conjunctival thickness. This system is a spectral domain OCT platform that takes 27,000 axial scans per second and has a 5-μm axial resolution. After turning off the room light, one experienced examiner (BL) positioned the patient's chin on the chin rest and asked the participant to press his head firmly against the forehead rest without blinking. The examiner then directed the scanned eye's upper nasal gaze and imaged the lower temporal conjunctiva to the 7:30 position in the right eye, approximately 3 to 5 mm away from the corneal limbus. We considered that observational interference from the rectus muscle tendons, underlying Tenon's capsule, and pinguecula in the conjunctival structure, could be minimized by choosing this location for OCT imaging. ⁴ The cross-sectional conjunctiva images were acquired using the Anterior Segment 5 Line Raster scanning protocol while the scan lines were rotated to image a cross-section perpendicular to the corneoscleral limbus at the scanned location. ³ This mode scans through five parallel lines of equal length at 3 mm and separated by 250 μm. Each line is composed of 4096 A-scans, and each five-line raster scan took approximately 0.75 seconds (in the public domain at www.meditec.zeiss.com; Carl Zeiss Meditec, Inc.). To ensure the tissue was measured at right angles, the examiner adjusted the area of the eye visible in the iris viewpoint with the chin-rest control arrows. After image capture, the individual line with the greatest clarity of detail was selected for the conjunctival structure analysis using the high-definition image analysis protocol. The conjunctival image should have well-defined anterior and posterior surfaces and the least amount of motion artifacts, especially within the area where the measurement caliper was to be placed. If the images obtained after scanning were not satisfactory because of eye movement, the scan was repeated. An engineer from Carl Zeiss Meditec, Inc. frequently helped us ensure that the HD-OCT device (Carl Zeiss Meditec, Inc.) worked properly. 

Three conjunctival thickness parameters were measured using central corneal thickness measurement software on the OCT device (Carl Zeiss Meditec, Inc.): the conjunctiva epithelial thickness, conjunctiva stromal thickness and conjunctiva full-thickness. On the cross-sectional image of the conjunctiva, the measurements were made at positions with apparent tissue landmarks, which were identified by differences in brightness between the tissues and visual clues such as the spaces under the conjunctiva; these landmarks were used to identify the endpoint of the conjunctival stroma. ⁴ During each conjunctival thickness measurement session, the experienced examiner took five independent measurements of the temporal conjunctiva in the right eye, and the mean value was used for analysis. 

The study was approved by the Institutional Review Board of Putuo Hospital, which is affiliated with Shanghai Traditional Medicine University, and was performed in accordance with the Declaration of Helsinki. Written consent was obtained from each participant. 

The three conjunctival thickness parameters are presented as the mean, minimum and maximum values, and SDs. The average age and average value of each parameter were compared between the male and female groups using a two-tailed Student's t-test. The average of each parameter was compared among the eight age groups (<20, 21–30, 31–40, 41–50, 51–60, 61–70, 71–80, and >80) using a one-way analysis of variance, followed by the Student-Newman-Keuls (SNK) post hoc test. Bivariate Pearson correlation analysis was also applied to determine the effect of age on the three parameter values. The test was considered to be statistically significant at P < 0.05. The data were collected using spreadsheet software (Excel; Microsoft Corp., Redmond, WA), and all statistical analyses were performed using statistical analysis software (SPSS v. 10; IBM Corp., Armonk, NY) on a personal computer (Windows; Microsoft Corp.). 

Results

A total of 711 eyes of 711 healthy human subjects were enrolled, and all of them met the inclusion criteria. Twenty-one participants used pirenoxine sodium eyedrops for their immature stage cataract, and none used these eyedrops within 2 hours before the OCT imaging. The measurement protocol with the HD-OCT device (Carl Zeiss Meditec, Inc.) was well tolerated by all subjects, and the bulbar conjunctival thickness measurements were successfully obtained from the high quality OCT images. The average age of the 711 participants was 46.5 years (SD, 21.6 years; range, 8–85 years). The male group consisted of 344 subjects, with an average age of 46.2 years (SD, 21.5); and the female group consisted of 367 participants, with an average age of 46.7 years (SD, 21.7). No significant difference was found between the two sex groups regarding age (t = 0.33, P = 0.74). The subject numbers and age groups are shown in Table 1. 

Table 1. 

 

Bulbar Conjunctival Thickness Measurements in 711 Normal Subjects

	Mean	SD	Minimum	Maximum
Epithelial thickness, μm	42.4	7.4	28	76
Stromal thickness, μm	197.7	32.5	88	259
Full thickness, μm	240.1	29.8	140	304

In the lower temporal bulbar conjunctiva OCT images of the 711 normal subjects (Fig. 1), the superior, continuous, narrow, and low reflective layer corresponded to the thin conjunctiva epithelial layer, and the middle high reflective layer, which is separated from the underlying progressively lower reflective layer by a narrow demarcation, corresponded to the conjunctiva stromal layer. ⁴ The bulbar conjunctiva epithelial thickness and conjunctiva stromal thickness were then measured between the different tissue landmarks. The bulbar conjunctiva full-thickness was the sum of the conjunctiva epithelial and stromal thicknesses. 

Figure 1. 

 

Bulbar conjunctival epithelium thickness measurements in different age groups of 711 normal subjects (μm). In the Age Group axis, groups 1 to 8 correspond to ages <20, 21 to 30, 31 to 40, 41 to 50, 51 to 60, 61 to 70, 71 to 80, and >80 years. Bivariate Pearson analysis revealed a correlation coefficient of 0.35, P < 0.01.

Figure 2. 

 

Bulbar conjunctival stromal thickness measurements in different age groups of 711 normal subjects (μm). In the Age Group axis, groups 1 to 8 correspond to ages <20, 21 to 30, 31 to 40, 41 to 50, 51 to 60, 61 to 70, 71 to 80, and >80 years. Bivariate Pearson analysis revealed a correlation coefficient of −0.70, P < 0.01.

The results of the conjunctival thickness measurements are summarized in Tables 1 through 5. No significant differences were found between the male and female groups for any of the parameters (Table 2). The Bivariate Pearson correlation analysis revealed a significant linear correlation between each of the parameter values and age (Figs. 1–3). When we divided the subjects into different age groups, significant differences were found among the eight groups for each of the parameters with the one-way analysis of variance test (Tables 3–5). The Student-Newman-Keuls post hoc test revealed significant differences between the age groups younger than 20 years and 20 to 30 years, and the groups aged 51 to 60 and 61 to 70 years in the epithelial thickness parameter. In the stromal and full thickness parameters, the average value decreased approximately throughout the entire lifetime. 

Figure 3. 

 

Bulbar conjunctival full-thickness measurement results in different age groups of 711 normal subjects (μm). In the Age Group axis, groups 1 to 8 correspond to ages <20, 21 to 30, 31 to 40, 41 to 50, 51 to 60, 61 to 70, 71 to 80, and >80 years. Bivariate Pearson analysis revealed a correlation coefficient of −0.67, P < 0.01.

Table 2. 

 

Bulbar Conjunctival Thickness Measurements by Sex in Groups of 711 Normal Subjects

	Male	Female	t*	P
Epithelial thickness, μm
Range	30–68	28–76
Mean ± SD	42.3 ± 6.9	42.5 ± 7.9	0.30	0.77
Stromal thickness, μm
Range	102–256	88–259
Mean ± SD	198.1 ± 32.3	197.4 ± 32.7	0.30	0.77
Full thickness, μm
Range	140–304	141–303
Mean ± SD	240.46 ± 29.6	239.8 ± 30.1	0.24	0.81

*  Two-tailed Student's t-test.

Table 3. 

 

Bulbar Conjunctiva Epithelial Thickness Measurements in Different Age Groups of 711 Normal Subjects*

Age†	Subject Number	Mean, μm	SD, μm	Minimum, μm	Maximum, μm
<20 ya	93	42.1	4.5	32	52
21–30 yb	99	39.3	3.8	31	51
31–40 yb	100	39.0	2.7	33	48
41–50 yb	95	40.0	4.4	30	52
51–60 yb	97	38.9	6.0	29	60
61–70 yc	104	48.0	9.8	28	76
71–80 yc	102	47.6	9.1	30	68
>80 yc	21	47.8	7.9	36	64

*  One-way ANOVA F = 36.7; P < 0.001.

†  Mean scores of bulbar conjunctiva epithelial thickness followed by the same letter do not differ, but means scores followed by another letter were significantly different between age groups according to SNK post hoc test after the one-way ANOVA test.

Table 4. 

 

Bulbar Conjunctiva Stromal Thickness Measurements in Different Age Groups of 711 Normal Subjects*

Age†	Subject Number	Mean, μm	SD, μm	Minimum, μm	Maximum, μm
<20 ya	93	224.3	12.4	188	259
21–30 yb	99	218.5	6.3	202	234
31–40 yb	100	213.6	5.1	200	228
41–50 yb	95	212.4	13.8	170	232
51–60 yc	97	198.6	24.1	120	256
61–70 yd	104	167.8	36.1	88	244
71–80 ye	102	162.0	30.4	100	224
>80 ye	21	157.5	28.5	108	204

*  One-way ANOVA F = 122.7; P < 0.001.

†  Mean scores of bulbar conjunctiva stromal thickness followed by the same letter do not differ, but mean scores followed by another letter were significantly different between age groups according to SNK post hoc test after the one-way ANOVA test.

Table 5. 

 

Bulbar Conjunctiva Full-Thickness Measurements in Different Age Groups of 711 Normal Subjects*

Age†	Subject Number	Mean, μm	SD, μm	Minimum, μm	Maximum, μm
<20 ya	93	266.4	13.6	228	303
21–30 yb	99	257.8	6.6	241	277
31–40 yb	100	252.6	5.3	240	268
41–50 yb	95	252.4	15.4	206	276
51–60 yc	97	237.5	23.8	176	304
61–70 yd	104	215.8	34.5	140	297
71–80 ye	102	209.7	29.0	142	276
>80 ye	21	205.3	28.1	156	245

*  One-way ANOVA F = 95.3; P < 0.001.

†  Mean scores of bulbar conjunctiva full-thickness followed by the same letter do not differ, but means scores followed by another letter were significantly different between age groups according to SNK post hoc test after the one-way ANOVA test.

Discussion

In the report by Feng and Simpson, the mean bulbar conjunctiva epithelial thickness in 13 healthy subjects was 44.9 ± 3.4 μm. ² Francoz and associates described a mean bulbar conjunctiva epithelial thickness of 42.0 ± 7.5 μm in their group of individuals aged 20 to 39 years and of 42.2 ± 7.9 μm in their group aged 41 to 66 years. ³ These values were all slightly higher than our measured results. We propose the reason for this difference is the systematic error of measurement results in the two different study groups: their subjects were from a Canadian population sample aged much younger than the participants in our present study were. 

As hypothesized, the three conjunctival thickness parameters did not differ between sexes. However, a previously unreported conjunctiva epithelial thickness variation curve was revealed from the present data: the thickness value decreased significantly after 20 years of age, was maintained at a relatively low level, and then increased sharply after 60 years of age. Because the number of subjects in the different sex and age groups all exceeded 90, with the exception of the eldest group, we considered the sample size of this study reliable in terms of the age comparisons of the individuals younger than 80 years. The bulbar conjunctiva epithelium is well recognized to consist of multiple layers of cylindrical epithelial cells. The superficial and intermediate epithelial cell layers contain round or oval mucus-secreting goblet cells. ⁶ In 1978, Abedl-Khalek and associates examined 49 biopsies of the bulbar conjunctiva by light and transmission electron microscopy from patients with no apparent conjunctival disease. ⁷ They found no significant morphological changes in the specimens from 33 patients between the ages of 50 to 79 years. However, in the 16 persons over the age of 80, they found mild conjunctiva epithelial atrophy in some biopsies and apparent hyperplasia in others, which resulted in variations in the epithelial thickness. They also found the presence of abnormal “hyaline bodies” in 25% of the elderly samples, which may be due to the intracytoplasmic degeneration of goblet cells in the epithelium. They attributed these morphological changes in the conjunctiva epithelial cell to a response to changes in their environment, such as a reduction in the tear quantity or quality in the elderly. ⁷ In other reports, alterations in the tear film, neuronal innervation, and reflex blinking were demonstrated to result in the abnormal proliferation and differentiation of the conjunctival epithelium in dry eyes. ^8,9 Therefore, due to increased aggravation by environmental pollutants in recent years, we speculate that the rising trend of conjunctiva epithelial thickness, which occurred as early as age 60 in the present study, was also an age-related alteration due to epithelial cell hyperplasia and an augmentation of hyaline bodies after an increase in desiccating stress. In the population aged between 20 and 60 years, the relatively stable bulbar conjunctiva epithelial thickness shown in the present study, which corresponds to previous morphological observations, ⁶ is consistent with the findings of Francoz and associates. ³ Finally, the physiological mechanisms underlying an abrupt change in the conjunctiva epithelial thickness at age 20 still remains poorly understood. In the future, more detailed morphological studies on the conjunctival epithelium in teenagers may help provide clear explanations. 

In our previous study, only 18 normal conjunctivas were examined in vivo, ⁴ which was insufficient to document age changes in the conjunctiva stromal thickness. As such, to our knowledge, this is the first reported study to reveal conjunctiva stromal thickness changes with advancing age in healthy subjects. The conjunctiva stromal thickness decreased to an almost significant degree throughout the entire lifetime, resulting in a full-thickness decrease from the teenage period to the elderly period. We speculate that subtle inflammation may produce this downtrend. In the report by Abdel-Khalek and associates, more inflammatory cells, lymphocytes, plasma cells, and polymorphonuclear leukocytes were identified in the biopsies of conjunctival stroma from the healthy elderly group than the younger group. ⁷ In some types of conjunctival inflammation and conjunctivochalasis, the overexpression of matrix metalloproteinases was found, correlating with their increased proteolytic activities. ^10,11 As the activity of matrix metalloproteinases gradually surpasses that of their tissue inhibitors, excessive degradation of the conjunctival matrix occurs, thereby causing the conjunctiva stromal thickness to decrease. However, as no direct evidence between the conjunctival extracellular matrix degradation and proteolytic enzyme activity was previously observed in healthy conjunctiva biopsies, the mechanism of conjunctiva stromal “atrophy” is still speculative and inspires much debate. 

This study suffers from several specific limitations. First, as emphasized in our previous study, ⁴ the ruler we used in the thickness measurements is calibrated for measuring corneal tissue, and inherent systematic bias in our conjunctival thickness values was inevitable. Nevertheless, there are no reported data describing the refractive index of the conjunctival tissue. It is therefore impractical to adjust the obtained values in the equation, and we cannot verify that the optical thickness measured by the OCT was correctly converted to the physical conjunctival thickness. For this reason, these measurement data should be considered to be preliminary and subject to further refinement to improve accuracy. Second, we chose the temporal region conjunctiva for the thickness measurements, which may not represent the data for the entire conjunctiva. Third, although we excluded subjects with previous ocular surgery, ocular surface diseases, history of trauma or glaucoma, lid abnormalities, or abnormal ocular surface findings on slit-lamp and funduscopic examinations, 21 participants who were using pirenoxine sodium eyedrops for their cataracts were enrolled. Because the number of these participants was quite small and no previously published studies have examined the effects of pirenoxine sodium eyedrops on the conjunctival thickness, we considered the usage of this eyedrop to barely have an impact on our study results. However, we could not exclude subjects who ever used drops (such as steroids) for other ocular diseases, especially in the elderly subjects. For example, a 70-year-old male may have once tried steroids in his 20s. These eyedrops may have affected the conjunctival structure, and future studies on the effects of eyedrops on different conjunctival thicknesses with age-matched controls who never used eyedrops will help explain the increased thickness of the conjunctiva in subjects aged older than 60 years. Nevertheless, we established a conjunctival thickness database for healthy Chinese subjects, which may further facilitate the earlier detection of conjunctival disease progression before the onset of functional changes and may aid in the evaluation of the treatments for conjunctival diseases. 

Acknowledgments

Supported by the Putuo Technology Committee Fund for Independent Innovative Studies (2012PTKW B-126). The authors alone are responsible for the content and writing of the paper. 

Disclosure: X. Zhang, None; Q. Li, None; M. Xiang, None; H. Zou, None; B. Liu, None; H. Zhou, None; Z. Han, None; Z. Fu, None; Z. Zhang, None; H. Wang, None