Article Text

Download PDFPDF

Panoramic view of human corneal endothelial cell layer observed by a prototype slit-scanning wide-field contact specular microscope
  1. Hiroshi Tanaka1,
  2. Naoki Okumura2,
  3. Noriko Koizumi2,
  4. Chie Sotozono1,
  5. Yasuhiro Sumii3,
  6. Shigeru Kinoshita4
  1. 1Department of Ophthalmology, Kyoto Prefectural University of Medicine, Graduate School of Medicine, Kyoto, Japan
  2. 2Department of Biomedical Engineering, Faculty of Life and Medical Sciences, Doshisha University, Kyotanabe, Japan
  3. 3Konan Medical, Inc, Nishinomiya, Japan
  4. 4Department of Frontier Medical Science and Technology for Ophthalmology, Kyoto Prefectural University of Medicine, Kyoto, Japan
  1. Correspondence to Professor Shigeru Kinoshita, Department of Frontier Medical Science and Technology for Ophthalmology, Kyoto Prefectural University of Medicine, 465 Kajii-cho, Hirokoji-agaru, Kawaramachi-dori, Kamigyo-ku, Kyoto 602-0841, Japan; shigeruk{at}


Purpose To observe the most peripheral region of the corneal endothelial cell (CEC) layer as long as optically recordable by use of a prototype slit-scanning wide-field contact specular microscope and produce a panoramic image to evaluate the variation of CEC density with ageing.

Design Observational case series study.

Methods This study involved 15 eyes of 15 normal healthy subjects divided into three groups according to age: A (20–40 years), B (41–60 years) and C (>60 years). The corneal endothelial layer of each eye was recorded in a horizontal direction, from nasal to temporal, with a slit-scanning wide-field contact specular microscope (Konan) and endothelial cell density (ECD) in three specific regions (central, mid-peripheral, and peripheral) was automatically calculated via built-in analysis software.

Results Corneal endothelial images from near the surgical limbus to limbus in all eyes were clearly recorded and panoramic images were made by combining still images. ECD in groups A, B and C were 2809±186, 2717±91 and 2580±129 cells/mm2 at the centre, 2902±242, 2772±97 and 2604±187 cells/mm2 at the mid-periphery and 2893±308, 2691±99 and 2533±112 cells/mm2 at the periphery. Significance differences in ECD was found between groups A and C in all regions and groups between B and C at mid-peripheral region.

Conclusions A prototype slit-scanning wide-field contact specular microscope enabled us to record the endothelial layer from the surgical limbus to limbus of the cornea and compare specific areas among subjects, and showed that ECD in each region of the cornea decreases with ageing.

Trial registration number UMIN000021264, Results.

  • Cornea
  • Imaging
  • Diagnostic tests/Investigation
View Full Text

Statistics from


It is well known that corneal endothelial morphometric analysis in vivo plays an important role in understanding the condition and dynamics of corneal endothelial cells (CECs).1 ,2 Although that analysis can be calculated via contact specular microscopy, non-contact specular microscopy and confocal microscopy, the area of analysis is much more limited; that is, an area approximately 3 mm from the centre of the cornea. Hence, analysis of a wider area of CEC layer in vivo allows for a deeper understanding of the cell dynamics.

Since 1968, several contact and non-contact clinical specular microscopes have been developed to observe and obtain images of human CEC layer at high magnification in vivo.3 ,4 Qualitative and quantitative analyses of CECs have provided information regarding changes in cell density, cell size, coefficient of variation (CV) of area and cell morphology (ie, the rate of hexagonal shape cells (6A) with ageing). Although a number of reports have investigated the difference of endothelial cell density (ECD) between the centre and peripheral regions of the cornea using a contact and non-contact specular microscope in vivo5–8 and histology in vitro,9–11 to the best of our knowledge, this is the first report to evaluate corneal ECD near the surgical limbus by comparing with in vivo wide-field CEC layer images.

In this study, investigation of the most peripheral CECs in vivo was performed via the use of a prototype slit-scanning wide-field contact specular microscope that is able to provide live imaging of CEC layer and location data. The purpose of this study was to elucidate the CEC dynamics over a wide area in vivo by comparing the ECD, 6A and CV with ageing via slit-scanning wide-field contact specular microscopy.



Approval for this study was obtained from the Ethics Committee of Kyoto Prefectural University of Medicine, Kyoto, Japan, and the study protocol followed the tenets set forth in the Declaration of Helsinki. Written informed consent was obtained from all subjects prior to their involvement in the study.

Volunteers from the local community were recruited from July 2015 to October 2015. In order to insure that normal healthy subjects were selected, each volunteer received a complete ophthalmologic evaluation via slit-lamp biomicroscopy, as well as intraocular pressure, refraction, best spectacle-corrected visual acuity examination. Exclusion criteria included a history of contact lens wear, prior ocular trauma or surgery, or ocular or systemic diseases that may affect the cornea.

In total, this study included 15 normal subjects (six males and nine females; age range 25–77 years) who were divided into the following three groups according to age: group A (20–40 years old), group B (41–60 years old) and group C (>60 years old).

Instrumentation used for examination

In this study, examination of each subject was performed using a prototype KSSP slit-scanning wide-field contact specular microscope (Konan Medical, Nishinomiya, Hyogo, Japan) with a 1/2-inch camera including a 400 000-pixel charge-coupled-device sensor with 9.9× magnification, and a 40× magnification contact cone lens made from fluorite with numerous slits on both sides (figure 1A). To record the corneal endothelial layer, the layer is illuminated through the slits on one side of the device and the endothelial cell images are reflected back through the slits on the opposite side. Briefly, oxybuprocaine 0.4% eye drops were used to anaesthetise the right eye of each subject. The subjects were then asked to concentrate on the fixed light with their left eye, and manually focused on the CEC layer images at the rate of 60-frames per second were recorded with the slit-scan technique while the right eye was swept in the horizontal direction from the nasal to temporal side with the cornea being flattened by the contact lens. Focused still images were then extracted from the MPEG-2 file and manually connected together to produce a panoramic image. Overlapping portions of the images, judged from the morphology of CECs or dark band from the Descemet fold, are combined to form the panorama image (figure 1B). The centre and peripheral points were detected from the panoramic images, and the mid-point between the centre and periphery was defined as the ‘mid-peripheral’ point. Then, the ECD, 6A and CV of each point in a 0.3 mm×0.3 mm area were calculated via the computer algorithm (Konan Medical).

Figure 1

Images showing the prototype slit-scanning wide-field contact specular microscope (left) and proper measurement method (right) (A). A panoramic image produced by combining still images (0.5 mm×0.7 mm) from the recorded endothelial cell layer (B).

Statistical analysis

Statistical analysis of the data was performed and basic descriptive statistics were calculated on all of the gathered data, with the values reported as mean±SD. Differences in each parameter across age groups were tested by Friedman test. Comparisons between groups were conducted via the Steel-Dwass test. A p value of <0.05 was considered statistically significant.


Continuous CEC layer images, from end-to-end, in the cornea in all subjects were able to be clearly recorded by use of the slit-scanning wide-field contact specular microscope and were able to be connected into panoramic images (figure 2). Six specific areas of the cornea were identified from the panoramic images and calculated by the computer algorithm (figure 3). The slit-scanning wide-field contact specular microscopy results are summarised in the table 1. Our results showed a mean±SD corneal ECD as follows: group A: 2809±186 cells/mm2 at the centre of the cornea, 2902±248 cells/mm2 at the mid-peripheral region (the mid-point between the centre and peripheral region) and 2893±308 cells/mm2 at the peripheral region; group B: 2717±91 cells/mm2 at the centre of the cornea, 2772±92 cells/mm2 at the mid-peripheral region and 2691±99 cells/mm2 at the peripheral region; and group C: 2580±129 cells/mm2 at the centre of the cornea, 2604±184 cells/mm2 at the mid-peripheral region, and 2533±112 cells/mm2 at the peripheral region. The correlation between ECDs versus age is represented by scatter plots (figure 4).

Table 1

Mean endothelial cell density (ECD, cells/mm2), the rate of hexagonal shape cells (6A) and coefficient of variation (CV) of area in each region in the examined groups

Figure 2

Typical example images including the anterior segment of the eye (left) and endothelial cell layer images (right) of the subjects in the three age groups.

Figure 3

Images showing the measurement method of corneal endothelial cell density in the centre, mid-peripheral and peripheral regions.

Figure 4

Scatter plot of the centre (A), mid-peripheral (B) and peripheral (C) region corneal endothelial cell density versus age.

In each group, ECD at the mid peripheral region had a tendency to be higher than that at the centre and peripheral regions, however, there was no significance (p>0.05). The age comparison results showed significance in ECD between groups A and C at the centre (p<0.05), the mid-peripheral region (p<0.05), and the peripheral region (p<0.05) and groups B and C at the mid-peripheral region (p<0.05). Significance was found in CV between groups A and C (p<0.01) and group B and C (p<0.05) at the centre. There was no significant difference in the rate of hexagonal shape (6A) was found between the regions in comparison with age.


In this study, a new slit-scanning wide-field contact specular microscope was used to successfully obtain a panoramic image of the optically recordable area of the corneal endothelium cell layer from near the surgical limbus to limbus in vivo in healthy subjects over a broad range of ages. Using the obtained results, we were able to identify specific areas of analysis and compare ECD between generations. The results showed a difference of ECD in each region of the cornea in relation to age in normal healthy subjects.

Use of the new wide-field contact specular microscope enabled us to observe corneal endothelium, from end-to-end, with ease, as the unique functions of this new instrument allow clinicians to do the following: (1) to record and observe wide-field (0.7 mm×0.5 mm) pictures of CEC layer in real time, thus preventing clinicians from overlooking a local abnormality, (2) to record continuous images of a CEC layer, even in thick corneas, at the edge by manually focusing on the layer of corneal endothelium and by only fixating on one point without depending on patient eye movement, (3) to record sharper images than can be obtained via the use of conventional contact or non-contact specular microscopes, because slit-scanning decrease the corneal scattering. With conventional specular microscopes, the condition of corneal endothelium across the entire cornea cannot be calculated, but only predicted, from the information of several regions depending on the target location. In fact, conventional non-contact specular microscopy allows a clinician to evaluate only the mid-peripheral region, whereas in vitro studies allow for examination of the cell-dense peripheral region close to the Schwalbe's line.5 ,11 In contrast to the conventional instruments, the new wide-field contact specular microscopy allows for observation of the corneal endothelium in the more peripheral region in real time, as long as the cornea is transparent. Moreover, a focal change and anomaly of corneal endothelium can be detected on the monitor. In terms of reproducibility when using this device, we previously checked the intra-observer and inter-observer variability (data not shown).

It is vital to be able to monitor corneal endothelial conditions over an extensive area of the cornea post intraocular surgery. For example, the primary postoperative complication seen in patients post angle-supported or iris-fixated phakic intraocular lens (IOL) implantation is endothelial cell loss,12–14 and a primary concern post corneal transplantations such as Descemet's stripping automated endothelial keratoplasty (DSAEK)15 and Descemet's membrane endothelial keratoplasty (DMEK)16 ,17 is being able to know the dynamics of the transplanted corneal endothelium. As the findings of this study show, as long as the cornea is transparent, wide-field contact specular microscopy allows for the corneal endothelial layer to be observed post these treatments in both wide and specific areas.

One of the greatest improvements that this instrument offers is the ability to identify the area being measured. Although still images at several points on the cornea can be obtained by non-contact specular microscopy, those images lack specific reproducible locations. Using slit-scanning wide-field contact microscope, clinicians can detect the relative location to compare the adjacent cells and Descemet's membrane folds.

It should be noted that one of the limitations of this study was the small number of subjects examined. In this study, the difference of ECD at the centre of the cornea per year of age was 0.21%, similar to that in previous reports.18 ,19 No significant difference of ECD in relation to ageing was found at the mid-peripheral area, yet a significant difference was found at the centre and peripheral regions. Anna and associates reported that ECD in the central and peripheral regions has difference with ageing, and that ECD in only older adults showed a significant difference between the central and peripheral areas.6 Taken together, these results suggest that compared with the central and mid-peripheral areas, endothelial cell loss in the peripheral area is most affected by ageing. All of the subjects involved in this study had no corneal epithelial damage, and it should be noted that when using a contact specular microscope, it is important to be careful to not inflict any ocular surface damage to the patient. Moreover, care should be taken with patients in whom the corneal epithelial layer can easily be detached due to endothelial dysfunction. In such cases, it might be better for the patient to wear a contact lens while undergoing examination by wide-field contact specular microscopy.

In conclusion, the prototype slit-scanning wide-field specular microscope used in this study allowed for a successful topographical distribution analysis of CECs in normal healthy subjects and for a comparison via a panoramic image. Furthermore, slit-scanning wide-field contact specular microscopy allows for observation of CECs post intraocular surgery such as phakic IOL implantation and CEC transplantation (eg, DSAEK or DMEK) to reveal the dynamics of CECs in vivo.


The authors wish to thank Mr John Bush of Kyoto Prefectural University of Medicine for his excellent and thorough review of the manuscript.


View Abstract


  • Contributors Design of the study (HT, NO, NK, CS, YS and SK), Acquisition, analysis or interpretation of data (HT), Drafting of the work (HT), and Revising the work (NO, NK, CS, YS and SK), Approval of the final version (HT, NO, NK, CS, YS and SK), Agreement for all aspects of the work (HT, NO, NK, CS, YS and SK).

  • Competing interests YS is an employee of Konan Medical, Inc.

  • Patient consent Obtained.

  • Ethics approval Kyoto Prefectural University of Medicine.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.