Choosing the specular microscope that fits your clinic
Assessing two devices in a real-world setting
Specular microscopy is a valuable in-clinic tool for patients with corneal endothelial disease or who are undergoing corneal examination prior to other surgeries. Research data published earlier this year could help clinicians gain a better understanding of what to look for in a specular microscope, and how to interpret inter- and intra-device repeatability results in a way that benefits their existing practice framework.
Ivo Guber, MD, FMH, FEBO, FICO, is a consultant ophthalmic surgeon at the University of Geneva, Switzerland. He is also the owner and medical director at Augenchirurgie.ch in Winterthur, Switzerland. Dr Guber and colleagues published an article earlier this year in Medicina1 which provided real-life evidence of inter-device variances in specular microscopy. These data were also the subject of a prior lecture2 delivered by Dr Guber, first at the South African Society of Cataract and Refractive Surgery in 2022.
Importance of the specular microscope
The investigators’ findings brought attention to two specular microscopy measurements, corneal endothelial cell density (ECD) and central corneal thickness (CCT). Before choosing a specular microscope for their practice, a clinician should investigate the accuracy of the data produced by the machine. Reliable and reproducible data are paramount, and because each device is unique, practitioners should perform research to assess whether the types of data rendered by a device will fit their clinical needs. Compatibility and comparability with pre-existing devices in the clinic are also important. Dr Guber also encouraged clinicians to consider whether they have specific handling needs. Easy-to-use controls and quickly legible displays are important and may require more serious consideration if the physician plans to delegate tasks to trainees, nurses or other staff members.
When clinicians are choosing a specular microscope, financial aspects of the decision deserve particular attention. The four devices showcased in Dr Guber’s presentation ranged from approximately €19,800 to €29,100, and he highlighted the upfront price and amortization period could both be detrimental to smaller clinics. Dr Guber used his own clinic as an example: based on the price of the specular microscope used, it took approximately 1 year, or 283 individual specular microscopy measurements, to offset the cost of the machine.2 Any practitioner should assess whether these investments will adequately serve the needs of a clinic’s patient base.
Two devices: Nidek CEM-530 and Tomey EM-4000
Study investigators assessed two commercially available specular microscopes, the CEM-530 (Nidek, Japan) and the EM-4000 (Tomey, Germany), in a real-life clinical setting. These two devices are comparable in price; Dr Guber reported the Tomey EM-4000 priced at €19,800 and the Nidek CEM-530 at €20,800.2 Both devices perform non-contact specular microscopy and have integrated non-contact pachymetry. The Nidek device has a manual alignment; endothelial measurement is self-triggered. The Tomey device has central auto-alignment and automatic analysis of the endothelium.
Figure 1. Intradevice cell density and central corneal thickness comparisons.
Patients presented for ophthalmic examination prior to planned cataract surgery. Investigators included 112 eyes of 56 patients (mean age, 61.05 years [range, 22-85]; 24 female, 32 male). Each eye was measured 3 times with each device, and staff recorded ECD and CCT values for each measurement.
Masked researchers, unrelated to the clinic, statistically analysed the results using a Friedman test, ANOVA, a Wilcoxon signed-rank test, a Mann-Whitney test and a paired t-test, according to sample distribution determined by a D’Agostino-Pearson normality test.
Intra-device repeatability values
According to the authors, both specular microscopes had consistent intra-device results (Figure 1). Dr Guber and colleagues wrote, “Tomey EM-4000 measured an ECD of 2390 ± 49.57 cells/mm2 (range 799-3010 cells/mm2) and a CCT of 546 ± 5.104 μm (range 425-615 μm), while the Nidek CEM-530 measured an ECD of 2417 ± 0.09 cells/mm2 (range 505-3461 cells/mm2) and a mean CCT of 546.3 ± 4.937 μm (range 431-621 μm).”
Figure 2. Interdevice comparisons of the Nidek CEM-530 and Tomey EM-4000. (All data1 courtesy of Kecik M, Kropp M, Thumann G, Pajic B, Guber J, Guber I.)
The intra-device measurements were in statistically significant accordance for each eye (Tomey, OD vs OG, P = .0796; Nidek, OD vs OG, P = .9910; Tomey measurement: 1-3, P = .7972; Nidek measurement: 1-3, P = .6207).
Inter-device differences and theories
The ECD values registered on the Nidek device were slightly higher than those registered on the Tomey (Figure 2a); the CCT (Figure 2b) values also had some inter-device discrepancies. “Inter-device variability in terms of endothelial cell count and central corneal thickness showed a reduced endothelial cell count with the Tomey compared to the Nidek device with an average of 121.2 cells (4.87%) less (* P = .0175),” the authors wrote.2 “Measured central corneal thickness was lower in the Tomey with a difference of 16.75 μm (3.07%) (* P = .0125).”
Guber and colleagues theorised that a number of factors could account for the Nidek’s higher ECD values registered throughout the study. The difference could be accounted for by a slightly larger endothelial capture field. The photographic range on the Tomey specular microscope is 0.25 mm by 0.54 mm, and the range on the Nidek is 0.25 mm by 0.55 mm. The study investigators wrote, “As the average diameter of a human corneal endothelial cell is 20 μm or 0.02 mm, [the Nidek device] is able to include more cells in the photograph for analysis.” The devices may also use different methods to calculate cells at the edge of the imaging frame, and automated algorithms may include or exclude “edge” cells differently between devices.
Additionally, the capture field of a specular microscope is small in comparison to the total corneal endothelium. Specular microscopes may have automatic centering, but when taking repeated measurements, the photograph will not be microscopically aligned with previous images. Cell density variations drive a need for repeated imaging of both central and peripheral corneal zones: it is recommended clinicians image at least 20% of the endothelial cell surface, which commercially-available specular microscopy devices can only achieve with repeated imaging. “The difference between both devices is relatively small (4.87%), and an argument can be made that this bears little clinical significance,” the study’s authors emphasised. “Indeed, a study on corneal donor tissues proposed 5% as a threshold of clinical significance for endothelial cell count.”
Conclusion
Both specular microscopes provided reliable ECD and CCT measurements in a clinical setting, and Guber described the intra-device repeatability for both machines as “excellent.” However, the inter-device variability was statistically significant, and the Nidek device provided higher values for the ECD count in the cases presented. The inter-device comparability is clinically acceptable. Dr Guber advised physicians to take note when comparing results between devices, especially in larger clinics or hospital centres, or in multicentre studies pertaining to corneal endothelial tissues.
References
1. Kecik M, Kropp M, Thumann G, Pajic B, Guber J, Guber I. Intradevice repeatability and interdevice comparison of two specular microscopy devices in a real-life setting: Tomey EM-4000 and Nidek CEM-530. Medicina. 2024; 60(7):1110.doi:10.3390/medicina60071110
2. Guber I. Finding the best specular microscope for your clinic. Lecture presented at: South African Society of Cataract and Refractive Surgery; November 3-5 2022; Cape Town, South Africa.
Ivo Guber, MD, FMH, FEBO, FICO | E: ivo.guber@augenchirurgie.ch
Guber is a specialist in cornea, cataract and refractive surgery and the leading eye surgeon at Augenchirurgie.ch eye centres in Switzerland. He is also a senior consultant at the University Hospital of Geneva, Switzerland.
