Cristina Nicula ¹ Karin Ursula Horvath ² Ariadna Patricia Nicula ³ Adriana Elena Bulboacă ⁴ Sorana D. Bolboaca ^5* Dorin Nicula ⁶

• ¹ Department I-Medical Specialties, „Iuliu Hațieganu” University of Medicine and Pharmacy, Cluj-Napoca, Romania
• ² Department of Ophthalmology, “George Emil Palade” University of Medicine, Pharmacy, Sciences and Technology, Târgu Mureş, Romania, Târgu Mureş, Romania
• ³ Emergency County Hospital Targu Mures, Târgu-Mureş, Romania
• ⁴ Department of Physiopathology, “Iuliu Hațieganu” Medicine and Pharmacy University, Cluj-Napoca, Romania, Cluj-Napoca, Romania
• ⁵ Department of Medical Informatics and Biostatistics, Iuliu Hatiegan University of Medicine and Pharmacy, Cluj-Napoca, Romania
• ⁶ Oculens Clinic, Cluj-Napoca, Romania

Purpose: The study aimed to determine the stability of topographic and tomographic indices measured with Pentacam and to evaluate the biomechanical parameters measured with Corvis ST in the diagnosis of subclinical (sKCN) and clinical keratoconus (KCN). Methods: This is a single-center cohort study with a retrospective review of topographic and tomographic indices and biomechanical parameters on adult patients with subclinical keratoconus KCN (sKCN), clinical keratoconus (KCN), and healthy subjects (control group). The area under the receiver operating curve (AUC) was used to identify the cut-off values for evaluated indices able to distinguish between subjects with sKCN and those with KCN. Results: Seventy-six patients (76 eyes) in the sKCN group, 74 patients (132 eyes) in the KCN group, and 70 patients (140 eyes) in the control group were analyzed. Evaluated participants had similar age but with male predominance in the sKCN group (p=0.0070). Significantly higher values in the KCN group of Front Kmax, ISV, IVA, KI, IHD, BAD_D, and lower values of TL and PRC (with excellent accuracy AUC>0.9) were observed in the differentiation of KCN by controls. Similarly, excellent accuracies were obtained by Front Kmax, ISV, IVA, KI, IHD, KISA, I-S, BAD_D, and RMS total with higher values in the KCN group and PRC and ARTmax with lower values in patients with KCN as compared to those with sKCN. Only FrontKmean (AUC=0.946, Se=85.6%, Sp=90.4%, p<0.0001) and I-S (AUC=0.96, Se=84.1%, Sp=97.3%, p<0.0001) proved accurate and not shared with differentiation of sKCN or KCN by normal eyes. FrontKmean proved good for case-findings (0.806 [0.742 to 0.871]) and screening (0.712 [0.645 to 0.778]). I-S performed excellent for case-findings (0.826 [0.764 to 0.888]) and good for screening (0.758 [0.700 to 0.817]). Conclusions: Subclinical and clinical KCN shared common Pentacam parameters with excellent or good accuracy in distinguishing subjects with and without pathology, but FrontKmean and I-S Pentacam proved excellent or good for case-finding and screening and are not shared with differentiation of the sKCN or KCN by the normal eyes. Furthermore, differentiation of sKCN by normal eyes could be done with KISA and CBI parameters, but only CBI is not shared with KCN.