Background:
Keratoconus is a progressive degenerative
disorder of cornea in which structural changes in the
cornea cause it to become thin and conical in shape.
There are no gold standard criteria for diagnosing
Keratoconus. KISA% is one of the popular criteria for
diagnosing Keratoconus. The new criteria were
reported in the year 2015, using Kmax
2
and TP
4
to
diagnose Keratoconus. The sensitivity and specificity of
this Kmax2
/TP were 99.5% and 95.7% respectively.
Materials and Methods:
In this retrospective study, we collected
information of 418 eyes (209 Keratoconus +209 ages
matched controls). The necessary data was collected
from the Visante AS OCT and Zeiss atlas topography
in a tertiary eye hospital, Udupi. The demographic data
of the patients was drawn from the system. . Further
details on visual acuity, refractive error, slit lamp
findings and fundus examinations were obtained from
the MRD. All parameters included in the study were
taken from the last 3 year data available in the
instruments. Keratoconus diagnosis had been
confirmed based on clinical and topographical findings.
Results:
The parameters were entered in the IBM SPSS
statistics version 24 for the analysis. KISA %[( K*I-
S*AST*SRAX*)*100/300] and Kmax
2
/T were calculated
and values are obtained. With these values, kappa
analysis was done to find the agreement between the
two criteria for diagnosing Keratoconus. It is found
that the r= 0.807 (p<0.001). This showed that the two
criteria have good agreement for Keratoconus
diagnosis.
All eyes were categorized under Amsler Krumeich
and CLEK’s classification systems. Interclass
Correlation statistics was used to find the level of
agreement between Amsler Krumeich and CLEK
classification system in diagnosing Keratoconus
Conclusion:
KISA% and Kmax2 /TP showed good agreement
between them in diagnosing Keratoconus. Hence both
criteria can be used interchangeably for the diagnosis
of Keratoconus. Amsler Krumeich and CLEK
classification system showed a good level of agreement
for Keratoconus diagnosis.
