CXL patient selection enhanced by using routinely obtained additional data
A new keratoconus progression scoring system called the Dutch Crosslinking for Keratoconus (DUCK) score may identify those who do not require corneal cross-linking (CXL) treatment more accurately than conventional scoring based on maximum change in keratometry alone, according to a study carried out at two centres in
The DUCK score is based on five routinely assessed clinical variables. The prospective cohort study, published in JAMA Ophthalmology (Wisse et al, doi:10.1001/jamaophthalmol.2019.0415), showed that adhering to the new scoring system rather than maximum keratometry, was associated with a 23% reduction in the overall rate of CXL treatment, without increasing the risk of disease progression.
“A proper selection of CXL candidates reduces unnecessary burden and risks to the patient. I think this is the most important message, since the procedure is inevitably painful and carries around a 1-to-5% risk of secondary infection. CXL can also bring about unwanted changes in manifest refraction that can profoundly affect uncorrected visual acuity. That is especially relevant for mild cases of progressive keratoconus who are not dependent on visual aids,” Robert P. L. Wisse MD, PhD, University Medical Centre Utrecht, Utrecht, the Netherlands, lead author of the study, told EuroTimes in an interview.
He added that research he and his associates conducted in 2015 showed that the mean total costs for one eye treated with CXL were €1,754.06 or $1,929.47.
Unnecessary treatments reduced
The study involved a total of 504 eyes of 388 patients referred for keratoconus between January 1, 2012, and June 30, 2014, to two academic treatment centres
The patients included a longitudinal discovery cohort of patients referred to the University Medical Center Utrecht, and a validation cohort referred to the Maastricht University Medical Center. All eyes had data available on disease progression at 12 and 24 months of follow-up. Analysis began in March 2017.
Baseline patient characteristics of the discovery cohort and the validation cohort were comparable in terms of mean age (26.8 years vs 26.3 years), sex (65% vs 72% men) and maximum keratometry (53.5D vs 52.7D).
Dr Wisse and his associates compared the DUCK scores to the conventional criterion of a 1.0D increase in maximum keratometry occurring within the previous 12 months. They found that if the decision to treat had been based on increase in maximum keratometry, 180 eyes (36%) would undergo CXL. That compared to only 138 eyes (27%) when using the DUCK scoring system.
DUCK score requires only routine examination
The DUCK score is based on routinely assessed clinical parameters; namely, age, visual acuity, refraction error, keratometry and subjective patient experience. Each criterion is scored from 0 to 2 points. An overall score of 5 or more is an indication for CXL.
“I think the beauty of the DUCK score is in the fact that it is device independent, and as such accessible to many eye care practitioners. New tech is often expensive, and only available for specialised centres,” Dr Wisse said.
Dr Wisse and his associates considered changes to uncorrected distance visual acuity (UDVA) the most sensitive parameter for visual function, since unreported adjustments to spectacle or contact lens prescriptions will not affect the measurements.
In the DUCK scoring system, a loss of less than one line of UDVA counted as zero points, a change of one or two lines counts as one point and a change of two lines counts as two points.
Age was an important parameter, since younger patients tend to have more progressive disease and conversely there tends to be less progression in older patients. Age younger than 18 years counts as two points, 18-to-35 years 1 point and older than 35 years counts for 0 points.
Regarding maximum keratometry, a change of less than 1.0D counts as zero points in the DUCK scoring system, a change of 1-to-2D counts as one point and a change of more than 2.0D counts as two points. The same changes in spherical equivalent are accorded the same score as with changes in keratometry.
Visual quality is based on the patient’s subjective assessment of their visual experience. No impact on daily life counts as zero points, mild impact counts as 1 point and severe impact counts as 2 points.
Some false negatives
Dr Wisse noted that in 95 eyes, keratoconus progressed despite fitting neither the maximum keratometry criterion nor the DUCK score for CXL treatment. The addition of other risk factors may yield fewer false negatives, he said. For example, research is yielding insights into the way that inflammatory activation affects the ocular surface, opening a promising new diagnostic avenue.
A structural assessment of ocular inflammation, enriched with biomarkers, should make the identification of suitable patients even better.
He noted that Rohit Shetty and his associates have showed that proper control of extraocular inflammation can reduce keratoconus progression. In the current study, the researchers assumed inflammation is controlled, in accordance with the Dutch centres’ clinical protocol.
“Corneal thickness is another factor that might be considered; from a biological standpoint, the tissue thinning is the archetypal result of keratoconus. However, it has no direct effect on visual acuity and its measurement is device-dependent. Moreover, corneal thickness can vary, based on prior contact lens use, for instance. Maybe a future validation stud y can answer this question,” Dr Wisse added.
Robert P. L. Wisse: R.P.L.Wisse@umcutrecht.nl