Prognostic factors for the progression of keratoconus.

PURPOSE: The progression of keratoconus to a stage where penetrating keratoplasty (PK) is required for visual rehabilitation has considerable implications for affected patients. To assist with counselling, the authors have attempted to identify which factors measurable early in the course of the disease may indicate the likelihood of subsequent surgery. METHODS: The authors reviewed the records of all patients who attended a single center over a 7-year period for contact lens management of their keratoconus. The influence of clinical variables on the time taken for the worst eye to progress to PK was evaluated by actuarial methods and multivariate analysis. RESULTS: Included in the study were 2723 patients with a mean follow-up for unoperated eyes from the first visit of 4.5 years (range, 3 months to 28 years). Data were available for multivariate analysis in 2363 patients. At the end of the study period, 757 eyes (21.6% of all patients) had been grafted. The number of eyes progressing to PK was independently related to both the maximum and minimum keratometry, a corneal cylinder of more than 1.9 mm, the Snellen acuity, the racial group (P < 0.0001), and the age at presentation (P = 0.0006). Sex, laterality, systemic atopic disease, maternal or paternal age at birth, joint hypermobility, and a family history of keratoconus were not statistically related to outcome. Progression to PK in one eye increased the risk of progression in the contralateral eye (P < 0.0001) and a linear model of disease progression is proposed. CONCLUSIONS: Several clinical variables can be measured in patients at the presentation of keratoconus that influence the probability of a subsequent PK.

Life expectancy in keratoconus.

It is observed that few patients over the age of 60 regularly attend the keratoconus clinic at Moorfields Eye Hospital. The hypothesis that patients with keratoconus have a shorter life expectancy owing to underlying connective tissue related disease was tested. From patient records a sample of 337 keratoconus patients aged at least 40 years by 1991 were identified, of which 279 were living, 13 were deceased, and 45 were untraceable. The mortality rate for keratoconus patients was compared with that of the general population using actuarial English life tables. Results show no significant difference between the general population mortality rate and that of the keratoconus sample even with adjustment for social class. Possible explanations for the non-attendance of older patients are discussed.

Practical Bayesian analysis of a simple logistic regression: predicting corneal transplants.

The Bayesian analysis of a logistic regression model is described using an example of predicting the need for a corneal transplant in keratoconus. Controversy over the use of subjective prior information in Bayesian methods is avoided by a formulation representing negligible prior information. Simple computational procedures are described, and it is argued that the results are more accurate, clearer and make fuller use of the information contained in the data. Analysis of more complex models is considered. In particular, it is argued that classical methods as implemented in the computer package GLIM can be used as approximations to Bayesian methods, particularly at the initial stage of model selection.

Predictors for likelihood of corneal transplantation in keratoconus.

A group of 70 patients who had attended the Keratoconus Clinic at Moorfields Eye Hospital over a 10-year period were analysed retrospectively. At the end of this period 23 eyes had undergone corneal transplantation. This gave an overall probability of 16%, various information obtained at the initial visit was then analysed for a possible relationship with the probability of needing a transplant within ten years. The only significant relationship found was the initial corneal radius as determined by the back optic radius of the first contact lens. A graph was constructed plotting probability of corneal transplant against each radius at the first visit, which could assist clinicians in advising patients.

Post keratoplasty dimensional and refractive change in contact lens and spectacle corrected cases.

Two groups of patients who had corneal transplantations for keratoconus were analysed retrospectively. One group had been visually corrected with corneal contact lenses and the second group had worn spectacles or used no optical correction. The reduction in post graft corneal astigmatism with time following removal of sutures was examined by the method of regression in groups. The results showed that the reduction in corneal astigmatism was significantly greater in the contact lens wearing group.