Study supports importance of genetic effects in myopia

October 15, 2004

Baltimore-Researchers from the Johns Hopkins Bloomberg School of Public Health studying a well-defined, older population found refractive error and myopia were highly heritable.

Baltimore-Researchers from the Johns Hopkins Bloomberg School of Public Health studying a well-defined, older population found refractive error and myopia were highly heritable.

The investigation included 759 siblings in 241 families from the Salisbury Eye Evaluation Study in eastern Maryland. Robert Wojciechowski, OD, MSc, a doctoral student in the department of epidemiology, reported the results at the annual meeting of the Association for Research in Vision and Ophthalmology.

Heritability, calculated using multivariate linear regression and estimated as twice the residual between-sibling correlation after adjusting for age, gender, and race, was found to be relatively high-61%. Logistic regression models were used to estimate odds ratios (ORs) of myopia given a myopic sibling relative to having a non-myopic sibling. Those analyses were performed using four different cut-off thresholds for defining myopia (-0.5, -1, -1.5, and -2 D), and all four models yielded similar ORs that ranged from 2.25 to 3. Relative risk calculations also showed that for all four myopia definitions, individuals having a myopic sibling were at about a twofold greater risk of being myopic compared with the general population.

The population for the myopia genetics study was assembled by inviting probands in the population-based Salisbury Eye Evaluation and their siblings residing within a 100-mile radius to participate.

"In investigating heritability of refractive error and familial aggregation of myopia, it is important to study an unselected cohort such as was available from this population-based eye study. Selectively recruiting patients with myopia would provide a biased, inaccurate measurement of what proportion of myopia is due to genetic factors," Dr. Wojciechowski said.

Of the 759 participants, about 60% were females and nearly three-fourths were white. The subjects ranged in age from 48 to 90 years and had a mean age of 73.4 years. Mean refractive error ±SD for the group was 0.65 ±2 D. With myopia defined as-1 D, about 14% of the population was myopic, and the patients with myopia had a median refractive error of -2.59 D.

All of the subjects underwent refractive error measurement in each eye, and mean spherical equivalent was calculated for each subject for use in the analyses. A visual acuity (VA) better than 20/40 was defined as plano (or zero refractive error). Individuals with VA better than 20/40 with glasses had refraction determined by lensometry. A full non-cycloplegic refraction was performed if VA was 20/40 or worse with or without glasses.

If the individual was pseudophakic in one eye, the refraction from the phakic eye was used. Refractive error data for patients with binocular pseudophakia were obtained from presurgical records.

Covariates in addition to age, sex, and race that were incorporated in the multivariate model included height, weight, body mass index, and Wilmer cataract grade, but none of those factors was statistically significantly associated with myopia. The analyses did find that myopia decreased by about 0.33 D per decade of age within its population's age group.

"Some previous studies have reported an inverse correlation between height and refractive error and an association between nuclear cataract and myopia," Dr. Wojciechowski said. "We found a nonsignificant trend for taller individuals to be more myopic, but no significant relationship between cataract and refractive error, perhaps because we did not have data on cataract grade for an appreciable proportion of our population that previously underwent cataract surgery."

Other study results Although the study found the likelihood that a person was myopic if his or her sibling was myopic was robust over the range of myopia definitions investigated, Dr. Wojciechowski noted that previous studies looking at populations with higher levels of myopia (< -6 D) have shown the risk of having myopia if an individual's sibling has that high level of refractive error is even greater.