OBJECTIVE: To assess whether gonioscopy performed on Samoyed puppies (6-14 weeks of age) is a reliable predictor of their gonioscopy results as adults (>11 months of age).
METHODS: A retrospective medical record search was performed to identify Samoyed dogs that had undergone gonioscopy as a puppy (6-14 weeks of age). A second prospective gonioscopy examination was performed as an adult (>11 months of age) and the results were compared. Gonioscopy was assessed on two characteristics: percent of iridocorneal angle (ICA) unaffected by pectinate ligament abnormalities (PLA) and ICA width, with PLA results categorized into unaffected ( ≥ 75% open), moderately affected (50%-74% open), and severely affected (<50% open) and the width categorized into wide, moderate, narrow, and closed groups. A multivariate model was created that considered factors such as PLA, angle width, sex, puppy age, neuter status, and time between examinations.
RESULTS: The eyes of 77 Samoyed dogs (154 eyes) met inclusion criteria. When assessing PLA, 90% of dogs had the same categorization as a puppy and as an adult. When assessing ICA width, 53.2% of dogs had the same width classification as a puppy and as an adult. There is strong evidence that variation between puppy and adult results increases with time between examination (p = .03).
CONCLUSIONS: Puppies unaffected by PLA and with a wide ICA angle, are likely to retain these features as adults. Selection of puppies with desirable gonioscopic traits may help breeders to choose dogs likely to retain these traits as adults for breeding. This may reduce the incidence of glaucoma in the breed. Puppy examinations should be performed in both eyes. Examination of adults prior to breeding remains prudent.

OBJECTIVE: Canine primary closed-angle glaucoma (PCAG) is a complex disease caused by multiple genetic factors. A c.590G>A variant in OLFML3 was recently reported to be a candidate for pectinate ligament abnormality (PLA) and PCAG in the Border Collie. We investigated the association of this variant with PLA and PCAG in Border Collies from the United Kingdom.
METHODS: The OLFML3 variant was genotyped in 106 Border Collies comprising 90 with normal eyes (controls) and 16 with PLA (n = 11) and/or PCAG (n = 5) (cases). Genotyping was performed in an additional 103 Border Collies to estimate variant frequency within the population. To investigate the association of the variant with disease in other breeds, genotyping was performed in 337 non-Border Collies with PLA and/or PCAG.
RESULTS: Of the 90 controls, 71 were homozygous for the wild-type allele, two were homozygous for the variant, and 17 were heterozygous. Of the 16 cases, three were homozygous for the wild-type allele, 11 were homozygous for the variant, and two were heterozygous. The association of the variant allele with disease was significant (P = 1.1 x 10-9 ). We estimated the frequency of this variant to be 4.4% within the United Kingdom Border Collie population, and it was not identified in clinically affected dogs of any other breed.
CONCLUSIONS: This study confirms the association of the OLFML3 variant with PLA and PCAG in Border Collies from the United Kingdom. DNA testing for the variant and selective breeding can reasonably be expected to result in a reduction of PLA and PCAG prevalence in the breed.