Incremental Revisions across the Life Span of Ophthalmic Devices after Initial Food and Drug Administration Premarket Approval, 1979-2015

Purpose: To characterize the frequency, nature, and regulatory mechanisms by which ophthalmic devices are iteratively modified after initial Food and Drug Administration (FDA) Premarket Approval (PMA).

Design: Retrospective cross-sectional analysis using publicly available FDA data.

Participants: Ophthalmic devices initially approved via the FDA's PMA pathway between January 1, 1979 and December 31, 2015.

Methods: We used the FDA's PMA Database to identify and characterize initial approvals and subsequent postmarket modifications to Class III ophthalmic devices. The FDA Recalls Database was used to identify associated safety events.

Main outcome measures: Median iterated life span (timespan across which modifications occurred after initial PMA) and median number of supplements approved per device, by device type, and overall, stratified by regulatory pathway and modification type.

Results: Between 1979 and 2015, the FDA approved 168 original ophthalmic devices via the PMA pathway and 2813 subsequent modifications. More than one third (n = 64; 38%) of original approvals were intraocular lenses. Overall, devices underwent a median of 11 postmarket modifications (interquartile range [IQR], 3-24.8) across a median 10.0-year iterated life span (IQR, 4.1-16.7). The majority of devices (n = 144; 86%) underwent more than 1 postapproval modification, including more than 1 design modification (n = 84; 50%). The median number of changes altering device design or labeling was 3.5 (IQR, 1-9). Although manufacturing alterations (n = 834 of 2813; 30%) were the most frequent type of revision, changes involving device design (n = 667; 24%) and labeling (n = 417; 15%) were common. Recalled devices underwent more frequent postapproval modifications per year (median, 1.4; IQR, 0.7-2.3; mean, 1.5; 95% confidence interval, 1.1-1.9) in the period preceding recall than did nonrecalled devices (median, 0.5; IQR, 0.2-1.1; mean, 0.8; 95% confidence interval, 0.7-1.0) across their market approval period (P < 0.001).

Conclusions: Most ophthalmic devices approved via the FDA's PMA pathway have undergone extensive revisions, including serial design and labeling changes, since their initial approvals, often without supporting clinical data. Ophthalmologists should take into consideration that cumulative revisions may render the clinical evidence that supported an original FDA approval less relevant to newer device models.