AAPS 104th Annual Meeting

Title: The 2019 FDA Breast Implant Recall And Subsequent Reporting Surge: A Controlled MAUDE Database Analysis

Background:

•In 2019, the FDA recalled Allergan BIOCELL textured breast implants for their association with breast implant-associated anaplastic large cell lymphoma (BIA-ALCL).

•This decision was followed by a significant amount of patient and media attention regarding the safety of breast implants

•Whether the recall selectively increased reporting of the malignancy or instead triggered a broader attention to implant-related complications remains poorly characterized.

Is the recall associated with an increase in all breast implant reports driven by non-malignant complications, rather than a true increase in reported BIA-ALCL events?

Methods:

•Data Source & Study Sample: 119,942 breast implant device reports and 118,508 matched adverse event narratives from 2010–2024 were analyzed.

•NLP-Based Narrative Classification: Natural language processing classified narratives into BIA-ALCL, rupture, capsular contracture, prophylactic removal, and litigation-related reports.

•Natural Control Comparison: Non-recalled manufacturers, including Mentor and Sientra, were used as natural controls to distinguish recall-specific effects from media-driven reporting behavior.

•Manufacturer-Stratified ITS Analysis: Interrupted Time Series analysis compared Allergan versus Mentor and Sientra using manufacturer-stratified Poisson regression with robust standard errors.

•Monthly Time-Series Modeling: Monthly-resolution ITS models were fitted with Newey–West standard errors.

•Sensitivity Analyses: Sensitivity analyses included COVID-era exclusion, alternative intervention points, & litigation adjustment.

•Multiple Testing Correction: Holm–Bonferroni correction was applied across seven event types to control for multiple comparisons.

FDA Maude Database

•119,942 Breast implants

118,508 matched narratives --> Natural language processing (BIA-ALCL, Rupture, Capsular contracture, Prophylactic removal): Interrupted Time Series Analysis (Sensitivity analysis: COVID, alternative intervention timing, litigation-adjusted)

Report surges:

•Allergan (IRR= 126.2) (p <0.001)

•Mentor (IRR = 239.7) (p <0.001)

•Sientra (IRR = 193.5) (p <0.001)

Results:

•Reports surged across all manufacturers and increased monthly (p < 0.001)
•Increase was driven by capsular contracture (IRR = 172.8) and rupture (IRR 63.6), p < 0.001).
•While BIA-ALCL reports increased in absolute number (IRR = 21.9), it declined proportionally from 1.6 to 0.7% of total reports.
•Prophylactic removal accelerated over time (p < 0.001)

Discussion:

•External events (recall, litigation, media) can drive temporary increases in reporting independent of true complication rates

•Rise in prophylactic removal likely reflects shifts in patient risk perception and clinician responsiveness

Conclusion:

•MAUDE data capture both clinical events and sociobehavioral responses during this period

AAPS 104th Annual Meeting

The 2019 FDA Breast Implant Recall And Subsequent Reporting Surge: A Controlled MAUDE Database Analysis

Poster Presenter

Authors

Affiliations

Part of Topic