Determining Breast Implant Prevalence: A Population Study of Italian Chest Radiographs

Abstract

Abstract Background Current breast implant prevalence within the general population remains elusive. An accurate prevalence is critical to serve as the denominator for any assessment of breast implant-related complication. The purpose of this manuscript is to assess this prevalence in women aged 20–70 years in Italy. Materials and Methods Eight reviewers, demonstrating a mean sensitivity of 87.0% and specificity of 97.0%, were recruited for retrospective identification of implants on chest radiographs from a tertiary academic hospital in a major urban setting. Three final reviewers were selected, and they assessed all eligible chest radiographs collected between January and December 2019. The hospital-based population was compared to epidemiological data at a local, regional and national level to demonstrate homogeneity of age structures using the phi correlation coefficient. Results We identified 3,448 chest X-rays which yielded 140 implants, with an overall prevalence of 4.1% for women aged 20–70. Implants were bilateral in 76% of cases and unilateral in 24%. They were placed cosmetically in 47.1% cases and used for reconstruction in 52.9% cases. Phi correlation coefficient found no differences across hospital-based, local, regional and national populations. Conclusion A validated method was performed to estimate implant prevalence from an academic hospital in a major urban setting at 4.1% and was used to estimate national prevalence in Italy. The implications of this epidemiologic study may reach across national borders for improved understanding of breast implant epidemiology and in predicting the total number of patients within a given population that may be affected by device complications. Level of Evidence IV This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266.