Full Citation
- Citation (Vancouver Style): Weltz TK, Peng S, Larsen A, Bak EEF, Tran JVQ, Gudjonsdottir LR, Hemmingsen MN, Ørholt M, Mielke LV, Trillingsgaard J, Elberg JJ, Hölmich LR, Jensen LT, Vester-Glowinski P, Bjarnsholt T, Trivedi U, Clemens MW, Li X, Sørensen SJ, Herly M. Breast implant surface texture is associated with distinct implant microbiome profiles in humans [1]. Acta Biomaterialia [2]. 2026;219:117–125 [2].
- DOI: https://doi.org/10.1016/j.actbio.2026.06.038
Disclosure
Dr. Epstein is a paid investigator in the Motiva U.S. FDA clinical trial and is compensated by Establishment Labs to train surgeons on the Preservé breast augmentation technique. He discloses this relationship in all public communications involving Motiva or Preservé. Dr. Epstein owns stock in Establishment Labs.
Authors and Affiliations
- Authors: Tim K. Weltz, Shuang Peng, Andreas Larsen, Erik E.F. Bak, John V.Q. Tran, Linda R. Gudjonsdottir, Mathilde N. Hemmingsen, Mathias Ørholt, Louise V. Mielke, Jesper Trillingsgaard, Jens J. Elberg, Lisbet R. Hölmich, Lisa T. Jensen, Peter Vester-Glowinski, Thomas Bjarnsholt, Urvish Trivedi, Mark W. Clemens, Xuanji Li, Søren J. Sørensen, and Mikkel Herly [3].
- Note: Tim K. Weltz and Shuang Peng contributed equally and share first authorship [4].
- Affiliations: [5]
- a: Department of Plastic Surgery and Burns Treatment, Copenhagen University Hospital, Rigshospitalet, Inge Lehmanns Vej 8, Copenhagen 2100, Denmark.
- b: Department of Biology, Section of Microbiology, University of Copenhagen, Universitetsparken 15, Copenhagen 2200, Denmark.
- c: Department of Pediatrics, West China Second University Hospital, Sichuan University, Key Laboratory of Birth Defects and Related Diseases of Women and Children, Ministry of Education, Chengdu, China.
- d: Sculpture by, Copenhagen, Denmark.
- e: AK Nygart, Copenhagen, Denmark.
- f: Amalieklinikken, Copenhagen, Denmark.
- g: Department of Plastic Surgery and Reconstructive Surgery, Copenhagen University Hospital, Herlev and Gentofte Hospital, Copenhagen, Denmark.
- h: Department of Immunology and Microbiology, University of Copenhagen, Copenhagen, Denmark.
- i: Department of Clinical Microbiology, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark.
- j: Department of Plastic Surgery, Division of Surgery, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
- k: Institute of Microbiology, College of Life Sciences, Zhejiang University, China.
Study Design and Evidence Level
- Study Type: Retrospective clinical cohort study utilizing explanted breast implants [6] [7].
- IRB Approval: Approved by the Regional Health Ethics Committee (H-20,015,276) and the Danish Data Protection Agency (P-2020-176, P-2020-177) [6].
- Sample Size: 221 patients with 391 breast implants [8].
- Groups: [9]
- Smooth (n = 50)
- Motiva SmoothSilk (n = 36)
- Mentor Siltex (n = 119)
- Eurosilicone Cristalline (n = 39)
- Allergan Biocell (n = 73)
- Other textures (n = 74)
- Follow-up / Implantation Time: Median time of implantation was 14 years (IQR 5.8–19.0, range 0.3–42.0) [10].
- Level of Evidence: Not reported.
Objectives
- To assess how breast implant surface design influences microbial colonization at the implant-tissue interface [11].
- To provide new insights into the interaction between host, material, and microbes relevant to the foreign body response [11].
Background Context
- Surface texture is a key design feature of silicone breast implants that has long been implicated in modulating the foreign body response, including the risk of capsular contracture [12].
- Experimental studies suggest that rougher implant surfaces can support a higher bacterial load [13].
- Textured breast implants are associated with the development of a rare lymphoma, breast implant-associated anaplastic large cell lymphoma (BIA-ALCL), with the highest risk observed for implants with the roughest surface textures [14] [15].
Methods
- Patient Population & Inclusion: Patients who underwent revisional surgery after a previous breast augmentation from 2019 to 2023 across five different plastic surgery departments and private clinics [7]. Implants without identifiable manufacturer details but with obvious surface texturing were included in an 'Other textures' category [16].
- Interventions & Sample Collection: During revisional surgery, the entire breast implant was aseptically collected in a sterile container immediately after explantation and stored at −80 degrees Celsius until analysis [17].
- DNA Extraction and 16S rRNA Amplicon Sequencing:
- Bacteria were isolated from the entire implant surface and dissolved in 100 mL sterile PBS buffer using alternating cycles of 30 s of manual agitation and ultrasonication for a total of three minutes [18].
- The suspension was concentrated into 1 mL of wash buffer, and DNA extraction was performed using the NucleoSpin 96 Soil DNA Isolation Kit on the epMotion 5575 robotic platform [18].
- PCR amplification targeted the hypervariable V3-V4 region of the bacterial 16S rRNA gene using modified broad-range primers 341F and Uni806R [19].
- Paired-end sequencing was performed on the Illumina MiSeq platform with the MiSeq reagent kit v3 (2 × 300 bp) [19].
- Bioinformatics: Raw fastq files were demultiplexed using Illumina software, and primers/adaptors were removed using cutadapt [20]. Amplicon sequence variants (ASVs) were identified using the DADA2 algorithm on the QIIME2 platform [20]. Taxonomic classification was mapped to the SILVA v132 reference database, and taxa were aggregated at the genus level [20].
- Contamination Control: Negative and positive controls (67 in total, representing 8.8% of all samples) were included [21]. Contaminants were identified and removed using the decontam method, resulting in the removal of 154 taxa [21].
- Outcome Measures: Within-group alpha diversity was assessed using observed richness, Pielou's Evenness Index, and the Shannon Index [22]. Beta diversity was analyzed using Bray-Curtis dissimilarity [23].
- Absolute Bacterial Quantification: A subset of 20 samples representing all implant surface groups was analyzed using digital PCR (dPCR) with a pan-bacterial 16S rRNA gene assay on the QIAcuity One system [24] [25].
- Statistical Methods: Differences in alpha diversity and differential abundance of the five most abundant genera were tested with generalized linear models (GLMs) [22] [26]. All GLMs adjusted for foreign body response severity (mild vs severe) and implantation time [26]. Generalized estimating equations (GEEs) were implemented to account for clustering between two implants from the same patient in bilateral surgeries [26]. Beta diversity significance was tested with PERMANOVA (adonis2) [23]. Multiple testing was adjusted using the Benjamini-Hochberg false discovery rate (padj < 0.05) [27].
Results
- Cohort Characteristics: The study analyzed 391 breast implants from 221 patients [8]. Severe foreign body response (Baker grade III/IV) was present in 194 implants, and mild foreign body response (Baker grade I/II) was present in 197 implants [28].
- Microbiome Composition: The most abundant bacterial genera overall were Staphylococcus (27%) and Corynebacterium (7.2%) [29].
- Alpha Diversity:
- Alpha diversity gradually decreased with increasing average surface roughness [30].
- Allergan Biocell implants had significantly lower diversity than smooth implants (Shannon index 1.7 [95% CI 1.5–1.9] vs. 2.0 [95% CI 1.7–2.5], padj=0.02) [30].
- This decrease was primarily driven by a trend towards lower evenness of the observed genera (Pielou's Evenness 0.60 [95% CI 0.54–0.66] vs 0.69 [95% CI 0.55–0.82], padj=0.08), as observed richness was similar between the two groups (26.1 [95% CI 20.8–31.4] vs 24.2 [95% CI 13.0–35.4], padj=0.82) [30].
- Eurosilicone Cristalline closely resembled the diversity of the Allergan Biocell group but had the highest observed richness (42.6 [95% CI 27.4–57.8]) [31].
- Mentor Siltex diversity fell between the rougher groups and the smoothest groups [31].
- Motiva SmoothSilk closely resembled the smooth group across all alpha diversity indexes, despite a large difference in median implantation time (smooth 16 years vs Motiva SmoothSilk 1.2 years) [32].
- Beta Diversity:
- The smooth surface group had a significantly different microbiome composition compared with all textured groups except for Motiva SmoothSilk [33].
- Allergan Biocell differed from all groups except Eurosilicone Cristalline [34].
- Mentor Siltex was significantly different from both the smoother groups (smooth and Motiva SmoothSilk) and the rougher groups (Allergan Biocell and Eurosilicone Cristalline) [34].
- Absolute Abundance (dPCR):
- The dPCR assay showed a mean valid partition rate of approximately 97% [35].
- Absolute bacterial abundance varied substantially, with a mean abundance of 29.6 copies/μL (95% CI 9.9–49.3, range 1.1–229.0), corresponding to a mean recovered 16S copies per implant of 5920 (95% CI 1980–9860, range 220–45,800) [35].
- Total bacterial abundance increased across surface groups according to average surface roughness [35].
- Differential Abundance:
- Staphylococcus was significantly more abundant on textured surfaces compared to smooth surfaces: Mentor Siltex (padj=0.01), Eurosilicone Cristalline (padj=0.01), and Allergan Biocell (padj<0.001) [36].
- No bacterial genera differed significantly between the Smooth and Motiva SmoothSilk groups, and no clinically relevant genera differed between Allergan Biocell and Eurosilicone Cristalline [37].
Conclusions
- Breast implants harbor distinct microbiome profiles according to their surface texture [38].
- Differences in microbiome diversity and composition follow a gradient according to average surface roughness [38].
- Rougher textures exhibit lower microbial diversity due to decreased evenness and a significantly higher relative abundance of Staphylococcus compared with smooth implants [38].
- These texture-associated differences in the implant microbiome may contribute to differences in the host foreign body response [38].
Key Terminology and Definitions
- Mild Foreign Body Response Severity: Defined as the absence of capsular contracture (Baker grade I/II) [26].
- Severe Foreign Body Response: Defined as the presence of capsular contracture (Baker grade III/IV) [26].
- Observed Richness: The number of bacterial genera detected per sample [22].
- Pielou's Evenness: How uniformly the present bacterial genera are distributed [22].
- Shannon Index: A metric that includes both richness and evenness to assess diversity [22].
Limitations
- Implantation Time and Severity Variation: The time of implantation and the number of implants with severe foreign body responses varied between groups, which could have influenced the microbial profiles [39].
- Taxonomic and Functional Resolution: 16S rRNA sequencing limits taxonomic resolution and lacks functional information on the present genera [40].
- Contamination Risk: Potential contamination during sampling and laboratory processing can never be entirely ruled out despite extensive negative controls [40].
- Implant Fill Assessment: Due to the small number of non-silicone filled implants (4.1%) in the cohort, the study was unable to assess implant fill as an independent determinant of the microbiome [40] [41].
- Absolute Abundance Sample Size: Absolute bacterial abundance was assessed only in a small pilot subset of 20 implants using dPCR [25] [42] [43].
Funding and Conflicts of Interest
- Funding: Funded by the Research Fund of Rigshospitalet, the Novo Nordisk Foundation (grant numbers 052198, 0058322), and the Lundbeck Foundation (grant number R413-2022-621) [44]. The funders had no role in study design, data collection, analysis, decision to publish, or manuscript preparation [44].
- Conflicts of Interest:
- Lisbet R. Hölmich received a research grant from Mentor/Johnson & Johnson (not relevant to the current project) [45] [46] and reports a relationship with Johnson & Johnson that includes funding grants [47].
- Mikkel Herly received financial support from the Lundbeck Foundation [47].
- Søren J. Sørensen received financial support from the Novo Nordisk Foundation [47].
- No other authors declared competing interests [46] [47].
Plain-Language Patient Summary
This study analyzed the bacterial communities (microbiomes) on 391 breast implants removed from 221 patients to determine if the implant's surface texture affects bacterial growth [8]. The researchers found that implants with rougher surfaces had different, less diverse bacterial profiles and a significantly higher abundance of Staphylococcus bacteria compared to smooth implants [30] [36] [38]. These findings suggest that the surface texture of a breast implant directly shapes the local bacterial environment, which may help explain why different implant designs lead to different bodily reactions or complications [38].
References
[1]“File: Breast implant surface texture is associated with distinct implant microbiome profiles in humans.pdf,” File.
