Autoimmune disease and breast implants: systematic review of outcomes

Abstract

Given the use of breast implants for both cosmetic and reconstructive purposes is growing in the United States, an evaluation of factors that may affect the outcome of breast implant surgery is needed. A systematic review was conducted to evaluate the question: Does a personal or family history of autoimmune disease affect outcomes in breast implant surgery? The literature search yielded 2,425 records, but after removal of duplicates, abstract screening, and full-text assessment, only two studies met the inclusion criteria for the final review. Both studies provided level III evidence and the average Methodological Index for Non-Randomized Studies (MINORS) score was 16.5 (range: 15–18 out of 24), indicating a fair level of evidence overall. This systematic review found no evidence to support that a diagnosis of an autoimmune disease and/or a family history of autoimmune diseases will lead to poor surgical outcomes in breast implant surgery. Further study is warranted.

INTRODUCTION

The cosmetic and reconstructive use of breast implants continues to grow in the United States, with 299,715 cosmetic breast augmentations and 88,005 implant-based breast reconstructions performed in 2019 alone [1]. According to the National Institutes of Health (NIH), autoimmune disease is also on the rise, affecting 23.5 million Americans, 75% of whom are women [2,3].

Autoimmune disease occurs as a result of the immune system attacking the body’s own organs, tissues, and cells. Some of the more common autoimmune diseases include type 1 diabetes, rheumatoid arthritis, systemic lupus erythematosus, and inflammatory bowel disease. By depositing antibody-antigen complexes and inflammatory mediators, these diseases can affect a single organ or multiple organ systems [4–6]. Manifestations are wide ranging, including arthralgias, soft tissue abnormalities, and cardiac, pulmonary, renal, gastrointestinal, and neurologic changes [7,8]. Spontaneous wounding and delayed healing have also been described [9].

Due to these effects on the body, it has been postulated that the presence of autoimmune disease may lead to poor surgical outcomes in general [10]. In the setting of breast implant surgery, however, the effect of personal or family history of an autoimmune disorder on surgical outcomes, remains unclear. The most recent FDA guidance document states that the safety of breast implants in patients with autoimmune disease was simply not evaluated in breast implant premarket clinical studies [10]. Yet now, at a time when the safety of breast implants comes once again under scrutiny, concerns have been raised relating to both the risk of acute, perioperative complications as well as the development of chronic systemic symptoms in the population of women with autoimmune disease.

The purpose of this systematic review was thus to evaluate existing evidence to determine whether patients with a personal and/or family history of an autoimmune disorder, have worse perioperative outcomes in the setting of either cosmetic breast augmentation or implant-based breast reconstruction.

METHODS

On April 21, 2020, we performed a systematic review of peer-reviewed literature to identify studies examining outcomes of patients with a personal history and/or family history of autoimmune disease who are undergoing/have undergone breast implant surgery. We searched three databases: Embase.com (Elsevier), Ovid MEDLINE ALL, and Scopus (Elsevier). The search terms fell into two main categories, combined using the AND operator: (1) breast implants/prosthesis and (2) autoimmune disease. The other terms used for the first category were augmentation, augmentation mammaplasty, and tissue expander. Additional terms used for the second category were vasculitis, connective tissue disease, lupus erythematosus, mixed connective tissue disease, scleroderma, Hashimoto’s thyroiditis, inflammatory bowel disease, celiac disease, psoriasis, rheumatoid arthritis, Crohn’s disease, Addison disease, Sjogren’s disease, ulcerative colitis, multiple sclerosis, myasthenia gravis, type 1 diabetes, and Graves’ disease. The search strategies (Supplementary Material S1) were designed in collaboration with a medical librarian and reviewed by a second medical librarian using the Peer Review of Electronic Search Strategies (PRESS) checklist [11]. A systematic review management software (Covidence, Veritas Health Innovation, Melbourne, Australia) was used to export the search results (Supplementary Material S1) from the three databases. This systematic review was not prospectively registered in the PROSPERO database.

Two reviewers independently reviewed all abstracts after removal of duplicates; they then screened the full-text articles (Figure 1). Studies with a specific cohort of patients with a personal diagnosis of and/or family history of an autoimmune disease diagnosed in the preoperative period were included. All studies with unknown preoperative autoimmune diagnosis or no identifiable autoimmune cohort, case reports, and/or cases studies were excluded. Non-English and animal studies were excluded. Data were extracted independently and included patient history, surgical history, and complications.

Figure 1.

PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) chart.

Due to the lack of randomized controlled trials in this review, we used the Methodological Index for Non-Randomized Studies (MINORS) to assess the quality of the studies [12]. MINORS is a valid instrument composed of 12 items for comparative studies, resulting in a score ranging from 0 to 24 (the higher the score, the better the quality of the study). Two authors scored each article; an average summary score for all papers was calculated. Risk of bias present in each study was determined at the outcome level.

RESULTS

The literature search yielded 2,425 records (Figure 1). After duplicates were removed, 1,173 records were screened. Full-text articles for 33 studies were assessed for eligibility. After full-text assessment, only two studies met the inclusion criteria for the final review. Both studies provided level III evidence [13]. The average MINORS score was 16.5 (range: 15–18 out of 24), indicating a fair level of evidence overall. The remaining studies were excluded due to wrong study design (n=10) and wrong study population (n=21).

Chen et al published an analysis of outcomes following breast reconstruction in women with connective tissue disease (systemic lupus erythematosus, scleroderma, Sjogren syndrome, systemic sclerosis, rheumatoid arthritis, Raynaud phenomenon, psoriatic arthritis, sarcoidosis, and mixed connective tissue disease) [8]. The authors performed a retrospective analysis of the National (Nationwide) Inpatient Sample database, including all female patients 18 years or older who were admitted for immediate breast reconstruction. The analysis specifically evaluated outcomes in two subgroups: autologous tissue and direct-to-implant reconstruction. Of the 23,048 patients in the implant group, 431 had connective tissue disease. Patients with connective tissue disease were significantly older, more often African American or Hispanic, and had higher illness severity indices and comorbidities (p<0.001). Connective tissue disease patients were also more likely to have had neoadjuvant chemotherapy (8.1% vs 5.2%, p=0.007) and used anticoagulants (3.5% vs 0.7%, p<0.001). The use of antithrombotic/antiplatelet agents, aspirin, non-steroidal anti-inflammatory drugs, and steroids was similar between groups. The study enrolled patients over a 5-year period, but the mean follow-up is not stated. Outcomes analyzed included postoperative infection, wound disruption/complication, bleeding, seroma, pulmonary embolism, deep venous thrombosis, and length of stay. On multivariable analysis, connective tissue disease was an independent risk factor for wound disruption (OR 4.084; 95% CI 2.101–7.939) in patients who had direct-to-implant reconstruction. Of note, connective tissue disease was not found to be a risk factor for wound dehiscence in the autologous reconstruction group. Connective tissue disease was not determined to be an independent risk factor for any additional complications.

The strengths of the Chen et al study include the large sample size and the collection of data from multiple institutions. A major limitation is only using inpatient ICD-9 codes for data extraction, which may underappreciate the incidence of wound disruption. In theory, the sequelae of wound disruption in direct-to-implant breast reconstruction can range from wound care to explantation to failure of the reconstruction. However, patients who had complications diagnosed and managed in the outpatient setting would not have been included in the analysis by Chen et al, which only counted inpatient data. Another limitation is that the study group only included breast reconstruction patients; therefore, conclusions may not be generalizable to the cosmetic population.

The smaller, single institution study by Shuck et al compared outcomes of oncologic breast surgery and reconstruction in patients diagnosed with connective tissue disease (psoriasis/psoriatic arthritis, rheumatoid arthritis, lupus, scleroderma, mixed connective tissue disease, and seronegative polyarthritis) with patients without such a diagnosis [9]. Four groups were analyzed: ablative surgery only, autologous reconstruction, implant reconstruction (direct-to-implant or the immediate placement of tissue expanders), and revision surgery (including but not limited to tissue expander exchange to permanent implant). In all, 33 patients with connective tissue disease had 112 procedures. Nineteen of the 33 patients (58%) were undergoing treatment with disease-modifying antirheumatic drugs (DMARDs). In the implant group, 20 patients with connective tissue disease were compared with 134 control patients. No significant difference in complications was found between groups; details regarding reconstructive failure or implant loss were not given. In the revision group, revision surgeries in 40 patients with connective tissue disease were compared with 105 revision surgeries in the control group. No significant difference in postoperative infection, the only outcome examined, was found. Details on the use of DMARDs in each group were not included. Follow-up was not stated.

The Shuck et al study is limited by the small sample size and the heterogeneity of the population within a single institution [9]. Additionally, the retrospective nature of the assessment does not allow for an expansive outcomes assessment. For example, in the implant group, the only complications examined were infection, seroma, hematoma, delayed wound healing, and overall complications. We do not have a sense of the severity of each of these complications or reconstruction loss. In the revision surgery group, only infection was examined, substantially limiting the strength of the conclusion that there is no difference between groups. The impact of comorbidities on the outcomes evaluated was not examined.

DISCUSSION

This review underscores the paucity of data available to support the concern that patients with a personal history of autoimmune disease are at risk for a poor surgical outcome after cosmetic breast augmentation or implant-based breast reconstruction. In a systematic review of 2,425 articles, only two met the inclusion criteria; both were high-quality studies. One article concluded that patients with autoimmune disease have a higher risk of surgical and medical complications but not wound dehiscence. The other article showed that autoimmune disease does not affect surgical outcome. Both studies have strengths and weaknesses, as described above.

There were no data found to support the idea that patients with a family history of autoimmune diseases are at risk for a poor surgical outcome [10]. It is assumed that the concern surrounding “a family history” of autoimmune diseases relates to the underlying pathogenesis of these diseases. Although autoimmune diseases have a clear genetic component [5,14–18], environmental factors can also play a role in the development of autoimmune disease in an individual [4]. Therefore, raising concerns about postoperative complications in women with a family history of autoimmune diseases who undergo cosmetic or reconstructive breast surgery may be premature. Further investigation is warranted here.

It is possible the treatment of autoimmune diseases, not the autoimmune diseases themselves, may affect surgical outcomes. Steroid use has been shown to affect all three phases of wound healing: inflammatory, proliferative, and remodeling [19]. Therefore, the use of oral steroids may delay wound healing after surgery. Interestingly, a study that included data from 635,265 patients extracted from the National Surgical Quality Improvement Program (NSQIP) database examined the association between chronic preoperative steroid use and postoperative complications [20] and found that wound dehiscence increased by an odds ratio of 3.3 among the 20,434 who reported preoperative steroid use. Superficial surgical site infections, deep surgical site infections, organ/space surgical site infections, and mortality all significantly increased with the use of steroids (OR 1.724, 2.353, 2.469, and 3.920, respectively). In a large retrospective review of 94,140 plastic surgery cases also extracted from NSQIP [21], multivariate logistic regression analysis showed that surgical complications (OR 1.3; p=0.0452) and medical complications (OR 1.8; p=0.0002) were more likely to develop in chronic steroid users (n=1,661) than in nonusers (n=92,479); chronic steroid use was defined as regular administration of oral or parenteral corticosteroid medications 30 days preoperatively or at the time of being considered a candidate for surgery. The rate of wound complications, however, was not significantly different between groups. Breast implant-based plastic surgery procedures performed in chronic steroid users included delayed insertion of breast prosthesis following mastopexy or mastectomy or during reconstruction (n=69, 4.2%) and immediate insertion of breast prosthesis following mastopexy or mastectomy or during reconstruction (n=54, 3.3%).

DMARDs may also lead to wound healing problems. DMARDs improve symptoms of autoimmune diseases through immunomodulation (T-cell suppression, antimetabolite therapy, and tumor necrosis factor alpha inhibition) [9,22–26]. Similar to steroid use, however, DMARDs have been shown to increase infections and delay wound healing after surgery [27,28]. In patients with inflammatory rheumatic diseases, the risk of infection after orthopedic surgery was significantly elevated in patients taking conventional DMARDs (n=756; p<0.001) [28] but was not significantly different between patients without inflammatory rheumatic diseases (n=47,887) and those with inflammatory rheumatic diseases but not taking DMARDs (n=451).

In our systematic review, the effect of medications for the treatment of autoimmune diseases on postoperative outcomes could not be determined. Chen et al reported a similar amount of steroid use for both the connective tissue disease group and the control group [8]. Shuck et al reported 58% of patients using DMARDs in the total connective tissue disease group [9] but did not provide a breakdown of patients on medications specific to the implant and revision surgery groups. Additionally, the type of DMARDs used were not noted in this study. Therefore, we cannot determine from this systematic review whether one form of autoimmune suppression has any greater impact on outcomes than others. Research on the use of steroids and/or DMARDs and surgical outcomes in breast implant surgery is needed.

In conclusion, this systematic review does not provide adequate evidence to support that the diagnosis of an autoimmune disease and/or a family history of autoimmune diseases will lead to poor surgical outcomes in breast implant surgery.

KEY POINTS.

1. There is a lack of research involving surgical outcomes in breast implant surgery and autoimmune disease.

2. Current evidence does not definitively indicate a diagnosis of autoimmune disease will lead to poor surgical outcome in breast implant surgery.