Results and Discussion
Initially, our query yielded 566 studies related to breast implant illness, its related syndromes, or silicone breast implant explantation. After abstract and full text review, eight studies met our inclusion and exclusion criteria for breast implant illness outcomes studies. Three additional studies were added after manual citation extraction and full text review, for a grand total of 11 studies (Figure 2). The primary findings, limitations, and year of publication of these studies are summarized in Table 1. The following discussion summarizes the conclusions of these key studies while highlighting their limitations. Finally, some critical themes arise from these investigations and are subsequently reviewed here.
Resolution of Symptoms With Explantation
Of the 11 studies identified, six studies[23–28] identified by our search protocol explicitly reported on outcomes following implant explantation. Though heterogenous in their study design and execution, these outcomes-based studies share a unifying theme in that there seems to exist a group of patients who experience improvements in symptoms after removal of their breast implants. The included studies report 17 to 84 percent of patients improved with implant removal.[23–31] However, there was not consensus regarding which subset of patients improved with explantation. The literature reviewed here suggests that a subset of these patients does not respond at all, while others respond initially before symptoms recur or progress. Complicating the conversation further, clear long-term outcome data at more than 1 year is lacking, with most studies stating an average follow-up of 1 to 7 months[23–26] and a single study with 2.7-year average follow-up. This is particularly problematic, as breast implant illness is a notoriously insidious disease process with highly variable natural history.
Peters et al., in their prospective study of 100 consecutive women requesting silicone implant explantation during the U.S. Food and Drug Administration moratorium, reported that, in patients without a rheumatic or autoimmune condition, more than 80 percent saw improvement with implant explantation. This is in contrast to those with documented autoimmune disease who saw no improvement. Importantly, they found that in those with fibromyalgia or other rheumatic condition, they saw an initial "euphoric" improvement immediately following surgery followed by recurrence of symptoms after a 6- to 12-month interval. Similarly, Colaris et al. examined 100 patients with silicone implant illness syndrome and reported a 50 percent initial improvement in symptoms with explantation, followed by 26 percent experiencing return of symptoms, though, the authors do not explicitly define their follow-up interval. Such a temporally dependent clinical response rate serves to highlight the importance of sufficient follow-up in the clinical study of breast implant illness.
In addition to this temporal relationship, several studies suggest that differing subsets of patients respond differently to implant removal. In a recently published retrospective study of patients requesting implant removal at a single institution composed primarily of breast implant illness patients, Wee et al. found an association between explantation and symptom improvement on survey. Further, they reported that patients with body mass index of more than 25 and those with capsular contracture had statistically significantly greater improvement compared to thinner patients and those without clinically significant capsular contracture on presentation. These effects seem to persist in the postoperative period as queried after 30 days; though there was a significant proportion of patients lost to follow-up: only 52.7 percent completed the survey after 30 days. Further limiting clinical interpretation of these findings, only 8.4 percent of the study group had a more than 6-month follow-up, and fewer than 2.9 percent had a 1-year follow-up.
The senior author (R.J.R.) of this present review previously performed a prospective study of patients requesting explantation secondary to concerns about systemic health concerns related to silicone breast implants. Patients reported a decrease in musculoskeletal complaints after explantation up to the study follow-up interval of 6 months. Further, those patients who reported worse musculoskeletal complaints before explantation saw the greatest improvement after explantation. Interestingly, this prospective study conflicts with our prior findings from a study of this population that found those with milder symptoms reported greater improvements in quality of life after explantation when asked to recall the history of their clinical response retrospectively.
While several studies demonstrated improvement in many patients with explantation, others showed that patients did not improve. The challenge is determining who will benefit from explantation with sustained improvement. Understanding the subjective and sometimes vague nature of the symptomology that makes up breast implant illness, the above studies highlight the importance of timing in study design, with a minimum follow-up time and prospective design necessary to accurate capture clinical response.
A subset of breast implant illness patients responds to implant removal, while others do not.
Response varies over time, with some recurring after 6 to 12 months.
Accurate study requires prospective, controlled design with more than 12-month follow-up.
Defining the Illness
Breast implant illness describes an inconsistently defined constellation of symptoms associated with silicone breast implants that has persisted relatively unchanged through the silicone breast implant moratorium to the subsequent moratorium lift and today (Figure 1). In 2014, Colaris et al. performed an observational, cross-sectional study of 100 patients with silicone implant incompatibility syndrome to a historical premoratorium cohort from 1985 to 1992. They observed that, while symptoms are varied, the described complaints and prevalence of breast implant illness have remained largely unchanged over the past 30 years, despite evolving breast implant manufacture practices and disease management strategies. Recently, some authors have suggested management requires en bloc capsulectomy akin to that used in the treatment of anaplastic large-cell lymphoma, despite the distinct lack of quality published studies related to this practice.
One of the great challenges in the study of breast implant illness is that the presenting systemic symptoms are generally subjective, often vague, and decidedly nonspecific. Quantifying myalgias, arthralgias, "brain fog," dry eyes, sleep, mood disturbances, and other such factors becomes a challenge for clinical study. Most studies have, therefore, been survey-based and often rely on self-reported symptoms. Complicating the picture further, breast augmentation patients often endorse many of these symptoms in general without breast implant illness. Fryzek et al. carried out a comparative study using the Swedish national implant registry to study these symptoms. They queried a large study group of 2500 patients having undergone breast augmentation and compared them to an even larger control group of 3500 who had undergone breast reduction surgery. Both groups were surveyed for the 28 systemic symptoms descriptive of breast implant illness. They found that both breast surgery groups endorsed some symptoms, but breast augmentation patients were statistically significantly more likely to experience 16 of the 28 queried symptoms, with an overall higher average number of symptoms (3.0 versus 2.6, respectively). They did not find significant differences between silicone versus saline breast augmentation patients. Thus, outside of any defined illness, many breast surgery patients experience some of the nonspecific systemic symptoms associated with breast implant illness at baseline. Breast augmentation patients, regardless of silicone or saline, are overall even more likely to report these symptoms.
Breast implant illness management trends have evolved over the past 30 years, empirically mirroring that of the unrelated condition BIA-ALCL.
Breast augmentation patients report higher nonspecific systemic symptoms at baseline, irrespective of specific diagnosis. Thus, studying this population requires careful attention to appropriate control group selection.
The Missing Link: Limitations in Study Design
Of the 11 identified outcomes studies, only four used a comparative study design (Table 1). Across these studies, control groups were inconsistently implemented. Lee et al. compared 50 breast implant illness patients undergoing explantation and capsulectomy to non–breast implant illness patients undergoing implant exchange and capsulectomy in a prospective, patient-reported outcomes questionnaire study. On histopathologic and microbiologic examination of the explanted capsules, they found significantly higher rates of synoviocyte metaplasia and 6-fold higher chronic infection, with Propionibacterium acnes being uniquely found in the breast implant illness cohort breast capsules. Though unable to demonstrate causation, these findings support the idea that the breast capsule may contribute to the disease process. Unfortunately, it was the only study to-date evidencing a link between the capsule and breast implant illness. Interestingly, no study has yet evaluated the effect of capsule management on breast implant illness symptom improvement, despite a growing number of surgeons purporting the necessity of en bloc capsulectomy.
Another ubiquitous limitation of our understanding of breast implant illness is that selection and sampling bias were present in all the identified breast implant illness-outcomes studies. Most existing studies suffer from significant selection bias in that they either have a significant proportion (>40 to 50 percent) lost to follow-up or have insufficient follow-up duration to capture the natural history of the disease process. Nine out of the 11 identified studies have mean follow-up times of 7 months or less, despite the earliest included study (Peters et al.), demonstrating that many patients have return of symptoms after several months, necessitating longer-term follow-up to capture the natural history of the disease process. Such selection bias limits the internal validity of these study findings, as the primary outcome variable is measured inadequately with insufficient follow-up.
Further, studies to-date suffer from sampling bias that threatens external validity, as the largest cohorts come from self-reported study populations of patients who, by definition, express symptomatology related to their implants—whether real or perceived. As a result, executing a statistically sound study design with appropriate control and experimental groups remains inherently challenging. In the context of these limitations, the existing data become difficult to interpret, and the task of establishing causation and formulating best management strategies for breast implant illness becomes conjecture at best.
The importance of the capsule to breast implant illness pathology has not been established.
Accurate study requires careful a priori experimental and control group selection.
Investigation of sustained response and outcomes requires a minimum follow-up of at least 6 months, but preferably more than 12 months. Shorter follow-up biases study results to suggest explantation leads to false-positive clinical response.
Plast Reconstr Surg. 2022;149(4):638e-645e. © 2022 Lippincott Williams & Wilkins