Chapter on Connective Tissue Disease

The Institute of Medicine report on silicone breast implants reviewed 2,262 peer reviewed articles and 1077 other articles, book chapters, and written materials. All these articles are listed in the report. Although this sounds incredibly impressive, one has to wonder how carefully all these articles were read, and by whom. Remember that Committee members are essentially doing this work in their "spare time" in addition to their regular jobs. Usually most of the work is done by the staff, which included the Study Director, two research associates, and a project assistant. That would still average out to more than 800 articles each if no article was read by more than one staff person.

Presumably, the more important articles were read by more than one staff person. Perhaps this overwhelming number of articles is part of the reason that the chapter on the epidemiological studies of connective tissue diseases is so short - only 14 pages of the 400 pages of the report. Part of the chapter is devoted to atypical symptoms, but this review will focus on the classic diseases.

The IOM report's conclusions regarding connective tissue disease are virtually identical to those of the scientific panel appointed by Judge Pointer to study breast implants, stating that "there is insufficient evidence" to conclude that implants are associated with "defined connective tissue disease."

Given the results of the research that has been done, this is a reasonable conclusion.

However, the section on connective tissue diseases is surprisingly skimpy, considering that these diseases are the greatest source of controversy regarding breast implants. This chapter could not be considered critical by scientific standards - it describes 17 studies, but doesn't criticize them. In fact, it often leaves out such crucial information as the length of time that the women in the studies had implants at the time that their health was evaluated. Some studies included women who had implants for only a few months.

Even including women with implants for 1-2 years is questionable, since so many of these diseases take many years to develop. In it's overall review of the studies, the IOM report's only mention of shortcomings consists of the statement that a previous scientific review panel and the authors of the studies themselves "have pointed out problems with individual epidemiological studies." However, the authors conclude that, despite these problems, "the consistency of results among many reports is impressive."

The one specific study that was criticized was the study conducted by Dr. Charles Hennekens at Harvard, a very large epidemiological study that found an increase in connective tissue diseases among women health professionals with implants. The authors assumed that these results were biased because the proportion of women reporting implants was twice as high as expected, and because the symptoms were based on self-report rather than medical records.

Is the sample biased? Some plastic surgeons and patients point out that health professionals, especially nurses, might be especially likely to have breast implants because they have greater access to medical treatment. In fact, nurses sometimes volunteer to serve as "test subjects" for plastic surgeons as a way to get free surgery.

The criticism of self-reported medical problems is legitimate: in medical research, it is better to have self-reported illnesses verified by medical exams. However, the women in this study were health professionals and therefore could be considered quite well qualified to report their own symptoms -and yet, that fact was not mentioned in the IOM report. The criticism of the Hennekens study, in the absence of criticisms of any of the other studies, is difficult to justify on a scientific basis. Many scientists would argue that using self-reported illnesses made by health professionals is a less serious shortcoming for implant studies than the use of hospitalization records to evaluate illnesses, and yet the studies that relied on hospitalization records were considered valid by the IOM report. Since many women with early stage connective tissue disease would not be hospitalized, they would not have been identified as ill in those studies. This would be a particular problem for the women who had implants for short periods of time and therefore would not have had time to develop full-blown symptoms even if the implants made them ill.

The chapter does not mention that the value of the studies of scleroderma are limited by the inadequate number of women with implants in those studies. Scleroderma is a very rare disease, and tens of thousands of women would need to be studied to detect even a doubling or tripling of risk. Similarly, no concern was expressed about how long it takes for scleroderma symptoms to be worrisome enough for patients to seek medical care and therefore be diagnosed.

After ignoring all these shortcomings, the authors conclude that "given the repeated finding of no elevated risk, the evidence supports the conclusion that there is no association, and therefore no justification for the use of resources in further epidemiological exploration of such an association."

So, in addition to failing to mention the shortcomings of the studies that have been done, the report concludes that researchers need not waste their time by doing better studies in the future. This last statement, based on 17 flawed studies, will be questioned by any unbiased researcher.

What Kind of Research is Needed? Are breast implants causing connective tissue disease? If we want to answer that question, we need research that: 1. Studies large numbers of implant patients for long periods of time, using a random or unbiased sample.

2. Focuses on women with ruptured implants, comparing them to women whose implants did not break, and to women who were similar but had no silicone implants.

3. Uses appropriate measures of illnesses, preferably including medical examinations.

It would be important to also compare women with saline breast implants to those with silicone gel implants, and mastectomy reconstruction patients with augmentation patients.