Additional medical records were obtained for 9 of the 10 suspected

polymyositis/dermatomyositis (PM/DM) cases. A CDC rheumatologist reviewed these and found 8 to be relatively typical PM/DM cases that are confirmed (6 definite and 2 probable) using the same criteria that Oddis used in his study?. The ninth case is not PM/DM, and we continue to seek records on the tenth case. Of the eight confirmed cases, seven occurred in white females and one in a white male, at least seven were hospitalized, three had a history of breast implants, two had a history of thyroid disease, and one died due to complications of the treatment.

1. Bohan A, Peter JB. Polymyositis and dermatomyositis. N Engl J Med 1975;292:344-347,403–407.)

2.

Unpublished manuscript: Oddia CV, Conte CG, Steen VD, Medager TA.

Incidence (survey) of hospital-diagnosed polymyositis-dermatomyositis: A twenty year study of cases in Allegheny County, Pennsylvania, 1963 to 1982)

CDC staff have reached a conclusion that the potential association between IBC and PM/DM is worthy of further investigation. There are several reasons for this.

First, using Dr. Chiacchierini's January 3, 1990 memo "Estimation of PM/DM cases from Zyderm/Zyplast exposure," which is based on reasonable assumptions, the seven cases among white females is much higher than the 4.253 cases expected is close to achieving the statistically significant value of nine cases. CDC staff favors this analysis, because it assumes a follow-up period of 1.1 years. Our experience suggests that effective follow-up would be worse than the company's surveys indicate.

In addition, Dr. Chiacchierini's estimates use the Oddis data, which include "possible" PM/DM (16% of the Oddis group). We believe PM/DM is too uncertain a category to be used for this investigative purpose. The Oddis manuscript did not allow possible cases to be separated from the others in enough detail for Dr. Chiacchierini's estimate. If one restricts the analysis to definite and probable cases only (and omits possible cases), the expected number would be lower than 4.253, and the number of observed cases that indicate statistical significance might drop from nine to eight). If Dr. Chiacchierini is willing to redo his analyses using only definite and probable cases, CDC staff will try to get the necessary age/sex/race specific data from Dr. Oddis.

Second, there may be underreporting of PM/DM following use of IBC because the occurrence of PM/DM subsequent to use of IBC is not established.

Dermatologists might not be notified of PM/DM by their IBC patients (median duration from last IBC use to PM/DM diagnosis among these eight cases was 6 months, and the range was 0 to 24 months), and rheumatologists diagnosing PM/DM would have no reason to link this rare disease to previous use of IBC.

Third, six of the eight cases had onset in the third to sixth month after their first exposure to IBC (the other two had onset during the first and ninth month after first exposure), suggesting a possible temporal clustering related to IBC use. If these were background cases unrelated to IBC, we would expect a more uniform distribution over time. While the temporal clustering is suggestive, it cannot be considered definitive.

Fourth, there is ample evidence in immunological models that foreign materials can elicit autoimmune problems. Therefore, it might not be surprising that the injection of a foreign (beef) product might result in a disease like PM/DM.

Thus, CDC staff view this potential association as an important problem warranting further investigation. We are willing to continue to provide help in the investigation by reviewing medical records, providing epidemiologic advice, and reporting additional IBC-associated PM/DM cases, but would urge the FDA to use its case finding and regulatory functions to determine if there are more IBC-associated cases of PM/DM. If additional cases are found, it may be necessary to look at other options, including more extensive analytic investigations.

This letter is a follow-up to the meeting held on August 29, 1990 between
representatives of the Food and Drug Administration (FDA) and Collagen
Corporation. The meeting was held to discuss the Notice of Adverse Findings
letter received by Collagen Corporation on August 9, 1990. in which the agency
took issue with the manner in which we have been interpreting the
application of the MDR regulation (21 CFR 803) to our products. At the outset of
the meeting, it was agreed that its purpose was to develop a set of criteria
which will govern our prospective MDR reporting (as well as the retrospective
This letter is meant to
two year analysis requested in the August 9 letter).
document our understanding of these criteria. Following the agency's
concurrence, these criteria will be used to determine whether an adverse
event is MDR reportable:

1.

2.

Causality

After much discussion regarding what constitutes a causal relationship. or a possible causal relationship. it was determined that unless a health care professional states that he/she thinks that the device did not or probably did not cause or contribute to the adverse event, a causal relationship will be assumed.

Antibiotic Treatment

Although the firm argued, and still believes, that antibiotic treatment
for adverse events such as abscess formation or infection is to "relieve
temporary impairment or damage". the agency representatives felt that
antibiotic treatment was to "preclude permanent damage or
impairment". Therefore, it was agreed that antibiotic treatment of an
adverse event, regardless of its discussion in the product labeling, would
constitute an MDR report. It was clarified later in a phone
conversation with Bryan Benesch that administration of "topical"
antibiotics would not trigger an MDR report.

With regard to abscess formation, it was agreed that treatments such as
steroid injections and incision and drainings are "medical intervention
to relieve temporary damage to body structure", and therefore do not
require an MDR repon since the possibility of abscess formation is
discussed in the package insen. Similarly, use of antihistamines,

3.

steroids (topical or systemic), various compresses, or NSAIDs for relief of pruritis, erythema, or pain in the treatment of an adverse event. would also not require an MDR report.

Scars

The discussion regarding scar formation following an adverse event
centered around two main issues: the permanence of the scar and the
severity of the scar. It was agreed that in the initial healing stages a
scar may be present. However, since some scars are transient, this does
not automatically represent "permanent damage to body structure".
Therefore, it was decided that a report of a scar that did not have 6
months to heal, did not yet represent "permanent damage", and was not
MDR reportable since the product labeling discusses the possibility of
scar formation. If, however, the scar is reported to be existent greater
than 6 months from initial formation, an MDR report would be required.
unless the reporting health care professional states that the scar is
cosmetically insignificant.

We believe that the discussion above fairly and accurately summarizes the agreements which we reached at our August 29 meeting with respect to the criteria for MDR reporting of incidents associated with our injectable collagen products. If your staff agrees with the description of the MDR reporting criteria set forth above, we would appreciate receiving a written statement to that effect at your earliest convenience.

As soon as that written statement is received. we shall immediately commence a re-review of our adverse event records for the last two years, as requested in the August 9 NAF letter. Because of the extensive quantity of records that must be reviewed utilizing the newly agreed upon reporting criteria, we request that we be provided with 180 days from the receipt of your written statement to conclude that re-review. Although we shall of course make every reasonable effort to conclude the re-review in less than the requested amount of time, the magnitude of the task may make that difficult.

We appreciate the courtesy extended to us by the CDRH staff at our meeting. We now believe that we have a much clearer understanding of the manner in which the staff interprets 21 CFR 803. and although we may not agree with that interpretive approach in its entirety, we are prepared to move ahead with it both prospectively and, as agreed, retrospectively for two years. We look forward to hearing from the agency on this matter in the near future so that we might begin to implement the commitments made in this letter. Please do not hesitate to contact me if any questions arise regarding this issues discussed in this letter.