Medical Device Daily Washington Editor

It's one thing to get a warning letter from FDA, but it's an entirely different matter when that warning letter states that a contract sterilization service omitted data from its reports to customers. However, that is precisely what FDA has alleged in a Sept. 27 warning letter to Steris (Mentor, Ohio) for a June inspection of its contract sterilization plant in Grand Prairie, Texas.

Steris was the target late last year of pressure from FDA to withdraw all of its System 1 sterilizers over unapproved changes to the 510(k) device's design, and the firm eventually capitulated despite that the agency could not document that any adverse events were attributable to the SS1 sterilizers (Medical Device Daily, Nov. 1, 2009). The latest warning letter is the only one listed in the agency's warning letter database for the company's Isomedix service in Grand Prairie, but FDA has inked warning letters to Steris's corporate operations six times since 1998.

FDA led the warning letter with a citation under finished device acceptance requirements, alleging specifically that the Steris plant had “no procedures to address how many relative humidity (Rh) monitors are placed in a sterilization load and how your firm accepts or rejects Rh data recorded during routine sterilization cycles.“ FDA states in the warning that 30 of 71 customer records from between Nov. 21, 2009 and May 16, 2010, included incomplete humidity data in reports to the customers. In one instance, the customer's specifications for humidity were for levels at between 34% and 70%, but one of the monitors registered a humidity reading of only 28.5%, which the letter alleges was not reported to the customer.

After listing another such example, FDA states that the company's humidity monitors “record relatively humidity every [redacted] during the EtO (ethylene oxide) processing, but your firm fails to have a procedure to address what time increment you will record and read Rh data for acceptance or rejection.“ Although the warning letter makes note of two pieces of correspondence from the plant's manager intended to address the inspectional findings, FDA does not address any fixes suggested by the company other than to address a gap analysis undertaken by the firm to address several citations.

The situation with humidity sensors migrated to a citation for corrective and preventive action. The warning letter states that the firm's CAPA procedure requires that a CAPA would be undertaken for non-conforming product, but that Steris undertook no such efforts for two instances of humidity readings that fell outside of specifications.

The fourth citation in the warning letter hints at the possible origin of the string of problems. According to FDA, Steris did not follow its own procedures for re-qualifying its ethylene oxide sterilizers, stating that the company conducted only one qualification cycle. However, the warning letter's listing of the required number of cycles is redacted. According to FDA, another problem with the qualification work was that temperature monitors used in the re-qualification process had not been calibrated within a redacted number of months of the procedure.

As for the firm's responses to the inspectional findings, FDA states that the calibration issue was adequately addressed by means of a re-qualification procedure for any equipment that was affected by monitors that may have been out of calibration, but FDA asserts that Steris “still needs to assess the potential effects of the out-of calibrated Rh monitors had on the product that has been released.“

For the balance of the findings, Steris is said to have undertaken a gap analysis for sterilizations undertaken up to June 30 from an unspecified starting date, but the warning letter indicates that as of the date of the warning, the agency was in possession of analytical data from only Jan. 1 to June 30. FDA reiterates that Steris did not address its out-of-calibration humidity monitors to its customers, and states that the agency would “need to evaluate your firm's results of the gap analysis review, your firm's course of action as a result of the analysis review, and the response from your customers once they are notified of your review results.“ The warning letter notes that a follow-up inspection “would also be required to assure that corrections are fully implemented and being followed to avoid future deviations.“

Steven Norton, director of communications at Steris, told Medical Device Daily that the warning letter “noted documentation and process issues related to GMP standards at that facility. We are working with FDA to resolve those issues,“ he said, adding that “no safety or efficacy issues“ had arisen as a result of conditions at Grand Prairie.

Norton added that there were no interruptions of services in connection with the corrections undertaken, and that the company is checking things at the 17 other contract sterilization facilities operated in the U.S. by Steris. “We're proactively taking a look at the other facilities to make sure those conditions are not being replicated at other sites,“ he said.

OsteoSymbionics cited for design controls

Some warning letters are well detailed while others are short, even though they're anything but sweet. The Nov. 30 warning letter to cranial implant maker OsteoSymbionics (Cleveland) is one of the latter, due in part to the fact that no correspondence from the company is discussed.

FDA states that it found fault with the company's design controls for cranial plates on four counts, including that the company failed to establish a design and development plan and failed to establish design inputs. The warning letter specifies that six of 18 design verification tests omitted a summation of test outcome despite that company procedures mandate a report comparing actual outcomes to expected outcomes.

In a following citation, the warning letter asserts that the company had not validated “the mixing, curing, casting, and cleaning processes used to manufacture the cranial plate implants,“ and did not validate sterilization of the cranial implant for more than one cycle. The sterilization procedure was also said to have been validated only for one humidity set point, 55%, despite that the implant's instructions for use indicated a humidity range of 30%-80%.

In another citation notable for its brevity, the warning letter states that one of the company's procedures for corrective and preventive action “does not list manufacturing processes and non-conformances as data sources to be analyzed to identify existing and potential causes of nonconforming product, or other quality problems.“ No other information is provided.

The warning letter also says that each of the 11 engineering change notices reviewed by the FDA investigator lacked information as to why “no verification and/or validation were required“ for the change. FDA closed the warning letter by noting simply that the company had failed to maintain a device master record for the cranial implants.

At press time, the firm had not returned a call for comment.

Mark McCarty, 703-268-5690