|
< 1 min reading time
Dear all, For devices of similar indications as product code KNW (Gastroenterology-urology biopsy instrument), several microbiology testing labs have mentioned that 3 samples is sufficient for sterilization and cleaning efficacy validation for a 510(k) submission based on previous submissions. Does anyone have examples or advice for justification of similar sample sizes for reprocessing efficacy validation? Additionally, the FDA guidance on reprocessing medical devices states, “The number of simulated use cycles that you use [for validation studies] should be scientifically justified”. For devices of similar indications as product code KNW microbiology labs have mentioned that 6 simulated use cycles per device is commonly used. Does anyone have examples of scientific justification for the number of simulated use cycles? Any feedback or thoughts on this topic would be greatly appreciated, Kind regards, Tom source: https://www.linkedin.com/groups/78665/78665-6102525848164855810 Marked as spam
 Posted by Asked on February 8, 2016 12:00 am
187 views
|
Meet your next client here. Join our medical devices group community.
Private answer
Dr. William Landgraf
The use of 6 samples for assay acceptance has been used routinely by the FDA. It stems from the recognition that 6 assays will yield a Std Dev. which is equivalent to a 95% Confidence Interval. We have used it at Syntex Labs and elsewhere for acceptable / approvable NDA submissions
Marked as spam
|
|
Private answer
Thomas Cronin
Dr. Landgraf,
Thank you kindly for your response. I am unclear of its applicability. Are you addressing the first question and suggesting that 6 samples are needed opposed to 3? For the second question, although there will be 6 cycles, the sterility of the device will only be measured once (after the 6th cycle). Regards, Tom Marked as spam
|
|
|
Private answer
Dr. William Landgraf
the use of 6 samples related to use of conventional analytical methods. I'm not sure it relates to sterilization issues or cleaning efficiency validations. As I stated before 6 assays will yield a 95% confidence interval derived from the Std Dev.
It seems logical to me, that one would use similar requirements for other assays for qualification. This use of 6 replicates is stated and used in the USP for their qualification needs. We have always used it for NDA submission and haven't had any negative feed back from the FDA Marked as spam
|
|
|
Private answer
Thomas Cronin
Dr. Landgraf,
Thank you for following up, your input is greatly appreciated. Regards, Tom Marked as spam
|
|
Private answer
Dr. Patrick Druggan
The standard deviation for any sample represents an estimate that covers ONLY 67 % of the population. To calculate a 95 % confidence interval you need a constant value from the t-distribution for n-1 degrees of freedom. With a sample size of three (or 2 df), the t value would be 6.2, do the 95 % conficence interval would be mean ± 6.2 SD. For a sample size of 6 (5 df) the t value drops to 3.16, as there is less uncertainty in your estimate. The sample size of six is the minimum standard in all analytical methods and is expressed in ICH Q. In reality, there is little gain in reduction onuncertainty beyond a sample size of six until you reach n = 10 with 9 degrees of freedom.
Marked as spam
|
