No process can prove that every unit in a batch is sterile, because proving it would mean testing every unit destructively. Sterility is therefore expressed as a probability: the sterility assurance level, or SAL, is the probability that a single unit remains non-sterile after the sterilization process. The standard for terminally sterilized products is an SAL of 10−6: no more than one unit in a million.
How 10−6 is achieved
Microbial killing by heat follows a logarithmic curve: each fixed interval of exposure kills the same fraction of survivors. The interval that reduces the population tenfold is the D-value. A cycle is designed by starting from the number of organisms on the product before sterilization (the bioburden) and adding enough D-values to bring the expected survivors down to 10−6 per unit. In the common “overkill” approach the cycle delivers at least a 12-log reduction of a highly resistant reference organism, expressed for steam as an F0 of at least 12 minutes at 121 °C, without needing to know the actual bioburden precisely. See what is bioburden?
Terminal sterilization vs. aseptic processing
| Terminal sterilization | Aseptic processing | |
|---|---|---|
| What happens | The product is filled, sealed and then sterilized in its final container | The product is sterile-filtered and filled into sterilized containers under controlled conditions; nothing is sterilized afterwards |
| Assurance | A calculated SAL, typically 10−6 | No SAL can be calculated; assurance comes from validated facilities, filters, procedures and simulation runs (media fills) |
| Applies to | Heat-stable products such as water in glass vials | Heat-sensitive products |
Bacteriostatic water is heat-stable and is usually terminally sterilized by steam in its sealed vial, with sterile filtration used earlier in the process; the manufacturer’s method is described in its documentation. See how bacteriostatic water is manufactured and sterile filtration explained.
Why the sterility test cannot prove sterility
The USP <71> sterility test on a batch samples a small number of units, typically 20. If one unit in a thousand were contaminated, a 20-unit sample would detect it only about 2% of the time. The test is therefore a gross check that the process did not fail badly, and the real assurance comes from the validated process and its records. This is why a certificate of analysis reports the sterility test as a pass, while the manufacturer’s process validation is what stands behind the sterile claim. See sterility testing (USP <71>) and how to read a COA.
What SAL means for a laboratory
The assurance applies to the sealed vial as manufactured. It says nothing about a vial once opened, and it is why the closure has to be intact for the claim to hold. See container closure integrity and is bacteriostatic water sterile?
All BAC Water Supply products are supplied for laboratory research use only and are not for human or veterinary use.