Sterility Testing And Sterile Filtration Are Not The Same Record

The two terms get used as if they were interchangeable, and they describe activities at opposite ends of a process. Sterile filtration happens to a bulk solution before it reaches a container. A sterility test happens to sealed containers afterwards, on a sample drawn from the finished lot. The first is a control step. The second is a verification step performed on the output of everything that came after the first.

That gap matters because most contamination events in a filled lot do not originate at the membrane. They originate downstream of it: at a tubing connection, at a filling nozzle, at a stopper that was not seated, in the air above an open vial. A filtration record is silent about all of those, because the filter was doing its job correctly while they happened.

Neither record is available for Aurum material. We do not perform or claim sterility testing, and no lot is represented as sterile. What follows is a description of what the two records are and what each one can support, not a description of our own material.

What a sterilising-grade filtration record contains

A membrane earns the sterilising-grade designation through a bacterial challenge, not through its nominal pore rating. The reference method challenges the membrane with 10^7 colony-forming units of Brevundimonas diminuta per square centimetre of effective filtration area and requires a sterile filtrate. A 0.22 micrometre rating is shorthand for having passed that challenge, not a sieve dimension that was measured directly.

Because the membrane cannot be inspected after use without destroying it, integrity is inferred from a gas-liquid test. A bubble point determination wets the membrane and raises gas pressure on the upstream side until gas breaks through the largest wetted pore. Hydrophilic 0.22 micrometre membranes are commonly specified with a minimum bubble point in the range of 3.0 to 3.8 bar using water at 20 degrees Celsius, with the exact figure set by the membrane manufacturer's validation data for that material and lot.

What a sterility test actually measures

The compendial sterility test is a growth test. Containers are opened under aseptic conditions and their contents are either passed through a membrane that is then transferred to media, or inoculated directly into media. Two media are used because they cover different organisms: a thioglycollate medium incubated at 30 to 35 degrees Celsius, and a soybean-casein digest medium incubated at 20 to 25 degrees Celsius. Incubation runs for not less than 14 days, and the result is binary: visible growth, or no visible growth.

Two controls sit alongside it and are as important as the test itself. A growth promotion control demonstrates that each medium lot supports growth of specified challenge organisms, which is what makes a no-growth result meaningful. A method suitability check demonstrates that the material under test does not itself inhibit growth, because a solution that suppresses the organism produces a clean-looking result for the wrong reason.

The sample size is the limit nobody quotes

Compendial sampling for a lot of more than 500 containers is 20 containers. That number is fixed by the method, not scaled to the risk, and its detection power can be calculated directly. If 0.1 percent of a 10,000-container lot were contaminated, the probability that at least one of 20 randomly drawn containers carries the contamination is 1 minus 0.999 raised to the 20th power, which is about 2 percent. Ninety-eight times out of a hundred, that lot passes.

This is why sterility is handled in pharmacopoeial practice as something built by process control and confirmed in part by the test, rather than something the test establishes on its own. The filtration record, the environmental monitoring record and the container closure integrity record carry most of the weight. The sterility test is the last and statistically weakest of them.

Where bioburden enters the picture is upstream: a low and characterised microbial load before filtration is what makes a single sterilising-grade membrane a defensible control point in the first place.

The failure mode: a clean filter record and a contaminated vial

The characteristic failure is a post-use bubble point that passes while a single vial in the lot grows. The membrane was intact, the pressure held, the record is genuine. Contamination entered at a joint between the filter housing and the filling line, or from a stopper that sat exposed, and the organism was never in front of the membrane at all.

It shows up as growth in one container out of twenty, in one medium, often late in the 14-day incubation because the initial count was low. And it is untraceable through the filtration record, because nothing in that record was ever measuring the place where the failure occurred. Investigating it requires the environmental and assembly records, which is precisely why those are kept.

What neither record covers

Neither test says anything about bacterial endotoxin. Endotoxin is a heat-stable lipopolysaccharide fragment that survives filtration and passes a 0.22 micrometre membrane freely, and dead organisms release it. A container can be sterile by growth test and still carry endotoxin, which is why that is measured by a separate assay with its own units.

• A filtration record is an equipment and parameter record. It does not describe the filled container.
• A sterility test is a growth result on a sample. It does not describe containers outside that sample.
• Neither addresses endotoxin, which requires a separate quantitative assay.
• Neither addresses container closure integrity, which is tested by physical or microbiological methods of its own.

Aurum's own position, stated plainly

We do not perform sterility testing and do not claim it. We do not perform endotoxin testing and do not claim it. Our published figures are identity and purity figures: purity by reverse-phase HPLC with ultraviolet detection at 214 nm, and identity by mass spectrometry. Those are chemical records. They contain no microbiological information of any kind, and no arrangement of them produces a sterility statement.

A related distinction applies to the purity figure itself, which is HPLC peak area percentage and not net peptide content by mass.

Common questions

Does a 0.22 micrometre filter make a solution sterile?

It removes organisms that the membrane retains, under the pressure, area and time conditions the membrane was validated for. Whether the container downstream of it is sterile is a separate question answered by separate records.

Is a sterility test the stronger of the two records?

It examines the right object, the finished container, but on a sample of 20 containers it has weak detection power for low contamination rates. Its value comes from sitting alongside process records, not from replacing them.

Why two different media and two temperatures?

Coverage. Thioglycollate medium at 30 to 35 degrees Celsius supports anaerobes and aerobes, soybean-casein digest medium at 20 to 25 degrees Celsius supports fungi and aerobes. A single condition would miss a class of organisms.

Can a lyophilized powder be sterility tested?

Yes, by dissolving the contents in a suitable diluent and passing that through a membrane, or by direct inoculation. The method suitability check matters more here, because the diluent and the solids can both affect growth.

Does a filtration record expire?

The record does not, but its applicability is bounded by the validated run conditions. Filtration beyond the validated time or above the validated differential pressure falls outside the conditions the retention claim was established under.