Container Closure Integrity Testing

A vial seal is a compression joint. An elastomeric closure is held against the glass sealing surface by an aluminium crimp, and the residual seal force stored in the compressed rubber is what keeps the interface closed. Integrity is therefore a mechanical property of an assembly, and it can be lost without any visible damage: a chipped sealing surface, a stopper seated 0.3 mm high, a crimp applied with the wrong residual force, or a rubber formulation that has taken a compression set.

The older regulatory convention took a 0.2 micrometre leak path as the threshold of concern, on the reasoning that it corresponds to the size of the smallest organisms of interest. USP <1207> replaced that fixed figure with a product-specific maximum allowable leakage limit, which is set from what the specific contents can tolerate over their labelled storage period rather than from a single universal number. A lyophilised solid and an aqueous solution in identical vials do not have the same limit.

Both parts of that shift matter to how a result should be read. A deterministic method reports a leak rate in standard cubic centimetres per second or a pressure change in millibar, against a calibrated reference. A probabilistic method reports that dye was, or was not, seen.

The methods, and what each one actually measures

Headspace analysis is the method with the most direct bearing on a lyophilised vial, because such a vial normally leaves the freeze-dryer holding either reduced pressure or an inert backfill. Frequency-modulation spectroscopy shines a tunable diode laser through the headspace and reads the absorption line of oxygen or water vapour, so it reports the internal gas state without opening the container. A vial that has slowly equalised with room air has a headspace oxygen fraction approaching 21 percent, which is a number rather than an impression.

Why dye ingress fell out of favour

Dye ingress is intuitive and cheap, and its result depends on variables that are not controlled. Whether dye passes a marginal leak path depends on the surface tension and wetting behaviour of the dye solution, on whether the path is wetted or gas-filled, on the applied pressure differential and its duration, on the immersion orientation, and on the reader's eyesight when the vials are inspected. Two laboratories running what they describe as the same procedure can reach different conclusions on the same containers.

It also cannot produce a leak rate, so it cannot be compared against a maximum allowable leakage limit. A dye result establishes that a defect large enough for dye to pass under those particular conditions either was or was not present. It says nothing quantitative about anything smaller.

The failure mode: integrity lost only at low temperature

Butyl rubber closures have a glass transition temperature of roughly minus 55 to minus 65 degrees C, measured by dynamic mechanical analysis. Below it the elastomer stops behaving like a rubber and behaves like a stiff glassy solid: it loses the elastic recovery that lets it conform to small irregularities in the glass sealing surface. A container that seals correctly at 20 degrees C can therefore open a transient leak path while held at minus 80 degrees C, and close it again on warming.

This is the failure mode that ordinary testing is worst at catching, because the test is almost always performed at ambient temperature and the container passes. Published helium leak work on vial and stopper assemblies at cryogenic temperatures has shown measurable leakage appearing during the cold hold and disappearing on return to ambient, with no permanent defect to find afterwards. Detecting it requires the measurement to be made at temperature, not before and after. The contributing variables are the closure formulation, the residual seal force from crimping, and how far below the elastomer's glass transition the hold goes.

What integrity does and does not imply

• An integral seal establishes that gas and liquid exchange with the environment is below the measured limit. It does not establish that the contents were ever sterile.
• A retained vacuum is consistent with an intact seal, and its absence is ambiguous, because some containers are backfilled with inert gas by design.
• Integrity is a property of the assembled unit at the moment of measurement. Crimping, handling, thermal cycling and shipping all act on the seal afterwards.
• A single unit result is not a batch result. CCIT is applied as a validated sampling scheme, and the sample plan is part of the claim.

The second item is the one most often over-read. Vacuum retention is a useful informal signal and not a substitute for a calibrated method.

Where our own position stops

Container closure integrity data for those vials is not among the specifications Aurum publishes, and CCIT records for them is not among the specifications Aurum publishes. We also do not perform or claim sterility testing or bacterial endotoxin testing, so nothing in our published record speaks to the microbiological state of a container's contents. What we can describe is the container specification and the visible state of a unit on receipt: an intact crimp, a stopper seated flush, no chip on the sealing surface, and a headspace that still pulls inward. Those are observations, not a validated integrity result, and we will not present them as one.