Steam sterilizer validation: Bowie-Dick, heat distribution and heat penetration

A porous-load steam sterilizer looks simple on paper: remove the air, let saturated steam reach every surface, hold it at temperature, dry the load. Validation is about proving each of those steps happens in every corner of every load you run. Most failed qualifications we are asked to look at did not fail because the machine was bad. They failed because a test was set up in a way that could not show what it was meant to show.

Here is how the main tests fit together, with the points where things usually go wrong.

The standards in the background

Two documents do most of the work. EN 285 covers the large steam sterilizer itself and the tests the manufacturer and user carry out on it. ISO 17665 covers the development, validation and routine control of the moist heat process. Pharmacopoeial texts and your own regulator's expectations sit on top of these, but if your protocol follows EN 285 and ISO 17665 sensibly, you are on firm ground.

Air removal: Bowie-Dick and vacuum leak

Air is the enemy of steam sterilization. A pocket of air inside a pack insulates the surface underneath it, and a load can show the right chamber temperature while the inside of a towel pack never sees steam.

The Bowie-Dick test checks that air removal and steam penetration work in a standard test pack. It is run daily in routine use, at the start of the day, in an empty chamber. A uniform colour change on the indicator sheet means the air was removed. A pale patch in the centre means it was not.

The vacuum leak test checks that air is not leaking back into the chamber during the vacuum phases. EN 285 sets the limit at a pressure rise of no more than 1.3 mbar per minute. A machine that passes Bowie-Dick in the morning but has a slow leak can still fail by the afternoon, which is why the two tests go together.

Heat distribution: the empty chamber

Heat distribution studies are run with an empty chamber, or with a minimal load, and sensors spread through the usable space, including the corners, near the door and near the drain. The drain is usually the coldest point and is where the reference sensor for the cycle sits.

For a 121 °C cycle, the sterilization temperature band is 121 to 124 °C. During the holding time every sensor should stay inside that band, and the spread between sensors should be small. EN 285 also limits the equilibration time, the time from the drain reaching sterilization temperature to every point in the load reaching it: 15 seconds for chambers up to 800 litres, 30 seconds for larger ones.

Heat penetration: the real loads

Heat penetration is where validation earns its keep. Sensors go inside the items that are hardest to heat: the centre of the biggest wrapped pack, inside a filter housing, the lumen of tubing, the cold spot of a container. Each defined loading pattern needs its own study, typically three runs.

Biological indicators, usually spores of Geobacillus stearothermophilus with a population of around 10⁶, go next to the sensors in the worst-case positions. A sensor tells you the temperature was reached. A spore strip tells you it was enough.

F0 and lethality

F0 expresses the lethality of the cycle as an equivalent number of minutes at 121.1 °C, with a z-value of 10 °C. It is a useful way to compare positions and runs, and most validation systems calculate it automatically. It does not replace the requirement to hold the temperature band for the full holding time. A load that reaches F0 of 15 through a long heat-up but never holds at 121 °C has not run the cycle that was validated.

Sensors and calibration

Every sensor should be calibrated against a traceable reference before the study and verified again after it, at a temperature close to the process temperature. If a sensor drifts during the study, the data from it cannot be used, and losing the coldest position to a drifted sensor is a common reason for repeating a run.

What usually sends a study back

• Loading patterns in the protocol that do not match how the production team actually loads the chamber.
• The worst-case item chosen by guesswork rather than by a short heat-up trial.
• Post-study calibration skipped or done at the wrong temperature.
• Dryness not checked, so wet packs show up in routine use.
• No requalification trigger defined for changes to the load, the steam supply or the controller.

Get the loading patterns agreed with production before the protocol is approved. It is the single change that saves the most repeat runs.