Microbiological monitoring

According to the European GMP guidelines for sterile manufacturing à (based on 91/356/EC and 91/412/EC), microbiological monitoring in class A areas requested with high frequency. Microbiological monitoring is carried out by means of a combination of settling plates, active/volumetric air sampling and sampling of surfaces with the use of swabs or contact plates. Monitoring should be carried out routinely. In chapter 1116, USP25 presents a schedule in which class 100 areas and the supporting areas, immediate adjacent to the class 100 cleanroom should be monitored at "each operating shift" for a minimum of 1m3 of air. In addition, does it mention the preference to sample more than the 1m3? Where the controlled environment of a class 100 cleanroom, according to the FDA Guideline on Sterile Drug Products called a critical area - is only allowed to contain 1 cfu per 10 cubic feet (which is 3.5 cfu/m3) of air, "several cubic meters of air should be tested if results are to be assigned a reasonable level of precision and accuracy." There is also so much more to consider when discussing a class A cleanroom according to GMP1 where the action level is set to be 1 cfu. ISO 14698 ã the new standard under discussion, also mentions the need for frequent microbiological monitoring in the operational state and indirectly states in Part 2 that frequency of monitoring and sample volume must be high enough to assure the significance of the retrieved data. Naturally the "sampling frequency is also dependant on many factors such as the type of manufacturing process or product, facility design, amount of human interventions and of coarse historical profiles of the microbiological environmental data. No single sampling scheme is appropriate for all environments"ä.

An easy solution? In pharmaceutical practice there is a growing interest towards continuous monitoring or high volume monitoring over a longer period of time, where the average production run often comes close to three or four hours. At first sight, a logical solution would be the increase of sampling frequency during a production run, which unfortunately has some unwanted adverse effects to the situation. These effects would be the increase of human interventions associated with a growing number of individual samples, and with that a growing risk of contamination. There is also the averaging of multiple samples over a shorter time interval, which is often not accepted, leaving longer sampling times as the sole solution. Unfortunately there are some problems associated with this. Problems such as inconvenience to the sample "taker" and the people around, dehydration of media, drying and successive death of microorganisms sampled in the early stages of the monitoring session as well. In addition, the growing use of isolator techniques add some distinct problems to the equation.

As always, speed is limited… A negative effect associated with an increase of speed, could be a decrease of sampling efficiency due to the increase of bounce off and dynamic draft, especially on smaller particles, when increasing the air velocity. According to the 19th century Navier-Stokes equations, air will bounce back from the surface of the collection medium, pushing back the smaller particles that would otherwise collide with the surface. This effect is strongest on smaller particles and increases with an increase in air velocity. Every impaction instrument would somehow be affected in this way. Therefore, by the use of a higher air speed, bigger particles are also affected, potentially leading to lower efficiencies. Every particle size therefore, has its optimum speed/efficiency point. Dehydration of media is another serious issue. Every living organism must find in its environment, all of the substances required for energy generation and cellular biosynthesis. Likewise microorganisms have specific nutritional requirements for their survival and propagation. Next to these chemicals and elements, factors such as temperature, oxygen (presence or absence) and water are of imperative importance. "Water is the solvent in which the molecules of life are dissolved, and the availability of water is therefore a critical factor that affects the growth of all cells. The availability of water for a cell depends upon its presence in the atmosphere (relative humidity) or its presence in solution or a substance (water activity). The water activity (Aw) of pure H2O is 1.0 (100% water). Micro-organisms live over a range of from 1.0 to 0.7 Aw" å. This means that bacteria are very dependent on a sufficient supply of water. Therefore dehydration of the sample media has a downward effect on the survival and growth of the microorganisms collected. The validation results of a Veltek SMA atrium impaction sampler system æ, clearly underline the above statements. When testing the growth potential of four different strains of organisms, being Bacillus subtilis, Escherichia coli, Aspergillis niger and Candida albicans at different sampling times, a clear relationship between sampling time, media volume and air velocity was shown. With a 25ml media filled standard petri dish, sampling volumes of about 3m3 still showed good growth. Increasing the media volume to 32ml, sampling volumes of 6m3 were realised and still showed good growth of the tested organisms. Extending the validated maximum sampling time by 15 minutes and therefore the sampled volume or air, poor growth for all tested organisms resulted. Next to the volume of media, the choice of media supplier also made a difference. Some 32ml media allowed a three-hour sampling time (in this case approx. 6m3), where others allowed sampling times of up to four hours. Tests were carried out at room temperature (21°C) and 60% humidity, where plates where inoculated post sampling and incubated for 72 hours å. The effect of room temperature and humidity is relatively small compared to the volume of air and air velocity. Decreasing the time, by increasing the speed of air intake, does not seem to give much relief. Further increase in speed might have an adverse effect on sampling efficiency and dehydration of the media. The ISO14698 Part 1 Draft International Standard says that the impact velocity should be a "compromise between 1) being high enough to allow the entrapment of viable particles down to approximately 1µm and 2) being low enough to ensure viability of viable particles by avoiding mechanical damage or the break-up of clumps of bacteria or micromicetes" ç. So drastically increasing the sampling speed, could result in a risk of decreasing the viability chances of viable microorganisms, and therefore potentially a decrease in the sampling efficiency. Filter units do not have the problems associated with dehydration of media, but show the same risks of mechanical damage and on top of that have a higher risk of successive death because of direct dehydration of the bacteria. A higher speed would make life in the lab more difficult when trying to collect all sampled microorganisms from the filter. Increase in time Prolonging the sampling time, if and when technically and physically possible, increases the inconvenience drastically when using the widely used handheld impaction sampling devices. Often these instruments are limited in sampling time and therefore in volume. Many of them do not even allow for longer times or volume settings. Some of these units will really have to be 'in the hand' of the sample taker, which in itself increases inconvenience when extending the sampling time. Another effect can be on the battery run and recharge time, which these units use regularly. The noise created by the fans and pumps of the sampling devices in general might also attribute to the inconvenience, when systems would require running for a longer time. As discussed earlier, the increase in sampling time would have a negative effect on dehydration of media and organisms.

Isolators Isolators introduce their own series of practical problems. They normally require systems that are capable of sampling from a remote location, since instruments in general do not handle VHP in the 'pass-thru' very well. In addition, the limited space inside the isolator dictates the use of a small sampling device. The use of barrier wrapping of the media as often necessary in isolators to protect from VHP, seems to slightly contribute to some dehydration. Systems that seem to be able to handle most of the discussed problems reasonably well, are systems that allow controlling units and pumps to be well hidden away from the work floor. Simple or advanced control of the sampling cycle, with very little handling by the personnel, near or at the point of sampling is preferred. Allowing large volumes of air to be sampled without dehydration problems over a long period of time, shows high efficiency on sampling viable particles with high viability after collection.