Sequential sampling of airborne organisms

One typical form of aseptic filling in the pharmaceutical industry, alongside the use of isolators or blow–fill-seal (BFS) technology, is the aseptic filling of liquid pharmaceutical products on filling lines. According to the revised supplemental EU GMP Guide, entitled "Manufacture of Sterile Medical Products" and published in September 2003, pharmaceutical filling processes must be subject to microbiological monitoring. The EU GMP Guide states that: "Where aseptic operations are performed, monitoring should be frequent using methods such as settle plates, volumetric air and surface sampling. Sampling methods used in operation should not interfere with zone protection." For aseptic preparation it says: "Handling and filling of aseptically prepared products should be done in a grade A environment with a grade B background".¹ This means that the filling area must be microbiologically tested and that monitoring must take place during ongoing operations, but without hampering the filling process or, in the worst case scenario, contaminating it. According to the EU GMP Guide, filling areas are classified as grade A environments, i.e. a maximum bacterial load of <1 cfu/m³ air is permitted. In order to achieve this detection limit it is necessary to sample 1m³ of air. A system for sampling airborne organisms in filling line environments should thus meet the following requirements: a collection procedure should be selected that guarantees that the sterilisable collection unit (e.g., gelatin filters on a metal filter holder) at the filling point are located in a place separated from the air sampler (remote collection procedure). This not only guarantees strict separation between class A and background, but protects the filling area from any potential contamination. The monitoring assembly should be a fixed installation and allow isokinetic collection in the filling area under laminar flow conditions. In addition, it should be possible to sample the air consecutively at various sites along the filling line during operation (sequential multiple sampling). Several recent projects have confirmed the practicability of the Gelatin Membrane Filter Method. This monitoring system consists of the MD8 airscan air sampler from Sartorius (see fig. 1) and several metal filter holders for the gelatin filter disposables (see also fig 1) that are connected to the air sampler through valved piping or tubing. Each valve is allocated by means of a tubing connection to a collection head and can be driven by a PLC control system. The MD8 airscan air sampler is connected to the PLC control system by a remote control interface (see test set-up fig 2). The connection between the remote control interface and PLC control makes it possible to display the respective sampling status (sampler standby, measurement on, end of measurement or error message). This test arrangement allows sampling to take place in the filling area while the sampler remains outside the critical sector. During the measurement, the MD8 airscan air sampler draws a defined air volume through each of the selected gelatin filters at the different filling points. When the sampling of airborne organisms is completed, the gelatin filters are transferred directly to a Petri dish containing a suitable culture medium and are incubated. Afterwards, the colonies grown are counted and analysed with respect to the number of cfu/m³ air. The water-soluble gelatin filter has a pore size of 3µm. It is sterilised by gamma irradiation and is capable of retaining spores of Bac. subtilis niger by up to 99.9995% (at an inlet velocity of 0.25m/s²) and coli phages T3 by up to 99.94% (at 80% relative humidity³). The sampler can be calibrated on-site in the system and is optimised for critical areas such as class A and B cleanrooms (filling lines as described here), isolators⁴ and BFS machines⁵. The following examples illustrate a proposed concept for the determination of airborne micro-organisms on filling lines (installed in class A cleanrooms) using a collection head (fig.3) with reusable gelatin filter unit located at the filling line which, in turn, is located in a class A cleanroom. A series of collection heads on the line enable sequential monitoring during filling without the risk of contamination. After collection the gelatin filters are aseptically removed through the barrier (separation of class A from background) through glove systems and placed on agar plates and incubated. During operation several similar filling line set-ups are operated in parallel. One MD8 airscan is used per filling line and serves several collection points. The MD8 air samplers are connected to the collection heads through valves with tubing and piping (inner diameter of the tubing and valves of at least 25mm) ranging up to 20m in length. A remote control interface is integrated into the system of each MD8 airscan and filling line. All lines are operated by a central PLC control system. The MD8 airscan air sampler, the remote control interface and all valves are located in a so-called unclassified "grey area". The remote control displays the current status of the measurement by means of different coloured lamps (green = standby, yellow = measurement on or end of measurement, red = error message). The procedure for sampling airborne organisms at the different collection points along the filling line is started by the PLC control system equipped with a control panel, located in a class C cleanroom. Monitoring usually takes place once per shift and per filling point during the shift. During the first shift the measurement is carried out after in-process control (IPC) release; during the last shift, the air is measured in the last 60 mins of the filling. The MD8 airscan is set to an airflow rate of 6m³/h and a collection time of 10 mins. This is equivalent to a total collected volume of 1m³ in 10 mins at an inlet velocity of about 0.45m/s on the gelatin filter (this equals isokinetic sampling at rates of around 0.45m/s, as are common for laminar flow systems), and which is consistent with the requirements set down in the EU GMP Guide. The culture medium used consists of a tryptic soy agar (TSA). Incubation is carried out at 30-35°C for 48-72 hours, then at 20-25° C for 5-7 days. In general, no microbes are found at filling points in class A rooms (class 100). Parallel to the active sampling of airborne organisms, settle plates are placed around each filling line while the system is being set up, and then placed continually during production (two to four per line). The action level for microbiological monitoring for all filling lines is above or equal to 1cfu/m³. As part of the validation of this monitoring method, proof must be provided that an air flow amount of 6m³/h can be collected by suction during a period of 10 mins. This is required because the length of piping, the inner diameter of the pipes and the pressure loss over the gelatin filter membrane can all have a major impact on the flow rate. In this regard, it may be necessary, with the help of a calibration device, to perform calibration and, if appropriate, perform adjustment of the air sampler after it has been incorporated in the system. Moreover, collection must be possible without the danger of secondary contamination in order to avoid false-positive results. Among other things, this can be achieved by using sterile gelatin filters, by making all system components around the collection head sterilisable and by taking the gelatin filter sample out of the critical area (class A) using glove systems that prevent contamination of the sample or of the cleanroom. The Gelatin Membrane Filter Method is currently a recognised method for sampling airborne organisms in isolators and BFS machines. Furthermore, the method is ideally suited for monitoring the air around filling lines in class A cleanrooms. The concept of operating the MD8 airscan air sampler by remote control and with long piping systems allows the air sampler to be spatially separated from the collection points, thereby making it possible to consecutively monitor the microbiological quality of the air at filling lines in the filling area and at several sites during ongoing operation. Compliance with the guidelines is given and no hazards jeopardise the filling procedure. In the future, it is anticipated that this type of microbiological monitoring on filling lines, similar to isolator and BFS technology, will gain importance, and that the Gelatin Membrane Filter Method with all its positive benefits will increasingly be used in these forms of aseptic pharmaceutical filling applications.