- Explain how action levels are used to maintain a controlled environment
- Define the 5 steps to DMAIC
- Explain the 5 why’s technique for finding a root cause
- Discuss the proper documentation for documenting a deviation
In this week’s 483 we’re going to talk about investigating microbial contaminations inside the cleanroom to find out possible sources and remediate the issue. We’re also going to talk about writing up that investigation and the resulting corrective action/preventive action or CAPA as a result.
In this 483 the firm failed to adequately evaluate and test for possible causes to the contaminations.
The first observation
In this first observation, the FDA notes that the PIC identified one possible cause of the contamination, but failed to test that theory; they didn’t re-test the air in that area of the buffer room to see whether or not it was in fact the cause.
It’s a little difficult for us to perform a full investigation with the limited information we have from the 483 alone but we do know that there were 15 CFU found in a particular location; which exceeds the action level for their ISO 7 class buffer room. Here are the action levels we should refer to found in USP <797>:
USP <797> doesn’t give too much direction on how to go about doing an investigation either.
If levels measured during viable air sampling exceed the levels in Table 5, an attempt must be made to identify any microorganisms recovered to the genus level with the assistance of a microbiologist.USP <797> Data evaluation and action levels
<797> only gives us possible corrective actions for us to take once we’ve figured out the source of the contamination:
If levels measured during the viable air monitoring program exceed the levels in Table 5 for the ISO classification levels of the area sampled, the cause must be investigated and corrective action must be taken. Data collected in response to corrective actions must be reviewed to confirm that the actions taken have been effective. The corrective action plan must be dependent on the cfu count and the microorganism recovered. Some examples of corrective action include process or facility improvements, personnel training, cleaning and disinfecting, or HEPA filter replacement and/or repair.USP <797> Data evaluation and action levels
There’s a system I like to use to investigate issues like this and it’s borrowed from Six Sigma called DMAIC (pronounced Duh-May-Ick). DMAIC is an acronym for:
In the define step you’re going to do just that, define the problem. You’ll want to document how the problem was identified and create a problem statement that describes the issue in detail and what needs to occur for the problem to be considered remediated.
In this 483 we would define the problem as being a contamination of 15 CFUs (colony forming units) were found to be in the ISO 7 buffer room in one particular location.
You’re then going to set up a way to collect data that will help you identify the root cause of the problem. In this case, the data measurement collection tool we would be using would be continued air monitoring. The key with viable air sampling is that you identify the colony to the genus level at least. This will give you the origin of the microorganisms to help you narrow down where they may be coming from. But there also could be other measurements you may want to take once you’ve properly analyzed the problem.
To analyze the problem there’s a few tools that you can use. One is called the “5 Why’s.” This is simply asking the question WHY 5 times to get to the ultimate root cause of the issue.
- Why did this contamination occur? – There was equipment in the room that was running that didn’t need to be turned on.
- Why was this the cause of the contamination? – The equipment has a fan that unnecessarily disturbs the airflow
- Why does it disturb airflow during compounding operations – It doesn’t need to be turned on
- Why doesn’t it need to be turned on – it’s not necessary for compounding
- Why is the equipment running then? This piece of equipment will be turned off during compounding operations
But I wouldn’t just stop here. This gets down to one possible root cause but when it comes to contamination there are up to 6 possible causes. For this I would fill out what’s called a Fishbone (or Ishikawa) diagram.
This is where the firm didn’t go far enough to investigate the issue. As you can see there’s quite a number of factors that can contribute to a contamination and finding out the root cause is really important.
Unfortunately, the 483 didn’t detail whether or not the 15 CFU were identified so this makes the investigation nearly impossible. Remember without knowing at least the genus of the microorganisms found, there’s no way to tell where they came from (people, materials, equipment, etc.)
In the improve step you’ll want to test your possible causes of contamination. If you’ve determined it’s equipment, maybe the HEPA filters in your SEC need to be checked/changed. If it’s people, perhaps the gowning procedures aren’t sufficient for lowering the shedding of particles from your operators. Are materials being wiped off and disinfected before they’re being brought into the cleanroom? Again, all of these possible causes would be identified by the fishbone diagram.
You’ve now narrowed down that it was a number of factors that contributed to the contamination; let’s just say for this case we found that materials and equipment weren’t being wiped off and disinfected prior to entry into the cleanroom. Also, let’s say that the operators needed to change their procedure for gowning, using more protective covering so that no skin is exposed while in the cleanroom.
You then need to measure each of these solutions in some way to show that the root causes identified were in fact the issue and that the cleanroom is now under a state of control.
First and foremost, you’ll use viable air sampling, at the very least testing in that one location that had 15 CFU. If any CFUs are detected, then identify them to see what might still be causing the contamination.
Next you could ensure everyone is properly wiping off materials prior to entry into the cleanroom using a checklist for cleanroom entry. Also, maybe you’d want to sample the operators fingertips and thumb as well as parts of their gown, post-compounding (prior to them disinfecting their gloves but at the end of the procedure), to ensure the operator is using proper aseptic technique, not contaminating their gowns and using the proper behavior when inside the cleanroom.
The results are in and each of the new policies and procedures you’ve implemented worked. There have been zero CFUs detected in that location for several weeks in a row now.
However, you’re only halfway done. You need to document all of this information in the form of a corrective action/preventive action documentation form.
Corrective Action & Preventive Action
Depending on where you look you may see that there are either 8 or 10 phases to a corrective/preventive action process.
In the CAPA, you’re going to use all of the data you collected from DMAIC to document how you remediated the deviation and how you’re going to check the effectiveness of the solution(s).
Free Templates for Download
I’ve made a simple DMAIC worksheet for people to fill out to follow along with and implement solutions to test for possible causes to a contamination event. But DMAIC can be used for any number of problems, when you have a deviation of any kind I suggest going through the DMAIC process to identify the root cause of the problem, test possible solutions, implement the ultimate solution and continued monitoring to ensure control over the issue.
I’ve also included a Corrective Action Preventive Action form for download which will allow you to officially document the deviation, how it was remediated and what policies, procedures and training resulted from the investigation and analysis.