Cleaning validation is a favorite subject of many pharmaceutical persons including inspectors. The cleaning validation is performed by considering below criteria;

1. Visual cleanliness
2. 10 PPM
3. Toxicity
4. Dose based   &
5. Permitted daily exposure criteria

Many organizations perform cleaning validation on the basis of a worst-case approach, by evaluating the worst-case product.

This means, cleaning validation is performed only on three consecutive cleaning operations after manufacturing worst-case products, while cleaning after manufacturing of all product batches other than these worst-case product batches, is evaluated or checked only on the basis of visual clean criteria.

If we consider visual clean criteria, it is an indirect test method used for routine monitoring of the cleaning operation.

Hence, there should be a maximum focus on the visual clean criteria. But, the point is, this aspect is least focused by many organizations resulting in non-conformance. As a case study, we referred to many inspectional observations by FDA. Surprisingly, we found that many firms received the inspectional observations as,

The inspector found multiple pieces of equipment that were identified as clean, however, observed, apparent residue on the interior of the equipment utilized in the manufacture of drug products.

In another case, the inspector raised the concern that the firm failed to provide a proper study and execution plan for validation and verification of the cleaning.

In some cases, the inspector raised the concern that the firm failed to adequately clean and visually inspect the equipment. There are many cases similar to these observations.

Now, the question comes to our mind, how we can make this criterion scientifically justified? For this we need to understand some aspects and be ready with a solution, of course through the standard operating procedure or pre-approved protocol. For example-

If equipment is observed as dirty, what is the action plan? Should we directly go for recleaning?

Here, I would like to tell you that, the conclusion can be drawn on the basis of visual cleaning itself. If equipment still remains visually unclean after following the cleaning procedure, it indicates that the cleaning procedure is not suitable to remove the residue of the previous product or detergent up to the acceptable limit. In such case, we should not directly go for recleaning. First of all, this deviation/ failure needs to be investigated as per the site procedure. On the basis of investigation, risk assessment, impact assessment, and evaluation, there may be a need of revisiting the existing cleaning procedure.

The next aspect is, how many people should inspect the equipment for visual clean criteria as it is an organoleptic test?

Considering this is an organoleptic test, the decision will not be conclusive if depend on a single person. Hence, I suggest involving an odd number of personnel. For example 7 for validation of the visual clean criteria. Let me tell you, routine cleaning is also checked visually by 3 persons which is also an odd number which includes the production operator, supervisor, and QA supervisor. So, during the study, If any of the persons raise a concern about visual cleanliness, while other persons do not have any concern about visual cleanliness, In such case, the dirty part identified by the person should be sampled by a suitable method and tested for residue. If residue is detected within the acceptance limit, the visual clean criteria are valid.

If residue is detected above the acceptance limit, the visual clean criteria and also the cleaning procedure is not valid and the firm should move to alternate cleaning and indirect test method.

 The next question is How the personnel are inspecting the equipment?

The inspection procedure should be so elaborate that, the inspector should inspect the equipment from a specified height, distance, and angle and with or without the use of an additional light source which can enable the person to detect the residue. Under this scenario, the person’s eyesight also plays a prime role. Hence, the eyesight of the personnel involved in the visual inspection of equipment should also be checked.

Another question normally comes into the picture is, whether personnel eyesight is suitable to detect the contaminant at an acceptable level?

To evaluate this, one may perform the spiking study. Where, the product residue at the worst case limit should be spiked on the stainless steel plate surface, dried, and examined visually. It is good practice to store spike plates in a frame along with clean plates, so as to provide easy reference for the persons who are performing visual inspection of the clean equipment.

After visual clean criteria, there are other criteria also which are stated earlier. The further important point is the problems faced during cleaning validation and during practical implementation, and how we can solve these problems.

10ppm criteria – This criterion means, 10ppm of the previous product residue is acceptable in the next product. Many peoples calculate 10 ppm considering the batch size of all products. Here I suggest you, take the minimum batch size product and calculate 10ppm.

There is the scenario, that the theoretical worst-case limit is so stringent that it is not possible to evaluate practically. This can be managed by a variety of means. There are many ways to handle these issues. Some ways are –

–         You can go for the second smallest batch size for considering the worst-case limit. However, for the first smallest batch size product, you decide through SOP not to manufacture the worst-case product before this product. And of course, you can validate the primary worst-case product separately.

–         You can check for actual common surface area and not consider all equipment areas. Because, many peoples are considering all equipment contact surface area for calculation resulting in a very stringent limit. But in such cases, you need to have equipment train just like you prepare product train.

–         You can work out the second worst-case product. In this case, you need to validate the first worst-case product also, but by putting procedural control, you can decide not to manufacture specific next products in the future. In this way, you can validate your worst-case products also as well as your next worst-case product also.

–         And of course, if possible, you can try to develop an alternate analytical method. In such cases, this much discussion may not require.

For further details, you can contact the author.