Member Spotlight: David Artiss, ISPE Founding Member
When Dave became involved with the pharmaceutical industry in the early 1970’s, pharma engineers had limited resources for guidance when confronted with a problem. Dave joined ISPE in its first year to help the industry develop this type of support. We talked to Dave about his experiences during the past 40 years, and his thoughts about the changes that have been made during that time.
Wow! You have been with the Society for 40 years! Can you give me a brief summary of your involvement with ISPE and share some of the more interesting experiences and successes you have had with the Society during your lengthy tenure?
When I started my involvement with the pharmaceutical industry in the early 1970’s, pharmaceutical engineers had limited resources for help and guidance when confronted with a problem. One resource that would have been of great value was to be able to discuss issues and concerns with their peers in other companies. However, very few companies would approve of their engineers admitting to any problems to any outside company. Therefore, experience that would be invaluable to others in the industry was not widely shared.
At that time the main source of piping materials, components and construction methods was the food and dairy industry, which were using the 3A standards. However, there is a significant difference between, for example, a hot WFI system that is in continuous use and a milk transfer piping system that is shut down, often daily, and frequently and cleaned and sanitized.
During the mid-1970s the pharmaceutical industry was going through a major upheaval caused by the FDA emphasis on the total control of the critical systems, from design through construction, start-up and qualification, and there was no national / international forum for the discussion of engineering issues and the dissemination of new ideas. Many pharmaceutical companies had to develop new methods to ensure compliance with FDA concerns and recommended practices. I had the good fortune at that time of providing the consulting engineering component of the Process Development group of a progressive pharmaceutical company that was determined to upgrade their systems to the best degree possible.
One significant area of concern was the assurance that welded systems were suitable and that all welds, including those in barely accessible locations, met all the required quality parameters for critical pharmaceutical piping systems. While there were some “semi-automatic” welding machines in use, they were used mainly for bench or easily accessible welds and could not be used in tight spaces or for elevated piping. We had to look elsewhere for better equipment, and we found that the aerospace industry had fully automatic welding equipment that provided complete external purge coverage and could be used in tight spaces. We researched and purchased a fully automatic autogenous welding machine, but we had to have special welding heads made to fit the standard OD tubing sizes.
The need for weld quality assurance required suitable weld inspection equipment and procedures. These were either unavailable or poorly suited to determine that the welds met pharmaceutical acceptance criteria. Inspection equipment currently in use in other industries did not meet our needs. We persuaded borescope manufacturers to develop instruments that could inspect welded 20’ long sections and welds that were not accessible by other inspection means. This type of inspection equipment has been improved over the years and is now standard practice in the pharmaceutical industry.
Also during this period, we initiated the process of electropolishing of the interior of stainless steel piping, not just to produce a shiny finish but a means of permitting the inspection of the quality of the longitudinal weld seam. The electropolishing removed the mechanical polish and exposed the weld and any inclusions in the material. We found that many welds in tubing and piping components did not meet the weld quality specifications and the tubing was rejected. After several rejections the tubing manufacturers improved their procedures to provide a better quality product to the pharmaceutical user.
For a subsequent client I had the privilege of being the consulting engineer and project manager for the first two facilities that installed PW, WFI and PS systems that had no manual welds. Every weld was made using the fully automatic welding process with 100% internal weld inspection and documentation. This construction method gained favor over the years and is now standard practice in the pharmaceutical industry. In order to fully utilize the automatic welding process special piping fittings with extended straight lengths / tangents were required. I persuaded one company to manufacture these “special” fittings and used them in subsequent installations; these fittings are now standard in the pharmaceutical industry.
At that time the smallest diameter OD tubing that was available was 1” but many small scale processes needed a smaller size. This led to the development of ¾” and ½” diameter tubing and smaller welding machine heads were developed for these sizes.
ISPE grew out of a need to provide a forum for the discussion and dissemination of these new technologies and methods to the entire pharmaceutical industry.
I have been active in the dissemination of the following concepts by means of presentations, training and involvement in special interest groups and at meetings of ISPE as well as to the PDA, PMA (now PhRMA) and to the FDA.
- PW, WFI and PS systems design, construction and validation
- Automatic welding
- Borescope weld inspection
- The value of electropolishing of stainless steel components and tubing
How did you come to dedicate your career to the pharma industry?
I started my career in the paper, mining, and petro-chemical industries in both design and construction of complex piping systems with emphasis on water treatment / purification systems, steam and condensate systems. In these long-established industries, the design and construction materials and methods were well established, and the petro-chemical industry had published and readily available recommended practices for the design and construction of almost every conceivable piping design project. In early 1970s I decided that it was an ideal time to work in a different and more technically challenging industry, so I took a position as a contract piping and facilities engineer for a large pharmaceutical company.
It was during this time that the FDA took a different approach to the oversight of the pharmaceutical manufacturing industry. Prior to this, the main emphasis of quality control was aimed at end product testing. But in 1976 the FDA developed a proposed cGMP for Large Volume Parenterals (LVPs) which was aimed at assuring total control of the manufacturing process from design through construction, operation, sampling, monitoring and maintenance of critical equipment and systems. The FDA emphasized that the only way to assure that the small sample of finished product that was tested was truly representative of the whole batch was to validate the entire process. The FDA proposed regulations were very detailed to the point of specifying, for example, that stainless steel type 316L be used for critical utility and process piping systems. These proposed cGMPs were not distributed but were discussed with some pharmaceutical manufacturers. Based on the initial comments the proposals were modified slightly, for example “type 316L stainless steel” became “stainless steel (nonrusting grade)” and then the proposals were issued in 1976 with 120 days for public comments. There were not many US Large Volume Parenteral manufacturers in the mid 1970s and while they were objecting to the very specific recommendations, the FDA began requesting comments as to why the proposed Large Volume Parenteral regulations should not also apply to Small Volume Parenteral (SVP) drug products. There were many, many more Small Volume Parenteral than Large Volume Parenteral manufacturers.
The industry, through the major professional associations PDA and PhRMA on behalf of their members, argued against these very restrictive regulations, but there were few who would publicly admit that their systems were not in technical compliance with the proposed regulations. While many pharmaceutical manufacturers addressed these proposed regulations there was little industry-wide discussion as to how to comply. The PDA and PhRMA are excellent industry associations, they provide a forum for the discussion of many of the challenges facing the industry, but there was a need for a forum more focused on specific engineering concerns related to the industry and with discussion of methods. This need led to the formation of ISPE. These proposed cGMPs were never promulgated into law, but many of the recommendations are in use today as accepted practice.
What are your thoughts on the state of the pharmaceutical industry today as opposed to how it was in 1980?
Many of the methods in use were based on age-old practices that have been improved by the application of scientific methods. A significant advance was the use of analytical methods for the determination of water quality, in place of the wet chemistry methods previously in use.
ISPE’s Good Practice Guides are very helpful.
I have always advocated that the qualification process should include Design Qualification and Construction Qualification in addition to the traditional IQ, OQ and PQ. I am pleased to see that design qualification is now a general recommendation.
Making the construction and installation activities part of a formal Construction Qualification provides more assurance of construction compliance. Contemporaneous documentation of incoming materials control, materials control during construction, construction procedures and physical installation will provide documentation that is suitable for subsequent qualification purposes.
What do you, as one of the original Members of ISPE, hope to see happen with the organization in the coming years?
Continue with the advancement in design and communication of new technologies and the implementation of existing and well proven methods.
There are certain what I call “magic numbers” that are in use in the pharmaceutical industry. The original source of these numbers is often shrouded in the past, but these numbers are used to support current design practices, for example:
- Pipeline velocity and the misunderstanding that the FDA term “turbulent flow” means the same as turbulent flow as defined by Reynolds Number
- Pipeline dead-legs definition and rationale
- Sterilization requires a minimum of 121ᵒC
There have been many individual ISPE members and scientists who have publicly questioned or evaluated these numbers but there may be a need for an in-depth review by an industry group such as ISPE, to finally distinguish between myth and scientific fact and to document the rationale for these legacy numbers.
Anything else you would like to share with the ISPE Community?
Systems design considerations should not only include equipment and construction costs. The cost of monitoring, sample processing, sanitization, and maintenance should also be considered in the design selection and not left to the operations / facilities to calculate.
It should always be remembered by engineers that microorganisms do not follow the laws of physics.
Risk assessment should be more specific to the equipment and systems features and limitations rather than be mainly a statistical analysis.
Design rational should be documented. The thought process and decision basis should be recorded, so that reviewers or future systems evaluation will know why the system was designed in a particular manner. This will provide a baseline to be used when considering design changes in the future.
Many times, when auditing facilities and critical utility systems I have encountered designs that appeared to be so different that one might question the engineer’s state of mind. But whoever did the design had a reason that seemed to be a good idea at the time. I try to determine the reason why, as this may still be a valid design consideration and we would not want to change the design and risk unexpected repercussions, without that knowledge. If there had been a written rationale for the design that is easy to understand, this would have been very helpful. A good place for this would be the URS or FDS.
