Cold shoulder on cold water
"You need to balance the utilities with the production volumes," says our wet GMP Agent-In-Place. "When production volumes
go up, so must utilities capacity. We had a cold water-for-injection (WFI) system that started with hot (80 °C) WFI and then
cooled to 5 °C. Its design capacity was 500 gallons per hour. As production volumes went up, the cold WFI usage requirements
went up as well.
"The primary control for the system was to maintain the pressure in the cold WFI system. Although the system would turn on
the cooling when demand for cold WFI rose, there was no limit of usage to 500 gallons per hour. To maintain the pressure,
more hot WFI was fed into the cold WFI system. At high usage, the hot WFI would not be sufficiently cooled, and the temperatures
of the cold WFI would rise. The temperature charts showed this fluctuation in temperatures. The Quality Department reviewed
these charts, but no investigation or corrective actions were performed until after an FDA inspection. Then, the fecal matter
struck the rotating ventilator."
"We had an old WFI system, and things started to go wrong," reports our GMP Agent-In-Place. Particulates were found in the
water, and it was determined that the shaft of one of the original equipment pumps (which was 30+ years old) was grinding
down. That pump was repaired with a sleeve on the shaft, installed with set screws. Other older pumps were repaired similarly
to prevent this from recurring. (Hey, we were being proactive, weren't we?)
"The shafts were 30 years old, and no longer round," our Agent groaned, "so all the sleeves starting wearing and discharging
particulates! As a result, the site was down for four weeks waiting for delivery and extensive postinstallation flushing to
remove the particulates. Now, the shafts are on a preventive maintenance program."
"I was doing a routine audit of a contract manufacturer," notes our GMP Agent-in-Place, "and I cited issues with the WFI system.
"The WFI monitoring data were frequently out of limits for total count and for conductivity. When I brought this to the attention
of the audit host, his response was, 'The products were tested and met their specifications. These monitoring tests are really
just an alert limit, no action is necessary. You are worked up over a nonissue.' The formal response said basically the same
thing: no action would be taken.
"The next year, I performed a follow-up audit. When I looked at the WFI system, I realized that a totally new system had been
installed! The engineering and contracting process would have taken more than a year, and my host clearly knew and lied about
the planned changes, and actions had been planned and taken.
"When I asked the host about the changes, he indicated that the end-product testing supported the product that was manufactured
in both cases and said 'So, what is your problem?'
"So, what other information (such as the batch record) was a lie? I wondered."
Pharmaceutical Technology's monthly "Agent-in-Place" column distills true-life cautionary tales from the secret files of Control, a senior compliance
officer. If you have a story of clueless operators, oblivious management, inopportune lapses of judgment, or Murphy's Law
in action, please send it to Control at:
We won't use any names, but if we do use your tale of disaster, courage, or just plain weirdness, Control will send you a
coveted Pharmaceutical Technology t-shirt.