Lab Frustration (Long rant)

Typically there will be sample valve(s) at one or more locations for RO membrane permeate for taking samples for analysis. A direct feed system is most simple and has the least chance of introducing pathogens from fomites like fingers or from air entering an atmospheric storage tank.

The sampling technique and the environmental status of the space where it is performed are critical. Between the typically “O” ring sealed valve and the outlet to a sample container there is a volume that can remain with a water droplet. If the room humidity is high, a droplet can remain long enough to start growing bacteria from the seeded ambient air or from bacteria already present in the permeate.

So to be cautious, this space must be sterilized and rinsed, prior to taking a sample. Not doing so might cause an erroneous high result for total plate count. Doing the process but not rinsing properly could cause a false negative.

Of course sterile procedure must be followed in filling a sample tube. Are the hands and the gloves still sterile? Are you wearing a mask or shield? Is the sample tube closure sterile?

Then there is the time delay and expense of sending the sample to a certified lab as quickly as possible, hopefully in a cold state. There might not be that much food for bacteria in the RO permeate, but there is plenty to feed bacteria.

So several days go by before the lab results are returned to the clinic. If the results are too high, the clinic will probably take more measures for system disinfection even as they wonder if the analysis was correct, and also decide to collect and send more samples for analysis. By this time, maybe a week has gone by since the original sample was taken. What records of the time should be analyzed?

If a direct feed system is well designed and operated properly, a sudden high TPC bacteria result and/or a high endotoxin level does not have that many potential causes. The first possible cause is a leak between the feed water to the RO membrane and the permeate. The most likely cause is leakage through an axial “O” ring between the RO element and the pressure vessel housing it. Lots of things could cause this but typically are the result of quick movement between the membrane element and the housing. Air that has entered the feed water, or sudden starts or stops, can contribute to this outcome. A face seal connection to the housing would mostly eliminate this risk.

Another possible event that could cause a leak would be a channeling through some portion of the seal between the membrane cast on thin sailcloth and the permeate fabric or netted carrier. The adhesive used is typically polyurethane but epoxy is also in use.

A vacuum condition in the loop could also cause contamination. Many connections fail to prevent ingress in a vacuum challenge.


Imagine if you will a situation wherein the cost of a pyrogen test is less than the cost of incubation on a media. Let us hope that this analysate can be produced synthetically rather than relying on the horseshoe crab. Nevertheless, at this time, the crab has a wonderful future in helping mankind understand and control pyrogens. It can yield its copper-based blood without dying for our analysis needs and continue to live. Still, mankind seems to be dedicated to destroying our environment no matter what. Sea life is not considered valuable.


It is clear that a test for pyrogens with its quick results would be chosen rather than culturing bacteria with its long delay and numerous chances for contamination.

Since the growth of a floc to measure the pyrogen level requires a very stable, vibration free location, some clinics would not be able to conduct the tests with confidence. This could be due to vibrations from highways, railroads, frequent earth tremors or building movements induced by wind or occupancy.

If a dialysis water system includes a vented storage tank, it is possible to imagine a colony of bacteria growing on a surface above the water line. If this colony falls into the water it could induce a sudden rise in both TPC and pyrogens. Likewise, any poorly designed and maintained water loop could suddenly show contamination from a dead leg growth that is released by hydraulic shocks.

There are also multiple possible ways to contaminate and possibly seed bacteria into the permeate when conducting chemical disinfection or membrane cleaning processes.

If there were confidence in the procedure, chemically disinfecting every week would further detract from justifying frequent bacteria analysis.

I would agree that performing some number of bacteria analyses (especially on the dialysate) is justified.

On line monitoring of the clinics patient’s hematocrits and blood temperatures are other measures important for detecting possible pyrogen threats.