Validation of 'non-culture' methods to detect bacterial contamination of platelets

A colleague in Kentucky reports that their blood bank is currently doing validation of methods to detect bacterial contamination in random platelet concentrates, including swirling, pH, and glucose measurements. During the validation of the aforementioned methods, they have been using bacterial species and dilutions found in an August 2002 Transfusion article by Werch JB et al entitled 'Detecting bacteria in platelet concentrates by use of reagent strips', and have innoculated each trial random platelet with one of four common bacteria, including coagulase negative Staph, Klebsiella pneumonia, Serratia marcesans, and Staph aureus. Due to inventory management challenges in their region, their platelets only had an average of 48 hours until expiration. The question that they would like answered is should they hold the "study" platelet concentrates past the normal expiration dates to test for bacterial growth at 72 hours, in order to properly validate their detection system? And if the pH, glucose, or swirling tests do not give results indicative of bacterial growth within the expiration dates, should they hold the platelets past the expiration dates and continue with the testing for 24 to 48 more hours in order to obtain appropriate validation data?

A knowledgeable 'blood banker from a small town in North Carolina' responds: "In order to validate the use of metabolic markers of bacterial overgrowth such as pH and/or glucose, it is important to first define what is normal with the platelet products in use and the test platform. This can be accomplished most easily by studying a number of outdated products (e.g. twenty 'day 6' platelets). Taking the mean value minus 2 standard deviations is one way to set such a cutoff. Once you have established what is considered normal with sterile units, you can then go on to validate the ability to detect units that are actual contaminated with bacteria. Ideally, this should be done with inoculated day zero platelets. However, given that such platelets are generally not readily available, one could inoculate an indate unit and test it over the next few days. However, one should not exceed the age for which you have defined your limits of normal. Therefore, it is not reasonable to inoculate a day 3 or 4 unit and follow it until days 7 or 8 of storage as the pH and glucose may already be below your established cutoffs for sterile indate units. The way around this problem is to inoculate the freshest units you can obtain, with FAST-growing organisms such as Bacillus cereus or Staph aureus. Bacillus cereus will typically reach plateau growth within 24 hours. A note of