Effects of Storage Time on Total Protein and Globulin Concentrations in Bovine Fresh Frozen Plasma Obtained for Transfusion

Proverbio, D.; Spada, E.; Baggiani, L.; Bagnagatti De Giorgi, G.; Roggero, N.; Belloli, A.; Pravettoni, D.; Perego, R.

Serum proteins have been reported to remain stable in storage [16] [17] [18] . The effect of duration and temperature of storage on canine plasma and serum constituents has been documented [16] . Most analytes show no or very mild change in canine plasma [16] and concentrations of glucose and total proteins in human samples have been shown to be unaffected by repeated freeze-thaw cycles [27] . Our results concur with results of previous studies that evaluated the effects of freezing and storage on routine assays of total protein and protein fractions in other animals [16] [17] [18] . In those studies, however, blood samples were collected by means of direct venipuncture, using a syringe and needle, and placed in glass or plastic tubes (either empty or containing lithium-heparin), conditions which do not mimic those used for obtaining and storing plasma for transfusion. In our study, when the value of protein fractions was expressed as a percentage of TP, no difference was found between fresh plasma and thawed plasma at any storage time. However, when the values were expressed in g/dL, significant differences were seen, with a number of analytes occurring at higher concentrations in frozen/thawed serum than in fresh serum, for example, mean concentration of TP between 0 months and 1 month (6,4%), 0 and 6 months (6.4%), and 0 and 12 months (3.5%); ALB concentration between 0 months and 1 month, 0 and 6 months, and 0 and 12 months (7.6%); and beta-2 concentration between 0 and 6 months and 0 and 12 months (7.5% and 13.2%, resp.). Reynolds et al. [17] reported similar results in a study examining effects on canine plasma protein after repeated freeze-thaw cycles. In our study the change in the mean values exceeded "accepted value of within subject biological variation" and "total error allowed for people" [28] , and in the case of beta globulin the percentage of variation exceeded 10%. In accordance with the interpretation of Cray et al. [18] we accepted that differences greater than 10% of the minimal difference are likely to have clinical significance in the interpretation of the biochemical analysis.