The "unfrozen fraction" hypothesis of freezing injury to human erythrocytes: a critical examination of the evidence.

This paper examines the experimental evidence presented by Mazur and his colleagues to support their hypothesis that the survival of slowly frozen human red blood cells is primarily dependent on the fraction of water that remains unfrozen, rather than on the high concentrations of sodium chloride produced by the formation of ice. This hypothesis is in direct conflict with the general belief that freezing injury under such conditions is caused by the concentration of solutes in the solution surrounding the cells: if the "unfrozen fraction" hypothesis is true, then much of the evidence supporting that belief must be dismissed as mere coincidence. We have reexamined Mazur's data, and have suggested an alternative explanation--that cells which are initially suspended in solutions that are not isotonic differ in their susceptibility to subsequent freezing and thawing, shrunken cells being more resistant and swollen cells more susceptible than normal cells. If this is true then the data can be explained without invoking a direct effect of the unfrozen fraction, solely on the basis of changes in the concentration of the solution surrounding the cells. We cite other experimental evidence, obtained in the absence of freezing, that red blood cells do indeed possess the required property. We further argue that the known effects of variations in cooling and warming rate, and in hematocrit, are able to account for the features observed by Mazur and his colleagues in their three published studies.