Various approaches are available to investigate cryoprotectant toxicity, ranging from theoretical work in organic chemistry to  cryopreservation of complete animals. Because resuscitation of complex organisms after cryopreservation is not feasible at the moment, such investigations need to be confined to viability assays of individual cells and tissues or ultrastructural  studies.

One simple model that allows for “high throughput” investigations of cryoprotectant toxicity are red blood cells (erythrocytes). Although the toxic effects of various cryoprotective agents may differ between red blood cells, other cells, and organized tissues, positive results in a red blood cell model can be considered the first experimental hurdle that needs to be cleared before the agent is considered for testing in other models.  Because red blood cells are widely available for research, this model eliminates the need for animal experiments for initial studies. It also allows researchers to investigate human cells. Other advantages include the reduced complexity of the model  (packed red blood cells can be obtained as an off-the-shelf product) and lower costs.

Red bloods cells can be subjected to a number of different tests after exposing them to  a cryoprotective agent. The most basic test is gross observation of the red blood cells in a cryoprotectant solution. When high concentrations of a cryoprotectant are used (such as in vitrification), a stepwise approach is necessary to avoid  osmotic  injury. If a cryoprotectant solution is extremely toxic rapid hemolysis will follow, which often can be observed by a noticeable change of the color of the solution,  red cell debris sinking to the bottom of the test tube, or negligible difference between the pellet (if there is one at all) and the supernatant after centrifugation. But if the researcher is interested in agents that are not extremely toxic, or wants to compare variants of  similar agents with each other, quantitative methods and detailed observations are required using respectively spectrophotometry and light microscopy.

In 1996, Bakaltcheva et al. used the red blood cell model for an elegant and thoughtful study  of cryoprotective toxicity. The authors did not only use spectrophotometry to measure hemolysis  but also used microscopy to study the morphology of the red blood cell after exposure to various agents at different temperatures. The results of these different measurements were in turn correlated with each other in order to determine if there are general properties  affecting cryoprotectant toxicity. The authors propose that reduced toxicity can be achieved by keeping the dialectric constant of the medium and membrane close to that of an aqueous solution without solutes.  These findings can also explain why cryoprotective mixtures  of various agents (such as DMSO and formamide) can reduce toxicity.  A general rule of thumb for formulating vitrification agents with reduced toxicity seems to be to maintain most properties of water but eliminating the posibility of ice formation. It should not be a surprise that such an approach has guided the choice of solvents in areas such as cryoenzymology.