OxyFile #289

Further Evaluation of the Therapeutic Index of Ozone Used in 

V. Bocci, F. Corradeschi, Silvia Silvestri, E. Luzzi and L. Paulesu

Institute of General Physiology of University of Siena 53100 Italy

On the basis that ozone is a very reactive and potentially toxic gas, 
it has been a common wisdom to use a fairly narrow range of ozone 
concentration among 5 and 40 ug/ml of blood.  This was based on the 
evaluation of a very low rate of hemolysis in citrated blood and on 
the assumption, based on empirical data, that low ozone 
concentrations are immunostimulatory while high concentrations are 
suppressive.  After having clarified that an important mechanism of 
action of ozone is to induce cytokine production by blood munonuclear 
cells, we could define a reliable end-point and correlate ozone 
concentrations and cytokine levels after a suitable incubation of 
blood.  The main aim of this research was to achieve an effective 
immunostimulation with the least cellular toxicity.  During the last 
few years we have extensively evaluated the toxic effects by 
measuring the level of ozone-induced hemolysis (below 3,5%), possible 
formation of metahemoglobin (always absent), morphologic damage 
(absent below 80 ug/ml of ozone) as evaluated by electron microscopy, 
plasma levels of lipid hydroperoxides (increasing 3 fold after 
ozonization with 90 ug/ml ozone/ml of blood but rapidly returning to 
baseline levels) and intraerythrocytic reduced glutathione levels 
never below 10% and rapidly restored.  Unexpectedly and contrary to 
the ozone dosage usually used in autohemotherapy (from 5 to 40 ug/ml 
O3/ml of blood), we have found that we could raise the ozone levels 
and the most effective concentrations without toxicity are ranging 
between 50 and 80 ug/ml depending upon the individual plasma levels 
of anti-oxidant compounds.  The concept of correlating the production 
of cytokines of blood mononuclear cells versus ozone concentration 
has thus represented a crucial advantage and it has become an 
indispensable end-point.

It is planned to discuss several types of pathology where this 
improved approach may be beneficial.

Publish Date: 1994