Transdermal patches (adhesive patches placed on the skin) may also be used to deliver a steady dose through the skin and into the bloodstream. Testosterone-containing creams and gels that are applied daily to the skin are also available, but absorption is inefficient (roughly 10%, varying between individuals) and these treatments tend to be more expensive. Individuals who are especially physically active and/or bathe often may not be good candidates, since the medication can be washed off and may take up to six hours to be fully absorbed. There is also the risk that an intimate partner or child may come in contact with the application site and inadvertently dose himself or herself; children and women are highly sensitive to testosterone and can suffer unintended masculinization and health effects, even from small doses. Injection is the most common method used by individuals administering AAS for non-medical purposes. 
“Mr. Dubanovich sustained carbon monoxide poisoning and anoxic brain injury [in a 2007 house fire],” Long said in the letter. “… Mr. Dubanovich has been participating in outpatient rehabilitation… dealing with a combination of issues from traumatic brain injury (TBI) rehabilitation, cognitive rehabilitation, post-traumatic stress disorder… in recovery from opiate addiction… breathing problems… sleep apnea… Hepatitis C [treatment]… counseling for depression and anxiety.”
The BAP00089 study (BACH) was conducted in Europe and Canada, and included 1032 severe CHE patients who had no response or a transient response (initial improvement and worsening of disease despite continued treatment) to potent topical corticosteroids or were intolerant of potent topical corticosteroids. All phenotypes of CHE were included; approximately 30% of patients had hyperkeratotic only CHE, however the majority of patients had multiple phenotypes. Essentially all patients had signs of skin inflammation, comprising of erythema and/or vesicles. Treatment with alitretinoin led to a significantly higher proportion of patients with clear/almost clear hands, compared to placebo. The response was dose dependent (see Table 1).