Citrus oil components (from cold-pressed and terpeneless oils), which contribute substantially to flavor of orange juice, were shown to be absorbed into various polymeric materials [low density polyethylene (LDPE), high density polyethylene (HDPE), polypropylene (PP), and surlyn (S)] used in aseptic packaging. Equilibration of absorption took place within 4 days, while extraction of the oil constituents occurred in 24 hr when higher temperatures were used. Substantial absorption of oil constituents by LDPE and S occurred while highly crystalline polymers (PP and HDPE) did not absorb much. The degree of absorption also depended on the nature of the oil constituents. Absorption was higher for terpene constituents followed by sesquiterpenes and aldehydes. Swelling factors showed that highly crystalline polymers (HDPE and PP) swelled less than crystalline polymeric materials. Citrus oil affected the crystallinity of the polymers by lowering their melting point and percent crystallinity.

A commercially available mammary implant containing a soft polyester polyurethane foam shell has been examined for possible diamine hydrolysis products. Fourier Transform Infrared (FT-IR) analysis has shown ester and aromatic structures present in the intact foam shell. Vigorous hydrolysis conditions using NaOH have produced a mixture with large amounts of toluene diamine (TDA) present as determined by Gas Chromatography-Mass Spectrometry (GC/MS), and GC/FT-IR. There are numerous clinical reports of the degradation of this implant shell after implantation with incorporation of fragments into phagocytic cells at the implantation site. Recent reports have discussed late pain and other allergic reactions to this implant. Release of the known toxic material, TDA, is consistent with these observations and strongly suggests that an alternative material be used in this application.

Hum et al. raise important questions of risk analysis when key data is lacking, and we would welcome further discussion of this issue by other biomaterials scientists.Let us clarify our position by first stating that polyurethanes are a highly useful and promising group of biomaterials.They should continue to be used in appropriate applications. However, we also feel that the breast implant coating application described in our article was inappropriate for the polyurethane chosen. A combination of several adverse conditions (which basic literature in our field clearly discuss) leads one to expect problems. The conditions are:(1) Long-term application (hence long-term stability (2) High surface area (foam), hence more rapid surface In regular commercial use, 90% of all polyurethanes available in the market are polyester-based polyurethanes,. . . and they have a relatively low cost. However, for medical applications, polyester-based polyurethanes should not be used. Polyester-based polyurethanes undergo hydrolytic degradation when they are implanted in the body. Therefore, polyester-based polyurethanes can only be used when a prosthesis is going to be implanted for very short periods of time. For instance, for one to three hours. If the prosthesis is going to implanted for a week or longer, polyester-based polyurethanes should not be used because they will depolymerize which causes a vaiety of physiological implications. For medical applications, polyether-based polyurethanes. . . are by far the best choice.It is therefore surprising that the implant manufacturer would select this high surface area foam: polyester rather than polyether-based polyurethane.The question was raised about the use of the words "normal hydrolysis" to describe our experiments at 150°C in NaOH. Such issue is reasonable and we welcome this opportunity to be more specific. We are aware that these are extreme conditions. They were originally chosen simply to identify the components present as stated in our article. It was interesting that a biodegradable foam shell was implanted without a disclosed composition. We then went beyond simple polymer identification and concluded that "TDA is released from normal hydrolysis." We should have clearly restated that it was normal in the sense of the reaction course (diamine products), but under conditions of accelerated kinetics (much faster in our laboratory than in vivo). We stated this also on page 313 by reference to biologically initiated hydrolysis studies at lower temperatures and pH. This was a standard accelerated test method to avoid the reaction time needed at 37°C and pH = 7. Is there a question that hydrolysis is likely or are some novel products being proposed? Urethane linkages normally hydrolize to release diamines. Anything else is unexpected (i.e., not normal) but may be possible. The desire to avoid these products has also been noted by Dr. Syzcher, who states that: "Aromatic polyurethanes have been shown to form small amounts of a carcinogen, MDA, when improperly processed. Only alip...