Polyurethanes are made of long molecular chains. These chains are composed of different segments. There are "hard" segments and "soft" segments. The hard segments group together and forum semi-crystalline regions within the polymer chain. When you pull on a polyurethane molecular, the soft segments, which are long and spindly, uncoil while the hard segments align in the direction of the pull. The reorientation of the hard segments and semicrystalline nature are what give PU its good mechanical properties.
These soft segments are subject to chemical attack through hydrolysis or oxidation. Basically the break the chains apart and the actual molecules of polyurethane break down and you end up with fractured tubing.
There's moisture in the air. If you don't dry the resin before making the tubing, the moisture in the resin becomes energetic when it goes through the extruder (which is melting the resin to make tubing) and the tubing that you get out has a very low molecular weight and has less resistance to further degradation and will eventually fall apart. The moisture attacks the previously mentioned soft segments.
Various companies make modifications to polyurethane by replacing the chemicals that are normally used for the soft segments with other chemicals that are usually intended to be less subject to the chemical attack. Aortech replaced some of the soft segments with a silicone based material. Presumably you need to dry the resin.
My guess is that the failure mechanism isn't due to moisture from the extrusion but is caused by the catalyst used to make the polymer - probably tin. My guess is that the catalyst itself is attacking the chain and that your silicone soft segments are melting away.