I'm guessing the fact that it's an inert gas is important as it won't be a risk (in case of fire or toxicity) in case the enclosing is broken.
Also I think the density was presumed to be at room temperature, reasonable pressure. Under these circumstances He might provide the least drag. Having a pressurized enclosure might make the drive more expensive as stronger materials need to be used, and it would also make it more dangerous in case of mishandling, so I think there was a limited range of pressure they could maintain in the enclosure.
Nitrogen is also inert enough for such use, it's easily contained and it doesn't require such measures.
>>>Having a pressurized enclosure might make the drive more expensive as stronger materials need to be used, and it would also make it more dangerous in case of mishandling<<<
Actually containing He at all would require these measures anyway, as He diffuses through solids much faster than air and as such cannot be easily contained. To quote the wikipedia article;
>>>One industrial application for helium is leak detection. Because helium diffuses through solids three times faster than air, it is used as a tracer gas to detect leaks in high-vacuum equipment (such as cryogenic tanks) and high-pressure containers. The tested object is placed in a chamber, which is then evacuated and filled with helium. The helium that escapes through the leaks is detected by a sensitive device (helium mass spectrometer), even at the leak rates as small as 10−9 mbar·L/s (10−10 Pa·m3/s). The measurement procedure is normally automatic and is called helium integral test. A simpler procedure is to fill the tested object with helium and to manually search for leaks with a hand-held device.
Helium leaks through cracks should not be confused with gas permeation through a bulk material. While helium has documented permeation constants (thus a calculable permeation rate) through glasses, ceramics, and synthetic materials, inert gases such as helium will not permeate most bulk metals.<<<
Further, high pressure containers are dangerous as they might have an outward failure. That is unlikely to be the case with a drive that is at say .5 atm as the pressure is directed inwards, so the worst case over is most probably a drive filled with air or a slightly crumpled drive. (as a thought experiment think about how submarines implode instead of exploding when the hull encounters pressures beyond what it can handle)
Also I think the density was presumed to be at room temperature, reasonable pressure. Under these circumstances He might provide the least drag. Having a pressurized enclosure might make the drive more expensive as stronger materials need to be used, and it would also make it more dangerous in case of mishandling, so I think there was a limited range of pressure they could maintain in the enclosure.