I think you're going to be surprised. Channel stuffing probably had a minor effect on the 4th quarter results
if you think destocking was minimal in Q4, then the monthly end user sales come to 170/3 = 56.67M per month. if you assume they hit this run rate right out of the gate after launch (which we know did not happen at the retail level based on script data - and retail accounts for a non-trivial amount of total sales (35% or so), then end user sales in Q3 were 113M. that means there was still 292-113 = 179M in inventory at end of Q3. This equates to 179/56.67 = 3.15 months or 13.5 weeks of inventory. so for your assumption to be correct NVS must be maintaining this same level of inventory at the end of Q4.
i don't think NVS is maintaining 13+ weeks inventory. In fact, I think destocking accounted for close to a month's worth of sales in Q4. Here's how i do the math:
Q3 end user sales about 140M (65M first 39 days and 75M next 30 days*), leaving 292-140 = 152M in the channel Q4 end user sales 240M, of which 170 sold to wholesalers and booked, and 70M drawn down from inventory, leaving 152-70=82M worth of stock in the channel, or just over 1 month's worth of inventory (240M/3=80 per month in Q4)
this equates to a run rate of 960M annually, which is consistent with a 40% market share discounted 15% from 2009 US lovenox sales (2,800M x .85 x .4 x .25 = 238M). i think this estimate is even a tad conservative since NVS hinted they may be slightly above 40%, and we know the lovenox market grew in volume from 2009
*calculated using script sales over the quarter, with assumption that scripts account for 35% of total sales and hospital sales stable after 1 week post launch