Having now read the guidance (and much of the evidence as well), it's pretty clear that NICE probably didn't discuss this at all. I say this because issues of how to detect smokers weren't something that NICE asked the University of Sheffield to look at - their reviews were in different areas.
There are clearly pluses and minuses for directing health professionals to do a CO test. As quoted by Coalition above, the evidence does suggest that there's a problem of pregnant smokers lying so that fewer smokers get picked up. The evidence also suggests that health professionals lack the guts (my words) to even ask pregnant women if they feel it might be uncomfortable. There are TWO problems with not using the test and it's not just about pregnant smokers.
If you take a look at the guidance, MW are not recommended to ask women. (Although it would probably be better if they were recommended not to ask women.) If you don't ask, they don't lie, and the CO test isn't a test of truthfullness.
Some of those currently lying about smoking are likely to be doing so because they don't want to quit; others would like to quit and are probably embarrassed. The guidance is written to suggest that every pregnant smoker should get support to quit, in the knowledge that many won't take it up or even try. NICE's public health experts are making the judgment that there are enough embarrassed smokers that it'll be worthwhile. (It's very likely to be effective enough to make money spent on these services efficient - it probably improves health over time and reduces NHS spending so long as most people try. The figures are based on the assumption that 70% of thsoe successfully quitting during pregnancy will start again within a year.)
Personally, I don't much care about the affronted shrill minority who seem to be the majority on this thread - annoyance should be largely forgotten quite quickly by any sensible person! (It's not a relevant concern for me, but as an anti-smoker I'd be annoyed to be tested. I'd remember it, but it wouldn't actually bother me on a day-to-day basis and it wouldn't be a major factor in judging whether or not the NHS did a good job or not. A dead baby would be.)
The whole irony in this is that NICE has been (little c) conservative in what it's said and how paternalistic it would choose to be. The evidence (which is mostly from the US) suggests the most effective thing to do is probably to pay pregnant smokers to give up. The public health experts were cautious not to emphasise this type of treatment and probably for a good reason. THAT'S paternalism to possibly get upset about!