Adding nicotine patches to varenicline had no beneficial or detrimental effect on urges to smoke, withdrawal discomfort, abstinence rates, or adverse effects profile.
Several issues need to be considered when interpreting these results.
The study had only a short-term follow-up. This however, would limit the generalisability of the results primarily if the results were positive. It is unlikely that a lack of effect during the acute withdrawal period when stop-smoking medications exert their main impact could change into a significant effect later on.
The sample size was sufficient to detect small differences in withdrawal ratings and craving and also a clinically meaningful difference in short-term abstinence rates. We cannot rule out a possibility of a subtle effect on abstinence rates detectable on a large sample. Such an effect seems unlikely though, because it would presumably be mediated by lowering of withdrawal discomfort and craving, and these parameters were not affected.
The negative results cannot be attributed to low compliance with medication use, as almost all participants used both medications during the first week after TQD when any beneficial effects would be expected to be the strongest. We used 16 h/15 mg patch which has extensive evidence of its efficacy . The 24 hour patch has been shown to have stronger effects on morning urges to smoke , but both types of patches have the same effect on smoking cessation outcomes . It could be argued that other short-acting NRT formulations, which can be used opportunistically could be more effective. This is possible, but any gains in potentially higher efficacy of short-acting NRT products are usually undermined by the fact that oral NRT products and nicotine nasal spray have more side effects and are less user-friendly then patches and generate lower adherence. Good sustained adherence to oral NRT products and nasal spray usually requires a period of supervised frequent and regular use . Where the expectation is that the product will be used only occasionally, as a supplement to another treatment, the adherence is likely to be low. Nevertheless, our results should be generalised to alternative NRT products with caution. There also remains a possibility of type II error, i.e. that there is a difference but the trial did not detect it.
Adding nicotine patches to varenicline did not increase the incidence of nausea or of any other adverse events. There was a trend for the active patch group to report more abnormal dreams, but the results overall raised no concerns regarding the safety of combining the two treatments.
One possible interpretation of the lack of synergy between the two medications is that NRT and varenicline achieve their effects via similar target mechanisms, which overlap to a large or even full extent. Varenicline may act on a more limited range of nicotinic receptors than nicotine itself, but these seem to include those involved crucially in the rewarding effects of smoking. By blocking such receptors, varenicline may be limiting any potentially beneficial effects of NRT as well. E.g. nicotine patches normally alleviate weight gain in continuous abstainers  but they had no such effect here.
Nicotine patches did not improve outcomes significantly in the subgroup of smokers who did not have a strong response to varenicline early on. This finding is more tentative than the main result because there was a trend in favour of nicotine patches and the study was not powered for sub sample analyses. Whether or not any patch effects are blocked by varenicline more in varenicline responders than in non-responders, an interesting question arises as to whether varenicline non-responders may benefit from NRT in the absence of varenicline. If this were the case, treatment efficacy could be improved by switching smokers who show no reaction to varenicline during the pre-quit period over to NRT. Future studies should evaluate this notion further.
The trial results have a practical implication. Finding a positive effect would indicate that a large-scale study with a long-term follow-up is warranted, but it would not provide a definitive proof of efficacy. A negative result on the other hand provides an indication that such a trial is unlikely to yield strong positive results. There is a widespread interest in combining NRT and varenicline in the hope of improving treatment outcomes. The results of this study suggest that such practice may not be productive or economical, although further trials would be useful to exclude the possibility of type II error.