RATIONALE: Using nicotine replacement therapy (NRT) whilst smoking, prior to quitting, is called preloading. Two reviews have estimated the effect of preloading on abstinence, but need updating. Neither investigated possible mediators or moderators of the effect, which could have implications for individual treatment plans. OBJECTIVES: To update the nicotine preloading efficacy estimate and test four hypotheses: (1) Efficacy is mediated through reduced smoking reward, (2) efficacy is mediated through increased NRT adherence post-quit, (3) efficacy is mediated through increased confidence, and (4) behavioural support modifies efficacy. METHODS: Randomised controlled trials were included that allocated cigarette smokers attempting to quit to either a preloading or control condition. A Mantel–Haenszel fixed-effect model was used to calculate risk ratios from quit rates at short- and long-term follow-ups. We carried out sub-group analyses and synthesised the data available on possible mediators and moderators qualitatively. RESULTS: Eight relevant studies were included, with 2,813 participants. The risk ratio (RR) for short-term abstinence was 1.05, 95% confidence intervals (CI) = 0.92, 1.19, and for long-term abstinence 1.16, 95% CI = 0.97, 1.38. There was a marginal benefit of using nicotine patch rather than gum for preloading, significant at short-term follow-up, and no significant benefit of more intensive pre-quit behavioural support. CONCLUSIONS: We found a weak non-significant effect of nicotine preloading on abstinence. None of our mediational hypotheses received strong support, however evidence suggests that efficacy was enhanced by the patch over acute NRT. Future research needs to investigate the mechanisms of preloading by carrying out mediational analysis.