Smokers who stop smoking shortly before surgery (recent quitters) seem to have worse surgical outcomes than early quitters. Therefore, concerns have been expressed about stopping smoking within 8 weeks of surgery. This has generated considerable uncertainty in the media and even in hospitals where smoking cessation advice is an important priority.
Researchers therefore conducted a 9 trial meta-analysis in which they say the concerns are unfounded and that all patients be encouraged to stop smoking. The results from some of the previous studies may simply indicate that recent quitting is less beneficial than early quitting, and not that it is risky.
Looking at the data however, we can see why there may be some confusion. The risk of an overall postoperative complication is unclear (relative risk, 0.78; 95% confidence interval, 0.57-1.07). What this says is that the researchers are 95% certain that the average risk for an adverse outcome is somewhere between 43% lower (0.57) and 7% higher (1.07). This is the confidence interval. It tells us where the average risk may lie. The average, or relative risk, is in the middle of the confidence interval at 0.78 (a 22% reduction). I’m getting the reduction percentages by subtracting from 1.00. For example, 1.00 – 0.78 = a 22% reduction.
It is however, even worse for those studies that focused on pulmonary complications (RR, 1.18; 95% CI, 0.95-1.46). This suggests an average risk increase of 18% (Relative Risk ,1.18), but the confidence interval is large (and spans 1.00) so we can’t say. The 95% interval ranges from 5% reduction (0.95) up to to a 46% increase in risk (1.46).
The researchers conclusions were “the concern that stopping smoking only a few weeks prior to surgery might worsen clinical outcomes is unfounded. There is no evidence to suggest that health professionals should not be advising smokers to quit at any time prior to surgery.”
Which sounds very positive and almost conclusive. What they are actually saying is that nobody knows, but the evidence that it worsens outcomes is not there.
I’ve been a little bit sneaky here. I’ve used this post more to point out how researchers reports findings and make their conclusions rather than to highlight the stopping smoking dilema. My concern with this and many similar papers is that they should just come out and say ‘we’ve still got no idea.’ Because at the end of the day they haven’t, and the pulmonary findings actually look more likely to support the concern than not. We just don’t know.
This is a clear case of absence of evidence, not a finding one way or another. The impression you may get from reading the abstract however, is that we’ve sorted this one out, continue as you were. Larger research trials will get sorted eventually.
When you see the phrase “There is no evidence…” you are being told that they don’t know. I think an appreciation of the findings from that perspective enables us to make better decisions. My final point is that smoking is a massive global health problem and a top priority for all your patients and equivocating around surgical complications must not be used as their excuse to continue.
