Intended for healthcare professionals

Editorials

Smoking and psoriasis

BMJ 1994; 308 doi: https://doi.org/10.1136/bmj.308.6926.428 (Published 12 February 1994) Cite this as: BMJ 1994;308:428
  1. H C Williams

    O'Doherty and MacIntyre first drew attention to the strong link between smoking habits and palmoplantar pustular psoriasis in 1985.1 Other studies have suggested that smoking might be an important risk factor for chronic plaque psoriasis as well.*RF 2-5* Most of these studies have recorded information on smoking and psoriasis simultaneously but have also inquired into smoking habits before the onset of the skin disease. There seems little doubt that smoking and psoriasis are associated, but two critical questions remain unanswered: to what extent is the association a causal one, and how important are the implications for the public?

    Though four out of six independent studies have suggested a twofold to threefold increased risk of psoriasis in smokers,*RF 2-7* these might all be misleading because of confounding by alcohol consumption - which is known to be associated with the initiation and severity of psoriasis.*RF 4,6,8-10* Only two of the reports linking smoking and psoriasis included a control for drinking behaviour.2,4 In these studies, the increased risk of psoriasis in smokers did persist after adjustment for alcohol intake, but the problem of residual confounding remains a concern, especially as both alcohol and smoking have been measured only in terms of recall of events in the distant past.11 Further similar case-control studies seem unlikely to resolve this issue unless they are done in societies where alcohol is forbidden and smoking is common.

    Two studies showed a dose-response effect with the number of cigarettes smoked,4,5 but one other did not.2 A finding of functional neutrophil abnormalities in patients with psoriasis who smoked adds some biological plausibility to the link,12 but other studies could be called on to support either arm of the hypothesis.11 Anecdotal reports suggest that stopping smoking has little effect on the course of psoriasis, but factors that affect incidence may not affect chronicity. Intervention studies are needed to examine the effect of smoking on chronicity of the disease and response to treatment.

    If smoking does cause psoriasis the impact on public health could be considerable. A twofold risk of psoriasis in smokers is quoted in most studies. Around 30% of adults smoke, so about a quarter of all cases of psoriasis could have been precipitated by smoking. For palmoplantar pustular psoriasis as much as half of the cases could be related to smoking. Patients with psoriasis seem to have a high incidence of cancers of the lung and larynx, and one explanation could be smoking.13 Smoking might possibly be more important for initiating psoriasis in women (K Poikolainen, personal communication, 1993), whereas alcohol might be more important in men.6 It is also important to consider psoriasis of early and late onset separately, as the two forms may have different causes.

    Any progress in identifying modifiable risk factors for this common and socially disabling disease is to be welcomed - and it is more useful than reports of temporary remission with expensive and potentially toxic drugs. Despite the high heritability of psoriasis, manipulation of environmental risk factors such as smoking, alcohol, exposure to streptococci, trauma, drugs, and emotional stress open up the possibility of substantial prevention of the disease in the future. Already the evidence is good enough for clinicians routinely to ask about smoking and alcohol habits in all patients with psoriasis and especially those with palmoplantar pustular psoriasis. Prospective clinical studies are needed to see whether stopping smoking makes any difference to the course of established psoriasis. Quite possibly people might be more willing to change their habits because of a current and socially debilitating skin disorder than because of long term risks of cancer and ischaemic heart disease.

    References

    View Abstract