IS THIS FOR REAL? CRAZY!

[dweezil1968]

JOURNAL OF REFRACTIVE SURGERY 2008; 24(7):737

Cigarette Smoking is Negatively Associated With Keratoconus
By Eberhard Spoerl, PhD; Frederik Raiskup-Wolf, MD; Eberhard Kuhlisch; Lutz E Pillunat, MD
PURPOSECigarette Smoking is Negatively Associated With Keratoconus
By Eberhard Spoerl, PhD; Frederik Raiskup-Wolf, MD; Eberhard Kuhlisch; Lutz E Pillunat, MD
PURPOSE

To investigate a correlation between cigarette smoking and keratoconus.
METHODS

Patients with keratoconus who were treated with corneal collagen cross-linking from June 2006 to November 2007 were asked about their smoking habits. A person smoking a minimum of two cigarettes per day for more than 1 year was classified as a smoker.
RESULTS

A total of 180 patients with keratoconus (mean age 28�9 years [range: 15 to 41 years]) were asked about their smoking habits. One hundred seventy-one (95%) were non-smokers and only 9 (5%) were smokers (95% confidence interval, 2.31 to 9.28). Using the chi-square test, a significant correlation was found between non-smokers and keratoconus (P<.001).
CONCLUSIONS

In this group of patients with keratoconus, few were smokers. Cigarette smoke contains toxic substances. Consequently, people are advised not to smoke. However, we speculate that the by-products of cigarette smoke may lead to cross-linking of collagen, which in the cornea, may prevent the development and progression of keratoconus. [J Refract Surg. 2008;24:S737-S740.]

[Andrew MacLean]

Small sample and P of less that 0.001?



I don't think I'll start smoking

Andrew

[GarethB]

In adition, with any paper you need more than the abstract otherwise you only have a small part of the story.

[rosemary johnson]

Hmmm, I can think of several problems with that, as outlined in the abstract:
1. as already mentioned, it is a very small sample size.
2. it is a sample biased towards those with only mild KC - because they were having CXL. It does not attempt to survey those with KC that is more severe already, so they are no longer suitable fr CXL
3. it is biased (on the same grounds) towards a youngish sample of the population, and possible also (if they have known for a while about having an eye condition) more health-conscious sample of the population than average.
4. KC is also associated with other health problems, in that people with KC are statistically more likely to have allergies, asthma, eczema, hay fever, etc, which might also have an impact on people's likelihood of smoking. I also remember reading that people with KC were on average more intelligent than the average population - what bearing this may have on smoking, I couldn't begin to speculate!
Now, maybe if one reads the full paper, they have got and cite general demographic information on the rates of smoking inthe general population, and maybe they even have got demographic smoking info on a typically youngish population with a higher than aaverage incidence of other long-term health conditions in general and of asthma/allergies/hayfever in particular.
Meanwhile, it might be very interesting - at the next KC group conference or AGM - to conduct a stra poll of members: ask for a show of hands (or secret ballot slips of paper, if people might not like to admit to it in public!) as to whether those present: 1. smoke cigarettes now; 2. used to smoke cigarettes but have given it up; 3. have never smoked cigarettes.
That would be a skewed result too, since the group regulars tend to be those with more severe cases of KC, but I wouldn't be surprised if the results were a bit different.
The last line of the abstract, about the possibiblity of tobacco smoke triggering crosslinking, sounds more interesting that the statistics. Anyone know any more about that?
Rosemary
who has never smoked cigarettes, but does occasionally ejoy a nice cigar after dinner. Or did, before the asthma got worse.

[Lia Williams]

And smoking and contact lenses are not compatible.

I avoid smoky situations to stop smoking getting in my eyes. Which I guess others to do. So one could say that the wearing of contact lenses is a way of giving up smoking!

Lia

[GarethB]

Those who attend AGM's and the socials will know which comitte member always makes a swift exit for a smoke

[Alison Fisher]

Oh boy.

Before I left home both my parents were heavy smokers. Can't say it did much when my KC was in its early stages other than irritate the heck out of me.

[DJ Smak]

weird because I think the opposite, that my smoking actually makes it worse.

[Matthew_]

Small sample and P of less that 0.001?

Andrew, I'm a bit rusty but doesn't that mean simply that the results are statistically significant (P<0.5). Ie that non-smoking and KC are correlated.
I think that may be the only the thing the study has going for it When you consider that smokers are in a minority and KC sufferers are also a minority, could it not simply be the case that coincidentally there are similiarly small proportions of smokers and KC sufferers in the (very small) sample. You cannot assume that correlation means cause and effect. The hypothesis that the two are related should be rejected and it hasn't been (Type II error?).

Even it were true; KC versus Lung Cancer? I'll stick with what I have got, thanks!

[rosemary johnson]

Well, I'm a bit rusty too......
but as I understand the report, they have corrected for smokers being a minority, but are finding that people with KC are even less likely to be smokers than people in the population as a whoel are.
This may well be true - but they don't seem to have allowed for the probability that having KC may itself make one less likely to take up smoking than an "average" person - or more motivated to give up if you're diagnosed after starting.
Factors like smoke irritating contact lenses, being of necessity more health-conscsiou and spending more time than average in hospitals/clinics may all make one less likely to be a smoker.
In other words, it may not be non-smoking that causes KC, but KC that causes non-smoking. They don't seem to have considered that.
Or at least, it isn't in the synopsis.
Rosemary