The thirdhand smoke scam

Now that the thirdhand smoke story has been reported around the globe, it's time to look at the study which led to headlines such as:

Third-hand smoke causes cancer, study shows risks to babies and toddlers

This is not your average piece of epidemiological number-crunching. It involved some real lab work which, when written down, is largely unintelligible to the layman*. Journalists rarely bother to read scientific studies at the best of times, but what chance do they have with paragraphs like this?

Laboratory experiments using cellulose as a model indoor material yielded a > 10-fold increase of surface-bound TSNAs when sorbed secondhand smoke was exposed to 60 ppbv HONO for 3 hours. In both cases we identified 1-(N-methyl-N-nitrosamino)-1-(3-pyridinyl)-4-butanal, a TSNA absent in freshly emitted tobacco smoke, as the major product. The potent carcinogens 4-(methylnitrosamino)-1-(3-pyridinyl)-1-butanone and N-nitroso nornicotine were also detected. Time-course measurements revealed fast TSNA formation, with up to 0.4% conversion of nicotine. 

And that's from the abstract - the bit that summarises the study for the casual reader. So what does it actually say?

To put it in something close to simple terms, the experiment involved putting nitrous acid (HONO) in contact with nicotine. The nicotine had been absorbed into surfaces (hence 'thirdhand smoke'). In the real-life experiment, this surface was the glove compartment of a truck driven by a heavy smoker (presumably the cabin of a truck was chosen because it is the smallest space a smoker might work in). In the other experiments, they used cellulose substrates. 

The aim was to see if the reaction created tobacco-specific nitrosamines (TSNAs), specifically NNK, NNA and NNN, some of which are believed to be carcinogenic.

The scientists found no trace of NNN in any experiment. In the glove compartment, they found levels of NNK that were barely above the detectable level (less than 1 ngcm-2). Even in extreme experimental situations, in which cellulose substrates were exposed to pure nicotine vapour, NNK levels failed to reach 5 ngcm-2.

They found somewhat larger measurements of NNA (20 ngcm-2 in extreme experimental conditions) but levels were much lower in the real-life conditions of the truck (1 ngcm-2). This was all rather academic anyway because, as the authors admit:

"NNA carcinogenicity has not been reported." 

In other words, the one TSNA they did manage to find in any quantity doesn't cause cancer. 

There is nothing obviously wrong with the way the chemistry was done here. The paper simply shows that nitrous acid (HONO) molecules will react with absorbed nicotine (just as it would with free-floating nicotine) to produce TSNAs. The more HONO in the room, and the more nicotine on the surface, the more the reaction will occur (of course).

The problem (and it's a big problem) is that mixing nitrous acid with nicotine is an experiment with virtually no practical application. If your house or car is full of nitrous acid then you have more to worry about than it reacting with absorbed nicotine. As the authors point out near the top of the 2nd column, 1st page:

"The main indoor sources of HONO are direct emissions from unvented combustion appliances, smoking, and surface conversion of NO2 and NO." 

NO2 and NO themselves are products of unregulated combustion. So you'll only be exposed to high concentrations of HONO if you're exposed to the products of combustion - ie. you're a peasant in a smoke-filled hut, you live in a very polluted city like New Delhi, or you are in fact smoking a cigarette. The combustion products themselves are carcinogens, and are present in much higher concentrations than the TSNAs. Any surface reaction is negligible. Your problem is the nitrous acid, not the TSNAs.

Is this kind of surface reaction likely to take place in the home? Not at all. Nitrous acid concentrations in the average Californian home are 4.6 parts per billion (ppb). This is 14 times lower than the 65 ppb concentrations used in this experiment (which indirectly compares with EPA limits for NO2 of 53 ppb). The chances of HONO and nicotine reacting to create detectable, let alone harmful, concentrations of TSNAs outside of a laboratory experiment are zilch.

In summary:

• The researchers used concentrations of nitrous acid 14 times higher than would be found in a normal environment

• Even at the unrealistic levels found in the experiment, there is no evidence that such doses are harmful to humans

• The main TSNA produced is not a carcinogen

• The weakest results were found in the real-life conditions, with measurements barely exceeding detectable levels in the smallest conceivable workplace of a heavy smoker

• Any effect from the TSNAs is negligible compared to the effects of the nitrous acid itself 
  

* And I thank my bio-chemist friend JPM for his assistance in making it intelligible to me.