Cancer risks of hookah (shisha, narghile) tobacco use require further independent sound studies


  • Conflict of interest: The authors have no financial or personal conflict of interest to declare.

Cancer risks of hookah (shisha, narghile) tobacco use require further independent sound studies

Dear Sir,

In the Letter on the risk for pancreatic cancer associated with the use of water pipes [type of device and use (tobacco, molasses-tobacco, etc.) however not specified], Fouad et al. regret that there would be a dearth of published studies on the health effects of this form of smoking.1 This is not exact as, for instance, a recent work from Turkey has recalled important pioneering works on hookah smoking carried out in the 1980s and 1970s by independent researchers.2 Contrary to a common misperception (influenced by the popular press too often interested in “fresh” scientific publications only), biomedical research on the multi century-old hookah has not begun at the dawn of the 21st century (year 2002).3 Dozens of valuable and relevant studies have remained for too long under the dust of shelves in medical libraries. Although the question here is not that of the quantity but rather that of the quality of the work, it is noteworthy that an impressive number of articles have been published on this issue since highly funded research centres dedicated to the fight of this form of smoking were set up 7 years ago. A neologism (“waterpipe” in one word) has even been created for this purpose.4

Three references are cited by Fouad et al. to support the important claim that “the longer duration of a Water Pipe Smoking session leads to a much higher yield of tar, nicotine, carbon monoxide, polycyclic aromatic hydrocarbons, and heavy metals than cigarette smoking. The water pipe smoker may therefore inhale the equivalent of 100 or more cigarettes”.1 However, one of these references (Chaouachi, Int J Environ Res Public Health 2009) is actually a critique of this statement and shows that the levels and number of toxic substances in cigarette and hookah smoke are completely different from each other. What is unfortunately not discussed is the fact that the elevated yields of toxic chemicals, as reflected in the other cited paper (Shihadeh and Saleh, Food Chem Toxicol 2005), were generated through the use of an unrealistic narghile smoking machine drawing one puff every 17 sec for a full hour with the charcoal in the same position. Such a set-up is supposed to mimic the “average” narghile smoker. Yet, while “averaging” a complex human behaviour (i.e. reducing its wide variations to periodic puffs drawn every 17 sec) over a short span of time (5 min, for instance) would be mathematically and epistemologically acceptable, extending this process to 60 min, as if a narghile smoker were a robot, represents a gross methodological error.4, 5 Furthermore, it should be noted that not only all these chemicals are not mutagenic but also no details are given about their biokinetics, i.e. how they do reach and enter the pancreas.

Consequently, rapid equivalences implying an identical chemical composition of two different smokes are scientifically unacceptable. This applies not only to mainstream smoke (the one inhaled by the smoker) but also to side-stream smoke (strikingly visible with cigarettes although almost inexistent with hookahs).5 In a recent article based on the above-mentioned controversial smoking machine, a tunnel was used to collect side-stream carcinogenic emissions that were measured only 1 min after the quick-lighting charcoal was lit and put atop the bowl of the pipe.6 Among other biases (ageing and dilution processes, ventilation holes of the cigarette not blocked), the tunnel was disproportionate in size given the obvious spatial differences between a thin 10 cm long cigarette and a portly 50 or 60 cm high hookah. Then, in the real world, users always wait up for several minutes before setting about smoking. During this time, highly toxic particles, appearing as a visible black plume of smoke at first, are emitted from the (non-natural) charcoal.4, 6

From Roffo's pioneering interest in narghile smoking 70 years back to that of El-Aasar and his colleagues 3 decades ago, and leaving aside other independent researchers, particularly in Asia and Africa (Wynder, Hoffmann, Sanghvi, and others),4, 5, 7–9 we know that the smoke from cigarette and water pipes are qualitatively and quantitatively very different from each other. This is particularly true when the given water pipe works with moassel, the fashionable tobacco-molasses mixture, so that equations such as those disseminated by the media from the work of “waterpipe” experts (e.g. 1 hookah session = 10, 20, 50, 100 up to 900 cigarettes) make absolutely no sense. Indeed, smoke from a shisha using flavoured moassel is mainly made up of water and glycerol (no biological activity) and is far less concentrated in chemicals (hundreds vs. thousands) than cigarette smoke.4 This situation bears some relation with the working temperature at stake which, in both cases, is different by hundreds of degrees Celsius. Interestingly, tar temperature is known to be a determinant of tobacco smoke carcinogenicity. Furthermore, the biological findings of the first aetiological study on Carcino-Embryonic Antigen levels in hookah exclusive smokers (i.e. smokers in remote villages of the Punjabi countryside who have been smoking only hookah), are in agreement with this chemical fact.10, 11 They are also in agreement with numerous historic studies and recent measurements of biomarkers in body fluids of shisha smokers in Germany.12 At an international anti-tobacco conference this year, a not yet published study on the levels of many toxic and carcinogenic substances (namely cadmium, lead, chromate, arsenic, metabolites of benzene, polycyclic aromatic hydrocarbons, acrolein, and acrylamide) was announced. Unsurprisingly, the levels were not higher than among cigarette smokers and, most of the time, not very different from controls. Even the level of NNAL (4-(methylnitrosamino)-1-(3-pyridyl)-1-butanol), metabolite of the potent carcinogen NNK (4-(methylnitrosamino)-1-(3-pyridyl)-1-butanone), was not different from that of controls. The researcher, who presented these negative results to the public surprisingly stated that they were “disappoint[ing] from a scientific point of view” and, for this reason, refrained from detailing all of them.12 Unfortunately, neither the Pakistani and German studies nor any of two dozens of historic publications pointing to the other direction, i.e. the absence of a link, or a weak association, between cancer and hookah smoking, were cited by Fouad et al.1, 4

This does not mean that hookah smoking is safe. Certainly the bulk of historic studies show that, globally taken, hookah use (particularly under its traditional aspects) is safer than other forms of smoking. However, if today's researchers, in contrast with their precursors of the past decades, do suspect any unnoticed hazards, and we believe they may exist, then the most practical and expected response should be to study them immediately. For instance, few scientists interested in this issue have noted that shisha smoking has spread to the whole world, not only because of its flavours but, above all, because of a recent technical innovation which is the use of quick-lighting charcoal instead of the one traditionally used, i.e. natural charcoal. Indeed, in many villages of Pakistan, dried animal faeces, called “gobar”, are also used for this purpose. For centuries, resorting to these natural products has implied a great “hassle” as when people prepare a barbecue outdoors. Now, from a public health perspective, the problem is that, despite repeated recommendations,5, 10 no study to date, since a pioneering one in 1993,13 has analysed the chemical composition of diverse types of charcoal widely present in the world market, against that of natural charcoal.

Instead of limiting research to a too narrow segment of the literature, perhaps the wisest and first thing to do before engaging in a praiseworthy international research collaboration for research on the hookah cancer hypothesis, would be to take stock, with no exception, of all relevant references of the world literature. In this respect, Fouad et al. can be referred to a critical comprehensive review of hookah health effects recently published by an independent research team in Africa. Because of the great importance of this task, the study was split in two parts.14, 15 Tobacco smoking is dangerous and leads to many diseases, particularly various types of cancer. Even if the evidence of an association with cancer has been historically rather weak for hookah smoking, the cancer hypothesis can be studied provided that the above condition is fulfilled. Debate is the main condition of scientific progress for the direct benefit of public health in the present case.

Finally, one of the references cited by Fouad et al. (Maziak et al., Tob Control 2004) contains several statements implying a direct association between water pipe (type and products not specified) smoking and lung and bladder cancer. These statements are in fact the exact opposite of those in the (cited) sources: Rakower and Fatal (Br J Cancer 1962) and Bedwani et al. (Int J Cancer 1997), respectively. Notably, Prignot et al. (Int J Tuberc Lung Dis 2008), also cited by Fouad et al., have pointed out that “these data indicate an increased risk […] but not bladder cancer (Egypt)”. Despite this clarification by antismoking researchers themselves, it is amazing that this serious error can be found again in supposedly peer-reviewed publications.16

Yours sincerely,

Kamal Chaouachi, Khan M. Sajid.