Intended for healthcare professionals

CCBYNC Open access

Sugary drink consumption and risk of cancer: results from NutriNet-Santé prospective cohort

BMJ 2019; 366 doi: (Published 10 July 2019) Cite this as: BMJ 2019;366:l2408

Related BMJ Opinion

“Sin taxes”—the language is wrong, but the evidence is clear

What's in a name? That which we call sugar by any other name would taste as sweet

Dear Editors

I note with interest the conclusions of the cohort study linking sugary drinks with the risk of overall cancer and breast cancer.

I notice, however, the interesting handling of what are considered to be sugar sweetened beverages in which the researchers do not distinguish the type of sugar added to sweeten the drinks; the researcher's description of "median sugar level" in units of g/100mL make it apparent to me that there is an assumption of what type of sugar was added without confirmation of its actual origin, for example, cane sugar or sugar beet.

The manner of treatment of this information is of special interest particularly when the NutriNet-Santé cohort study from 2009 is near the height of the debate of the role of fructose, particularly in High Fructose Corn Syrup (HFCS) in the causation of obesity and related metabolic conditions.

Of interest, it is acknowledged that HFCS is used to sweeten many soft drinks in the US whereas it appears that cane sugar is the predominant sweetener used in the manufacture of beverages in Europe. There was (and still are) concerns about the role of fructose in obesity as studies indicate that fructose may be a carbohydrate with greater obesogenic potential than other sugars (ref 1), possibly via excessive fructose metabolism by the liver. It is possible that the non-discriminatory use of the HFCS label on all fructose-containing corn syrup (which is originally mostly glucose) confuses many researchers and clinicians since many US food manufacturers use HFCS 42 (42% fructose, 58% glucose) for cereals/processed food and HFCS 55 for soft drinks; these fructose to glucose ratios are not dissimilar to regular cane sugar. However it is well known that HFCS 90 sweeteners are used in soft drinks from other countries like Japan; it is possibly used in the US as a "speciality application" which may include canned fruits, confectionery and dessert syrups.

Also of interest, typical sugar composition in apple juice is 60% fructose, 27% glucose, and 13% sucrose. This disproportionally high representation of fructose compared to glucose is not seen in other common types of fruit juices (for example, orange or grape). Again there is no obvious attempt to differentiate the type of fruit juices (rather than "median sugar level" as sugar content) in the determination of dietary sugar in this cohort study.

Thus it is particularly frustrating to find that no attempt has been made to differentiate the consumption of sugar type in this study, particularly when the participants in the NutriNet-Santé study were invited to undergo biological sampling and clinical examination, including bioimpedance measurements, to determine visceral fat index. It is well known that excessively disproportionately high fructose consumption is suspected to be a significant driver in lipogenic hepatic processes which contribute to body adipose tissue production.


Competing interests: No competing interests

11 July 2019
Shyan Goh
Orthopaedic Surgeon
Sydney, Australia