Letter by Chiu Regarding Article, “Sugar- and Artificially Sweetened Beverages and the Risks of Incident Stroke and Dementia. A Prospective Cohort Study”
To the Editor:
In their recent study published in Stroke, Pase et al1 concluded that artificially sweetened beverage (ASB) consumption was associated with a higher risk of stroke and dementia. Additionally, the authors also asserted that sugar-sweetened beverages (SSBs) were not associated with stroke or dementia.
The findings in their study do not necessarily support their conclusions.
The authors stratified Framingham Heart Study cohort subjects based on their history of consumption of either ASB or SSB. The authors noted hazard ratios of 2.96 for ischemic stroke, and 2.89 for Alzheimer disease for subjects who consumed ASB, compared with subjects who did not consume ASB. There was no increased hazard in the incidences of ischemic stroke or Alzheimer disease for subjects with SSB consumption compared with those with no SSB consumption.
It is important to note that the cohort of combined strata of ASB subjects and the cohort of combined strata of SSB subjects each comprised the entire Framingham Heart Study cohort. Therefore, the number of incident events of stroke and dementia in combined consumption and nonconsumption ASB subjects would be identical to the number of events found in combined consumption and nonconsumption SSB subjects.
Demonstrating risk relationships using hazard ratios can have its pitfalls. No doubt the authors have demonstrated that subjects who consumed ASB had close to 3× the risk of developing stroke and dementia, compared with those who did not consume ASB. The lack of an increased hazard of incident stroke and dementia in SSB consumption subjects means that the rate of incident stroke and dementia in SSB consumption subjects were no higher than in SSB nonconsumption subjects. Given these findings, the presence of an increased hazard ratio in ASB strata comparisons may have resulted from an unusually low incidence in ASB nonconsumption subjects (reference subjects). Alternatively, the lack of a significant effect of SSB consumption may have resulted from an unusually high incidence of stroke and dementia in the SSB nonconsumption groups (reference subjects).
The only findings that would have strongly supported the authors’ conclusions would have to include 3 conditions: (1) a similarly low rate of incident stroke and dementia in both ASB nonconsumption and SSB nonconsumption subjects, (2) a persistently low rate of incident stroke and dementia in SSB consumption subjects, and (3) a higher rate of incident stroke and dementia in ASB consumption subjects. Clearly, these conditions had not been met with the results of this study.
It is interesting to note that when the authors conducted mediation analysis on the effect of diabetes mellitus on the association of ASB consumption and incident all-cause dementia and Alzheimer disease, and on the effect of hypertension on the association of ASB consumption and incident all-stroke, the association effects were attenuated. In fact, as Wersching et al2 correctly noted in the accompanying editorial in the same issue, such association effects were no longer (statistically) significant. Had such results been included in the main results of the study, the authors might not have concluded that ASB increases the risk of stroke and dementia.
Finally, one is reminded that the study did not compare groups of subjects who exclusively consumed ASB or SSB. The extent of the overlap of ASB and SSB consumption, if any, is unclear from the data. This may potentially confound any analysis of the association of type of beverage consumption with the incidence of stroke and dementia.
Overall, Pase et al presented intriguing results with their study. I would only argue, however, that their conclusion of ASBs being associated with a higher risk of stroke and dementia, and of SSBs not being associated with stroke and dementia, had not been supported by the results of the study.
Edward K. Chiu, MD, PhD
Division of Cardiology
Department of Medicine
Stroke welcomes Letters to the Editor and will publish them, if suitable, as space permits. Letters must reference a Stroke published-ahead-of-print article or an article printed within the past 4 weeks. The maximum length is 750 words including no more than 5 references and 3 authors. Please submit letters typed double-spaced. Letters may be shortened or edited.
- © 2017 American Heart Association, Inc.
- Pase MP,
- Himali JJ,
- Beiser AS,
- Aparicio HJ,
- Satizabal CL,
- Vasan RS,
- et al
- Wersching H,
- Gardener H,
- Sacco RL