Finally, proof that aspartame will destroy your kidneys?

Hot on the heels of claims about a study that supposedly correlated aspartame with lymphoma, but that turned out to be "weak", a 2011 study linking diet sodas to “increased kidney decline” has been making the rounds of the anti-aspartame sites. I am not sure why it took until early 2013 for it to start popping up on the various blog sites.The way that the conspiracy sites tell it, the study was damning and demonstrated that aspartame will pretty much kill your kidneys.

Some headlines:

• “Diet Soda, Aspartame Shown to Destroy Kidney Function”

• “Aspartame alert: Diet soda destroys kidney function”

• “Aspartame and Kidney Damage”

What does the study say?

But what does the study actually say and how does it fit into the larger body of evidence?

As of this writing, the full text of the study("Associations of sugar and artificially sweetened soda with albuminuria and kidney function decline in women" by Julie Lin, published in Clinical Journal of the American Society of Nephrology) is available for free, so anybody can read it without a subscription.

The results:

Consumption of ?2 servings per day of artificially sweetened (diet) soda was independently associated with eGFR decline ?30% (OR 2.02, 95% CI 1.36 to 3.01) and ?3 ml/min per 1.73 m2 per year (OR 2.20, 95% CI 1.36 to 3.55). No increased risk for eGFR decline was observed for <2 servings per day of diet soda. No associations were noted between diet soda and MA or sugar soda and MA or eGFR decline.

With the simplified conclusion of:

Consumption of ?2 servings per day of artificially sweetened soda is associated with a 2-fold increased odds for kidney function decline in women.

Basically, the study looked at the results of two tests that are associated with a decline in kidney function. One of them, Estimated Glomerular Filtration Rate or eGFR, had apparently a larger decline for diet than for sugar sodas. The other measure, microalbuminuria or MA, did not have a statistically significant effect for either sugar or diet drinks.

Looking deeper into the study content itself, their Table 3 presents the OR values for eGFR decrease (using <1 day to represent OR=1.0) but they do not present a direct comparison between sugar-sweetened and diet drinks for each level of consumption. They show the 1 to 1.9/day (OR 0.9) and >=2/day (OR 2.02) for diet, but only show >=1/day (OR 1.56) for sugar-sweetened. The authors note:”...only 3% of the women in our study consumed ?1 sugar soda per day”, versus 22% for >=1 for diet soda, which would obviously generate a much smaller sample set. But it seems like it should have been possible to calculate the >=1 per day for diet soda to at least get a better comparison (and perhaps just show it in the footnotes of the table). Forgetting about any sort of comparison (i.e. pretending that we have essentially two separate studies), the granularity makes sense because it was only after 2/day of diet that the potential problem manifests. But, if the two were nearly identical at >=1,it would imply an issue with soft drinks in general, not specific to diet drinks.

But in any case, there is still the 2-fold (100%) increase in risk from >=2 per day which should be looked at. Also note that from 1 to 1.9 per day actually has an OR < 1 (the authors note “No association between lower levels of artificially sweetened soda intake and eGFR decline was seen, implying a threshold effect rather than one that increases linearly”). I doubt that actually means diet beverages have a “protective” effect. But what must also be noted is the 56% increase for >=1 of sugar sweetened. The study is touted as showing diet beverages to “destroy” kidneys, but nobody talks about the increased risk from sugar-sweetened soft drinks. And if I am at all right and combining the diet data to produce a >=1 figure, it would very possibly come very close to the sugar number.

A little statistical support

As I am certainly no statistical expert, I actually consulted a friend who provides statistical support for research at one of the UC schools, and the relevant parts of his response were (emphasis mine):

It is odd that they would evaluate and report the non-equivalent intakes...my guess is that they are justifying this non-equivalency by trying to show "proof of concept" that artificially sweetened soda is, at higher intakes, associated with increased risk.

But you are right that the odds ratio for artificially sweetened drink does not appear to be statistically different than the odds ratio for sugar soda. You can tell by the confidence intervals (CI) they report. The 95% CI is 1.36 to 3.01 for artificially sweetened, meaning that we can be 95% certain that the true OR in the population is somewhere between 1.36 and 3.01 (this range is due to the fact that we are dealing with a sample - the range would be smaller if the sample was larger and/or variance in the outcome was smaller)... the OR for sugar soda, by contrast, is 1.56 (with CI = .84 to 2.91), which is a value that is captured within the CI for artificially sweetened soda and thus not statistically different from it... So what they are saying is that, on its own, drinking more than 2 artificially sweetened sodas a day is associated with increased risk...But they cannot say that this risk is statistically greater than the risk of drinking 1 sugar soda a day...What does the OR look like for 2 or more sugar sodas a day? Seems like that would be important to know....

So this seems to mostly confirm my hunch that perhaps there should be more highlighting the result that consumption of carbonated soft drinks (sugar-sweetened or no) is correlated with a decline in kidney function, in women at least. The authors themselves actually point out that the diet link was not hypothesised so the study was not actually designed to test for that effect. Further studies may nullify it (or strengthen it). This study is in no sense a smoking gun that diet drinks are going to destroy your kidneys.

I want to quickly note that there were two previous studies (both mentioned by the current Lin study). A 2004 study based on the National Health and Nutrition Examination Survey found no association between diet sodas and kidney damage but some association with sugary sodas. A separate 2009 study based on the Multi-Ethnic Study of Atherosclerosis (MESA) also found no association of sugar-sweetened sodas for people with pre-existing chronic kidney disease (CKD).

In all cases the authors mention that it could be other factors in the lifestyle or the beverage itself which relate to the kidney decline, because the studies were not designed to actually figure out what is causing the problem. One crucial thing is that aspartame is never identified as the crucial factor in any of the studies.

Conclusion

So it seems that the analysis here should be tempered -- if you consume a lot of sodas (sugary or not), you might be at increased risk for kidney damage versus not consuming soda. But this is not how the anti-aspartame sites present the data, and that is the problem. If anything, this is a great case for moderation, both in consumption and analysis. As is frequently the case with these sort of studies, it seems to warrant further studies specifically intended to isolate the reason for the decline, such as an ingredient common to both sugar-sweetened and artificially-sweetened.

REFERENCES
Lin, Julie, and Gary C. Curhan. "Associations of sugar and artificially sweetened soda with albuminuria and kidney function decline in women."Clinical Journal of the American Society of Nephrology6, no. 1 (2011): 160-166.

Shoham, DA, et al. Sugary soda consumption and albuminuria: Results from the National Health and Nutrition Examination Survey, 1999-2004. PloS ONE 3(10): e3431. doi:101371/journal.pone.0003431.

Bomback, AS, et al. Sugar-sweetened beverage consumption and the progression of chronic kidney disease in the Multi-Ethnic Study of Atherosclerosis (MESA). American Journal of Clinical Nutrition, 2009, 90:1172-8.

UPDATE: Clarified the intro.