Over the last couple weeks I have come across medical news headlines that sounded something like this: “Baking Soda — An Inexpensive Treatment Aid for Autoimmune Diseases Like Arthritis”
I won’t even get into the fact that arthritis is most commonly understood to mean osteoarthritis which is NOT an autoimmune disease, but let’s not digress too much.
The slightly more reasonable news outlets sounded at least a little skeptical: “Drinking baking soda could be an inexpensive, safe way to combat autoimmune disease”
The headlines seemed alluring, and the news articles themselves seemed to at least represent some interesting findings from a recently published study in a well respected journal: The Journal of Immunology.
But medical news headlines in general are often very deceiving and there are many well respected researchers and writers such as Stephan Guyenet PhD, Sarah Ballantyne PhD and Denise Minger who make a living deciphering research studies themselves, presenting accessible and accurate representations of the literature to an interested lay audience. We need people like them!
And we must recognize that medical news outlets, while very unlikely to be malicious in nature, will often NOT represent the real findings of a study or present them in the context in which the authors intended.
Needless to say, I was intrigued, curious and worried when I saw these headlines, so I did the first reasonable thing I could think of: get the paper itself.
Now this isn’t always easy as many academic papers are not open access, and as such, these papers require payment or an affiliation with some academic institution (payee) to obtain. I am lucky enough to have such an affiliation as part of my family medicine residency training, and was able to get the full paper itself.
This is where I would like to stop and make a few points to help you triage and understand the relative reliability of a medical news source making references to “recent study X”.
Red Flags for Misinterpretation of Research
- The study as referenced in the news article is not open access. You may ask why does this matter? If the paper isn’t open access or isn’t readily available then it is more likely to go unread by the author or news team reporting its findings. This is just common sense: if you make something harder to read, then mostly likely it won’t be read thoroughly.
- The title of the paper isn’t mentioned in the news article. Researchers in general do a very good job of trying to explain the primary purpose of a study in the title. If the title is rather technical or says something that may make the news headline confusing then it will likely be left out of the news article.
- The news article contains paraphrased comments taken from the abstract of the paper. Once again, a fairly obvious point related to point #1, but if the news article only has comments based on descriptions/text in the freely available abstract then you should be a little worried that they did not read the paper thoroughly. Abstracts are great and very helpful, but should not be treated in isolation.
I do not profess to be an expert of medical literature. I haven’t been alive on this planet long enough to be an expert in anything, but I want to share my curiosity and express some insights gleaned from reading numerous articles and specifically offer some reflections from the research study in question.
In order to try to facilitate a more accessible discussion, I have organized the remainder of this article into four sections:
- My concerns with the study
- What the study did not find
- What the study did find (for science geeks)
- Final reflections and recommendations
So without further adieu, shall will dig into the heart of this discussion?
CLICK TO EXPAND // The science behind how baking soda reacts in the body
My Concerns with the Study
First things first, what was the actual title of the article referenced in all of the medical news outlets?
Whoa, that’s a mouthful.
Revisiting our red flags, I already mentioned this study is not open access and it has a title that is nearly indecipherable to most in a common lay audience (and even medical) — a pretty good reason to keep the title out of the news article wouldn’t you think? So yes, the title was not in many news articles for a now obvious reason.
Before we go further, another point I should make is that the title of this paper and the title of most news headlines are rather different — another red flag that should make you a little suspicious of the medical news report.
So what does the title even mean? Let’s break it down.
- NaHCO3 is the chemical formula for sodium bicarbonate- essentially baking soda.
- Oral: simply refers to something taken by mouth.
- Splenic: refers to the spleen, a multi-faceted organ in the human body responsible for immune function and recycling of red-blood cells amidst many other functions.
- Anti-inflammatory: this may seem obvious, but like its counterpart inflammation, is often misrepresented and misunderstood. We will revisit this in detail shortly.
- Cholinergic: refers to the neurons or parts of the nervous system that utilize acetylcholine as its primary neurotransmitter or communicating molecule.
- Transmitted: communicated, usually by nerves, but there are other ways that cells communicate with each other too.
- Mesothelial cells: these cells are epithelial cells (barrier cells) that act as a lining for most organs in the body, heart, lungs, abdominal cavity (including spleen), etc.
CLICK TO EXPAND // The science behind cholinergic neurons and mesothelial cells
So now that we have deciphered the title, what exactly is the title trying to say or what can we reasonably infer from the title as presented?
I would suggest the following:
Baking Soda (sodium bicarbonate) at some unknown amount when taken by mouth (by unknown subjects, perhaps mice, rats, or humans) initiates a presumed anti-inflammatory immune response originating in the spleen via a signaling pathway that involves cholinergic neurons/nervous tissue and a specialized subset of cells (mesothelial cells).
Seems reasonable enough.
I hope you see from the title and my summary, (and I know it’s just the title) that conclusions reporting baking soda to be a significant aid to autoimmune disease are more than a bit of a stretch and should elicit at least some form of a pause and curious inquiry.
We will come back to the title when we describe in more detail the main findings of the study, so just keep this tucked away for later. Back to my main concerns.
1. The study was initially designed to explore mechanisms of immune system modulation (pro-inflammatory vs. anti-inflammatory responses) related to kidney disease, both acute kidney injury and chronic kidney disease (CKD). Also, the initial description and study is in rat models, not humans.
Just within the introduction of the paper, the authors begin to explore some potential pathways of immune system activation or pathways that can be activated (possibly by sodium bicarbonate) to create an anti-inflammatory response to help prevent damage to the kidney as part of an acute injury or as part of a progressive chronic disease state.
This is incredibly important research and valuable to study, but this is not directly studying any mechanisms of autoimmunity in humans, and additionally the authors make very general and perhaps vague statements such as:
As inﬂammation has been associated with CKD progression, we speculated that NaHCO3 may act to protect the kidneys by reducing inﬂammation.
The above statement was made in between more specific statements regarding potential cellular mechanisms of immune activation and anti-inflammatory pathway activation such as:
Oral NaHCO 3 intake promotes M2 macrophage polarization by activating splenic anti-inﬂammatory pathways.
Inflammation is not a monster or something you “catch” like a virus. Let me explain.
Inflammation is actually an intricate but coordinated system of neural (nervous tissue) and molecular signaling both locally and systemically employed by the body with the main purpose of detecting, assessing and responding to a perceived local disturbance or disruption in homeostasis. This disturbance, for example, could be a traumatic injury: mechanical disruption of cellular tissue, the presence of a non-self entity (bacteria, virus) or the loss or cessation of normal homeostatic signaling.
We can categorize this system of signaling in primarily two ways
- The primary immune cells involved
- The duration or “time course” of signaling (aka: acute vs. chronic)
The acute inflammatory response occurs over a short time scale, usually days to a week or two while the chronic inflammatory response occurs over weeks to months and even years.
The body also utilizes different cell types to carry out the acute and chronic inflammatory responses. This is a perfect example of specialization as well as the important structure and function relationship.
CLICK TO EXPAND // The science behind the acute inflammatory response
Inflammation is a response that can be characterized by the activation of different immune cell types as well as an incredibly intricate set of signaling molecules or systems attempting to tell the immune cells where and how to respond.
So you see, inflammation is not a monster, but a complicated and often misrepresented response, not a disease.
Okay, back to my main concerns with the paper.
2. The authors appear to have mistakenly misrepresented the baseline data (before ingestion of bicarbonate solution) by the human subjects and do not provide the raw data for the human subjects after ingestion of either bicarbonate or control solution.
This is a big problem. Let’s explore why.
I will point out first the issue with misrepresented or “incorrect” baseline data.
Taken from the study itself:
Baseline ﬂow cytometry values of all subjects, before ingesting NaHCO3 or NaCl in solution, are presented in Table IV. Prior to any treatment, the percentages of blood leukocytes that were TNFa + neutrophils, M1 macrophages, or M2 macrophages were all signiﬁcantly higher in the NaHCO 3 TXT group when compared with baseline values obtained in the NaCl TXT group (Table IV).
Basically what this is saying is that at baseline there was actually a difference in the percentages of certain immune cells between the baking soda group and the salt (NaCl) group. While the authors state this clearly, this is a major confounding variable as the two populations they are seeking to compare- those who drink a baking soda solution and those drinking a NaCl or salt water solution are different at baseline with respect to some of the key immune cells being followed.
But there’s even more that doesn’t fully add up.
In that statement they say that the immune cell percentages are higher in the baking soda group when compared to the salt group, but what does their table show?
Taken from the paper:
I bring your attention to the values in question:
Now I am not a maths whizz but it looks to me like the percentages in the baking soda group are lower, not higher, as the authors state, in all categories when compared to the NaCl group including the two noted as statistically significant (TNF-alpha, and M1 macrophages). Now this could be a simple reversal mistake, I am not sure, but it is a big mistake nonetheless. If we assume that this data is correct, however, by having significantly different baseline levels of certain immune cells, we still cannot reliably draw conclusions on whether any changes witnessed from before ingestion to after between the group are statistically significant or meaningful.
Additionally, as previously mentioned before, there is no table showing the post-ingestion raw data or values for immune cell types. The authors provide multiple graphs of percentage change in different immune cells, but percentages are unitless values and in this case were standardized from a zero baseline that wasn’t originally statistically equivalent. This brings me to concern three, which you may have already picked up from the screenshot.
3. The study involved 12 human subjects.
Now I don’t know about you, but even for something as seemingly innocuous as baking soda, no matter what the authors concluded, I would be very hard pressed to start recommending baking soda to my patients after a trial starting from a rat model that then involved 12 human subjects.
I know we have to start somewhere, but this is an incredibly small sample size.
What the Study Did NOT Find
1. There was a misrepresentation and association of rheumatoid arthritis with the findings of this study.
Note: This is not the fault of the authors, but of news media outlets and shows you why you cannot just accept the medical news summaries.
The first time rheumatoid arthritis is mentioned in the study is on page 12:
Inﬂammatory macrophages and excessive TNF-a production have been implicated in the pathology of a broad range of disease states, including rheumatoid arthritis (17), cardiovascular disease (18), atherosclerosis (19, 20), irritable bowel disease (21), type 2 diabetes (22), and neurodegenerative diseases as well as others (23–26).
The second time rheumatoid arthritis is mentioned in the study is on page 13:
Furthermore, in humans, efforts to stimulate the cholinergic anti-inﬂammatory pathway chronically by implanting stimulating electrodes on the vagal nerves have shown promise in patients with rheumatoid arthritis (17).
This excerpt has nothing to do with what was actually done by the authors in this study. While the authors provide speculation that they are able to stimulate some aspect of this cholinergic “anti-inflammatory” response (a broad and vague statement), in this study they did not test human subjects with rheumatoid arthritis and did not induce this pathway via electrode stimulation of the vagus nerve.
Yet, despite this, nearly all major news outlines took this language and made statements about how this intervention would be a helpful intervention for autoimmune diseases like rheumatoid arthritis.
2. This study involved healthy subjects consuming baking soda (sodium bicarbonate).
This was not a therapeutic intervention. We cannot draw reliable conclusions from this study that this will help people with relative states of chronic inflammation such as chronic autoimmune disease.
CLICK TO EXPAND // Further explanation on the lack of evidence for using sodium bicarbonate as a therapeutic intervention
CLICK TO EXPAND // A detailed explanation of the study's findings
Final Reflections and Summary
If you have stuck with me for this crazy journey, I believe we have arrived at some final, primary conclusions.
- I will not recommend individuals with autoimmune disease to start taking baking soda.
- Taking a whole body systems approach, I will choose to start instead with a dietary intervention such as the Paleo Autoimmune Protocol combined with a thorough investigation into any underlying gut disturbances.
- This study involved only 12 subjects and contained some flaws in the reporting of their data that limit our ability to make reasonable conclusions.
- I encourage everyone to remain cautious about medical news stories and look for our red flags to identify sources that may be distorting the findings of the study.
Thanks for sticking with me and I hope that you can see the forest through the trees and begin to use some of the techniques provided here to cultivate a greater awareness of the distortion within medical news media and how to become your own curious and not cynical citizen scientist.