POSTHOC ANALYSIS: Does drinking diet soda increase the risk of stroke and heart attack?

I posted a report which was picked up by media from one of the findings based on a study presented at the annual International Stroke Conference in Los Angeles recently.

While the conference centered on many breakthroughs and landmark findings in the management of stroke, the study done by Hannah Gardner from the University of Miami Miller School of Medicine grabbed the media limelight.

The study, which followed more than 2,500 New Yorkers for nine or more years, found that people who drank diet soda every day had a 61 percent higher risk of vascular events, including stroke and heart attack, than those who completely eschewed the diet drinks. In this study, researchers surveyed 2,564 north Manhattan residents about their eating behaviors, exercise habits, as well as cigarette and alcohol consumption. The study volunteers were also given physical check-ups that included blood pressure measurements and blood tests for cholesterol and other factors that might affect the risk for heart attack and stroke. The increased likelihood of vascular events remained even after Gardener and her colleagues accounted for risk factors, such as smoking, high blood pressure and high cholesterol levels. Pointing the finger more squarely at diet drinks, the researchers found no increased risk among people who drank regular soda.


Obviously, this report sent the beverage companies as well as the US FDA scrambling at the preliminary report that could potentially damage a multitrillion dollar industry.

The question that looms is - Does drinking diet soda daily increase the risk of vascular events?

First of all, a prospective study that is randomized and well-controlled in design using a very large number of subjects would be impossible to do. Not only because of the enrollment criteria for such a study but based on this survey, it would pose ethical risks for conducting such a study that can put people enrolled in the study at risk for a vascular event (stroke or heart attack). Let me put it simply - it's like doing a prospective study on whether smoking causes cancer or not. All the data for smoking causing cancer are based on retrospective data. Which means that the investigators looked at patients with lung cancer and then looked back into their history of smoking and voila! The studies point to the fact that patients who had a smoking history had a higher risk of developing lung cancer (and other forms of cancer) than those who did not smoke. Remember. The key word here is HIGHER RISK. Which means that even if you did not smoke, you still have a chance of getting lung cancer in your lifetime even if you lived in a plastic bubble.

So why not design a prospective study? I mentioned the ethical dilemma here. Patient safety as well as patient risk should never be compromised in the design of a scientific study. The prospective risk studies are unethical to conduct. You cannot randomize one group of subjects to 1 pack a day for 20 years and compare them to another group of subjects who do not smoke and then follow them up for a period of 20 years! This is unacceptable and cannot be conducted in humans.

Hence, we're left with looking at studies designed in the way Gardner and colleagues did their study on diet soda and the risk of vascular diseases.

Second is the fact that while the authors had statistically analyzed the data by removing some other risk factors such as increased cholesterol and hypertension and history of heart disease in their study, a historical data is difficult to verify and can be unreliable particularly when you're doing a survey.

I have taught epidemiology even before epidemiology even was a formal course in the Philippines. So I would know what I am talking about when I say that the difficulty in conducting surveys is the veracity of the response of subjects. Reliability can range from 0-100% and the person conducting the survey (enumerator) would have to be in the cast of "Lie to Me" (the TV series) in order to spot whether the subject being surveyed is truthful or not. However, the considerably large number of subjects in this study would possibly account for outliers. But like all surveys and retrospective studies, we take everything as face value and analyze the data given what we have.

This is how studies on linking smoking and lung cancer, marijuana intake and lung cancer, sun exposure and skin cancer, HPV infection and cervical cancer, hepatitis B and liver cancer were born. No one does a prospective study.

Which goes to the analysis on other factors which MAY have contributed to the higher risk of stroke and heart attack of patients drinking diet soda.

In this diet conscious era, we all like to look like supermodels. Yet want to eat like pigs! (Ouch, did I hurt anyone?) So some beverage company comes us with a slogan that we can go ahead and drink all the soda we want without the guilt feeling of having to consume more calories. But the human mind (and the gastronomic capacity) will always reason out that since we're having a can of Coke Zero then we have saved enough calories to go ahead and devour a chicken and go for the unlimited rice galore at Mang Inasal!

So let's do the math.

A regular coke will contain the following calories depending on the amount you take in:

8 fluid oz. = 105 calories
12 =155 calories
16=200 calories

And I'm talking about the bottled beverage. A can of coke is about 140 calories while coke from the dispenser will vary from each fast food outlet. For example, an 8 oz coke from Wendy's is about 160 calories while 8 oz coke from McDonald's is 250 calories. That's because the dispensers will have varied caramel-sweetener-water mixture.

And while we're busy counting calories from the beverage drink, we forget the food we eat.

Let's do the math again. This time we drop the regular coke and exchange it with Coke Zero (0 calories) and go for the extra rice.

In short-grain white rice there are:
approx 267 calories in 1 cup steamed or boiled

In medium-grain white rice there are:
approx 242 calories in 1 cup steamed or boiled

In long-grain white rice there are:
approx 205 calories in 1 cup steamed or boiled

In medium-grain brown rice there are:
approx 218 cals in 1 cup steamed or boiled

In long-grain brown there are:
approx 216 calories in 1 cup steamed or boiled

In basmati rice there are
approx 190 calories in 1 cup steamed or boiled.

In wild rice there are:
approx 166 calories in 1 cup of steamed or boiled

In Chinese restaurant steamed rice there are:
approx 199 calories in 1 cup (loosely packed).

So you've balanced everything off with the same junk you disposed to the junk you took in. Talk about zeroing it out then adding more!

Put that in the Pinoy's perspective where noodles are served in parties as viand and not as main dish and here's what you get with eating pasta or noodles together with your rice:

Without sauce, butter, or oil, there are:

BY THE OUNCE OR 28g
approx 38 calories in each ounce or 28g of cooked fresh pasta
approx 44 calories in each ounce or 28g of (enriched) cooked pasta
approx 35 calories in 1 each ounce or 28g of (enriched) whole wheat pasta

BY THE SERVING
approx 133 calories in a 3½ oz or 100g serving of cooked fresh pasta
approx 152 calories in a 4 oz or 114g serving of cooked fresh pasta
approx 304 calories in an 8 oz or 227g serving cooked fresh pasta
approx 608 calories in a 1 pound or 454g of cooked fresh pasta

BY THE CUP
approx 175 calories in 1 cup (5 oz) of (enriched) cooked whole wheat spaghetti
approx 182 calories in 1 cup (4.1 oz) of cooked small shells
approx 190 calories in 1 cup (5 oz) of cooked fresh spaghetti
approx 212 calories in 1 cup (4.7 oz) of cooked spiral shaped
approx 221 calories in 1 cup (5 oz) of (enriched) cooked elbow shaped
approx 221 calories in 1 cup (5 oz) of (enriched) cooked spaghetti

And I could go on and on, but the take home message is that while we are gullible enough to believe advertising gimmicks (all the models for diet sodas are reed thin), the studies do not take into consideration the daily dietary practice of people who exchange another serving of food for the zero calories in a beverage drink.

Finally, is the fact that artificial sweeteners and coloring have long been studied to be carcinogenic in mice and laboratory animals.

While it may be a fact that aspartame or food coloring (caramel) will need to be consumed in larger amounts in humans compared to animal studies to probably cause cancer or vascular problems, the same science is true for the amount of nicotine exposure to animals before there is lung damage in humans. But the US FDA will not comment on this issue because undermining the diet soda industry will definitely send the stocks of Coca-Cola tumbling down and mark the end of the era of an American beverage industry that supports American Idol. More so, not at this time when the American economy is still reeling from a financial deluge.

So what do the studies show?

If you plan to go on a diet because of nutritional or health reasons - please count the calories. If you need to take something to drink, don't just go for the zero calories but with artificial sweeteners. There's always water which has zero calories as well. This study may not have enough data to shake the rabid eating and drinking habits of people, but it's a wake-up call that we need to improve the overall dietary habit in order to lead healthier lives.

After all - you are what you eat!

[Calorie facts from righthealth.com and Nutrition Center of the Philippines]

Fraud in medicine

I received a text message the other day from the mom of a concerned patient asking me if I had been following CNN's Anderson Cooper's show regarding the relationship between MMR and Autism. And if I did, what was my opinion?

How did this controversy ever start? Who is Dr. Andrew Wakefield? What was the study all about?

Dr. Wakefield is (or was until he was stripped off his license to practice for fraud and conflict of interest) a British surgeon and researcher who published a paper in February 1998 linking autism and vaccines in the prestigious journal the Lancet. The study was done in 12 children who had bowel problems and diagnosed to have autism. Of the 12 children, 8 were supposedly received MMR vaccine while 1 developed measles infection prior to the onset of the autism symptoms. After the publication, many parents became indignant to the pharmaceutical companies that they opted to defer or delay MMR (measles, mumps, rubella) immunization for their children. Subsequent decline in vaccine coverage led to outbreaks and deaths among afflicted children.

Let me answer some queries of parents.

1. "If there was not enough proof on the findings of Wakefield, why did a prestigious journal like Lancet publish his study?"

I served as Editor-in-Chief for 15 years with the Santo Tomas Journal of Medicine (Official Publication of the Faculty of Medicine and Surgery of the University of Santo Tomas) and Editor-in-Chief for 8 years with the Philippine Journal of Pediatrics (Official Publication of the Philippine Pediatric Society). These, aside from various positions in the editorial boards and reviewer of several manuscripts published locally and internationally, form the basis for my right to critique the publication of Wakefield and how the public and media wrongfully perceived ONE article.

Let me remind everyone that the journal is an avenue for sharing data and findings of various researchers. Unlike the newspapers or mainstream media, each scientific article is unique in the sense that it is a complex structure based on the design of the study and the statistical treatment on the given data. Studies can be anywhere from case reports to complex cohort studies and the interpretation of said studies may be thwarted unnecessarily by those NOT trained to interpret results and findings from the published literature.

All studies are peer-reviewed. Which means that those who reviewed the article agree that the paper is worthwhile publishing. It does NOT imply that we agree with the findings. For a simple case-series (12 cases do not make a formidable study and is the one of the weakest of all types of studies) like Wakefield's, many other researchers worldwide attempted to duplicate his study by either doing larger case-series or case-control studies BUT were NOT able to duplicate his findings. It is up to the astute physician or reader to critique and interpret the published studies well. Unfortunately, misinterpretations by media men who are not even trained to discern a good study from a fraudulent one ends up as sensationalized journalism.

Simple studies like Wakefield's are published to trigger the medical and research community to do larger and extensive studies in order to demonstrate positive findings in favor of the his data. Which is the core of research. One study DOES NOT make the rule. One study should be able to trigger prospective studies on a larger scale basis in order to demonstrate either reproducibility and fact OR fraud.


2. Why was the article published initially then called fraudulent later on?

With so many researches in whatever field of endeavor (business, IT, medicine, science, pharmacy, engineering, etc.) it is difficult to make 3 or 4 people conclude that what he or she is reviewing is fraudulent or not.

Anne Hudson and Faith McLellan published a book "Ethical Issues in Biomedical Publication" [The Johns Hopkins University Press, 2000] that tackles ethical dilemmas in scientific publication. And misconduct in research (like misconduct in various forms of business and media) can range anywhere from misuse and manipulation of data to conflict of interest to publication bias.

Plagiarism and fabricated data has been the most clear example of publication scandals. Even our Supreme Court was not spared from such moronic acts. The most well known cases of scientific misconduct on publications are the Darsee and Slutsky cases.

John R. Darsee was a Harvard student and was only 31 years old when he joined the Cardiac Research Laboratory of world renowned cardiologist and editor of cardiology textbooks of medicine, Dr. Eugene Braunwald. The mentors of Darsee at Emory University tagged him as a brilliant and outstanding physician.

"...Until May 1981, when he was observed faking lab results of experiments on dogs on two separate occasions, Darsee seemed to be living up to the promise of his recommendations. He worked hard, published extensively, and gained Braunwald's deep trust and respect. Indeed, when accusations of misconduct were brought against Darsee, Braunwald at first suspected his accusers of being motivated by jealousy. And when Darsee confessed to falsifying data but insisted that he had done so on only one occasion, Braunwald believed him, accepting a "single, bizarre act" as an explanation for Darsee's behavior. It was months before Braunwald was persuaded that Darsee had engaged in widespread pattern of misconduct, which had begun when he was an undergraduate biology major at Notre Dame and continued during his years as a cardiology resident and fellow at Emory. It took years to investigate Darsee's research thoroughly and to determine how many of his publications were based on fabricated data....When the dust settled, investigators had revealed that Darsee's nine Harvard papers, co-authored by Braunwald and Braunwald's research deputy Robert A. Kloner, were all based on fabricated data and had to be retracted. These papers had been published in prestigious journals, such as the American Journal of Cardiology and Proceedings of the National Academy of Sciences. In addition, most of the 10 papers and 45 abstracts that Darsee co-authored with researchers at Emory could not be validated had had to be retracted..."

Robert A. Slutsky was a mirror reflection of Darsee. Slutsky started as a radiology resident and unsalaried associate clinical professor at the University of California San Diego. It was startling that Slutsky was producing 1 paper every 10 days for several years. Even if you were not seeing patients and were just reviewing procedures in radiological science, it would be practically impossible to get that many researches in a span of days.

"...Despite this extraordinary output, which should at least have attracted critical attention to his work, Slutsky's fabrications went undetected until early 1985, when he was being considered for appointment as an associate professor of radiology at UCSD. Then, a member of the departmental committee evaluating Slutsky's work noticed duplicate data in two of Slutsky's articles. Questioned about the duplication, Slutsky abruptly resigned his appointments at UCSD...In the investigation that followed, a ten-membered UCSD faculty committee determined that, of Slutsky's 137 publications, 77 were valid, 48 were questionable, and 12 were fraudulent...."

These are not isolated cases and perhaps there is more in the published medical literature. Which is why we teach medical students to be able to discern the quality of the published literature. Not everything you read is the golden rule.


3. Are you absolutely sure there is no link between vaccines and autism?

No I am not. And there is no such a thing as a drug being absolute free of side effects. Even if John Lloyd winks back at you on an advertisement on Biogesic being safe, it is moronic AND ironic that the public and media are willing to believe some lackey actor endorsing a drug to be devoid of side effects when in fact, Paracetamol (the generic name of Biogesic) is the number one drug worldwide that causes drug-induced hepatitis.

Which means that we must remain vigilant about drugs and vaccines in the market. Vaccines remain to be the major discovery of the 20th century. Without vaccines, more people would have died. Every drug or vaccine that comes out in the market are not fool proof or should I say, without side effects. During the initial phases of the studies, we only use a few thousand patients. So if the drug or vaccine, say, causes a side effect of autism in 1 out of every 100,000 patients that receive it, you will not be able to detect that in the initial clinical trials.

Add to that the benefit:risk ratio. Which means that without the vaccine 1 in every 100 children afflicted with measles will die. 1 in every 10 child afflicted with measles will require hospitalization. 1 in every 10,000 children afflicted with measles will have neurologic sequelae. Therefore on a public health scale, the implementation of a vaccination program is more cost effective and will have greater impact on health benefit for the majority rather than the few who MAY purportedly develop autism.

After reviewing the case of Wakefield, the following conclusions were arrived at:

1. He received money from lawyers of parents (to the tune of close to 500,000 pounds) for his work on the paper. Parents of the 12 children in the study of Wakefield were recruited through a lawyer in the UK to prepare a lawsuit on the manufacturers of MMR.

2. No other study was able to duplicate the study of Wakefield. In addition, Wakefield was not a pediatrician nor a developmental specialist. All he did was take the word of the parents that their child was autistic and had received MMR and concluded that they developed problems after the vaccine was administered. On thorough review of the data of the 12 patients in Wakefield's study, many of them had developmental problems even BEFORE they received the vaccine. In addition, what explains the fact that there were 4 patients that never received the vaccine and YET were diagnosed to have "autism"? One was said to have developed measles infection. If this line of thinking were true then that should put all patients who develop measles infection at risk of developing autism as well.

3. I have patients diagnosed to have autism and have never received any form of vaccine (let alone MMR) whatsoever. What explains their autism?

The cause of autism remains unknown today. There are several theories, but no one knows the exact cause. But with the evolution of science in leaps and bounds, we may eventually be able to unfold the mystery regarding autism.

The increasing cases in autism is currently not related to MMR vaccine. In spite of the precipitous drop in MMR vaccination in the UK and the US after the Wakefield publication, evidence shows that the number of autism cases had still increased. Are we seeing an epidemic? Is it environmentally related? Or is it genetic? Or perhaps it is because we have greater awareness of the diagnosis of autism that we are able to label patients as autistic earlier on.

I pity actually the parents of autistic children who are dying to defend the stand of Wakefield. And while I commiserate with them I cannot condone their line of thinking on what they believe. An open mind and understanding on the spectrum of autism should be their focus so that they can begin to care for their children afflicted with this disorder. They need to focus on providing care and helping out in the dissemination campaign on appropriate education and understanding on how we can help families with autistic children. I am not asking them to embrace the disease and walk away. I am simply asking them to look beyond one man's fraudulent publication and belief and focus on their children who need their care and attention more than defending Wakefield's ludicrous claims.

Media should also desist from providing biased reporting. If they do not understand the published literature, they should ask someone with knowledge and expertise to interpret it for them at a fair airtime and not blab blab blab on their one sided interpretation! A little knowledge is deadlier than having no knowledge at all.

In the history of pharmacology, it is best to remember that all drugs are double bladed swords. One man's medicine may be another one's poison. After all, science is fallible.

[Reference:
Jones, AH: "Changing traditions of authorship" in Ethical Issues in Biomedical Publication, Ed: Anne Hudson Jones and Faith McLellan. The Johns Hopkins University Press, Baltimore USA, 2000: 7-12

Photo from infowar.com]