by Steve Hickey
A Mean Regression
One strange and not fully explained phenomenon of clinical trials is that the first time an experiment is done, the results can be excellent, but subsequent experiments produce less dramatic improvements with the treatment. This trend is a form of regression to the mean, in which extreme results tend to become more average over time. Regression to the mean can result in wrongly concluding that an effect is due to treatment when it is due to chance. It is also an explanation of how the data from clinical trials can apparently support the existence of the placebo effect.60 A group of people selected for a clinical trial is not representative of the population but biased, by definition, as they are sick. Often patients will get better with time, or regress to the mean, over the period of the trial. In most trials, regression will be indistinguishable from a placebo effect.
In 1886, British statistician Francis Galton (1822–1911), a half cousin of Charles Darwin, first described “regression towards mediocrity,” which now has the more politically correct name of regression to the mean.61 Galton measured the height of adults and their parents. When the parents’ height was greater than average, the children were shorter than the parents. However, when the parents were shorter than average, the children tended to be taller than their parents. Flight instructors provide a more recent example: when instructors praised a trainee pilot for an excellent landing, typically their next landing attempt was poor.62 It is easy to see how the instructors thought that praising the trainees was counterproductive, but the real explanation was regression toward the mean.63
In some studies, patients diagnosed with a particular disease may have another malady or could even be healthy.64 Clinical diagnosis can be subjective. In these cases, the experiment can give completely misleading results, as the subjects do not even have the disease being considered. Another source of error arises when patients receive more than one treatment at a time and the therapies interact.65 The results obtained could result from any one of the therapies or from some interaction between them. This multiple treatment interference is not easy to disentangle from the raw data obtained in an experiment and is a more particular problem when a large number of variables are considered.
Unreliable Results
The justification of both clinical trials and epidemiology is based largely on statistics, but many clinical studies provide unbelievable statistical results. Peter Gøtzsche, of the Nordic Cochrane Centre and Network, recently investigated statistics in clinical trials published in 2003.66 He found that many of the reported data were potentially biased. The results were calculated from small selections of the subjects studied or were derived from a biased collection of results. These potentially prejudiced results were reported without explanation in 98 percent of 260 trials. Gøtzsche was able to check the calculations for twenty-seven of the trials and found that a quarter of all results deemed “non-significant” were actually significant. Furthermore, four of the twenty-three reportedly significant results were wrong, five were doubtful, and another four were open to interpretation.
Significant results in abstracts should generally be disbelieved. Abstracts of papers are often the only section read in detail and frequently contain unsupported data.67 When a new drug is studied, the bias typically presents the drug in a more positive light. For example, in all but one of eighty-two studies, bias in the conclusion or abstract of comparative trials of two anti-inflammatory drugs consistently favored the new drug over the control drug.68 Based on this information, doctors might prefer to prescribe a profitable new drug when it actually has no benefit over existing medicine.
Many clinical trials are designed to produce significant results. A large number of statistical tests are often built into the trials. Indeed, trials which include more than 200 statistical tests are sometimes specified in protocols.69 A trial with 200 tests of identical drugs is almost certain to produce one or more significant results by chance alone. Thus, statistically significant improvements will be reported for a new drug even if it was identical to the one with which it was compared. By boosting the number of tests, a drug company can be fairly certain that it will get a result that will be useful for marketing the drug.
Additional measures are sometimes used to massage clinical results. Studies comparing study designs with the actual published reports of clinical trials have revealed selective use of data to overestimate the effectiveness of the treatment.70 This selective reporting bias occurs even in “high quality” government-funded studies.71 As a result, the medical literature represents a selective and biased subset of study outcomes72 and often promotes new drugs that have no greater benefit than existing treatments.
Meta-Analysis
Meta-analysis is a technique that is increasingly used to summarize results from a number of clinical trials. One recent analysis suggested that antioxidant vitamins increase the risk of death.73 However, the researchers selected a small number (68) of papers from the many (16,111) they initially considered. The selection was carried out by the researchers themselves with full knowledge of the results of the studies. Clearly, an unconscious bias on the part of the researchers could produce a misleading result.74 Consistent with media predispositions against supplements, this paper was given a high profile, both in the general news and in medical reporting. Vitamin C provides a particular target for these inaccurate accounts, and the steady over-reporting of vitamin C scare stories is likely to contribute to the public’s gradually eroding confidence in the medical establishment.
Conventional medicine’s failure to even consider the effects of massive doses of vitamin C on a range of illnesses is a result of assuming that it has been “proven” that only low doses are needed. As we have seen, researchers erroneously suggested that the body was saturated at only 200 mg and no more would be absorbed. With this background, it is understandable that the apparently wild claims for larger doses of vitamin C were considered unscientific. After all, doctors may have thought, it is only vitamin C, a harmless white powder, found in fruit. Even if a vitamin tablet makes you feel better, they could attribute this to a placebo effect.
The problem is that the case against vitamin C is weak, and both the size of the doses and the claimed efficacy are much greater than conventional medicine could apparently comprehend. Placebo-controlled clinical trials are a “gold standard” in medicine but are often viewed with suspicion by the public and scientists from more rigorous disciplines. The claims for the efficacy of vitamin C are objective, uniquely large effects and cannot be explained as placebo. Reliance on clinical trials and “proof” holds back advances in medicine. All medical treatments depend for their justification on the basic sciences of biophysics, biochemistry, and physiology. Vitamin C has such justification.
CHAPTER 5
The Need for Antioxidants
“The truth is, the science of Nature has been already too long made only a work of the brain and the fancy. It is now high time that it should return to the plainness and soundness of observations on material and obvious things.”
—ROBERT HOOKE, MICROGRAPHIA
Scientists suggest that long ago, before life evolved, the earth’s atmosphere contained little oxygen. However, as primitive microorganisms and plants progressed, they developed the capacity to use energy from sunlight to make their own food (glucose) from water and carbon dioxide, giving oxygen as a byproduct. This process is called photosynthesis and, directly or indirectly, it provides the energy for most life on earth. Over millions of years, the concentration of oxygen in the atmosphere grew higher, allowing the evolution of organisms that use oxygen to burn fuels to create the energy for life, as we do.
All life is based on chains of carbon, hydrogen, nitrogen, and oxygen atoms, which bond together by sharing electrons to form molecules. Unfortunately, oxygen has a penchant for a somewhat one-sided sharing. Oxygen’s strong electron pull distorts such organic molecules by stealing some of their electrons, a process known as oxidation. In an increasingly oxygen-rich atmosph
ere, organic molecules would combine with oxygen and their structure would be damaged.
A spectacular example of the process of oxidation familiar to us all is burning. For example, when we burn coal for heat or petroleum in car engines, the coal or gas is oxidized. The process does not have to be so dramatic, however. If cooking oil is exposed to the air, it deteriorates slowly, combining with oxygen. The same is true when a cut apple turns brown. This reaction requires energy, but is essentially similar to slow burning. The energy supply for our bodies also involves oxidation: we slowly metabolize or “burn” molecules from our food, producing chemical energy. As a result, every cell in our body needs a constant supply of oxygen to feed our internal biochemical fires. Antioxidants, such as vitamin C, are substances that help prevent damage to the body from oxygen.
Oxidation and Free Radicals
A cell is the smallest unit of life: it can maintain itself, grow, and reproduce. Most organisms, such as bacteria, exist as single-celled creatures, able to thrive without much cooperation with others. To create larger organisms, cells need to cooperate. Single-celled organisms do not have arms, legs, or eyes, thus have a limited scope for interacting with their environments. Large multicellular animals have the advantages of sight, organized movement, and thought, as a result of cooperation between cells. Living cells need oxygen to burn food and produce energy. The main exception to this rule is anaerobic bacteria, which cannot live in the presence of oxygen. However, most cells have evolved antioxidant defenses to provide essential barriers to the poisonous effects of too much oxidation.
Oxygen has the paradoxical property that, although it is essential to life, it can also be a deadly poison. We need oxygen to metabolize food, but it can also attack our own body tissues. One reason may be that our tissues are similar to the food we eat, since they are made of the same molecules: the building blocks of our bodies and cells are organic molecules consisting largely of water, proteins, and fats. Both meat and vegetables in our food are composed of these same materials.
To prevent such damage, our cells have evolved numerous antioxidants. An antioxidant is the name given to any mechanism or substance that prevents oxidation. Our cells use a substantial proportion of their energy to manufacture a large array of antioxidants, which shows how important these mechanisms are. Biological cells are full of antioxidants, which protect them from the oxidizing environment, but this protection takes energy. Thus, a sliced apple exposed to air turns brown, as its antioxidant defenses are overwhelmed by exposure to oxygen. Most diseases and features of aging also involve oxidation reactions.
Pure oxygen can kill cells and tissues, but the damaging process of oxidation does not require oxygen—many other molecules share this chemical property. Any substance that can steal electrons from other molecules is called an oxidant. For example, ultraviolet light and x-rays can knock electrons out of organic molecules, causing oxidation. Such ionizing radiation creates free radicals.
Free radicals are energetic molecules that can cause oxidation and tissue damage. Before we can understand what a free radical is, we must look more closely at the structure of atoms and molecules. Atoms contain a nucleus of neutrons and protons surrounded by a cloud of electrons. An electron has a negative electrical charge that generates a magnetic field as it spins. In molecules, most electrons exist in stable pairs, each spinning in opposite directions so that their magnetic fields cancel each other out.
A free radical is a reactive molecule with one or more unpaired electrons, which can be highly reactive.1 This means trouble, because these energetic molecules will grab electrons from other molecules. Since atoms and molecules in the body are constantly vibrating and moving about, free radicals can rip electrons right out of essential proteins or even DNA. Uncontrolled free radical reactions cause tissue damage too. What is worse, free radicals can produce chain reactions within our cells, as one molecule after another becomes oxidized.
Oxygen, as you might expect, is a free radical. The unique chemistry of oxygen is partly a result of it being stable despite having two unpaired electrons. It has two electrons spinning in the same direction, which makes it magnetic. Magnetic substances, like iron, have at least one unpaired electron. The phrase “reactive oxygen species” describes a number of free radicals and energetic molecules derived from oxygen. Some reactive oxygen species are not free radicals but rather highly energetic oxidants. These produce free radicals within tissues and include hydrogen peroxide, hypochlorous acid, ozone, and singlet (one-atom) oxygen. Reactive species can also be formed from nitrogen and from chlorine, as found in common household bleach.
Types of Free Radicals
Superoxide molecule is a reactive oxygen species. It is similar to the oxygen molecule (O2), but the superoxide molecule is oxygen packing an additional electron (·O2−). It does not live up to its dazzling name, however, as it is not very reactive. While superoxide reacts rapidly with free radicals, its reactions with organic molecules in the cell require an acidic environment.
The most abundant substance in our bodies is water. Water molecules consist of two atoms of hydrogen and one of oxygen (H2O). Several important free radicals are related to this simple molecule. The hydroxyl ion (·OH) is water that has lost a hydrogen atom. The dot (·) in the formula indicates that it has an unpaired electron. Hydroxyl ions are highly energetic and particularly damaging: they react instantaneously with almost any molecule they meet on their rapid diffusion through a cell. Without a large concentration of antioxidants, a hydroxyl radical can trigger a chain of damaging reactions.
If a hydroxyl ion steals an electron from an essential protein or DNA molecule, it becomes inert or unreactive. However, the molecule it stole the electron from now has an unpaired electron and has itself become a free radical. This sequence is called a free radical chain reaction. This chemical chain letter continues until one of the molecules changes its essential structure, disabling its function. Alternatively, it may be stopped by an antioxidant donating an electron to render the free radical safe and stop the damaging chain of events.
Addition of an oxygen atom to water forms hydrogen peroxide (H2O2). Hydrogen peroxide is well known for its action as an oxidizing agent in hair bleach, but in much higher concentration it can be used as rocket fuel. This potentially reactive molecule is present within our cells and tissues, fortunately at a low level. While hydrogen peroxide is normally considered harmful, recent research suggests that it is an essential part of signaling mechanisms used for cellular control. Higher concentrations of hydrogen peroxide act as an oxidizing agent, especially in the presence of iron or copper. This reaction with iron produces destructive hydroxyl radicals and free radical damage. The reaction of hydrogen peroxide with iron is given the name “the Fenton reaction” in general chemistry, named after the chemist Henry Fenton, who described the reaction at the end of the nineteenth century.
Antioxidants and Reduction
Antioxidants prevent oxidation damage by donating electrons to replace those lost to oxidation. This process of providing electrons is the reverse of oxidation and is called reduction.
Antioxidants are called free radical scavengers because they neutralize free radicals in the body. An antioxidant, such as vitamin C, can donate electrons to the free radical, thus stopping the free radical from stealing another electron. Vitamin C’s donated electron allows the maverick free radical electron to form a balanced pair.
Oxidation: losing an electron
Reduction: gaining an electron
In the example of a cut apple, we can prevent the browning by applying an antioxidant. If the apple is left untreated, oxygen in the air oxidizes the surface tissue, pulling out electrons and turning it brown almost immediately. A solution of vitamin C, or lemon juice, applied to the cut surface of the apple slows down the oxidation process. The surface of a treated apple will stay fresh and white for a considerable time. The addition of vitamin C provides a supply of antioxidant electrons, preventing damage and discoloration.
Antioxidants are often added to food, such as deli meats or bread, to maintain shelf life and freshness.
The apple experiment illustrates something more profound than the simple preservation of food. It shows how a living tissue can be damaged by oxidation and, more importantly, how an antioxidant can maintain tissue health. The uncut apple sealed within its skin is sheltered from the air and oxidation damage; its natural antioxidants can maintain the health of the unexposed tissue. When exposed, however, its antioxidants are insufficient to prevent the damage. The cut apple may be protected from oxidation by addition of an antioxidant. Similarly, our tissues have adequate antioxidant defenses to provide for a lifespan of about eighty years. However, these defenses are insufficient to prevent the damaging oxidation reactions in acute illness or chronic disease.
Our tissues maintain a controlled balance of reduction and oxidation, called the “redox state.” Cells produce antioxidants and antioxidant electrons continuously in order to prevent oxidation damage. When the supply of energy from metabolism is disturbed, the balance between oxidants and antioxidants breaks down. An example of this is the damage caused by heart attacks or occlusive strokes. In these conditions, an artery supplying the tissue is blocked by a blood clot, leaving the tissue deprived of oxygen. The cells are unable to generate sufficient metabolic energy, so they divert energy away from production of antioxidant electrons. If the artery opens again (for example, if the blood clot dissolves), oxygen rushes into the tissue and the cells find themselves right back in an oxidizing redox environment, but now with a shortage of antioxidants. As a result, heart or brain tissues that survived the original insult may die. The now abundant supply of oxygen triggers a massive burst of free radicals that kills and damages cells. This process is described as reperfusion damage and it may be inhibited by supplying suitable antioxidants.2