Perhaps the more rigorous may have to revise a few certainties. It is a little easier to explain the strange results of an American study carried out in Texas and published two years ago in one of the most respected medical magazines of the world, the New England Journal of Medicine. In an experiment intended to assess the effectiveness of a very common knee operation, carried out on 650,000 individuals a year in the United States, at a cost of US$ 5,000 each, people who underwent a false surgery, with three superficial cuts on the knee, improved just as much as those whose had been submitted to the real surgery, with the removal of worn out parts of cartilage.
The explanation? The improbable recovery of someone who had experienced the simulated operation is due to the conviction that surgery really would eliminate the pain in the knee, evidence that thought can manage to modify the workings of the body. It is the so-called placebo effect: something that in principle should not work, from the physical and chemical point of view – like the superficial cuts instead of surgery, or pills of flour instead of pills with an active ingredient – in practice can work and, astonishingly, eliminate pains, lower the blood pressure, alleviate anxiety, and diminish depression.
This phenomenon, which has been known about for almost two thousand years, since at least the Rome of emperor Marcus Aurelius, is only now beginning to be revealed from the biochemical and physiological point of view. It is the expectation that the surgery will be effective that alters the performance of the central nervous system, even though in practice it is merely a simulation. Activated by the imagination of the patient, some areas of the nervous system associated with the perception of pain become less active, while others, related to the inhibition of pain, are activated, according to recent studies. The non-treatment is thus a sort of inducing to error accepted by the brain itself: what has never been becomes being itself.
The nuances of the real
The recent discoveries rekindle a debate on ethics about whether, when, and how to use placebos in testing new medicines and treatments, or even as a palliative therapy in some chronic diseases, like migraine or gastritis, provided that they do not involve a risk of dying. Up until today, researchers have held it to be certain that the most efficient way of discovering the real capacity for curing of a new medicine or a new operation is to compare the real treatment with a placebo – in theory, as good as nothing.
Before releasing a medicine to be consumed by the population, the medicines control agencies – like the Food and Drug Administration (FDA), in the United States, which acts as a benchmark for Brazil, or the European Monitoring Center for Drugs and Drug Addiction, in Europe – require tests with human beings, with the use of placebos, the so-called double-blind clinical tests, in which neither the doctor nor the patient know who is getting medicine and who a placebo.
This is, incidentally, one of the few situations in which the use of a placebo meets with consent. The Declaration of Helsinki, a set of ethical standards that govern the use of placebos in studies with human beings, determines that in any medical study the best diagnostic or therapeutic treatment should be assured to all patients – including those from the control group, if there is one – and researchers may adopt a placebo only in exceptional cases.
For some time it has been suspected that placebos were more than something that simply does not exist, but that manages to do good – an imaginary remedy, in short. Derived from the Latin verb placere, which means to provide pleasure or to please, the word placebo appears in medical books at the end of the 18th century. But the concept of the placebo effect really gained weight after the Second World War, when medical research began to reveal alterations in the workings of the organism produced by pharmacologically innocuous substances.
In 1955, American anesthesiologist Henry Knowles Beecher published, in the Journal of the American Medical Association, a provocative article that became a landmark: The powerful placebo. Beecher analyzed 15 clinical studies in which a part of the volunteers was given a placebo to treat pain, cardiac disturbances, and gastric problems. From 21% to 58% of the patients, a figure that varied according to the problem shown, improved by taking only the compounds without any pharmacological action.
And one third became a magic number. Even today, doctors and pharmacologists believe that a proportion of people, similar to the one verified in the study of half a century ago, improve because of the placebo effect, not from the specific action of the active ingredient of the medicines, in spite of the advances of the chemical and pharmaceutical industries in this period. The most recent revelation about this effect shows the placebo in action on the nervous system in real time.
Using a nuclear magnetic resonance apparatus, the team led Jonathan Cohen, from Princeton University, in the United States, produced images of the brain in activity of people who took part in two distinct experiments, supposedly aimed at testing the effectiveness of a new analgesic – in actual fact, an innocuous cream.
Lying down in the inside of the apparatus, the participants had to quantify their pain when given a shock on the fist in the first test, and, in the other, they would rank the pain caused by the contact with a hot object on their forearm. The computer screen would show by means of colors the intensity of the activity of each area of the brain. In both experiments, presented in an article in Science of February 20 of this year, the people claimed to feel less pain the in the tests done after applying the cream.
The simple information that they had been given a dose of analgesic ointment – actually, this cream without any pharmacological action – was sufficient to diminish the activity of four regions connected with the perception of pain: the anterior cingulate cortex and the somatosensory cortex, the insula, and the thalamus. The lower the activity of these areas, the greater the pain relief. In contrast, these same areas remained more active in those who believed that they had been given another cream, without any action against pain – in both groups, the same inactive ointment was applied.
Cohen’s team planned the second experiment, in which the hot object stayed in contact with the skin for 20 seconds, so that it would be possible to accompany the evolution of the activity of the nervous system and the action of the placebo, right from before the painful stimulus – a phase called anticipation, when a lit-up sign with the words “Get ready!”, advised the volunteer that he would shortly come into contact with a hot object – until the pain ceases completely.
As soon as people were given this warning, the researchers found that there was a considerable increase in the activity of the prefrontal cortex – a part of the brain located above the eyes, which regulates other areas of the nervous system. Next, on the monitor of the computer that showed the entire brain, another region of the nervous system located in the mesencephalon, associated with the production of natural analgesics as potent as morphine, would light up. In the view of Jonathan Cohen, the coordinator of this research, the greater activity of the prefrontal cortex at this stage results from the expectation of relief and diminishes the workings of the areas responsible for the perception of pain. It was for this reason that the volunteers reported a 22% reduction in the intensity of the pain in the second experiment.
But the mere expectation that the treatment would work does not explain the results completely. In two other experiments carried out in Italy, Martina Amanzio and Fabrizio Benedetti, neuroscientists from Turin University, discovered another component of the placebo effect, conditioning – something like what the Russian physiologist Ivan Pavlov saw, when he trained famished dogs to salivate when they heard a bell ring.
In a similar way to the conditioned reflex of the dogs, the consumption of a given anesthetic repeatedly accustoms the body to react in the same way when an injection of anesthetic is replaced by a dose of water with salt – unknown to the patient, of course. The consequence is that the central nervous system makes the body itself produce compounds against pain, in an almost automatic reaction to a known stimulus, such as, for example, feeling one’s mouth full of water when the chocolate cake is ready.
One of the difficulties of the Italian team’s experiment was to draw up a test that would prove the power of conditioning. The researchers from Turin put together a complex battery of tests, with 229 people, separated into 12 groups, to assess the results better – in each group, a different strategy was adopted. Each participant went through five assessment sessions, in which the same exercises were repeated: they would press with their hands the levers of an apparatus with springs – a dynamometer –, while an airbag of an apparatus for measuring pressure, attached to the arm, blocked the passage of the blood to the volunteer’s forearm. With each push, the pain in the would increase rapidly and in a matter of minutes would reach an unbearable level.
Martina and Benedetti saw that people would tolerate the pain for a longer period if, ten minutes before the test, they were treated with an injection of morphine. The most shocking thing is that the resistance to pain was similar to the resistance achieved with morphine when the person was given an injection of salty water in the third session of exercises, after having taken the analgesic for two tests running. If, instead of salty water, the injection contained a drug that cuts the effect of morphine – naxolone – the tolerance to pain would diminish. It was a clear sign that the continued use of morphine would condition the nervous system to activate the areas that produce analgesics of the group of opioids, to which morphine belongs.
In a similar way, the placebo also induced the body to activate its own source of analgesics – but by a different mechanism – after the application of a dose of ketorolac, a potent non-opioid analgesic, instead of morphine. There was, however, a surprise: the resistance to pain increased even more when, while the placebo was being injected, the researchers induced the participants to believe that the innocuous compound was morphine or ketorolac. It was clear that the placebo was acting on the pain by distinct neurological mechanisms (opioid and non-opioid) activated partly by expectation and partly by conditioning.
In an even more curious experiment, Martina, Benedetti and another neuroscientist from Turin, Claudia Arduino, revealed one aspect of the placebo hitherto unimagined: its beneficial action may manifest itself in specific parts of the body, to which attention and expectation are directed. In a test that recalls a scene of torture, the researchers injected, simultaneously, in the backs of the hands and feet of each volunteer a dose of capsaicin, a substance that makes peppers feel hot to the palate.
In the following seconds, the participant would be given a slight shock in the foot or the hand, to assess the intensity of the pain. After the application on the right hand of an innocuous ointment, which they believed to be a new analgesic, the volunteers would claim to feel less pain only in that member. In an article published in the Journal of Neuroscience, the Italian group concluded that attention mechanisms may possibly also be involved in the effect generated by the placebo, since the expectation directed towards a given part of the body concentrated the benefits there only.
After helping to understand how our organism reacts when faced with the expectation of a cure, the evidence built up over the last few years that the placebo is more than an innocuous substances is also opening up some polemics: is it that our capacity for self-cure is large, or does the problem lie with the medicines available to the population, which are less effective that would be expected? Several studies show that innocuous substances are capable, in some diseases, of making people improve as much as medicines regarded as effective do, a distinction that seems quite complicated, particularly in the case of medicines used to treat depression.
In a polemical article called Listening Prozac but hearing placebo, published in Prevention and Treatment of June 1998, Irving Kirsch, from the University of Connecticut, and Guy Sapirstein, from Westwood Lodge Hospital, both in the United States, stated that innocuous substances were just as effective as antidepressives in the treatment of depression. It was a blow for medicines of proven effectiveness, which have a market turnover of billions of dollars world-wide. Perhaps for precaution, the editor of the magazine added an observation to the study, advising that the work used a polemical methodology for comparing studies carried out with different methods and criteria for treatment.
In the following year, physician Thomas Weihrauch, a director of the Pharmaceutical Research Center of the pharmaceutical company Bayer, in Germany, looked for signs of the placebo effect in various studies that assessed the action of five medicines produced by Bayer – against chest pains, anxiety, cerebral vascular accident, gastritis, and diabetes.
With the exception of the treatment for diabetes, the supposedly innocuous compounds showed an effectiveness that varied from case to case. Another even more unexpected finding was that, in a similar way to the medicines, the placebo substances caused, in the majority of cases, side effects like dryness in the mouth, tiredness, and mental confusion. In the conclusion of the work, published in Drug Research of 1999, the researcher warns that the doctors should make a very rigorous selection of the people to be treated, before prescribing a medicine without scientifically proven effectiveness.
A doubt then arises. If an inert compound works in part of the cases, can it still be regarded as a placebo? It depends on who answers. For the more skeptical, the placebo has no pharmacological action, period. In 2001, Danish researchers Peter Gotzsche and Asbjorn Hrobjartsson, from the University of Copenhagen and from the Nordic Cochrane Centre in Copenhagen, an international organization that analyzes sets of clinical studies in search of evidence of the effectiveness of treatments, presented in New England Journal of Medicine an article debunking the placebo effect.
After analyzing 130 clinical studies, the researchers found that, generally speaking, giving a placebo was equivalent to providing the sick person with no treatment at all. “We found little evidence that, in general, placebos show powerful clinical action”, Gotzsche and Hrobjartsson concluded. According to them, the benefits brought about by placebos seem very small, and they were observed only in the tests in which the assessment of the improvement was made by the patients themselves – that is to say, the analysis was subjective – or in the therapy of pain.
Not all think like this. There are those that advocate a reassessment of the very definition of placebo. “There is no one substance or one treatment that, for once and for all, can be defined as a placebo”, claims philosopher Zbigniew Szawarski, from Warsaw University, Poland, in one of the articles on the role of placebos in medical research published in the January issue of the Science and Engineering Ethics magazine. According to the philosopher, the reason is that the effectiveness of any chemical compound – innocuous or pharmacologically active – also depends on the characteristics of the medicine (such as color, shape, aroma), on the person who is taking it, on the relationship with the doctor, and even on the circumstances in which it is used.
Nikola Biller-Andorno, from the University of Göttingen, Germany, and a consultant in ethics to the World Health Organization, offers an alternative. “The dichotomy between an ‘active’ substance and a placebo is not a proper one, since placebos can produce an effect and part of the action of ‘active’ substances can derive from the ‘placebo effect’. Accordingly, therapy with a placebo should not be considered as an absence of treatment”, Andorno writes in one of the articles in Science and Engineering Ethics. “Instead of thinking of the employment of one or the other, it may be more appropriate to imagine how the placebo effect can be used to improve the effectiveness of a given therapy”, she proposes.
Neuroscientist Raúl de La Fuente-Fernández, from the University of British Columbia, in Canada, and from the Architect Marcide Hospital, in La Coruña, Spain, agrees with the need for altering the way of looking at the placebo. In his opinion, it is time to rethink the structure of scientific studies. “Recent observations indicate that the moment has arrived to plan appropriate investigations that use placebos”, the researcher says. For him, the placebo effect may be a mechanism that human beings have developed through natural selection.
“In a period when available active treatments did not exist, a capacity for responding to remedies with supposed curative properties could prolong survival”, comments La Fuente, who two years ago revealed now innocuous substances act on the nervous system of people with Parkinson’s disease, which causes the loss of control of movements and progressively kills the cells that produce dopamine.
In his article in January’s Science and EngineeringEthics, in which he comments on the biochemical evidence for the placebo effect, La Fuente launches the idea that this suggestive effect must have been greater, before the present-day medicines arose: “The power of faith healing may have diminished in modern times, as a consequence of the growing influence that scientific method, once established in the medical literature as the only valid tool for arriving at the truth, may have had in the mind of the population at large.”Republish