This might appear improbable in the context of a developing region that is by definition economically depressed, considering the high investment required (the actual cost of a hybrid PET/CT system is about US$2 million, $800,000 for a dedicated PET unit and more than $1.5 million for a 11 MeV cyclotron) even not taking into account the necessary training of chemists, engineers, physicists, physicians and technologists. However, there are several reasons that can be identified to explain our current reality, that many have defined as 'PET-mania' not without certain connotation of irrationality and despectiveness. The most obvious, and certainly not the most irrational one, appears to depend on the clinical usefulness of the method itself. In fact, many published studies demonstrate a favourable position of PET in the cost-effectiveness equation, particularly concerning oncologic applications that constitute currently 90% of clinical indications. In the United States, the reimbursement of PET studies by insurance companies in certain clinical situations including lung, breast, esophageal and colon cancers, lymphoma, melanoma, refractary epilepsy and myocardial viability is a good indicator of the effectiveness of the procedure and also reflects the impact that evolving detector and software technology has had on the overall clinical performance of the method. Furthermore, there is a continuous incorporation of new indications to the list as a result of incremental clinical evidence about the benefit that PET brings to the clinical management of patients. This fact has meant an economic viability to the growth of PET centers in the United States and other places of the world with more of a selling pressure by industry that has extended also into Latin America.

On the other hand, the practice of 'conventional' nuclear medicine had recently reached a 'plateau' level, leading to a period that we could define as 'creative abulia' with a few procedures namely cardiologic and bone scans representing nearly 80% of all studies performed in an average clinical nuclear medicine facility. This has been happening in spite of the large amount of functional information potentially supplied by other type of procedures such as renal studies, brain, gastrointestinal, etc. However, the application of these methods has gained only limited popularity or even their utilization has been steadily decreasing. In consequence, the introduction of a new technology (which is really a new version of a technology that is already more than three decades old) is waking up the 'nuclear siesta' that was involving great part of our scientific community with a few meritory exceptions. This has pushed away the ghost of extinction of our specialty, like many were ominously prognosticating, and at the same time has opened a new and exciting world of possibilities in patient care and research. Nuclear medicine is dressing itself with new and shiny 'molecular clothing' that all of us are anxious to experiment with, and nobody wants to be left behind this fast moving train.

Finally, there is a strong ethical reason for introducing PET. It is not only our duty but our obligation to offer our patients and referring physicians everything that our specialty can possess as a tool for the recovery of health, and if PET constitutes one such tool, our efforts must be directed to make this service available to the general population.

Now, in this new PET scenario it is possible to detect some aspects that would deserve a closer analysis. In the first place, we must clearly state that PET is a nuclear medicine technique as far as it constitutes a diagnostic imaging procedure that uses open sources of radiation. The coupling of a computerized axial tomography (CT) system no matter how advanced this may be, should be considered just an auxiliary device needed to provide anatomical information in order to precisely localize an abnormal functional finding, or for the assessment of a functional change in a previously observed structural abnormality. In fact, this is the objective of the fusion of functional and anatomical images acquired simultaneously, keeping in mind that this can also be achieved - although in a more imperfect manner - using separately acquired studies. By all means we must reject the concept that ¹⁸F-FDG or other positron emitting radiopharmaceuticals represent merely sophisticated 'contrast agents' that are only complementary of a high-quality CT study. Indeed, this might be heard from some radiologists in an attempt to justify their dominion in this area, which is not infrequent in many countries although still uncommon in Latin America. If we wish to keep control of PET technology we should fight the battle on all fronts, bringing up regulatory issues and improving legislation trying to fill in the gaps that may exist in this matter. This doesn't mean that we should advise the radiologist to jump off the ship, but that rather we must define who the captain is, and based on this definition we should work together in order to get the maximal benefit from the method. Beyond the specific, peculiar characteristics of different countries we believe that we should walk down the path of cooperation accepting the challenge of cross-education, with the nuclear physician being trained in structural methods and the radiologist incorporating functional imaging concepts for better benefit of both and in particular of the patient. After all, the 'fusion' of functional and anatomical images in our minds is what we have always been doing in nuclear medicine, sometimes with the help of the radiologist, and all of us are usually aware about the convenience of having available additional imaging information at the time of interpreting a radionuclide study. For this reason, dedicated PET alone should not be regarded as obsolete technology. Indeed, it can still have diagnostic impact in most clinical scenarios at a significant lower cost compared to PET/CT. Efforts should be made, however, towards the acquisition of a PET/CT machine whenever possible, due to the higher accuracy that it provides in terms of lesion localization.

Resources for health in Latin America are very limited specially in the public area where they should be distributed according to the potential social impact to be achieved by each investment. Clinical indications of PET are probably the same in any population, however the availability of this technology may have a very different meaning in different countries, depending on the incidence of various pathologies. For instance, the influence on public health that PET may represent is not the same for relatively more developed countries with longer life expectancy and higher prevalence of cancer as compared to others whose main health problems are infant mortality, malnutrition and infectious diseases. Both realities coexist in Latin America often in a same country, so allocation of resources should be done with maximum efficiency. We should remember that diagnostic sensitivity, specificity and accuracy of a method depend in part on the prevalence of the disease under investigation and for this reason, extrapolation of data from other populations is not recommended. This aspect is to be taken into account in order to make the necessary adjustments and reconsider the clinical utility of the technique in our region under a different light. Furthermore, the spectrum of clinical applications should not spread beyond those scientifically accepted, otherwise there is a risk of credibility declination which may be eventually difficult to revert. Many of us have witnessed the introduction of new technologies that are initially viewed as universal solutions, for later falling down to a 'near zero' level and finally reaching a logical equilibrium only supported by the scientific evidence accumulated in its favor. Hence, let us take a firm but cautious course of action climbing the steps one by one and promoting the utility of the method based on solid and robust information.

PET findings are relatively non-specific, specially using ¹⁸F-FDG since this is a metabolic tracer that is taken up in tumors as well as in infectious, inflammatory and even physiologic processes. Although this is not new in nuclear medicine (bone scintigraphy is a clear example of a non-specific yet very useful technique), it should be always kept in mind that a correct interpretation of a functional study can only be achieved with a knowledge of the clinical context. It is not unusual to see the utilization of PET as a screening method for the detection of occult neoplasias in asymptomatic subjects. This indication has no scientific support although we all know that its practice allows many projects to be economically viable, thus generating resources that in some way subsidize the correct utilization of the technology. However, the already mentioned non-specificity can lead to severe interpretative errors in this population and to the indication of other invasive diagnostic procedures with potential iatrogenic results. Eventually, the use of PET as a screening tool could be considered in high risk populations, although the point needs further investigation and should not be currently promoted for routine clinical practice until sufficient evidence becomes available.

Education of human resources is crucial. Installation of a PET/cyclotron unit is something big enough to have a significant impact among the local scientific community of a developing country. Beyond the specific application in health care, this new technology represents an opportunity for on-site high level training of basic scientists, physicians, and technologists. Of course, this has to be preceeded by the formal education and training abroad of some individuals being able to make the unit start working. For this stage of the PET process - training of human resources - the co-participation of public and private sectors is important. Local and foreign universities, commercial companies and some international organizations should play a significant and synergic role in this regard, under the catalytic influence of national atomic energy commissions and nuclear medicine prefessional societies. The International Atomic Energy Agency (IAEA) has been supporting PET projects worldwide and is starting to have also some participation in Latin America. We look forward to an increasing involvement of the IAEA to facilitate the dissemination of PET technology in our region, as it has been successfully doing with other nuclear medicine modalities. The Society of Nuclear Medicine (SNM) and the European Association of Nuclear Medicine (EANM) could also provide some educational models based on their own experience, which could well be adapted to our specific reality.

Quality control is another issue to think about. Technologic development has made the instrumentation progressively more complex, although we should recognize that the evaluation of performance parameters is much easier now with the availability of automatic procedures and electronic corrective measures. However, none of these are really useful in the absence of a comprehensive quality assurance (QA) program to be followed in a systematic and rigurous manner. Again, the need for a QA program is not new in nuclear medicine. On the contrary, it is something we always should have developed and kept running in our departments although this has rarely been completely accomplished in the region. We have now the opportunity to revalue this practice extending the QA concept to every step of the procedure (not just instrumentation) for which we need to convince ourselves that this will lead to a more confident interpretation of results and to an increased credibility of the clinical information that we provide.

Last but not least, PET is not only ¹⁸F-FDG. There are several other positron-emitting tracers labelled with ¹¹C, ¹³N and ¹⁵O which are of great scientific interest and promising clinical utility. Clinical applications other than oncologic should also be encouraged and explored, particularly in the areas of cardiology, neurology and infection/inflammation. Thus, we should think in producing all the range of positron isotopes when we are to design a cyclotron project and, considering the very short half-life of most of these isotopes, the proximity of the clinical site to the production site could be critical. Also, the use of ⁶⁸Ga which is a generator-produced positron-emitting radioisotope as an alternative to cyclotron-produced isotopes should not be overlooked in developing areas, since it may potentially represent an inexpensive, transient solution while a cyclotron is still not available.

In summary, we welcome PET to Latin America. It undoubtedly will contribute to the better care of our patients and to our desired growth of nuclear medicine as a specialty. However, these goals will have better chance of being achieved if we take into account some essentials, such as:

1. Defending PET as a nuclear medicine procedure by definition.
2. Establishing a precise and significant role for radiology in PET/CT, with mutual cooperation and cross-education of physicians and technologists being desirable.
3. Evaluating the potential impact of PET in national health systems without simply extrapolating international data.
4. Being aware of the limitations of the method and clearly stating the accepted clinical indications.
5. Educating and training of human resources with the co-operation of public and private sectors and international organizations.
6. Developing total quality assurance programs under which PET units should operate.
7. Trying to obtain the maximum yield of the technique through the utilization of multiple tracers and expanding the clinical applications of PET.

Most probably, many of the readers would like to add some points to the list above. If so, that would do nothing but contributing to the purpose of this editorial.

Dr. Fernando Mut

Jefe del Servicio de Medicina Nuclear, Asociación Española

Montevideo, Uruguay

E-mail: fmut@adinet.com.uy