A systematic search was conducted using the PubMed, Cochrane and ScienceDirect databases, as well as the following CPG development groups: Clinical Evidence, Scottish Intercollegiate Guidelines Network (SIGN), New Zealand Guidelines Group (NZGG), National Institute for Health and Clinical Excellence (NICE), Agency for Healthcare Research and Quality (AHRQ) and GuíaSalud. The following terms were used: “fibromyalgia” (MeSH), “guideline” (MeSH) and “clinical practice guideline”. In addition, we carried out hand searches for the most relevant articles and Google Scholar was checked. The searches were limited to literature published from January 2003 to December 2014.

Those CPG that met the following requirements: (a) be focused on the treatment of FM; (b) be based on scientific evidence (clinical trials, systematic reviews, meta-analyses); (c) include the level of evidence (LOE) and strength of recommendation (SOR), in addition to clear references; (d) involve a systematic evidence search; (e) be performed in an adult population; and (f) be written in English or Spanish. The latest available edition of each guideline was evaluated. Any complementary information provided by the guidelines was taken into account. Protocols designed by hospitals and/or departments were excluded because of their specificity and the difficulty in adapting them to other settings.

Once the CPG were identified, their recommendations were analyzed and grouped according to type of treatment: general, pharmacological, physiotherapeutic, psychological, or alternative therapy. Only explicit recommendations were considered: an affirmation or easily identified statement summarizing the evidence regarding the therapies that should be used, those that can be used and others for which advice concerning their use is provided.

The LOE and SOR of each was identified and translated to our own standardized evidence-grading system (Table 1), developed by harmonizing the proposals of the U.S. Preventive Services Task Force7 and the SIGN8 and GRADE9 systems. This classification makes it possible to compare guidelines in terms of the LOE and SOR of the recommendations.

Finally, the most relevant recommendations for the treatment of FM were identified on the basis of the evidence and of the variability among the guidelines, a complete recommendation has been formulated in line with the proposals in the most recent guidelines, and 2 independent collaborators reviewed the final text.

The CPG included evaluate the available evidence on general, pharmacological, physiotherapeutic, psychological, and alternative therapies. Only 7 therapies appear in all the guidelines: aerobic exercise, cognitive-behavioral therapy, amitriptyline, selective serotonin reuptake inhibitors (SSRI), serotonin and noradrenaline reuptake inhibitors (SNRI), anticonvulsive drugs and mild opioids. These recommendations show discrepancies in the LOE and SOR, except in the case of amitriptyline; however, in 1 guideline,18 the latter is grouped with the antidepressants. The SSRI, SNRI and anticonvulsive drugs are also analyzed as groups (without specifying any given agent) in 2 guidelines.14,18 Exercise and cognitive-behavioral therapy remain constant in all the guidelines issued after 2008, with the highest LOE.

We identified 22 types of therapy that had a high LOE (A1, A2) in at least 1 of the guidelines. The appearance of recommendations of high LOE varies among the different guidelines. The guideline of the European League Against Rheumatism (EULAR)14 is that which includes the fewest of these therapies, only 8 of the 22 recommendations, and the update of the German S3 guidelines (AWMF),3 with 19, is that which includes the highest number. The analysis of the types of therapy, together with their LOE and SOR (Table 2), enables the identification of the most relevant activities in the treatment of FM.

Recommendations with low LOE and SOR that appear in a single guideline were not included in the tables. For example, dietary supplements were evaluated in a meta-analysis,19 but were not found to be effective (LOE: C1, SOR: C). There are therapies that have been studied in randomized clinical trials that showed no significant differences or inconsistent results (LOE: C2) and, thus, their use cannot be recommended (SOR: C): craniosacral therapy, lymphatic drainage, hypnotic drugs, interferons, sodium oxybate, neuroleptic drugs, serotonin receptor antagonists (5-HT3), cannabinoids, thermotherapy and transcranial magnetic stimulation. In other therapies, although the LOE is the same (C2), the evidence is of poor quality or insufficient and, thus, does not support arguments either for or against their use (SOR: I): hyperbaric therapy, whole-body hyperthermia, galvanic baths, psychodynamic therapy, psychoanalysis, hormone therapy, local anesthetics, diet, dance therapy, melatonin, music therapy, cryogenic chamber, ultrasounds, laser, mindfulness, monoamine oxidase inhibitors, muscle relaxants, antiviral drugs, dopaminergic agonists, anxiolytics and potent opioids. In some cases, there is no identified evidence or the available literature is not useful (LOE: D, SOR: I) for intervention in quadrants: flupirtine, acetylsalicylic acid, acetaminophen and metamizol.

Two guidelines published in 201116,17 and another 2 issued in 20123,18 differ widely in the number, type of therapy and how their recommendations are expressed, even though they were written in the same year. The differences between guidelines in terms of the recommendations of high LOE are also important. Two of the guidelines,17,18 one from 2011 and the other from 2012, present only 50% of the 22 recommendations that were classified in at least 1 other guideline as having a high LOE, whereas a guideline published earlier15 presents 68.18% of the recommendations. These should appear in all the guidelines, or at least have continuity over time once they are identified, as the continuity in the information is essential for the updating and continuing education of the professionals.

These data cast doubt on the method of creating the guidelines. The CPG are at the peak of the information pyramid,22 they are the mode of transmitting the results of research to professionals and reducing the variability in practice.23 Thus, it is necessary to draft them in accordance with the highest quality and thoroughness possible. Therefore, when it comes to developing a clinical practice guideline, it is advisable that certain recommendations be followed, like those issued by the Institute of Medicine.23 Other models for creating, developing and implementing guidelines are the elements proposed in the Appraisal of Guidelines Research & Evaluation (AGRE)24 and those of the Guide Line Implementation Appraisal (GLIA).25

To cover every aspect of FM, it is necessary to approach treatment with an interdisciplinary strategy, which has been shown to be effective,26 but some approaches must be given priority over others. The evidence shows that nonpharmacological treatment is consolidated as the first line of action in FM. Some of these therapies have been shown not only to control the symptoms, but also improve the quality of life of the patients.27 In general, they are associated with little risk and can be maintained indefinitely. For this reason, we consider that they should be given priority in all the health care settings and at every level. The establishment of exercise as the basis of the treatment should be complemented by the necessary measures, both pharmacological and nonpharmacological, for individualized patient management.

The current view of chronic diseases, which includes not only biomedical, but behavioral aspects as well, has led to a change in the molds, whereby treatment and evaluation are approached from a more biopsychosocial point of view.28 Therefore, the boom in the use of cognitive-behavioral therapy in chronic diseases is not surprising. Despite the fact that the biopsychosocial model has been established in the health care and scientific settings for some time now,29 it seems to have had a lesser impact on the organization and administration of the health systems, just like the majority of the nonpharmacological treatments that have been dealt with in this review. This, together with the lack of organization and of clear protocols, leaves patients defenseless in their confrontation with the disease.30 What FM patients need is a larger number of specialized physiotherapists and psychologists incorporated into the present health care network, as well as an improvement in their integration with the other professionals to achieve a workable interdisciplinary structure.

When we analyze drug therapy, we find that the studies evaluated by the CPG chiefly report short-term results; however, occasionally, even over these short periods of time, important secondary effects develop. This indicates that the use of drug therapy should be reserved for specific situations, to respond to uncontrolled symptoms or to intense, irruptive pain. According to the Association of the Scientific Medical Societies in Germany, even amitriptyline, the medication most widely used and studied in this disease, has important secondary effects, and the methodological quality of the evidence that supports it is relatively low.3 Other drugs, such as antiepileptic agents and antidepressants, also share these risks, and the results they offer are, at best, moderate.

Serotonin and noradrenaline reuptake inhibitors, like milnacipran, provoke a clear controversy among the guidelines. This circumstance may be influenced by the fact that the United States Food and Drug Administration (FDA) has approved this compound, whereas, in Europe, approval of its use specifically for FM has not been granted. Nevertheless, a recent meta-analysis shows that, although its benefits are similar to those of duloxetine, it is surpassed by the latter even in patients with major depression (in whom its utility has not been demonstrated).31 The opinion of the authors in every case seems to be that duloxetine should be recommended as the drug of choice as compared with milnacipran. On the other hand, SNRI could be useful to relieve pain–if fluoxetine is used–and other symptoms, such as fatigue, depression, anxiety or the overall feeling of being unwell–using fluoxetine or paroxetine, but the Spanish guidelines prefer to avoid recommending them because of the low level and controversial evidence,16,17 whereas the German guideline would recommend them, without distinguishing one type from another, in a very specific patient population and for a limited time.3

With regard to muscle relaxants, the studies are of low quality, and the results show a limited efficiency. The dilemma with using medication of this type is the possibility of undermining the capacity of the patients to perform physical activity.

The drawback in treatment with opioids is that current studies on their use are, again, limited to covering a short period of time.32 The increase in the mortality and development of important secondary effects due to different factors33 is presently creating a real public health problem in the United States,34 where their use in nonmalignant chronic disease has proliferated. In view of current evidence, the benefit of opioids in FM is not great enough for patients to accept the collateral damage that they cause. There are other methods, with fewer secondary effects and the same or greater benefits, that can be useful, and affordable, in the treatment of FM.

Given the results reported in this review, it appears to be necessary to standardize the method for developing CPG. Incongruities, like those observed between the number and the LOE of the recommendations in the different guidelines are not acceptable. It is not necessary to resolve the same problem several times; if there are CPG that were developed previously, they can be adapted to our setting, supplemented or updated. In any case, it is necessary to take into account the efforts made by other authors to be efficient and congruent in the information conveyed to the health care community. It is inadmissible that in the information era and with the tools to standardize existing methodologies, there be so much difference in the contents in the guidelines analyzed in this study, and even over the years.

The recommendations made in this review are limited by the quality and thoroughness of the guidelines analyzed. There are limitations related to the information reported by the CPG; in some, it was taken into account whether the drugs had been approved and marketed in the country from which the guideline was issued when it came to establishing the LOE and SOR.3,17 In short, our recommendations summarize the key aspects of the guidelines included in this systematic review. They are not intended to supplant the clinical judgement of the professionals, but to serve as an aid in the principal aspects of managing this disease.

It may be interesting to know that one of the guidelines touches on the diagnosis and treatment of childhood FM syndrome,3 and concludes that there are no valid standardized criteria for diagnosing FM in children. The therapeutic processes presented are based on expert consensus due to the lack of solid evidence.

In conclusion, the variability among the existing CPG, both in the number and quality of the recommendations, continues to demonstrate the need for studies, like the present article, that review and summarize the topic. Despite these differences, there are certain effective recommendations for the management of FM that have been summarized in this text, facilitating access to the best practices. Therefore, the final step necessary for the proper management of FM is the commitment of the health care community to adhere to and implement these recommendations.

The authors declare that no experiments were performed on humans or animals for this study.

The authors declare that no patient data appear in this article.

The authors declare that no patient data appear in this article.