to the editor: In their article on fibromyalgia, Drs. Chakrabarty and Zoorob include and apparently endorse the 1990 American College of Rheumatology diagnostic criteria. However, these criteria leave many important clinical questions unanswered. Are these tender points specific? What if I push the appropriate way on 50 places on the patient's body, including the indicated tender points, and 12 of the 18 tender points are positive and 25 of the other 32 spots are positive? Does this patient have fibromyalgia, or just nonspecific, widespread tenderness? Do all other areas have to be negative for the tender point reactivity to be valid?

I believe that fibromyalgia is not a rheumatologic disease, but a chronic pain syndrome. The presumed legitimacy of the tender point concept seems to be the only factor that keeps fibromyalgia from being classified as a chronic pain syndrome. I would be interested to know how the authors feel about the usefulness of these arbitrary criteria in clinical practice.

to the editor: We read with interest Drs. Chakrabarty and Zoorob's article on fibromyalgia, but we were somewhat disappointed in the discussion of therapy and Table 2.¹ This table classifies medications as having strong, moderate, weak, or no evidence of effectiveness. However, there is little evidence to substantiate this classification. For example, amitriptyline and cyclobenzaprine (Flexeril) are listed as having strong evidence for effectiveness. Although short-term studies indicate benefit in up to one third of patients with fibromyalgia,² long-term, prospective studies have not shown any prolonged beneficial effect.³ In one study, initial improvement noted at six to 12 weeks was no longer present by 26 weeks.³

The moderate effectiveness claim for some medications also is questionable. The first medication listed as being of moderate effectiveness is duloxetine (Cymbalta). In a double-blind trial comparing duloxetine with placebo, patients taking duloxetine measured significantly better on the Fibromyalgia Impact Questionnaire (FIQ) total score, but there was no difference on the FIQ subscales of pain, fatigue, or morning fatigue.⁴ As pain is the major clinical feature of fibromyalgia, it is not clear how the authors defined effectiveness.

Current therapeutic options for fibromyalgia are many, but disappointing. One five-year prospective survey of 214 patients with fibromyalgia from 114 rheumatology practices showed that a total of 74 different medications were used for treatment, suggesting that no specific drug or class of drugs is especially useful for patients with fibromyalgia as a whole.⁵

to the editor: I would like to question a recommendation made by the authors of the article “Fibromyalgia”. The authors state: “The combination of amitriptyline and fluoxetine [Prozac] has been determined to be more effective than either agent alone, and physicians should consider this unless otherwise contraindicated.” The recommendation is based upon a double-blind, crossover study¹ of fluoxetine (20 mg), amitriptyline, and the combination of the two. Only 19 patients were included in this study; therefore, I do not believe a study with such a small population should guide treatment decisions.

Amitriptyline is metabolized through the cytochrome P450 2D6 pathway. In the United States, one person in 12 lacks this isoenzyme because of genetic mutations.² These persons are at high risk for adverse events caused by increased concentrations of the unmetabolized parent compound. Administration of a CYP450 2D6-inhibiting drug such as fluoxetine³ creates a similar high-risk situation in the presence of co-administered amitriptyline.

Although the cited study used low doses of both medications, the authors mentioned in the same paragraph another study, which found benefit from 80 mg of fluoxetine on the symptoms of fibromyalgia in women.⁴ Readers may mistakenly assume that the authors recommend high-dose fluoxetine in combination with amitriptyline.

Given the potential for causing unpredictably elevated levels of amitriptyline and scant evidence of benefit, it is unclear why a physician would risk using fluoxetine and amitriptyline together instead of other therapies.

to the editor: In the excellent review of fibromyalgia, the authors listed myofascial pain syndrome, chronic fatigue syndrome, and hypothyroidism as conditions to consider in the differential diagnosis of this common musculoskeletal disorder. Family physicians also should consider vitamin D deficiency when evaluating patients with diffuse truncal pain.^1–3

Vitamin D deficiency is common in North America because of inadequate sun exposure (especially for those who live above 39 degrees of latitude) and intake. It has been documented in postmenopausal women with osteoporosis, older adults, white preadolescents, Hispanic and black adolescents, exclusively breastfed infants, and pregnant women—in other words, the entire gamut of a family physician's practice.^1,3

Vitamin D deficiency can be easily diagnosed by obtaining a serum 25-hydroxyvitamin D level (the immediate metabolic precursor of the active 1,25-dihydroxyvitamin D form). Levels less than 30 ng per mL (75 nmol per L) are insufficient. In adults, treatment is 50,000 IU vitamin D weekly for eight weeks. After treatment, another 25-hydroxyvitamin D level can be obtained, and the patient retreated if necessary.³ However, symptoms may take weeks to months to remit.

in reply: These letters are a testament to the fact that fibromyalgia remains, to a large extent, a medical enigma. In our article, we aimed to analyze the scientific evidence on this topic and provide a comprehensive review for physicians.

In Dr. Tock's letter, he questions the legitimacy of using the American College of Rheumatology (ACR) criteria in the diagnosis of fibromyalgia. Since the establishment of these diagnostic criteria in 1990, they have remained a cornerstone in the diagnosis of this disease. Although the pathogenesis of fibromyalgia is far from clear and there is considerable overlap between patients with fibromyalgia and patients with other unexplained syndromes, the ACR criteria have been used consistently in the diagnosis of fibromyalgia.^1,2 The London Fibromyalgia Epidemiology Study demonstrates that in the general population, adults meeting the ACR criteria of fibromyalgia appear to have distinct features compared with those who have chronic pain but do not meet the criteria.³ A diagnosis of fibromyalgia can be made if the history alone is highly suggestive of the diagnosis, even if the ACR's tender point criteria are not met.⁴

The letter from Dr. Abeles and colleagues concerns the relative effectiveness of the treatment of fibromyalgia. Our intention in Table 2 was to categorize the quality of the evidence as presented by Dr. Goldenberg and associates. These categories relate to the types of studies conducted more than to the effectiveness of the treatments. There is evidence from a large number of randomized clinical trials that tricyclic antidepressants are effective in improving fatigue, pain, sleep, and overall well-being in patients with fibromyalgia. A meta-analysis that extracted data from 13 randomized, placebo-controlled trials found that, amitriptyline (studied in eight of these trials) was effective in treating many of the symptoms of fibromyalgia.⁵ Cyclobenzaprine (Flexeril) has been demonstrated to be effective in five randomized placebo-controlled trials, and it was found that patients receiving cyclobenzaprine were three times as likely to report overall improvement.⁶ Although duloxetine (Cymbalta) is a relatively new entrant into the treatment field of fibromyalgia, this serotonin/norepinephrine inhibitor has been evaluated in two multicenter randomized, controlled trials and found to be a safe and efficacious treatment for both the pain and functional impairment associated with fibromyalgia.⁷ Although there are no long-term studies showing prolonged beneficial effects, these medications have shown evidence of efficacy in numerous clinical trials.

In Dr. Adams' letter, he expresses concern with our recommendation for the combination of amitriptyline and f luoxetine (Prozac) therapy, which is based on a study of only 19 patients. We agree that small studies need to be interpreted with caution; however, this is not the sole basis of the recommendations.⁸ As mentioned above, readers should understand that this article does not constitute a practice guideline. Rather, it is a road map to a rapidly evolving set of literature. Administration of cytochrome (CYP) P450 2D6-inhibiting drugs increases the risk of adverse reactions, especially in patients taking certain medications; physicians prescribing any medication should be aware of the dosage, indications, interactions, and adverse effects of the medications used and should weigh the risks and benefits before prescribing the medication. A Cochrane Review on antidepressant and centrally active agents for the treatment of fibromyalgia is currently in progress.⁹

We thank Dr. Richardson for his excellent point about considering Vitamin D deficiency in the differential diagnosis of patients with fibromyalgia.