A Review of the Evidence for Overlap among Unexplained Clinical Conditions

Observant physicians for at least a century have described illnesses seen in clinical practice that share features such as fatigue and pain, disability out of proportion to physical examination findings, inconsistent demonstration of laboratory abnormalities, and an apparent association with stress and psychosocial factors. These clinical features, as well as clinician discomfort with making a diagnosis in the absence of objective abnormalities, have resulted in the creation of disturbing labels for affected patients, such as heart sink patients, hypochondriacs, and amplifiers (1-3). The patients' symptoms and syndromes have been called functional, somatic, medically unexplained, and psychosomatic. The existing literature, although dispersed across many disciplines, suggests that many unexplained clinical conditionsfor example, the chronic fatigue syndrome, fibromyalgia, and temporomandibular disordershare demographic, clinical, and psychosocial features, as well as objective findings. In fact, it has been suggested that the diagnosis assigned to patients with one of these illnesses depends more on chief symptom and clinician specialty than on the actual illness (4, 5). Furthermore, inadequate information is available on the cause, pathophysiology, natural history and prognosis, and medical management of these syndromes. We examine the nature and extent of overlap among selected unexplained clinical conditions, as well as the limitations of previous research, and suggest directions for future research. Methods Literature Retrieval We identified English-language articles through a search of the PubMed database from 1966 to January 2001. Search terms included chronic fatigue syndrome, fibromyalgia, irritable bowel syndrome, multiple chemical sensitivity, temporomandibular disorder, tension headache, interstitial cystitis, and post-concussion syndrome. Variants of these terms were also used; for example, for temporomandibular disorder, additional keywords included the preresearch diagnostic criteria (6) terms craniomandibular disorder, myofascial facial pain disorder, and atypical facial pain. Next, we used the terms for hallmark symptoms associated with the eight disorders (for instance, chronic fatigue for the chronic fatigue syndrome). Finally, the bibliographies of the retrieved reports and those of several additional articles that had been previously identified were examined for other articles. Although some relevant articles may have been overlooked in this process, it yielded almost 600 reports, 53 of which are included here. Reports were most commonly excluded because they were not original research (for example, reviews, commentaries, case studies), the methods or participants were inadequately described, or the comparison groups did not have an unexplained clinical condition. We could not conduct a meta-analysis because of the number of unexplained clinical conditions and possible comparisons, given a small number of studies overall; variations in study design and method (in particular, participant selection and case definitions); and the quality of the investigations. Inclusion Criteria We review two types of studies, both of which must have assessed patients with at least one unexplained clinical condition and report on symptoms, overlap with other unexplained clinical conditions, or physiologic markers. The first type (summarized in Tables 1 and 3) compared at least two groups of patients with different unexplained clinical conditions (for example, the chronic fatigue syndrome vs. fibromyalgia). Because of their comparative design, studies of two groups can more precisely estimate the nature and extent of overlap. Such studies provide clinical information with implications for practicing physicians (elucidating, for example, that fibromyalgia is common in IBS and vice versa); facilitate understanding of the unifying mechanisms, pathophysiology, and psychological underpinnings of these conditions; and allow researchers to better target areas of symptomatic or physiologic overlap. The second type of study (Table 2) assessed patients with one predominant condition for the co-occurrence of other syndromes (for example, frequency of interstitial cystitis in fibromyalgia). Examined together, the findings in Tables 1 and 2 represent the range of point estimates for syndrome overlap. For each study, information was extracted on the participant groups, diagnostic criteria applied, methods, and major findings. Role of the Funding Source The sources of the authors' funding had no role in the design, collection, or interpretation of the data or in the decision to publish the findings. Results Evidence for Overlap The authors of a recent review comparing the case definitions of 12 functional somatic syndromes for which published criteria exist (including the chronic fatigue syndrome, fibromyalgia, IBS, multiple chemical sensitivity, temporomandibular disorder, and tension headache) (5) noted substantial overlap among definitions, even for core symptoms: Eight contained bloating or abdominal distention, or headache; 6 contained fatigue; and 6 contained abdominal pain. It is therefore not surprising that comorbidity has been reported between some of these conditions. Nonetheless, certain syndromes demonstrate considerable overlap despite different case definitions. For example, criteria for a diagnosis of the chronic fatigue syndrome require debilitating fatigue and 4 of 8 specific symptoms, including muscle pain, in the absence of explanatory medical and psychiatric disorders (7). Fibromyalgia, on the other hand, is defined by 3 months of widespread musculoskeletal pain and tenderness to palpation at 11 of 18 specific body sites (8). Thus, similarities in formal criteria are unlikely to account for most of the overlap among syndromes. Overlap of Symptoms and Syndromes Tables 1 and 2 summarize the literature on co-occurrence of unexplained clinical conditions. These investigations are primarily focused on the overlap between the chronic fatigue syndrome, fibromyalgia, IBS, and temporomandibular disorder. While the comorbidity of the chronic fatigue syndrome and fibromyalgia has been relatively well characterized, their relationship to IBS, multiple chemical sensitivity, chronic headache, interstitial cystitis, and other unexplained conditions has only recently been systematically studied (4, 17). Table 1. Overlap of Symptoms and Syndromes among Unexplained Clinical Conditions in Studies Comparing Two or More Patient Groups Table 2. Overlap of Symptoms and Syndromes among Unexplained Clinical Conditions in Studies of a Single Patient Group Comparative investigations in referral clinic populations (Table 1) have reported that up to 70% of patients with fibromyalgia meet the case definition for the chronic fatigue syndrome (4) and, conversely, that 35% to 70% of patients with the chronic fatigue syndrome have fibromyalgia (4, 9). Moreover, 13% to 18% of patients with temporomandibular disorder meet fibromyalgia criteria (15, 23) and 75% of patients with fibromyalgia satisfy the case definition for temporomandibular disorder (15, 70). Other painful conditions, such as tension headache, also commonly co-occur with fibromyalgia (15). Furthermore, 32% to 80% of patients with fibromyalgia (9, 10, 21, 23), 58% to 92% of patients with the chronic fatigue syndrome (9, 23), and 64% of patients with temporomandibular disorder (23) also have IBS. Conversely, 32% to 65% of those with IBS meet the criteria for fibromyalgia (10, 21). Moreover, one study showed that the symptoms reported by patients with fibromyalgia and patients with interstitial cystitis were more similar to each other than to those reported by healthy controls (17). Finally, in 53% to 67% of persons with the chronic fatigue syndrome, illness worsens with exposure to various chemicals, and 55% of persons with fibromyalgia have symptoms consistent with multiple chemical sensitivity (4). In contrast, 30% of patients with multiple chemical sensitivity meet the criteria for the chronic fatigue syndrome (4). Taken together, these direct comparisons of patient groups generally demonstrate high rates of diverse clinical conditions regardless of the condition examined. Table 2 shows studies that assessed patients with a single unexplained clinical condition for the presence of at least one additional syndrome. Despite methodologic differences, these investigations confirm the results of those summarized in Table 1. For example, among patients with fibromyalgia, 21% to 80% have coexisting chronic fatigue syndrome (39, 43, 61), 36% to 60% have IBS (8, 27, 32, 40, 43, 48, 51), 10% to 80% have headaches (27, 32, 43, 57), 75% have mandibular dysfunction (35), and 33% have multiple chemical sensitivity (54). Many patients with the chronic fatigue syndrome also have IBS (63%) (49), multiple chemical sensitivity (41%) (64), and other unexplained illnesses (67). Less overlap with unexplained clinical conditions has been observed in patients with interstitial cystitis (53) or temporomandibular disorder (55). Of interest, these studies suggest that sex may influence the co-occurrence of unexplained clinical conditions. For example, IBS is more common among women (40%) than men (14%) with fibromyalgia (66), and fibromyalgia occurs more often in women (36%) than men (12%) with the chronic fatigue syndrome (12). Objective Findings As shown in Table 3, objective findings consist primarily of physiologic abnormalities on physical examination or laboratory testing. Across studies, tender points are the most common finding on physical examination among unexplained clinical conditions (9, 15, 17, 21, 23, 61, 71, 75). For example, the average number of tender points in patients with temporomandibular disorder or interstitial cystitis was higher than that among controls but lower than among patients with fibromyalgia (15, 17, 23). Most investigations of pathophysiologic similarities have foc