|Home | About | Journals | Submit | Contact Us | Français|
Unexplained clinical conditions share common features, such as pain, fatigue, disability out of proportion to physical examination findings, inconsistent laboratory abnormalities, and an association with stress and psychosocial factors. In this literature review, we examine the extent of the overlap among urological and non-urological unexplained clinical conditions characterized by pain, note the limitations of previous research, and suggest several possible explanatory models.
Using hallmark symptoms and syndromes as search terms, a search of 12 databases identified 1,037 full-length published articles in 8 languages from 1966 to April, 2008. The search focused on the overlap of chronic pelvic pain, interstitial cystitis, painful bladder syndrome, chronic prostatitis/chronic pelvic pain syndrome, or vulvodynia, with fibromyalgia, chronic fatigue syndrome, temporomandibular disorder, or irritable bowel syndrome. Information on authorship, type of case and control groups, eligibility criteria, case definitions, study methods, and major findings were abstracted.
The literature suggests considerable comorbidity between urological and non-urological unexplained clinical conditions. The most robust evidence for overlap was for irritable bowel syndrome and urological unexplained syndromes, with some estimates of up to 79% comorbidity between chronic pelvic pain and symptoms of irritable bowel syndrome. However, most studies were limited by methodological problems, such as varying case definitions and selection of control subjects.
Overlap between urological and selected non-urological unexplained clinical conditions is substantial. Future research should focus on using standardized definitions and rigorously designed and well-controlled studies to further assess comorbidity, clarify the magnitude of the association, and examine common pathophysiological mechanisms.
Functional somatic syndromes, medically unexplained symptoms, somatoform disorders, and unexplained clinical conditions are common names given to conditions characterized by a lack of clear physical or biological etiology or an inconsistent demonstration of laboratory abnormalities.1 These conditions are further characterized by symptoms such as pain, fatigue, sleep disturbances, and disability. The literature suggests that many of these conditions also share demographic characteristics, clinical course, and psychosocial profiles.1 The diagnosis given to a patient suffering from one of these conditions often depends on the hallmark symptom and the expertise of the treating clinician rather than on the illness itself.1 The frequency of co-diagnosis and the common features shared by unexplained clinical conditions have become topics of growing interest over the past few years.
Physicians have been challenged to diagnose and appropriately treat patients with such unexplained clinical conditions as fibromyalgia (FM), chronic fatigue syndrome (CFS), temporomandibular disorder (TMD), irritable bowel syndrome (IBS), chronic pelvic pain (CPP), interstitial cystitis (IC), painful bladder syndrome (PBS), chronic prostatitis/chronic pelvic pain syndrome (CP/CPPS), and vulvodynia. A recent review found that patients with IC often suffered from other unexplained conditions and symptoms.2 Although a substantial body of literature has examined the comorbidity of non-urological unexplained clinical conditions such as FM, CFS, and IBS, 3 the extent of the overlap between urological and non-urological unexplained clinical conditions, and the potential mechanisms for their co-occurrence, remain largely unknown.
The recent emphasis by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) on a multidisciplinary approach to the study of urological chronic pelvic pain syndromes, further underscores the need to examine and understand the overlap between urological and non-urological unexplained clinical conditions (http://www2.niddk.nih.gov/Research/ScientificAreas/Urology/MAPP). We have, therefore, conducted an exhaustive review of the literature on comorbidity between the most common urological pelvic pain syndromes (i.e., CPP, IC, PBS, CP/CPPS, and vulvodynia) and the non-urological conditions of FM, CFS, TMD, and IBS. Our review focused on overlap with these non-urological conditions for several reasons. First, our goal was to examine overlap among conditions with a significant pain component which ruled out overlap with psychiatric conditions such as panic disorder that also have been shown to co-occur with urological pelvic pain syndromes. Second, frequently comorbid pain conditions with known physical or biological etiology were not included since they would not satisfy the definition of unexplained clinical conditions. Third, these 4 non-urological conditions were the focus of most data-based articles in the existing literature and were of particular interest to NIDDK. Finally, these non-urological conditions differed in the location of pain, which could influence the extent of overlap between urological and non-urological unexplained clinical conditions.
Articles were identified by a search of the following databases: PubMed (1966-April 2008), CINAHL (1981-April 2008), CochraneCollaboration Reviews (1993-April 2008), Current Contents (2000-April 2008), EBCO Academic Search Premier (1975-April 2008), EMBASE (1974-April 2008), ISI Web of Science (1980-April 2008), PsychInfo(1967-April 2008), Science Citation Indexes (1996-April 2008), and Scirus, Scopus, and Google Scholar (April 2008).
Our initial search included the following search terms for the 5 urological unexplained clinical conditions: chronic pelvic pain, interstitial cystitis, painful bladder syndrome, chronic prostatitis, and vulvodynia, in combination with 4 non-urological unexplained clinical conditions: chronic fatigue syndrome, fibromyalgia, temporomandibular disorder, and irritable bowel syndrome. The Medical Subject Headings were combined with truncated keywords that described the symptoms of the various conditions. All possible combinations of the 2 sets of search terms were examined.
Next, we looked for all search terms individually and in combination with the words overlapping, overlap, comorbidity, and comorbid. In addition, we combined the terms for the hallmark symptoms associated with each non-urological condition with the hallmark symptoms associated with each urological condition (e.g., chronic fatigue and painful bladder). Bibliographies of relevant articles were examined to identify any other reports that may have been missed. Finally, members of the NIDDK’s Working Group on Urological Chronic Pelvic Pain, a group of experts in the fields of urology, rheumatology, epidemiology, and internal medicine (see Acknowledgments section for full list of Working Group members), were contacted to review the list of relevant articles and to locate other pertinent peer-reviewed publications. Together, these search strategies yielded 1,037 unique publications.
Articles included in this review examined the comorbidity or overlap of at least 1 urological condition or hallmark symptom with at least 1 non-urological condition or hallmark symptom. We focused on articles in English, Hebrew, German, French, Spanish, Korean, Japanese, and Russian, because they could be reviewed by the authors or members of the Working Group. No limits were placed on participant age, but all relevant articles pertained to adults, most of whom were women.
We reviewed 2 types of studies: comparison and single-condition. The first type compared two or more groups diagnosed with different unexplained clinical conditions (e.g., IC versus FM patients) to each other, or compared one group diagnosed with an urological unexplained clinical condition to a healthy control group (e.g., CP patients versus healthy volunteers). The single-condition studies examined participants with a specific unexplained clinical condition or symptom for the co-occurrence of other conditions (e.g., frequency of vulvar pain in IBS or frequency of IBS in vulvodynia). Comparison studies can more precisely estimate the extent of comorbidity or overlap than can single-condition studies, and they can highlight potential similarities in pathophysiology across conditions.
Studies that were initially excluded because they did not meet the overlap criterion were reviewed again for possible incidental findings on the overlap of urological and non-urological conditions that may have been reported as part of the data analysis or in the context of other study findings. The initially-excluded studies that were deemed relevant upon further review are included in the tables, as indicated.
Most (90%) of articles that we identified were excluded because they did not examine the overlap between urological and non-urological conditions. Other articles were excluded for the following reasons: the main patient group did not have an unexplained clinical condition (e.g., migraine or panic); the article focused on an unexplained clinical condition or symptom not considered here, or presented data on broad symptom categories that included conditions other than the ones considered here (e.g., functional digestive disorders); despite mentioning overlap, the article lacked relevant data; or the main topic was not pertinent to our review (e.g., abuse or pharmacological treatment approaches). Articles that did not present primary data, such as reviews, were not included in our analysis or tables even if they supported our findings and conclusions.
Based on the aforementioned inclusion and exclusion criteria, 41 relevant publications were identified and are reported here.1 For each study, we extracted information on study characteristics (e.g., name of first author and publication year), type of participant group (e.g., patients or community participants), group sample sizes, diagnostic criteria, methods, and major findings. The initial review findings were examined by a second reviewer who independently extracted the above information from the relevant publications. Members of the Working Group reviewed drafts of the findings for completeness of the literature review and contributed to the synthesis and interpretation of the findings. The final findings of the review were endorsed by, and represent the views of, the authors and the Working Group.
We found 25 publications on comparison studies that provided data on the comorbidity of urological and non-urological unexplained clinical conditions or symptoms (Table 1). Although some of the studies were published in the early 1990s, most (68%) were published since 2005. Of these 25 studies, 19 reported on patient populations and 6 reported on non-clinical samples derived from population surveys. The studies used diverse diagnostic approaches, including physician diagnoses, symptom-based definitions, and more rigorous, well-established criteria such as the NIDDK criteria for IC diagnosis or the American College of Rheumatology criteria for fibromyalgia. Study methods included self-report questionnaires on medical history, standardized surveys and interviews, physician and tender point examinations, and computerized reviews of International Classification of Diseases-9 codes.
Sixteen studies reported data on the overlap between urological unexplained clinical conditions and musculoskeletal conditions or symptoms such as FM or chronic widespread pain, joint pain, generalized pain, or myalgia.4–19 However, only 4 studies were specifically designed to examine the comorbidity of FM or chronic widespread pain with one of the urological conditions.5,8,14,15 About 9% to 12% of IC patients also experienced FM,5,14 and 23% to 27% of FM patients had symptoms consistent with IC.5 One of the earliest studies found that 68% of men and 66% of women with FM also reported pelvic pain symptoms.4 A twin study that controlled for genetic and environmental confounding found that approximately 23% of twins with chronic widespread pain and 11% of their pain-free co-twins reported urinary tract problems.9 Although the overlap between CP/CPPS and FM has not been directly examined, one study observed that 21% of individuals with CP/CPPS symptoms reported a history of musculoskeletal, rheumatological, or connective tissue diseases,11 and another noted a trend for higher manual tender point examination scores in CP/CPPS patients than in controls.16 Vulvodynia has also been described as co-occurring with FM.8,15
Only 3 studies, all of them on non-clinical populations, examined the overlap between urological unexplained clinical conditions and CFS. One study of twin pairs discordant for various definitions of chronic fatigue or CFS found that fatigued twins were up to 20 times more likely to have IC, and up to 6 times more likely to have chronic pelvic pain, than their non-fatigued co-twins.20 Conversely, CFS was reported more than twice as often by individuals with CP/CPPS symptoms than by asymptomatic age-matched controls.11 Likewise, individuals with vulvodynia more often reported CFS than did asymptomatic controls.15 We did not find comparative studies on overlap between TMD and any urological unexplained clinical condition.
Finally, more than half of the comparison studies (15 of 25) examined the overlap among urological unexplained clinical conditions and IBS. Several studies, using different methods, reported that 7% to 48% of patients with IC or symptoms of PBS also had IBS. 10,14,21,22 Others found that patients with IC were 11 times more likely to be diagnosed with IBS than were control participants.23 Further, 22% of men with CP/CPPS or its symptoms had IBS.11,19,22 Likewise, 37% of women with chronic pelvic pain met the criteria for IBS.24 Conversely, one study found that 35% of IBS patients, but only 14% of patients with inflammatory bowel disease, reported symptoms of chronic pelvic pain.25 In other studies, IBS was associated with increased probability of an IC diagnosis and increased urinary symptom scores.13,26,27 Vulvodynia has also been documented as comorbid with IBS.8,15,28
We found 16 publications that assessed individuals with a single unexplained clinical condition for the presence of at least 1 additional syndrome or hallmark symptom (Table 2). Despite methodological differences, the results of these investigations are generally consistent with those in Table 1. A study of patients with chronic pelvic pain symptoms found that 19% also had FM symptoms and 8% had CFS symptoms.29 These rates are similar to those noted in other publications of self-reported FM (17%) and CFS (9%) in individuals with IC30 and vulvar pain.31 Conversely, a recent study that used the American College of Rheumatology criteria for FM found that 11% of IC patients met the diagnostic criteria for FM.32 Additionally, 2 studies provided data on the overlap between TMD and at least 1 urological unexplained clinical condition. One study found that 13% of patients with TMD had a history of IC,33 and the other reported that almost 20% of women with vulvar pain reported TMD.31
Most of the single-condition studies examined the comorbidity of urological unexplained clinical conditions and symptoms with IBS. These studies reported that 30% to 75% of individuals with IC or IC symptoms had IBS or IBS symptoms,30,34,35 and 26% to 56% of IBS patients reported urinary problems.36,37 Likewise, 19% to 79% of patients with chronic pelvic pain had IBS or IBS symptoms.29,38–40 Among women with vulvar pain, 35% reported IBS.31
Our extensive review of the literature on urological and non-urological unexplained clinical conditions found a small but growing body of published reports documenting the comorbidity of these conditions. The most robust evidence for overlap exists for IBS and unexplained urological conditions, in terms of both the number of publications and the relative consistency of results across studies. This overlap may reflect either a publication bias or the anatomical fact that urological chronic pelvic pain conditions and IBS occur in a similar region of the body and involve visceral pain sensations. In contrast, the few studies on FM, CFS, and TMD suggest a more modest comorbidity with urological unexplained clinical conditions. Nonetheless, our findings generally support previous assertions of overlap among some or all of the urological and non-urological unexplained clinical conditions.2,3,41
The common sociodemographic and clinical features shared by these conditions, the symptom overlap in many of the case definitions, and the tendency to respond to the same therapies, have led several investigators to suggest that all unexplained clinical conditions are different manifestations of the same underlying pathophysiological process.1,3 Additionally, some experts have argued that the comorbidity of FM, CFS, TMD, and IBS could result from similarities in their case definitions.1 Although this explanation may apply to the overlap between urological pelvic pain conditions and IBS,41 urinary and pelvic pain symptoms are not part of the symptom criteria for FM, CFS or TMD. Thus, commonalities in symptom criteria cannot completely account for the overlap summarized in this review. Nonetheless, the heterogeneity of these conditions, especially sub-groups of patients with uncharacteristic symptoms, can account, in part, for the substantial overlap between these conditions.
Despite decades of research on unexplained clinical conditions, their etiology and pathophysiology remain elusive. We identify 3 dominant perspectives on the mechanisms for these conditions: 1) physiological processes, particularly neuroendocrine, immunological, and neurotransmitter dysfunction in the central nervous system; 2) victimization, abuse, and trauma; and 3) psychological distress, psychiatric disorders, stress tolerance, and recovery from stress.3 Although abnormalities have been demonstrated across several domains that are consistent with all of these perspectives, most studies have examined only single hypotheses. In addition, many affected patients do not exhibit neuroendocrine abnormalities, do not suffer from depression or other psychiatric conditions, and have not been abused or victimized. Organ-based approaches (i.e., examining the bladder, prostate, or skeletal system) have similarly failed to provide useful insights into the etiologies of, or effective treatments for, these conditions. In general, research on urological chronic pelvic pain conditions and similar disorders has not been able to distinguish epiphenomena from underlying pathophysiology.
A multidimensional conceptual model seems better suited to guide research and understanding, both of the conditions themselves and of their co-occurrence.1 A multidimensional model can encompass the range of predisposing (e.g., genetic and environmental influences, physiological perturbations), precipitating (e.g., trauma, infections), and perpetuating (e.g., central sensitization, perceptual amplification) factors, as well as predictors of chronicity (e.g., psychological and sleep disturbances) that are necessary but not individually sufficient for developing 1 or more unexplained clinical conditions (see Figure). A strength of this conceptualization is that it incorporates the 2 most prominent overarching theories of pain in order to facilitate the understanding and treatment of unexplained clinical conditions. The diathesis-stress model takes into account both the predisposing biological and genetic factors as well as the environmental and life events that interact to trigger behaviors and disorders.42 Alternately, the gate control theory of pain and its more recent expansion emphasize the role of the brain and its interaction with cognitive and emotional factors in pain transmission and modulation.43 The multidimensional model presented here combines these overarching theories to offer multiple hypotheses that test similarities in vulnerability factors and stressors, as well as the central affective, sensory, cognitive, motor, inhibitory, and autonomic responses involved in pain processing. Elements of this model has spurred a small but growing body of literature, especially in pain sensitivity and its enhanced perception, and central sensitization of pain in FM and CFS,44,45 and support the presence of basic similarities across these conditions. A more broadly based conceptual model can facilitate interdisciplinary linkages among the basic, clinical, and population sciences to investigate their etiology, clinical manifestations, comorbidity, and prognosis.
Conceptually driven investigations of the mechanisms that may predispose, precipitate, perpetuate, and predict these conditions can improve our understanding of urological unexplained clinical conditions and their comorbidities. Given the prominence of pain across the syndromes of interest, an examination of the central sensitization hypothesis is crucial to understanding their substantial comorbidity. Further, twin and family studies can clarify the role of shared genetic and environmental factors in their genesis and perpetuation.
Our review discovered several methodological shortcomings in the literature that undermine the strength of our conclusions and limit comparability across studies. First and foremost, study participants were drawn primarily from tertiary care clinics, hampering the generalizability of findings both to primary care patients and to the general population. Clinical ascertainment and other biases associated with the study of treatment-seeking populations are well known,46 and can be an important source of spurious comorbidity. Anxiety, poor coping, stressful events, and other psychological factors also may play a role in health-care seeking in IBS and FM.47 Thus, the comorbidity of urological and non-urological unexplained clinical conditions may be an artifact of healthcare use.
Second, we found inconsistencies in assessment. Although research on unexplained clinical conditions may be hampered by a lack of diagnostic markers, all of these conditions have established case definitions that are designed to facilitate symptom-based diagnoses as well as comparability across research studies. Nonetheless, many different methods were used to assess the conditions of interest, ranging from the use of established case definitions to self-report of physician diagnoses and review of computerized ICD-9 codes. Several investigations relied on medical history questionnaires or self-reported diagnoses without independent confirmation by medical record reviews or clinical examinations. This reliance calls into question the accuracy of all rates of conditions based on self-report, but especially those that require physical examination findings (e.g., FM) or exclusion of specific medical disorders (e.g., CFS). In addition, studies based on physician diagnoses in medical records may underestimate comorbidity if a patient attends one clinic for primary symptoms but is discouraged from seeking specialty care for other symptoms, as may occur in health maintenance organizations. Studies also differed in inclusion and exclusion criteria. All of these factors affect the degree of overlap among urological and non-urological unexplained clinical conditions and make comparisons across studies problematic.
Third, with the exception of CP/CPPS in men, most participants with unexplained clinical conditions were women. It has been debated whether certain urological unexplained clinical conditions, particularly IC/PBS and CP/CPPS in men, represent somewhat different manifestations of the same disorder and whether the risk factors and manifestations of IC/PBS differ in men and women.48 Sex differences clearly do exist in the epidemiology, symptomatology, physiology, and psychological features of FM, CFS, and IBS,49 raising the possibility that the overlap between urological and non-urological unexplained clinical conditions is sex-specific. Clearly, there are too few studies in men to fully examine any sex differences in overlap that may exist. Studies with large samples of men are needed to examine the role of sex in the co-occurrence of these conditions. Likewise, more data are needed to examine the association of these conditions across different cultural, racial, and ethnic groups. Finally, in the studies that compared 2 or more groups, the control groups consisted variously of healthy individuals, those with other chronic illnesses, or controls whose health status was unreported. Only 2 investigations controlled for genetic and common environmental effects by using co-twin control methods,9,20 and most studies did not use multivariate analytic techniques to adjust for demographic or other differences that could reduce the strength of the association between comorbid conditions.
Our extensive review of the published literature on the overlap between urological and non-urological unexplained clinical conditions suggests that comorbidity is frequent. However, these studies suffer from several methodological shortcomings that undermine the strength of our conclusions and limit comparability across studies. Future studies examining the overlap should adhere to the established research diagnostic criteria for each condition. Large scale, rigorously designed, and well-controlled studies in both clinical and community populations can yield more definitive answers to questions of overlapping symptoms and syndromes. Additionally, longitudinal cohort designs are needed to examine their risk factors, their temporal onset, and their prognosis. A recent study published since this review was undertaken found significant associations between antecedent non-urological unexplained clinical conditions and incident cases of IC/PBS,50 further underscoring the need for longitudinal studies. Other studies of urological unexplained clinical conditions should focus on health- and behavior-related characteristics – for example, quality of life, disability, or health care use – that can estimate the added burden of co-occurring syndromes on meaningful outcomes such as functional status, employment, and health care costs. Examining possible mediators for comorbidity – for example, abuse, victimization, or psychological distress – can elucidate the relationship between overlapping conditions and identify subgroups at risk for developing urological chronic pelvic pain syndromes and related conditions. Well-controlled studies to assess comorbidity and pathophysiological mechanisms can advance our understanding and improve the treatment of these conditions.
This work was supported in part by grant numbers U01 DK082325 from the NIDDK (Buchwald, Afari), R21 AR053963 (Buchwald) from NIAMSD, and R01AR51524 from NIAMSD (Buchwald, Afari). These grants supported the time of Drs. Buchwald and Afari in the design and conduct of the review, collection of publications, interpretation of the data, and preparation of the manuscript. John Kusek, PhD, Leroy Nyberg, PhD, and Christopher Mullins, PhD, of the primary funding agency reviewed drafts of the manuscript. In addition to the authors, the NIDDK Working Group on Urological Chronic Pelvic Pain included Daniel Clauw, MD, Jordan Dimitrakov, MD, John Kusek, PhD, Christopher Mullins, PhD, Leroy Nyberg, PhD, Christopher Payne, MD, Cecilia Peñacoba, PhD, Michael Pezzone, MD, Michel Pontari, MD, Jeannette Potts, MD, and John Warren, MD. The authors wish to thank the members of the Working Group for help in identifying relevant publications and reviewing drafts of this manuscript. We also wish to thank Dayron Rodriguez, Michael D. Smith, Jaeseop Lee, and Michael O’Leary of Harvard Medical School; Almudena López of University Rey Juan Carlos, Madrid, Spain; and Debra Sprague of the University of Washington for support in conducting the literature review and the preparation of this manuscript.
1A complete list of references for these publications is available from authors by request.
Portions of this paper were presented at the NIDDK symposium on Unexplained Chronic Pelvic Pain Syndromes on June 16–17, 2008, in Bethesda, MD.
None of the authors have any conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript.
Dr. Buchwald has full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Ms. Bullones Rodriguez made substantial contributions to the conception and design, acquisition of data, analysis and interpretation, drafting of the manuscript, and technical support. Dr. Afari made substantial contributions to conception and design, acquisition of data, analysis and interpretation, drafting of the manuscript, critical revision of the manuscript for important intellectual content, and obtaining funding. Dr. Buchwald made substantial contributions to conception and design, critical revision of the manuscript, obtaining funding, and supervision.