|Home | About | Journals | Submit | Contact Us | Français|
To examine the prevalence and timing of nonbladder conditions in a community cohort of women with symptoms of interstitial cystitis/bladder pain syndrome (IC/BPS).
As part of the Rand Interstitial Cystitis Epidemiology (RICE) study, we identified 3397 community women who met a validated case definition for IC/BPS symptoms. Each completed a survey asking if they had a physician diagnose them as having irritable bowel syndrome, fibromyalgia, chronic fatigue syndrome, migraines, panic attacks, or depression. If a positive response was received, subjects were asked to provide the age of symptom onset. All subjects were also asked to provide the date of IC/BPS symptom onset.
A total of 2185 women reported a diagnosis of at least one of the nonbladder conditions. Onset of bladder symptoms was not consistently earlier or later than the onset of nonbladder symptoms. Depression tended to occur earlier (P < .05), whereas fibromyalgia generally occurred later (P < .05). Mean age of onset was lowest for migraine symptoms, depression symptoms, and panic attacks symptoms, and greatest for fibromyalgia and chronic fatigue syndrome symptoms. Mean age of irritable bowel syndrome and IC/BPS symptom onset was between these other conditions.
These findings confirm the common co-occurrence of IC/BPS with chronic nonbladder conditions. In women with IC/BPS symptoms and coexistent nonbladder conditions, bladder symptoms do not uniformly predate the nonbladder symptoms. These observations suggest that phenotypic progression from isolated bladder symptoms to regional/systemic symptoms is not a predominant pattern in IC/BPS, although such a pattern may occur in a subset of individuals.
Interstitial cystitis/bladder pain syndrome (IC/BPS) is known to commonly coexist with other regional and systemic pain conditions, such as irritable bowel syndrome (IBS), fibromyalgia (FM), chronic fatigue syndrome (CFS), and migraine headaches,1-5 as well as with mental health disorders, such as depression and panic disorder.2,6-8 Based on these associations, it has been suggested that IC/BPS may progress from an organ-specific phenotype (isolated bladder symptoms) to a regional phenotype (bladder symptoms plus IBS) to a systemic phenotype (bladder symptoms plus IBS plus FM, CFS, migraines, etc.).5 However, very little longitudinal data exist to support this hypothesis. Furthermore, it is not clear whether concomitant mental health conditions typically precede IC/BPS symptoms. More knowledge about the relative timing of these associated conditions may provide important information about relevant patient subgroups that may aid in treatment approaches or may provide clues about the etiology of the symptoms. The purpose of this study was to examine the prevalence and timing of nonbladder conditions in a community cohort of women with IC/BPS symptoms.
As part of the Rand Interstitial Cystitis Epidemiology (RICE) study, a national telephone survey of 146 231 U.S. households was conducted to identify adult women with IC/BPS symptoms. These symptoms were classified using 2 RICE IC/BPS case definitions (high sensitivity and high specificity). The methods have previously been described.8,9 The survey identified 3397 women who met the high sensitivity definition, of whom 1469 also met the high specificity definition.
All women completed a survey asking if they had previously been diagnosed by “a doctor” with any of the following 7 conditions: interstitial cystitis or painful bladder syndrome, IBS, FM, CFS, migraines, panic attacks, or depression. If a positive response for 1 of the 6 nonbladder conditions was received, subjects were then asked to provide the age of symptom onset. All subjects were also asked to provide their current age, the date of onset for their IC/BPS symptoms, and whether they had an IC/BPS diagnosis. These variables were used to derive the age of symptom onset and to determine the order of onset of the conditions.
First, the proportion of women with each self-reported diagnosis was determined. Second, pairwise correlations were calculated among each of the 7 diagnoses; tetrachoric correlations10,11 were used so that these correlations were independent of the prevalence of the diagnosis. Standard Pearson product moment correlations for dichotomous measures become smaller as the prevalence of the dichotomous measure decreases. Tetrachoric correlations estimate the strength of the underlying association between the dichotomous measures that is independent of prevalence. Third, for each pair of diagnoses, we then calculated the proportion in which the symptoms of 1 diagnosis occurred before the other diagnosis, and determined whether this proportion significantly differed from 0.5 using a 1-sample chi-square test of proportions. Only women who reported having both diagnoses were included in each analysis. These analyses were then repeated, restricting to those with IC/BPS diagnoses. The mean age of diagnosis was calculated for subsets of women with each diagnosis. In each woman with more than 1 diagnosis, we recorded which diagnosis had the earliest symptom onset. For each diagnosis, we then calculated the proportion of cases in which the symptoms of that diagnosis began before the other concomitant diagnoses. Weights accounting for sample design and nonresponders were used for all statistical analyses.9 All statistical analyses were conducted using SAS software, version 9.1.3 (SAS Institute, Cary, NC).
A total of 2185 women (64% of the cohort) reported a diagnosis of at least one of the nonbladder conditions, as follows: depression (45%), IBS (22%), panic attacks (23%), migraines (18%), FM (11%), and CFS (6%). In addition, 534 women reported a diagnosis of interstitial cystitis or bladder pain syndrome. Weighted proportions for these diagnoses are provided in Table 1 in bold. Actual sample size may vary based on those missing age of symptom onset, with no more than 7% missing in any given subset.
Tetrachoric correlations between pairs of diagnoses are provided in Table 1 in italics. The range of values is 0 to 1, with 0 indicating no correlation and higher values indicating stronger associations between conditions. The highest correlations were seen for FM and CFS (0.708) and for depression and panic attacks (0.679). The lowest correlations were seen for IC/BPS and panic attacks (0.068), IC/BPS and migraines (0.164), IC/BPS and depression (0.167), and for panic attacks and migraines (0.175). All tetrachoric correlations between IC/BPS and the other conditions were statistically significant except for panic attacks.
Temporal associations between the symptom onset of the various conditions are provided in Table 1 in gray. The proportions represent the conditions for which column symptoms started before the row symptoms, in women who had both sets of symptoms. This analysis demonstrates that migraine symptoms and depression symptoms typically started before the other symptoms in most patients (P < .05 for all such comparisons, with migraine starting earlier than depression more often than the reverse, also P < .05). There was no consistent temporal pattern for the other conditions; with only one sequence being statistically significant-IBS was twice as likely to occur before FM than vice-versa (P< .05). The analyses presented in Table 1 were also performed in the subset of RICE women who met the more restrictive high specificity definition (n = 1469), and the results were essentially the same (data not shown).
If the analysis is limited to those women who reported physician diagnosis of interstitial cystitis or painful bladder syndrome (n = 534), the results change somewhat (Table 2). All 7 queried diagnoses were more common in this subgroup, which may reflect a greater exposure to medical care. As in the entire cohort, migraine symptoms tended to occur before most other symptoms (significantly earlier than FM and IC/BPS). However, in this smaller group, FM symptoms typically started after the other symptoms (P < .05 for all temporal associations except CFS). Other temporal relationships did not yield consistent patterns. In particular, symptom onset of the mental health conditions (panic attacks and depression) was not consistently earlier or later than the symptoms of the other conditions. In this cohort, IC/BPS symptoms tended to occur after migraine (69%) and IBS (59%) symptoms, but before panic attacks (39%) and FM (36%) symptoms.
In the entire cohort of 3397 women, there were 1413 identified with 2 or more diagnoses. In these women, migraine, depression, and IC/BPS symptoms were more likely to occur first, whereas IBS, panic attacks, CFS, and FM symptoms were less likely to occur first (data not shown). Almost exactly half (51%) of the RICE IC/BPS cohort who had nonbladder diagnoses reported that these symptoms started before the IC/BPS symptoms. IC/BPS and IBS had a similar age of symptom onset (32.4 and 32.1, respectively). Mean age of onset was lowest for migraine symptoms (23.8 years), depression symptoms (29.3 years), and panic attack symptoms (30.8 years), and greatest for FM (37.5 years) and CFS symptoms (35.4 years) (Table 3).
These findings confirm previous observations that IC/ BPS is often accompanied by chronic nonbladder pain conditions and mental health disorders.3, 5, 12 The rates of these nonbladder conditions in our study are consistent with those previously reported (Table 4). Although our study did not include a control group, previous studies with controls have demonstrated each of these conditions to be significantly more common in patients with IC/BPS.3-5,12 When examining the co-occurrence of these conditions, the highest correlations were seen for depression and panic attacks (0.679), and for CFS and FM (0.708). These findings are expected, as they have been previously demonstrated to be closely associated conditions.13-15 Most of the remaining correlations were only moderate (~0.3) or low, including all correlations between IC/BPS and the other conditions. It should also be noted that just over a third of the RICE cohort (n = 1212, or 37%) reported none of the nonbladder conditions. Therefore, whereas these diagnoses are common in patients with IC/BPS, they are not universal.
A unique aspect of our study is that we obtained retrospective data regarding the age of symptom onset for a variety of diagnosed conditions. One important observation from these data are that the bladder symptoms do not show a consistent tendency to precede the nonbladder symptoms. These findings are consistent with those of Warren et al,3 who identified a high rate of antecedent nonbladder syndromes (including all 6 conditions examined in our study) in women with new diagnoses of IC/BPS. Taken together, our findings indicate that about one third of women with IC/BPS have isolated bladder symptoms, about one third have bladder symptoms which precede the nonbladder symptoms, and about one third have nonbladder symptoms that precede bladder symptoms.
However, we note that symptoms of IC/BPS and IBS occur around the same age, with FM and CFS occurring later. This suggests that many patients may begin with a regional pain syndrome in the pelvis (bowel or bladder) and then progress to systemic symptoms. This progression may be bidirectional between adjacent organs (bladder and bowel) with some patients beginning initially with IBS and then progressing to IC/BPS and finally FM and CFS, and other patients beginning with IC/BPS and progressing to IBS and then to FM and CFS. In this cohort of women with IC/BPS symptoms, a few additional temporal trends are evident. Migraine headaches, if they occur, tend to occur before other symptoms (including bladder symptoms). This is also true for depression, but to a lesser extent. Conversely, FM symptoms and CFS symptoms, if they occur, tend to occur later than the other symptoms (including bladder symptoms), as well as having a strong tendency to co-occur. Because the FM and CFS symptoms are more systemic in nature, this does suggest that there may be a transition from local/regional symptoms to more systemic symptoms in some patients.
The temporal relationships observed in the entire cohort (Table 1) were not always consistent with those observed in the subset of women with an IC/BPS diagnosis (Table 2). For instance, in the entire cohort, migraine and depression symptoms occurred significantly earlier in almost all comparisons, whereas in the smaller subset with an IC/BPS diagnosis, fewer of these temporal relationships were significantly different. Conversely, FM symptoms occurred significantly later in most comparisons in the subset with an IC/BPS diagnosis, but in fewer of the comparisons in the entire cohort. In both of our temporal comparisons, symptoms of IBS, IC/BPS, CFS, and panic attacks did not consistently occur earlier or later than other symptoms. It is notable that, despite the smaller sample size, the same number of comparisons in the subgroup reached statistical significance compared with the entire cohort (12 in each). This suggests that the observed differences may be due to factors other than the different sample sizes. For instance, women with a reported diagnosis of IC/BPS would be expected to have had more interactions with healthcare professionals. This is supported by the higher rate of all 6 nonbladder diagnoses in the subgroup with an IC/BPS diagnosis. It is possible that an IC/BPS diagnosis is a marker for a different, or more severe, phenotype, because more severe symptoms may lead to more healthcare seeking and more assigned diagnoses. An analysis of healthcare seeking in the RICE cohort and its relationship to symptom severity is underway.
The relationship between mental health disorders and IC/BPS is a rather sensitive subject to many, as certain patients have experienced poor care and frustration due to the misconception that the IC/BPS symptoms are simply a manifestation of a “hysterical” personality. Fortunately, this antiquated thinking now seems to be rare. Our data suggest that mental health conditions (particularly depression) do predate IC/BPS symptoms in a substantial number of women, but an almost equal number first experienced these symptoms after the IC/BPS symptoms. The relationship between mental health disorders and IC/BPS is poorly understood. In some patients, it may be a reaction to the chronic pain symptoms. However, a biological relationship may also exist that could predispose women to a variety of conditions. For instance, it is recognized that patients with depression are more likely to report pain than nondepressed patients, that patients with chronic pain are more likely to report depression symptoms, and that the presence of concomitant pain and depression makes both conditions more difficult to treat.16 Furthermore, exacerbation of pain tends to worsen depression symptoms, and vice versa.17 Depression and pain also seem to share common neuroanatomic and neurochemical pathways.17 More specifically, serotonin, norepinephrine, and corticotrophin-releasing factor have been suggested as mediators of both depression and bladder pain or dysfunction.18-20
A number of limitations to this study deserve discussion. First, the analysis of all conditions is based on self-reported medical diagnoses and symptoms. Previous studies have demonstrated that self-reported diagnoses are reasonably accurate for many chronic medical conditions and malignancies.21-23 However, to our knowledge, the accuracy of self-reporting for the diagnoses examined in this study is unknown. Previous analyses have demonstrated that the RICE community women report significant and bothersome symptoms, which are equivalent to those observed in a clinic cohort of established patients with IC/BPS.24 Furthermore, approximately 90% of these women have sought care for their bladder symptoms (data not shown). Given the bothersome nature of these symptoms, it is reasonable to assume that these women may remember details about symptom onset more accurately than women with more mild symptoms. Second, the age of symptom onset may have been subject to recall bias. Third, we did not perform a comprehensive assessment of the symptoms of the nonbladder disorders. Such an analysis would be able to identify individuals with relevant but undiagnosed symptoms. Fourth, the results may be biased due to greater medical provider familiarity with certain conditions (eg, depression, migraines). This could explain why migraines and depression were identified as early conditions compared with chronic fatigue and fibromyalgia. Finally, because the study was limited to those with IC/BPS symptoms, it describes the sequence of diagnoses not among the entire population with any of the 7 diagnoses, but rather the sequencing among those with IC/BPS symptoms. Despite these limitations, we feel that these data do provide valuable new information about the temporal association between IC/ BPS and its associated conditions.
Community women with IC/BPS symptoms are commonly diagnosed with nonbladder pain conditions and mental health disorders. When 2 or more of these disorders coexist, the onset of bladder symptoms was not consistently earlier or later than onset of nonbladder symptoms. Migraine and depression symptoms tended to occur early, whereas fibromyalgia and chronic fatigue symptoms occurred later. Onset of IBS symptoms was not clearly earlier or later than IC/BPS symptoms. These findings suggest that phenotypic progression from isolated bladder symptoms to more regional/systemic symptoms is not a predominant pattern in IC/BPS. More definite data about IC/BPS symptom progression requires prospective longitudinal studies.
Funding Support: National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) (grant U01 DK070234).
Financial Disclosure: The authors declare that they have no relevant financial interests.