Comorbidities of bladder pain syndrome/interstitial cystitis: a population-based study

Authors


Herng-Ching Lin, School of Health Care Administration, Taipei Medical University, 250 Wu-Hsing St., Taipei 110, Taiwan. e-mail: henry11111@tmu.edu.tw

Abstract

Study Type – System prevalence (cohort)

Level of Evidence 2a

What's known on the subject? and What does the study add?

Compared with the general population, patients with BPS/IC often experience difficulties in performing normal activities owing to physical limitations, decreased energy, greater pain and impaired social functioning

With the exception of metastatic cancer, separate conditional logistic regression analyses in this study suggested that subjects with BPS/IC were consistently more likely than subjects without BPS/IC to have all the medical comorbidities investigated. When compared with subjects without BPS/IC, subjects with BPS/IC had particularly higher odds of comorbid neurological diseases, rheumatological diseases and mental illnesses.

OBJECTIVE

  • • To explore the comorbid medical conditions of patients with bladder pain syndrome/interstitial cystitis (BPS/IC) in Taiwan using a cross-sectional study design and a population-based administrative database.

SUBJECTS AND METHODS

  • • The study included 9269 subjects with BPS/IC and 46 345 randomly selected comparison subjects.
  • • Conditional logistic regression analyses were performed to calculate the odds ratio for each of the 32 medical comorbidities (hypertension, congestive heart failure, cardiac arrhythmias, blood loss anaemia, peripheral vascular disorders, stroke, ischaemic heart disease, hyperlipidaemia, hepatitis B or C, migraines, headaches, Parkinson's disease, rheumatoid arthritis, systemic lupus erythematosus, ankylosing spondylitis, pulmonary circulation disorders, chronic pulmonary disease, diabetes, hypothyroidism, renal failure, fluid and electrolyte disorders, liver diseases, peptic ulcers, deficiency anaemias, depressive disorder, psychoses, metastatic cancer, solid tumour without metastasis, alcohol abuse, drug abuse and asthma) between subjects with and without BPS/IC.

RESULTS

  • • With the exception of metastatic cancer, the subjects with BPS/IC had a significantly higher prevalence of all the medical comorbidities analysed than subjects without BPS/IC.
  • • With the exception of metastatic cancer, separate conditional logistic regression analyses suggested that subjects with BPS/IC were consistently more likely than subjects without BPS/IC to have any of the medical comorbidities investigated in this study.
  • • When compared with subjects without BPS/IC, subjects with BPS/IC had particularly higher odds of comorbid neurological diseases, rheumatological diseases and mental illnesses.

CONCLUSION

  • • Our results indicated that subjects with BPS/IC had an increased prevalence of multiple comorbidities.
Abbreviations
BPS

bladder pain syndrome

IC

interstitial cystitis

OR

odds ratio

NHIRD

National Health Insurance Research Database

NHI

National Health Insurance

ICD

International Classification of Diseases

RA

rheumatoid arthritis

DD

depressive disorder.

INTRODUCTION

Bladder pain syndrome/interstitial cystitis (BPS/IC) is a chronic pain syndrome of unknown aetiology that primarily affects women and has no known cure [1]. It is characterized by suprapubic pain and may be accompanied by voiding problems including urgency, frequency and nocturia [2]. Compared with the general population, patients with BPS/IC often experience difficulties in performing normal activities owing to physical limitations, decreased energy, diminished health-related quality of life, greater pain, and impaired social functioning [3–5]. To better understand the clustering of diseases that affect patients with BPS/IC, the present study set out to explore the comorbid medical conditions of patients with BPS/IC in Taiwan by using a cross-sectional study design and a population-based administrative database.

SUBJECTS AND METHODS

We used data sourced from the National Health Insurance Research Database (NHIRD). This database is derived from the Taiwan National Health Insurance programme (NHI) and is provided to scientists in Taiwan for research purposes. Taiwan inaugurated its single-payer nationalized healthcare system in 1995, which has consistently held very high coverage rates, with 22.6 million of Taiwan's 22.96 million population (∼98.4%) being enrolled in the NHI programme in 2007. Each year, the Taiwan National Health Research Institute collects data from the NHI programme and sorts it into the NHIRD files, including registration files and original claims data for reimbursement. The NHIRD permits researchers to follow-up on the medical services used by each enrollee, beginning with the initiation of the NHI programme. Numerous studies employing the NHIRD have been published in internationally peer-reviewed journals [6,7]. In addition, previous studies have validated the completeness and accuracy of the NHIRD [8,9].

The present study was exempt from full review by the Institutional Review Board of Taipei Medical University because the NHIRD consists of de-identified secondary data released to the public for research purposes.

STUDY SAMPLE

Our cross-sectional study included a study and a comparison group. For the study group, we selected 9269 patients aged ≥18 years, who had received their first-time diagnosis of BPS/IC (International Classification of Diseases [ICD]-9-CM code 595.1 [chronic IC]) in ambulatory care visits (including outpatient departments of hospitals and clinics) between 1 January 2006 and 31 December 2007. Since administrative datasets are often criticized with regard to their diagnostic validity, the present study only included patients who had at least three diagnoses of BPS/IC coded in their ambulatory care claims, with at least two of those diagnoses being made by a certified urologist. In Taiwan, the first diagnosis reflects the clinician's assessment for the condition in question. The second diagnosis indicates the presence of the condition based on the outcome of the clinical assessment. The third diagnosis indicates that the patient received at least one prescription medication for the treatment of the condition. We furthermore ensured that among these prescription medications, every one of the BPS/IC subjects included in the present study received a prescription for sodium hyaluronate. Under the NHI programme, prescriptions for Cystistat are tightly regulated. In addition to reviewing the patient's medical history, performing a complete pelvic examination, and getting a positive result from a potassium sensitivity test, according to the regulations of the NHI, in order to prescribe Cystistat an attending physician is required to perform a hydrodistention test and a cystoscopy must demonstrate at least one severe abnormality according to the diagnostic criteria proposed by the European Society for the Study of IC/PBS [10]. Although this test is invasive despite the trend toward less invasive diagnostic criteria used abroad, it is considered to be the ‘gold standard’ for the diagnosis of IC/BPS [11]; therefore, it is very unlikely that anyone without a clear diagnosis of BPS/IC would be eligible for this treatment, and equally unlikely that misdiagnosed cases were included in the present study.

We likewise retrieved the comparison group for this study from the NHIRD. We first excluded patients who had previously visited ambulatory care centres with a diagnosis of BPS/IC and those <18 years old. We then randomly selected 46 345 comparison subjects to match the study subjects in terms of sex, age group (18–29, 30–39, 40–49, 50–59, 60–69 and >69 years), monthly income (NT$0-NT$15 840, NT$15 841-NT$25 000, ≥ NT$25 001 [US$1 = NT$33 in 2007]), geographic location (Northern, Central, Eastern, and Southern Taiwan), and index year using the SAS program proc surveyselect (SAS System for Windows, Version 8.2). We defined NT$15 840 as the first income level threshold as this is the government-stipulated minimum wage for full-time employees in Taiwan. For the study group, the year of index date was the year in which the study subjects received their first diagnosis of BPS/IC, while for the comparison group the year of index date was simply a matched year in which comparison subjects had used a medical service. For the comparison group, we assigned the date of their first use of medical services occurring during that matched year as the index date. As a result, 13 278 patients were included in the study.

COMORBIDITIES

We selected 32 medical comorbidities for analysis in the present study. Of these, 22 were selected based on the Elixhauser Comorbidity Index [12], which includes 30 comorbidity measures. We only chose 22 from the Elixhauser Comorbidity Index because of the low prevalence in Taiwan for the other eight comorbidity measures (valvular disease, neurological disorders, paralysis, weight loss, obesity, coagulopathy, lymphoma and HIV). We also selected 10 highly prevalent medical conditions in Taiwan including stroke, ischaemic heart disease, hyperlipidaemia, hepatitis B or C, migraines, headaches, Parkinson's disease, systemic lupus erythematosus, ankylosing spondylitis and asthma [13].

We only counted the 32 comorbid conditions if the condition occurred in an inpatient setting or appeared in two or more ambulatory care claims coded <1 year before and after the index date.

STATISTICAL ANALYSIS

We used the SAS statistical package (SAS System for Windows, Version 8.2, Cary, NC, USA) to conduct all the statistical analyses. Pearson chi-squared tests were performed to compare the prevalence of medical comorbidities between subjects with and without BPS/IC. We further used separate conditional logistic regression analyses (conditioned on sex, age group, monthly income and geographic region) to calculate the odds ratio (OR) for each of the 32 medical comorbidities between subjects with and without BPS/IC. A two-sided P value of <0.05 was considered to indicate statistical significance.

RESULTS

Of the 9269 subjects with BPS/IC and randomly matched 46 345 subjects without BPS/IC, the mean (sd) age was 50.3 (16.7) years; ∼50% were <50 years old. Table 1 shows the distribution of sociodemographic characteristics of the sampled subjects.

Table 1. Demographic characteristics of patients with BPS/IC and controls in Taiwan in 2006–2007 (N= 55 614)
VariablePatients with BPS/IC, N= 9269Comparison subjects, N= 46 345 P
n (%) n (%)
Sex  1.000
 Male1685 (18.2)8 425 (18.2)
 Female7584 (81.8)37 920 (81.8) 
Age  1.000
 18–29 years1020 (11.0)5 100 (11.0)
 30–39 years1477 (15.9)7 385 (15.9)
 40–49 years2205 (23.8)11 025 (23.8)
 50–59 years1977 (21.3)9 885 (21.3)
 60–69 years1126 (12.2)5 630 (12.2)
 ≥70 years1464 (15.8)7 320 (15.8)
Monthly income  1.000
 ≤ NT$15 8403893 (42.0)19 483 (42.0)
 NT$15 841–25 0003884 (41.9)19 460 (41.9)
 ≥ NT$25 0011492 (16.1)7 402 (16.1)
Geographic region  1.000
 Northern4449 (48.0)22 261 (48.0)
 Central2150 (23.2)10 729 (23.2)
 Eastern2447 (26.4)12 234 (26.4)
 Southern223 (2.4)1 121 (2.4)

Table 2 shows the prevalence of medical comorbidities between subjects with and without BPS/IC. Subjects with BPS/IC had a significantly higher prevalence of all medical comorbidities, except metastatic cancer, than subjects without BPS/IC.

Table 2. Prevalence of medical comorbidities in patients with BPS/IC and in the comparison group (N= 55 614)
ComorbidityPatients with BPS/IC, N=9269Comparison subjects, N= 46 345 P
n (%) n (%)
Cardiovascular   
 Hypertension3478 (37.5)14 812 (32.0)<0.001
 Ischaemic heart disease273 (3.0)750 (1.6)<0.001
 Hyperlipidaemia3137 (33.8)10 595 (22.9)<0.001
 Congestive heart failure747 (8.1)2 274 (4.9)<0.001
 Cardiac arrhythmias1292 (13.9)3 283 (7.1)<0.001
 Blood loss anaemia119 (1.3)315 (0.7)<0.001
 Peripheral vascular disorders119 (1.3)224 (0.5)<0.001
 Stroke1291 (13.9)4 090 (8.8)<0.001
Neurological   
 Migraines889 (9.6)2 008 (4.3)<0.001
 Headaches2414 (26.0)8 156 (17.6)<0.001
 Parkinson's disease265 (2.9)629 (1.4)<0.001
Rheumatological   
 RA1320 (14.2)2 754 (5.9)<0.001
 Systemic lupus erythematosus69 (0.7)135 (0.3)<0.001
 Ankylosing spondylitis312 (3.4)636 (1.4)<0.001
Pulmonary   
 Pulmonary circulation disorders19 (0.2)48 (0.1)0.010
 Chronic pulmonary disease2498 (27.0)7 164 (15.5)<0.001
 Asthma1230 (13.3)3 338 (7.2)<0.001
Endocrine   
 Diabetes, uncomplicated1809 (19.5)6 656 (14.4)<0.001
 Diabetes, complicated782 (8.4)2 713 (5.9)<0.001
 Hypothyroidism1583 (17.1)3 833 (8.3)<0.001
Renal   
 Renal failure383 (4.1)1 201 (2.6)<0.001
 Fluid and electrolyte disorders473 (5.1)1 146 (2.5)<0.001
Gastrointestinal   
 Liver diseases2290 (24.7)6 592 (14.2)<0.001
 Peptic ulcers502 (5.4)1 256 (2.7)<0.001
Viral/infectious   
 Hepatitis B or C814 (8.8)2 393 (5.2)<0.001
Haematological   
 Deficiency anaemias1200 (13.0)3 295 (7.1)<0.001
Mental illness   
 DD1488 (16.1)2 507 (5.4)<0.001
 Psychoses894 (9.7)1 764 (3.8)<0.001
Oncological   
 Metastatic cancer49 (0.5)184 (0.4)0.073
 Solid tumour without metastasis880 (9.5)2 439 (5.3)<0.001
Other   
 Alcohol abuse49 (0.5)112 (0.2)<0.001
 Drug abuse216 (2.3)524 (1.1)<0.001

Table 3 shows the ORs with corresponding 95% CIs for each comorbidity for the sampled subjects. Separate conditional logistic regression analyses (conditioned on sex, age group, monthly income and geographic region) consistently suggested that subjects with BPS/IC were more likely to have any of the medical comorbidities investigated in the present study than subjects without BPS/IC, with the exception of metastatic cancer. In particular, when compared with subjects without BPS/IC, subjects with BPS/IC had higher odds of comorbid neurological diseases, rheumatological diseases and mental illnesses.

Table 3. Adjusted OR of medical comorbidities in subjects with BPS/IC compared with subjects without BPS/IC (N= 55 614)
 Subjects with BPS/IC vs those without, OR, 95% CI P
Cardiovascular  
 Hypertension1.28 (1.22–1.34)<0.001
 Ischaemic heart disease1.85 (1.61–2.23)<0.001
 Hyperlipidaemia1.73 (1.65–1.81)<0.001
 Congestive heart failure1.70 (1.56–1.85)<0.001
 Cardiac arrhythmias2.13 (1.98–2.28)<0.001
 Blood loss anaemia1.90 (1.54–2.35)<0.001
 Peripheral vascular disorders2.68 (2.14–3.35)<0.001
 Stroke1.68 (1.57–1.79)<0.001
Neurological  
 Migraines2.35 (2.16–2.55)<0.001
 Headaches2.02 (1.93–2.11)<0.001
 Parkinson's disease2.14 (1.85–2.48)<0.001
Rheumatological  
 RA2.64 (2.46–2.83)<0.001
 Systemic lupus erythematosus2.57 (1.92–3.44)<0.001
 Ankylosing spondylitis2.50 (2.18–2.87)<0.001
Pulmonary  
 Pulmonary circulation disorders1.98 (1.16–3.37)<0.001
 Chronic pulmonary disease2.02 (1.92–2.13)<0.001
 Asthma1.97 (1.84–2.11)<0.001
Endocrine  
 Diabetes, uncomplicated1.45 (1.37–1.53)<0.001
 Diabetes, complicated1.48 (1.36–1.61)<0.001
 Hypothyroidism2.30 (2.16–2.45)<0.001
Renal  
 Renal failure1.62 (1.44–1.82)<0.001
 Fluid and electrolyte disorders2.12 (1.90–2.37)<0.001
Gastrointestinal  
 Liver diseases1.98 (1.88–2.09)<0.001
 Peptic ulcers2.06 (1.85–2.29)<0.001
Viral/infectious  
 Hepatitis B or C1.77 (1.63–1.92)<0.001
Haematological  
 Deficiency anaemias1.95 (1.82–2.09)<0.001
Mental illness  
 Depression3.35 (3.13–3.59)<0.001
 Psychoses2.70 (2.48–2.94)<0.001
Oncological  
 Metastatic cancer1.33 (0.97–1.83)0.102
 Solid tumour without metastasis1.89 (1.74–2.05)<0.001
Other  
 Alcohol abuse2.20 (1.57–3.08)<0.001
 Drug abuse2.10 (1.79–2.47)<0.001

DISCUSSION

The results of the present study show that patients with BPS/IC were more likely than matched controls to have suffered from numerous comorbid conditions within 1 year of their BPS/IC diagnosis. The comorbidity profile of BPS/IC was extremely wide, and included an increased likelihood of all the investigated conditions with the exception of metastatic cancer. The three most prevalent comorbid conditions were rheumatological and neurological diseases as well as mental illnesses.

Consistent with previous studies, the present study found an increased prevalence of rheumatoid arthritis (RA), systemic lupus erythematosus, and ankylosing spondylitis among patients with BPS/IC [14–18]. The strongest association was detected between RA and BPS/IC (OR 2.64). RA is a chronic destructive inflammatory disease of the synovial joints that occurs in 0.5–1.0% of the adult population worldwide [19,20].

While RA is well known to be an autoimmune disease, there is also substantial evidence pointing to the autoimmune character of BPS/IC [21], and it is therefore possible that the increased prevalence of RA and other rheumatological diseases found among patients with BPS/IC in the present study may involve the presence of shared underlying autoimmune disturbances. BPS/IC has been found to have many characteristics that support the hypothesis that it is an autoimmune disease [22]; immune competent cells such as lymphocytes, plasma cells and mast cells have been shown to accumulate in the detrusor muscle and lamina propria of individuals with BPS/IC [23]. Perhaps most importantly, a number of studies have shown that autoantibodies are present in patients with BPS/IC. One such study found various antinuclear antibodies to be present in 36% of patients with BPS/IC [24], and some antibodies that were previously assumed to be secondary have also been found that recognize urinary epithelial cells in the urine and serum of patients with IC/BPS [25].

As the only way to prove that a disease is not autoimmune is to demonstrate a non-autoimmune mechanism for its pathogenesis [21], and to date no such mechanism has been identified for RA, the classification of BPS/IC as an autoimmune disease remains speculative. One alternative approach to accumulating evidence regarding the autoimmune character of BPS/IC proposed by van de Merwe [21] is to demonstrate its positive association with other autoimmune diseases.

A previous cross-sectional study was conducted on the association between RA and BPS/IC. It surveyed the clinical records of 129 patients with classic BPS/IC, and 93 patients with non-ulcer BPS/IC in Sweden and found RA to occur in 13% of the patients with classic BPS/IC and in 4% of those with non-ulcer BPS/IC [14]. The 13% value observed among the patients with classic BPS/IC is close to the 14.2% value detected in the present study. Furthermore, based on the diagnostic criteria used in selecting our patient pool, the patients with classic BPS/IC in that study better match the BPS/IC subjects analysed in the present study.

The present study also found an increased prevalence of mental illnesses among the patients with BPS/IC. Patients with BPS/IC were 2.7 times more likely than matched controls to suffer from psychoses, and 3.35 times more likely to suffer from depressive disorder (DD). The prevalence of DD among patients with BPS/IC in the present study was 16.1%. As far as we know, only one study by Watkins et al. [26] reported a population prevalence of probable DD in women with BPS/IC symptoms. In their study, they found 34.8% of women with BPS/IC symptoms to have reported a probable DD, compared with a point prevalence of 4.9–5.4% among the general population (5.9–6.7% of women) [27,28]. In addition, the diagnosis of BPS/IC and depressive symptoms in their study was based on a self-reported questionnaire without the validation of clinical specialists. This may have undermined the strength of their findings; however, since they relied on questionnaires to identify women with probable DD, it is likely that the group of women who they identified as having a probable DD was actually a composite mixture of women with different depression diagnoses. Owing to the above-mentioned factors, the association between BPS/IC and DD has been left unclear to date. We believe that the results of the present study probably more closely approach the actual clinical prevalence of DD among patients with BPS/IC because of our large study design and strict diagnostic criteria.

The final major category of diseases detected to have notable increases in prevalence among patients with BPS/IC included neurological conditions. The three conditions investigated were migraines, headaches and Parkinson's disease. Although the present study is notable in that, to the best of our knowledge, it is the first to detect an increased prevalence of Parkinson's disease among patients with BPS/IC [14], the strongest association was found between BPS/IC and migraines (OR 2.35). One previous study investigating chronic pain syndromes found that 6.5% of patients with migraine headaches had BPS/IC [29], and another study reported that, among 207 BPS/IC cases and 117 controls, only 11.2% (13) of the age-matched controls had migraine headaches compared with 29.9% (61) of the cases [30]. Although the present study reported a significantly lower prevalence of migraine among patients with BPS/IC (9.6%), as stated above, we believe that our greater statistical power, research design, and strict diagnostic criteria should have yielded estimates more closely resembling the actual clinical prevalence. Current research suggests that the pain attributed to chronic pain disorders differs mechanistically from acute pain, with the most predominantly contributing factors stemming from the CNS. What were once understood to be a diverse range of disorders, including migraine, irritable bowel syndrome and IC/BPS, are more recently coming to be recognized as different manifestations arising from a common set of CNS processes, with neurotransmitters analogically acting as the ‘volume control’ of the pain processes involved [31]. Dopamine has been considered to play an important role in the pathogenesis of migraine for the past 30 years, with current research now associating certain dopamine-related gene variants with migraines [32–34]. The CNS control of micturition also involves many transmitter systems including dopamine, and abnormalities in these systems may underlie the increased prevalence of migraine among the patients with IC/BPS in the present study [35].

The principal strength of our study was our use of a population-based dataset, which included 9269 BPS/IC cases. This enabled us to trace all of the cases of RA and IC/BPS during the study period. The large sample size afforded a considerable statistical advantage which allowed us to detect subtle differences. This research thus provides fundamental data exploring a wide spectrum of comorbidities among patients with BPS/IC.

While the present results provide large-scale epidemiological data to the existing studies regarding the cluster of diseases surrounding BPS/IC, they must be seen in the light of a number of limitations. First, we used ICD coding to source both the BPS/IC diagnoses, as well as all the diagnoses of comorbid conditions analysed in this study.

Second, although we excluded all the comparison patients who had previously been diagnosed with BPS/IC, it is still possible that some comparison patients were suffering from BPS/IC but did not seek medical attention; however, if such a bias exists in the data, the results of our analysis would be biased towards the null, resulting in more conservative estimates.

Third, a causal association cannot be determined by our study. The present study is primarily a hypothesis generating work and secondarily an aid to clinical practitioners treating patients suffering from BPS/IC.

Fourth, individual information about lifestyle and dietary factors such as smoking, alcohol consumption, and obesity, was not available through the administrative dataset and we were therefore unable to adjust for their possibly confounding effects.

Fifth, Han Chinese constitute the vast majority of enrollees under the Taiwan NHI programme; therefore, the application of our findings among other ethnicities should be exercised with caution.

Sixth, numerous studies in the literature have suggested that aetio-pathological associations exist between IC/BPS and other medical conditions that we did not assess in this investigation. Some of these include fibromyalgia, irritable bowel syndrome, chronic fatigue syndrome, vulvodynia and pelvic floor dysfunction. Their absence from our consideration represents a major limitation of this study.

Finally, the study's design cannot preclude the possibility of ascertainment bias; patients with BPS/IC may be more likely to receive examination for clinically asymptomatic comorbidities than patients without BPS/IC. Likewise, those patients with comorbid conditions, by virtue of their greater exposure, are more likely to report BPS/IC than those who have less contact with physicians.

In conclusion, to the best of our knowledge, this is the first study to investigate the prevalence of a broad-spectrum of medical comorbidities associated with BPS/IC using a large population-based dataset from Taiwan. Our results indicated an increased prevalence of multiple comorbidities among patients with BPS/IC.

This study is based in part on data from the National Health Insurance Database provided by the Bureau of National Health Insurance, Department of Health, Taiwan and managed by the National Health Research Institutes. The interpretations and conclusions contained herein do not represent those of the Bureau of National Health Insurance, Department of Health, or the National Health Research Institutes.

CONFLICT OF INTEREST

None declared.

Ancillary