Clinical Phenotyping of Women With Interstitial Cystitis/Painful Bladder Syndrome: A Key to Classification and Potentially Improved Management

J. Curtis Nickel,*,† Daniel Shoskes‡ and Karen Irvine-Bird

From the Department of Urology, Queen’s University at Kingston, Kingston, Ontario, Canada, and the Department of Urology, Cleveland Clinic, Cleveland, Ohio

ABSTRACT

Purpose: We have proposed a clinical phenotype system (UPOINT) to classify patients with urological pelvic pain to improve the understanding of etiology and guide therapy. We examined the relationship between UPOINT and symptoms in patients with interstitial cystitis/painful bladder syndrome.
Materials and Methods: Patients with interstitial cystitis/painful bladder syndrome were classified in each domain of UPOINT, that is urinary, psychosocial, organ specific, infection, neurological/systemic and tenderness. Symptoms were assessed using the Interstitial Cystitis Symptom Index, Pain/Urgency/Frequency score and visual analogue scale for pain/urgency/frequency. Clinically relevant associations were calculated.
Results: The mean age of 100 consecutive patients with interstitial cystitis/ painful bladder syndrome was 48 years, median symptom duration was 7 years and median Interstitial Cystitis Symptom Index score was 12.8. The percent positive for each domain was urinary 100%, psychosocial 34%, organ specific 96%, infection 38%, neurological/systemic 45% and tenderness 48%. All patients were included in at least 2 domains, with 2 domains for 13%, 3 domains—35%, 4 domains—34%, 5 domains—13% and 6 domains—5%. The number of domains was associated with greater symptom duration (p _ 0.014) but not age. The number of domains was also associated with poorer general interstitial cystitis and pain symptoms (Interstitial Cystitis Symptom Index p _ 0.012, pain p _ 0.036) but not with frequency or urgency. The psychosocial domain was associated with increased pain, urgency and frequency, while tenderness was associated with increased Interstitial Cystitis Symptom Index score, pain/urgency/frequency score and urgency. The neurological/systemic domain was associated with increased Interstitial Cystitis Symptom Index score while the infection domain was not associated with any increased symptoms.
Conclusions: The UPOINT phenotyping system can classify patients with interstitial cystitis according to clinically relevant domains. Increased symptom duration leads to a greater number of domains, and domains that function outside of the bladder (psychosocial, neurological, tenderness) predict a significant impact on symptoms. We hypothesize that the UPOINT system can direct multimodal therapy and improve outcomes.

Key Words: cystitis, interstitial; phenotype; classification

Abbreviations and Acronyms
CP-CPPS _ chronic prostatitischronic pelvic pain syndrome
IC _ interstitial cystitis IC/PBS _ interstitial cystitis/ painful bladder syndrome
ICSI _ Interstitial Cystitis Symptom Index
MAPP _ Multidisciplinary Approach to Pelvic Pain
NIH _ National Institutes of Health
PUF _ Pain/Urgency/Frequency
UCPPS _ Urologic Chronic Pelvic Pain Syndromes
UPOINT _ urinary, psychosocial, organ specific, infection, neurogenic/systemic, tenderness
VAS _ visual analogue scale

Submitted for publication October 31, 2008.
* Correspondence: Department of Urology, Queen’s University, Kingston General Hospital, 76 Stuart St., Kingston, Ontario, Canada K7L 2V7 (telephone: 613-548-2497; FAX: 613-545-1970; e-mail: jcn@queensu.ca).
† Financial interest and/or other relationship
with Merck Frosst Canada, GlaxoSmithKline, Ortho-McNeil, Medtronic, Allergan, Farr Laboratories, Triton Pharma and Plethora Solutions.
‡ Financial interest and/or other relationship with Triurol.

0022-5347/09/1821-0155/0 Vol. 182, 155-160, July 2009
THE JOURNAL OF UROLOGY® Printed in U.S.A.
Copyright © 2009 by AMERICAN UROLOGICAL ASSOCIATION DOI:10.1016/j.juro.2009.02.122

          Clinicians who strive to practice an evidence-based approach to treating patients with interstitial cystitis/painful bladder syndrome are frustrated. Many promising oral, intravesical and surgical therapies for IC/PBS1 often fail in clinical practice, and when subjected to large randomized placebo controlled trials.^2,3 Only recently have we become aware that patients with IC/PBS are not a homogenous group with perceived bladder pain associated with urinary frequency and urgency, but rather a group of individual patients with widely differing clinical phenotypes. This realization has led NIH to fund the MAPP Study group to explore basic science (particularly biomarker and etiological studies) and epidemiology to better understand the differences in this heterogeneous group of patients. It is hoped that phenotyping patients may explain why some have persistent symptoms despite specific IC/PBS therapy and that the concept eventually may be applicable to direct better treatment plans.
           We recently described the clinically practical UPOINT phenotyping classification system for patients
diagnosed with UCPPS (CP-CPPS and IC/PBS).⁴ UPOINT is a 6-point clinical classification system that categorizes the phenotype of patients with UCPPS into 6 clinically identifiable domains including Urinary, Psychosocial, Organ Specific, Infection, Neurological/Systemic and Tenderness (muscle). This clinical classification system is not necessarily based on etiology (but may be eventually) but remains flexible (will incorporate new epidemiology, therapeutic and biomarker research as it becomes available). It is proposed that this new clinical tool for urologists will be used to direct individually based therapy. UPOINT has been recently evaluated and validated in males with CP-CPPS,⁵ and in this report we examine the concept in 100 consecutive female patients evaluated at an interstitial cystitis clinic.

METHODS

Patients
The patient population included 100 consecutive female patients with IC/PBS evaluated at a tertiary interstitial cystitis clinic (JCN). All patients were diagnosed on the basis of chronic (more than 6 months) pelvic pain, pressure or discomfort perceived to be related to the urinary bladder accompanied by at least 1 other urinary symptom such as urgency or frequency. Therefore, patients would have mostly fulfilled the diagnostic criteria described in the IC Data Base Study,⁶ the recent European Society for the Study of IC/PBS definition of IC/bladder pain syndrome,⁷ as well as the most recent definition of IC/PBS described at the NIH CPPS consensus meeting in Baltimore in December 2007.⁸ Urine must be sterile at diagnosis and assessment, and urinary cytology must be negative. All other diseases that could cause pelvic symptoms were excluded with our standardized history and physical examination which included a pelvic examination. Most patients underwent cystoscopy, usually under local anesthesia, while those with an equivocal diagnosis underwent an anesthetic challenge test (with alkalized lidocaine).⁹
UPOINT Domains
Patients were categorized into the clinical domains (details have been described previously).4 The urinary domain included patients reporting bothersome urinary frequency, urgency, nocturia, incontinence and/or dysuria. It was expected that most if not all patients would be included in this domain due to the criteria used to make the clinical diagnosis of IC/PBS.
           The psychosocial domain included patients determined to be clinically depressed (or with a recent history of depression), those with identifiable maladaptive coping mechanisms (eg catastrophizing) or those who had problems with social interactions. This was a clinical assessment based on history and focused interview.
           The organ specific domain included patients who reported pain with bladder recycling (typically pain with bladder filling and temporary relief with voiding), pain on bladder filling detected with low volumes of irrigation
fluid, glomerulations and/or Hunner’s ulcers noted during cystoscopy (under local or general anesthesia), and/or those with typical inflammation confirmed on bladder biopsy. Patients with equivocal findings (usually those with severe pelvic floor pain in whom it is difficult to decide if the problem is bladder pain, pelvic muscle pain/spasm or both) were assessed before and after an anesthetic challenge test (200 mg lidocaine alkalized with 8.4% sodium bicarbonate for a final solution volume of 10 cc instilled in an empty bladder and then drained by catheter after 10 to 15 minutes).^9,10 The Appendix provides a detailed description and interpretation of the intravesical alkalized lidocaine challenge test. It was expected that most patients would be included in this domain due to the criteria used to make the clinical diagnosis of IC/PBS.
           The infection domain included patients who were confirmed to have significant bacteriuria with typical uropathogenic bacteria in the previous 2 years associated with exacerbation of baseline symptoms and return to baseline symptoms following appropriate antimicrobial therapy. IC/PBS was diagnosed and UPOINT was assessed only when culture was sterile for a period of 3 months.
           The neurological/systemic domain included patients with a concurrent diagnosis of irritable bowel syndrome, fibromyalgia, chronic fatigue syndrome, vulvodynia or any other condition that suggested neuropathy or neural upregulation. The tenderness domain included patients with pelvic floor or lower abdominal muscle/ligament tenderness and or pain, including but not restricted to specific trigger points during standardized abdominal and pelvic examination.
Measures and Analysis
All patients completed the ICSI,¹¹ the PUF questionnaire,¹² and VAS measures of Pain, Urgency, Frequency which described these symptoms on a scale from 0 (no symptom) to 10 (the worse symptom imaginable). The ICSI score appeared normally distributed and, therefore, parametric statistics were used. ANOVA was used for comparison of multiple groups and Bonferroni’s Multiple Comparison test (Bonferroni correction applied comparing p value to the alpha level 0.05/k) was then used to compare pairs of groups. For comparison of category data the nonparametric Kruskal-Wallis test was used. Symptom duration and severity, and the number of positive domains were compared using Spearman rank correlation and reported as the r statistic with 95% confidence intervals and p value. The unpaired Student t test was used for comparison between scores in patients positive or negative for each domain. Significance was set at p _0.05 and no a priori power calculation could be performed since, as this is the first analysis of this kind to our knowledge, we could not know the range or variance of results.

DISCUSSION
Women diagnosed with IC/PBS were easily categorized into the UPOINT phenotype system using a traditional but standardized clinical assessment. It is not surprising that almost 100% of our IC/PBS cohort was included in the urinary and organ specific domains. Unlike male CP-CPPS, in which approximately 52% and 61% were categorized in the urinary and organ specific domains, repectively,⁵ the diagnosis of IC/PBS by definition includes women with pain perceived to be localized to the bladder, almost always associated with urinary frequency and urgency. In our clinic the diagnostic algorithm for such a diagnosis includes objective findings of pain on bladder recycling at time of local cystoscopy (when necessary confirmed with the anesthetic challenge test), glomerulation or Hunner’s ulcers on hydrodistention under general anesthesia. Thus, most of our patients were categorized in the organ specific domain, while all were included in the urinary domain. Treatment for these domains would be empirical trial of anticholinergics or pyridium for the urinary domain (some treatments may be considered for multiple domains), while organ specific therapies would be traditional IC/PBS bladder centric therapies including oral and intravesical hepinaroids, hydroxyzine for those with allergic history, hydrodistention under general anesthesia and perhaps surgical ablation of Hunner’s ulcers. It is likely that future treatment trials will show a differential treatment effect for subsequent subcategories of the organ specific domain (eg based on cystosopic and/or biopsy findings).
          Remarkably 13% of patients only had these 2 domains, and this could explain the failure of therapies directed against those domains to the exclusion of the other identified domains.^2,3 In fact, the other 87% were categorized into 3 (35%), 4 (34%), 5 (13%) or all 6 (5%) domains. The symptom severity measured by the ICSI (but not the PUF questionnaire) and reported pain severity increased as the number of domains experienced by the patients increased. The duration of symptoms but not age was strongly associated with an increase in the number of domains. These findings might suggest a progression from the 2 definitive urinary and organ specific domains to multiple domains as the duration of symptoms increased, but only natural history studies could confirm this impression.
           A surprising finding in our cohort of 100 consecutive patients with IC/PBS was the fact that 38% were included in the infection domain. It is generally conceded that cystitis associated with infection is not compatible with a concurrent diagnosis of IC. It is only when one realizes that we included an unselected population of women with an IC/PBS diagnosis made at a time when the urine was sterile, that more women with IC may experience exacerbation secondary to uropathogenic bacteriuria than previously recognized. In fact, women with bacteriuria or recent documented bacterial cystitis are routinely excluded from most IC studies. However, in our cohort when assessment was performed when the urine was sterile, we showed that being included in the infection domain did not impact any symptom parameter. Although the only randomized placebo controlled trial assessing long-term sequential and combination antimicrobial strategy in IC suggested that it may be generally beneficial, the adverse events make empirical antibiotic therapy unfeasible.¹³ We believe that antimicrobial therapy should be restricted for documented uropathogenic bacteriuria, which may be associated with a flare-up of baseline symptoms.
           The major psychosocial domain considerations identified in patients with IC/PBS include depression, maladaptive coping mechanisms (in particular catastrophizing) and social problems.¹⁴ The psychosocial domain was associated with greater pain, urinary urgency and frequency. Cause and effect between pain (and IC/PBS symptoms) and psychosocial parameters cannot be assessed in this study. However, it is likely that in individual patients one parameter can significantly impact the other. A recent cohort study of female patients with IC/PBS from 3 IC/PBS clinics examining the unique and shared associations of patient quality of life, IC/PBS symptoms, pain and psychosocial factors, showed that depression and greater helplessness/catastrophizing were associated with diminished quality of life.¹⁵ The data provide support for the clinical assessment, targeting and management of psychosocial parameters to improve not only patient adjustment but also symptoms and quality of life. Depression can be treated medically, maladaptive coping can be modified with cognitive behavioral therapy and social dysfunction can be treated with counseling.
           For patients to be included in the neurological/ systemic domain, the clinician must first identify conditions outside the bladder that imply chronic pelvic neuropathy, central sensitization and/or a diagnosis and symptoms of 1 of the other associated conditions (the most prevalent would include irritable bowel syndrome, fibromyalgia, chronic fatigue syndrome and possibly vulvodynia).16 Interestingly being included in this domain was only modestly associated with more severe ICSI scores, suggesting that although this domain is likely associated with significant quality of life issues (this needs to be evaluated in further studies), it does not impact pain and urinary symptoms as much as we originally hypothesized. Suggested therapy would include medical neuromodulation (amitriptyline, gabapentinoid therapy), surgical neuromodulation (implantable nerve stimulator) and/or therapy specifically directed toward an identifiable associated medical condition (for example directed therapy for irritable bowel syndrome).
           Patients identified in the tenderness domain had pelvic, perineal and/or abdominal muscle (or ligament) pain/tenderness, and in many cases discrete trigger points on physical examination.¹⁷ Many of these patients probably represent a group with pelvic floor dysfunction. This domain had a significant impact on general symptoms measured by the ICSI and PUF (this was the only domain associated with higher PUF scores) as well as urinary urgency. Therapy for this domain would include counseling, focused exercises, muscle relaxants and various forms of physical therapy. It is likely that the most successful therapy for this important domain would incorporate a combination of such approaches along with directed pelvic floor physiotherapy.
          One of the difficulties for urologists using the UPOINT classification system is our lack of knowledge and experience in classifying patients into the psychosocial and neurological/systemic domains, and then describing the actual individual patient phenotype within the domains. We initiated a deep, multicenter (10 urology sites comprise the Phenotypic Management of Pelvic Pain Research Group) biopsychosocial phenotyping case-control study (using 22 validated questionnaires) to better define these 2 phenotype domains.¹⁸ We will determine the prevalence of various subphenotypes within these domains (for example examining depression and catastrophizing within the psychosocial domain, and irritable bowel syndrome, vulvodynia and fibromyalgia within the neurological/systemic domain), and their impact on patient symptoms and quality of life. The group plans to develop clinically simple and applicable questionnaires to better quantify these domains, but until those are developed domain identification will be based on standard clinical assessment.
          One of the potential limitations could be considered a major strength in that the study was based on an unbiased real-life clinical practice cohort, not limited to exclusion criteria developed for specific clinical trials. The UPOINT system of classifying patients according to phenotypes and then directing tailored therapies to individual patients really just formalizes and guides what many clinicians are already doing.¹⁹ Most importantly the UPOINT approach to IC/PBS is not static but rather is extremely flexible. It will easily incorporate new epidemiology and basic science findings (including biomarkers) as proposed in the recently initiated NIH MAPP project. We cannot be sure if there should be other domains or whether some of the domains should incorporate more categories in such an enigmatic condition as IC/PBS.²⁰ Future research in etiological mechanisms, epidemiology and biomarkers will allow subcategorization in the specific UPOINT domains.
           Another obvious major limitation is that there are no evidence-based indications for specific therapies directed toward the various phenotypes. Additionally, since a number of treatments have multiple modes of action that may benefit patients with IC/PBS, specific treatments cannot be targeted exclusively to a particular phenotype. For example, amitriptyline has neuromodulatory, antinociceptive and antidepressant effects, mechanisms that could prove effective for more than 1 domain. The therapeutic suggestions outlined in this article are only best evidence-based and await further confirmation. However, this will likely come from prospective reallife clinical practice studies as the practical obstacles of testing such dynamic multimodal treatment hypotheses in randomized, controlled trials may be insurmountable. That said, researchers should consider phenotypically based inclusion and exclusion criteria to best match the specific treatment to related UPOINT domains.

CONCLUSIONS
Patients with IC/PBS with longer symptom duration have more UPOINT domains but prospective research is needed to clarify the time course of symptom development for the different domains. Most women with IC/PBS have UPOINT domains outside the bladder which likely is why bladder specific treatments often fail. Finally patients with positive psychosocial and tenderness domains had worse symptom scores, and future studies should address whether specific treatment for these domains will enhance the efficacy of traditional IC/PBS therapies.

APPENDIX
Intravesical Alkalized Lidocaine Anesthetic Challenge Test
The following test, based on the data from a proof of concept study,10 a randomized placebo controlled trial9 and clinical experience, is suggested to allow the clinician to assess the degree that bladder associated pain is responsible for the patients’ pelvic pain syndrome.
A. Procedure:
Step 1: A focused pelvic examination assessing bladder (bimanual examination), perineum, pelvic floor and painful trigger points is performed and documented.
Step 2: 12 or 14Fr Foley catheter is inserted into bladder and urine drained.
Step 3: The bladder is filled with saline by gravity feed (80 cm water pressure) until patient has discomfort and/or pain. The bladder is drained and volume documented (functional bladder volume/capacity).
Step 4: 5 cc 4% lidocaine is instilled into bladder followed directly by 5 cc 8.4% lidocaine and catheter clamped for 10 –15 minutes.
Step 4: Solution is drained and bladder refilled, similar to Step 3 except that volume is held for 5 minutes (hydrostatic bladder dilation) and then drained. Bladder volume after anesthetic challenge is calculated.
Step 5: Cystoscopy is performed and glomerulations, Hunner’s ulcers etc are documented.
Step 6: The focused pelvic examination, similar to Step 1, is performed again and the assessment documented.
B. Interpretation:
Organ specific domain is confirmed if one, two or all of these conditions are met:
1. Pelvic examination after anesthetic challenge test results in significant reduction or amelioration of perceived bladder pain (anterior vaginal wall and bimanual examination). This will further allow for better assessment of contributions of other perineal/pelvic muscles and trigger point contributions to pelvic pain.
2. The bladder volume after anesthetic challenge should be considerably more than the functional bladder volume attained before the challenge. While the actual increase in volume to confirm an organ specific domain has not been validated, the authors suggest that it should be at least a 50% increase.
3. The presence of significant glomerulations, submucosal and/or mucosal bleeding, and/or Hunner’s ulcers confirms an organ specific domain.