Weissman MM, Gross R, Fyer A, Heiman GA, Gameroff MJ, Hodge SE, Kaufman D, Kaplan SA, Wickramaratne PJ. Interstitial Cystitis and Panic DisorderA Potential Genetic Syndrome. Arch Gen Psychiatry. 2004;61(3):273-279. doi:10.1001/archpsyc.61.3.273
Evidence from a genetic linkage study had suggested a possible syndrome
in some families with panic disorder (PD). This syndrome includes bladder
problems (possibly urinary interstitial cystitis [IC]), thyroid disorders,
chronic headaches/migraine, and/or mitral valve prolapse. In 19 multiplex
families with PD, one marker (D13S779) on chromosome
13 gave a logarithm of odds score of more than 4 when individuals with any
of the syndrome conditions were analyzed as affected. Families with the bladder
problems yielded the highest logarithm of odds scores. These findings were
replicated in an extended sample of 60 families. Whereas PD had been well
characterized by direct interview, the urologic problems had been found only
via medical history checklists and records. A case review by a board-certified
urologist suggested they could be IC.
To determine whether patients diagnosed as having IC by urodynamics
and/or cystoscopy and their first-degree relatives (FDRs) have increased rates
of the syndrome conditions, thus validating that the bladder problems observed
in the linkage study could be IC and providing further support for the panic
Case-control and family history study.
Two metropolitan urology clinics.
One hundred forty-six probands (67 with IC and 79 with other urologic
disorders) and 815 FDRs.
Main Outcome Measures
Lifetime rates of syn-drome conditions in probands and FDRs who were
blind to urologic or psychiatric diagnoses in the proband.
Compared with patients without IC, patients with IC had a significantly
higher lifetime prevalence of PD (controlling for age and sex) (odds ratio,
4.05; 95% confidence interval, 1.22-13.40; P = .02)
and a higher lifetime prevalence of any of the syndrome disorders (controlling
for age and sex) (odds ratio, 2.22; 95% confidence interval, 0.89-5.54; P = .09). First-degree relatives of probands with (vs without)
IC were significantly more likely to have PD, thyroid disorder, urologic problems,
and any of the syndrome disorders (controlling for age and sex of the relative
and sex of the proband) (adjusted odds ratio, 1.95; 95% confidence interval,
1.13-3.38; P = .02). These results in relatives were
not influenced by PD in probands, and did not change substantially when controlling
for the proband-relative relationship, modeling age as a categorical (vs continuous)
variable, or excluding FDRs with PD. There were no interactions between proband
IC status and sex of the relative.
The increased frequency of seemingly disparate disorders in patients
with IC and their FDRs is consistent with the genetic linkage findings in
families with PD. These findings suggest that the bladder problems observed
in the linkage study may be IC. The hypothesis that there is a familial, possibly
pleiotropic, syndrome that may include IC, PD, thyroid disorders, and other
disorders of possible autonomic or neuromuscular control deserves further
We report results from a study designed to follow up genetic linkagefindings of a previously unidentified syndrome in some multiplex familieswith panic disorder (PD).1 This syndrome includedPD, various urinary problems (possibly interstitial cystitis [IC]), thyroiddisorders, mitral valve prolapse (MVP), and/or chronic headaches/migrainein some family members. The genetic linkage study had selected the familiesbased on initial evidence of at least 3 family members affected with PD.2,3 While the information on PD was obtainedfrom careful interviews, information on the bladder problems and other medicalconditions was obtained only through medical records and history checklists.By using these less-than-definitive methods for the medical conditions, logarithmof odds scores of 4.2 on chromosome 13 (marker D13S779)were obtained in the first 19 multiplex families with PD and these other medicalconditions.1 These findings were further supportedin a genome scan performed on an extended sample that included the initial19 families and an additional 41 families, for a total of 60 multiplex familieswith 587 individuals.4
We present findings from a new study aimed to follow up the bladder/kidneyproblems. The bladder/kidney problems yielded the highest logarithm of oddsscores in the genetic linkage study, but the diagnostic meaning was unclear.Our records were reviewed by a urologist (S.A.K.) who believed that they couldbe IC. We examined the rate of PD and the other syndrome disorders in a sampleof patients diagnosed as having IC by urologists using urodynamics and/orcystoscopy with bladder distention.
The specific hypothesis tested is that probands with IC and their first-degreerelatives (FDRs), compared with control subjects and their FDRs, will havean increased rate of the syndrome disorders, including PD, thyroid disorders,MVP, and/or severe headaches/migraine. These findings would partially validatethat the bladder problems seen in the linkage study could be IC and also addfurther support for the panic syndrome identified in the linkage study.
There are several hypotheses implicating various neurobiological mechanisms,such as autonomic dysregulation, in the cause of PD.5,6 Itis possible from a developmental perspective that a putative gene(s) wouldgive rise to additional dysfunctions in other areas, such as the bladder,in which regulation involves smooth muscle function or corticotropin-releasinghormone (CRH). Therefore, these seemingly unrelated conditions might be explainedby mutations in certain genes that exert effects on multiple aspects of physiologicaland anatomical features and lead to pleiotropic phenotypes in probands andtheir families. Pleiotropy is defined as the phenomenon by which a singlegene is responsible for several distinct and seemingly unrelated phenotypiceffects.7
Panic disorder is a highly familial complex genetic disorder8 with heritability, based primarily on twin studies,of about 49%.9 The lifetime prevalence of PDis about 1% to 3% cross-nationally, with a mean age of onset in early adulthood.Women have higher rates of PD than men, and the clinical manifestation ofPD is similar across diverse cultures.10 Symptomsinclude recurrent episodes of sudden unpredictable apprehension and associatedautonomic manifestations involving the cardiorespiratory system (shortnessof breath, chest pain, and palpitations), neurologic symptoms (dizziness,paresthesias, and trembling), gastrointestinal symptoms (nausea and abdominaldistress), other autonomic symptoms (hot flashes, chills, and sweating), andcognitive symptoms such as fear of dying. Treatment established through controlledclinical trials includes a range of pharmacological agents, most commonlyselective serotonin reuptake inhibitors, and behavioral approaches.11 None are fully effective, and PD is associated withincreased rates of medical use and suicide attempts.12
Interstitial cystitis is a chronic debilitating bladder syndrome ofunknown cause, and there is no generally accepted treatment.13 Therehave been several attempts to estimate the prevalence of IC in the UnitedStates.14 The largest most systematic set ofdata, based on self-report from the National Household Interview Study, showsa lifetime prevalence of 0.5% after weighting of the US population by age,race, and sex.15 It is more commonly foundin females, with a median age of onset at 40 years.16,17 Therole of genetic susceptibility has not been thoroughly investigated. One smalltwin study18 found considerably higher concordancein monozygotic vs dizygotic twins. Five of the 8 monozygotic and none of the26 dizygotic twins had confirmed IC.18 Preliminaryfindings from a family study18 suggest higherrates of IC in the FDRs of IC patients vs population controls.
Interstitial cystitis symptom presentation varies, but most commonlyincludes urinary frequency and urgency, nocturia, severe pain on bladder filling(typically relieved with voiding), and sterile urine.19,20 Interstitialcystitis encompasses a major portion of the chronic pelvic pain syndrome,21,22 which includes many urologic patientswith bladder and/or pelvic pain, irritative voiding symptoms, and negativeurine culture and cytologic test results.23 Interstitialcystitis in males is characterized by impairing clinical symptoms typicalof chronic prostatitis (pain on voiding and erectile dysfunction) withoutevidence of leukocytes or bacteria cultured in the prostatic secretions.24- 26 A syndrome remarkablyanalogous to IC, called feline IC, occurs in domestic cats. Studies27- 29 of cats and humanssuggest central nervous system involvement, including subtle abnormalitiesof the hypothalamic-pituitary-adrenal axis and a significant increase in tyrosinehydroxylase immunoreactivity in the locus coeruleus.
Available treatments for IC are based primarily on observational dataand a few clinical trials. Treatments included are cystoscopic hydrodistentionof the bladder, amitriptyline hydrochloride, antihistamine (oral hydroxyzinehydrochloride), pentosan polysulfate sodium, and intravesical dimethyl sulfoxidetherapy.30,31 These treatmentsmay improve symptoms, but there is insufficient information to know whethertreatment modifies the long-term course.32 Interstitialcystitis is considered a local manifestation of a systemic disease, possiblyan autoimmune disorder, but this is controversial. Systematic studies33,34 of large samples of patients withIC have found increased rates of autoimmune diseases, migraine headaches,and hypothyroid disease. Recently, the National Institute of Diabetes andDigestive and Kidney Diseases requested research applications in basic cellular,molecular, and genetic studies of IC.
Eligible participants were English-speaking patients aged 18 to 70 yearsfrom 2 urology clinics in New York City, headed by Columbia University–affiliatedboard-certified urologists (D.K. and S.A.K.), based on their urologic diagnosis,independent of the other syndrome disorders.
Cases were women with IC and men with chronic prostatitis (with normalurine sediments and sterile urine and prostatic fluid), considered the maleequivalent of IC.22 Controls were patientswith bladder diseases that have well-established, diagnosable, underlyinganatomical causes. Controls included men and women with noninvasive bladdercancer or detrusor instability, a condition characterized by involuntary contractionsof the smooth muscular coat of the bladder. In women, the condition is usuallysecondary to cystocele (a condition in which the bladder base descends belowthe inferior ramus of the symphysis pubis either at rest or with straining,due to a defect in the anatomical support of the bladder), and in men, itis secondary to benign prostatic hypertrophy. Also included among controlswere women diagnosed as having cystocele and men with benign prostatic hypertrophyor prostate cancer.
To avoid selection bias, computer-generated lists of all patients diagnosedas having IC were provided to us by assistants (a medical student and a secretary)to the urologists who were unaware of our study hypothesis. For controls,we were given lists of cases that met the control inclusion criteria. No informationon psychiatric or other conditions was included, just the urologic diagnosis.As required by the institutional review board, all patients were first senta letter signed by their urologist inviting them to participate in the study.They were given an opportunity not to participate by calling in and declining.Those who did not call in within 2 weeks were followed up and invited to participate.The patients received no other information about the study from the urologistsor the investigators before receiving the letter. When subjects were calledback, they were told that we would ask health questions about themselves andtheir FDRs. All of the calls to the subjects for recruitment and interviewwere made by an assistant who was blind to the urologic diagnosis and didnot have access to the patients' medical records. Sixty-seven (72.8%) of the92 available cases and 79 (43.4%) of the 182 available controls agreed toparticipate in the study. All participants gave informed consent and agreedto provide family history information.
Urologic diagnoses were made by a board-certified experienced urologist(D.K. or S.A.K.), before this study, blinded to the study interview data.Urodynamic tests with cystometry and cystoscopy (with hydrodistention) reports35 were available for 126 (86.3%) of the patients (27.4%underwent both tests, 36.3% underwent urodynamic tests only, and 22.6% underwentcystoscopy only). Of the 9 patients who underwent neither test, 8 underwentultrasonography and other tests and 1 underwent ultrasonography only. Medicalrecords were missing for 9 of the 146 patients. Patients were assessed bya trained research assistant, under the supervision of a psychiatrist (R.G.).Both were blinded to the urologic diagnosis and were not affiliated academicallyor geographically with the urology clinic. Lifetime PD was determined by diagnosticassessment, similar to that used in the genetic linkage study.2 First,the anxiety section of the Schedule for Affective Disorders and Schizophrenia–LifetimeVersion,36 a structured psychiatric interview,was administered.37 The interviewer also wrotea narrative case history for patients who reported any psychiatric symptoms.The data were then used by the same 2 senior clinicians who diagnosed PD inthe linkage studies (M.M.W. and A.F.), blinded to the urologists' and theinterviewer diagnoses, to confirm the psychiatric diagnosis using the best-estimateprocedure.38 Psychiatric symptoms and informationon the medical conditions in the FDRs were obtained from the patients usingan extended version of the Family History Screen, a brief validated instrumentfor collecting lifetime psychiatric history on an informant and FDRs.39 There is a well-documented underreporting of diagnosisusing family history methods, because informants may not have sufficient informationabout all the relatives' symptoms to meet diagnostic criteria. Therefore,we used broader criteria for PD in the FDRs, including panic attacks. Medicalexaminations were not available on FDRs, but we used the same methods to obtainmedical conditions in the relatives as had been used in the linkage study.1
t Tests and χ2 testswere used to test for group differences in demographic characteristics. Comparisonswere made between the 2 proband groups and between the FDRs of these 2 groups.
Separate logistic regression analyses were used to evaluate the associationof IC with each of the 4 disorders comprising the putative syndrome in probands(PD, MVP, thyroid disorder, and chronic headaches/migraine), with IC as anindependent variable (1 indicates present; and 0, absent) and each syndromedisorder as the outcome (1 indicates present; and 0, absent). We also createda binary outcome variable to capture the presence of any of the 4 syndromedisorders. Because probands with IC were younger and had more females thancontrols, age and sex were entered as covariates in all analyses. Each analysiswas performed twice, with age entered as a continuous variable and also asa 5-level categorical variable based on quintiles of the age distribution,because there was no a priori assumption about how age might be acting asa confounder. We tested the interaction between sex and proband IC statusin all analyses. Statistical significance was set at the 5% level (P<.05, 2-tailed).
Separate logistic regression analyses were used to evaluate the associationof proband IC with each of the 5 disorders comprising the syndrome in FDRs(PD, MVP, thyroid disorder, chronic headaches/migraine, and urologic [bladderor kidney] problems), with proband IC as the independent variable (1 indicatespresent; and 0, absent) and FDR's status on each syndrome disorder as theoutcome (1 indicates present; and 0, absent). We also created 2 binary outcomevariables to capture the presence in relatives of any of the 5 syndrome disordersand the presence of MVP, thyroid disorder, chronic headaches/migraine, and/orurologic problems. Panic disorder was omitted from this second definitionto rule out the possibility that the presence of PD explained the findings.All parameters were estimated using generalized estimating equations40 to adjust for the nonindependence of observationsamong relatives from the same family. We did not use survival or Cox proportionalhazards regression models because we did not have the age of onset in relatives.Statistical significance was set at the 5% level (P<.05,2-tailed).
To rule out the effects of PD, which is an independent familial disorder,we performed all of the previously described analyses on a restricted setof FDRs (those with probands who did not have PD). We also controlled forthe following variables: FDR age (because FDRs of IC probands were youngerthan FDRs of non-IC probands), FDR sex (to further reduce potential confounding),sex of the proband informant (because of differential reporting by sex), andthe relationship between FDR and proband (using a 3-level variable coded asparent, child, or sibling). Each analysis was performed twice, with age enteredas a continuous variable and also as a 5-level categorical variable basedon quintiles of the FDR age distribution, because there was no a priori assumptionabout how age might be acting as a confounder. We tested the interaction betweenFDR sex and proband IC status in all analyses. Statistical significance wasset at the 5% level (P<.05, 2-tailed).
Patients with IC were younger than controls (mean ± SD, 44.8± 13.3 vs 60.3 ± 9.6 years; P<.001),and most were women (83.6% vs 41.8%; P<.001).Cases and controls did not differ in race or ethnicity (88.1% vs 77.2% white),number of completed school years (mean ± SD, 15.2 ± 2.5 vs 14.8± 2.4), or percentage employed (74.6% vs 63.5%). All analyses wereadjusted for sex and age. We report the analyses adjusting for age as a continuousvariable, because the results were nearly identical when adjusting for ageas a categorical variable.
There was more than a 4-fold higher risk of PD and more than a 6-foldhigher risk of thyroid disorder among patients with IC compared with controls,and more than a 2-fold increased risk of having any of the disorders composingthe syndrome (Table 1). Ratesof headaches/migraine and MVP did not differ significantly between groups.There was no significant interaction between proband sex and IC status inany of the analyses (P range, .13-.83).
Eight hundred fifteen FDRs were identified (315 of IC probands and 500of controls). There were no sex differences in relatives by proband group.Relatives of probands with IC were younger (mean ± SD, 50 ±22 vs 54 ± 22 years; t784 = 2.47, P = .01) and had fewer children. All analyses were adjustedfor age. As with the proband analyses, we report the FDR analyses adjustingfor age as a continuous variable, because the results were nearly identicalwhen adjusting for FDR age as a categorical variable.
We first examined the prevalence of the syndrome disorders in all theFDRs of all probands, including probands with PD (data not shown). As expected,based on numerous family studies,41 PD wasfamilial. The odds of the outcome disorder in the FDRs of IC patients wereincreased more than 2-fold for PD, thyroid disorder, urologic problems (excludingbladder cancer), and the syndrome disorder whether PD was or was not includedin the definition. There was no significant interaction between FDR sex andproband IC status in any of the analyses (P range,.11-.62).
We next restricted our analysis to FDRs of probands without PD to determinewhether the syndrome was being transmitted independent of PD in the proband(Table 2). We also controlledfor sex of the informant to ensure that bias was not introduced by unequalrepresentation of female informants in the IC groups. The results did notchange substantially. The odds of the outcome disorder in the FDRs of IC patientswere increased more than 3-fold for PD, more than 2-fold for thyroid and urologicproblems, and nearly 2-fold for the syndrome whether PD was or was not includedas a syndrome disorder.
These results, in a sample of patients carefully diagnosed as havingIC, show an increased risk of PD and of the syndrome in IC patients and theirFDRs. These findings, together with findings from the genetic linkage study,1 while still tentative, suggest that in a subgroupof patients with PD, IC and other seemingly disparate disorders may be partof the same syndrome.
Previous studies found an increased comorbidity of PD with cardiovascularproblems,42 chronic headaches/migraine,43 MVP,44 and thyroiddisorders.45 It is usually assumed that theseassociations are spurious (ie, a misclassification due to an overlap of symptoms).None of the family studies of PD have determined whether medical disordersthat coaggregate with PD have an increased familial risk independent of PDin relatives of probands with IC. Our findings suggest that the associationsmay be a result of a shared cause, such as genetic susceptibility. Our findingsmay also explain some of the association observed between MVP and autoimmunethyroid disorders46,47 and migraine.48
A pleiotropic gene might give rise to any of several plausible biologicalmechanisms shared by IC, PD, and the other syndrome disorders. There are relateddata, although speculative, that may explain our findings or suggest morespecific hypotheses (the study of Weissman et al1 hasdetails). Autonomic dysregulation is implicated in the cause of PD.5,6 The bladder's function involves smoothmuscle function regulated through innervation from autonomic nuclei,49,50 so that changes in autonomic tonemight lead to voiding difficulties, as seen in patients with IC. Stress, whicharouses the noradrenergic system, was shown to be associated with symptomexacerbation in IC patients.51 Tricyclic antidepressants,mostly amitriptyline, which inhibit central norepinephrine, may be effectivein some patients with IC.52 Animal model datashow that cats with IC have increased plasma norepinephrine concentrations.53 Autonomic mechanisms have also been considered ascausative factors in other nonpsychiatric syndrome disorders. Mitral valveprolapse in patients with PD is usually the mild noncalcified type; its causehas been related to a more general dysautonomia.44 Migraineinvolves abnormal dilation of cerebral blood vessels, an action that is underautonomic control.54
Another way that autonomic reactivity might be involved in IC and PDis via neurogenic inflammation. Human and animal model data indicate a defectin the bladder's cytoprotective glycosaminoglycan lining that could allowpenetration of various substances that can activate bladder mast cells. Mastcell–derived proinflammatory and vasoactive molecules may, in turn,contribute to the pathogenesis of IC. Bladder mast cell activation is mediatedand augmented by neurotransmitters and neuropeptides, such as serotonin,55 and serotonergic imbalance is implicated in PD. Mastcells might also play a central role in the pathogenesis of migraine and immune-mediatedthyroid disorders.56
The common genetic susceptibility possibly shared by the disorders composingthe syndrome might be linked to the Barrington nucleus. This pontine nucleusimplicated in urination49 links parasympatheticpreganglionic neurons with prosencephalon-projecting nuclei, thus providingan anatomical substrate for coregulation of pelvic visceral symptoms and mentalactivity in the prosencephalon.57 The Barringtonnucleus contains numerous CRH neurons that project to the spinal parasympatheticnucleus innervating the bladder.58 Dysregulationof the hypothalamic-pituitary-adrenal axis plays a role in the cause of PD,59 and causes enhancement of central secretion of CRH.Increased CRH secretion from the Barrington nucleus might, in turn, inhibiturination and cause voiding problems typical of IC.
The strengths of our study include a sizable sample, state-of-the-arturologic diagnoses, reliable assessment of PD, and information on FDRs. Alimitation of the family history approach is that a patient with PD may bemore likely to report PD in relatives.60 Wetook care of this potential bias by also restricting our analysis to relativesof probands who did not have PD. Another limitation is the low response rateamong controls. However, this potential selection bias could not distort ourfindings in the FDRs, even if PD was not equally distributed between respondersand nonresponders in the control group, because the strategy we adopted ensuredthat results in relatives were independent of PD status in probands. Otherlimitations include the lack of medical assessment of MVP, thyroid disorder,or headaches/migraine. Moreover, we only assessed FDRs, and the genetic linkagestudy included multiplex families spanning several generations. Thus, disordersin the extended family may have been missed by including only FDRs. Pleiotropywould not require that all elements of the expression of the phenotype bepresent in an individual. Finally, while our results concerning the increasedrisk of the syndrome in the IC patients and their FDRs are consistent withthe findings of the genetic linkage study, a family history study alone cannotvalidate a genetic syndrome or confirm pleiotropy.
Potential clinical implications of this finding include identificationof new pharmacological interventions for IC, targeting specific neurotransmitterreceptors.61 Selective serotonin reuptake inhibitors,which are effective in PD patients, might inhibit serotonergic activationof mast cells and modulate exaggerated bladder activity through down-regulationof central postsynaptic serotonin receptors.62,63 Inaddition, trials of CRH antagonists as anxiolytic agents in PD patients arein process.64 These novel agents might be effectivealso in treating IC. Urologists should be aware of the increased prevalenceof PD, a treatable disorder, among their IC patients. Future research shouldinclude efforts to replicate the family aggregation and genetic findings andclinical trials with selective serotonin reuptake inhibitors for IC.
It is likely that the range of syndrome disorders is larger than wehave identified, including disorders shown to be associated with IC, suchas fibromyalgia, celiac disease, and irritable bowel syndrome, which in turnhave been associated with PD and/or migraine and other disorders of possibleautonomic or neuromuscular control, leading to the speculation that many orall of these conditions share underlying pathophysiologic features. Also,PD frequently co-occurs with mood and other anxiety disorders that have alsobeen associated with several other syndrome conditions. The phenotype of PDmay be broader than identified in this study. There is increasing recognitionthat phenotype hunting in family and genetic studies of psychiatric disordersmay profit from assessment of a wide range of medical and psychiatric disorders.65
Corresponding author and reprints: Myrna M. Weissman, PhD, Departmentof Psychiatry, College of Physicians and Surgeons, Columbia University, Divisionof Clinical and Genetic Epidemiology, New York State Psychiatric Institute,1051 Riverside Dr, Unit 24, New York, NY 10032 (e-mail: firstname.lastname@example.org)
Submitted for publication January 13, 2003; accepted September 22, 2003.
This study was supported by a National Alliance for Schizophrenia andDepression Senior Investigator Award (Dr Weissman); and grants MH2874 (DrWeissman) and MH35792 (Dr Fyer) and training grant 5T32-MH13043 (Dr Gross)from the National Institute of Mental Health, Rockville, Md.