Excluding confusable diseases in patients with presumptive diagnosis of interstitial cystitis: A large patient cohort study.

S‐JO conceptualized the study; HJJ, YJK, and S‐JO created the data and the methodology; HJJ and S‐JO performed the formal analysis and wrote the original draft; HJJ, YJK, MSC, SJJ, and S‐JO performed the investigation, and reviewed and edited the manuscript; and MSC, SJJ and S‐JO supervised the study.

Of the patients who visited the outpatient clinic with pelvic pain, a total of 646 patients (164 men, 25.4%; 482 women, 74.6%) (mean age 60.8 ± 12.1 years) underwent an outpatient observation cystoscopy with a suspected IC. The average time from the first onset of pain to the visit to our hospital was 21.8 ± 33.2 months. At the first visit, microhematuria (red blood cells >5 per high‐power field) and pyuria (white blood cells >5 per high‐power field) were detected in 123 (19.0%) and 228 (35.3%) patients, respectively. Thirty‐nine (6.0%) patients showed positive urine culture findings. Nine (1.4%) patients showed atypical cells in urine cytology (Table  1 ). Five patients (0.8%) had positive urine tuberculosis test results. Imaging studies showed that 42 (6.5%) patients had abnormalities, of which 17 (2.6%) showed bladder wall thickening on CT. Twelve (1.9%) patients had urinary stones either in the bladder ( n  = 4) or in the upper urinary tract ( n  = 8). Two (0.3%) patients were suspected to have bladder tumors on imaging. One patient had bladder invasion and endometriosis on CT. In addition, prostate cancer was suspected in one (0.2%) patient (prostate‐specific antigen 2.8 ng/mL) and pleural tumor in another patient (0.2%) (Table  1 ). All the above patients then underwent observational cystoscopy to clarify any concurrent bladder pathology suggestive of IC. Imaging studies also showed that six (0.9%) patients were suspected to have urinary tuberculosis. Baseline characteristics in patients with presumptively diagnosed with interstitial cystitis ( n  = 646). a Abbreviations: AFB, acid fast bacillus; BPH, benign prostatic hyperplasia; HoLEP, holmium laser enucleation of the prostate; HPF, high‐power field; MTB & NTM PCR, Mycobacterium tuberculosis and nontuberculous mycobacteria polymerase chain reaction; PSA, prostate‐specific antigen. All variables are presented as mean ± standard deviation or number of patients (percentage). Body mass index is calculated as weight in kilograms divided by the square of height in meters. Of the 646 patients, 275 (42.6%) showed hyperemic bladder lesions on cystoscopy, suggestive of IC. In addition, bladder cancer was found in 3 (0.5%) patients (cases 2, 4, and 6 in Table  3 ). Of those who showed atypical hyperemic lesions on initial observational cystoscopy, four had to take anti‐tuberculosis medication immediately after the clinical diagnosis of urinary tuberculosis. One (0.2%) of the four patients with abnormal radiologic findings in the urinary tract had negative urine tuberculosis test results. However, the patient was clinically diagnosed with tuberculosis and was treated with anti‐tuberculosis medications. Observational cystoscopy confirmed bladder tumors in three patients and bladder stones in three patients. One patient with a history of holmium laser enucleation of the prostate for benign prostatic hyperplasia presented with gross hematuria and pelvic pain. There was a bleeding papillary mass in the prostatic fossa with no active lesion in the bladder. A biopsy of the mass revealed a prostatic transitional cell carcinoma (TCC). These patients were excluded from the IC diagnosis. Of the 275 patients, 254 (39.3%) underwent cystoscopic surgeries, including a bladder biopsy, together with a TUF for ulcerative lesions, diagnostic bladder hydrodistension, or both. Two patients only underwent hydrodistension because bladder biopsies performed at another hospital immediately before visiting our hospital had been negative. Pathologic examination of the bladder biopsy specimens revealed chronic inflammation in 236 (36.5%) patients. However, eight (1.2%) patients (cases 1, 3, 5, 7–11 in Table  3 ) were found to have bladder tumors, including bladder carcinoma in situ (CIS) in six patients and papillary TCC in two patients (Table  3 ). Additionally, one patient (0.2%) was reported to have prostate adenocarcinoma on concomitant transrectal prostate biopsy. During the follow‐up period after the initial cystoscopic surgery, a TUF for ulceration, therapeutic bladder hydrodistension, or both, together with a bladder biopsy, were performed if pelvic pain persisted or worsened. Of the 66 patients who underwent subsequent secondary cystoscopy, two were additionally diagnosed with bladder cancer (cases 12 and 13 in Table  3 ) (Figure  1 ). Of the patients who had severe intractable pain despite conservative treatment, 21 (3.3%) underwent supratrigonal or total cystectomy with urinary diversion or bladder replacement using the bowel segment. Pathologic examination of the cystectomy specimen showed bladder tuberculosis in one patient (case 7 in Table  2 ). Flowchart for clinical pathway for IC/BPS (Bx, Biopsy; IC/BPS, interstitial cystitis/bladder pain syndrome; TB, tuberculosis). 1 Four patients began taking tuberculosis medication immediately after cystoscopy, and two patients subsequently underwent bladder biopsy and began taking tuberculosis medication. 2 There are a total of two patients who underwent bladder biopsy at another hospital and received hydrodistension and transurethral fulguration of the ulcer as the first procedure at our hospital. 3 TB from bladder replacement ( n  = 1). Results of urinary tuberculosis‐related evaluation. Normal (IVP); Multiple perfusion defected areas in right kidney with renal pelvicalyceal wall enhancement, suggesting APN (CT) Abbreviations: AFB, acid fast bacillus; APN, acute pyelonephritis; CT, computed tomography; IVP, intravenous pyelography; MTB & NTM PCR, Mycobacterium tuberculosis and nontuberculous mycobacteria polymerase chain reaction; RBC/HPF, red blood cells per high‐power field; TB, tuberculosis; WBC/HPF, white blood cells per high‐power field. Overall, urinary tuberculosis was identified in seven (1.1%) of the 646 patients during the examination (Table  2 ). Urinary tuberculosis was confirmed in five of the seven patients based on positive urine AFB test results (Table  2 ). Five of the six patients whose findings were suspicious for urinary tuberculosis at baseline imaging showed positive tuberculosis in the urine AFB test (Table  2 ). Fourteen (2.2%) patients had genitourinary tract malignancies (13 with bladder tumors and one with prostate adenocarcinoma) (Table  3 ). Results of urinary tract malignancy‐related evaluation. Abbreviations: CIS, carcinoma in situ; CT, computed tomography; IVP, intravenous pyelography; TCC, transitional cell carcinoma; TRUS, transrectal ultrasonography. The biopsy is conducted when hydrodistension or transurethral fulguration is performed after imaging study, cytology test, and cystoscopy.

Most IC guidelines list urinary tuberculosis and genitourinary cancer as confusable diseases; therefore, they must be excluded from IC during the diagnosis process. The American Urological Association guidelines 1 , 6 do not mention the need for imaging studies. In the earlier East Asian guidelines, 3 urine cytology and cystoscopy were included as basic tests, whereas pelvic imaging studies and bladder biopsy were optional. In the most recent East Asian guidelines, 5 ultrasonography was included as a basic test. The 2008 European Association of Urology (EAU) guidelines did not include urine cytology and imaging studies as basic diagnostic tests. 4 However, ultrasound has been included in recent diagnostic guidelines for the urologic phenotyping of pelvic pain, 21 Most of these guidelines are based on the predominant disease patterns in high‐income countries. Therefore, special care should be taken to exclude confusable conditions in areas where certain diseases, such as tuberculosis, are rare. Of the seven patients finally treated with anti‐tuberculosis medication, 5 (71.4%) had positive urine tuberculosis test results. The percentage of patients suspected of having urinary tuberculosis on imaging examination was 0.9% ( n  = 6). One patient showed acute pyelonephritis on CT. In total, five patients showed tuberculosis in both urine tuberculosis tests and imaging examinations. Urinary bladder tuberculosis is difficult to diagnose but appears in imaging studies as caliectasis, hydronephrosis, ureteral stricture, and contracted bladder. 22 , 23 Although there were very few cases of urinary tuberculosis, this study reminds us of the importance of imaging studies in the screening phase in areas where tuberculosis is commonly found. As the detection rate of urine AFB smears and cultures in the diagnosis of tuberculosis is poor, methods with increased sensitivity have been developed, such as urine AFB fluorescence staining and MTB and NTM PCR. It is noteworthy that a 43.8‐year‐old female patient (case 7 in Table  2 ) was not suspected of having tuberculosis in either the imaging study or urine tuberculosis test; however, a granuloma consistent with tuberculosis was found in the bladder specimen after cystectomy. Malignancy of the urinary tract should be ruled out before IC diagnosis. Bladder cancer has been reported to be diagnosed in 0.32%–1% of patients, with a follow up of 29–72 months. 13 , 24 , 25 We found that bladder cancer was diagnosed in 2.2% of the patients. In this study, abnormalities were found in 42 patients on all imaging studies, including CT, but only two patients were suggestive of bladder cancer. Of the 17 patients who showed bladder wall enhancement or a thickened bladder wall, three were later diagnosed with bladder cancer (bladder CIS in two cases and papillary TCC in one case). Nine patients had atypical cells in voided urine cytology. On the other hand, bladder tumors were diagnosed in three patients only through initial observation cystoscopy. Therefore, no single test can completely exclude urinary tract malignancies. Imaging studies, cytology, and cystoscopy should complement each other to exclude urinary tract malignancies. Although the pathology in the initial bladder biopsy was reported as chronic inflammation of the bladder, there were two cases in which a CIS was diagnosed in a subsequent bladder biopsy during follow up (cases 12 and 13 in Table  3 ). It can be assumed that the amount of biopsied tissue was insufficient or that no representative bladder tissue was biopsied. This suggests that malignancy may not be ruled out with a single biopsy; therefore, a repeat bladder biopsy may be needed during follow up. Similar results have been reported in other studies. 13 , 14 , 15 Bladder tumors are found in many cases after the diagnosis of IC. The authors claimed that IC was a precursor lesion of bladder tumors. 24 , 25 However, in our experience, it is more likely that a malignant tumor was not adequately ruled out at the initial diagnosis. Based on the lessons learned from this study, we suggest that more detailed tests should be optionally included in the basic evaluation for IC. As recently recommended by the EAU 21 and the East Asian guidelines, 5 pelvic imaging studies should be performed. Because ultrasonography has limitations, the authors argue that applying more precise tools, such as pelvic CT, should be considered more seriously in cases with atypical clinical findings. We also argue that a bladder biopsy is necessary in cases other than those showing a typical Hunner lesion and that a repeat biopsy may be required if necessary. Most previous IC studies 13 , 24 , 25 had a relatively short‐term follow up. The strength of this study is that it was a large‐scale patient cohort study in which patients were observed over a long period. This study was limited by its retrospective nature. Additionally, the study was conducted at a single center; therefore, it may be difficult to generalize the results. As consistent diagnostic tests and regular follow ups were performed by a single urologist (S‐JO) under the protocol of electronic medical record order sets, we believe that patient selection bias was minimized. In conclusion, our results show that continuous efforts to rule out bladder tumors or tuberculosis are still essential in the follow up of patients with suspected IC, even if these diseases are not excluded at the initial examination. Imaging studies are necessary to rule out tuberculosis.

Interstitial cystitis (IC) is a chronic pelvic pain or discomfort associated with the urinary bladder that severely affects the quality of life of patients. 1 , 2 , 3 , 4 It is often accompanied by lower urinary tract symptoms, such as daytime frequency, nocturia, and urgency. During the diagnostic process, confusable diseases must be excluded as the cause of symptoms. 1 , 2 , 3 , 4 IC is diagnosed based on the exclusion criteria. 5 Exclusion diagnoses included malignancies and infections, including urinary tuberculosis. 4 There is no single optimal test for diagnosing IC. The most important test in the diagnostic process of IC is cystoscopy 6 , 7 to identify Hunner lesions. The classic description of Hunner lesion is “a reddened area with blood vessels radiating toward a central scar.” 4 However, the extent, location, and appearance of Hunner lesions vary from person to person, leading to no objective, standardized criteria for Hunner lesions. Therefore, correct cystoscopic detection of lesions is challenging, especially in medical institutions where patients with IC are not commonly encountered. 8 Additionally, there is currently no universally recognized biomarker for diagnosing IC. 5 Therefore, it is challenging to diagnose IC accurately. As a result, IC is often misdiagnosed as other diseases, 9 , 10 , 11 , 12 such as malignancy, 13 , 14 or overactive bladder. 15 To our knowledge, observational studies over a long period regarding the exclusion of confusable conditions are scarce. Here, we aimed to analyze the results of excluding confusable diseases in a patient population with presumptive IC experience over the last 15 years in actual clinical settings.

The authors have no conflicts of interest.

We retrospectively reviewed the electronic medical records of consecutive patients who presented to Seoul National University Hospital (SNUH) with pelvic pain suggestive of IC between October 2005 and December 2019. This study was approved by the Institutional Review Board of SNUH (IRB No. 2401‐148‐1506). Due to the retrospective nature of this study, the requirement for informed consent was waived by the Institutional Review Board. Patients underwent a diagnostic pathway for IC according to published guidelines. 2 , 3 , 4 , 16 , 17 The inclusion criteria were patients with pain or discomfort in the suprapubic, pelvic, urethral, vaginal, or perineal areas perceived to be related to the urinary bladder and concomitant lower urinary tract symptoms with increased daytime frequency, nocturia, and urgency. The exclusion criteria were age <18 years, confusable diseases such as benign or malignant genitourinary tumors, urinary tract infections, radiation cystitis, symptomatic urethral diverticulum, urinary stones, pregnancy, any significant gynecologic disease including vaginal candidiasis, pelvic inflammatory disease, urogenital prolapse, ovarian cyst, adenomyosis, endometriosis, and gynecologic malignancies. When the patient first visited the hospital for pelvic pain, detailed information was obtained regarding the time since the pain onset, the pain location, the presence of underlying diseases, and medical history. A physical examination; urinalysis—urine culture, uroflowmetry with postvoid residual volume measurement, and a frequency‐volume chart for 3 days 18 ; and urine cytology were performed. Serum prostate‐specific antigen was measured in men aged >50 years to rule out prostate cancer. If more than 10 5 /colony‐forming units of bacteria were detected in the urine culture, antibiotics were administered, and the test was re‐examined. A tuberculosis test was performed on urine specimens to identify genitourinary tuberculosis. In the last 15 years, the methods for diagnosing tuberculosis at SNUH have changed. Urine acid‐fast bacilli (AFB) smear and culture, urine AFB fluorescence stain (Aerospray® TB Slide Stainer, ELITechGroup Inc, Logan, UT, USA), and urine Mycobacterium tuberculosis (MTB) and nontuberculous mycobacterium (NTM) polymerase chain reaction (PCR) (Applied Biosystems 2720 Thermal Cycler, Thermo Fisher Scientific Inc., Waltham, MA, USA) were performed. Ultrasonography, intravenous pyelography, or computed tomography (CT) were performed as imaging studies to rule out other confusable diseases. When necessary, urodynamic studies were performed according to the International Continence Society (ICS) standardization protocol. 19 Observational cystoscopy was performed on an outpatient basis if ulcerative IC was suspected or if other diagnoses had to be ruled out. If an abnormal bladder mucosal lesion was identified during cystoscopy, the patient was hospitalized for cystoscopic surgery. A bladder biopsy, together with transurethral fulguration (TUF) of the ulcer lesion, was performed under general or spinal anesthesia. Diagnostic bladder hydrodistension was performed in most patients. The patient was placed in the lithotomy position. A rigid 24‐Fr cystoscope was inserted through the urethra. After performing bladder‐washing cytology, the bladder was carefully observed, mapped, and recorded to identify bladder mucosal lesions. Hydrodistension was performed according to the following protocol. 20 Normal saline was placed 80 cm above the pubic symphysis of the patient, and the bladder was filled to maximum anesthetic bladder capacity. Bladder hydrodistension was maintained for 8 min and then drained. After drainage of normal saline, mucosal changes such as glomerulation, cracks, and fissures were confirmed. A cold cup biopsy was performed on three representative hyperemic bladder lesions initially identified before hydrodistension. Bladder tissue samples were immediately preserved in formalin glass containers. A TUF was performed on the bladder mucosal lesions. After initially excluding confusable diseases, follow up was performed at regular intervals of 3 or 4 months on an outpatient basis. During follow up, patients were treated with medications such as analgesics, pentosane polysulfate, or both. If the pain was persistent or severe, the patient was hospitalized, and additional surgeries, such as TUF for bladder lesions along with bladder biopsy, therapeutic bladder hydrodistension, or both, were performed under general or spinal anesthesia. For intractable pain, cystectomy was performed with either bladder replacement or urinary diversion. All variables were expressed as means, standard deviations, and percentages. Baseline data were analyzed using descriptive statistics. Statistical significance was set at P values less than 0.05. All data were analyzed using SPSS for Windows (Version 27.0, SPSS, IBM Corp., Armonk, NY, USA).