The potassium sensitivity test: a predictor of treatment response in interstitial cystitis


Suresh K. Gupta, Urology, Arrowe Park Hospital, Wirral, Merseyside, UK.



To determine whether the potassium sensitivity test (PST) can be used to predict the response to treatment with intravesical sodium hyaluronate in patients with interstitial cystitis.


Thirty-eight patients diagnosed with interstitial cystitis were recruited; each had a PST, carried out in a double-blind fashion, followed by six weekly doses of intravesical sodium hyaluronate. The patients were assessed before and after treatment using a self-administered interstitial cystitis symptom index (CSI) and problem index (PI). The clinical response was defined as none (<25% improvement in clinical symptoms), mild (>25%), moderate (50–75%) and excellent (>75%).


The PST was positive in 23 and negative or indeterminate in 13 patients; two patients withdrew from the study. Overall 20 of 36 (55%) patients reported an improvement after six doses of intravesical sodium hyaluronate, but 17 (74%) with a positive PST improved, compared to only five (22%) with a negative test (P = 0.03). There was an improvement in the CSI after treatment in both groups, but a significant improvement in the PI only in patients with a positive PST (P = 0.01). The magnitude of change for the CSI and PI was significantly greater in the positive than in the negative group (CSI, P = 0.043; PI, P < 0.001). There were no major complications. Three patients complained of pain after the test, and two with a positive and one with a negative PST developed a urinary tract infection.


Although the role of the PST in the diagnosis of interstitial cystitis requires further clarification, the test helps to predict the response to treatment with glycosaminoglycan-substitution therapy.


potassium sensitivity test


cystitis symptom index


problem index


National Institute of Arthritis, Diabetes, Digestive and Kidney Disease.


Interstitial cystitis is a chronic debilitating clinical syndrome characterized by irritative voiding symptoms in the form of frequency, nocturia, urgency and pain. Previously it was considered a rare condition that only affected women, but recent epidemiological studies show that it is more prevalent than previously thought [1]. It is also being more commonly diagnosed in men, and some studies suggest that a proportion of patients diagnosed with chronic pelvic pain and chronic abacterial prostatitis may share a common pathophysiology, and have a variant of interstitial cystitis [2]. Several possible causal factors, including autoimmune response, mast cell activation, neuropathic changes, ‘toxic’ urine and altered urothelial permeability, have been proposed [3–9]. Studies in the last decade strongly suggest the possibility of a defect in the bladder's mucous lining of glycosaminoglycan in a large subset of patients [8,9]. Based on the epithelial permeability model, Parsons et al.[10] developed a test in which bladder mucosal integrity is tested by instilling a potassium solution intravesically.

Glycosaminoglycan-replacement therapies, e.g. sodium pentosan polysulphate and intravesical sodium hyaluronate, are now being increasingly used as a first line of treatment. However, the results are variable and the outcome of the treatment is unpredictable [11–13]. Logically, the treatment with glycosaminoglycan-replenishing drugs should be more effective in patients with interstitial cystitis who have epithelial dysfunction. The aim of the present prospective study was to assess whether the potassium sensitivity test (PST) can be used to predict the response to treatment with intravesical sodium hyaluronate.


Over a year, 38 patients with symptoms suggestive of interstitial cystitis or chronic pelvic pain of unknown origin were recruited into the study. The local ethics committee approved the protocol and informed written consent was obtained from all patients participating in the study. Interstitial cystitis was diagnosed according to the definition of National Institute of Arthritis, Diabetes, Digestive and Kidney Disease (NIADDK), except for cystoscopic findings. The important criteria for diagnosis were significant urinary frequency, urgency, pelvic pain and persistently negative urine analysis. Patients had no apparent cause for chronic urinary symptoms, no NIADDK exclusion criteria and persistent symptoms for ≥ 9 months. Patients with confirmed UTI, active genital herpes, a previous history of pelvic radiation and those who had previously received oral pentosan polysulphate or intravesical sodium hyaluronate were excluded from the study.

Each patient completed a self-administered interstitial cystitis symptom index (CSI) and problem index (PI) devised by O’Leary et al.[14]; both these instruments have been designed for assessing interstitial cystitis and were validated for self-administration.

The PST was performed as described by Parsons et al.[15] except for two important differences. First, we used 0.3 m potassium chloride as the test solution instead of 0.4 m, to avoid excessive provocation of symptoms. Second, both the patients and the investigator were unaware of the identity of the solutions, as the pharmacy supplied 40 mL of sterile water and of 0.3 m potassium chloride in bottles labelled A and B. Each patient had three consecutive bladder instillations of 5 min each (the patient being unaware of the solution type). The first instillation was always normal saline, the second and third being either solution A and B. Each solution was instilled intravesically and slowly over 2–3 min to minimize the volume-induced provocation of urinary symptoms. The solution was left for 5 min and the patient asked to grade the pain and urgency on an analogue scale of 0 (no pain or urgency), 1–3 (mild pain or urgency), 4–6 (moderate pain or urgency), 6–8 (severe pain or urgency) and 9–10 (excruciating pain or urgency). Between each instillation with test solution the bladder was rinsed with distilled water. The pharmacy disclosed the identity of the solutions at the end of the study. The test was considered positive if pain or urgency with potassium solution was 2 points more than with normal saline, and negative if none of the solutions caused any pain or urgency. If a patient reported pain and urgency with both solutions, the test results were considered indeterminate, unless the potassium solution was more provocative. The result of the test was not disclosed until the patient's response to treatment was assessed, to avoid any patient- or investigator-based bias.

At the end of the test the bladder was rinsed with saline, 50 mL of sodium hyaluronate was instilled into the bladder and the catheter removed. The patient was asked to retain the solution in bladder for 40 min, during which they were asked to lie on their back, each side and front for 10 min each. All patients received six weekly doses of sodium hyaluronate and those who responded to the treatment were given a choice of continuing treatment with monthly instillations of sodium hyaluronate. At the end of the weekly instillations, each patient completed the CSI and PI, with any change from baseline assessed as the primary outcome variable. The validated self-administered index consists of eight questions which measure the severity of symptoms, e.g. frequency, urgency, nocturia and pain, and assesses how problematic these symptoms are for the patient. The maximum score sum for CSI is 20 and for the PI is 16. The patients were also requested to grade the clinical improvement as no improvement (<25%), mild (25–50%), moderate (50–75%) and excellent (>75%).

The Mann–Whitney U-test was used to compare differences in age, duration of symptoms, baseline CSI and PI, and the improvement after treatment in the CSI and PI in patients with a positive and negative PST. The cystoscopic findings of the two groups were assessed using the chi-square test for two proportions. The CSI and PI before and after treatment were compared using Wilcoxon's test for paired samples.


Thirty-eight patients were initially recruited but two withdrew because they had adverse effects; one had a positive and the other a negative PST. Selected baseline characteristics are shown in Table 1. There was no significant difference in two groups in age, duration of symptoms, cystoscopic findings, and the CSI and PI before treatment.

Table 1. 
The baseline data
N patients2313 
Men/women 3/20 3/10 
Mean (sd):
Age, years49.39 (15.6)44.31 (9.22)0.31
Duration of symptoms, months32.87 (19.07)31.33 (18.7)0.74
Cystoscopic findings (typical changes), n/N 11/19 7/120.9
Mean (sd):
CSI12.72 (2.83)13.54 (2.75)0.55
PI10.83 (2.72) 11.46 (2.10)0.58

A significant clinical improvement was reported by 17 of 23 (74%) patients with a positive PST, vs three of 13 (23%) with a negative PST (P = 0.03). Four patients with a positive test reported excellent, eight moderate and five a mild improvement in symptoms. Three patients with a negative test reported a moderate improvement. There was an improvement in the CSI in both groups, but a significant improvement in the PI only in those with a positive PST (P = 0.01). The change in the CSI and PI after treatment (Table 2) was significantly greater in those with a positive than in those with a negative PST (CIS, P = 0.043, PI, P < 0.001).

Table 2. 
The CSI and PI in patients with a positive or negative PST, before and after intravesical sodium hyaluronate therapy
Mean (sd) indexPST
N patients2313
 before12.72 (2.83)13.54 (2.75)
 after9.61 (3.69)12.08 (3.63)
 change−3.08 (3.34)−1.46 (1.89)
 before10.83 (2.72)11.46 (2.10)
 after7.52 (3.64)11.38 (2.60)
 change−3.3 (2.75)−0.08 (1.66)

Of the 20 patients who responded to the treatment at the end of 6 weeks, only 12 wished to continue with monthly instillations for 6 months; nine of these 12 maintained a good response. Two of four patients with no maintenance treatment remained in remission, and four could not be contacted.

There were no major complications; three patients complained of pain (two PST-positive) and two withdrew from the study. Three developed a UTI (two PST-positive) and one had mild haematuria (PST-positive), which resolved spontaneously.


There is no specific treatment for interstitial cystitis because of the uncertain cause and pathogenesis. Instead, several treatments, including tricyclic antidepressants, anti-inflammatory drugs, glycosaminoglycan-replacement therapy, intravesical instillations of dimethyl sulphoxide and BCG have been used, with variable results. Most patients are treated empirically and the outcome at best is uncertain.

Parsons et al.[10] challenged a group of patients with interstitial cystitis, and normal controls, with intravesical potassium solutions. They reported greater sensitivity to potassium in patients with interstitial cystitis than in controls, indicating increased bladder epithelial permeability to potassium in these patients. Furthermore, potassium levels in solutions retrieved after bladder instillation with potassium solution were lower in controls pretreated intravesically with protamine sulphate, which temporarily damages bladder epithelium. The obvious implication is that the PST can identify a subset of patients with interstitial cystitis who have predominantly an abnormality of the bladder epithelium. The epithelial abnormality might be caused by defective glycosaminoglycan, which forms a barrier to prevent urine and bladder contents leaking into the urothelium and damaging underlying muscle and nerves [4,16]. Heparinoid therapy either in the form of oral pentosan polysulphate or intravesical sodium hyaluronate is thought to coat bladder epithelium, thereby restoring epithelial impermeability and reducing the symptoms of interstitial cystitis [16]. However, the results reported from treatment with heparinoids are less than satisfactory. In a recent study, only 11% of patients remained on pentosan polysulphate, with most patients withdrawing because there was no benefit [17]. The results with intravesical sodium hyaluronate are also unpredictable. Morales et al.[12] studied the response of weekly intravesical hyaluronic acid in 25 patients with interstitial cystitis. They reported an encouraging 56% (complete plus partial) response rate at 4 weeks and 71% at week 12. However, in a more recent study by Porru et al.[13] only 30% of patients improved with intravesical sodium hyaluronate.

The present study shows that significantly more patients with a positive PST responded to treatment with intravesical sodium hyaluronate than those in whom the PST was negative. Furthermore, the magnitude of change was much greater in the positive than the negative group. The most likely explanation is that a positive PST identified a subset of patients with interstitial cystitis with predominant epithelial abnormalities, who were more likely to respond to glycosaminoglycan-replacement therapy. In patients with a negative test probably other causes rather than epithelial dysfunction were responsible for the interstitial cystitis symptoms. We think that the present results are more robust than those from previous reports of the PST, as we increased the number of test solutions from two to three, diminishing the possibility of the patient identifying the potassium solutions by chance.

In the only other study relating the result of the PST to treatment outcome, Teichman and Nielson-Omeis [18] retrospectively reviewed 38 patients with interstitial cystitis, who were treated with either intravesical heparin or oral pentosan polysulphate and antidepressants. They reported significant improvements in pain score, frequency and nocturia in patients with a positive rather than a negative test. The main drawback of that study was that it was retrospective, and all but one patient had antidepressants in addition to epithelial substitution treatment.

In the present study, 20 of 36 (55%) patients responded to the treatment, which is similar to the response in previous studies. However, stratifying patients by the PST result gave a response rate of 74% in patients with a positive test, vs 22% in those with a negative or indeterminate PST (P = 0.03). Instead of assessing individual symptoms, we used the CSI and PI before and after treatment to assess the severity of the disease [6]. The advantage of this index is that it considers not only the severity of symptoms (CSI) but also the degree to which patients are bothered by them (PI); this may be a crucial factor when assessing therapy.

The assessment after therapy showed an improvement from baseline in the CSI and PI in both PST-positive and -negative patients. However, the improvement in the PI in PST-negative patients was minimal, and the improvement in both indices was significantly better in patients with a positive than in those with a negative PST. None of the patients was taking antidepressants or any other specific treatment for interstitial cystitis. Therefore any improvements in symptoms can be attributed to sodium hyaluronate instillation, although a spontaneous improvement in symptoms remains an unlikely possibility. This clearly implies that intravesical sodium hyaluronate is much more likely to be clinically effective in patients with a positive PST.

The role of cystoscopy and hydrodistension for diagnosing and managing interstitial cystitis is not resolved. Most studies suggest that hydrostatic distension does not provide consistent or reliable diagnostic information [19,20]. The primary aim of the present study was not to assess the significance of cystoscopy in predicting the outcome of treatment. Nevertheless, we retrospectively analysed the data to determine whether cystoscopic findings could have been a useful predictor of the response. Cystoscopy before treatment showed typical changes of interstitial cystitis in 11 of 19 patients with a positive test compared to seven of 12 with a negative test (P = 0.9). On reviewing the treatment response, 11 of 18 patients who reported an improvement and seven of 12 with no improvement showed typical changes of interstitial cystitis on cystoscopy before treatment. Thus cystoscopy findings would not have predicted the outcome of treatment in this study.

The present study shows that glycosaminoglycan-replacement therapy can be selectively offered to patients based on the result of their PST. The current cost of one intravesical instillation of sodium hyaluronate is UK £120 excluding the nursing and day-surgical unit costs. Patients also face the inconvenience of weekly visits to hospital, and loss of working time and earnings. A month's supply of pentosan polysulphate costs UK £75–150 and it takes at least 4 months to assess whether the patient is responding to the treatment. Therefore, it is sensible to treat patients selectively on the basis of a PST; not only will it be financially prudent, but it would also reduce the number of ineffective treatments. However, the treatment should not be completely withheld from patients with negative PST, as some patients would benefit from this despite a negative test, as in the present study. We suggest that patients with intractable symptoms can be offered this treatment as long as they understand that the chances of a successful outcome are much lower.

One important limitation of the present study is that few of the patients opted for maintenance therapy, so we cannot comment on the long-term result of the treatment. The main reason for noncompliance was the reluctance of some patients to continue intravesical instillations once the symptoms had improved. Two of the patients who stopped after six treatments requested that the treatment was reinstated after 3 months, when their symptoms began to deteriorate.

In conclusion, patients with interstitial cystitis and a positive PST have a significantly greater chance of improving with glycosaminoglycan-substitution treatment. The PST may identify a subset of patients with interstitial cystitis with predominant epithelial dysfunction and who are more likely to respond to this particular treatment. We recommend using the PST before initiating glycosaminoglycan-substitution treatment.


We are grateful to Mr Richard Stephenson, Mr Paul Kutarski and Mr Andrew Cliff, consultant urologists at Arrowe Park Hospital, for allowing their patients to be included in the study. We also thank Miss Helen Wong, statistician and clinical effectiveness co-ordinator at Clatterbridge Hospital, for help with the statistical methods used.


None declared.