Correspondence: Dominik Abt M.D., Department of Urology, St. Gallen Cantonal Hospital, Rorschacherstrasse 95, St. Gallen 9007, Switzerland. Email: firstname.lastname@example.org
Intractable hematuria is a common and severe complication in patients with inoperable bladder carcinoma. The aim was to provide an overview of therapeutic options for such cases, and analyze their effectiveness and risk profile, so a systematic literature search of peer-reviewed papers published up to September 2012 was carried out. Various options are available to treat hematuria in patients with inoperable bladder cancer; these include orally administered epsilon-aminocaproic acid, intravesical formalin, alum or prostaglandin irrigation, hydrostatic pressure, urinary diversion, radiotherapy, embolization and intraarterial mitoxantrone perfusion. These treatment options are associated with different prospects of success, risks and side-effects. Well-designed and large studies comparing options are completely lacking. Despite various treatment options, management of intractable hematuria in patients with inoperable bladder cancer remains a challenge, and most of the reported methods should be seen as experimental. Interventional radiology and alum instillation seem to be suitable alternative options for patients who, after critical consideration, cannot be treated by irrigation, transurethral resection or palliative cystectomy.
Bladder carcinoma often causes recurrent and intractable hematuria. Reasons for bleeding might be sloughing of tumor mass, side-effects of radiation, cyclophosphamide-induced hemorrhagic cystitis and other sources of hematuria, such as prostate cancer, simultaneous ureteropelvic cancer or severe infection. Treatment is often hindered by drugs that dilute the blood and by comorbidities.
Intractable bleeding in advanced bladder carcinoma is often not curable by irrigation of the bladder by a catheter. If transurethral resection and coagulation are not able to control bleeding, palliative cystectomy with urinary diversion is an effective treatment of choice. However, surgery might not be possible because of comorbidities, or the patient might refuse it. Alternative palliative methods to reduce bleeding and the frequency of transfusion, and alleviate pain are required to improve quality of life in such cases.
The present review explores alternative treatments for patients with bleeding complications of bladder cancer who are not suitable for surgery or who refuse it. Practical experience with the treatments available, and their effectiveness and risk profiles are discussed.
A comprehensive literature search for articles published up to September 2012 was carried out in PubMed using different combinations of the following search terms: “bladder cancer”, “hematuria”, “epsilon-aminocaproic acid”, “formalin”, “alum”, “prostaglandin”, “hydrostatic pressure”, “urinary diversion”, “radiotherapy”, “arterial occlusion”, “embolization”, “mitoxantrone” and “palliative cystectomy”. We reviewed the publications thus selected, as well as literature cited in these publications.
Table 1 gives an overview of the most investigated and most promising treatments revealed by the literature search. It is evident that quite a wide range of methods for the management of intractable hematuria have been used, a fact that underlines the difficulty of finding suitable treatment options for patients not qualifying for or refusing surgery. However, there is a lack of large, well-controlled trials for all treatment options described; studies have generally been limited to descriptions of small numbers of cases, or were of experimental character.
Table 1. Treatment of intractable hematuria
Chemoperfusion with mitoxantrone
It should be noted that hyperbaric oxygen therapy, although used to treat radiation cystitis and cyclophosphamide-induced hemorrhagic cystitis, was not used to treat hematuria caused by bladder carcinoma in any of the articles found. Thus, hyperbaric oxygen therapy is not further considered in the present review.
Orally administered epsilon-aminocaproic acid
EACA is a synthetic lysin that competitively inhibits fibrinolysis induced by plasminogen and plasmin. When given orally, the drug is absorbed rapidly and 80% is secreted unchanged in the urine within 24 h. The literature search found only studies reporting small numbers of cases and one experimental study on the use of EACA. The patient populations in these studies were heterogeneous with regard to the cause of hematuria, which included radiogenic cystitis, infections, hemorrhagic disorders and urothelial cancer. There was no empirical data regarding duration and dosage of EACA treatment.
The use of EACA in hematuria was first described by Vega et al. in a patient with sickle-cell trait in 1971. Stefanini et al. described nine patients with hematuria of various causes treated with approximately 150 mg/kg/day EACA for up to 21 consecutive days. The authors reported that hematuria was controlled effectively in all cases without overt clinical reactions.
Side-effects of EACA were rare, but serious. Thrombotic complications, myopathy, rhabdomyolysis, and renal and hepatic failure were all reported.[4, 5]
Intravesical formalin treatment
Intravesical formalin treatment causes precipitation of cellular proteins of the bladder, and leads to occlusion and fixation of teleangiectatic tissue and small capillaries.[6, 7] Treatment of inoperable carcinoma of the bladder by formalin instillation was first described in 1969 by Brown. In 24 patients with advanced carcinoma suffering from hematuria and strangury, a 10% formalin solution was instilled into the bladder over 15 min. Relief of hematuria was seen in 22 patients within 1–5 days, with a mean duration of 4 months, without general complications; the method was thus deemed by Brown to be safe and effective. Subsequently, the method has been widely used with predominantly good success rates. For example, Fair reported on 14 patients treated with 1% formalin instillation; 10 patients were responsive to the first instillation and a further two to the second instillation. Cessation of hematuria was achieved in the remaining two patients by another treatment with 2% formalin. However, severe side-effects were subsequently described in other studies, commonly leading to the discontinuation of formalin instillation in patients with persistent gross hematuria. In a study of 10 patients, Giannakopoulos et al. reported 40% renal failure, 40% clinically significant reduction of bladder capacity (<100 mL), 30% urinary incontinence, 30% urgency and nocturia, and one case with retroperitoneal fibrosis. A high rate of renal failure (>60%) was also reported by Ferrie et al. According to other reports, formalin treatment caused ureteral stenosis, hydronephrosis and vesico-ureteric reflux. Although the incidence of complications appears to be lower if formalin solutions of ≤4% are used, the effectiveness of this treatment seems to be inferior.[9, 11-13] Furthermore, instillation is painful and requires general or spinal anesthesia, and a catheter has to be left in the bladder after the procedure for bleeding control. There are no studies comparing the effectiveness of formalin instillation to catheterization alone. Formalin treatment is no longer common in everyday clinical practice. Cystoscopy and cystography should always be carried out first to exclude the existence of blood clots or vesico-ureteric reflux. In the latter case, if formalin treatment is carried out, prior insertion of a ureteral occlusion catheter is necessary.[9, 12] Choong et al. provide a recommendation for the careful use of formalin (Table 2).
Table 2. Protocols for treatment of hematuria by intravesical irrigation or instillation
Cystoscopy and cystography are carried out first to exclude the existence of blood clots or vesico-ureteric reflux. Insertion of a ureteral occlusion catheter is carried out in the case of reflux.
As spinal or general anesthesia has to be used, coagulation of major bleeding vessels can be carried out.
External areas of skin and mucosa are protected with Vaseline. Vagina is packed to prevent catheter leakage.
Low initial formalin concentrations (1–2%) should be used.
Irrigation is carried out for 10 min under gravity at <15 cm H2O.
Alternatively instillation is carried out under gravity at <15 cm H2O with catheter left open at a level just above the pubic ramus.
Contact time should be limited to 15 min.
Cystoscopy with evacuation of blood clots is recommended before treatment.
Using a 1% alum solution; 50 g of alum is dissolved in 5 L sterile water and used to irrigate the bladder at 250–300 mL/h.
Using the 1% solution or a stock solution of 400 g of potash of alum (McCarthy's) in 4 L sterile hot water, 300 mL of the stock solution is added to 3 L of 0–9% saline through a sterilizing filter and the bladder is irrigated with up to 30 L of this solution in 24 h.
Cystoscopy, clot evacuation and insertion of a three-way catheter are carried out.
Instillation of the bladder can be carried out using 50 mL of 4–8 mg/L carboprost tromethamine for 1 h. Bladder is drained and another 50 mL are maintained for 1 h. After draining the bladder it is irrigated with normal saline now. This can be repeated up to four times a day with a 24 h course consisting of 400 mL carboprost tromethamine administered within 8 h.
Dosage can be increased to 10 mg/L if there is no improvement.
Alternatively continuous irrigation of the bladder can be carried out using 8–10 mg/L carboprost tromethamine at 100 mL/h for 10 h.
Alternative approaches to achieve intravesical hemostasis were investigated as a result of the high complication rate associated with formalin treatment. The results of alum irrigation treatment were first published by Ostroff and Chenault in 1982. Alum (either aluminum ammonium sulfate or aluminum potassium sulfate), like formalin, also causes protein precipitation. However, alum has lower cell penetrability than formalin, so that its effect is limited to the cell surface and interstitial spaces, and cells remain viable. Systemic absorption of alum is also lower.[16, 17] Alum excretion proceeds through a renal route, and increased serum levels can result in prolonged prothrombin times.
Ostroff and Chenault reported complete cessation of hematuria over various periods using a 1% alum solution in six patients with massive bladder hemorrhage (2 due to carcinoma). No side-effects were observed, and treatment could be given without anesthesia. After this initial publication, several studies on small numbers of patients were reported (Table 3),[19-22] with success rates between 66% and 100% using 1% alum solutions. Unfortunately, no hard definitions of success and relapse were provided, except by Arrizabalaga et al. and Goswami et al. Also, no comparisons to standard bladder irrigation alone were made in any of the aforementioned studies. Schootstra et al. reported satisfactory results after treating 16 patients with 0.5% alum solution (full text not available). The lack of well-designed studies in adequate numbers of patients hampers drawing overall conclusions on all this data.
Table 3. Cases treated with intravesical 1% alum irrigation
Suprapubic pain in all patients leading to one dropout; 2 patients with vesical tenesmus; 3 with spasms leading to pericatheter leak; mild pyrexia in 4 patients
Irrigation rate 3–10 mL/min; irrigation stopped 6 h after cessation of bleeding; clear definition of response; significant change in serum aluminum levels and thromboplastin time after a period of 24 h, not associated clinical side effects; average volume of irrigation fluid: 10.5 L
Bladder spasms and suprapubic pain are common during alum treatment. These symptoms are obviously caused by the acidity of the alum solution, but can be effectively managed by antispasmodics. Serious side-effects of alum irrigation have only been reported in isolated cases. Encephalopathy and acute aluminum intoxication occurred predominantly in patients with renal dysfunction.[24-26] Although renal dysfunction is not a clear contraindication, Shoskes et al. reported the death of a patient with renal dysfunction who was treated twice within a 3-month period with 1% alum given over 48 h.
To avoid systemic side-effects, serum aluminum can be monitored during treatment. Signs of clinical toxicity seem to occur at a mean serum aluminum concentration of 7.4 μmol/L, and surveillance of patients is recommended when concentrations exceed 3.7 μmol/L.[29, 30] Goswami et al. found only a moderate increase of serum aluminum (from 1.68 μmol/L at baseline to 3.36 μmol/L on treatment) in a prospective study on 12 patients with normal renal function and without clinical evidence of aluminum toxicity.
On balance, a 1% alum solution can be considered to be a treatment option, at least in patients without renal dysfunction. Measurement of serum aluminum would not appear to be necessary provided usual concentrations and amounts of alum irrigation are used. A recommendation for the “ideal” application is given in the review by Choong et al. (Table 2). All authors recommend cystoscopy with evacuation of blood clots before treatment.
Intravesical instillation of prostaglandin
Prostaglandins can cause constriction of vascular smooth muscle cells and aggregation of platelets. Only one study compared the effect of intravesical instillation of PGF2 alpha with alum on hematuria caused by bladder cancer: Praveen et al. treated 10 patients with a dose of 1 mg PGF2 alpha daily for a maximum of 5 days. Six patients had complete cessation of macroscopic hematuria; two patients had partial control. Failure of control was seen in two patients. There was no significant advantage in efficacy or safety of PGF2 treatment compared with alum. In both groups, patients had only local side-effects (bladder spasms, catheter blockage). The authors concluded that high cost, low availability and stringent storage conditions are drawbacks for routine use of PGF2 alpha and recommended it only as an alternative method where alum treatment is not successful. A comparison of these treatments to bladder irrigation with normal saline was not carried out. Several studies have been carried out on the use of prostaglandin for the treatment of hemorrhagic cystitis caused by radiation, cyclophosphamide and bone marrow transplantation. Prostaglandin was effective in these studies, although the detailed mechanism of prostaglandin action on the bladder epithelium remains unclear. A recommendation for the use of prostaglandin also can be found in Table 2.
Hydrostatic bladder distention
Intravesical hydrostatic pressure for the treatment of bladder carcinoma was first described by Helmstein. In 1972, he reported on 43 patients thus treated to produce necrosis of the tumor. Holstein et al. were the first to publish a study using this method for the control of bladder hemorrhage in 1973. Six patients were investigated. Under epidural anesthesia, a balloon was inserted into the bladder, which was filled with fluid for 6 h to the level of the systolic blood pressure. Sustained hemostasis was achieved in three patients; two patients showed temporary improvement and one patient did not respond. In further studies, similar effectiveness was achieved using intravesical hydrostatic pressures of 10–20 cm H2O above diastolic blood pressure, accompanied by a reduced risk of side-effects as compared with those seen at higher pressures.[33, 34] In a study on 49 patients treated for urothelial carcinoma using hydrostatic bladder distention, England et al. reported a 100% success rate of hemostasis in nine patients with gross hematuria. Likewise, Hammonds et al. reported successful hemostasis in eight out of eight patients treated with pressures of 115–145 cm H2O for 6 h. Over a decade later, in 1986, Antonsen et al. studied the effect of hydrostatic bladder distention on both hematuria and tumor necrosis in 18 patients with severe bleeding. A total of 12 patients showed cessation of hematuria within 1–5 days after treatment, with a recurrence of bleeding after a mean of 3 months. Despite these good response rates, no further publications on the Helmstein bladder distention technique were found by the literature search. In the initial study, Helmstein observed two cases of bladder rupture and one perforation through an ulcerating tumor. Bladder rupture was only a very rare complication when pressure was limited to 10–25 cm H2O above diastolic blood pressure. Abdominal pain, nausea, temporary incontinence and pyrexia were reported more commonly.[34, 36, 37]
Information on hydrostatic bladder distention is unfortunately limited to these few studies. In all cases, catheters for bladder irrigation were inserted after the procedure until the urine became clear; no studies compared the success rate of hydrostatic bladder distention followed by bladder irrigation with bladder irrigation alone. Bearing these limitations in mind, and in view of the need for anesthesia when using this technique, hydrostatic pressure treatment of intractable hematuria is of doubtful value.
In their 2009 review, Ghahestani et al. suggested that urinary diversion might be a treatment option for intractable hematuria caused by tumor sloughing. By excluding the bladder from the urinary tract, internal blood tamponade can be carried out. Contact of the bladder mucosa with urine urokinase, with the enhanced risk of bleeding, is also prevented. This procedure might also make the bladder accessible for other treatment options, such as formalin instillation, with a reduced risk.
Only one study was found by the literature search where urinary diversion was used: Pomer et al. treated 16 patients with severe intractable hemorrhagic cystitis after radiotherapy (2 of them with bleeding tumors) by cutaneous ureterostomy. A total of 11 patients remained hemorrhage-free and three had only slight intermittent hematuria. No cases of relapse requiring treatment were observed during follow up.
No literature was found where the outcome of urinary diversion without cystectomy (e.g. ileal conduit, pouch, ureterocutaneostomy) was described for cases of intractable hematuria. The option of urinary diversion alone in high-risk cases or patients refusing cystectomy is thus debatable, but retention of large tumor masses in the absence of cystectomy might also pose a risk of further complications. Palliative cystectomy with consequent urinary diversion might be a highly invasive intervention, but it is considerably easier and faster than curative cystectomy.
Irradiation has been examined for the control of hematuria in patients with bladder carcinoma since the late 1960s. In 1979, Chan et al. reported on seven patients, five with intractable hemorrhage and two with ureteral obstruction, whose medical conditions prevented surgery. Using up to three doses of 10 Gy with 12 × 12 cm portals, they found prompt cessation of bleeding in all cases for 2–8 months (mean 5 months). Side-effects were mild (diarrhea, nausea and vomiting for 3–4 days after therapy), and there were no long-term complications. Fossa and Hosbach treated 39 patients with advanced bladder carcinoma with radiotherapy (3 Gy × 10 over 2 weeks) palliatively, and reported a marked improvement of hematuria.
In 1999, Jose et al. treated 65 patients with weekly 6-Gy fractions up to a total of 30–36 Gy while investigating a low toxicity regimen for palliative bladder radiotherapy in patients with poor performance status. There was good tumor response and local control, but just seven of 14 patients with hematuria were controlled by the treatment. The main acute toxicity was urinary frequency (up to hourly at the peak of the reaction) in seven patients, and urinary obstruction in one. Late toxicity amongst the 16 patients who were evaluable after 1 year included four patients with persistent frequency, one with severe hematuria and one with a bladder capacity <100 mL. One patient experienced late bowel and bladder morbidity.
Information on the use of low-dose radiotherapy for the treatment of intractable hematuria is restricted to these few studies. The rate of success is unclear: while tumor control should lead to control of hematuria, the latter can also be a side-effect of the irradiation itself. Additionally, the method seems too elaborate to come into routine use.
In 1974, Hald and Mygind were the first to report on unilateral hypogastric artery embolization for the treatment of intractable hematuria in a patient with bladder carcinoma. They occluded the hypogastric artery and stopped bleeding using pieces of the patient's vastus lateralis muscle. They reported disappearance of the tumor at cystoscopy and only moderate side-effects (tenderness and induration of the gluteal region and the back of the thigh, and transient urgency). Subsequent studies using embolization were reviewed by McIvor in 1982, who described 35 patients with an overall response rate of 92% regardless of time to relapse. In 1988 Appleton et al. reported results on eight patients with bladder hemorrhage as a result of carcinoma who were treated with selective embolization. Effective control was achieved in six patients and partial control in two patients, without clinically relevant side-effects.
Selective embolization now appears to be a widely used method of treating persistent hematuria caused by carcinomas of the bladder. Its popularity might be a result of the perceived low risk of serious side-effects, although information on the frequency of complications is lacking. There have only been a few reports of severe complications: Hietala and Sieber reported necrosis of the bladder after embolization of an internal iliac artery in patients with severe pelvic trauma, but similar complications were not seen after treatment of hematuria. The risk of gluteal necrosis, particularly known by patients undergoing embolization for pelvic trauma, can be avoided by selective embolization of the anterior trunk of the internal iliac artery, sparing the gluteal branches.[49-51] The most common problems are pain in the gluteal region and the back of the thigh, as well as transient voiding dysfunction, as initially reported by Hald and Mygind. The complication rate appears to have been further reduced by selective and super-selective embolization brought about by technical improvement of catheters and occlusion.[46, 52-54] Randomized or prospective studies have still to be carried out, however.
Intra-arterial chemoperfusion with mitoxantrone
Mitoxantrone is a chemotherapeutic agent that causes DNA cross-linking and blocking of the cell cycle. Mitoxantrone perfusion for the treatment of intractable hematuria was described by Textor et al. in 2000. In a non-randomized clinical trial, 15 patients treated with standard arterial embolization were compared with 15 patients treated by selective and super-selective perfusion of the tumor-supplying vessels with 20 mg/m2 body surface area mitoxantrone over 1–2 h, with a mean of two applications each. Complete control of the hemorrhage was achieved in 12/15 of the patients with embolization and 14/15 of the patients with chemoperfusion. Hemorrhage stopped within 24 h in embolization patients and after 4–15 days (mean 10 days) in chemoperfusion patients. The authors reported recurrence of hemorrhage in 4/13 embolization patients and 3/14 of chemoperfusion patients. Major complications were only observed in embolized patients (7/22). Side-effects of mitoxantrone were temporary nausea, vomiting and pyrexia. The authors described a significantly reduced occurrence of post-therapeutic pain as a major advantage of mitoxantrone chemoperfusion (6/31 compared with 20/22 in the embolization group) and concluded that intra-arterial chemoperfusion using mitoxantrone is an effective therapy in patients with intractable gross hematuria. However, because of the delayed effect of chemoperfusion, they recommended standard embolization therapy in patients with life-threatening bleeding, which might have led to selection bias in this non-randomized study. Mitoxantrone chemoperfusion seems to be a valuable therapeutic option, and it is therefore all the more surprising that the literature search did not find any other studies evaluating this method for treatment of hemostasis.
Alternative techniques for hemostasis in advanced bladder cancer have been poorly studied in patients who cannot be treated with the well-established methods of irrigation, endoscopic coagulation and cystectomy. The reasons might lie in the sporadic use of these treatments and the short life expectancy of the patients. Even so, it is remarkable that all nine methods reviewed above have not been systematically investigated in studies comparing them with the efficacy and safety of the well-established approaches, in particular with irrigation alone.
Comparison of the success rates of the different published studies is limited by the lack of generally accepted definitions of success and relapse, and the difficulty in assessing the severity and cause of bleeding. Table 4 attempts to summarize published response rates, treatment periods and complications.
Table 4. Summary of therapeutic options for intractable hematuria
Reported response rate (%)
Reported duration of treatment
Complications (incidence and severity combined)
Depending on method
Up to several weeks
Alternative methods for hemostasis in intractable hematuria are almost always described as options for patients where anesthesia and surgery pose a high risk. However, invasiveness, complexity and risks of some of the alternative approaches do not always seem to be lower. Thus, intravesical formalin treatment and hydrostatic bladder distention cannot be carried out without anesthesia, are time-consuming, must frequently be carried out more than once and carry substantial risks. For these reasons, they are rarely used in routine clinical practice.
The use of orally administered EACA appears at first sight to be practicable and simple. However, its effectiveness in treating recurrent hematuria is uncertain based on the results of the available studies, and its possible risks should not be underestimated in view of the fact that the relevant patient population is very susceptible both to thromboembolic events and to renal complications. Thus, screening for risk factors and seeking advice from a hematologist are necessary for every patient for whom administration of EACA is considered.
Hypofractionated radiotherapy has been reported by some authors to result in almost complete cessation of hematuria caused by bladder carcinoma. Critical examination of the literature showed, however, that several applications are usually necessary, and that treatment might be required for up to several weeks. This might be suitable for treatment of subacute and recurrent hematuria, and prophylaxis of bleeding, but would not seem appropriate for cases of acute bleeding, or for patients receiving palliative cancer therapy.
Alum instillation seems to be a promising therapeutic option for treatment of intractable hematuria. Alum can be easily instilled without anesthesia and can thus be used to supplement bladder irrigation, which is carried out routinely in almost every case of gross hematuria. The success of alum instillation appears very promising and is associated with low risk, except for patients with renal dysfunction, where caution is required. In contrast, prostaglandin instillation for the treatment of hematuria caused by bladder carcinoma has hardly been investigated. However, because of its advantageous risk profile, prostaglandin instillation might be considered as an alternative for alum where response is inadequate or in patients with renal dysfunction.
In the case of inoperable patients with acute or hyperacute hematuria, the possibility of interventional radiology seems to be a very interesting option. No anesthesia is required, and cessation of bleeding occurs quickly and can be maintained. Although embolization has been reported to have serious side-effects, the complication rate (particularly post-interventional pain) appears to have been reduced by selective and super-selective embolization brought about by technical improvement of catheters and occlusion. Intra-arterial chemoperfusion with mitoxantrone might be a promising variant of conventional embolization, although it has been insufficiently studied. Based on the results to date, it would not appear to be the method of choice in hyperacute bleeding because of the greater time required to achieve hemostasis as compared with embolization. However, there are fewer side-effects associated with this treatment.
The present discussion shows that relevant alternative treatment options for intractable hematuria are available. However, the effectiveness and risk profiles of these methods remain uncertain because of the lack of information. Thus, standard treatment procedures for recurring or intractable gross hematuria always have to be considered during the process of treatment. Patients initially refusing surgery might reconsider this option in the face of prolonged hospitalization or repeated, painful bladder tamponade. Patients initially classified as inoperable because of their poor general state condition might be optimized by adequate medical therapy and prepared for surgery. Thus, our proposed decision tree (Fig. 1) for the treatment of hematuria in advanced bladder cancer also includes standard treatments.
In summary, treatment of intractable hematuria in patients with inoperable bladder carcinoma is still a challenging and often frustrating matter. We suppose the decision to use alternative treatment options must be suited to each patient, and their particular problems and risks. In our opinion, alum instillation and interventional radiological treatment seem to be the most reasonable solutions where, after critical consideration, the patient cannot be treated by irrigation, transurethral resection or palliative cystectomy.