Overview of Bladder Cancer Surveillance
Most cases of noninvasive bladder cancer remain clinically indolent with even a modicum of urologic intervention. Some cases progress, however, and most patients experience recurrence at least once in the follow-up period, creating the need for accurate tumor surveillance.
The American Urological Association recommends surveillance every 3-6 months for 3 years and at least yearly thereafter.  Similarly, the US National Comprehensive Cancer Network recommends cystoscopy and urinary cytology every 3-6 months for 2 years and then at increasing intervals as appropriate. 
The recommendation for cystoscopic tumor surveillance every 3 months dates back to at least 1936. Bladder cancer surveillance standards arose based primarily on expert opinion rather than as an evidence-based standard.
The origin of the traditional timing (ie, beginning 3 months following bladder tumor removal) is not clear. One theory is that it may have arisen as the timeframe believed to be required for healing after tumor resection. Cystoscopic findings prior to this time would have been difficult to interpret because of numerous factors, including incomplete healing and edema, among others.
Patients who have undergone radical cystectomy require routine surveillance to monitor for local recurrence or the development of metastatic disease. Abdominal and pelvic CT scanning and chest radiography should be performed annually. Some patients with more adverse pathology at the time of cystectomy (eg, locally advanced disease, lymph node metastases) may require more frequent imaging.
The retained male urethra is at risk for cancer recurrence after radical cystoprostatectomy. Urethral recurrence occurs in approximately 7% of patients after cystoprostatectomy.
Cancer involving the prostate (urothelium or stroma) at the time of cystoprostatectomy is the most significant risk factor for urethral recurrence. Monitoring the retained urethra has historically included periodic urethral cytology with subsequent biopsy, if indicated. However, some small studies have suggested monitoring with urethral washings does not confer a survival benefit. 
Gross hematuria or bloody urethral discharge requires immediate workup. A positive urethral cytology or biopsy finding warrants immediate urethrectomy.
Dipstick Studies and Urine Microscopy
Dipstick and microscopic examination of the urine allows detection of hematuria or infection. Hematuria suggests the likelihood of bladder cancer recurrence.
Infection should delay instrumentation because of the risk of sepsis and because of the concern that inflammatory changes might further complicate evaluation of the urothelium.
According to one review, the value of conventional cytology appears to have diminished in the last decade. The cause for this decline is unclear and does not appear to be a change in the criteria for determining malignancy through cytological examination, because specificity did not change appreciably during the interval. The authors theorized that a decrease in the specialization of cytopathologists reading urine specimens might be responsible.
Irrespective of the cause, the sensitivity of cytology in the published literature has diminished, as is shown in the data reported by Halling and associates in their review of the literature. According to the authors, the sensitivity of conventional cytology for grade 1 tumors was 37% before 1990 but only 11% afterward. Similarly, the sensitivity for grade 2 tumors fell from 75% to 31%. 
Of concern is that although it is widely believed that cytology might miss low-grade tumors but is the criterion standard for high-grade tumors, this sensitivity has apparently changed. Based on the Halling et al review of the literature, cytology found 94% of grade 3 tumors in the earlier era, but after 1990, cytology reportedly detected only 60% of even high-grade tumors. 
Fortunately, although the sensitivity of cytology has clearly declined, the specificity remains high and approaches 100%. Therefore, a positive cytology result should be regarded as a true positive; aggressive investigation for occult disease in both the lower and upper urinary tract should ensue.
A host of newer tumor markers and molecular diagnostic indicators has been used, as follows:
DNA ploidy and image analysis
Chromosomal aneuploidy or polysomy
Bladder tumor–associated antigen/analytes
Nuclear matrix proteins 
Monoclonal antibodies (ImmunoCyt)
Hyaluronic acid and hyaluronidase
Fibrin/fibrinogen degradation product
Phenotypic antigens, including Lewis X, M344, DD23, and T138Ag
Growth factors and receptors, including epidermal growth factor receptor, autocrine motility factor, basic fibroblast growth factor
Oncogene and tumor suppressor genes
Molecular cytology (fluorescence in situ hybridization [FISH])
A recent review reported the following sensitivities and specificities for commonly available tumor markers  :
BTAstat - sensitivity 57-83%, specificity 46-73%
Nuclear matrix proteins - sensitivity 47-100%, specificity 60-70%
ImmunoCyt - sensitivity 86%, specificity 79%
Accu-Dx - sensitivity 52-81%, specificity 75-90%
Fluorescence in situ hybridization
Detection of specific DNA alterations known to be associated with bladder cancer is possible using multitarget FISH. DNA probes (stains) hybridize with abnormal chromosomal sites and may be visualized using fluorescence microscopy. According to several published articles, FISH has significantly greater sensitivity than conventional cytology while maintaining the known high specificity of cytology.
The DNA probes chosen for available FISH testing are based on the highest-yielding combination of chromosomal abnormalities. Three of these are centromeric enumeration probes, which allow rapid determination of aneuploidy of chromosomes 3, 7, and 17, the most commonly related to bladder cancer. The fourth probe is used to label the 9p21 locus, known to be the site of a significant tumor suppressor gene. Loss of this tumor suppressor gene is also related to cancer recurrence and progression. A positive result is defined as a gain of 2 or more chromosomes (3,7, or 17) in 4 cells, isolated loss of 9p21 in 12 cells, or isolated gains of only 1 chromosome in 10% or more cells. Most FISH-positive patients develop recurrent urothelial carcinoma within 1 year.
Among patients with bladder cancer in whom cytology results were negative, atypical, and suggestive, FISH detected 60%, 89%, and 100%, respectively, allowing identification of cancer in most patients in whom cytology failed to detect cancer recurrence.
Data from the Mayo Clinic indicate that a conversion of FISH results from positive to negative during bacillus Calmette-Guérin (BCG) therapy indicates a response, whereas a failure to convert indicates a high likelihood of cancer persistence following BCG therapy. 
Instilling 5-aminolevulinic acid into the bladder a few hours prior to cystoscopy allows accumulation in malignant sites often not visible during white-light (normal) cystoscopy. When illuminated with a light ranging from 375-445 nm, up to one fourth of small malignant areas may be missed during routine cystoscopy. This has been proven by biopsy studies.
Of concern is that these areas may be more likely to harbor higher-grade cancer than those identified otherwise. Carcinoma in situ (CIS) is missed by conventional cystoscopy 22% of the time.
Halling et al similarly found a significant number of cancers that cystoscopy failed to definitively identify.  This draws into question whether early recurrences are truly recurrent cancer or simply incompletely removed cancer.
Transitional cell carcinoma (TCC) is a field change disease, meaning that the entire field of transitional cells is prone to the DNA changes leading to cancer. Therefore, the entire urothelium should be monitored, especially in higher-risk individuals. 
The frequency of upper tract imaging for post-treatment surveillance is still not standardized. However, many authors recommend annual urography, especially in high-risk patients.
Patients at high risk should undergo intermittent evaluations during the surveillance period. Patients with positive urine cytology results or positive findings from FISH or other bladder tumor indicators who have no tumor identified in the bladder to explain the positive test result should undergo repeat upper tract evaluation.
Excretory urography (ie, intravenous pyelography [IVP]; now limited in availability in many institutions, having been replaced with newer CT imaging techniques), retrograde pyelography, and ureterorenoscopy can help detect upper tract synchronous or metachronous tumors. The accuracy of these tests increases in the order listed, with radiographic imaging missing up to three fourths of small upper tract tumors in some series if read by a radiologist instead of the urologist present in the operating room during retrograde contrast injection.
Office-based cystourethroscopy has a role in some patients with upper tract TCC treated with nephron-sparing surgery.
CT urography using digital reconstruction of CT images to create a view of the ureters potentially can blend the advantages of CT (ie, speed, visualization of renal parenchyma, visualization of nonurological structures) with the advantages of excretory or retrograde urography (ie, visualization of the upper tract lumen). An An increasing number of referral centers have the ability to perform this modality.
Some centers also use CT intravenous pyelography. Kidney-ureter-bladder (KUB) imaging is performed, followed by a noncontrast CT scan of the abdomen and pelvis. Contrast is then injected intravenously, followed by vascular and excretion-phase CT abdominopelvic imaging. The patient then undergoes anteroposterior and oblique abdominal radiography and postvoid radiography.
Complications of Cystoscopy
Complications of surveillance cystoscopy are as follows:
Urinary tract infection
In a patient with spinal cord injury, autonomic dysreflexia can occur as a response to bladder distention, leading to potentially life-threatening hypertension. This may be prevented with nifedipine or terazosin in some cases, but careful monitoring is requisite. Any signs of headache, tremors, or hypertension should lead to immediate cessation of the procedure and emptying of the distended bladder. For this reason, most patients with spinal cord injury should probably undergo cystoscopy in the operating room with anesthesia monitoring.
Surveillance in Patients with In Situ Carcinoma
Following successful treatment of initial or recurrent carcinoma in situ (CIS) with intravesical bacillus Calmette-Guérin (BCG) vaccine, chemotherapy, or both, patients are monitored at regular intervals with cystoscopy and urine cytology, usually every 3 months for the first 1-2 years and every 6 months thereafter. Intravenous pyelography (IVP) is also usually performed every 6-12 months. This follow-up continues for a minimum of 5 years.
Following cystectomy and urinary diversion, cytology is performed every 3 months for the first 1-2 years and every 6 months thereafter. In patients with a urostomy or continent diversion, a catheterized specimen is obtained from the stoma, not the urostomy bag, because contact with the bag and the stagnant nature of urine in the bag confounds the cytologic findings.
If the patient did not undergo a urethrectomy, a urethral wash for cytology should also be performed on this schedule. Patients with an intestinal neobladder should provide a voided specimen. IVP or loopography should be performed every 6-12 months to evaluate the upper urinary tract, as should CT scanning, chest radiography, and serum chemistry studies to rule out metastatic disease. After 2-3 years, a vitamin B-12 level should be checked because many of these patients develop deficiency of this vitamin.
Surveillance After Radical Cystectomy
A group from the University of Texas MD Anderson Cancer Center retrospectively reviewed their post–radical cystectomy surveillance protocol for 382 patients and concluded that a stage-specific approach was most appropriate.  With a median follow-up of 38 months, 25% of patients experienced recurrences, with a median time to recurrence of 12 months. The 4 most common sites of recurrence (in decreasing order of incidence) were the lung, pelvis, bone, and liver. Seventy-four percent of recurrences were asymptomatic, and 43 of the 72 asymptomatic recurrences were detected with chest radiography or liver function serum tests.
Only 5% of patients with pT1 disease had subsequent metastases, and all were identified with chest radiography or liver function tests. Among 10 patients who were found to have asymptomatic intra-abdominal recurrences based on CT scan findings, 9 had pT3 disease. Patients with pT2 and pT3 disease had recurrence rates of 20% and 40%, respectively. All recurrences in patients with pT2 or pT3 disease occurred within 24 months.
Based on these findings, the group recommended that surveillance should include the following:
Liver function tests
Alkaline phosphatase assessment
The group recommended scheduling surveillance according to the patient’s disease stage, as follows:
pT1 disease - Annually
pT2 disease - Every 6 months for 3 years, then annually
pT3 disease - As with pT2 disease, but starting at 3 months, with CT scanning at 6, 12, and 24 months
All patients with transitional cell carcinoma (TCC), particularly those at higher risk of recurrence (eg, because of distal ureteral involvement at cystectomy, multiple recurrent bladder tumors, carcinoma in situ [CIS]), should undergo upper tract radiographic studies every 1-2 years.
Adherence to Surveillance
Only 40% of patients adhere to the recommended schedule of bladder cancer surveillance.  Failure to undergo standard surveillance has been attributed to numerous issues. Advanced age and lower-risk tumors are associated with a failure to follow guidelines, as are lower economic status and urban dwelling.
Costs of Surveillance
Although the incidence of bladder cancer is less than that of prostate cancer, expenditures are almost twice as high for bladder cancer because of its chronic nature and the need for long-term surveillance. According to the US Agency for Health Research and Quality, annual expenditures are $2.2 billion for bladder cancer versus $1.4 billion for prostate cancer. This suggests a close assessment of surveillance techniques and standards is appropriate.
Artificial neural networks using various tumor markers similar to those described by Parekattil et al may be more cost effective for detecting recurrence and progression than the current screening protocol of cystoscopy and conventional cytology at predetermined intervals.  Using more effective markers may allow scheduling of cystoscopy on a more logical and targeted schedule than is currently the default.
Nonlinear surveillance strategies have been shown in at least one model to actually decrease the time to detection of tumor recurrence while optimizing the utilization of resources.  In nonlinear surveillance, the follow-up schedule is determined by patients’ individual risk profiles, with longer intervals between visits for low-risk patients and shorter intervals for high-risk patients. The risk profile is based on the patient's prior recurrence rate and tumor stage and grade.
A fresh look at surveillance strategies is in order based on the lack of evidence on which current standards are based and on new findings regarding the ability to predict cancer recurrence using neural networks, prediction models, and improved diagnostic tests, including molecular diagnostic evaluations. As this process progresses, the conventional surveillance protocol will likely change. Using tumor stage and grade in conjunction with improved surveillance methods, resources may be focused on patients at risk of recurrence and progression.
Specific concepts likely to help focus surveillance in the near future include improved endoscopic techniques that can identify otherwise imperceptible malignancy and molecular diagnostic tests that can identify malignant change prior to anatomic transformation. However, until such time, cystoscopy in conjunction with some form of cytology (either conventional or molecular cytology, ie, FISH) is likely to remain the mainstay in surveillance at currently accepted intervals.