Spinal Cord (2010) 48, 257–261

& 2010 International Spinal Cord Society All rights reserved 1362-4393/10 $32.00
www.nature.com/sc

ORIGINAL ARTICLE

Bladder cancer in spinal cord injury patients

JF Kalisvaart, HK Katsumi, LD Ronningen and RM Hovey
Department of Urology, University of California, Irvine, CA, USA

Study design: Retrospective review.

Objective: Spinal cord injury is a known risk factor for bladder cancer. The risk of bladder cancer has
been reported at 16–28 times higher than the general population. Earlier studies have identified
indwelling catheters as risk factors. We examined the characteristics of bladder cancers in a spinal cord
injury (SCI) population.
Setting: Long Beach VA Hospital Spinal Cord Injury Unit, Long Beach, California
Methods: We reviewed SCI patients seen and diagnosed with bladder tumors between January 1983
and January 2007. Data collected included time since diagnosis, method of diagnosis, form of bladder
management, pathologic type, treatment of the tumor, and outcome.
Results: A total of 32 patients with bladder cancer were identified out of 1319 seen. Tumors found
were 46.9% squamous cell carcinoma (SCC), 31.3% transitional cell carcinoma (TCC), 9.4%
adenocarcinoma, and 12.5% mixed TCC and SCC. The primary form of bladder management was
44% urethral catheter for a mean of 33.3 years, 48% external catheter for a mean of 37.4 years, and 8%
intermittent catheterization for a mean of 24.5 years. Nineteen patients had a known method of cancer
detection with 42% found on screening cystoscopy.
Conclusions: The pathologic makeup of the tumors is similar to that reported earlier. Over 50% of
patients diagnosed with bladder cancer in our population did not have an indwelling catheter. This
suggests that the neurogenic bladder, not the indwelling catheter, may be the risk factor for bladder
cancer. Urologists should consider diligent, long-term screening of all patients with SCI for bladder
cancer and not just those with indwelling catheters.
Spinal Cord (2010) 48, 257–261; doi:10.1038/sc.2009.118; published online 15 September 2009

Keywords: spinal cord injuries; urinary bladder neoplasms; catheterization

Introduction
Bladder tumors in the spinal cord injury (SCI) patient are a A total of 32 patients with tumors were identified out of
well-recognized phenomenon. Although controlled prospec- approximately 1319 patients seen and followed by urology,
tive studies are lacking, several retrospective studies have and data was collected on each patient including the time
suggested that the relative risk of a bladder neoplasm is since their diagnosis, management of their bladder, the
16–28 times that of a normal population.1–4 SCI patients are number and method of identification of the bladder tumors,
also known to have a different spectrum of disease than the the method of treatment of the tumors, and the end result of
general population with a significantly higher proportion of their disease.
squamous cell carcinoma (SCC) compared with transitional
cell carcinoma (TCC), and a lower age and a higher stage at
diagnosis.5 We examined the risk factors and characteristic of
Results
bladder tumors in an SCI population.
A retrospective review of patients seen at the Spinal Cord
Materials and methods Injury Unit of the Long Beach Veterans Administration
Hospital found 32 patients with bladder tumors out of 1319
All SCI patients at the Long Beach Veterans Administration
patients seen and followed by urology. Of these tumors, 15
diagnosed with bladder tumors between January 1983 and
(46.9%) were SCC, 10 (31.3%) were TCC, 3 (9.4%) were
January 2007 were reviewed.
adenocarcinoma, and 4 (12.5%) had mixed TCC and SCC.
Correspondence: Dr JF Kalisvaart, Department of Pediatric Urology, The tumors were identified for a mean of 34 (range 16–62)
Emory University, 5445 Meridian Mark Road, Suite 420, Atlanta, years after the initial SCI (Table 1).
GA 30342, USA.
Overall, 25 patients with tumors had information available
E-mail: jonkalisvaart@gmail.edu
Received 9 April 2009; revised 3 August 2009; accepted 15 August 2009; regarding their bladder management. Of these, 4 patients
published online 15 September 2009 (16%) managed their bladders with clean intermittent
 Bladder cancer in spinal cord injury patients
JF Kalisvaart et al
258

Table 1 Patient demographics by tumor type

All tumors SCC TCC Adeno TCC + SCC

Total (%) 32 (100) 15 (46.9) 10 (32.3) 3 (9.4) 4 (12.5)

Time in years from injury to diagnosis (range) 34 years (0–58) 44 (14–55) 50 (0–58) 41 (34–53) 45 (33–55)
Underwent cystectomy (%) 27 (84) 14 (93.3) 7 (70) 3 (100) 3 (100)
Died (%) 23 (71.9) 11(73.3) 7 (70) 2 (66.7) 2 (66.7)
Years after diagnosis (range) 3.8 (0–20) 5.1 (0–20) 1.7 (0–5) 5 (2–8) 3 (0–6)
Tobacco use (%) 21 (70) 7 (50) 9 (90) 3 (100) 1 (50)
Pack years (range) 58 (10–180) 44.7 (18–100) 65.6 (10–180) 77 (25–150) 10

Abbreviations: SCC, squamous cell carcinoma; TCC, transitional cell carcinoma.

Table 2 Method of bladder management

All tumors SCC TCC Adeno TCC + SCC

Total 25 9 9 3 1
Foley (%) 14 (56) 2 (22.2) 7 (77.7) 3 (100) 2 (66.7)
Years (range) 26.7 (1–53) 31 (30–32) 21 (1–53) 30 (20–50) 37.5 (20–55)
CIC (%) 4 (16) 1 (11.1) 1 (11.1) 0 (0) 1 (33.3)
Years (range) 17.3 (3–33) 16 17 NA 33
SPT (%) 2 (8) 1 (11.1) 1 (11.1) 0 (0) 0 (0)
Years (range) 2.5 (2–3) 3 2 NA NA
EDD (%) 14 (56) 5 (55.6) 5 (55.6) 2 (66.7) 1 (33.3)
Years (range) 32 (1–55) 42.8 (18–50) 41 (32–50) 10.5 (1–20) 33

Abbreviations: SCC, squamous cell carcinoma; TCC, transitional cell carcinoma.

catheterization for a mean of 17.3 years (range 3–33 years), information was available, one (11.1%) used clean inter-
14 patients (56%) had an indwelling urethral catheter to free mittent catheterization for 16 years, two (22.2%) used an
gravity drainage for a mean of 26.7 years (range 1–53 years), indwelling urethral catheter for a mean of 31 years (range
2 patients (8%) had a suprapubic catheter for a mean of 2.5 30–32 years), one (11.1%) used a suprapubic tube for 3 years,
years (range 2–3 years), and 14 patients (56%) managed their and five (55.6%) used external catheters for a mean of 42.8
bladder with an external catheter for a mean of 32 years years (range 18–50 years). A total of 7 out of 14 patients
(range 1–55 years). Several patients used multiple forms of (50%) used tobacco for a mean of 44.7 pack years (range
bladder management over the course of their disease 18–100 pack years).
(Table 2).
All 32 patients had information available about the stage
Transitional cell carcinoma
of their tumor. A total of 6 (20.6%) had Stage 1, 3 (10.3%)
Of the 10 patients with TCC, 7 (70%) eventually underwent
had Stage II, 11 (37.9%) had Stage III, 9 (20.6%) had Stage IV,
cystectomy and 7 out of 10 died at a mean of 1.7 years after
and 3 (10.3%) had CIS (Table 3).
diagnosis (range 0–5 years) with 2 dying of their disease.
A total of 9 of the 32 patients (28.1%) were alive as of
All 10 patients had staging information available with 5 patients
January 2007 with 23 (71.9%) having died. Of those patients
(50%) having Stage I disease, 4 patients (40%) with Stage III
who are deceased, a cause of death could be identified for 21
disease, and 1 patient (10%) with Stage IV disease. A total of
of them with 9 (42.8%) having died of their bladder tumor
9 patients had information regarding bladder management
at a mean of 2.4 years after initial diagnosis (range 0–9)
available and among them 1 (11.1%) used clean intermittent
(Table 1).
catheterization for 17 years, 7 (77.8%) had an indwelling
A total of 21 out of 30 patients (70%) who had information
urethral catheter for a mean of 21 years (range 1–53 years),
available used tobacco for a mean of 58 pack years (range
1 (11.1%) had a suprapubic catheter for 2 years, and
10–180 pack years) (Table 1).
5 (55.6%) had an external catheter for a mean of 41 years
(range 32–50 years). Several patients used multiple forms of
bladder management over the course of their SCI. Nine patients
Squamous cell carcinoma
(90%) used tobacco for a mean of 65.6 pack years (range 10–180
Of the 15 patients with SCC, 14 (93.3%) eventually under-
pack years).
went a cystectomy and 11 of the 15 died at a mean of 5.1
years after diagnosis with 6 dying because of their disease
(range 0–20 years). A total of 13 patients had staging Adenocarcinoma
information available with 1 patient (7.6%) with Stage II Of the three patients with adenocarcinoma all three (100%)
disease, 6 patients (46.1%) with Stage III disease, 3 patients underwent cystectomy and two eventually died at a mean of
(23.1%) with Stage IV disease, and 3 patients (23.1%) with 5 years after diagnosis (range 2–8 years) with none dying of
CIS. Of the nine patients for whom bladder management their disease. All three patients had information regarding

Spinal Cord
 Bladder cancer in spinal cord injury patients
JF Kalisvaart et al
259

Table 3 Stage of bladder tumors at diagnosis (15.7%) had findings on annual radiologic workup (ultra-
sound or CT scan), which led to cystoscopy and two patients
Tumor type (total patients Stage Number of
(10.5%) had a scrotal abscess, which led to cystoscopy. The
with known staging) patients (%)
date of the last cystoscopy was available on 15 patients, and a
Overall (32) 1 6 (20.6) mean of 1.6 years (range 0.7–5.5 years) had passed since their
2 3 (10.3) last cystoscopy.
3 11 (37.9)
4 9 (20.6)
CIS 3 (10.3)
Squamous cell carcinoma (13) 1 0 (0)
2 1 (7.6) Discussion
3 6 (46.1)
4 3 (23.1) Bladder tumors in the SCI patient are a well-recognized
CIS 3 (23.1) phenomenon. Most earlier studies have shown that the
Transitional cell carcinoma (10) 1 5 (50) relative risk of a bladder neoplasm in the spinal cord injured
2 0 (0) population is 16–28 times higher than that of the general
3 4 (40)
4 1 (10) population.1–3 Recently, this increase has come into some
CIS 0 (0) question with a study showing no increase in cancer rates
Adenocarcinoma (3) 1 1 (33.3) among a spinal cord injured population.4 The patient
2 1 (33.3) population in this study did not undergo formal screening
3 0 (0)
4 1 (33.3) and had a much lower rate of tobacco use than in other
CIS 0 (0) studied populations, but it does suggest that further, well-
Mixed transitional and 1 0 (0) controlled studies need to be performed to determine the
squamous cell carcinoma (3) true incidence of bladder neoplasms in spinal cord injured
2 1 (33.3)
3 1 (33.3) populations.
4 1 (33.3) SCI patients are also known to have a different spectrum of
CIS 0 (0) disease than the general population with a significantly
higher proportion of SCC compared with TCC, a lower age at
diagnosis and a higher stage.5
staging and one patient (33.3%) each had Stage I, II, and IV We present one of the largest series of SCI patients with
disease. All three patients (100%) had a history of Foley bladder tumors with 32 patients examined out of a total
catheter use for a mean of 30 years (range 20–50 years) and population of 1319 patients seen. In this study, several items
two patients (66.7%) also had a history of external catheter of interest were noted. First, the distribution of tumors seen
use for a mean of 10.5 years (range 1–20 years). All three was similar to earlier studies with SCC representing approxi-
patients used tobacco for a mean of 77 pack years (range mately 1/3 of the tumors and TCC representing almost 1/2
25–150 pack years). (Table 1). Interestingly, adenocarcinoma represented 9.4%,
which is significantly above the quoted rate of 2% in the
general population.6 This increase has also been seen in
Mixed TCC and SCC
earlier studies with West et al.5 reporting adenocarcinoma
All three (100%) of the patients with mixed tumors under-
representing 10% of their bladder tumors. One of the known
went cystectomy and two of the three died at a mean of 3
risk factors for adenocarcinoma of the bladder, much like in
years after diagnosis (range 0–6) with one dying of their
other bladder malignancies, is chronic irritation. The
disease. Three patients had staging information with one
increase in adenocarcinoma could potentially be due to
patient each (33.3%) having Stage II, III, and disease.
chronic irritation from catheterization, chronic urinary tract
One patient (33.3%) had a history of clean intermittent
infections, bladder stones, or perhaps an intrinsic change in
catheterization for 33 years, two patients (66.7%) had a
the neurogenic bladder.6
history of Foley catheter use for a mean of 37.5 years (range
Also seen in this study are a high number of patients with
20–55 years), and one (33.3%) used an external catheter for
bladder tumors who did not have an indwelling catheter
33 years. Two of the patients had information available
(Table 2). Earlier studies have identified potential risk factors
regarding tobacco use and one of the two (50%) used tobacco
for bladder tumors in the SCI population. These have
for 10 pack years.
included indwelling catheter for 410 years, frequent UTIs,
and frequent bladder stones.2,5–7 In patients with SCC in our
Method of detection of tumors study, however, 55% used an external catheter for a mean of
The method of detection of the tumor was known in 19 40 þ years with only 22.2% having had an indwelling
patients. Eight patients (42.1%) had screening cystoscopies catheter. Some of these patients may have had an indwelling
with their annual examination, which detected a tumor. catheter early in their injury, which was not documented,
Seven patients (36.8%) reported hematuria. Five of these had but with 40 þ years of using an external catheter, the
a cystoscopy based solely on the hematuria, and two majority of their bladder management had been with an
reported hematuria at their annual screening cystoscopy. external catheter. Earlier studies have suggested that the
One patient (5.2%) had a positive voided cytology, which led nature of the neurogenic bladder itself may lead to an
to a cystoscopy and discovery of his tumor. Three patients increased risk of tumors because of the interaction of the

Spinal Cord
 Bladder cancer in spinal cord injury patients
JF Kalisvaart et al
260

bladder mucosa with a high volume of urine that is often This diligence in screening remains especially important
seen in the neurogenic bladder.5 The current findings seem because of the aggressive and rapidly growing nature of
to support this. bladder tumors in spinal cord patients. In our study, the
The recommendations for screening in this population tumors tended to be aggressive with 58.4% of the tumors
have also been somewhat controversial and earlier recom- presenting in Stage III or IV and 43% of the patients with
mendations have varied widely. Current VA guidelines for available information died of their tumor at a mean of 2.4
SCI patients state that a cystoscopy is to be performed every years after diagnosis. Furthermore, patients had a mean of
10 years for patients with indwelling urethral catheter, only 1.6 years since their last cystoscopy indicating that,
suprapubic tubes, or who use tobacco. They also recommend despite surveillance and early detection, these tumors are
that cystoscopy be performed for indications including new also fast growing (Table 3). Both of these findings have been
onset retention, recurrent UTIs, stones, or hematuria.8 noted in earlier studies.3,11,13–16 The reason for the rapid and
Others have advocated screening SCI patients with an aggressive nature of these tumors is unknown. In other
indwelling catheter after 8–10 years of use with yearly populations with bladder cancer, approximately 25% will
cystoscopy, cytology, and possible biopsy.9 Recent studies present with muscle invasive disease,6 whereas in this study
by Razdan et al. in 200310 and Kitahara et al. in 200611 show over 1/2 had advanced disease at presentation. In other
that the majority of urologists are performing annual populations, bladder cancer is thought to be due to a general
cystoscopies for patients with long-term indwelling cathe- field defect, often because of a systemic irritant such as
ters, and 75–88% perform them if the patient has an tobacco leading to the growth and development of cancers.
indication such as hematuria or bladder stones. There has The mechanism may be different in SCI patients with
also been some interest in the use of urine tumor markers for chronic, focal irritation of the bladder because of stones,
screening. A paper by Davies et al. in 200512 confirmed that UTIs, catheters, or urinary stasis having a function in
cystoscopy is presently the best screening tool for bladder creating more aggressive tumors.
cancer in SCI population. In a prospective study looking at
screening with urine cytology and urine tumor markers in an
SCI population, they found three patients with bladder Conclusion
cancer over a 5-year period on cystoscopy, none of whom Our study suggests that an indwelling catheter is not the sole
had positive cytology, BTA stat, or survivin assay. source of the increased risk of bladder neoplasm that appears
At our institution, we attempt to perform a yearly screen- to exist in the spinal cord injured population. Although it is
ing cystoscopy on all patients with indwelling catheters for a relatively small study, it suggests that a high index of
any length of time, especially those who have had one in suspicion needs to be maintained with spinal cord patients
place for 410 years. We also perform a cystoscopy on any with or without indwelling catheters. Furthermore, tumors
patient with hematuria, recurrent urinary tract infections, can happen, and are most common, several decades after the
or any other urologic complaints indicating need for a initial injury suggesting that a vigilant attitude toward
cystoscopy (that is, frequency, change in bladder function- annual screening cystoscopies needs to be maintained for
ing, clogging of catheter, etc). We do not routinely take the life of the patient.
random bladder biopsies, but are aggressive in biopsying any
abnormal appearing areas and frequently take cytologies.
This study suggests that screening needs to be continued Conflict of interest
diligently and indefinitely in spinal cord patients as bladder
tumors were identified for a mean of 34 years after their The authors declare no conflict of interest.
initiating injury. The length of time since the injury may
itself be a risk factor for the development of bladder tumors.
References
As the bladder is exposed chronically to relatively static
urine, irritation from an indwelling catheter, bladder stone, 1 Kaufman JM, Fam B, Jacobs SC, Gabilondo F, Yalla S, Kane JP et al.
Bladder cancer and squamous metaplasia in spinal cord injury
and/or UTIs, it is going to develop changes in the cells.
patients. J Urol 1977; 118: 967–971.
As these cells undergo metaplasia to cope with these 2 Groah SL, Weitzenkamp DA, Lammertse DP, Whiteneck GG,
irritants, it is inevitable that eventually some of them will Lezotte DC, Hamman RF. Excess risk of bladder cancer in spinal
undergo dysplasia and begin the process of tumorogenesis. cord injury: evidence for an association between indwelling
catheter use and bladder cancer. Arch Phys Med Rehabil 2002; 83:
As with any similar process, the longer the irritant is present,
346–351.
the more likely that dysplasia and tumors will develop. 3 Bejany DE, Lockhart JL, Rhamy RK. Malignant vesical tumors
Furthermore, performing an annual cystoscopy in following spinal cord injury. J Urol 1987; 138: 1390–1392.
asymptomatic patients with risk factors such as an 4 Subramonian K, Cartwright RA, Harnden P, Harrison SCW.
Bladder cancer in patients with spinal cord injuries. BJU Int
indwelling catheter, recurrent urinary tract infections or
2004; 93: 739–743.
bladder stones remain important as 6 patients out of 19 5 West DA, Cummings JM, Longo WE, Virgo KS, Johnson FE, Parra
diagnosed tumors (31.5%) would have been missed if RO. Role of chronic catheterization in the development of
cystoscopy was performed solely for symptoms. As such, it bladder cancer in patients with spinal cord injury. Urology 1999;
53: 292–297.
may be prudent to perform a cystoscopy in high-risk groups
6 Messing EM. Urothelial tumors of the bladder. In: Walsh PC, Retik
at a higher frequency than that recommended by the VA AB, Vaughn Jr ED, Wein AJ (eds). Campbell-Walsh Urology, 9th
guidelines. edn. Saunders: Philadelphia, 2002. pp 2407–2447.

Spinal Cord
 Bladder cancer in spinal cord injury patients
JF Kalisvaart et al
261

7 Shokeir AA. Squamous cell carcinoma of the bladder: pathology, 12 Davies B, Chen JJ, McMurry T, Landsittel D, Lewis N, Brenes G
diagnosis and treatment. BJU Int 2004; 93: 216–220. et al. Efficacy of BTA stat, cytology and surviving in bladder
8 Spinal Cord Injury and Disorders System of Care Procedures. VHA cancer surveillance over 5 years in patients with spinal cord
Handbook 11761. Department of Veterans Affairs, Veterans Health injury. Urology 2005; 66: 908–911.
Administration: Washington DC, 2005. 13 Richie JP, Waisman J, Skinner DG, Dretler SP. Squamous
9 Stonehill WH, Dmochowski RR, Patterson AL, Cox CE. Risk carcinoma of the bladder: treatment by radical cystectomy. J Urol
factors for bladder tumors in spinal cord injury patients. J Urol 1976; 115: 670–672.
1996; 155: 1248–1250. 14 Larsen LD, Chamberlin DA, Khonsari F, Ahlering TE. Retro-
10 Razdan S, Leboeuf L, Meinbach DS, Weinstein D, Gousee AE. spective analysis of urologic complications in male patients with
Current practice patterns in the urologic surveillance and spinal cord injury managed with and without indwelling urinary
management of patients with spinal cord injury. Urology 2003; catheters. Urology 1997; 50: 418–422.
61: 893–896. 15 Navon JD, Soliman H, Khonsari F, Ahlering TE. Screening
11 Kitahara S, Iwatsubo E, Yasuda K, Ushiyama T, Nakai H, Suzuki I cystoscopy and survival of spinal cord injured patients with
et al. Practice patterns of Japanese physicians in urologic squamous cell cancer of the bladder. J Urol 1997; 157: 2109–2111.
surveillance and management of spinal cord injury patients. 16 Bickel A, Culkin DJ, Wheeler Jr JS. Bladder cancer in spinal cord
Spinal Cord 2006; 44: 362–368. injury patients. J Urol 1991; 146: 1240–1242.

Spinal Cord