NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Clinical Governance Research and Development Unit (CGRDU), Department of Health Sciences, University of Leicester. Referral Guidelines for Suspected Cancer in Adults and Children [Internet]. London: Royal College of General Practitioners (UK); 2005 Jun. (NICE Clinical Guidelines, No. 27.)

  • This publication is provided for historical reference only and the information may be out of date.

This publication is provided for historical reference only and the information may be out of date.

Cover of Referral Guidelines for Suspected Cancer in Adults and Children

Referral Guidelines for Suspected Cancer in Adults and Children [Internet].

Show details

14Urological cancers

General recommendations


A patient who presents with symptoms or signs suggestive of a urological cancer should be referred to a team specialising in the management of urological cancers, depending on local arrangements. [D]

Specific recommendations

Prostate cancer


Patients presenting with symptoms suggesting prostate cancer should have a digital rectal examination (DRE) and prostate specific antigen (PSA) test after counselling. Symptoms will be related to the lower urinary tract and may be inflammatory or obstructive. [C]


Prostate cancer is also a possibility in male patients with any of the following unexplained symptoms: [C]

  • erectile dysfunction
  • haematuria
  • lower back pain
  • bone pain
  • weight loss, especially in the elderly.

These patients should also be offered a DRE and a PSA test.


Urinary infection should be excluded before PSA testing, especially in men presenting with lower tract symptoms. The PSA test should be postponed for at least 1 month after treatment of a proven urinary infection. [C]


If a hard, irregular prostate typical of a prostate carcinoma is felt on rectal examination, then the patient should be referred urgently. The PSA should be measured and the result should accompany the referral. Patients do not need urgent referral if the prostate is simply enlarged and the PSA is in the age-specific reference range1. [C]


In a male a patient with or without lower urinary tract symptoms and in whom the prostate is normal on DRE but the age-specific PSA is raised or rising, an urgent referral should be made. In those patients whose clinical state is compromised by other comorbidities, a discussion with the patient or carers and/or a specialist in urological cancer may be more appropriate. [C]


Symptomatic patients with high PSA levels should be referred urgently. [C]


If there is doubt about whether to refer an asymptomatic male with a borderline level of PSA, the PSA test should be repeated after an interval of 1 to 3 months. If the second test indicates that the PSA level is rising, the patient should be referred urgently. [D]

Bladder and renal cancers


Male or female adult patients of any age who present with painless macroscopic haematuria should be referred urgently. [C]


In male or female patients with symptoms suggestive of a urinary infection who also present with macroscopic haematuria, investigations should be undertaken to diagnose and treat the infection before consideration of referral. If infection is not confirmed the patient should be referred urgently. [D]


In all adult patients aged 40 years and older who present with recurrent or persistent urinary tract infection associated with haematuria, an urgent referral should be made. [C]


In patients under 50 years of age with microscopic haematuria, the urine should be tested for proteinuria and serum creatinine levels measured. Those with proteinurea or raised serum creatinine should be referred to a renal physician. If there is no proteinuria and serum creatinine is normal, a non-urgent referral to a urologist should be made. [C]


In patients aged 50 years and older who are found to have unexplained microscopic haematuria, an urgent referral should be made. [C]


Any patient with an abdominal mass identified clinically or on imaging that is thought to be arising from the urinary tract should be referred urgently. [C]

Testicular cancer


Any patient with a swelling or mass in the body of the testis should be referred urgently. [C]


An urgent ultrasound should be considered in men with a scrotal mass that does not transilluminate and/or when the body of the testis cannot be distinguished. [D]

Penile cancer


An urgent referral should be made for any patient presenting with symptoms or signs of penile cancer. These include progressive ulceration or a mass in the glans or prepuce particularly, but can involve the skin of the penile shaft. Lumps within the corpora cavernosa not involving penile skin are usually not cancer but indicate Peyronie’s disease, which does not require urgent referral. [D]



Prostate cancer

The total registrations of newly diagnosed cased of prostate cancer in 2001 in England and Wales was 26,027 per 100,000 population. Prostate cancer is rare in men below 50 years of age, with only 0.5% of cases occurring in men in this age group. Incidence rises steeply with increasing age until peaking in those aged 85 years.

Figure 3. 2001 Registrations of prostate cancer in England and Wales. (77).

Figure 32001 Registrations of prostate cancer in England and Wales. (77)

Testicular cancer

In England and Wales in 2001 there were 1,997 cases of newly diagnosed testicular cancer. Incidence is low in those aged under 20 years, but increases steeply and peaks between the ages of 30–34 years. Almost 50% of cases occur in those aged under 35 years and 80% in those aged under 45 years.

Figure 4. 2001 Registrations of testicular cancer in England and Wales. (77).

Figure 42001 Registrations of testicular cancer in England and Wales. (77)

Cancer of the kidney

There were 5338 new diagnoses of cancer of the kidney registered in 2001 in England and Wales. Of those cases 3,281 were in males and 2,057 in females. Cancer of the kidney is rare below the age of 40 years and increases with age in both sexes. The disease peaks between ages 80–84 years in males. In females the incidence is generally much lower but peaks within the same age group.

Figure 5. 2001 Registrations of cancer of the Kidney in England and Wales. (77).

Figure 52001 Registrations of cancer of the Kidney in England and Wales. (77)

Cancer of the bladder

In 2001 there were 11,403 cases of bladder cancer in England and Wales. Of those 8,101 were in males, and 3,302 in females. Incidence is low below the age of 50 years in both sexes, after which incidence increases sharply in males and more gradually in females.

Figure 6. 2001 Registrations of cancer of the bladder in England and Wales. (77).

Figure 62001 Registrations of cancer of the bladder in England and Wales. (77)


Prostate cancer

Age specific mortality is similar to incidence, increasing with age from 50 years onwards. In 2002 in England and Wales there were 8,973 deaths from prostate cancer. Age distribution is shown in Figure 724.

Figure 7. 2002 Mortality rates from prostate cancer in England and Wales. (78).

Figure 72002 Mortality rates from prostate cancer in England and Wales. (78)

Testicular cancer

Age specific mortality from testicular cancer varies vary little across the age groups. In 2002 in England and Wales there were 120 deaths from testicular cancer. Age distribution is shown in Figure 8.

Figure 8. 2002 Mortality rates from testicular cancer in England and Wales. (78).

Figure 8

2002 Mortality rates from testicular cancer in England and Wales. (78).

Cancer of the kidney

Mortality is low in both sexes in those aged under 40 years. In 2002 in England and Wales, there were 1,035 deaths from cancer of the kidney in females and 1,749 in males.

Figure 9. 2002 Mortality rates from Kidney cancer in England and Wales. (78).

Figure 92002 Mortality rates from Kidney cancer in England and Wales. (78)

Cancer of the bladder

Mortality is low in those aged under 60 years in both sexes in England and Wales. In 2002 in England and Wales there were 1,501 deaths from bladder cancer in females and 2,919 in males. Age distribution is shown in Figure 10.

Figure 10. 2002 Mortality rates from cancer of the bladder in England and Wales. (78).

Figure 10

2002 Mortality rates from cancer of the bladder in England and Wales. (78).

Review of cancer referral audits

The review(13) identified 43 relevant clinical audits. The proportion of two week referrals found to be in accordance with the symptoms in the guidelines(2) who were found to have cancer ranged from 70% to 100% (14 audits). The proportion of patients referred under the two week system who were subsequently found to have cancer ranged from 13% to 40% (15 audits). The proportion of patients with cancer who had been referred under the two week system ranged from 0% to 39% (six audits). The proportion of two week referrals assessed as clinically appropriate ranged from 78% to 91% (four audits).

14.1. Symptoms and Signs

14.1.1. Key Clinical Question

In people attending primary care services with urological problems, which symptoms and signs and other features including family history when compared with the ‘gold standard’ are predictive of a diagnosis of cancer, and which are not?

14.1.2. Evidence Question

In people attending primary care services with symptoms and signs that might be associated with upper urological cancers, which symptoms and signs and other features including family history, when compared with the ‘gold standard’, are predictive of a diagnosis of cancer, and which symptoms and signs are not? Are any non-clinical features associated with a diagnosis of cancer?

14.1.3. Evidence Statements

Prostate cancer is rare below the age of 45 years, but the incidence rises steeply thereafter. (III)

Testicular cancer can occur at almost any age, but is most common below the age of 40. (III)

Renal cancer is rare below the age of 35, but increases in incidence thereafter. It is more common in males. (III)

Bladder cancer is rare below the age of 50, but increases in incidence thereafter. It is more common in males. (III)

Prostate cancer often presents with symptoms of urinary outflow obstruction. Other presenting symptoms include urinary tract infection, and features of metastasis, such as bone pain. (III)

Most prostate cancers can be palpated on digital rectal examination by the primary care professional, but an abnormality on examination may be caused by conditions other than cancer. (III)

The most common presenting feature of testicular cancer is enlargement of the testis, which may or may not be associated with pain. (III)

Testicular cancer may first present with features of metastasis, for example with back pain or breathlessness. (IV)

Penile cancer is rare, and presents as an area of induration, erythema or warty growth. (III)

Bladder cancer commonly presents with macroscopic haematuria. (III)


The urological cancers considered in this guideline included prostate, testicular, kidney, and bladder. Carcinoma of the scrotum is rare and was not included. There were few primary care studies dealing with urological cancer (especially so for testicular and penile cancers). In addition, many of the studies dealing with prostate cancer concentrated on examinations and tests for screening rather than on the assessment of patients presenting to primary care with symptoms. Although the scope of this guideline excluded papers dealing with screening, the quantity of primary evidence in this area was so limited that some screening focused reviews have been included, to enable cautious extrapolation to symptomatic patients.

(NICE, 2001)(265)

This document was classified by NICE as guidance and was commissioned by the Department of Health and the National Assembly for Wales to provide advice to health professionals on the appropriate referral of patients from general to specialist services. A consensus method was used to generate the advice. A multidisciplinary panel was established for each topic considered, and selected research evidence was considered.

One of the topics considered was urinary tract outflow symptoms. The advice recommended that patients be offered a prostate specific antigen (PSA) test with the reasons for doing the test being explained and the patient counselled with regard to the possible consequences. Patient information on PSA tests can be obtained from the National Electronic Library for Cancer ( Immediate referral was advised if the patient has acute urinary retention or evidence of acute renal failure; urgent referral was advised if the patient has (a) visible haematuria, (b) there is a suspicion of prostate cancer based on the findings of a nodular or firm prostate, and/or a raised PSA, (c) culture negative dysuria, (d) they develop chronic urinary retention with overflow or night-time incontinence. Referral to be seen soon was advised if the patient has recurrent urinary tract infection or microscopic haematuria. Referral within an appropriate time was advised if the patient has chronic renal failure or renal damage, or symptoms have failed to adequately respond to treatment in primary care. Use of a scoring system such as the WHO International Prostate Symptom Score was encouraged.

(SIGN, expected 2005)(230)

SIGN is publishing guidelines on the management of transitional cell carcinoma of the bladder. The guidelines are directed at management in secondary care and do not consider identification and referral of suspected cases in any detail. However, the guidelines do recommend that patients with frank haematuria should be seen within two weeks by a specialist, and that those with occult asymptomatic (microscopic) haematuria should be seen within six weeks.

(Lobel et al, 1998)(266)

These guidelines were developed by an international group, and included reference to 89 original articles, although the methods of guideline development were not described in detail. The guidelines gave detailed consideration to initial assessment in primary care, but did state that all patients with gross haematuria should be examined and referred to a urologist for assessment for possible bladder tumour. Patients with asymptomatic microscopic haematuria should be referred if they are aged over 50 years. In those under aged 50, the guidelines were uncertain, but noted that the incidence of cancer in this group was 5% with asymptomatic microscopic haematuria and 10.5% with symptomatic microscopic haematuria.

(Mickisch et al, 2001)(267)

These guidelines were prepared by the European Association of Urology following a literature search using Medline, with articles being graded by a panel of experts. The presenting features include haematuria, palpable tumour and flank pain. However, presentation with clinical features is becoming less common and many cases are being diagnosed at the asymptomatic stage. The majority of tumours are diagnosed by abdominal ultrasound performed for various reasons.

Secondary studies
Prostate cancer
(Muris et al, 1993)(268)

In this review, publications were identified from Medline dated 1982 to 1991, and those included involved studies of patients with complaints in which rectal examination was indicated. Eight studies met the inclusion criteria, two of which involved men attending outpatient departments because of prostate related symptoms. These studies included a total of 325 patients. The sensitivity of rectal examination in detecting prostate cancer was 98% and 92% in the two studies, specificity was 53% and 48%, and likelihood ratio 2.09 and 1.77.

(Selley et al, 1997)(269)

This was a systematic review of the diagnosis, management and screening of early localised prostate cancer. From the included studies of digital rectal examination (DRE), it was concluded that 50–95% of localised prostate tumours are palpable and could be detected by DRE. A proportion of the lesions detected on palpation are benign, and include benign prostatic hyperplasia (BPH), retention cysts, prostatic calculi, prostatic atrophy, fibrosis associated with prostatitis, and non-specific granulomatous prostatitis. False positive rates on DRE are as high as 40–50%.

The sensitivity of DRE ranged from 44% to 97% in the four studies reporting this, and specificity from 22% to 96%. The reasons for these variable findings were probably related to the different sizes of the studies, case selection and variable final diagnostic criteria.

(Fowler et al, 2000)(270)

The aim of the study was to determine whether features used to detect prostate cancer are different in black and white American men. The study subjects were 179 black and 357 white men who had undergone prostate biopsy 1992–1999 at one medical centre. The patients had an abnormal DRE, a PSA of less than 4ng/ml and no history of prostate surgery. Cancer was detected in 38 black (21%) and 65 white (18%) men. There was no difference in the overall or PSA stratified cancer detection rate.

Testicular cancer

No relevant articles dealing with the diagnosis of testicular cancer in primary care were identified, and therefore this review was included.

(Gospodarowicz, 1999)(271)

Testicular cancers are uncommon, occurring most commonly in men aged 15 to 35 years. The majority are primary germ cell tumours (GCT). Although the incidence of germ cell tumours has doubled in the past 30 years, the mortality has declined.

There is considerable geographic and ethnic variation in incidence of germ cell tumours, it being less common in non-white men. Men with a history of cryptorchidism have an approximate five-fold risk. Family clusters have been reported, and patients with XY gonadal dysgenesis are at increased risk. Prior testicular cancer is also a risk factor for cancer in the surviving testis. Patients with tumours most commonly present with painless testicular enlargement. Up to 45% have testicular pain. Less common presentations include features of metastasis, for example back pain and dyspnoea.

Penile cancer

No other relevant articles were identified, specifically no studies of presentation in primary care, and therefore this review was included.

(Burgers, 1992)(272)

This was a comprehensive review of penile cancer. Cancer of the penis is rare, accounting for only 0.4–0.6% of all male malignancies in the US (incidence 0.2/100,000 males/year). Squamous cell carcinoma accounts for at least 95% of cases, sarcomas being the most common non-squamous type. It usually presents in the sixth decade of life, with a mean age at diagnosis of 58. There is an association between absence of circumcision and penile cancer, but the precise aetiology is unclear. The possible role of pre-malignant conditions has not been clarified.

Presentation is varied, ranging from innocuous areas of in-duration, erythema or warty growth to obvious extensive carcinoma with sloughing. The earliest symptoms include itching or burning, and ulceration which progresses to a lump, mass or nodule if left untreated. Pain is usually minimal in relation to the other features. It can occur at any anatomical site; 48% develop in the glans, 21% prepuce, both (9%), coronal sulcus 6%, shaft <2%.

Bladder/renal cancer
(Buntinx, 1997)(273)

This was a systematic review of studies of the diagnostic value of macroscopic haematuria in diagnosing urological cancers in primary care. Studies were sought using Medline and FAMLI databases. 14 studies were selected, but none had been undertaken in primary care, most being based on chart reviews in hospital settings of referred patients. The findings are summarised in Table 2.

Table 2. Pooled sensitivity of macroscopic haematuria for the diagnosis of urological cancers.(273).

Table 2

Pooled sensitivity of macroscopic haematuria for the diagnosis of urological cancers.(273).

Penile cancer
(Micali et al, 2004)(2)

In this review, a Medline search and meeting abstract books were searched to identify articles on penile cancer. Relatively few studies reported the clinical findings at presentation, but on the basis of the studies identified, the clinical features of squamous carcinoma of the penis were described as either small ulcers that both enlarge superficially and infiltrate deeper tissues, or as circumscribed areas of warty growth that subsequently progress to a mass or nodule and undergo necrosis or ulceration. In both cases, erythema, induration, bleeding and secondary infection are reported as common findings. The majority of cases occur on the glans, foreskin. And/or coronary sulcus, and initially may be asymptomatic or more often cause itching, burning or pain. The development of penile cancer has been reported as following the occurrence of several predisposing factors, including leukoplakia, genital lichen sclerosus, human papillomavirus, and chronic balanitis.

Primary studies
Prostate cancer
(Haid et al, 1994)(274)

This study involved 99 men who had undergone transrectal ultrasound (TRUS) at a US hospital. The records were reviewed to extract information on findings from digital rectal examination (DRE), prostate biopsy reports, and PSA levels. With biopsy as the gold standard, 32 (32.3%) of the 99 had carcinomas.

Among those with carcinoma, 24 (77.4%) (of 31 with data) had a palpable nodule on rectal examination, the mean PSA was 32.5, and 15/31 had an abnormality on transrectal ultrasound (48.4%). Among those who did not have carcinomas, 52/64 had a palpable nodule (81.2%), the mean PSA was 8.4, and 26/65 had an abnormality on ultrasound (40.0%).

For DRE sensitivity was 77.4% while specificity was 18.8%. The positive predictive value (PPV) amounted to 31.6% and negative predictive value (NPV) totalled 63.2%.

Whereas TRUS sensitivity was 98% while specificity was 60%. The PPV amounted to 36.6% and NVP totalled 70.9%.

Finally PSE sensitivity was 93.3% while specificity was 21.2%. The PPV amounted to 40% and NVP totalled 84.6%.

(Brett, 1998)(275)

A consecutive series of 241 men aged 50–79 attending a solo general practice in Perth, Western Australia, in 1996 were invited to take part in the study. Of these, 211 gave consent, and were offered both digital rectal examination and PSA tests. A prostate was regarded as abnormal on examination if there was evidence of nodularity, induration, asymmetry or absence of median sulcus. 199 (91.0%) were found to have a normal prostate, and 19 (9.0%) abnormal. The PSA test was regarded as normal if results were in the 0–4ng/ml range. 191 (90.5%) were in the normal range, and 20 (9.5%) were abnormal.

Of the 211 patients, 182 were normal on both tests, 29 having an abnormal finding on one or other test. From the 29, 11 biopsies were performed, with prostate cancer detected in three (27.3%). Twelve patients opted for various reasons not to undergo biopsy (eight had had biopsies in the past), and six were not biopsied because of poor health.

Table 3Summary findings of DRE and PSA test in 211 males aged 50–79 consulting one general practitioner.(275)

Age (yrs)NumberNormal DREAbnormal DRENormal PSAAbnormal PSA
(Mansson et al, 1999)(149)

This study was a retrospective case series, the cases being patients identified from a cancer registry and from one district in Sweden (Kungsbacka). The medical records of all patients were reviewed for information about initial symptoms, diagnostic procedures, outcome of diagnostic procedures, level of care, and doctor’s delay. The study collected information about new cases of prostate cancer presenting 1980–1984. There were 86 cases, an age standardized incidence of 103/100,000 per year. The symptoms at presentation are shown in Table 4.

Table 4. Initial symptoms of prostate cancer reported at the first visit and all the symptoms presented at the first visit.(149).

Table 4

Initial symptoms of prostate cancer reported at the first visit and all the symptoms presented at the first visit.(149).

(Gjengsto et al, 2004)(172)

This study was undertaken in a early prostate cancer clinic in Norway, and included 872 patients referred by their general practitioner between 1997 and 2000. A total of 360 (41.3%) were diagnosed as having prostate cancer on biopsy. The median age was 63.7 years (range 40–86 years). Of the 373 patients who had consulted their general practitioner because of lower urinary tract symptoms, 34.3% were found to have cancer. Among the 462 patients who consulted without urological symptoms, the frequency of cancer was higher – 51% for those undergoing a health check, 41.9% for those with non-urological disease and 42.1% among those concerned about having cancer. The general practitioner gave a raised PSA as the reason for referral in 647 cases (222 or 34.3% had cancer), elevated PSA and suspicious DRE in 185 (125 or 67.6% had cancer), and suspicious DRE alone 24 (7 or 29.2% had cancer). Of those found to have cancer, 79.2% had no family history; 11.7% had a first-degree relative history, 6.4% a second-degree relative history, and 1.9% a first and second degree relative history.

Bladder/renal cancer
(Summerton et al, 2002)(276)

The patients in this case series were 363 people referred to an open access haematuria clinic in the UK. Patients were aged between 18 and 80, and the final diagnosis was established by cystoscopy and radiological assessment, supplemented by review of the records to check for any changes in diagnoses over time. Information was collected prospectively about clinical features and comorbidities at first clinic attendance. Cases were classified into urological and non-urological cancers, and urological and non-cancerous/normal groups. The associations between clinical features and diagnoses were explored using a variety of statistical techniques, including logistic regressi on.

172 patients had macroscopic haematuria and 186 microscopic haematuria. Of the 363 referred patients, no abnormality was detected in 260, 42 had benign prostatic disorders, 12 had strictures or stenoses, 13 had calculi, and 36 had urological cancers (28 of which were bladder cancers, two prostate cancers, five renal cancers, and one had both renal and bladder tumours). In multivariate analysis, the variables tending to be associated with urological cancer were older age, male sex, macroscopic haematuria (especially if a single episode), poor stream, history of urinary tract infection and smoking.

Table 5Log-likelihood ratios for normal conditions versus urological cancer (N=296).(276)

VariableNormal conditions (%) N=260Cancer (%) N=36Log likelihood ratios
Age in years
</= 4555 (22.1)1 (2.8)−.21
46–5969 (27.3)6 (16.7)−.05
60–7490 (35.6)15 (41.7)0.02
75+38 (15.0)14 (38.9)1.0
Male126 (48.5)28 (77.8)0.05
Female134 (51.5)8 (22.2)−.08
Type of haematuria
microscopic159 (61.9)3 (8.6)−2.0
macroscopic98 (38.1)32 (91.4)0.9
Timing of macroscopic blood
Beginning12 (13.0)7 (21.9)0.5
Throughout25 (27.2)13 (40.6)0.4
End14 (15.2)3 (9.4)−0.5
Unsure41 (44.6)9 (28.1)−0.5
Number of episodes
061 (23.5)2 (5.6)−1.4
159 (22.7)15 (41.7)0.6
265 (25.0)5 (13.9)−0.6
3+75 (28.8)14 (38.9)0.3
History of UTI
0195 (75.0)29 (80.6)0.1
128 (10.8)7 (19.4)0.6
2+37 (14.2)0 (0)−2.4
0–1157 (60.4)18 (50.0)−0.2
2–488 (33.8)15 (41.7)0.2
5+15 (5.8)3 (8.3)0.4
No231 (88.8)24 (68.6)−0.3
Yes29 (11.2)11 (31.4)1.0
Poor stream
No221 (85.0)25 (71.4)−0.2
Yes39 (15.0)10 (28.6)0.7
No171 (65.8)21 (60.0)−0.1
Yes89 (34.2)14 (40.0)0.2
(Bruyninckx et al, 2003)(277)

The study was undertaken in a network of Belgian general practices and included all patients attending with macroscopic haematuria during 1993–1994. 83 general practitioners took part. Patients were followed up for 18 months to determine the final diagnosis.

409 patients attended with macroscopic haematuria and 126 patients were diagnosed during the same period as having urological cancer. The mean age of patients with macroscopic haematuria was 57 years, but the age of those with cancer was 72 years. 13% of those with haematuria were younger than 40 years and 53% older than 60 years.

In 87 patients (70 males, 17 females) bladder cancer was detected, and in 39 (23 males, 15 females, one sex unknown) other urological cancers were detected. 75 of the 126 patients reported macroscopic haematuria in the weeks before diagnosis, giving a sensitivity for a diagnosis of any urological cancer of 59.5% (95% CI 50.4–68.1%). The PPV of macroscopic haematuria for the diagnosis of urological cancer was 10.3% (95% CI 7.6–13.7%). The occurrence of haematuria with dysuria or increased frequency of micturition did not change the likelihood of cancer (see Table 6).

Table 6. Probability (expressed as a percentage) of urological cancer for macroscopic haematuria and additional signs and symptoms.(277).

Table 6

Probability (expressed as a percentage) of urological cancer for macroscopic haematuria and additional signs and symptoms.(277).

Risk Factors

(Morganstern, 1998)(278)

This review provides a summary of risk factors for urological cancers. Age is the principal risk factor for prostate cancer. Risk factors for the development of bladder cancer in addition to age include cigarette smoking, and occupational exposure among dye, rubber, textile and leather workers. The risk of bladder cancer with tobacco appears to be dose-dependent and partly reversible with smoking cessation, although the risk associated with occupational exposures appear to be relatively long lasting.

Most cases of renal cell carcinoma are sporadic, although a small proportion are familial and related to mutations on chromosome 3p and Von Hippel-Lindau disease. There is a moderate, dose-dependent risk associated with cigarette smoking; Increased risk is also associated with excess body weight, hypertension and/or antihypertensives, increased parity, and a variety of occupational exposures including asbestos, petroleum products, and dry cleaning solvents. Acquired cystic kidney disease with renal insufficiency also poses a risk.

(Zeegers at el, 2003)(279)

This review sought to determine the risk of prostate cancer among relatives of affected patients. Studies published up to 2002 were included, following a search in various bibliographic databases, and a random effects meta-regression model was used to undertake a meta-analysis.

33 studies were included. From the pooled findings, the relative risk among first-degree family members was 2.53 (95% CI 2.24–2.85). The risk for second-degree relatives was only slightly elevated (1.68, 95% CI 1.07–2.64). Among first-degree family members, the risk increased with the number of affected relatives and decreased with increasing age of the affected relative.

(Huyghe et al, 2003)(280)

Following a Medline search for articles published 1980 to 2002, 30 studies were included to identify trends in the incidence of testicular cancer. A trend towards an increased rate over the last 30 years was observed in the majority of industrialized countries, including North America, Europe and Oceania. There were marked differences between nearby countries, for example 2.5/100,000 in Finland and 9.2/100,000 in Denmark, as well as among regions in the same country. From the limited information available about incidence in ethnic groups, the incidence among white men in the US has increased, but this is not the case among black Americans. Worldwide, only Maori were found to have an incidence as high as that among white males.

14.2. Investigations

14.2.1. Key Clinical Question

Should any investigations be undertaken in primary care, before referral?

14.2.2. Evidence Question

In patients attending primary care services with symptoms that may be caused by cancer, which investigations when compared with the “gold standard” are predictive of a diagnosis of cancer, and which are not?

14.2.3. Evidence Statements

The PSA test is moderately sensitive and specific. (III)

Other than tests of microscopic haematuria, currently available urine tests for tumour markers are insufficiently sensitive for use in primary care. (III)

Secondary studies

Prostate cancer
(Watson et al, 2002)(281)

This guidance was published by the NHS Cancer Screening programme, and intended as advice for primary care professionals. It made clear that the PSA test is not diagnostic, and abnormal results require further investigation before a diagnosis is reached. It pointed out that 20% of men with clinically significant prostate cancer will have a normal PSA. It was advised that before having a PSA test, men should not have

  1. an active urinary tract infection,
  2. ejaculated in the previous 48 hours,
  3. exercised vigorously in the previous 48 hours, or
  4. had a prostate biopsy in the pervious 6 weeks. If practical, a PSA should be done before a DRE or delayed until a week after the DRE.

Age-specific cut-off values recommended by the Prostate Cancer Risk Management Programme were given as follows:

  • aged 50–59 years ≥ 3.0 ng/ml
  • aged 60–69 years ≥ 4.0 ng/ml
  • aged 70 years and over > 5.0 ng/ml
(Price et al, 2001)(282)

This study reported a comprehensive review of factors that influence the use of PSA in screening programmes. It identified nine studies that had investigated the effect of prior DRE on PSA levels. In the majority of studies DRE was followed by a small but statistically significant increase in PSA. In most cases this increase is not clinically significant, but in a small number the increase would have clinical significance. The review concluded that serum for PSA testing should be drawn before DRE, but if such an examination has already been performed, PSA testing should ideally be delayed for one week. The review also recommended use of age-related reference ranges to determine decision points for total PSA in men younger than 60 years as this will increase cancer detection in those likely to benefit from treatment.

(Roddam et al, 2004)(283)

In this modelling study, the impact of biased and nonequimolar assays on the decision to recommend prostate biopsy was investigated. Small deviations in bias and nonequimolarity were found to lead to significant increases in the number of false-positive biopsy recommendations, highlighting the importance of using a calibrated, unbiased equimolar assay to measure serum PSA values.

(Garnick, 1996)(284)

This review was one of a series concerned with aspects of prostate cancer. Articles published 1992–1996 were sought in a search of Medline. A largely qualitative analysis of the identified articles was undertaken.

Most of the initial screening studies that had assessed an abnormal PSA had used 4.0ng/ml as the upper limit of normal. Several studies considered methods of refining interpretation of the PSA test. The PSA density refers to a numerical ratio determined by dividing the PSA serum value by the volume of the prostate gland as determined by transrectal ultrasonography. This gives the PSA value per gram of prostate, and densities of 0.15 or more may strongly indicate the presence of cancer. However, estimation of the volume of the prostate gland is subject to error. Prostate-specific antigen velocity refers to the rate of change in the PSA value over time. A value that continues to increase over time may signal cancer. Two studies of the value of PSA velocity were included in the review, and they indicated that a change of more than 0.75ng/ml per year should be regarded with a high degree of suspicion. Recent studies have also suggested that the upper-limit of normal PSA value varies by age, being lower in younger than older men. Some preliminary studies have been undertaken of the potential role of the relative percentage of free PSA and PSA bound to serum proteins.

(Selley et al, 1997)(285)

This was a systematic review of the diagnosis, management and screening of early localised prostate cancer. PSA is a protease produced almost exclusively by prostatic epithelium. The normal range is between 0–4ng/ml, although some men with cancer have values in the normal range, and high values can be caused by conditions other than cancer. Reports of PSA sensitivity range from 57–99%, and specificity from 59–97%. The gold standard test used in studies of PSA testing is prostate biopsy, but in the primary studies not all men with elevated results would have undergone biopsy. Therefore, the true number of cancers cannot be accurately determined. The review found that evidence to support use of PSA density was equivocal, and that further research was needed into the role of PSA velocity, free and bound PSA and age-specific reference ranges for PSA normal values.

Bladder/renal cancer
(Lokeshwar and Soloway, 2001)(286)

In this systematic review, Pubmed was used to identify relevant articles. The tests included were urine cytology, haematuria detected by dipstick, and tests currently undergoing evaluation, including human complement tests, nuclear mitotic apparatus protein testing, cytology plus immunofluorescence, telomerase testing and the hyaluronic acid and hyaluronidase test.

Urine cytology was reported to have a sensitivity of 35–40% (range between studies 16–60%) for detecting bladder cancer. Haematuria can be caused by many conditions other than cancer, and therefore the specificity for cancer is low, but the sensitivity was reported to be 67–90%. There is insufficient evidence available to determine which of the other tests, or which combination of tests, can be recommended as non-invasive methods of detecting bladder cancers.

14.3. Delay and Diagnostic Difficulties

14.3.1. Key Clinical Question

What influence do age, gender, social class and ethnicity have on the differential delay at presentation?

What diagnostic difficulties do primary care practitioners themselves report in determining whether a woman/man who presents with urological symptoms/signs may or may not need urgent referral with suspected cancer?

14.3.2. Evidence Question

In people attending primary care services with symptoms or signs that might be explained by urological cancer, which psychosocial and socio-demographic factors are associated with delayed presentation? Which factors influence delay by patient and which delay by provider?

What diagnostic difficulties do primary care practitioners themselves report in determining whether a person who presents with urological symptoms/signs may or may not need urgent referral with suspected cancer?

14.3.3. Evidence Statements


This is little or no evidence on factors associated with delay in the diagnosis of testicular or penile cancer. (III)

The occurrence of macroscopic haematuria is associated with shorter patient delay in seeking advice. (III)

The occurrence of haematuria is associated with shorter delay by doctors. (III)

Diagnostic Difficulties

No evidence could be identified in order to create evidence statements.



Most of the evidence found related to bladder cancer. Based on the few studies identified, it appeared that there was no relationship between patient delay and age, gender, level of education, perceived seriousness of initial symptoms, or civil status. The type of symptom appeared to be the most important factor in determining patient delay, with haematuria as the main symptom that prompted the patient to seek urgent medical advice.

Doctor delay appeared to be longer for women than for men. Doctor delay was also greatly influenced by the nature of the presenting symptoms, with shortest delay for patients presenting with haematuria and pain than with haematuria only, and longest when urgency was the only symptom. Patient age also appeared to influence doctor delay, with younger patients being diagnosed significantly earlier than those aged 70 years and above.

The effect that delay in diagnosis may have on overall survival is an area of interest for researchers. Patients whose cancers are diagnosed early do not appear to have a clear survival advantage.

Secondary studies

No papers were identified

Primary studies
Testicular cancer
(Khadra et al, 2002)(287)

The aims of this UK study were to investigate the level of awareness of testicular cancer and practice of testicular self examination (TSE) in general practitioners’ male attendees, and to see if awareness of TSE was related to age, marital status, education, ethnicity, social class, knowing someone with testicular cancer, having attended a men’s health clinic and having heard of a testicular cancer awareness campaign.

The authors recruited men from two UK general practices, one inner city and one suburban. Questionnaires were issued to consecutively attending male patients (N = 202) between the ages of 18 and 50 years.

Although 91% of men claimed to be aware of testicular cancer, only 26% knew both the age group most affected (25–34 years) and that testicular cancer can be curable if detected early.

Forty-nine per cent of responders had carried out TSE in the past year, but only 22% did so according to recommendations, i.e. feeling for lumps on a monthly basis. TSE was associated with age >35 years, white ethnicity, having correct knowledge of testicular cancer, knowing someone with testicular cancer, having attended a men’s health clinic and having heard of a testicular cancer awareness campaign.

TSE was suggested by the media to 56% of those who examined themselves and by a nurse or general practitioner to only 16%. Forty-eight per cent of those carrying out TSE had received written instructions, and 10% had received a testicular examination by their general practitioner. Only 3% had attended a men’s health clinic in the past. Of those 103 responders not carrying out TSE, 71% said they did not know what to do, 27% said they were too busy and 2% were afraid they might discover a lump. Eighty-five per cent (169/199) of the men were keen to find out more about TSE and 67% (136/202) would attend a men’s health clinic if one were set up in their general practitioners surgery.

Bladder cancer
(Mansson et al, 1993)(288)

This Swedish study was undertaken to investigate various factors that may play a role in patient and doctor delays in the diagnosis of bladder cancer. The authors examined the clinical records of all patients with a diagnosis of bladder cancer as gathered from a regional tumour registry. Variables extracted from the records included onset date and specific pattern of symptoms, date and place of first medical consultation, referral patterns, investigations, and date of diagnosis, amongst others.

A questionnaire was sent to patients to explore how seriously the patients viewed their first symptoms of bladder cancer, their experiences of previous serious or protracted illness, and their habitual perception of bodily changes and level of general education. Patient delay was defined as the time lag from the patient’s first awareness of symptoms to the first medical consultation, and doctor delay as the interval from that consultation to establishment of correct diagnosis. In cases referred for further investigation, two phases of doctor delay were distinguished, viz. time from first consultation to first referral letter (time lag A) and from first referral to diagnosis (time lag B). The date of diagnosis was defined as the date of a first positive pathologic report.

The clinical records of 343 patients were examined, and 203 patients completed the questionnaire (88.6% of those eligible). Macroscopic haematuria was the commonest symptom bringing the patient to the doctor. Urgency was more common in advanced than in superficial cancer (51% vs.34%, P<0.002). No correlation was found between presence of haematuria and tumour category.

161 (67%) patients initially consulted a in primary care (mostly a general practitioner) and 51 (15%) a private practice (mostly a general practitioner or gynaecologist). The remaining 118 patients presented at a hospital. Three patients (1%) never sought medical advice and were diagnosed at postmortem examination.

The median patient delay was 15 days (mean 141, range 0–2,857). There was no relationship between this delay and age or gender.

The type of symptom was an important factor in patient delay. Haematuria prompted the patient to seek medical advice more quickly than either urgency or pain (median 5 vs. 45 and 38 days respectively, P<0.001). Although the difference was not statistically significant, median patient delay was longer in patients with advanced cancer than in those with superficial tumour.

Amongst the responders to the questionnaire, no correlation was demonstrable between patient delay and level of education, perceived seriousness of initial symptoms, or civil status.

The median doctor delay was 62 days overall. It was longer for women than for men (76 vs. 59 days, P<0.05). The health service first consulted was a major factor in doctor delay (P<0.001), delay varying from a median of 78 days for patients initially seen in a primary care unit to a median of 21 days when the patient directly attended a department of urology, but the longer median delay was not due to delayed referral to a specialist, since in the total series doctor delay phase A was only six days, whereas phase B was 47 days (indicating considerable waiting time in the referral system).

The use of urine cytology and intravenous urography in general or private practice was associated with some, but not significant, shortening of doctor delay. The nature of the presenting symptoms influenced doctor delay, with delay being shorter with haematuria plus pain than with haematuria only, and longest when urgency was the only symptom (median 44, 53 and 114 days, P<0.001).

Patient age was associated with doctor delay. The median delay was less in patients younger than 70 years than in older patients, viz. 54 and 69 days (P<0.01).

(Wallace et al, 2002)(289)

The aim of this UK study was to assess the effect on survival of delays in the diagnosis and treatment of bladder cancer (dividing the delay from onset of symptoms to first treatment into several components, comprising patient delay, general practitioner delay, and two or more periods of hospital delay. The authors sought to collect data prospectively on all newly diagnosed cases of urothelial cancer (N=1,537) in the West Midlands from January 1991 to June 1992. The data collected included the dates of onset of symptoms, first referral by the general practitioner, first hospital appointment and first definitive treatment. Clinical details collected included the presence or absence of haematuria (macroscopic or microscopic), the number, size and type of tumours, and the findings on bimanual examination. Details of patient characteristics were also collected. In addition, patients were asked to complete a questionnaire on their smoking and occupational history.

Delay times were calculated as follows: date of onset of symptoms to date of first referral by a general practitioner (delay 1); date of general practitioner referral to date of first attendance at hospital (delay 2); date of first hospital attendance to date of first treatment by TURBT (delay 3); date of general practitioner referral to date of first treatment (hospital delay = delay 2 + 3); date of onset of symptoms to date of first treatment (total delay = delay 1 + 2 + 3). The date of first definitive treatment was the date of the diagnostic TURBT.

The median (IQR) Delay 1 was 14 (0–61) days. Patients with a longer delay were more likely to present with a higher stage tumour (P=0.04). Patients with an unknown haematuria status were more likely to have a shorter delay (P<0.001). No other patient or tumour characteristics showed a significant difference above or below the median delay. Delay 1 had a significant effect on survival; patients with a delay of <14 days to referral had an improved survival of 5% at 5-years compared with those who had a delay of >14 days (P=0.02). Adjusting for tumour stage, there was a trend for patients with a shorter Delay 1 to have a better survival (P=0.06).

The median Delay 2 was 28 (7–61) days. Patients known to have had macroscopic haematuria (N=1032) were more likely to have a shorter delay than those known to have had microscopic haematuria (N=70); patients with an unknown haematuria status were more likely to have a longer delay (P<0.001). There were no other significant differences in patient or tumour characteristics above or below the median delay. Patients who had a shorter Delay 2 had a significantly worse survival (P=0.001). Survival by Delay 2 after adjusting for tumour stage similarly showed that patients with a shorter Delay 2 had significantly worse survival (P=0.001).

The median total delay was 110 (62–209) days. Longer delays were significantly associated with women (P=0.05), younger patients (P=0.03), non-smokers (P=0.04) and patients with a low risk of occupational exposure (P=0.04). No other patient or tumour characteristics showed significant differences above or below the median delay. The total delay had no effect on survival (P=0.17); this was also true after adjusting for tumour stage (P=0.43).

For prognostic factors, there was no survival difference for sex (P=0.92), haematuria (P=0.39) and number of tumours (P=0.13), both in the log-rank analysis and Cox regression models.

(Mommsen et al, 1983)(290)

The purpose of this Danish study was to elucidate causes of delay in the diagnosis of bladder cancer.

The authors interviewed patients (N=212) with newly diagnosed bladder tumour admitted consecutively to a department of oncology and radiotherapy during a three year period beginning in 1977. The interview concerned symptoms, some demographic variables and the time intervals under study (phases A, B, and C). Phase A covered the interval between onset of the presenting symptom and the consultation with the general practitioner. Phase B was the interval between this consultation and the patient’s first examination at the local hospital. Phase C was the interval between the hospital examination and initiation of definitive treatment.

The presenting symptom was haematuria, which commonly was painless, in 79% of the patients. The interval from onset of symptoms until treatment averaged 28 weeks (median = 15 weeks). The general practitioner delay comprised half of the total delay.

Half of the patients consulted their general practitioner within a week after onset of the presenting symptom. A higher percentage of men than of women delayed 13 weeks or more.

Fewer women than men (62% and 82%) were referred to hospital within 12 weeks of the index consultation with the general practitioner (χ2 = 8.97; d.f.=1; P <0.005). Of the patients with haematuria, 13% of the men but 35% of the women were referred to hospital after 13 weeks or more (χ2 =9.70; d.f.=1; P<0.005). Cystitis as the presenting symptom was associated with later referral to hospital than haematuria; this was most pronounced for men (χ2=12.56; d.f.=1; P<0.005).

(Wallace et al, 1999)(291)

The authors of this UK study examined the relationship between delay in presentation of patients with bladder cancer and tumour stage and material deprivation.

Data on delay periods to treatment, tumour characteristics, occupation and postcodes were collected for patients with urothelial cancer presenting to a Regional Cancer Intelligence Unit. The Townsend material deprivation score was derived from the patient’s postcode (the score assesses four variables measuring unemployment, overcrowding, wealth and income).

A delay of < two weeks in referral to hospital was associated with a 6% improvement in survival (P = 0.018); shorter delays to hospital appointment correlated inversely with survival (P< 0.001). The overall delay time and delay to hospital admission did not correlate with survival. The deprivation scores showed no correlation with delay times, smoking or T-category of tumour. Material deprivation was correlated with low tumour grade (P = 0.004) and better survival (P = 0.02).



The age-specific cut-off PSA measurements recommended by the Prostate Cancer Risk Management Programme are as follows: aged 50–59 years ≥ 3.0 ng/ml; aged 60–69 years ≥ 4.0 ng/ml; aged 70 years and older ≥ 5.0 ng/ml. (Note that there are no age-specific reference ranges for men aged over 80 years. Nearly all men of this age have at least a focus of cancer in the prostate. Prostate cancer only needs to be diagnosed in this age group if it is likely to need palliative treatment.)

Copyright © 2005, National Collaborating Centre for Primary Care.
Bookshelf ID: NBK45775


  • PubReader
  • Print View
  • Cite this Page
  • PDF version of this title (3.0M)

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...