This site is intended for health professionals only

At the heart of general practice since 1960


Cancer expert Dr Nick Brown looks at recent papers that have caught his eye

Cancer expert Dr Nick Brown looks at recent papers that have caught his eye

Colorectal cancer

NICE guidance advises urgent referral of patients aged over 40 with six weeks of rectal bleeding and diarrhoea, and referral of patients aged 60 or more with rectal bleeding for six weeks with anal symptoms or diarrhoea.

Should we follow NICE guidance on when to refer?


Risk in primary care of colorectal cancer from new onset rectal bleeding: 10-year prospective study. BMJ 2006; 333: 69-70

Method The percentage of participants in whom colorectal cancer or colonic adenoma was identified, following the investigation of all patients aged 45 or more who presented with a new onset of rectal bleeding in a rural four-doctor practice with 4,426 patients. In this prospective study, all patients were treated, according to a protocol, by rigid sigmoidoscopy followed by barium enema, flexible sigmoidoscopy or colonoscopy.

Results A total of 15 colorectal cancers (5.7 per cent) and 13 adenomas (4.9 per cent) were identified in the 265 patients studied, giving an incidence of neoplasia of one in 10 patients presenting with these symptoms. This represented 40 per cent of all colorectal neoplasms diagnosed during this period, as overall the practice had 38 colorectal cancers and 33 adenomas during this period, giving an overall incidence of 84 in 100,000. This is in line with national statistics apart from fewer than expected incidence in the oldest age group (>75).

Conclusions This paper challenges the current NICE guidelines.

What I will do now I will offer the option of referral for further investigations for all patients aged over 45 with new onset rectal bleeding.

What are the most significant predictors of colorectal cancer?


Predicting colorectal cancer risk in patients with rectal bleeding. BJGP 2006;56: 763-767

Methods Symptom data were collected using a self-completed questionnaire on 604 patients referred with rectal bleeding to an open-access community-based nurse flexible endoscopy clinic in southern England. Cases with known ulcerative colitis and patients on warfarin were excluded and the bowel symptom data were analysed.

Results Some 22 cases were diagnosed with colorectal cancer (3.6 per cent). There was a non-significant increased incidence in males, and older age groups were more likely to have colorectal cancer. The presence of either dark blood, or blood mixed with stool, were the only symptoms associated with a higher risk of colorectal cancer. While the presence of haemorrhoids reduced the likelihood of colorectal cancer, it was still present in three (2 per cent) of these patients.

Conclusions In a population of patients with rectal bleeding referred for further investigations, the most significant predictor of a diagnosis of cancer is age. The only other significant predictor is patient-reported rectal bleeding mixed with stool. Where patients have an isolated symptom of bright red rectal bleeding, the presence of haemorrhoids does not significantly reduce the risk of the presence of cancer.

What I will do now I will pay particular attention to older patients with colorectal symptoms, to patients with blood mixed in with stool and I will attach little diagnostic importance to the presence of haemorrhoids.

Prostate cancer

Screening of asymptomatic men using PSA is not recommended in the UK and therefore most cancers are diagnosed after presentation to primary care with symptoms. Genitourinary symptoms are a common presentation in general practice – but what are the most significant symptoms?

Which clinical features are most likely in prostate cancer?


Clinical features of prostate cancer before diagnosis. BJGP 2006; 56: 756-762

Methods All patients aged 40 and over with a diagnosis of carcinoma of the prostate were identified from the cancer registry at the Royal Devon and Exeter Hospital and 21 participating practices for a five-year period. Each was matched with five male controls and the primary care records for the two years prior to diagnosis, and those of the controls, were examined for reported symptoms.

Results Some 248 cases of prostate cancer were identified, of which 31 were excluded due to factors such as cancer recurrence, presence of other cancers or inability to access the GP record. Some 6.5 per cent of patients had no previous GP consultations within this two-year time scale.

Urinary retention, frequency, hesitancy, nocturia, weight loss, impotence, and haem-aturia were independently associated with prostate cancer. Most of the cases had lower urinary tract symptoms and this was the main reason that their prostate cancer was uncovered.

The first four of these symptoms probably represent enlargement of the prostate gland and of these, urinary retention has the strongest association with prostate cancer, but this clinical scenario is usually managed by secondary care. The other three symptoms were also recorded in 6-9 per cent of controls. The association of impotence and prostate cancer was a surprising finding not previously recorded in primary care.

Weight loss is a non organ-specific symptom of all forms of cancer and haematuria has an association with all forms of urological cancer. An abnormal rectal examination had a stronger association with prostate cancer than any of the individual symptoms.

Conclusions GPs can discriminate between a benign and malignant feeling prostate on examination. The discovery of cancer of the prostate in a patient with symptoms alters the management and there seems little advantage in delaying treatment in patients who are symptomatic with cancer of the prostate. Such patients should all have a PSA test and the relative risk of having prostate cancer can, if necessary, be discussed with the patient using the data from this paper.

What I will do now The paper has a useful predictive table I will use to inform patients when they present with symptoms. All patients presenting with any of the symptoms should have a PR examination and a PSA test and this should not be considered to be a primary screening diagnostic process.

Lung cancer

Lung cancer is one of the common cancers in men and women. But GPs may see only one new case per year, while the core symptoms, and in particular cough and dyspnoea, are very common. Initial investigation for lung cancer is a chest X-ray and this accounts for 25 per cent of diagnostic procedures or 236 per 1,000 patients per year – but how effective are chest X-rays in excluding or diagnosing lung cancer?

How important are chest X-rays in lung cancer diagnosis?


Chest radiology in general practice: indications, diagnostic yield and consequences for patient management. BJGP 2006; 56: 574-8

Methods Some 792 patients referred by 78 GPs for chest X-rays to three hospitals in the Netherlands were studied and the outcome of change of patient management was assessed by means of questionnaires filled in by the GPs before and after the investigation.

Results Clinically relevant abnormalities were found in 24 per cent of all chest X-rays and management changed in 60 per cent of patients following X-ray, including fewer referrals to a specialist, fewer prescriptions of antibiotics and increased reassurance to patients.

Conclusion Chest X-ray is an important diagnostic tool for GPs and is cost-effective.

What I will do now Chest X-rays continue to be a useful diagnostic tool in investigating patients with symptoms and signs of possible lung cancer, and I will continue to follow the recommendations of the UK referral guidelines for suspected cancer.

How should GPs interpret negative chest X-rays if symptoms persist?


Negative chest X-rays in primary care patients with lung cancer. BJGP 2006; 56:570-573

Methods Some 260 cases of primary lung cancer living in the Exeter PCT boundary were identified within a four-and-half-year period from the cancer registry and from primary care records. The records were then examined for the reporting of any symptoms and also the results of any chest X-rays.

Results Some 66 per cent of the cases included in the study had had at least one chest X-ray requested from primary care in the 12 months prior to diagnosis. Some 23 per cent of these X-rays were reported as negative to cancer – 10 per cent were reported as completely normal and 13 per cent were abnormal with no malignancy suspected. The proportion of normal X-rays reported increased the further from the date of diagnosis and this was particularly significant over six months. Negative X-rays occurred with all the common symptoms of cancer apart from hoarseness.

Conclusion Most negative chest X-rays will be true negatives, but in the presence of continuing symptoms, doctors should not be reassured by a negative result, particularly if it was over six months ago.

What I will do now There is a significant false negative rate for chest X-ray for the detection of early cancer and where symptoms persist, I will instruct the patient to inform me and I will take further action.

What should GPs do in suspected lung cancer in the face of waiting lists for chest X-rays?


Delay in diagnosis of lung cancer in general practice. BJGP 2006; 56: 863-8

Methods Some 142 patients diagnosed with lung cancer were identified from the cancer registry in the county of Aarhus in Denmark during a six-month period. A review of delay was performed using patient and GP interviews to determine the delay from the presentation of the first relevant symptom to referral to secondary care and the reasons for this delay determined.

Results Some 92 patients participated and the median delay was 33 days. Cough was the most common symptom in 31 per cent of patients but 17 per cent had atypical symptoms. Some 34 per cent of patients had a chest X-ray requested, of which 21 per cent had no suspicion of cancer on the chest X-ray, The most frequent recorded reason for system delay was waiting for X-rays.

Lack of specific follow-up instructions was a principal reason for delay in the 10 per cent of patients who waited up to seven months for follow-up. A false negative chest X-ray was also a principal cause of prolonged delay of diagnosis. Blood tests only helped the diagnostic process in a few cases and more often contributed to further delay.

Conclusion One-third of patients who contacted their GP with signs and symptoms of lung cancer were referred immediately to secondary care and these patients had the shortest diagnostic delay. Patients referred for chest X-ray experienced some delay and this was most marked in those with a false negative chest X-ray result. Lack of specific follow-up instructions was an important and avoidable source of delay which was quite significant in a number of patients.

What I will do now The organisation and processing of X-rays may lead to delay and if I suspect a diagnosis of lung cancer on purely clinical grounds, then I will directly refer to a chest physician rather than ordering a chest X-ray. The effectiveness of all courses of action require good patient understanding of follow-up arrangements and precise time scales and I will ensure that the patient clearly understands what to expect.

Nick Brown is a GP and GP trainer in Chippenham, and former primary care cancer lead for Kennet and North Wiltshire PCT before reorganisation

Competing interests None declared

Rate this article 

Click to rate

  • 1 star out of 5
  • 2 stars out of 5
  • 3 stars out of 5
  • 4 stars out of 5
  • 5 stars out of 5

0 out of 5 stars

Have your say