- Email: [email protected]
Brain Tumours: a Retrospective Audit
Clinical Oncology (2003) 15: 7–9 doi:10.1053/clon.2002.0133
Original Article Urgent 2-week Referrals for CNS/Brain Tumours: a Retrospective Audit D. S. N. A. Pengiran Tengah*, P. O. Byrne†, A. J. Wills‡ *Department of Neurology, Derbyshire Royal Infirmary, Derby DE1 2QY, U.K.; †Department of Neurosurgery, ‡Department of Neurology, University Hospital, Queen’s Medical Centre, Nottingham, NG7 2UH, U.K. ABSTRACT: Objective: To assess the effectiveness of the 2-week referral system for CNS/brain tumours and to contrast this with the number of patients with neurological cancers identified independently of this system. Method: Retrospective casenote review of patients referred to emergency neurology clinics pre-implementation of the 2-week referral system. Retrospective review of GP referral letters via this system and comparison to Department of Health referral guidelines. Review of corresponding casenotes to determine the actual neurological diagnosis. Identification of patients with CNS/brain tumours diagnosed independently of this system from a local CNS cancer register. Results: Over a 3-month period pre-implementation of the referral system, of 12 patients referred as emergencies, none had CNS/brain cancer. Forty-three patients were referred via this system over a 9-month period to neurology departments of a teaching hospital and a district general hospital. Thirty per cent of the referrals did not follow the Department of Health guidelines. Only 9% actually had CNS tumours (two astrocytomas, two cerebral metastases). The remainder were diagnosed with chronic daily headache (10), epilepsy (5), migraine (3), demyelination (2), essential tremor (2), other (17). During this period at least 69 neurological cancers were identified independently of the 2-week system. Conclusion: These guidelines may increase diagnostic precision if adhered to rigidly. Inappropriate referrals have extended already lengthy outpatient waiting times in other specialities. We suggest early re-consideration of these guidelines and further study for earlier identification of CNS cancer. Pengiran Tengah D. S. N. A. et al. (2003). Clinical Oncology 15, 7–9 2002 The Royal College of Radiologists. Published by Elsevier Science Ltd. All rights reserved. Key words: Audit, CNS/brain tumour, 2-week referral Received: 25 March 2002
Revised: 5 June 2002
Introduction
In response to the Government’s White Paper entitled ‘The New NHS – Modern, Dependable’, the Department of Health (DoH), in association with the Royal Colleges, produced referral guidelines for suspected cancer aimed at general practitioners (GPs) ‘to facilitate appropriate referral between primary and secondary care . . .’.[1] Thus, patients who GPs suspect of having cancer, are guaranteed assessment by a specialist within 2 weeks. These guidelines were devised by Working Parties (chaired by an expert in the relevant cancer area and consisting of a multi-disciplinary team) based on expert judgement and consensus of best available published literature. Unfortunately, the evidence base for pre-emptive diagnosis of neurological cancers is not broad. Although malignant CNS tumours are Address for correspondence: Dr AJ Wills, Consultant Neurologist, Department of Neurology, University Hospital, Queen’s Medical Centre, Nottingham, NG7 2 UH, U.K. Tel: 0115-9709141; Fax: 0115-9709493; E-mail: [email protected] 0936–6555/03/010007+03 $30.00/0
Accepted: 26 June 2002
uncommon, early detection in some tumour types might have implications for prognosis [2].
Subjects and Methods
Our cohort of patients was referred to the Departments of Neurology at Queen’s Medical Centre (QMC), Nottingham, a tertiary referral centre, and the Derbyshire Royal Infirmary (DRI), Derby, a large district general hospital (serving a population of 570,000), via the 2-week urgent referral system during a 9-month period (July 2000–April 2001) from implementation of the system. We reviewed the referral letters of these patients and compared them with criteria set out in the guidelines produced by the DoH (Table 1). Patient casenotes were then studied to determine the specialist diagnosis. Prior to implementation of this system, traditionally GPs were able to contact the Neurology Department in Derby or Nottingham directly either by telephone or facsimile regarding patients that they deemed needed to
2002 The Royal College of Radiologists. Published by Elsevier Science Ltd. All rights reserved.
8
Table 1 – Referral guidelines for suspected brain/CNS cancer1 Brain tumours: Guidelines for urgent referrals v Subacute progressive neurological deficit developing over days to weeks (e.g. weakness, sensory loss, dysphasia, ataxia) v New onset seizures characterized by one or more of the following – Focal seizures – Prolonged post-ictal focal deficit (longer than 1 h) – Status epilepticus – Associated inter-ictal focal deficit v Patients with headache, vomiting and papilloedema. v Cranial nerve palsy (e.g. diplopia, visual failure including opticiandefined visual field loss, unilateral sensorineural deafness). Consider urgent referral for: Patients with non-migrainous headaches of recent onset, present for at least 1 month, when accompanied by features suggestive of raised intra cranial pressure (e.g. woken by headache; vomiting; drowsiness) 1
Department of Health. Referral Guidelines for suspected cancer. http://www.doh.gov.uk/cancer/referral.htm 2000.
be seen urgently. However this system was not specifically designed to detect CNS cancers but to offer early diagnosis in ‘serious’ neurological disease. If this referral was felt to be urgent by the Consultant Neurologist, the patient could be seen at the weekly emergency neurology clinic. Therefore as a comparison we investigated patients who were referred to this clinic. All patients with CNS cancer in this region are listed in a local CNS cancer register and we were able to ascertain the number of patients with CNS cancer not referred by this system. Results
Forty-three patients were referred during this 9 month period (32 referred to QMC and 11 to DRI). Thirteen (30%) of the 43 referrals did not follow the DoH referral guidelines. Examples included patients with headache but no features of raised intra-cranial pressure, two patients with new onset tremor and a patient with glove-and-stocking paraesthesiae. Four of the 43 (9%) were found to have cancer (two with malignant astrocytoma and two with cerebral metastases from presumed lung primaries). These four patients all had headache. In addition, one patient had altered consciousness and personality change, one had unsteady gait and weight loss, one had visual field loss and clumsiness of the left hand and one had debility plus abnormal visual fields and optic discs. These all conformed to the referral guidelines. The rest of the patients had a variety of diagnoses. These were: chronic daily headache (10), epilepsy (5) and migraine (3) demyelination (2), essential tremor (2), cervical myelopathy (2), ENT causes (2), cerebrovascular disease (2), hypertension (1), trigeminal sensory neuropathy (1), Alzheimer’s disease (1), depression (1), peripheral neuropathy (1), hepatic encephalopathy (1), no fixed diagnosis given but space-occupying lesion ruled out (5).
At least 69 patients with CNS cancer, within the same catchment area, were identified independently of this system according to our regional CNS cancer register. Prior to implementation of the 2-week referral system, 12 patients were referred as emergencies within a 3–month period, with diagnoses that included epilepsy (3) and chronic daily headache (2). None had CNS/brain cancer.
Discussion
Clearly, the White Paper arose in response to public concern at delays in cancer diagnosis and treatment. However we believe that particularly for CNS/brain cancers, these guidelines are potentially harmful. This is mainly because CNS cancers present in a diverse manner and eliciting a neurological history and interpreting neurological signs can be challenging. GPs may receive little neurological training. Member of Parliament, Desmond Swayne, highlighted the need for improved neurological training within primary care in the recent House of Commons Hansard Debates focusing on neurology [3]. Moreover this system is open to misuse as demonstrated by 30% of referrals not adhering to referral guidelines. CNS cancers often present late and have a poor prognosis at the time of presentation. In high-grade glioma, early diagnosis does not substantially affect survival. Conversely non-glial tumours, e.g. CNS lymphoma, are often amenable to chemotherapy and may even be curable [2]. Our efforts in improving cancer mortality in this field should therefore be concentrated on more effective diagnosis. This should include thorough epidemiological studies and an adequate evidence base for earlier diagnosis. The DoH guidelines are necessarily broad but this is potentially problematic. Thus, subacute sensory loss developing over weeks is highlighted as a potential symptom of underlying malignancy but could equally be caused by carpal tunnel syndrome. In Scotland this potential ambiguity has been addressed by the addition of the condition ‘in the absence of previously diagnosed or suspected alternative disorders’. The guidelines are also slightly ambiguous. It is noteworthy that in the DoH guidelines mental state change is not mentioned as a separate category for referral, though an alternative interpretation would be that mental state change is a subacute neurological deficit (See Table 1). However in a specimen referral form and in the body of the DoH guidelines, memory loss and personality change is included as a presenting feature of neurological cancer. With regards to other specialities, this system may have its merits, albeit at the expense of increasing waiting times for routine referrals. A recent study looking at 25 gastrointestinal 2-week referrals showed that 40% of these patients had ‘serious illness’ (15% with malignancy and 25% with serious non-malignant disease) [4]. A study looking at the effectiveness of this
2- / system in breast cancer actually found a drop in cancer detection rate and an increase in both urgent and non-urgent waiting times [5]. Prior to implementation of the referral system, we had no fixed system for cancer referral although GP referrals were vetted and those deemed urgent were seen early (within 1 week). Arguably a hit-rate of 9% is sufficient particularly as analysis of 3 months of emergency clinics pre-implementation did not identify any cases of CNS cancer. However, our primary concerns are insufficient manpower to meet the 2-week recommendation and that the lengthy waiting times in Neurology outpatient clinics may be extended further. The latter is particularly worrying in light of the fact that 69 patients with CNS cancer were identified independently of this referral system. The DoH guideline document correctly stresses the importance of audit, and states that information collected will inform subsequent revisions. However, we have shown that with CNS/brain tumours this system is falling short of expectations and this needs to be addressed. We propose early re-assessment of the 2-week policy, with particular reference to ‘hit-rate’, effects on routine waiting times and more thorough study of cancers diagnosed
9
independently of the system. There may be a case for excluding certain specialities such as neurology from this system. Perhaps, a method of closer communication between primary and secondary care would be a more efficient and effective use of resources. Acknowledgements A portion of this work was presented at the Association of British Neurologists Autumn Meeting, Durham, England; 12–14 September 2001. References 1. Department of Health. Referral Guidelines for suspected cancer. http://www.doh.gov.uk/cancer/referral.htm 2000. 2. Rampling R. Cancer Scenarios: An aid to planning cancer services in Scotland in the next decade-Brain and CNS tumours. http:// www.show.sct.nhs.uk/sehd/publications/csatp/csatp-19.htm 2001. 3. Swayne D. House of Commons Hansard Debate. London: House of Commons; 2001 6 March 2001. 4. Moreea S, Green J, MacFie J, Mitchell C. Impact of a two-week waiting time standard on the gastroenterology service of a District General Hospital. Br J Surg 2001;88(Suppl 1):68. 5. Khawaja AR, Allan SM. Has the breast cancer ‘two week wait’ guarantee for assessment made any difference? Eur J Surg Oncol 2000;26:536–539.
Original Article Urgent 2-week Referrals for CNS/Brain Tumours: a Retrospective Audit D. S. N. A. Pengiran Tengah*, P. O. Byrne†, A. J. Wills‡ *Department of Neurology, Derbyshire Royal Infirmary, Derby DE1 2QY, U.K.; †Department of Neurosurgery, ‡Department of Neurology, University Hospital, Queen’s Medical Centre, Nottingham, NG7 2UH, U.K. ABSTRACT: Objective: To assess the effectiveness of the 2-week referral system for CNS/brain tumours and to contrast this with the number of patients with neurological cancers identified independently of this system. Method: Retrospective casenote review of patients referred to emergency neurology clinics pre-implementation of the 2-week referral system. Retrospective review of GP referral letters via this system and comparison to Department of Health referral guidelines. Review of corresponding casenotes to determine the actual neurological diagnosis. Identification of patients with CNS/brain tumours diagnosed independently of this system from a local CNS cancer register. Results: Over a 3-month period pre-implementation of the referral system, of 12 patients referred as emergencies, none had CNS/brain cancer. Forty-three patients were referred via this system over a 9-month period to neurology departments of a teaching hospital and a district general hospital. Thirty per cent of the referrals did not follow the Department of Health guidelines. Only 9% actually had CNS tumours (two astrocytomas, two cerebral metastases). The remainder were diagnosed with chronic daily headache (10), epilepsy (5), migraine (3), demyelination (2), essential tremor (2), other (17). During this period at least 69 neurological cancers were identified independently of the 2-week system. Conclusion: These guidelines may increase diagnostic precision if adhered to rigidly. Inappropriate referrals have extended already lengthy outpatient waiting times in other specialities. We suggest early re-consideration of these guidelines and further study for earlier identification of CNS cancer. Pengiran Tengah D. S. N. A. et al. (2003). Clinical Oncology 15, 7–9 2002 The Royal College of Radiologists. Published by Elsevier Science Ltd. All rights reserved. Key words: Audit, CNS/brain tumour, 2-week referral Received: 25 March 2002
Revised: 5 June 2002
Introduction
In response to the Government’s White Paper entitled ‘The New NHS – Modern, Dependable’, the Department of Health (DoH), in association with the Royal Colleges, produced referral guidelines for suspected cancer aimed at general practitioners (GPs) ‘to facilitate appropriate referral between primary and secondary care . . .’.[1] Thus, patients who GPs suspect of having cancer, are guaranteed assessment by a specialist within 2 weeks. These guidelines were devised by Working Parties (chaired by an expert in the relevant cancer area and consisting of a multi-disciplinary team) based on expert judgement and consensus of best available published literature. Unfortunately, the evidence base for pre-emptive diagnosis of neurological cancers is not broad. Although malignant CNS tumours are Address for correspondence: Dr AJ Wills, Consultant Neurologist, Department of Neurology, University Hospital, Queen’s Medical Centre, Nottingham, NG7 2 UH, U.K. Tel: 0115-9709141; Fax: 0115-9709493; E-mail: [email protected] 0936–6555/03/010007+03 $30.00/0
Accepted: 26 June 2002
uncommon, early detection in some tumour types might have implications for prognosis [2].
Subjects and Methods
Our cohort of patients was referred to the Departments of Neurology at Queen’s Medical Centre (QMC), Nottingham, a tertiary referral centre, and the Derbyshire Royal Infirmary (DRI), Derby, a large district general hospital (serving a population of 570,000), via the 2-week urgent referral system during a 9-month period (July 2000–April 2001) from implementation of the system. We reviewed the referral letters of these patients and compared them with criteria set out in the guidelines produced by the DoH (Table 1). Patient casenotes were then studied to determine the specialist diagnosis. Prior to implementation of this system, traditionally GPs were able to contact the Neurology Department in Derby or Nottingham directly either by telephone or facsimile regarding patients that they deemed needed to
2002 The Royal College of Radiologists. Published by Elsevier Science Ltd. All rights reserved.
8
Table 1 – Referral guidelines for suspected brain/CNS cancer1 Brain tumours: Guidelines for urgent referrals v Subacute progressive neurological deficit developing over days to weeks (e.g. weakness, sensory loss, dysphasia, ataxia) v New onset seizures characterized by one or more of the following – Focal seizures – Prolonged post-ictal focal deficit (longer than 1 h) – Status epilepticus – Associated inter-ictal focal deficit v Patients with headache, vomiting and papilloedema. v Cranial nerve palsy (e.g. diplopia, visual failure including opticiandefined visual field loss, unilateral sensorineural deafness). Consider urgent referral for: Patients with non-migrainous headaches of recent onset, present for at least 1 month, when accompanied by features suggestive of raised intra cranial pressure (e.g. woken by headache; vomiting; drowsiness) 1
Department of Health. Referral Guidelines for suspected cancer. http://www.doh.gov.uk/cancer/referral.htm 2000.
be seen urgently. However this system was not specifically designed to detect CNS cancers but to offer early diagnosis in ‘serious’ neurological disease. If this referral was felt to be urgent by the Consultant Neurologist, the patient could be seen at the weekly emergency neurology clinic. Therefore as a comparison we investigated patients who were referred to this clinic. All patients with CNS cancer in this region are listed in a local CNS cancer register and we were able to ascertain the number of patients with CNS cancer not referred by this system. Results
Forty-three patients were referred during this 9 month period (32 referred to QMC and 11 to DRI). Thirteen (30%) of the 43 referrals did not follow the DoH referral guidelines. Examples included patients with headache but no features of raised intra-cranial pressure, two patients with new onset tremor and a patient with glove-and-stocking paraesthesiae. Four of the 43 (9%) were found to have cancer (two with malignant astrocytoma and two with cerebral metastases from presumed lung primaries). These four patients all had headache. In addition, one patient had altered consciousness and personality change, one had unsteady gait and weight loss, one had visual field loss and clumsiness of the left hand and one had debility plus abnormal visual fields and optic discs. These all conformed to the referral guidelines. The rest of the patients had a variety of diagnoses. These were: chronic daily headache (10), epilepsy (5) and migraine (3) demyelination (2), essential tremor (2), cervical myelopathy (2), ENT causes (2), cerebrovascular disease (2), hypertension (1), trigeminal sensory neuropathy (1), Alzheimer’s disease (1), depression (1), peripheral neuropathy (1), hepatic encephalopathy (1), no fixed diagnosis given but space-occupying lesion ruled out (5).
At least 69 patients with CNS cancer, within the same catchment area, were identified independently of this system according to our regional CNS cancer register. Prior to implementation of the 2-week referral system, 12 patients were referred as emergencies within a 3–month period, with diagnoses that included epilepsy (3) and chronic daily headache (2). None had CNS/brain cancer.
Discussion
Clearly, the White Paper arose in response to public concern at delays in cancer diagnosis and treatment. However we believe that particularly for CNS/brain cancers, these guidelines are potentially harmful. This is mainly because CNS cancers present in a diverse manner and eliciting a neurological history and interpreting neurological signs can be challenging. GPs may receive little neurological training. Member of Parliament, Desmond Swayne, highlighted the need for improved neurological training within primary care in the recent House of Commons Hansard Debates focusing on neurology [3]. Moreover this system is open to misuse as demonstrated by 30% of referrals not adhering to referral guidelines. CNS cancers often present late and have a poor prognosis at the time of presentation. In high-grade glioma, early diagnosis does not substantially affect survival. Conversely non-glial tumours, e.g. CNS lymphoma, are often amenable to chemotherapy and may even be curable [2]. Our efforts in improving cancer mortality in this field should therefore be concentrated on more effective diagnosis. This should include thorough epidemiological studies and an adequate evidence base for earlier diagnosis. The DoH guidelines are necessarily broad but this is potentially problematic. Thus, subacute sensory loss developing over weeks is highlighted as a potential symptom of underlying malignancy but could equally be caused by carpal tunnel syndrome. In Scotland this potential ambiguity has been addressed by the addition of the condition ‘in the absence of previously diagnosed or suspected alternative disorders’. The guidelines are also slightly ambiguous. It is noteworthy that in the DoH guidelines mental state change is not mentioned as a separate category for referral, though an alternative interpretation would be that mental state change is a subacute neurological deficit (See Table 1). However in a specimen referral form and in the body of the DoH guidelines, memory loss and personality change is included as a presenting feature of neurological cancer. With regards to other specialities, this system may have its merits, albeit at the expense of increasing waiting times for routine referrals. A recent study looking at 25 gastrointestinal 2-week referrals showed that 40% of these patients had ‘serious illness’ (15% with malignancy and 25% with serious non-malignant disease) [4]. A study looking at the effectiveness of this
2- / system in breast cancer actually found a drop in cancer detection rate and an increase in both urgent and non-urgent waiting times [5]. Prior to implementation of the referral system, we had no fixed system for cancer referral although GP referrals were vetted and those deemed urgent were seen early (within 1 week). Arguably a hit-rate of 9% is sufficient particularly as analysis of 3 months of emergency clinics pre-implementation did not identify any cases of CNS cancer. However, our primary concerns are insufficient manpower to meet the 2-week recommendation and that the lengthy waiting times in Neurology outpatient clinics may be extended further. The latter is particularly worrying in light of the fact that 69 patients with CNS cancer were identified independently of this referral system. The DoH guideline document correctly stresses the importance of audit, and states that information collected will inform subsequent revisions. However, we have shown that with CNS/brain tumours this system is falling short of expectations and this needs to be addressed. We propose early re-assessment of the 2-week policy, with particular reference to ‘hit-rate’, effects on routine waiting times and more thorough study of cancers diagnosed
9
independently of the system. There may be a case for excluding certain specialities such as neurology from this system. Perhaps, a method of closer communication between primary and secondary care would be a more efficient and effective use of resources. Acknowledgements A portion of this work was presented at the Association of British Neurologists Autumn Meeting, Durham, England; 12–14 September 2001. References 1. Department of Health. Referral Guidelines for suspected cancer. http://www.doh.gov.uk/cancer/referral.htm 2000. 2. Rampling R. Cancer Scenarios: An aid to planning cancer services in Scotland in the next decade-Brain and CNS tumours. http:// www.show.sct.nhs.uk/sehd/publications/csatp/csatp-19.htm 2001. 3. Swayne D. House of Commons Hansard Debate. London: House of Commons; 2001 6 March 2001. 4. Moreea S, Green J, MacFie J, Mitchell C. Impact of a two-week waiting time standard on the gastroenterology service of a District General Hospital. Br J Surg 2001;88(Suppl 1):68. 5. Khawaja AR, Allan SM. Has the breast cancer ‘two week wait’ guarantee for assessment made any difference? Eur J Surg Oncol 2000;26:536–539.