Headache Suspicious of Cancer

​For information on headaches where brain tumour is not likely to be the cause see https://apps.nhslothian.scot/refhelp/Neurology/Headache opens a new window

​Background to the Pathway

The Edinburgh direct access CT head route for headache has been available for >10 years. This is not based on any specific protocol. Cases with headache were audited. (Zienius, K., Chak-Lam, I., Park, J. et al. Direct access CT for suspicion of brain tumour: an analysis of referral pathways in a population-based patient group. BMC Fam Pract 20, 118 (2019). https://doi.org/10.1186/s12875-019-1003-y) https://bmcfampract.biomedcentral.com/track/pdf/10.1186/s12875-019-1003-y.pdf opens a new window

• Over a 5-year period, 3257 head scans were performed (318 scans excluded).
• 53 patients (1.8%) were reported to have intracranial tumours
  • 42 were significant (diagnostic yield of 1.43%)
  • 11 were incidental.
• There were no false negative CT scans for tumour.
• 559 patients had non-tumour findings,
  • 31% of which were deemed clinically significant.
  • In 34% of the 559 patients (190) – referral for further imaging and/or specialist assessment from primary care was still thought warranted.

Waiting times for scans was about 3-4 weeks

Headache alone as a presenting symptom of cancer is rare. New headache alone PPV for brain tumour = 0.1.

Headache “plus” the addition of cognitive symptom increases the PPV to 5.9

https://bmjopen.bmj.com/content/bmjopen/9/8/e029686.full.pdf opens a new window

In 2017, a Edinburgh Protocol Based Referral Protocol was produced for cases with Headache (Suspicious of Cancer) “Plus” to stream more urgent cases for faster CT scans. Selection was based on GP opinion, but certain red flags were required to assess the value of these and a semantic verbal fluency score (how many words can you think of in a minute). These scans were done within a week. The SVFT did not affect the speed of the scan but all cases where a SVFT was performed had faster scans – as they were protocol complete – versus those referred through this pathway where SVFT was not complete.

The SVFT is a fast screen for cognition and has been used as a screen in dementia (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5619418/ opens a new window). A score <17 indicates concern for cognition (https://www.verywellhealth.com/verbal-fluency-test-and-how-it-screens-for-dementia-98629 opens a new window). The Sematic verbal fluency score has also been used as a screen for incipient cognitive problems in Parkinson’s Disease (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3655231/ opens a new window ), Multiple Sclerosis for attention, executive dysfunction and memory (https://www.frontiersin.org/articles/10.3389/fneur.2020.629183/full opens a new window) and also for brain tumour (https://www.researchsquare.com/article/rs-217313/v1 opens a new window ), where it showed a moderate to strong correlation between MMSE and ACE-R, but was more sensitive and scores <10 words in a minute correlate with lack of capacity to consent (https://pubmed.ncbi.nlm.nih.gov/24991034/ opens a new window and (https://link.springer.com/article/10.1007/s11912-019-0793-3 opens a new window).

The Edinburgh protocol was audited after year 1 (https://abstracts.ncri.org.uk/abstract/headache-suspicious-of-cancer-the-edinburgh-early-diagnosis-pathway/ opens a new window ) and after 3 years (https://www.bnosconference.co.uk/wp-content/themes/wordpress-template/pdf/bnos_2021_abstract_book.pdf p34-35 opens a new window). The findings have been presented at the British Neuro-Oncology Society Meeting (July 2021) and the National Cancer Research Institute Annual Meeting (Nov 2021). To summarise:

• From March 2017-Nov 2019 there were 669 PBR for Headache SoC (62% female; Mean age 53 years).
• Ave time to scan <7 days
• SVFT were completed in 381 (57%). Median SVFT was 17 animals/minute.
• 11 cases (2.9%) had cancer. 10 of the 11 (91%) SVFT <17 (median 10 animals)
• SVFT was <17 in 188 HA SoC cases overall – 5.32% had a tumour
• Other causes for a low SVFT <17 were: 12 not native English speaker; 6 PMH brain injury; 19 co-existing dementia/MCI; 5 small vessel disease; 4 cysts (pineal / arachnoid); 4 Giant Cell Arteritis; 2 Chiari 1 malformations.
• 53 with SVFT <17 had current or PMH Psychiatric or Pain Clinic conditions and were taking (opioid analgesics/ antidepressant /antipsychotics). None had a brain tumour.