Guidelines should be based on all of the evidence

Kennedy et al present results from a survey followed by a guideline on ‘crisis pack’ prescribing.1 As they note, the systematic literature review of terminal haemorrhage in patients with advanced cancer reported that almost all sources recommend the solitary usage of midazolam,2 and this strategy is supported by (among others) the latest versions of the Oxford Textbook of Palliative Medicine 3 and the Palliative Care Formulary.4 Indeed, apart from the survey results, there is nothing to justify the inclusion of morphine in their guideline. Terminal haemorrhage ‘is not typically a painful event,”1 is not usually associated with breathlessness,2 and the doses suggested (≤1/6th daily dose) are unlikely to produce sedation (as this is the typical dosage for ‘breakthrough pain’). Moreover, the guideline does not recommend an opioid for the community setting despite the availability of non-injectable ‘rapid onset opioids’ (eg, intranasal fentanyl). Finally, parenteral morphine is a controlled drug, and the time needed for the drug to be procured/administered is likely to be greater than the time taken for the patient to exsanguinate.2