Discussion

At its peak, the Covid-19 pandemic led to an unprecedented disruption of healthcare services and a necessary shift in priority towards identifying and treating Covid-19 infected patients.^{Reference Vose9} As such, routine cancer surveillance, two-week wait referrals and cancer surgery were suspended for variable periods of time. An audit of cancer diagnoses in Italy found a reduction of 39 per cent during the initial pandemic period in contrast to the previous 2 years.^{Reference De Vincentiis, Carr, Mariani and Ferrara10} The true impact of cancer diagnosis delay is yet to be fully determined in the UK as normal oncological services resume. However, UK-based modelling studies, unfortunately and somewhat unsurprisingly, predict a substantial rise in avoidable cancer deaths in the foreseeable future because of delayed diagnoses,^{Reference Maringe, Spicer, Morris, Purushotham, Nolte and Sullivan8} with two thirds of cases initially presenting with advanced disease prior to Covid-19.^{Reference Schenker, Arnold, Baumen, Heron and Johnson3}

Specialist input from palliative care teams is often sought by clinicians at an early stage to ensure optimal management of these patients. Management of this subset of patients is uniquely complex and challenging. By nature of anatomical location, head and neck cancer and its management may have significant adverse effects on the most basic human functions, such as breathing, eating and speaking.^{Reference Begbie, Douglas, Finlay and Montgomery11} Furthermore, cosmetic disruption of the head and neck, inability to verbally communicate and alternative routes of nutrition may amount to significant adverse psychological impact on both patients and their family members.^{Reference Simcock and Simo12} With oncological burden among all specialties expected to rise profoundly as a result of the Covid-19 pandemic, specialist services are likely to become more stretched and less readily available for routine palliative care management, resulting in an expectation that head and neck disease and incidents of major haemorrhage, including carotid blowout, may increase. Therefore, sound knowledge of anticipatory medications and catastrophic haemorrhage protocol is also fundamental.

An understanding of medical and surgical management of palliative head and neck cancer patients, as well as knowledge of complications associated with surgical and non-surgical management, are core competencies outlined in the intercollegiate otolaryngology curriculum.^{Reference Davis, Spraggs, Murray and Bussey6} However, our experience suggests that current coverage of palliative head and neck cancer in the specialty curriculum is relatively sparse and inconsistent, with many trainees relying on prerequisite knowledge and principles acquired from rotations during junior medical training. Our survey has shown a general lack of palliative care experience, with only 2.1 per cent having undertaken a postgraduate palliative care rotation. A lack of perceived confidence among the surveyed cohort was also demonstrated, with 55 (37.9 per cent) stating a lack of confidence, 58 (40 per cent) stating neutrality and only 32 (22.1 per cent) respondents indicating that they are confident in palliative head and neck cancer management.

Our novel survey, designed and ratified by five palliative care consultants utilising the Delphi Method, included eight multiple-choice questions based on real-life patient scenarios. They were specifically designed to test knowledge and decision-making surrounding key elements of palliative care management, including prescribing, do not attempt cardiopulmonary resuscitation considerations and end-of-life care decisions relating to invasive and non-invasive interventions.

Each of the three questions assessing prescribing knowledge around anticipatory medications were answered correctly by less than half of the surveyed cohort. Regarding prescribing of anticipatory medications on an ‘as required’ basis, 37.9 per cent (n = 55) correctly opted for subcutaneous bolus injections for symptom-control. National Institute of Health and Care Excellence (NG31)¹³ recommend either subcutaneous or intravenous administration; however, in the palliative head and neck cancer patient with poor nutritional status, intravenous access is likely to cause further distress, making subcutaneous injection the more appropriate option. Furthermore, in the event of a palliative head and neck cancer patient experiencing catastrophic bleeds, such as in cases of carotid blowout, the aim of therapy is to alleviate anxiety and distress.^{Reference Charlton, Hooke, Armstrong, Vandenberg and Hardy14} Because of its rapid onset of action and administration, midazolam is the most frequently recommended drug in cases of terminal haemorrhage.^{Reference Harris and Noble15} Considering preparation and administration time, intramuscular administration of 10 milligrams of midazolam was the most suitable option in our survey relating to catastrophic haemorrhage, selected by only 28.3 per cent (n = 41) of respondents.

With a 5-year survival between 19 and 59 per cent, prognosis for head and neck cancer patients is relatively poor, making do not attempt cardiopulmonary resuscitation a vital consideration. Do not attempt cardiopulmonary resuscitation decisions are made where cardiopulmonary resuscitation is clinically judged very unlikely to be effective and should always be discussed with the patient and family where possible.^{Reference Fritz, Slowther and Perkins16} Clinicians should be aware of the principles surrounding do not attempt cardiopulmonary resuscitation decision-making. Our survey question corresponding to do not attempt cardiopulmonary resuscitation decision-making was answered correctly by the majority, with 78.6 per cent (n = 114) acknowledging the need for a second professional opinion in cases where patients or relatives disagree with a do not attempt cardiopulmonary resuscitation decision taken and 71.1 per cent (n = 104) correctly determining that do not attempt cardiopulmonary resuscitation decisions should be discussed with patients unless there is a clear risk of causing psychological harm.^{Reference Fritz, Slowther and Perkins16}

Questions assessing end-of-life care decision-making, clinical application and invasive versus non-invasive intervention were answered with variable accuracy. In the scenario involving a relatively young patient with anaplastic thyroid cancer, expressing the desire for end-of-life care at home, most respondents (n = 115, 79.3 per cent) acknowledged the need to provide palliative care support at home in line with the patient's wishes. However, given the short history of dyspnoea and stridor signalling impending airway compromise, far fewer respondents (n = 43, 29.7 per cent) also opted for tracheal stenting as a necessary and life-preserving intervention. Similarly, in the scenario involving an actively bleeding fungating neck wound in a patient desiring comfort-focussed care, the majority (n = 74, 51 per cent) opted for do not attempt cardiopulmonary resuscitation, ward-based care with anticipatory medications and no further blood transfusions. However, 24.1 per cent (n = 31) correctly opted to follow this same management strategy with the exception of giving blood transfusions to provide symptomatic relief.

With regards to lasting power of attorney lasting power of attorney (LPA) the attorney is designated to make health and welfare decisions in the best interests of the person they represent, when a patient has lost capacity to make decisions for themselves.¹⁷ LPA decisions may only be overridden by a medical professional where the lasting power of attorney does not act in a patient's best interests or contradicts a previous advanced directive made by the patient. In the scenario we described, the patient's daughter, who has lasting power of attorney, wishes to avoid invasive intervention. Thus, in this case of new-onset stridor in a patient with significant co-morbidities, admission to the ward with best supportive care would be the most appropriate action, selected by 41.1 per cent (n = 60) of respondents. A similar number (n = 62, 42.8 per cent) also opted for CT of the base of skull to diaphragm. However, since invasive treatment is to be avoided based on the lasting power of attorney decision, any findings on CT scan would be unlikely to alter management.

The final question is based on a patient with advanced base of tongue squamous cell carcinoma (SCC), stepped down from intensive care following chest sepsis and a recent cerebral infarct, and now intermittently bleeding from his tumour. This question provoked the least discrepancy in responses, with the majority (n = 117, 80.7 per cent) correctly opting to determine a ceiling of care based on shared decision-making with the patient.

Although palliative care decisions are by no means binary and rely on sound clinical judgement applied on a case-by-case basis, our survey has evidenced global knowledge deficits, with an overall mean score of 5.0 out of 10 amongst the 145 participating clinicians of both middle grade and consultant level. Educational interventions will certainly help otolaryngologists manage increasing austerity and pressure as the National Health Service, as a whole, attempts to tackle enormous backlogs of cancer patients in the wake of the Covid-19 pandemic.

Ballou and Brasel (2019) noted that although palliative care is a central component of surgical practice, postgraduate education and training opportunities are limited.^{Reference Ballou and Brasel18} At present, there is a lack of consensus in the literature regarding the optimal modality through which palliative care education can be delivered. However, simple interventions, such as palliative care based workshops, have been shown to significantly improve palliative care knowledge and attitudes in the short term.^{Reference Pernar, Peyre, Smink, Block and Cooper19,Reference Harden, Price, Duffy, Galunas and Rodgers20} As well as group sessions, online modules were also favoured as a potential educational modality among our surveyed cohort (n = 94, 64.8 per cent and n = 71, 49 per cent, respectively). Online palliative care training has been trialled amongst undergraduate nurses and subjectively gauged to be successful.^{Reference Thrane21} The authors were unable to find evidence in the literature of this approach amongst surgical faculty but acknowledge that online training provides a convenient and accessible means by which education can be widely disseminated. Furthermore, simulation-based education is becoming increasingly commonplace, with the clear advantage of being able to learn and acquire valuable skills through practice without the risk of causing patient harm.^{Reference Bearman, Nestel and Andreatta22} Simulation-based education has been shown effectively to improve confidence and competencies in palliative care among other healthcare specialties and disciplines.^{Reference Goldonowicz, Runyon and Bullard23–Reference Renton, Quinton and Mayer25}