Palliative emergencies in the pre-hospital setting

Pain severity

In palliative care the presence of pain is highly prevalent, with up to 70% of patients living in a painful state during their end-of-life experience (Chang et al, 2007). The existence and severity of pain varies from patient to patient and from condition to condition but all palliative illnesses can cause pain for its sufferer, and therefore treatment in palliative patients should always bear in mind the existence of pain.

Pain management is a central component of palliative treatment and clinicians treating these patients need to be aware of many factors. For example:

All too often the task of acute pain management falls to paramedics working in the acute medical setting. When that happens, the assessment of the patient can only be acute in nature, and by clinicians with little training in pain assessment. Reviews looking at the accuracy of physician-led pain intensity have shown a tendency to underestimate pain by as much as 60–68% (Helfand and Freeman, 2009). And worldwide, few physicians are aware or have knowledge of pain management guidelines (Macpherson, 2006). This problem is further compounded by poorly funded medical centres that can often lack the required access to pain specialists (Macpherson, 2009), as well as poor pre-hospital pain management by paramedics.

A systematic review by Sampson and Goodacre (2014) looked at 83 independent studies with the aim of improving assessment, awareness, management, and documentation of pain, found that the evidence supporting changes in education and overall management displayed a high probability of bias in the study design, and more understanding is needed about the theories underlying the interventions. This review agrees with the premise that the knowledge underpinning the assessment in pain management is based on weak evidence and that more research needs to be done.

The Chapman review looked at the assessment and treatment of cancer patients and concluded that it is frequently undertreated due to the barriers of poor assessment, insufficient knowledge of pain mechanisms and treatment options (Chapman, 2012). Chapman also emphasised the dynamic nature of both the pain and disease process, and the individual patient's needs, and gave findings that patients will often be reluctant to report pain for varying reasons (Carr, 2007; Chapman, 2012).

Pain specialists and palliative care experts are able to relieve the majority of discomfort or reduce the severity of pain for these patients by adopting a ‘total pain’ approach. This was first introduced by Cecily Saunders, and looks at the physical, emotional, social, and spiritual distress caused by palliative conditions (Saunders, 2006). However, because only a small percentage of doctors or physicians have a familiarity with the problems endured by palliative patients, many will either go untreated or under-treated (Macpherson, 2009). The evidence base for clinical trials looking to improve the level of treatment is weak due mainly to the clinically unstable nature of these patients and the difficulties in recruitment (MacMillan and Weitzner, 2003). This, alongside the ethical dilemmas commonly associated with pain management trials, has meant that many guidelines come from the World Health Organization's (WHO) analgaesic ladder for treating mild to severe pain (WHO, 2004).

Pain assessment tools

A literature review in 2008 indicated the urgent need for a consensus regarding pain assessment in palliative care patients (Hjermstad et al, 2008). The literature review identified 11 different assessment tools and found that none were ‘fully inclusive’, and that no studies were conducted in the last few days of life, thus giving a selection bias to healthy patients. The main argument for this fragmentation of assessment tools is the constant need for new instruments appropriate to the findings of the clinical needs or a particular area of research pain, especially in regards to advanced cancer states (Carr et al, 2004; Caraceni et al, 2005; Hølen et al, 2006). This review highlighted the need for a general consensus, although the dynamic process of pain and disease progression should also be taken into account and that a pain assessment tool should be adaptable for all patients (Chapman, 2012).

All assessment tools have been shown to have strengths and weaknesses. The Multidimensional Continuous Pain Assessment Chart (MCPAC) looks mainly at pain intensity (with medication, quality of life, sleep, and mobility taken into consideration) and uses a chart to clinically observe patients over time, an option paramedics do not have. This tool was developed because existing tools were considered too cumbersome and not suited for global pain assessment (Bercovitch et al, 2002). The tool was, however, developed within a clinic and with other clinicians with little to no involvement from patients or other professionals. A weakness to this tool is the lack of a validation process, either internal or external, or cross-cultural verification to see if it meets the need of all patients in all cultures. This weakness was partially recognised with the Pain Control in Palliative Care Questionnaire (PC PCQ) tool, as this tool takes into account the patients beliefs in terms of cause and consequence of the illness. But as with the MCPAC tool, the PC PCQ tool does not look at any cross-cultural differences or problems that arise thereof (Böstrom et al, 2004). The PC PCQ tool notes the treatment effect of medications and the history of the actual pain itself. It asks the question: who/what do you do/call when pain is not under control? This places an emphasis on the support offered by relatives and friends during the palliative stage of illness instead of trained experts. It also gives it a broader scope of potential treatment effects and sense of well-being—noting the work of Cecily Saunders—than the MCPAC tool. But as with the MCPAC tool, there seemed to be no involvement from patients or other professionals during its development, and therefore being totally clinician led. The issue of cross-cultural differences was addressed by Steel et al after a literature review highlighted the need for a more inclusive assessment tool (Steel et al, 2003). Steel et al developed the Resident Assessment Instrument for Palliative Care (RAI-PC) tool which places an emphasis on pain intensity, temporal patterns, location of pain, the quality of pain, and its effect. It was developed as part of a field test which included a pain assessment and tool assessment study. It also performed a cross-cultural test for validity within various cultures.

As there are many different assessment tools, none of which are fully inclusive, the need for a consensus has been realised. Further research should be conducted to formulate a fully inclusive pain assessment tool which addresses the needs of all patients and all needs, and recognises palliative pain as a progressive disorder which requires constant assessment.

Seizure frequency

As either the result of medications or disease progression, seizures can play an uncomfortable role in the care of palliative patients. The fear of further seizures is a common factor in palliative patients that have already experienced a seizure, and this places a burden on physicians to prescribe rescue medications or a prophylactic drug to combat the seizure. Studies have found that around 45% of patients with a primary brain tumour can suffer frequent seizures and a further 15–30% will develop seizures during the progression of their disease (Moots et al, 1995; Glantz and Recht, 1997).

Anti-convulsant drugs

Evidence suggests that medications given to palliative patients, regardless of their disease, to treat their various conditions, whether chemotherapy drugs or opioids, can potentially cause seizure activity, so further adding to the problems of the patient. Guidance for paramedics regarding the treatment of seizures are outlined in the UK Ambulance Services Clinical Practice Guidelines (it must be remembered that these are only guidelines and paramedics are not governed by them), with the end result being either: a) refer to the clinical pathway, or b) treat and take to hospital (Association of Ambulance Chief Executives, 2013). However, any guidance on end-of-life care in the guidelines is still very limited (Brady, 2014) and therefore, despite advances in university training, paramedics will rarely treat palliative patients in the same way they would other conditions.

The evidence base for the treatment of seizures in palliative patients is in itself weak. There is no Cochrane review concerned directly at seizure activity in palliative patients, and very few relevant studies looking directly at this issue. Therefore the guidelines for treatment have stemmed from studies of epileptic patients and their seizure activities and then transferred to the palliative setting. No study has looked at the validity of this procedure or its impact on palliative care.

A few studies have addressed the issue of seizure activity associated with a specific disease and found that medications themselves can cause the seizures. Krouwer et al studied seizure activity in end-of-life patients suffering from a brain tumour and listed medications known to have the side effect of seizure activity (Krouwer et al 2000). Termed ‘non-neoplastic causes of seizures’, Krouwer et al stated that the opioid ‘meperidine’ can have the side effect of seizure activity and should be stopped if this occurs (Krouwer et al, 2000). In addition to this opioid, and perhaps surprisingly, the benzodiazepines diazepam, clonazepam, lorazepam, and midazolam can all exacerbate seizures in certain palliative patients. When this occurs the physician will switch to an alternative anti-convulsant drug to relieve the seizure activity.

Paramedic drugs for seizure termination

The efficacy and safety of diazepam (benzodiazepines) in terminating pre-hospital seizures has been investigated in various studies. A randomised, double-blind, placebo-controlled trial, The Prehospital Treatment of Status Epilepticus (PHTSE) study, compared rates of seizure termination prior to hospital arrival through the administration of diazepam (5 mg) or lorazepam (2 mg) by paramedics Alldredge et al, 2001; Lowenstein et al, 2001). Of the 205 patients recruited, none of which were palliative, the benzodiazepine group showed a significantly higher level of seizure termination than the placebo group (Alldredge et al, 2001). This study, however, was set in the acute setting with intravenous access available and no patients having a terminal illness. The study did favour intravenous lorazepam 59.1% (Lowenstein et al, 2001) over intravenous diazepam 42.6% (Alldredge et al, 2001) for treatment in epileptic patients, and all were transported to the emergency department. The validity of this study for palliative care is to be questioned as transportation to the emergency department following every seizure is not generally accepted by palliative patients and the ‘rescue’ nature of the treatment does not address the underlying issue. Neither can these results be generalised to other routes of administration as it relies on pre-hospital personnel successfully obtaining IV access.

Without the ability of intravenous administration the benzodiazepines used in the pre-hospital setting are often rectal diazepam and the patient's own buccal midazolam. Rectal diazepam is, commonly, the preferred route as it offers the advantage of simple patient/family education and reduced costs when compared to other routes. It is rapidly absorbed and achieves a therapeutic serum concentration within 10 minutes (Fitzgerald et al, 2003; Dieckmann, 1994). However, buccal midazolam is gaining popularity in the residential setting and hospices, as studies have found it to be more effective than rectal diazepam in terminating seizure activity, especially those suffering with refractory seizures (Scott et al, 1999; Rainbow et al, 2002). The RAMPART (Rapid Anticonvulsant Medication Prior to Arrival Trial) study compared intramuscular midazolam with intravenous lorazopam in 893 patients (Silbergleit et al, 2012). The study demonstrated a significant benefit in favour of intramuscular midazolam with fewer hospital admissions (73.4% versus 63.4%, respectively) attributed mainly to the administration consistency achieved through the intramuscular route (Silbergleit et al, 2012). The route of administration in the RAMPART study makes the findings more applicable to palliative patients' needs than those in the PHTSE study. The ease of administration and the better outcome of intramuscular midazolam allow for family and carers of palliative patients to control the seizure event without the need for hospital admission.

The implication for future paramedic practice highlighted from these studies is for the greater use of midazolam as the pre-hospital anti-convulsant drug of choice. The ease of administration and its overall effectiveness should allow for paramedics to better care for palliative patients suffering a seizure.

The complexity of the issue

According to the American Thoracic Society, shortness of breath is defined as: ‘a subjective experience of breathing discomfort that consists of qualitatively distinct sensations that vary in intensity’ (American Thoracic Society, 1999). It is a common symptom in end-of-life care, still difficult to treat, and causes great discomfort for the patients (LeGrand, 2002; Westphal and Campbell, 2002; Bausewein et al, 2008). The treatment appropriate for this symptom consists of both non-pharmacological and pharmacological interventions. A 2008 Cochrane review, now withdrawn due to an updated version being written, looked at the possible non-pharmacological approaches being used and which of the approaches are the most effective (Bausewein et al, 2008). The review stated that breathlessness is a complex symptom caused by many factors, with the experience unique to the individual (Bausewein et al, 2006), hence all patients may display differently, and with differing aetiologies. The review explored the different mechanisms of non-pharmacological intervention and the evidence base for their use. The review demonstrated that pulmonary rehabilitation has been shown to relieve dyspnoea and fatigue in patients with chronic obstructive pulmonary disease (COPD) (Lacasse et al, 2006), which in turn improves the emotional and self control aspect of the patient's condition. Lacasse reviewed 31 randomized controlled trials and found statistically significant improvements in quality of life when focusing on dyspnoea, fatigue, emotional function, and mastery of the condition. This study displayed a few weaknesses in its study design. The lack of any kind of blinding meant that a high level of detection bias is possible by the patients knowing that they were receiving treatment. The lack of blinding may have had a significant effect on anxiety felt by the patients, which is a commonly known factor in breathlessness (O'Driscoll et al, 1999). There was also a lack of involvement of bed-bound patients that are in the last weeks of life. This lack of research for these patients has been well documented mainly due to ethical boundaries (LeGrand, 2002). It has been written that the strategies designed to help breathlessness may not improve the symptom intensity but rather help the patient's ‘air hunger’, and help them to breathe normally and reduce anxiety and distress (Sherwood et al, 2005, Wouter, 2005).

Non-pharmacological treatment

Cold facial stimulation has been shown to relieve the symptoms of breathlessness according to previous studies (Schwarzstein et al, 1990). This can be achieved by seating the patient near an open window or in front of a fan. The physiological mechanism hypothesised for this reaction is thought to be the application of air, or any cold solution, to the face, nasal mucosa, or pharynx, which can positively alter ventilation. Schwartzstein et al measured the sensation of breathlessness in 16 normal subjects to test the hypothesis of breathlessness associated with loaded breathing (Schwarzstein et al, 1990). They subjected the participants to restricted breathing through an inspiratory resistive device and measured the findings on a modified Borg scale. The conclusion from this test was that cold air on the face significantly reduces breathlessness (p < 0.002) without causing a significant reduction in ventilation. The placebo of initiating the diving reflex through cold air being blown on the legs was added to the study which showed no positive signs for the subjects. The low numbers involved in this study and the fact that all patients were ‘normal subjects’ mean that a leap of faith needs to be used in order to transfer the results to the palliative patient. Also, the subjects knew that the test can be terminated if they chose it to be so, an option that palliative patients do not have, thus eliminating the anxiety factor commonly associated with breathlessness.

Pharmacological treatment—opioids

A variety of pharmacological strategies have been used in the treatment of palliative breathlessness. There is good evidence, mainly gained from non-malignant patients, for the effective use of opioids to help ease the condition and relax the patients (Jennings et al, 2002; Abernethy et al, 2003). Although these papers are now becoming outdated, Shearer et al recently submitted a protocol to the Cochrane library looking at opioids for treating dyspnoea in patients with chronic heart failure (Shearer et al, 2014), Barnes et al submitted a Cochrane protocol looking at opioids for the palliation of refractory breathlessness in adults with advanced disease and terminal illness (Barnes et al, 2014), and an updated version of the 2012 Cochrane review by Jennings et al on opioids for the palliation of breathlessness in advanced disease and terminal illness is being written (Jennings et al, 2012).

The Jennings systematic review included doubleblinded trials, randomised trials, and placebocontrolled trials for any disease states that cause breathlessness (Jennings et al, 2002). The review meta-analysed 18 studies and showed a statistically significant improvement in patients' breathing when being given oral opioids. The review did show that nebulised opioids are statistically no better than placebo, a view disputed by Westphal and Campbell (2002). Westphal and Campbell referenced a retrospective chart review of 54 patients suffering from chronic breathlessness and reported a 63% improvement in patients given nebulised doses ranging from 5 mg to 30 mg (Lord, 1997). The Jennings review aligned itself to the Davis study (Davis et al, 1994), which stated that oral opioids are better than nebulised morphine due to its ability to bind to opioid receptors throughout the respiratory system, a view backed by various case reports (Farncombe, 1997; Zeppetella, 1997). Davis et al concluded that 5 mg of oral morphine is equivalent to 50 mg of nebulised morphine, and postulated that the effect of the opioid was from systemic absorption with only 10% reaching blood circulation. The discrepancy between these reviews highlights the difference between statistical and clinical significance and the implications that these can have on future practice.

The shortcoming in all studies looking at opioid use for breathlessness is the lack of terminally ill patients. Only 6 patients out the 38 that were assessed in the Abernethy study had cancer (Abernethy et al, 2003), and of those none were opioid-naive. The study showed a 5–10% improvement on the visual analog scale for those taking 20 mg of modified-release morphine over a four-day period compared to the placebo. Although this small increase will help the patient to breathe easier, the question remains on the potential adverse effects of introducing morphine into a medication regimen and whether the gains justifies the cons. The study from Abernethy et al, as with most studies, did not note the adverse effects, and although it is clear that opioids help dyspnoea in palliative care, there remains uncertainty about starting dose, regimen, choice of drug and release status, and indications for parenteral use (Booth et al, 2008).

Pharmacological treatment—anxiolytics

The treatment of breathlessness with anxiolytics is a long established practice comprising of the administration of benzodiazepines or phenothiazines. The evidence for their use in palliative care is weak, with no Cochrane review or randomised-controlled trials published at this present time. Thus the evidence for prescribing phenothiazines comes from studies involving non-palliative patients and COPD sufferers (O'Neil et al, 1985; Man et al, 1986; Rice et al, 1987). This transfer of data from another discipline does not allow for aetiologies that may be responsible for dyspnoea in palliative patients and may not be relevant. A dose of 6.25 mg oral levomepromazine has been shown to benefit patients suffering anxiety (Joint Formulary Committee, 2013), but no recording of adverse events has been noted, and there is no study looking at the benefits for cancer patients. In cases of severe breathlessness, treatment with phenothiazines are preferred to benzodiazepines (McIver et al, 1994), and this is especially the case in end-of-life care (Dudgeon, 2006).

As with phenothiazines, the evidence base for the use of benzodiazepines is weak with no Cochrane review or random-controlled trials for their use in cancer patients. They are empirically given for anxiety by clinicians, and in some cases are co-administered with morphine for the control of cancer-induced dyspnoea (Navigante et al, 2006). A cochrane review by Simon et al identified seven studies which looked at the use of benzodiazepines for the relief of breathlessness and did not show a beneficial effect in patients with advanced cancer or COPD (Simon et al, 2011). This review had a high level of rigour as it included 200 participants and performed a metaanalysis in six out-of-the seven studies. The review concluded that, in addition to it being non-beneficial in the treatment for breathlessness, the administration of benzodiazepines had no effect on the prevention of breakthrough breathlessness. Therefore, although benzodiazepines are a good treatment for anxiety and have proven to be of benefit in normal patients, the effect of reducing breathlessness in palliative care where anxiety may be the cause is questionable.

The use of home oxygen

A common treatment for end-of-life dyspnoea is for the prescription of home oxygen. Evidence from randomised controlled trials suggests that the benefits from oxygen come not from the correction of hypoxaemia but instead from the flow of gas to the face (Bruera et al, 1992; Booth et al, 2011). Therefore the physiological and psychological basis for the benefits of oxygen therapy appears to be the same as the non-pharmacological treatment of a fan for palliative dyspnoea. This justifies the protocol, allowing for the prescription of oxygen only after the trial of a fan, even for intractable hypoxic cancer patients, a policy challenged by Breaden et al (2013). Braeden et al questioned that the evidence base for the policy stems from two systematic reviews and a multinational random-controlled trial, none of which involved palliative patients (Uronis et al, 2008; Abernethy et al, 2010; Uronis et al, 2011). This questions the validity of the protocols when treating the needs of end-of-life patients with evidence and research gained from a different discipline. To support their argument, Breaden et al held seven focus groups in different areas of Australia and challenged 51 experienced palliative care nurses for their views on the prescription culture for oxygen (Breaden et al, 2013). The final analysis from the focus groups was that oxygen therapy works and should be adopted as a firs-line treatment in palliative patients instead of waiting for the results of other trials. This showed that the protocols established from other disciplines may not be compatible with palliative care, and that to base a protocol solely on controlled trials can have significant shortcomings. Thus, for a protocol to be established it may, in some instances, be prudent to conduct qualitative research from experts in their field. A negative to the Braeden et al study is that the data was totally qualitative and at a high risk of bias from eminence-based practice instead of evidence based practice. Looking at all the evidence it is clear that more research needs to be conducted involving end-of-life patients.

Palliative experts have stated that more research involving end-of-life patients needs to be carried out, and with a possible reassessment of the current protocols for all aspects of palliative intractable dyspnoea.