Most studies originated in North America (n = 12) or in Europe (n = 3). Studies predominantly used quantitative (n = 9) or mixed (n = 4) designs. We found only surveys, mostly prospective (n = 11).

Study populations included patients, carers and health care professionals. The male gender predominated, and the median age of patients ranged from 51 to 62 years. Ten main themes were addressed: symptom burden; perspectives of life-sustaining treatment and comfort for patients, families and health care professionals; goals of care discussions; patient and family needs; quality of life; PC and survival; referral to PC, barriers and opportunities; integration of PC; outpatient care and cost-effectiveness analysis.

Symptom Burden

Few studies addressed directly symptoms in ESLD patients awaiting LT. An early PC intervention study has found that the most commonly reported symptoms were fatigue, sleep disorders and pruritus [9]. In another study patients reported lack of energy, pain, sleep disorders and drowsiness as the most frequent, severe and distressing symptoms [10]. Other frequent and severe symptoms were dry mouth, lack of concentration and itching. Lack of concentration was considered more distressing than dry mouth.

A qualitative interview study of patients and bereaved carers found deteriorating physical health was compounded by ongoing psychological issues, commonly relating to alcohol dependence, including depression and guilt related to the nature of the disease [11].

Attitudes toward symptom management, such as pain in ESLD patients, were evaluated and showed that 30–40% of participants, despite having moderate pain, reported using no pain medication. The patient who used it reported having less than 50% relief and were only moderately satisfied with their overall pain treatment. Moreover, although 8 out of the 11 patients who were not LT candidates took a strong opioid, only 2 out of the 9 patients awaiting LT were given the same pain relief treatment [12]. In a survey of LT service providers [13], nurses, postgraduate year 1 medical and surgical trainees were less likely to agree that LT providers were proficient in managing pain and depression than attending physicians. Also, postgraduate year 1 medical and surgical trainees (86%) and attending physicians (100%) were reported more likely to avoid using opioid medication to treat pain in these patients than nurses (62%) (p = 0.0001).

Perspectives of Life-Sustaining Treatment and Comfort for Patients, Families and Health Care Professionals

An intensive care unit prospective study that included 6 patients (all considered for LT), 19 family members and 122 health care professionals revealed that the decision trajectory and the reasoning behind the life-sustaining treatment and comfort care decisions are resumed by the metaphor “on the train” [14]. The use of this expression enabled patients and family members to conceptualize their experience of the decision-making process – patients and families are in the train, where each life-sustaining treatment decision is a “train station,” a place where they can theoretically choose to disembark but they feel that they have little choice to do it. In the same study, Hansen et al. [14] stated that 4 subthemes became evident during the patients’ stay in the intensive care unit: (i) for families, life was the obvious answer, the purpose of lifesustaining treatment decisions was to keep the patient alive and headed toward the goal of LT; (ii) excluding families from “minor” life-sustaining treatment decisions added to their limited preparedness for major ones that followed; (iii) multiple professionals, each with a narrow focus, infrequently explained to family members how the function of an organ interrelated with the function of other organs or what it meant in the context of the overall picture of the patient’s illness; (iv) different perceptions of the patients’ illness course among specialities, between professions, and between professions and patients/families.

A structured PC intervention study in the surgical intensive care unit showed that, during the intervention period, the number of patients without resuscitation status significantly increase over time amongst those who died (from 52 to 81%, p = 0.03) [15]. The mean length of stay decreased, both in those who died and those who survived in the intensive care unit. Withdrawal of ventilator, pressors and nutritional support increased significantly in the intervention period.

Goals of Care Discussions

Goals of care discussions seem to be less common in patients considered for LT [15], and we found different perspectives from different professionals in the survey of Beck et al. [13]: a vast majority of nurses (96%) and postgraduate year 1 medical and surgical trainees (91%) responded that it was appropriate to hold goals of care discussions during clinical visits with the patient’s primary care provider, while only about a third of attending physicians (31%) agreed (p = 0.0001). In contrast, more attending physicians (69%) thought it was appropriate for a patient’s hepatologist to have goals of care discussions during clinical visits, with 100% of nurses and 91% of postgraduate year 1 medical and surgical trainees agreeing.

PC intervention studies revealed different results: in the intensive care unit [15] setting goals of care discussion increased from 2 to 39%, while in an outpatient early PC intervention study [9] it did not significantly increase rates of advance directive documentation.

Patient and Family Needs

Patients with ESLD who die while awaiting transplantation seems to have greater unmet needs for PC than other decedents, even if all received relatively aggressive care during their terminal hospitalization [15].

Two studies addressed indirectly some patient and family needs [11, 21]. Physical health problems and ongoing psychological issues (related or not with alcohol and other dependences) were identified as associated to physical disability, financial insecurity and the risk of increasing isolation [11]. Some patients feared advance care planning would result in a loss of hope and referred uncertainties about disease trajectory. Regarding LT, being assessed unsuitable was considered as an uniquely difficult period [11]. Families were perceived to be in need of support themselves, requiring frequent reassurance as the patient’s condition deteriorated [21]. Caregiving was associated with psychological distress (particularly relating to ongoing alcohol use and hepatic encephalopathy) and deprioritization of individual needs [11].

Quality of Life

None of the selected articles assessed quality of life directly. Beck et al. [13] reported that patients and their families value goals of care discussions and were grateful for PC. Responses from the Family Quality of Dying and Death questionnaire, after a structured PC intervention in the surgical intensive care unit [15], revealed a trend toward improvements in “time with family and friends,” breathing comfort, spiritual services, presence at time of death and overall physician care.

PC, LT and Survival

The study of Medici et al. [20] reported that 4 of the 6 transplanted patients experienced an improvement of their MELD score during hospice stay, suggesting that this service can potentially provide effective care to terminally ill patients. None of the other studies revealed data regarding the impact of PC on prognosis.

Referral to PC, Barriers and Opportunities

The included studies revealed different rates of referral of ESLD patients to PC, ranging from 4.5 to 29.1% [13, 17, 18], with an increasing tendency over the past 10–15 years [13, 18, 19].

Specifically from patients awaiting LT who died, only 2% received PC consultation [13] and in a nation-wide US weighted sample of 59,687 patients dying in the hospital with decompensated liver disease from 2009 to 2013, 18,027 (29.1%) received a PC consultation during the hospitalization and 331 patients were awaiting or received LT (1.9%) [19]. A study about the utility of the MELD (Model for End-Stage Liver Disease) score in transplant candidates and simultaneous hospice referral described a 5% rate (8/157) of referral to hospice care [20].

Factors associated with low PC referral included African-American, Hispanic and Asian ethnicity [18, 19] and absence of insurance coverage [18]. Factors associated with increased referral to PC were older age [17, 18], Caucasian ethnicity [17], “do not resuscitate” status [18], treatment in a teaching hospital [18, 19], in medium [18] and large-sized hospitals [18, 19], presence of hepatocellular carcinoma [18], presence of metastatic cancer [18], higher MELD score at listing [17, 20] and at delisting [17] and higher Charlson comorbidity index [19]. Decedents who were listed or received LT during admission were also less likely to receive PC than those who were not listed for LT [19].

Late referrals to PC among patients who died awaiting transplantation were identified by Kathpalia et al. [17], who found that the median number of days between PC consultation and death was 4. Different perspectives were identified by Beck et al. [13] among different professional groups, with nurses (88%) and postgraduate year 1 medical and surgical trainees (63%) being more likely than attending physicians (44%) to report that there were patients for whom they wished they had consulted PC earlier (p = 0.001). Attending physicians were less likely than nurses or postgraduate year 1 medical and surgical trainees to agree that it is appropriate to consult PC when the patient is diagnosed with ESLD, when the patient is listed for transplantation or when a patient with ESLD is admitted to the hospital for any reason. In contrast, the majority of attending physicians agreed that it is appropriate to consult PC when the patient is imminently dying [13]. In another study, the majority of LT service surgeons felt that PC should have been introduced earlier in the patient course, but only a few considered that there was a delay in the institution of PC under their care [15].

Attending physicians, more than any other group, were identified by all respondents (including attending physicians themselves), as creating a barrier to involving PC in the care of their patients [13]. In addition, they were much more likely than nurses and postgraduate year 1 medical and surgical trainees to identify the patient and family members as creating a barrier to PC consultation. Many respondents cited lack of clear criteria for involving PC and difficulty prognosticating end of life in ESLD patients as significant barriers. In a qualitative study involving general practitioners from the UK [21], the main concerns identified were those relating directly to the condition (symptom management and the need to combine a PC approach with ongoing medical interventions), issues arising from patients’ social circumstances (stigma, social isolation and the social consequences of liver disease) and deficiencies in the organization and delivery of services.

Benefits of PC were verified on life-sustaining treatments [19], goals of care discussions [15] and symptom management reported in the early PC intervention study [9]. As a matter of fact, these studies found a 50% improvement of the initial moderate-to-severe pruritus, and of the general well-being, appetite, anxiety and fatigue. Pain, myalgia, sexual dysfunction, sleep disturbance and dyspnea also showed improvement, but this improvement did not reach statistical significance, and the early intervention also decreased depression (27.8%). An American study identified a significantly lower mean of total costs and length of stay for patients who received PC consultation in its terminal hospitalization, with an association of PC consultation with reduced hospital length of stay and reduction in nearly all procedures [19].

Integration of PC

A qualitative study on the incompatibility of health care services and end-of-life needs in advanced liver disease reported that patients relied on hospital services for most aspects of care, with general practitioners often being bypassed in decision-making and perceived as being unaware of ongoing developments [11]. On the other hand, general practitioners consider themselves as likely to have an established relationship with the patient and a greater understanding of their social situation and needs, whereas specialists offer expert knowledge on liver disease and treatment options [21]. They highlighted the importance that primary care physicians place on being able to provide a coordinating role but only when supported by members of the specialist teams. Managing complex and unusual symptoms, or judging when to introduce a PC approach, for instance, would benefit from this collaboration. Collaborative working with the support from specialist hospital clinicians was regarded as essential, with general practitioners acknowledging their own lack of experience and expertise in this area. Further training of general practitioners and PC professionals on issues specific to liver disease were referred in another study [22]. Improved awareness of supportive care on symptom control, advanced communication and prognosis discussion for all professionals, and better knowledge of how PC professionals can support general practitioners and liver specialists were also referred as important measures [22]. A study of proactive case finding to improve concurrently curative and PC in patients with ESLD showed that patients in the quality improvement project were more likely to be considered for LT (77.6%, p < 0,001), to have their transplant evaluation completed (22.4%, p = 0,01) and to receive PC (62.5%, p = 0.38) [23].

Outpatient Care

Of the 15 studies selected, 1 took place in the hospice [20], 8 studies exclusively included inpatients [13–19, 23] and 4 studies were developed in the outpatient setting [9– 12]. Of the 4 studies that included only health care professionals, 2 also considered the outpatient context [21, 22].

From the published results it is not possible to assess that there are differences between the needs of inpatients and those of outpatients. Baumann et al. [9] reported that after the early palliative care intervention in the outpatient setting, 50% of moderate to severe symptoms improved and 43% of patients showed improvement in clinically significant depressive symptoms, greater in those with more symptoms.

In another 2 studies [21, 22], many general practitioners felt unable to manage advanced cirrhosis in the community. All health care professional groups seem to wish to increase community provision of PC support and recognized that they required further training to improve their skills in caring for people with ESLD [22]. Service configuration was one of the key areas identified for future research [22].

Cost-Effectiveness Analysis

Based on the National Inpatient Sample (national data set across the USA) having been listed or receiving LT during the hospitalization was associated with higher overall costs compared to the unlisted status [19]. In this study, Patel et al. [19] reported that PC consultation was associated with lower costs and procedure burden at the end of life.

Discussion

We have identified 15 studies which met the inclusion criteria. Studies were quite diverse on themes, including symptom burden in ESLD patients to life-sustaining treatment decisions or integration of PC. There were also differences on the populations studied, all studies included patients with ESLD awaiting LT, but some also included noneligible LT patients. The authors found no randomized controlled trials. In accordance with previous literature male patients under 65 years old predominate [24].

In a global perspective, the studies considered in this review reported significant PC needs of ESLD patients awaiting LT and some benefits of PC but still an inadequate response to these needs, with challenging barriers to better integration of PC for patients awaiting LT.

Symptoms such as fatigue and sleep disturbance were the most frequent and severe symptoms identified [9, 10] similar to what has been described in a recent meta-analysis [25]. Pain was also reported in one of the included studies [10]. In contrast, breathlessness and sexual dysfunction were not mentioned as they had been in the same meta-analysis [25], with sexual dysfunction not even being evaluated, which suggests that it should also be included in future symptom assessments. Psychological needs were identified but apparently not in the same manner by all professionals, as recommended [11, 13]. A significant burden from psychological symptoms, such as depression, was also identified in more than 50% of ESLD patients included in an Italian survey [26], and anxiety was reported in pretransplantation patients ranging from 27 to 44% [27, 28]. The experience of waiting for a LT has considerable impact on patients [29]. Qualitative studies have described uncertainty as the cornerstone of this psychological process, with very diverse perspectives being reported by patients: some expressed joy and relief and regarded transplantation as an opportunity; others manifested a sense of danger at facing the likelihood of death and of developing cancer during the waiting period [30, 31]. Difficulty coping, loss of trust in physicians and medical, personal and social uncertainties have all been reported [30–32]. LT patients with a coexisting history of substance abuse, in particular, would probably benefit from special attention [33], since this population tends to be threatened by negative attitudes of health care professionals, has a significant risk of noncompliance to treatment and might have special concerns about both symptom management and caregivers’ support.

Despite the significant symptom burden, symptom management is still probably suboptimal as suggested by low rates of pain relief [12] and inappropriate professional skills [13]. Adequate medication in ESLD is challenging, since most drugs are metabolized in the liver, thus putting patients at a higher risk of adverse effects. As a result, physicians often resort to less aggressive symptom management. Moreover, physicians may avoid the prescription of opioids to those with a history of substance abuse and often fear the side effects of opioids (e.g., constipation and worsened encephalopathy) [34].

Although there were few studies that directly evaluate the impact of PC in LT patients, most of them identified potential benefits, not only in symptom management [9], but also in goals of care discussions [15], in life-sustaining treatment decisions [19] and even in direct and indirect health costs [19]. Similar results have been reported in patients awaiting other organ transplants [35–37]. There is an inherent difficulty in discussing goals of care and advanced directives with patients pursuing curative therapies, but as suggested by Potosek et al. [34] it is imperative to discuss those and to identify health care proxies, as encephalopathy can impair decision-making. Regarding cost-effectiveness analysis, more studies are required.

Only 2 studies indirectly addressed the quality of life of patients with ESLD [13, 15]. Both favored the PC approach which has a beneficial impact on it. However, robust studies that specifically assess quality of life are essential as one of the cornerstones of the overall PC approach.

The study of Medici et al. [20] reported an improvement of their MELD score during hospice stay, suggesting that this service can potentially provide effective care to terminally ill patients. Although no other study has directly assessed the impact of PC on the prognosis and survival of ESLD patients, it is expected that this might be positive, similarly to studies regarding other medical conditions [38].

Referrals of LT patients to PC seem to be increasing but are still quite low and probably occur only in very late stages of the disease, putting patients at risk for decreased quality of care at the end of life [13, 17–19]. Many barriers have been identified, considering patients, carers, health care professionals and services. The “on the train” perspective hinders the ability of all the interveners to consider a simultaneous approach of PC while awaiting transplantation [14]. Kathpalia et al. [17] considered that these patients represent a unique subgroup of patients with a terminal condition – by virtue of having ESLD – but await the promise of a cure through LT. In this setting, PC, which traditionally has been considered only for those “at the end of life,” may be perceived – by both the patient and providers alike – as unnecessary and unwelcome. More and appropriate education of health care professionals and training is probably necessary, as suggested in surveys analyzed here [13, 18, 19, 21, 22].

Integration of PC in LT patients seems to be considered essential by all parts [13, 15, 21, 22]. Transplant candidates not only benefit from being provided PC, since they suffer from a terminal condition with significant needs, but also require a complete and complex medical care even when their condition is very advanced [39]. Also, LT patients who are removed from the waiting list often experience withdrawal of specialty care, feelings of abandonment and likely imminent death [34], all of which must be addressed. Development of clinical triggers for PC consultation and effective collaborative approach may clarify better integration [21].

This review was limited by the restricted number of articles included, written only in English, which resulted in a data set that is limited to North American and European ESLD patients that may not take into consideration other relevant realities. Note also that important differences regarding design of the studies may have resulted in significant data heterogeneity, which is beyond the scope of this review since quality assessment was not evaluated.

Conclusions

This review found notable discrepancy between the need for PC and its use in ESLD patients awaiting LT. The belief that transplantation and PC are mutually exclusive still prevails. However, there is increasing evidence that the strategy of providing PC alongside disease-directed therapy while awaiting LT might benefit patients, carers and even the health system. This review indicates some potential areas for developing shared care models. Further research is required, with more robust studies, to better define the optimal strategy of the best way of PC integration for ESLD patients awaiting LT.

References