Palliative patients under anaesthesiological care: a single-centre retrospective study on incidence, demographics and outcome

Anaesthesiologists have been involved in the care of palliative patients since the beginning of modern palliative medicine. The field of anaesthesiology covers many parts of palliative medicine, e.g. pain management, use of sedative/anxiolytic medication, and treatment of critically ill patients. Also palliative medicine and anaesthesiology are similar to each other as both view integrating and coordinating therapies of different medical specialties as an important aspect of their specialty (e.g. physical and social therapists in pain management, spiritual care in the intensive care unit). Therefore some authors see anaesthesiologists as ideal physicians for the care of palliative patients [1, 2].

Besides the anaesthesiologist’s involvement in palliative medicine, palliative patients are also treated by anaesthesiologists in the intensive care unit [3], in pain management [4], and in pre-hospital emergency medicine [5]. Management of palliative patients in the anaesthesiologist’s primary workplace, the operating room, has not been in the immediate focus of the literature. Most articles on palliative patients in the operating room have been written from a surgical perspective with an emphasis on indication of surgical treatment, types of palliative surgical procedures, and surgical outcome [6, 7]. Anaesthesiological literature on the other hand has focused on describing the ethical problems in the care for palliative patients with do-not-resuscitate-orders in the perioperative period [8, 9]. Only few articles describe the perioperative medical management of palliative patients [10]. While incidences of palliative surgeries have been described in the literature [11, 12], no data to our knowledge exists on the incidence of palliative patients being under the care of an anaesthesiologist in the operating room or during other procedures.

Our study was undertaken to find out how often anaesthesiologists treat palliative patients during the periprocedural period and to describe the demographics, the anaesthesiological treatment, the medical problems, and the outcome of this patient group.

Of 17,580 patients under anaesthesiological care at our hospital in 2009, 276 could be classified as palliative patients (1.57 %) (Fig. 1). The palliative patients were older, had a higher ASA-classification, showed more often the presence of an oncological disease, and had a higher hospital mortality than the non-palliative patients (Table 2). All these differences were statistically significant (P < 0.001). Concerning the oncological disease, the primary tumor site in most patients was the gastrointestinal tract (27.9 %) followed by the head/neck region (16.6 %) and tumors of the lung (8.7 %). Anaesthesiological and procedural parameters of the palliative patients are shown in Table 3. Most contacts with palliative patients occurred in the operating room (68.5 %). In a total of seven cases palliative patients were treated in a resuscitation situation, since anaesthesiologists lead the cardiac-arrest-teams in our hospital. In comparison to the non-palliative patients, procedures in palliative patients were significantly more often urgent or emergency procedures. Patients undergoing emergency procedures had a significantly higher in-hospital mortality than non-emergency patients (33.3 % vs 14.8 %, P < 0.001) and a significantly higher rate of treatment in the ICU (45.0 % vs. 31.0 %, P = 0.047).

Of the 17,304 non-palliative patients, 12,621 (72.9 % vs. 60.9 % in palliative patients) were elective, 1,870 (10.8 % vs. 17.4 %) were urgent, and 2,813 (16.3 % vs. 21.7 %) were emergency cases (P < 0.001). Most procedures in palliative patients were directly related to the life-limiting disease; only 3.3 % had other reasons (e.g. myringotomy in a patient with otitis media, axillary-femoral bypass in a patient with lower limb ischemia). A substantial amount of palliative patients were treated after the regular working hours (7 a.m. to 4 p.m.) or on weekends (30.4 %). Palliative patients were rarely identified as palliative by the anaesthesiologist, who conducted the preprocedural interview and filled out the anaesthesia record (4.2 %). A do-not-resuscitate order was mentioned on only two anaesthesia records (0.7 %). Most palliative patients were treated by general surgery (29.7 %), internal medicine (24.6 %), and head and neck surgery (18.1 %) (Fig. 2a). In regard to the total number of patients under anaesthesiological care from one specialty (Fig. 2b), the highest rates of palliative patients were in radiation therapy (8.0 %), thoracic surgery (7.4 %), and internal medicine (6.1 %). Since receiving or being in need of a specific palliative therapy to alleviate symptoms was one of the criteria to define a patient as a palliative patient, all patients showed at least one symptom (range 1–6), with pain being the most prevalent symptom followed by gastrointestinal symptoms and nutritional problems (Table 4). 11.6 % of the palliative patients were seen by our hospital’s palliative care consultation service during their hospital stay, with 2.2 % being seen preprocedural.

Nearly 1 in 60 patients under anaesthesiological care in this university hospital could be classified as palliative. An important aspect of this article is how we define a palliative patient. While there is a definition for palliative care by the World Health Organization [14], there is no general consensus on how to define a palliative patient. We based the criteria of this study on the results of a systematic review by Van Mechelen et al. [13], who extracted the key elements for a definition of a palliative care patient. To interpret our study’s results, it is important not to treat “palliative” as equal to “non-curative”. There are numerous medical conditions, which cannot be cured but are stable and non-progressive, e.g. metastasized prostatic carcinoma in the elderly patient or chronic renal failure requiring dialysis. These patients would not require specialized palliative periprocedural treatment and were therefore excluded from the final analysis by our definition (Fig. 1). On the other hand, not all patients receiving palliative therapy were automatically included in the final analysis. Early palliative care becomes more and more available to patients; it is not restricted to patients with a far advanced disease anymore. Also we excluded all patients whose palliative status could not be recognized preprocedurally, because our study focuses on the anaesthesiologist’s assessment of the patient. For example a patient with a pancreatic tumour, that was seen to be irresectable intraoperatively, was not included in our analysis. If we had included these patients in our definition, we would have found a rate of 5.0 % palliative patients (875/17580 patients). It is difficult to compare our results to previous studies, because these studies investigated the rates of palliative procedures, resulting in rates in the range of 6 [12] to 12.5 % [11].

The comparability of our hospital mortality of 18.8 % in this patient population is limited by the fact that the other studies report 30-day or 90-day mortalities, which are 11 [12] and 6 % [15] respectively, again determined in patients undergoing palliative procedures. With the data available to us, it was not possible to determine survival rates or other mortalities than the hospital mortality.

Concerning the symptoms found, three out of the six most prevalent ones lie in the specific area of competence of an anaesthesiologist, pain, pulmonary symptoms, and airway problems. Also neuropsychiatric (e.g. delirium) and gastrointestinal (e.g. nausea and vomiting) symptoms are commonly seen and treated by anaesthesiologists postprocedurally. In the studied patient population anaesthesiologists could become more involved in the preprocedural stage and use their knowledge of symptom control to improve the patient’s well-being before the planned procedure and thereby taking over the role of a perioperative physician.

The results of our study also show, that palliative patients can be encountered in different and unexpected situations (e.g. as emergency cases, outside the operating room, after regular working hours). It is important for every anaesthesiologist to identify a patient as palliative and to recognize his medical problems to deliver optimum care. This seems to be either difficult or deemed as unimportant by the anaesthesiologist conducting the preprocedural interviews in our study, since only 4.3 % of the patients studied were classified as palliative on the anaesthesiological record. Identification of a patient as palliative could also lead to a higher involvement of a palliative consultation service, which has been shown to improve quality of care [16]. Palliative care in general should be initiated as early as possible [17]. We see the need for improvement at this point. The rate of patients seen by our hospital’s palliative consultation service is remarkably low, and patients are obviously seen too late. This indicates that the awareness for this patient group, and their potential benefit from specialised palliative therapy, is generally not high, also among the disciplines directly attending to these patients.

The importance of awareness for this patient group is underlined by the high number of patients undergoing invasive procedures at the end of life or at an advanced stage of disease [18, 19]. The number of these patients is likely to further increase in the future [20]. In this study we had seven patients undergoing cardiopulmonary resuscitation, showing that also in our patient population invasive procedures were undertaken.

Another point for improvement is the consideration of advance directives (AD) and do-not-resuscitate (DNR) orders in the periprocedural context. It is estimated that 25 % of oncological patients in Germany have formulated Ads [21], but these were only mentioned on 0.7 % anaesthesiological records of our patients. While the topic of ADs and DNR orders is widely covered in the international literature [8, 9], a discussion in Germany has yet to be started. There seems to be a low awareness of this problem, resulting in the low mentioning of ADs preprocedurally.

This study shows that anaesthesiologists treat palliative patients under varying circumstances. Palliative patients show a high range of symptoms needing the attention of specialist medical treatment in the periprocedural period, which can be ideally provided by anaesthesiologists. Awareness for this patient group needs to be heightened to adequately recognize and treat these patients. One way to achieve this goal would be to implement a “palliative pathway” for these patients. This would include identifying the patient status by the primarily treating specialty as early as possible, mentioning this status on the notification for the anaesthesiological evaluation, thereby setting the focus on optimized symptom control and recognition of therapeutic restrictions such as AD and DNR orders by the anaesthesiological caregivers, and also integrating palliative care specialists for continuing good symptom control postprocedurally and after discharge from the hospital.