Background
The total pain concept is most useful in palliative care as pain may occur on a physical, psychological, social and spiritual level [
1,
2]. It is well established that physical pain and psychological distress are connected [
3]. About two thirds of patients with advanced cancer suffer from pain, and more than half of those experience moderate to severe pain [
4]. The WHO cancer pain relief guidelines are the standard therapy for pain [
5,
6], which may achieve acceptable pain relief in over 50% of the treated patients [
7]. Thus, there is still a sizeable group of patients with pain refractory to pharmacological treatment [
8]. The N-methyl-D-aspartate (NMDA) receptor is an important structure in the conduction of pain signals [
9] and is relevant to pathological pain states [
10,
11].
Ketamine is a non-competitive NMDA-receptor antagonist that is effective in the treatment of refractory cancer pain [
12‐
16]. Reviewing the existing data for ketamine as an adjuvant to opioids for cancer pain, Bell et al. concluded that the data is not yet sufficient to evaluate the usefulness of ketamine [
17‐
19]. Other authors came to the same conclusion but also considered ketamine as a reasonable option for refractory, severe neuropathic and chronic pain [
20‐
22].
Ketamine has dose dependent analgesic and anesthetic properties with sympathomimetic side effects while preserving protective reflexes [
23]. It belongs to the WHO essential medicines [
24]. Ketamine binds at the phencyclidine binding site of the NMDA receptor and interacts with other receptor types like opioid and cholinergic receptors [
25,
26]. It is used in clinical practice as a racemic mixture in a 1:1 ratio of the
R- and
S-enantiomer of ketamine and as pure
S-ketamine [
27].
S-ketamine has the potency to block the NMDA-receptor about two times stronger than
R-ketamine [
28]. The anesthetic potency of
S-ketamine is about twice as high as racemic ketamine and three to four times higher than
R-ketamine [
29]. Ketamine has psychotomimetic side effects including perceptual distortion and cognitive disorganization [
30]. Ketamine also induces hallucinations and changes in mood and body image [
23,
29]. It is abused as a recreational drug and can cause addiction [
31]. The abuse of ketamine is associated with urological, neuropsychiatric, hepatobiliary and gastrointestinal complications [
32]. In the treatment of depressive or anxious patients, racemic ketamine (0.50 mg/kg), as an i. v. infusion lasting 40 min, has an anti-depressive and anxiolytic effect [
33,
34]. This effect begins a few hours after application, peaks at 24 h and lasts about one week [
35]. The effect can be maintained through repeated application of racemic ketamine [
36,
37]. There is also evidence of an anti-depressant [
38,
39] and anxiolytic effect of enantiomer
S-ketamine [
29]. Recent clinical studies indicate that nasal application of
S-ketamine in combination with an oral antidepressant reduces treatment-resistant depression [
40‐
44]. The United States Food and Drug Administration (FDA) recently approved
S-ketamine as a nasal spray to be used in conjunction with an oral antidepressant for the therapy of treatment-resistant depression [
45]. The positive psychological effect of ketamine is attributed to an induction of neuroplasticity, which reverses the negative effect of stress and depression on neural cells and synapses [
46]. Key enzymes in this process are BDNF (brain derived neurotrophic factor) [
47] and mTor (mechanistic target of rapamycin) [
48].
The rapid effect of ketamine on stress, anxiety and depression may be of huge importance for the treatment of psychiatric conditions of patients in palliative care. A total of 29% of patients in palliative care suffer from adjustment disorder, anxiety disorders or depression [
49]. Anxiety and depression are related to a lower quality of life [
50]. Depression and hopelessness are associated with a desire for hastened death [
51] and higher rates of suicide [
52]. Additionally, research data suggests that physical and psychological symptoms are interlinked. For example, there is a positive association of depression and anxiety with pain [
53] and with physical symptom burden in patients in palliative care [
54]. Moreover, anxiety and depression are associated with higher mortality in cancer patients [
55]. Current pharmacological therapy options include benzodiazepines for anxiety and antidepressants for depression [
56]. Benzodiazepines have a fast onset of action, but also limiting side effects like sedation, confusion, loss of coordination, addiction and paradox effects [
56,
57]. Antidepressants often need about six weeks to achieve remission and do not have an effect in one third of depressive patients [
58]. Six weeks is a long time for palliative care patients; too long for many. Thus, there is a need for fast-acting and reliable therapy options for these patients. Existing evidence points to a positive influence of racemic ketamine on depression and anxiety of patients in palliative care [
59,
60].
In our specialized palliative care unit (SPCU) we regularly use
S-ketamine (0.25 mg/kg i. v. infusion over 45 min) as an analgesic treatment for therapy refractory pain.
S-ketamine is favored over racemic ketamine because it has higher analgesic and anesthetic potency and it shows less psychotomimetic side effects [
29]. In a retrospective pilot study, we analyzed clinical routine data. We investigated whether an analgesic therapy with
S-ketamine has a positive impact on psychological distress caused by anxiety and depression of patients in palliative care compared to a control group. This research question is of high interest because, to our knowledge, there is a lack of data regarding the influence of
S-ketamine on psychological distress of patients in palliative care. Additional statistical calculations are performed to address variables with potential confounding influence, i.e. pain, need for physical care, received psychological support, received specialized palliative patient treatment, duration of anti-depressant therapy and medication with benzodiazepines, opioids and antidepressants. Furthermore, hints for longer lasting analgesic and psychotomimetic effects of
S-ketamine are examined. The main hypothesis is that compared to a control group, an analgesic
S-ketamine infusion reduces psychological distress caused by anxiety and depression. Additionally, it is hypothesized that even after taking the confounding variables into account, the positive effect of
S-ketamine on psychological distress remains.
Discussion
This retrospective pilot study provides the first evidence of a positive effect of
S-ketamine on the psychological distress of patients in palliative care. We find a multivariate effect on depression and anxiety with a primary effect on anxiety. Our result corresponds to earlier studies showing that ketamine racemate shows similar effects in patients in palliative care [
59,
60,
75]. To our knowledge, the effect of the purified enantiomer
S-ketamine on patients in palliative care has not been analysed previously.
S-ketamine is reported to have a positive effect on anxiety in surgical patients without palliative diagnosis [
29], and has recently been approved as nasal spray by the FDA – but only when used in conjunction with an oral antidepressant and only for the therapy of treatment-resistant depression [
45]. We hope that for patients with a life-limiting disease,
S-ketamine can be useful outside the FDA approval. This study may be a first step towards the approval to treat anxiety of patients in palliative care with
S-ketamine.
The positive effect of
S-ketamine was mainly on anxiety with no significant effect on depression. The influence of
S-ketamine on anxiety had consistently large effect sizes. Our data indicate that S-ketamine treatment may be effective in routine clinical practice. In our study,
S-ketamine reduced the global STADI values by a clinically relevant level in 5 out of 8 patients (Fig.
1). Thus, we estimate the number needed to treat is approximately 2. Further studies are needed to establish the effectiveness.
The influence of S-ketamine on depression showed mainly medium effect sizes. The significant effect with a large effect size of S-ketamine on psychological distress was mainly caused by the reduction in anxiety. However, the analyses also showed that the changes in anxiety and depression due to S-ketamine were similar. Thus, S-ketamine had an analogical effect on anxiety and depression. Even after taking the confounding variables into account, the significant effect on anxiety remained. There was also no evidence of persistent psychotomimetic side effects in the S-ketamine group until the next morning. Furthermore, there were no indications of a sustained pain reduction by S-ketamine until the next morning in the group comparison.
The pronounced effect of
S-ketamine on the anxiety of patients in palliative care may be related to the peculiarities of this group of patients. In a case report on two hospice patients receiving a single dose of ketamine racemate (0.50 mg/kg bolus per os) to treat psychological distress, there was a positive effect on anxiety and depression, with a more pronounced reduction in anxiety over the first four days [
60]. In addition, both patients experienced an improvement in pain perception with a maximum of four and eight days after ketamine administration. In a feasibility study, the effect of daily oral administrations over 28 days of ketamine racemate (0.50 mg/kg bolus per os) on anxiety and depression was investigated [
75]. There was a significant response (reduction of questionnaire scores > 30%) of anxiety to ketamine racemate after three days with a medium effect size (
d = 0.67). For depression, there was a significant response after 14 days with a large effect size (
d = 1.14). After 28 days a significant effect was sustained with large effect sizes for anxiety (
d = 1.34) and depression (
d = 1.34). However, pain was unchanged [
75].
The results of our work and the two hospice studies suggest that S-ketamine and ketamine racemate act primarily on anxiety in patients with a life-limiting disease. Whether this is a special pattern of action in these group of patients requires further clarification.
In our study, a positive effect of
S-ketamine on depression could not be identified in the group comparisons. However, the initial univariate group comparison (Table
4) found a medium effect size (
r = 0.32) for depression. A post-hoc sample size calculation with G*Power 3.1 [
76,
77] showed that a total of
n = 20 patients would be necessary to determine a significant effect on depression for a group-by-time interaction in a two-way mixed ANOVA. Thus, according to our data, a prospective study would need 20 patients or more.
In this study, the descriptive interpretation of the data suggests that more patients in the
S-ketamine group than in the control group died on the ward (Table
1). A causal relationship to
S-ketamine is not plausible for the following reasons: In general, about 60% of the inpatients in the SPCU die on the ward [
78]. Thus, the mortality of the
S-ketamine group can be considered average. Furthermore, a study by Irwin et al. [
75] showed that daily oral administrations of ketamine racemate for 28 days led to no serious adverse events. There were no changes in vital signs (blood pressure, heart rate and respiratory rate) during the course of their study. A mild increase of symptoms in 12.5% of patients were related to diarrhea, sleeping problems and restlessness. In addition, the patients showed a decrease in symptom burden related to gastrointestinal, neurological and psychiatric symptoms. Further studies on the effect of ketamine racemate on the mental health of hospice patients [
59,
60] and of psychiatric patients [
79] also report a low rate of adverse events. The most frequent adverse events in patients receiving ketamine racemate (0.50 mg/kg over 40 min i. v.) as a therapy of treatment resistant depression were drowsiness, dizziness, poor coordination and a strange or unreal feeling [
79]. These symptoms were mostly experienced in the first two hours after the beginning of the infusion, diminishing after four hours and practically ceasing after 24 h.
A descriptive synopsis of the data collected in this study suggests that the S-ketamine group was a group of patients with a higher symptom burden than the control group. The S-ketamine group showed, at T1, STADI T-values over the critical limit of 60. In addition, the S-ketamine group reported moderate pain at both time points. Furthermore, the S-ketamine group showed, at both points of measurement, a need for more care than the control group (as indicated by the AEDL score). On average, the S-ketamine group also had less psycho-oncological treatment. It is plausible that the reduced physical status of the S-ketamine group, which was manifested in increased need for care and increased mortality on the ward, reduced the possibility of participating in psycho-oncological interventions.
Limitations
The limitations of this study results from its retrospective design, which prevented randomization. Because of the retrospective design, the data is not optimal to measure the effect of
S-ketamine. The best interval to measure the maximum effects of ketamine or
S-ketamine is one day after administration of ketamine or
S-ketamine. In our study, there were several days between T1 and T2 in the
S-ketamine group (Fig.
1), which may have reduced the measured effect of
S-ketamine on anxiety and depression. However, the obtained data (Fig.
1) showed a stronger reduction of psychological distress caused by anxiety and depression one day after
S-ketamine administration than four days after
S-ketamine administration. Thus, these data are consistent with the time course of the effect of ketamine [
80]. Furthermore, the retrospective approach does not allow an evaluation of how the patients have perceived the effects of
S-ketamine and how they assess the benefits and risks related to
S-ketamine treatment. Non-randomization can lead to systematic bias. In this study, the group membership was systematic, because only the patients suffering from refractory pain received
S-ketamine. In this context, additional patient data indicates that the
S-ketamine group was a patient population with a higher symptom burden. Thus, the patients in the
S-ketamine group, who suffered from refractory pain, still had other physical and psychological symptoms, which distinguished them from patients in the control group. The STADI scores for anxiety and depression were significantly higher in the
S-ketamine group than in the control group at T1. To minimize statistical errors arising from the selection of the control group, we generated another control group. This alternative matching strategy takes psychological distress into account. Still, when using the alternative matching strategy, our results on the effect of
S-ketamine are essentially the same (see Additional files). To avoid other statistical errors, we calculated Cronbach’s alpha and test-retest-reliability to ensure good reliability of our instruments. For the analyses of variance, we ensured that all assumptions of normal distribution, homogeneity of variance and homogeneity of covariance matrices were met.
A further limitation of our study is the non-randomization and the small sample size, which makes it difficult to generalize the data. Despite this limitation, our results provide the basis for prospective studies, which will be needed as soon as S-ketamine is approved as nasal spray by the European Medicines Agency and other regulators around the world. Subsequent studies will provide an empirical basis for the treatment of anxiety and depression with S-ketamine of patients in palliative care. The first step would be prospective feasibility studies, including qualitative data if sample size is expected to be low. Further studies could include double-blinded, randomized and placebo-controlled trials. During the course of these studies, the questions on the pattern of effects, the optimal forms of application and the choice of medication S-enantiomer vs. racemate should be considered.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.