A Prospective Study of Inpatient Ketamine Subanaesthetic Dose Infusion in Chronic Refractory Pain

Background: Inpatient subanaesthetic ketamine infusion for 5 days may improve pain and reduce oral opioid usage in patients with chronic pain. Objective: This study aims to investigate pain and psychological outcomes of ketamine parenteral infusion (0.1 - 0.35 mg/kg/h or maximum 24 mg/hour) for 5 days in patients with chronic refractory pain. The secondary objective is to explore any prognostic pain and psychological factors associated with the successful response to the ketamine treatment. Methodology: A prospective longitudinal study of a small cohort (N = 35) of patients with heterogenous chronic refractory pain conditions was conducted from one week to two months follow-up. Results: Pain Severity was significantly improved from mean 6.5 to 5.1 (t = 3.77, p < 0.001, d = 0.6) at 1-week and 5.9 (t = 2.14, p = 0.042, d = 0.4) at 2-month; Pain Interference from mean 7.0 to 5.1 (t = 4.99, p < 0.001, d = 0.9) at 1-week and 6.1 (t = 2.16, p = 0.041, d = 0.4) at 2-month; Pain Self-Efficacy Questionnaire (PSEQ) from mean 17 to 24 (t = −3.37, p = 0.002, d = −0.6) at 1-week and 23 chronic refractory pain with significant opioid reduction and small improvements in all chronic pain outcomes, except anxiety, at 1-week and 2-month follow-up and with minimal severe adverse effects.


Introduction
The short-term analgesic and rapid anti-depressant effects of subanaesthetic ketamine infusion in chronic refractory pain have been widely reported in the literature [1] [2] [3] [4]. Current literature suggests that if used cautiously, ketamine infusion is safe, modestly effective, well-tolerated and can be used to treat opioid-induced tolerance or hyperalgesia in patients with chronic refractory pain [5] [6] [7] [8]. Expert consensus guidelines support the use of ketamine for chronic refractory pain but the level of evidence varies according to dose range, route of delivery and pain conditions [5].
Recent evidence from systematic reviews and meta-analysis of randomized controlled trials by Jasper [18], Orhurhu [19] and Michelet [20] confirmed subanaesthetic ketamine infusion provided a short-term analgesic response from 2 weeks [19] to 12 weeks [20] in chronic refractory pain. In contrast, controlled studies of multiday low dose ketamine infusion in patients with CRPS reported a long-term analgesic response of 3 months [9] [10] [21] and up to 3 years in a retrospective case series [22]. Long-term responses to ketamine were dose-related, with cumulative effects in repeated and multiday ketamine infusions were postulated to be dose-related [9] [10] [21] [22]. The reviewers hypothesize that active metabolites of ketamine (e.g. nor-ketamine) may play an important role in enduring effects on chronic pain through its neuromodulatory effects on the affective and motivational dimensions of chronic pain [23]. Ketamine may have novel applications as an adjuvant pharmacotherapy to facilitate psycho-behavioural modification in chronic pain, depression and substance use disorder.
Recently, there is evidence showing that intravenous ketamine infusion provides rapid antidepressant and anti-suicidal responses in patients with Treatment Resistant Depression (TRD) [24] [25] and in war veterans with chronic pain [26]. However, there are concerns for recreational misuse, tachyphylaxis, bladder toxicity and hepatotoxicity if ketamine is misused or abused long term  [27] [28] [29]. We postulate that chronic refractory pain is a complex pain condition predominantly mediated by central sensitization caused by excessive NMDA receptor activation resulting in cellular and subcellular changes in the central nervous system. Furthermore, researchers have shown that individual perception and/or experience of chronic pain may be influenced by psychosocial factors, such as self-efficacy, depression, anxiety, stress and catastrophizing. The amount of opioid use in a chronic pain patient may also reflect a patient's lack of pain coping skills, high levels of psychosocial stress, and/or deficiency in descending noxious inhibitory pathways. The primary objective of this study was to investigate the immediate (1-week) and longer term (2-month) responses of pain severity and pain interference after a subanaesthetic dose of ketamine infusion (0.1 -0.35 mg/kg/h or maximum 24 mg/hour) for 5 days as an inpatient treatment for chronic refractory pain. Our secondary objectives were to assess the change in oral Morphine Equivalent Daily Dose (oMEDD) and other psychological outcome measures (pain self-efficacy, depression, anxiety, stress and pain catastrophizing) following ketamine infusion. The final objective was to investigate any baseline predictive factors (e.g. pain phenotype classifications, pain severity & interference, oMEDD usage, psychological outcome measures) that may differ between responders and non-responders to ketamine infusion.

Study Design
This is a prospective open label longitudinal observational study of a cohort of inpatients with heterogenous refractory chronic pain conditions admitted for 5 days of subanaesthetic ketamine infusion (0.1 -0.35 mg/kg/hour, maximum 24 mg/h). Baseline data were collected before commencement of the infusion and the outcome measures collected at 1 week (at the end of infusion) and at 2 months follow-up. The study was approved by the Local Ethics Committee of the South Eastern Sydney Local Health District (HREC reference no. 14/136 (LNR/14/POWH/692).

Patient Selection Process
Patients with chronic refractory pain syndromes of heterogenous phenotypes referred to the Pain Management Clinic at the Prince of Wales Hospital having undergone multidisciplinary assessment who were offered an inpatient ketamine infusion treatment as part of management were invited to participate in the study from 2014 to 2019. Informed consent was obtained from all participants prior to the study commencement by the patient's attending pain physician. This included discussions of the risks and benefits of the ketamine infusion. Participants were generally offered a planned 5-day of inpatient admission for the ketamine infusion and all data related to the study were collected by the pain nurse involved in the study.
Inclusion criteria: 1) Patients able to provide written informed consent prior to the enrolment.
2) Patients diagnosed with severe debilitating chronic refractory (>3 months) pain unresponsive to conventional multimodal medical therapy or multidisciplinary management.
3) Patients with no contraindications or known allergy to ketamine. 4) Patient with heterogenous chronic primary and secondary pain disorders as described in IASP (International Association of Study of Pain). 5) Patient with severe chronic pain disorder that severely interfered with daily activities and psychological well-being (e.g. depression, anxiety and drug dependency).
Exclusion criteria: 1) Patients with unstable psychological or psychiatric conditions, e.g. untreated bipolar disorder, post-traumatic stress disorder, major depression with acute suicidality, severe personality disorder and psychotic illness.
2) Patients having recently undergone major interventional pain procedures, such as nerve blocks or implantable therapies within 3 months of enrolment.
3) Patients with ongoing litigation or legal proceedings with respect to their pain condition. 4) Patients with known drug dependency or substance use disorder specifically related to ketamine or other psycho-stimulant drugs. 5) Patient with previous severe reactions or allergy to ketamine. 6) Patient with severe cardiovascular, respiratory, hepatic, renal or other organ impairments.

Ketamine Parenteral Infusion Protocol
Ketamine was administered parenterally either by subcutaneous or intravenous infusion for 5 days, starting at a dose of 0.1 -0.2 mg/kg/hour according to patient body weight, age and general medical condition and increased gradually by 0.1 -0.2 mg/kg/day slowly daily until significant pain relief was achieved, or development of intolerable adverse effects or a maximum dose of 0.35 mg/kg/hour (or 24 mg/day) was reached.
During the course of inpatient ketamine infusion treatment, the pharmacological interventions were limited to change of opioids or dose (e.g. opioid rotation, opioid reduction) and titration of other non-opioid pain medications in order to optimize pain relief and reduction of medication related harms.
This dose range of ketamine is unlikely to cause any potential serious complications. Transient minor side effects may occur, such as dizziness, nausea, hallucination, visual disturbance, tachycardia and hypertension. These transient adverse effects usually subside with the cessation of the ketamine infusion. The duration of ketamine infusion was limited to 5 days in order to allow for therapeutic response while avoiding long-term neurocognitive, hepatic and urological adverse effects. Item 1 asked patients to rate the change in their Overall Status on a 7-point scale from −3 "Very much worse" to +3 "Very much improved". Item 2 asked patients to rate how much Pain Relief has been provided from "0% No relief" to "100% Complete relief". A successful response to the infusion was defined as a reported Pain Relief of 30% or more [33]. All outcome data (BPI, PSEQ, DASS-21, PCS, GPI, AEC, oMEDD) were collected at 1 week and recollected at 2 months after the ketamine infusion by the pain nurse who served as an independent observer. All data collected were entered into the Australasian online electronic Persistent Pain Outcome Collaboration (ePPOC) database. The data was then retrieved by the study investigators through an online login to the ePPOC system identifying the patient file, outcome data and date of service. At the end of the study, participants who withdrew from the study and their reasons were documented. Other relevant medical information including medications and doses were extracted from patient's medical records. was used to detect differences in outcome measures between admission and discharge and between admission and follow-up. The Student's independent samples t-test was used to detect differences between males and females on admission and on discharge, between responders and non-responders on admission and at discharge, and between those who completed the outcome measures at follow-up and those who did not. Pearson's Chi-squared test was used to detect whether females and males were equally represented in the responder category.

Statistical Analysis
Significance level for all tests was set at p = 0.05. Patients were classified according to the type of pain they described ( Figure  1). The largest group was nociplastic pain (N = 13) including Complex Regional Pain Syndrome (CRPS) and Fibromyalgia Syndrome (FMS). Another large group was neuropathic pain (N = 11) including peripheral and central neuropathic pain. (e.g. radiculopathy and spinal cord injury pain). The smaller groups were nociceptive pain (N = 4) including visceral pain and musculoskeletal pain, and lastly mixed pain (N = 7) including mixed neuropathic and nociceptive pain as shown in Figure 1.

Results
The outcome measures on admission and at 1-week were completed by all 35 patients. Due to incomplete data, only 25 patients (completers) were included in the analysis at the 2-month follow-up. There were no differences between completers and non-completers on any outcomes on admission (t 33 < 1.67, p > 0.105). At 1-week on discharge, oMEDD was found to be lower for completers (mean 96 mg SD 81) compared to non-completers (mean 188 mg SD 134) (t 33 = 2.504, p = 0.017). There were no differences between females and males on any outcome measures on admission or after 1-week on discharge (t 33 < 1.52, p > 0.138).
The results of statistical analysis of all pain and other outcome measures on admission, at 1-week on discharge and at 2-month follow-up are shown in Table   2.    The following secondary outcomes were all significantly improved at both 1-week on discharge and at 2-month follow-up: oral Morphine Equivalent Daily Dose (oMEDD) is shown in Figure 4; Pain Self-Efficacy (PSEQ) in Figure 5; Depression and Stress in Figure 6 and Figure 7; and Pain Catastrophization Score (PCS) in Figure 8. There was no improvement in Anxiety at any time point.
The opioid oMEDD reduction was significant at 1 week from mean dose 191 mg/day (SD 185 mg day) to 122 mg/day (SD 106 mg/day); and to 93 mg/day (SD 77 mg/day) at 2-month. The total mean oMEDD reduction of 191 mg/day to 93 mg/day was 36% at 1-week (p = 0.023) and 122 mg/day to 93 mg/day was 51% at 2-month (p = 0.016).
3) Responder Analysis (≥30% Pain Severity Reduction) The Global Perception of Improvement (GPI) items of Overall Status and Pain Relief were completed by 26 patients at 1-week on discharge, with 9 patients not     answering these items. As the Pain Relief item was not used at the two-month follow-up, it was replaced as the determinant of response by the BPI-Pain Severity score [32]. IMMPACT recommends that a moderately important clinically significant reduction in pain is 30% or more on the BPI-Pain Severity score [33] [34]. Therefore, a 30% cut-off was used to determine a successful response to the ketamine infusion at 1-week on discharge and at 2-month follow-up.  Table 3.

4) Global Perception of Improvement (GPI)
At discharge the GPI items Overall Status and Pain Relief were completed by 26 of the 35 patients. Overall Status was reported to be improved by 81% of these patients (very much improved N = 5, 19%; much improved N = 8, 31%; and minimally improved N = 8, 31%). Four patients reported no change (15%) and one reported being minimally worse (4%). The incidence of adverse effects related to treatment was categorized into central nervous system (CNS) (e.g. neurocognitive disturbance, hallucination, dizziness), gastrointestinal (e.g. nausea, vomiting, abdominal bloating), skin (e.g. redness on insertion site, itchiness, inflammation/infection), and others (e.g. hepatic, urological). A total of 28 patients completed the data of adverse effects at the end of infusion on discharge. The majority of patients experienced CNS-related adverse effects (n = 19), followed by gastrointestinal (n = 13), skin (n = 11) and others (n = 1). Some patients described multiple adverse effects on different systems. Only 4 out of 28 patients (14%) described no adverse effects as shown in Figure 9. Most adverse effects were mild in severity, transient and no treatment was required except reduction in infusion dosage. No patients were severely disturbed by neurocognitive adverse effects.

5) Adverse Effects
6) Validity of Measures as Inpatients Responses to some of the outcome measures on discharge are likely to be related to the limitations imposed by the experience of being an inpatient. For example, the BPI-Pain Interference scale asks patients to report how much pain has interfered during the past week with general activity, mood, walking ability, normal work, relations with other people, sleep and enjoyment of life. Change or the lack of change from admission to discharge should be viewed in this context.

Discussion
Approximately 20% of the adult Australian population experience persistent pain, and of these people, one third have high levels of disability [35] [36]. Globally, chronic refractory pain is the leading cause of years lost to disability, and poor physical and mental health [37]. Chronic refractory pain can be defined as persistent pain, which is resistant to conventional medical therapies, or neuropathic, nociceptive and other pain complicated by the development of central   [39]. They may lead to development of depression [40], anxiety, overuse of opioid analgesics with the risk of dependence and toxicity [41] and increase the risk of suicide [42].
Ketamine as a potent NMDA receptor antagonist has anti-neuropathic, anti-hyperalgesic, anti-depressant and opioid sparing effects [43] [44]. Recently, its metabolite Hydroxynorketamine (HNK) is postulated to produce rapid (within days and weeks) antidepressant and anti-suicidality effects [45] mediated by activation of the AMPA receptor [46]. As part of a multidisciplinary approach, ketamine as an adjuvant pharmacotherapy may be able to modify pain cognitions and coping strategies to improve pain management and quality of life [47]. This may be aided by ketamine's neuromodulatory effect on both the somato-sensory and affective-motivational dimensions of pain by enhancing neural activity in anterior cingulate cortex, orbital frontal cortex, insula and brainstem in healthy volunteers [48].
The pain numerical rating scale effect size of this small study was consistent with other randomized control trials [9] [10] [11] [12] in showing rapid onset of analgesic effect within 1-week (6.5 to 5.1, effect size −0.6, p < 0.001) and sus-  [49]. Another systematic review and meta-analysis of randomized controlled trials for chronic pain reported small effect up to 2 weeks after ketamine infusion with significant mean pain difference in pain scores, −1.83 points on 0 -10 numerical rating scale [19].
There was a reduction in opioid dose (oMEDD) from mean 199 mg to 122 mg at 1 week (p = 0.023, 36% reduction) and 93 mg at 2-month (p = 0.016, 51% reduction). This is consistent with that reported in another longitudinal study of ketamine infusion followed by long-term ketamine sublingual lozenge therapy [50]. The opioid usage as measured by the oMEDD was reduced by 59% following inpatient ketamine intravenous infusion for 5 days. The authors found that the use of ketamine lozenges was effective in maintaining the cessation of opioid therapy in 31% of their patients. Ketamine was found to be safe, and efficacious in facilitating opioid reduction in chronic refractory pain patients. Limitations of the study include the relatively small sample size; incomplete data; variable ketamine dosing (subanaesthetic dose); heterogenous pain phenotypes; and the absence of measures of temporal summation and quantitative sensations. Our study was not placebo-controlled and randomized, hence limiting generalizability. We did not control for unknown confounders (e.g. confounding effect of opioid rotation and reduction, and other non-opioid drug titrations used concurrently with the ketamine infusion, and the inpatient experience itself). Nevertheless, it is a pragmatic and clinically relevant treatment given in a supervised context as part of multidisciplinary pain management. Finally, this study found ketamine to be safe, well tolerated, and very efficacious in We recommend that future research should continue to use validated multidimensional measures, including patient-reported assessments on pain outcomes, determine the longer term sustainability of opioid dose reductions, and the effectiveness of repeated ketamine infusions. As ketamine infusions require several days of inpatient treatment, investigations could explore different treatment delivery modalities such as day-stay outpatient infusions to reduce patient risk and discomfort as well as hospital-related costs and length of hospital stay. Quantitative Sensory Testing (QST) and testing for the contribution of temporal summation as a surrogate measure for wind-up in central sensitization could also be explored as a predictor of response to ketamine infusion.

Conclusion
In this small study, inpatient ketamine subanaesthetic dose infusion contributed to improvements in all outcome measures, except anxiety, that were evident at 1-week on discharge and were maintained for those reporting at 2-month follow-up. There were significant improvements in pain severity and pain interference, depression and stress, pain catastrophization and pain self-efficacy. Oral opioid oMEDD reduction was 36% at 1-week and 51% at 2-month. There were no group differences between responders versus non-responders analysis in psychological outcome and pain classifications. Rates of adverse effects from ketamine infusion were high but these were mostly mild, requiring no specific treatment other than dose adjustment.