Sage Journals: Discover world-class research

Abstract

Introduction:

Mechanical ventilation (MV) is a common and often live-saving intervention on the Intensive Care Unit (ICU). The optimisation of sedation to mechanical ventilation is fundamental, and inappropriate sedation has been associated with worse outcomes. This scoping review has been designed to answer the question ‘What is known about the use of ketamine as a continuous infusion to provide sedation in mechanically ventilated adults in the intensive care unit, and what are the gaps in the evidence?’

Methods:

The protocol was designed using the PRISMA-ScR checklist and the JBI manual for evidence synthesis. Data were extracted and reviewed by a minimum of two reviewers.

Results:

Searches of electronic databases (PubMed, OVID, Scopus, Web of Science) produced 726 results; 45 citations were identified for further eligibility assessment, an additional five studies were identified through keyword searches, and 12 through searching reference lists. Of these 62 studies, 27 studies were included in the final review: 6 case reports/case series, 11 retrospective cohort/observational studies, 1 prospective cohort study, 9 prospective randomised studies.

Conclusion:

We found a lack of high-quality well-designed studies investigating the use of continuous ketamine sedation on ICU. The available data suggests this intervention is safe and well tolerated, however this is of very low certainty given the poor quality of evidence.

Keywords

Intensive care critical care sedation ketamine mechanical ventilation scoping review

Background

Mechanical ventilation (MV) is a common intensive care intervention (approximately 20 million patients per annum¹ worldwide) that accounts for much intensive care unit (ICU) resource utilisation. Despite efforts, patients requiring MV unfortunately still experience high mortality and morbidity.²

Most critically ill patients requiring MV are medically sedated with optimisation of sedation and analgesia being fundamental to management. International guidance has been aimed at improving outcomes³ owing to reported associations between deep sedation and negative prognostic markers,⁴ however these results have not been consistently demonstrated⁵ and agent selection is a balance of risks and benefits. Providing adequate sedation, maintaining comfort, reducing pain, and minimising agitation and delirium have been identified as top priorities for ICU research by the intensive care community through the James Lind Alliance.⁶

In a survey of ICUs in the UK, propofol combined with either alfentanil or fentanyl was the most common sedation-analgesia regime; 92.2% of units reported propofol as first line agent.⁷ A third of units reported using other non-ketamine sedative agents either ‘frequently’ or ‘very frequently’, including benzodiapines (29.4%), clonidine (35.3%), and dexmedetomidine (11.8%).

Most traditional sedatives (including propofol, benzodiazepines, and alpha-2-agonists) are associated with multiple significant and potentially problematic adverse effects, (commonly hypotension, bradycardia, and prolonged mechanical ventilation ). The predominant mechanism is attenuation of external stimulation, reduced sympathetic tone, and vasodilation.⁸

Sedatives, in particular benzodiazepines significantly increase the risk of delirium in ICU.^3,9 Delirium during critical illness is associated with higher mortality, longer ICU stay, and poorer long-term outcomes.¹⁰

Ketamine is an N-methyl D-aspartic acid (NMDA) receptor antagonist that has been used since the 1970s to provide cataleptic, amnesic, analgesic, and dose dependant anaesthetic effects.¹¹

Ketamine has been particularly successful in military¹² and pre-hospital¹³ settings owing to its ability to stimulate the sympathetic nervous system, preserving heart rate and blood pressure, whist avoiding respiratory suppression,¹⁴ as a result ketamine has become increasingly popular as an anaesthetic agent for emergency surgical procedures in hypotensive patients.¹⁵

Although ketamine has been in clinical practice for approximately 50 years; ketamine by continuous infusion remains an unlicensed and rarely used sedative to facilitate MV.⁷ Ketamine infusions have however been used successfully in mechanically ventilated patients on ICU in the context of severe asthma,¹⁶ utilising ketamine’s bronchodilatory properties.

Reluctance around more routine use of ketamine is likely multifactorial and may relate to lack of clinical familiarity (as demonstrated in two surveys of UK sedation practices^7,17), perceived contraindications such as raised intracranial pressure (ICP), and possible side effects such as ‘emergence reactions’.

Emergence reactions are psychomotor symptoms experienced by some patients when waking from ketamine anaesthesia. The reported incidence of these reactions varies from 5% to 30%.¹⁸

Milder manifestations such as floating sensations, disorientation, excitement, dysphoria, vivid dreams, and amnesia, through to more marked reactions including frank delirium, hallucinations, and agitation^18,19 are well documented in the context of general anaesthesia. To our knowledge these have not been examined in the context of the critically ill, where neuropsychiatric dysfunction and risk-benefit ratio differ markedly from those in general anaesthesia.

Ketamine was recently licenced for major depressive disorder (MDD) treatment following reviews that highlighted transient psychotomimetic effects with single-dose administration.^20,21

MDD and post-traumatic stress disorder (PTSD) are common amongst the survivors of ICU, with the prevalence of MDD ranging in the literature from 17% to 43%^22,23 and PTSD ranging from 21% to 35%.^22,24 Associations between perceived risk factors and PTSD have been extensively examined with varied results.²³ Duration of ICU delirium, early post-ICU depressive symptoms, the use of benzodiazepines, and physical and cognitive impairment during recovery and rehabilitation have all been identified as significant risks for developing MDD.^24
–26

To our knowledge there is currently no evidence suggesting administration of ketamine for the purpose of sedation during mechanical ventilation can influence longer-term psychological outcomes. This is of interest in this review given the prevalence of MDD and PTSD and could represent an area for potential future research including clinical trials specifically investigating ketamine and critical care recovery and rehabilitation.

As part of this scoping review we aim to establish what literature exists regarding the use of continuous ketamine sedation on ICU, particularly in relation to cardiovascular effects, delirium and the psychological aspects of ICU.

Methods

Aim

The aim of this scoping review is to review current literature in order to answer the following research question: ‘What is known about the use of ketamine as a continuous infusion to provide sedation in mechanically ventilated adults in the intensive care unit, and what gaps in the evidence exist?’

Objectives

To search relevant databases to identify peer-reviewed literature investigating ketamine’s efficacy, safety, and clinical effects when used as a continuous infusion in MV patients on ICU.

Protocol

This study was designed using the Preferred Reporting Items for Systematic Reviews and Meta-analysis extension for Scoping Reviews (PRISMA-ScR) checklist²⁷ and the JBI Manual for Evidence Synthesis.²⁸

The protocol is available from Research Square (https://doi.org/10.21203/rs.3.rs-910472/v1).

Search strategy and terms

Full details of the search strategy employed can be found in the appendix (see Appendix B)

A three-step literature search was used in accordance with JBI methodology for scoping reviews²⁸:

A preliminary search of PubMed was conducted using the terms: ‘(Ketamine) AND (Intensive Care) AND ((Sedation) OR (Analgesia))’.

Subsequent searches were conducted across all included databases (PubMed, Embase, OVID, Scopus, Web of Science) using words and phrases contained in the title, abstract, and index terms of papers retrieved during the preliminary search.

Reference list of identified reports and articles were analysed for additional sources and reports.

Eligibility criteria

Only full-text reports published in English in peer-reviewed literature were included.

Inclusion criteria:

- Adult patients (⩾18 years old)

- Use of continuous ketamine infusions for the purpose of primary or adjunctive analgesia or sedation during mechanical ventilation

Accepted methods included experimental and quasi-experimental designs (e.g. randomised controlled trials, non-randomised controlled trials, and interrupted time-series studies), analytical observational studies including prospective and retrospective cohort studies, case-control, case series, individual case reports and descriptive cross-sectional studies.

Systematic reviews and meta-analyses were not included, but their references were screened for appropriate studies that met the inclusion criteria. Conference abstracts, qualitative studies and opinion papers were not considered for inclusion in this scoping review.

Information sources

Searches were made across PubMed, Embase, OVID, Scopus, and Web of Science using the inclusion criteria above.

Screening of sources

Two reviewers (NDR and WW) independently screened titles and abstracts for inclusion. Full texts were obtained for relevant identified articles and citations were collated in EndNote X6.

Disagreements arising between reviewers at each stage of the selection process were resolved through discussion, without the need for an additional independent reviewer.

Data charting process

NDR and WW extracted data from all included publications using the data extraction tool developed by the reviewers (see Appendix A).

The data extracted included publication details, study design, setting, intervention details, primary and secondary objectives, and key findings.

Critical appraisal of sources

Sources were critically appraised using the relevant aspects of the CASP checklists (https://casp-uk.net/casp-tools-checklists/) by two independent reviewers (NDR and WW).

Synthesis of results

Results are presented with frequency counts of concepts, populations, and study characteristics, accompanied by a descriptive summary. Studies are grouped based on methodological design, along with outcome measures, and relevance to this review.

Results

Selection of sources of evidence

Searches of electronic databases produced 726 results. After duplicate removal and initial screening of titles and abstracts, 45 potentially relevant citations were identified for full-text retrieval and further eligibility assessment. Common reasons for exclusion at this stage were review publications, incorrect patient population (paediatric or intra-operative), incorrect intervention (non-continuous ketamine), and non-human studies.

Five studies were identified from keyword searches and searching reference lists of the full-text and review articles identified 12 further studies for eligibility screening.

Of the 62 studies undergoing eligibility screening, 35 were excluded: 10 were judged not to be original quantitative research (e.g. review articles, commentaries), 10 were unavailable in English or full-text form, and 15 either did not include the study population of interest or outcomes of interest (paediatric populations, intra-operative populations, participants not intubated and ventilated) (Figure 1). The remaining 27 studies are included in this review and are reported in Table 1.

Figure 1.

Searches, inclusion, and exclusion flow diagram.

Table 1.

Characteristics of included studies.

Author (year)	Country of origin	Population	Study Design	Sample size	Intervention (±comparator)	Dose/Duration of sedation	Primary and Secondary Outcomes	Key findings
Prospective Randomised Studies
1. Amer M (2021)	Saudi Arabia	Medical, surgical and transplant ICU patients	Active-controlled, pilot, feasibility clinical trial	40 (+43 control)	Ketamine + fentanyl (vs standard care – fentanyl ± propofol ± dexmedetomidine ± midazolam)	Dose: • 1st 24h–0.06mg/kg/h• 2nd 24h–0.12mg/kg/hDuration: • Maximum duration of 48 h	Primary outcome: • Feasibility of larger randomised controlled trialSecondary outcomes: • Duration of MV• Richmond agitation sedation scale (RASS)• ICU survival• ICU length of stay (LoS)• Hospital LoS• 28-day Mortality• CAM-ICU• Agitation/combative behaviour• Haemodynamics• Antipsychotic use• Significant adverse events (SAEs)/Adverse Drug Reactions (ADRs)	• Feasibility thresholds (recruitment/consent/protocol adherence) met• No significant difference in ventilator free days (p = 0.7)• Ketamine group achieved more target RASS scores at 24 h (67.5%vs 52.4%) and 48 h (73.5%vs 66.7%)• No significant difference in MAP• No significant difference in 28-day mortality (p = 0.79)• No significant difference in ‘rescue medication’ or antipsychotic use between groups• No SAEs, emergence reactions, confusion, or nightmares
2. Bourgoin A (2003)	France	Neuro ICU (head injury)	Prospective randomised double blind study	12 (+13 control)	Ketamine/midazolam (vs sufentanil/midazolam)	Dose: • Ketamine 4.92 ± 1.5 mg/kg/h + midazolam 0.098 ±0.03mg/kg/h• Sufentanil 0.008 ± 0.002 mcg/kg/min + midazolam 0.098 ±0.02mcg/kg/minDuration: • Ketamine/midazolam 6.2 ± 3.2 days• Sufentanil/midazolam 5.3 ± 3.8 days	Primary outcome: • intracranial pressure (ICP)/Daily ICP elevations and daily cerebral perfusion pressures (CPP)Secondary outcomes: • Cardiovascular parameters• Mortality• ICU LoS• SAEs/ADRs	• No significant variations in ICP (p = 0.28)• No significant variations in CPP between groups• No significant differences in MAP• Volume of fluid administered was significantly less in ketamine group (429 ±405ml vs 992 ±703ml; p = 0.02)• No significant difference in ICU LoS (21 ± 13vs 18 ± 13 days) or mortality at 6 months (2vs 1)
3. Christ G (1997)	Austria	ICU patients with catecholamine dependent heart failure	Prospective randomised study	13 (+12 control)	Ketamine/midazolam (vs sufentanil/midazolam)	Dose: • Ketamine 2.5 ± 0.9 mg/kg/h + midazolam 0.12 ±0.04mg/kg/h• Sufentanil 0.88 ± 0.33 mcg/kg/h + midazolam 0.15 ±0.07mg/kg/hDuration: • Unspecified	Primary outcome: • Cardiovascular and haemodynamic parametersSecondary outcomes: • Vasopressor/catecholamine dose	• Ketamine group had significant increase in MAP at 24 h (68 ±11mmHg to 77 ±14mmHg; p = 0.01)• Ketamine significantly increased mean pulmonary artery pressure at 24 h (29 ±8mmHg to 33 ±10mmHg; p = 0.04)• Ketamine significantly increased systemic vascular resistance index (p < 0.001)• No change in vasopressor/catecholamine dose
4. Gupta B (2022)	India	Surgical ICU	Prospective randomised controlled study	30 (+30 control)	Ketamine-dexmedetomidine (KD) vs dexmedetomidine only (DEX)	Dose: • KD – 0.4mg/kg/h + 0.45 mcg/kg/hDEX–0.7 mcg/kg/hDuration: • KD – 7.5 h±1.6DEX – 8.5 ± 1.2	Primary outcome: • Sedation Scale (unnamed − ‘0’ = no sedation, patient wide awake and alert; ‘4’ = deep sleep, difficult to rouse)Secondary outcomes: • Cardiovascular parameters at 30, 60, 120 min, 6 h• Pain scores at 1, 2, 4, 6 h, post-extubation• Requirement of additional analgesia (fentanyl 1 mcg/kg bolus)• Time to extubation• Satisfaction of patients• SAEs/ADRs	• Sedation scores were higher in KD group (p < 0.05)• Systolic BP was lower in DEX group at all time points (p < 0.05)• Heart rate was lower in DEX group at 30 min and 60 min (p < 0.05)• Pain scores were higher in DEX group at 2 h and 4 h (p < 0.05)• Time to extubation was less in KD group (not significant)• LoS were comparable.• More ADRs including hypotension and bradycardia requiring atropine in DEX group• No hallucinations in either group
5. Mogahd MM (2017)	Egypt	Post CABG	Prospective randomised study	35 + 35	Ketamine/ dexmedetomidine (KD) or ketamine/propofol (KP)	Dose: • 0.25 mg/kg/h• Duration:• Maximum 9 h	Primary outcome: • Fentanyl doseSecondary outcomes: • Ramsay sedation scale• Cardiovascular parameters• Duration of MV• SAEs/ADRs	• No significant ketamine related SAEs/ADRs in either group• Significant reduction in fentanyl dose• Reduced time to extubation in KD group vs KP group
6. Perbet S (2018)	France	ICU (unspecified)	Prospective Randomised, double-blind, controlled trial	80 (+82 control)	Remifentanil + propofol or midazolam + Adjunctive Ketamine (vs Placebo)	Dose: • 0.20 mg/kg/hDuration: • 8 days (3–16)	Primary outcome: • Total daily opiate consumptionSecondary outcomes: • Concomitant analgesic/sedative infusion• Delirium incidence and duration• Mortality• Duration of MV• ICU LoS	• No significant difference in opiate consumption (p = 0.548)• Reduced delirium incidence in ketamine group (21%vs 37%; p = 0.03) and shorter duration of delirium (3vs 5.3 days; p0.005)• Delirium incidence not significantly different between concomitant sedative choice (propofol vs midazolam; p = 0.93)• No difference in duration of MV, ICU LoS, or mortality.
7. Raj, A (2022)	India	Obstetric patients	Prospective randomised controlled study	33 + 34	Ketamine/dexmedetomidine (KD) vs ketamine/propofol (KP)	Dose: • Ketamine dose in KD + KP groups: 1mg/kg bolus + 0.25 mg/kg/h infusionDuration: • Maximum 12 h	Primary outcome: • Ramsay sedation scale• Secondary outcomes:• Cardiovascular parameters• ECG changes/arrhythmias• Pain scores• ICU length of stay• SAEs/ADRs including hyper-salivation and agitation	• No significant difference in sedation scores (p = 0.284)• KD group had significantly higher MAP at time points 30 m, 4, 6, 8, 10 h (p < 0.05)• Lower incidence of ADRs in KD group (not significant)• Higher additional analgesia requirements in KP group (p < 0.05)• Longer mean ICU LoS in KP group: 22.91 ± 4.03vs 30.44 ± 7.26 (p < 0.001)
8. Reese JM (2018)	USA	Septic shock	Prospective, non-randomised, open label, pilot study with retrospective cohort (control)	17 (+29 control)	Ketamine primary sedation (48 h) (vs retrospective cohort – standard care)	Dose: • 0.3 mg/kg/hDuration: • Maximum 96 h	Primary outcome: • Vasopressor requirements at 24, 48, 72, and 96 hSecondary outcomes: • Ketamine dose• Corticosteroid use• Concomitant analgesic/sedative infusion• Duration of MV• ICU and hospital LoS• Mortality	• Non-significant trend towards decreased vasopressor dose• Significantly fewer benzodiazepine opiate used in ketamine group (p = 0.015)• Significantly less opiate consumption in ketamine group (p < 0.001)• No significant differences in other secondary outcomes
9. Schmittner MD (2007)	Germany	Neuro ICU (SAH/TBI patients)	Prospective randomised pilot study	12 (+12 control)	Methohexitone/ketamine primary sedation (vs methohexitone/fentanyl)	Dose: • Maximum 2 mg/kg/hDuration: • 5 days	Primary outcome: • Cerebral haemodynamics (Intracranial pressure (ICP) and cerebral perfusion pressure (CPP))Secondary outcomes: • Bispectral index (BIS)• Cardiovascular parameters• Vasopressor dose• Gastrointestinal motility	• No significant changes in ICP between groups• Lower vasopressor dose in ketamine group (3.6 ± 5.1vs 12.8 ± 18.4 mcg/kg/h; not significant due to sample size)• No difference in gastrointestinal motility• No difference in BIS
Prospective Cohort/Observational Studies
10. Suleiman, A (2022)	USA	Surgical ICU	Prospective interventional (cohort) study	12	Low-dose continuous ketamine infusion (no comparator)	Dose: • Ketamine 0.3 mg/kg/h for 1 h + 0.mg/kg/h for 1 h	Primary outcome: • Inspiratory airflowSecondary outcomes: • Beta-gamma electroencephalogram power change• Respiratory/ventilatory parameters including work of breathing, airway resistance, lung compliance• Haemodynamic parameters	• Ketamine (5 mcg/kg/min and 10 mcg/kg/min) increased inspiratory airflow from 0.36 l/s to 0.47 l/s (p = 0.017) and 0.44 l/s (p = 0.01)• Ketamine increased beta-gamma electroencephalogram power compared to baseline (p < 0.01)• Low and high-dose ketamine increased minute ventilation (p = 0.022 + p=0.017) and tidal volume (p = 0.008 + p=0.012)• Work of breathing improved with ketamine 10 mcg/kg/min (p = 0.009)• No significant effects on HR or BP
Retrospective Cohort/Observational Studies
11. Atchley E (2021)	USA	Medical ICU	Retrospective cohort control	78 (+78 control)	Ketamine vs propofol or dexmedetomidine	Dose: • Maximum: Ketamine 2.4mg/kg/h Propofol 30 mg/kg/h Dexmedetomidine 0.4 mcg/kg/hDuration: • .Ketamine 67.7 h• Propofol/dexmedetomidine 51.0 h	Primary outcome: • Clinically significant negative haemodynamic events (hypotension or bradycardia)Secondary outcomes: • ICU LoS• Duration of MV• Vasopressor requirements• Mortality <12h after cessation of sedative	• Lower incidence of bradycardia (1.3%vs 14.1%; p = 0.001) and hypotension (24.4%vs 39.7%; p = 0.02) in ketamine group• Multivariate analysis found propofol and dexmedetomidine to be only factors associated with negative haemodynamic events• Higher vasopressor requirement prior to sedation in ketamine group (p = 0.001), and higher vasopressor requirement during sedation in ketamine group (p = 0.003)• Ketamine group experienced smaller percent change in MAP (25.3%vs 33.8%; p < 0.001) and absolute change in systolic BP (26.2vs 42.0 mmHg; p < 0.001)• Ketamine group had a longer duration of MV (247.7vs 137.3 h; p < 0.001) and longer ICU stay (14vs 10 days; p = 0.01)• There was higher mortality within 12 h of stopping ketamine compared to propofol or dexmedetomidine (25.6%vs 6.4%; p < 0.001)
12. Buchheit JL (2019)	USA	Surgical ICU	Retrospective cohort	40	Propofol/Opioid + Adjunctive low-dose ketamine (no comparator)	Dose: • Median: 0.3 mg/kg/h• Range 0.21–0.3 mg/kg/hDuration: • Median: 1.89 day• Range: 0.96–3.06 days	Primary outcome: • Slope of change of morphine equivalents 12 h post-initiation of ketamineSecondary outcomes: • Rate of concomitant sedatives• RASS pre- and post-ketamine• Critical Care Observation of Motor Activity (COMA) score• Delirium (CAM-ICU) score• Vasopressor dose• Cardiovascular parameters• SAEs/ADRs including hypersalivation	• Significant opioid dose reduction at 24 h from 6.66 mg/h to 0 mg/h (p < 0.001)• Significant reduction in opioids at time points 1 h and 6 h (p = 0.004)• Significant reduction in propofol at time points 1 h and 6 h (p = 0.014)• No significant difference in RASS (p = 0.476)• Significant reduction in vasopressor requirements (p = 0.019)• No differences in BP, HR, or respiratory rate• No significant ADRs including hypersalivation or psychomimetic events
13. Flinspach AN (2021)	Germany	Medical ICU (Covid-19/ARDS)	Retrospective monocentric analysis	8 (+48 control)	Midazolam/esketamine ± clonidine (vs any other combinations of sedatives used)	Dose: • Mean 0.86 ±0.75mg/kg/hDuration: • Not reported	Primary outcome: • Sedation combinations used during Covid-19Secondary outcomes: • Prone positioning• Sedation depth• Mortality	• Unusual high dosages of ketamine in mechanically ventilated patients with COVID-19 compared with literature
14. Garber PM (2019)	USA	ICU (Mixed)	Retrospective 2 centre observational study	104	Fentanyl ± propofol ± dexmedetomidine ± midazolam + Adjunctive ketamine (no comparator)	Dose: • Maximum median: 0.42 mg/kg/h (0.3–4.6 mg/kg/h)Duration: • Median: 90.5 h	Primary outcome: • Concomitant analgesic/sedative infusion dose 24 h post-ketamine initiationSecondary outcomes: • RASS• Vasopressor dose• SAEs/ADRs• Delirium (CAM-ICU)• ICU LoS• Hospital LoS• Mortality	• Reduction in concomitant analgesic/sedative agents at 24 h (−20% (-63.3% − 0.0%); p = 0.001)• Improved time spent within goal sedation range (7.1% pre-ketamine to 25% post-ketamine; p = 0.005)• No significant change in incidence of delirium (77.2% pre-ketamine vs 82.4% post-ketamine; p = 0.0641)• Reduction or cessation of vasopressors at 24 h
15. Garner O (2021)	USA	Medical ICU (Covid-19)	Retrospective observational study	59	Propofol or midazolam or dexmedetomidine or fentanyl + Adjunctive ketamine	Dose: • Median cumulative dose at 72 h 18–31 mg/kgDuration: • Maximum 72 h	Primary outcome: • Concomitant analgesic/sedative infusion dose post-ketamine initiationSecondary outcomes: • Use of paralysing agents• Vasopressor dose	• Reduced requirements of propofol at 24 h post-ketamine (34.2vs 54 mg/kg; p = 0.003)• No significant change to midazolam (p = 0.763), dexmedetomidine (p = 0.764), or fentanyl (p = 0.136)• Increased hydromorphone at all time points following ketamine introduction (p < 0.001)• Increase in cisatracurium (p = 0.009) but no increase in rocuronium• Reduced noradrenaline requirements at 24 h post-ketamine (38vs 62.8 mcg/kg; p = 0.028)
16. Groetzinger LM (2018)	USA	ICU (Mixed)	Retrospective observational study	91	Propofol or benzodiazepine or fentanyl or dexmedetomidine + ketamine (no comparator)	Dose: • Median 0.41 mg/kg/h (0.04–2.5 mg/kg/h)Duration: • Median 2.8 days	Primary outcome: • Concomitant analgesic/sedative infusion dose post-ketamine initiationSecondary outcomes: • Riker Sedation Agitation Scale (SAS)• Intensive Care Delirium Screening Checklist (ICDSC)• Cardiovascular parameters• Respiratory parameters• Use of additional sedatives/antispychotics• Duration of MV• ICU LoS• SAEs/ADRs	• 63% reduced or discontinued one or more pre-existing sedatives with 24 h of initiating ketamine• Ketamine associated with increased percentage of SAS scores in range compared to 24 h prior (61%vs 55%; p = 0.001)• No significant changes to haemodynamics• No increase in delirium incidence or antipsychotic use• No significant increase in ADRs
17. Jaeger M (2020)	USA	Medical ICU	Retrospective single centre cohort-control study	86 (+86 control)	Propofol or benzodiazepine or fentanyl or dexmedetomidine + Adjunctive Ketamine (vs any non-ketamine sedation)	Dose: • Median rate: 0.47 mg/kg/h (0.22–0.93 mg/kg/h)Duration: • Median 2.25 days (1.6–3.6 days)	Primary outcome: • Percentage of Richmond Agitation-Sedation Scale (RASS) scores at goalSecondary outcomes: • Pain scores• Vasopressor dose• Concomitant analgesic/sedative infusion• Use of additional analgosedation• Haemodynamics• ICU LoS• In-hospital mortality• SAEs/ADRs	• No significant difference in percentage of RASS scores at goal (78.7%vs 81.4%; p = 0.29)• No significant difference in pain scores (p = 0.53)• Significant improvement in MAP (93vs 100 mmHg; p = 0.005)• Ketamine associated with reduced noradrenaline dose (p = 0.03)• No significant difference in concomitant sedative doses• Significantly fewer intermittent benzodiazepines required in ketamine group (p < 0.0001)• Ketamine group had a significantly longer ICU LoS (8.8vs 5.2 days; p < 0.001)• No increase in delirium (p = 0.27)• No significant SAEs/ADRs
18. Pruskowski KA (2017)	USA	Trauma patients	Single-centre retrospective cohort study	36	Adjunctive ketamine (no comparator)	Dose: • 0.64–0.94 mg/kg/hDuration: • Maximum 72 h	Primary outcome: • Concomitant analgesic/sedative infusion dose post-ketamine initiationSecondary outcomes: • RASS• Antipsychotic use	• No significant difference in time spent in target RASS (p = 0.416)• Decreased opioid use following ketamine initiation (431.3 mg/72 h vs 272.5 mg/72 h; p = 0.026)• Significant reduction in propofol following ketamine initiation (35vs 22.8 mcg/kg/min; p = 0.002)• No significant difference in benzodiazepine doses (p = 0.735)• Significant increase in dexmedetomidine dose following ketamine initiation (0.7vs 0.9 mcg/kg/h; p = 0.002)• Significant increase in ziprasidone use post-ketamine initiation (p = 0.018)
19. Shurtleff V (2018)	USA	ICU (unspecified)	Retrospective cohort	39 (+40 control)	Standard care (unspecified) +Adjunctive ketamine (vs standard care)	Dose: • Median: 0.42 ±0.12mg/kg/hDuration: • Median: 43h (17–91 h)	Primary outcome: • Number of days without delirium (CAM-ICU)Secondary outcomes:• Delirium incidence/duration• RASS• Duration of MV• ICU LoS• Hospital LoS• Mortality	• No significant difference in delirium incidence (p = 0.274) or duration (p = 0.351)• Higher percentage pain scores at goal (p = 0.044)• Lower median percentage RASS scores at goal (p = 0.019)• Higher incidence of coma (16vs 6; p = 0.013), however no significant difference in coma-free days (8 vs 7; p = 0.511)• Significantly lower propofol requirements in ketamine group (p < 0.001)
20. Umunna BP (2015)	USA	ICU (unspecified)	Single-centre retrospective cohort study	30	Ketamine primary sedation (no comparator)	Dose: • Median: 2 mg/kg/h• Range: 0.5–4 mg/kg/hDuration: • Median: 59.6 h	Primary outcome: • Incidence of adverse eventsSecondary outcomes: • Duration of MV• ICU LoS• Motor Activity Assessment Score (MAAS)• Vasopressor requirements• Cardiovascular parameters• Concomitant analgesic/sedative infusion• SAEs/ADRs	• 6.7% incidence rate of tachyarrhythmias• 6.7% incidence rate of agitation• No significant change in vasopressor requirements• No incidences of hypersalivation• Frequency of adverse events similar to more common sedative agents (13%)• Provided effective sedation (average MAAS 1.9)
21. Von der Brelie C (2017)	Germany	Neuro ICU (SAH patients)	Retrospective observational study	41 (+24 control)	Midazolam/fentanyl + Adjunctive ketamine (vs Midazolam/Fentanyl)	Dose: • Not specifiedDuration: • Mean: 18.4 days• Range: 4–32 days	Primary outcome: • Incidence of adverse eventsSecondary outcomes: • Duration of MV• Duration of sedation• Cerebral infarctions• ICP• Mortality• RASS• Vasopressor dose	• No significant increase in ICP• Not associated with a higher rate of complications• Lower rate of cerebral infarctions• Reduction of vasopressor dose after ketamine initiation• 38 of 41 patients receiving ketamine (92.7%) achieved target RASS
Case Series/Reports
22. Achar MN (1993)	Kuwait	Refractory asthma	Case report	1	Propofol/midazolam + adjunctive ketamine	Dose: • 10–30 mg/hDuration: • 18 h	No primary/secondary outcomes listedResults/discussion focus on:• Sedation efficacy (not scored)• Respiratory parameters• Cardiovascular parameters• Arterial Gas values (PaCO2/PaO2/pH)	• Improved sedation• Improved gas exchange• Effective bronchodilator• Successful weaning from ventilator• Safe CVS profile in presence of acute myocardial injury
23. Benken ST (2017)	USA	Neuro ICU (head injury)	Case report	1	Propofol/Fentanyl/Midazolam + adjunctive ketamine	Dose: • Initiated at 1 mg/kg/h (maximum 5 mg/kg/h)Duration • 60 h	No primary/secondary outcomes listedResults/discussion focus on:• Use of concomitant sedatives• Management of agitation• CVS and respiratory stability	• Successful ventilator weaning, extubation, and transition of care
24. De Pinto M (2008)	USA	Trauma patient	Case report	1	Fentanyl/lorazepam/methadone + Adjunctive low-dose ketamine (no comparator)	Dose: • Dose range 0.036–0.078 mg/kg/hDuration: • 5 days	No primary/secondary outcomes listedResults/discussion focus on:• Use of concomitant sedatives• Management of pain in opioid tolerance• SAEs/ADRs	• May be helpful for the management of pain and sedation in critically ill patients• No unwanted psychomimetic ADRs
25. Hamimy W (2019)	Egypt	ICU (mixed)	Case series	14	Ketamine-propofol (ketofol 1:1 8mg/ml) sedation (no comparator)	Dose: • Median: 48 mg/h (38.4–60 mg/h)Duration: • Median: 16 h (range: 1–24 h)	Primary outcome: • Ramsay Sedation ScaleSecondary outcomes: • Cardiovascular parameters• ECG changes/arrhythmias• Vasopressor dose• Respiratory parameters• SAEs/ADRs	• Maintained sedation scores in range• No significant changes to HR or BP• No respiratory depression• No ADRs
26. Park GR (1987)	UK	Acute lymphatic leukaemia + pneumonia	Case report	1	Midazolam + Adjunctive ketamine (no comparator)	Dose: • Mean 1.18mg/kg/h (0.3–1.5 mg/kg/h)Duration: • 102.5 h	No primary/secondary outcomes listedResults/discussion focus on:• Bronchospasm /bronchodilation• Plasma concentrations of ketamine/metabolites• Sedation (not scored)• SAEs/ADRs including psychomimetic symptoms• Cardiovascular/respiratory parameters• Concomitant analgesic/sedative infusion	• Successful sedation from 5-day ketamine infusion• Increase in arterial pressure• Diminution of bronchospasm• No psychomimetic effects
27. Treu CN (2017)	USA	Opiate abuse MV patients	Case series	4	Adjunctive ketamine (no comparator)	• Dose: • 0.1–0.8 mg/kg/h• Duration:• 3 days	No primary/secondary outcomes listedResults/discussion focus on:• Use of concomitant analgesic/sedative infusion• Sedation score (SAS)• Cardiovascular parameters• SAEs/ADRs	• Useful to facilitate mechanical ventilation in patients with histories of opioid abuse• Increased oral secretions

Critical appraisal

Given the range of employed methodologies and apparent quality of evidence, sources were critically apprised using the relevant aspects of the CASP checklists (https://casp-uk.net/casp-tools-checklists/), which can be found in Table 2 in the appendix.

Table 2.

Critical appraisal of included studies.

Author (year)	Methods	Focused research question	Randomisation	Blinding	Baseline matching	Risk of bias	Generalisability	Comments
Achar MN (1993)	Case report	No	No	No	No	High	Low	• Single case report
Amer M (2021)	Active-controlled, pilot, feasibility clinical trial	Yes	Yes	Yes	Yes	Low	Good	• Randomised trial designed using the CONSORT guidelines• Blind to treatment assignment, however not blind to trial intervention once assigned to an arm, therefore not true double blind trial.• Feasibility/pilot study so not powered• Protocol available prior to study• Wide population• Ketamine as an adjunct in treatment group• Intervention only lasted 48 h then stopped• Relatively low adherence rate to protocol (75%)
Atchley E (2021)	Retrospective cohort	Yes	No	No	Yes	High	Medium	• Retrospective – high risk of bias and cofounders• Significant differences in time intervals for starting ketamine compared to other sedatives• Significant differences between cohorts at baseline e.g. coronary artery disease, and diagnosis (ARDS)• Reasonable sample sizes
Benken ST (2017)	Case report	No	No	No	No	High	Low	• Case report• A very specific case – Difficult to extrapolate to other ITU populations.
Bourgoin A (2003)	Prospective randomised double blind study	Yes	Yes	Yes		Low	Low	• Double blind• Only patients with severe head trauma with GCS less than 8 at risk of high ICP (low generalisability).
Buchheit JL (2019)	Retrospective cohort	Yes	No	No	No	High	Low	• Retrospective cohort study• Only surgical patients• Small sample size• Only included patients passing ‘spontaneous breathing trial’ safety check• Ketamine added to opioid infusion
Christ G (1997)	Prospective randomised study	Yes	Yes	Yes	Yes	Low	Low	• Prospective, randomised, controlled study• Well matched demographics• Small sample size• Group A had higher APACHE at baseline• Group A had lower LVEF at baseline
De Pinto M (2008)	Case report	No	No	No	No	High	Low	• Case report so has some limitation in generalisability/validity• A very specific case – Difficult to extrapolate to other ITU populations.
Flinspach AN (2021)	Retrospective monocentric analysis	Yes	No	No	No	High	Low	• Monocentric retrospective study.• In context of severe ARDS only – low generalisability.• Indication for triple sedative choice not clear
Garber PM (2019)	Retrospective 2 centre observational study	Yes	No	No	No	High	Low	• Recruitment open to broad range of patients – however most were male and surgical.• Retrospective – high risk of bias and cofounders• Outcome looked at change in dosage of other angiosedatives (propofol/fent/midaz/dexto) drug whilst on ketamine infusion. Only commented at 12 h intervals up to 72 h. Questions validity of administrated over e.g. 1 week.
Garner (2021)	Retrospective observational study	Yes	No	No	No	High	Low	• Recruited a specific subgroup of patients (Covid-19 ARDS)• Retrospective – high risk of bias and cofounders• Outcome looked at change in dosage of other angiosedatives• Only commented at 12 h intervals up to 72 h. Questions validity of administrated over e.g. 1 week.
Groetzinger LM (2018)	Retrospective observational study	Yes	No	No	No	High	Low	• Retrospective study – high risk of cofounders and bias• One centre study therefore low generalisability.• Unclear as to why ketamine started.
Gupta B (2022)	Prospective randomised controlled study	Yes	Yes	No	Yes	Low	Medium	• Randomised controlled trial• Unblinded• Un-validated sedation scores used• Surgical patients only
Hamimy W (2019)	Case series	No	No	No	No	High	Low	• Small sample (14) case series therefore limitation in generalisability/validity• No comparator group• Short duration of study period (<24h)
Jaeger M (2020)	Retrospective single centre cohort study	Yes	No	No	No	High	Medium	• Small sample size• Range of presentations (improving generalisability)• Retrospective study – high risk of bias and cofounders• Patients in he ketamine group actually had similar amounts of propofol.
Mogahd MM (2017)	Prosepctive randomised study	Yes	Yes	Unclear	Yes	High	Low	• Sample size of 35 patients.• Post CABG therefore low generalisability• Randomisation process unclear• Unclear if blinded.• Unclear if/how other variable were controlled for e.g. opiate analgesia.• Unclear duration of sedation.• No non-ketamine comparator
Park GR (1987)	Case report	No	No	No	No	High	Low	• Case report• Ketamine used for management in bronchospasm – Low generalisability.
Perbet S (2018)	Randomised, double-blind, controlled trial	Yes	Yes	Yes	Yes	Low	High	• Randomised, parallel-group, placebo controlled, double blind study• CONSORT checklist followed• Well matched groups• Ketamine as adjunctive – can’t draw conclusions about as primary agent• Single centre – no external validity• Infrequent CMA-ICU scores
Pruskowski KA (2017)	Single-centre retrospective cohort study	Yes	No	No	No	High	Low	• Single centre• Trauma patient demographic only.• Retrospective cohort study – unable to control for potentially cofounding factors• The trial authors admit that at the time benzodiazepines were used alongside ketamine continuous infusions.• No control
Raj A (2022)	Prospective randomised controlled trial	Yes	Yes	No	Yes	Low	Low	• Single centre• Randomised controlled study• No blinding• Limited to obstetric patients (low generalisability)
Reese JM (2018)	Retrospective cohort (control) and prospective, non-randomised, open label, pilot study	Yes	No	No	Yes	High	Medium	• Non-surgical patients only.• Significant difference in sample size between control and non-control groups.• The control group was a retrospective group (i.e. collected from paper charts – possibility of cofounding e.g. control group all may been have received different analgesics).• Absence of RAAS/sedation scores• Ketamine sedation only lasted 48 h
Schmittner MD (2007)	Prospective randomised pilot study	Yes	Yes	No	No	Low	Low	• No blinding after randomisation to either fentanyl or ketamine group.• Not clear how other potentially cofounding variables were controlled.• Unclear which population the patients presented from e,g. are they the same demographic/socioeconomic standing.• Seemingly no consideration for other clinical history of the patients or factors that influence their angiosedative requirements.
Shurtleff V (2018)	Retrospective cohort	Yes	No	No	No	High	Low	• Retrospective study limited by the inability to control potentially cofounding variables.
Suleiman A (2022)	Prospective interventional cohort study	Yes	No	No	No	High	Low	• Prospective cohort with small sample size (12)• Short duration of ketamine infusion (2 h) – cannot extrapolate to longer durations• Surgical patients only (poor generalisability)
Treu CN (2017)	Case series	No	No	No	No	High	Low	• Case series – very poor quality evidence• Only 4 patients• Specific patient group – opioid use disorders• Heavily sedated/severely agitated patients prior to initiation of ketamine
Umunna BP (2015)	Single-centre retrospective cohort study	Yes	No	No	No	High	Low	• Small sample size• No control group• Retrospective so absence of ability to control variables e.g. other analgesia received• Used MAAS not RAAS• Adverse events updated from nursing progress notes, would nursing and medical notes have led to greater AE identification.• ‘These dates were chosen as they coincided with the peak shortage of traditional agents’ – no drugs available hence used ketamine. Any other different practices ongoing?
Von der Brelie C (2017)	Retrospective observational study	Yes	No	No	No	High	Low	• Small sample size and specific presentation (low generalisability)• Retrospective so absence of ability to control variables e.g. other analgesia received• Comparison to standard care

Risk of bias was considered to be high in 20 of the 27 studies (74.1%) commonly due to lack of focused research question and unsatisfactory study designs (retrospective cohorts, lack of blinding, and lack of control groups).

Characteristics

The included studies, country of origin, aims, designs, measures and outcomes, as well as main findings can be found in Table 1.

Of the 27 articles relating to the use of continuous ketamine infusions for sedation or analgesia during MV, there were nine prospective randomised studies^29
–37, one prospective cohort study,³⁸, 11 retrospective cohort^39
–44 or observational studies,^45
–49 and six case reports or case series (Figure 2).^50
–55

Figure 2.

Designs of included studies (%).

A total of 987 participants received continuous ketamine infusions whilst undergoing MV. There were 497 controls or comparators. Sample sizes were generally small with 12 studies using a sample size of less than 30 participants per arm (range N = 1–104).

Mechanically ventilated patients were included from the following ICUs: Medical ICU (40.7%),^{36,41,45,47,50,54,55} Mixed, general, or unspecified ICU (39.0%),^{29,34,39,43,44,46,48,53} Neurosurgical ICU (8.2%),^30,37,49,51 Cardiac ICU (6.7%),^31,33 Surgical ICU (2.8%),^32,38,40 Obstetric ICU,³⁵ and Trauma ICU (2.6%).^42,52

The most frequent country of origin was the USA (55.5%), other countries included France, Germany, India, and Saudi Arabia, with only one case report coming from the United Kingdom.⁵⁴ The majority of studies (77.9%) were published between 2017 and 2022.

Intervention and comparators

Of the 27 studies included, 11 (40.7%) investigated the use of ketamine as a primary sedative,^{30
–33,35
–37,39,41,44,53} the remainder used ketamine as an adjunct to pre-existing sedative or analgesic therapies. Eleven (40.7%) directly compared ketamine primary sedation or adjunctive ketamine to non-ketamine regimes.^{29
–31,33,36,37,39,41,43,45,49} One study compared the addition of ketamine to placebo.³⁴

Doses and durations

Ketamine was administered as a continuous infusion in all studies. Doses across included studies (Table 1) varied greatly with a minimum reported infusion dose of 0.036mg/kg/h,⁵² and a maximum of 6.42 mg/kg/h.³⁰ The majority of studies used a dose range under 1 mg/kg/h with 12 authors reporting an average dose or dose range under 0.5 mg/kg/h. Seven studies used a range between 0.5 and 1 mg/kg/h, three used between 1 and 2 mg/kg/h, and four authors reported doses over 2 mg/kg/h. Only one author did not report the doses of ketamine.⁴⁹

Duration of continuous ketamine infusions also varied significantly across studies (Table 1), with some study designs stipulating a maximum duration, and others continuing until cessation of MV (or death of patient). The minimum reported duration was 2h³⁸ and the maximum was 36 days (864 h).⁴⁹ Six studies reported durations of infusions under 24 h, nine reported durations between 24 and 72 h, and ten studies reported durations over 72 h.

Reported outcomes and findings

The primary and secondary outcome measures, and significant findings relating to the review question are reported in Table 1 in the appendix.

The most commonly reported primary outcome was the effect of ketamine on dose or rate of other concomitant sedatives or analgesics (25.9%).^{33,34,40,42,46
–48} Other primary outcomes included sedation efficacy,^32,35,41,53 safety of ketamine infusions,^44,49 effects on neurological parameters e.g. intra-cerebral pressure,^30,37 cardiovascular effects for example, blood pressure and vasopressor requirements,^31,36,39,47 respiratory parameters for example, gas exchange and airway pressures,^38,50 feasibility data,²⁹ and effect on delirium.⁴³ Commonly reported secondary outcomes included cardiovascular parameters, vasopressor requirements, ICU length of stay, ventilator free days/duration of MV, adverse events, and mortality. None of the studies reported long-term patient-centred outcomes such as quality of life nor the incidence, severity, or duration of depression or PTSD.

Concomitant infusions

Seven of 27 studies (25.9%) investigated the effect of initiating ketamine infusions on the rate of concomitant analgesic and sedatives,^{33,34,40,42,46
–48} this was reported as secondary outcome in a further 12 studies (44.4%).

Initiation of ketamine was associated with a decrease in propofol doses in six studies^{40,42,43,46
–48} and a decrease in alternative sedatives (e.g. dexmedetomidine) in two studies.^46,48 Benzodiazepine requirements (intermittent or continuous) were reduced in two studies^36,41, opioid requirements were also significantly lower in three studies.^33,36,40 However, in some studies, there was no difference reported in the dose of concomitant sedatives⁴⁷ or opioid requirements.^34,47 One study reported an increase in dexmedetomidine doses following initiation of ketamine⁴² and another reported an increase in opioid requirements.⁴⁷ These inconsistencies were sometimes present even within studies. Pruskowski et al,⁴² for example, found that ketamine infusions led to a significant reduction in propofol (35vs 22.8 mcg/kg/min; p = 0.002), but also a significant increase in dexmedetomidine dose (0.7vs 0.9 mcg/kg/h; p = 0.002).

Sedation efficacy

Only four studies (14.8%) investigated the efficacy of sedation as a primary outcome,^32,35,41,53 however assessment of sedation using a validated scoring system (e.g. Richmond Agitation Sedation Scale (RASS) or Riker Sedation-Agitation Scale (SAS)) were reported as secondary outcomes in 11 additional studies (40.7%).

Three authors reported sedation scores when comparing a ketamine-based sedation to a non-ketamine sedation regimen. Amer et al²⁹ reported more target RASS scores at 24 h (67.5%vs 52.4%) and 48 h (73.5%vs 66.7%) when comparing ketamine and fentanyl to ‘standard care’. Gupta et al³² found sedation scores were higher in the ketamine-dexmedetomidine group compared to dexmedetomidine alone (p < 0.05), though used an un-validated score. Gupta et al also found patients reported worse pain scores in the dexmedetomidine group at 2 h and 4 h (p < 0.05). Jaeger et al⁴¹ however found no significant difference in the percentage of RASS scores within target range (78.7%vs 81.4%; p = 0.29) or in pain scores (p = 0.53) when comparing adjunctive ketamine to any other sedatives.

Additional studies reported an improvement in the time study participants spent within the RASS target range (7.1% pre-ketamine to 25% post-ketamine; p = 0.005).⁴⁶ An increase in the number of SAS sedation scores within target range was also observed after ketamine initiation when compared to 24 h period beforehand (61%vs 55%; p = 0.001).⁴⁸ Several studies reported that patients who received ketamine experienced adequate or appropriate sedation.^44,49
–55 Four studies reported no difference in sedation scores or bispectral index.^35,37,40,42

Only one study, conducted by Shurtleff et al,⁴³ reported lower median percentage RASS scores at target with adjunctive ketamine compared to standard care alone (70%vs 84%; p = 0.019). A higher incidence of coma was also observed in the ketamine group (16 vs 6; p = 0.013). Despite this finding there was no significant difference in coma-free days between groups (8 vs 7; p = 0.511), and those who received ketamine had a higher percentage of pain scores at target (99%vs 91%; p = 0.044).

Cardiovascular outcomes

Many investigators reported on the effect of ketamine on haemodynamics for example, mean arterial pressure (MAP), blood pressure (BP), or heart rate, as well as vasopressor requirements.

Several authors reported improved haemodynamics, including Christ et al³¹ who demonstrated a significant increase in MAP at 24 h (68 ±11mmHg to 77 ±14mmHg; p = 0.01), Raj et al³⁵ who found significantly higher MAP (p < 0.05), and Gupta et al³² who demonstrated higher systolic BP (p < 0.05) and heart rates (p < 0.05) when compared to dexmedetomidine. Atchley et al³⁹ found their ketamine cohort had a smaller percentage change in MAP (25.3%vs 33.8%; p < 0.001) and absolute change in systolic BP (26.2vs 42.0 mmHg; p < 0.001), whilst experiencing a lower incidence of bradycardia (1.3%vs 14.1%; p = 0.001) and hypotension (24.4%vs 39.7%; p = 0.02). Jaeger et al⁴¹ also reported significant improvements in MAP (93vs 100 mmHg; p = 0.005) and reduced noradrenaline dose (p = 0.03). Other authors similarly reported reduced vasopressor requirements.^{36,37,40,46,47,49}

These results were inconsistent, however, with some investigators such as Amer et al,²⁹ Groetzinger et al,⁴⁸ and Umunna et al⁴⁴ reporting no significant differences in vasopressor requirements or haemodynamics. Whilst Bourgoin et al³⁰ reported no significant difference in MAP, they found significantly less fluid was administered in the ketamine group (429 ± 405ml vs 992 ± 703ml; p = 0.02).

Respiratory outcomes

Suleiman et al³⁸ demonstrated improved respiratory dynamics including inspiratory airflow, increased minute ventilation, and tidal volumes, as well as improved work of breathing. Achar et al⁵⁰ reported improved gas exchange and bronchodilation following the addition of ketamine for refractory asthma.

Neurological outcomes

Three authors investigated effects of ketamine on intracranial pressure. All found no significant increases, changes, or variations in ICP^30,37,49 and Suleiman et al³⁸ reported Ketamine increased beta-gamma electroencephalogram power compared to baseline (p < 0.01).

Delirium

Perbert et al³⁴ found a reduced incidence of delirium in their ketamine cohort (21%vs 37%; p = 0.03) and a shorter duration of delirium (3vs 5.3 days; p0.005). Delirium incidence was not significantly different between concomitant sedative choice (propofol vs midazolam; p = 0.93). Garber et al⁴⁶ reported no significant change in incidence of delirium following initiation of ketamine (77.2% pre-ketamine vs 82.4% post-ketamine; p = 0.0641) and Jaeger et al⁴¹ found no difference in incidence of delirium between ketamine and non-ketamine groups (48% in the ketamine group vs 56% in the non-ketamine group; p = 0.27). Shurtleff et al⁴³ also reported no difference between ketamine and non-ketamine groups (74% in the ketamine group vs 85% in the non-ketamine group; p = 0.274).

Safety and adverse events

Over half of authors (51.9%) commented on the safety of ketamine; 85.7% of these authors reported no significant adverse reactions or events^{29,32,33,35,40,41,44,46,48,52
–54} and the absence of emergence or psychomimetic reactions was specifically reported in four studies.^29,40,52,54

Treu et al⁵⁵ reported increased oral secretions, and Umunna et al⁴⁴ reported a 6.7% incidence of tachyarrhythmia but with an overall incidence of adverse events similar to more commonly used sedatives. Atchley et al³⁹ reported a longer duration of MV (247.7vs 137.3 h; p < 0.001) and longer ICU stay (14vs 10 days; p = 0.01). Jaeger et al⁴¹ also found that ketamine infusion was associated with a significantly longer ICU length of stay (8.8vs 5.2 days; p < 0.001).

Discussion

Summary

In this thorough and comprehensive review of the literature we report 27 publications relating to the use of continuous ketamine infusions as an analgesic or sedative for mechanically ventilated patients on ICU.

There was a paucity of high-quality evidence with very few well-designed prospective studies that use patient-centred primary endpoints with only seven studies (25.9%) prospectively comparing ketamine as a primary sedative to alternative regimens.

We were unable to establish any studies that investigated the role of ketamine on the incidence, severity, or duration of depression or PTSD following ICU admission, indicating a potential target for future well-designed prospective clinical trials.

However, despite the lack of well-designed, well-powered studies, this review did provide reassurance around the safety of ketamine sedation with low numbers of ADRs or SAEs reported. Reports of reduced incidence of delirium with no reports of emergence reactions go some way to alleviate apprehensions around unwanted psychomimetic phenomena.

The low incidence of adverse effects was not restricted to particularly low-dose or short-duration regimes, with doses and durations of ketamine ranging significantly across studies.

The reported findings suggest a range of potential patient benefits, including: improved sedation and pain scores, reduced concomitant sedative infusions, reduced opiate requirement, and haemodynamic stability. However these results must be interpreted with caution given the poor quality of included studies and small sample sizes. Additionally, the retrospective nature of many of the studies increases the risk of recall bias, inclusion bias, and reporting bias.

Whilst this was not a dedicated review of the evidence surrounding the relationship between ketamine and ICP, there were three studies that reported the effect of ketamine on intracranial pressure. Interestingly, none of the studies reported a significant increase, change, or variation in ICP. This may help allay the concerns of clinicians with regard to the safety of ketamine in patients with raised ICP or brain injury, which is still regarded as a contraindication to ketamine use.

Questions remain with regard to the overall generalisability of the results presented in this scoping review as many of the studies were conducted in specific patient populations for example, post-cardiac surgery or traumatic brain injury.

Conclusions

This review highlights the lack of robust, high-quality evidence for the use of continuous ketamine infusions as a primary sedation on ICU for patients undergoing MV, however the data available indicates that continuous ketamine infusions on ICU appear safe and well tolerated and may convey some benefits to patients at the doses and durations investigated.

The multiple gaps in the literature identified in this review need to be addressed through a large, prospective randomised controlled trial comparing ketamine as primary sedation to a well-matched control group that incorporates patient centred outcomes and endpoints. Currently no such trials are registered on clinicaltrials.gov.

Limitations

This scoping review was developed using the PRISMA-ScR checklist, and a protocol was publicaly available prior to commencing the review.

This review was limited by the studies available: 10 studies were excluded for not being available in either full-text form or English, which may have potentially excluded some important articles.

Footnotes

Appendices

Appendix A.

Data extraction tool.

Scoping Review Details
Title
Objectives
Question
Population
Concept
Context
Evidence Source
Citation
Author(s)
Year of Publication
Origin/country of origin
Aims/purpose
Population and sample size
Methodology /methods
Intervention type/comparator
Outcomes
Key findings relating to scoping question

Appendix B – Search Strategy and Terms

Preliminary PubMed search terms: “(Ketamine) AND (Intensive Care) AND ((Sedation) OR (Analgesia))”

Subsequent searches of PubMed, Embase, Ovid, Scopus, and Web of Science using words / phrases and index terms from papers retrieved during initial search.

Additional sources were identified through searching reference lists of identified reports and review articles.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Nicholas David Richards

References

Adhikari

Fowler

Bhagwanjee

, et al. Critical care and the global burden of critical illness in adults. Lancet 2010; 376: 1339–1346.

Wunsch

Linde-Zwirble

Angus

, et al. The epidemiology of mechanical ventilation use in the United States. Crit Care Med 2010; 38: 1947–1953.

Barr

Fraser

Puntillo

, et al. Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med 2013; 41: 263–306.

Jackson

Proudfoot

Cann

, et al. A systematic review of the impact of sedation practice in the ICU on resource use, costs and patient safety. Crit Care 2010; 14: R59.

Aitken

Kydonaki

Blackwood

, et al. Inconsistent relationship between depth of sedation and intensive care outcome: systematic review and meta-analysis. Thorax 2021; 76: 1089–1098.

James Lind Allicance. Intensive Care Top 10 | James Lind Alliance 2020, https://http://www.jla.nihr.ac.uk/priority-setting-partnerships/intensive-care/top-10-priorities/ (accessed 01 September 2020).

Richards-Belle

Canter

Power

, et al. National survey and point prevalence study of sedation practice in UK critical care. Crit Care 2016; 20: 355.

Devlin

Mallow-Corbett

Riker

. Adverse drug events associated with the use of analgesics, sedatives, and antipsychotics in the intensive care unit. Crit Care Med 2010; 38: S231–S43.

Reade

Finfer

. Sedation and delirium in the intensive care unit. N Engl J Med 2014; 370: 444–454.

10.

Salluh

Wang

Schneider

, et al. Outcome of delirium in critically ill patients: systematic review and meta-analysis. BMJ 2015; 350: h2538.

11.

Kurdi

Theerth

Deva

. Ketamine: current applications in anesthesia, pain, and critical care. Anesth Essays Res 2014; 8: 283–290.

12.

de Rocquigny

Dubecq

Martinez

, et al. Use of ketamine for prehospital pain control on the battlefield: A systematic review. J Trauma Acute Care Surg 2020; 88: 180–185.

13.

Sibley

Mackenzie

Bawden

, et al. A prospective review of the use of ketamine to facilitate endotracheal intubation in the helicopter emergency medical services (HEMS) setting. Emerg Med J 2011; 28: 521–525.

14.

Rosenbaum

Palacios

. Ketamine [Text]: Treasure Island (FL): StatPearls Publishing, https://http://www.ncbi.nlm.nih.gov/pubmed/ (2019, accessed 01 February 2021).

15.

Morris

Perris

Klein

, et al. Anaesthesia in haemodynamically compromised emergency patients: does ketamine represent the best choice of induction agent? Anaesthesia 2009; 64: 532–539.

16.

Goyal

Agrawal

. Ketamine in status asthmaticus: A review. Indian J Crit Care Med 2013; 17: 154–161.

17.

Yassin

Terblanche

Yassin

, et al. A web-based survey of United Kingdom sedation practice in the intensive care unit. J Crit Care 2015; 30: 1–6.

18.

Pai

Heining

. Ketamine. Contin Educ Anaes Crit Care Pain 2007; 7: 59–63.

19.

Strayer

Nelson

. Adverse events associated with ketamine for procedural sedation in adults. Am J Emerg Med 2008; 26: 985–1028.

20.

McGirr

Berlim

Bond

, et al. A systematic review and meta-analysis of randomized, double-blind, placebo-controlled trials of ketamine in the rapid treatment of major depressive episodes. Psychol Med 2015; 45: 693–704.

21.

Corriger

Pickering

. Ketamine and depression: a narrative review. Drug Des Dev Ther 2019; 13: 3051–3067.

22.

Davydow

Desai

Needham

, et al. Psychiatric morbidity in survivors of the acute respiratory distress syndrome: a systematic review. Psychol Med 2008; 70: 512–519.

23.

Wintermann

G-B

Rosendahl

Weidner

, et al. Predictors of major depressive disorder following intensive care of chronically critically ill patients. Crit Care Res Pract 2018; 1: e1586736.

24.

Davydow

Gifford

Desai

, et al. Posttraumatic stress disorder in general intensive care unit survivors: a systematic review. Gen Hosp Psychiatry 2008; 30: 421–434.

25.

Wade

Howell

Weinman

, et al. Investigating risk factors for psychological morbidity three months after intensive care: a prospective cohort study. Crit Care 2012; 16: R192.

26.

Jackson

Pandharipande

Girard

, et al. Depression, post-traumatic stress disorder, and functional disability in survivors of critical illness in the BRAIN-ICU study: a longitudinal cohort study. Lancet Respir Med 2014; 2: 369–379.

27.

Tricco

Lillie

Zarin

, et al. PRISMA extension for scoping reviews (PRISMA-ScR): Checklist and Explanation. Ann Intern Med 2018; 169: 467–473.

28.

JBI. JBI Maunual for Evidence Synthesis. JBI, https://synthesismanual.jbi.global/ (2020, 1 July 2022).

29.

Amer

Maghrabi

Bawazeer

, et al. Adjunctive ketamine for sedation in critically ill mechanically ventilated patients: an active-controlled, pilot, feasibility clinical trial. J Intensive Care 2021; 9: 54.

30.

Bourgoin

Albanèse

Wereszczynski

, et al. Safety of sedation with ketamine in severe head injury patients: comparison with sufentanil. Crit Care Med 2003; 31: 711–717.

31.

Christ

Mundigler

Merhaut

, et al. Adverse cardiovascular effects of ketamine infusion in patients with catecholamine-dependent heart failure. Anaesth Intensive Care 1997; 25: 255–259.

32.

Gupta

Mhaske

Pai

, et al. A comparative study of sedo-analgesic effect of dexmedetomidine and dexmedetomidine with ketamine in postoperative mechanically ventilated patients. J Anaesthesiol Clin Pharmacol 2022; 38: 68–72.

33.

Mogahd

Mahran

Elbaradi

. Safety and efficacy of ketamine-dexmedetomidine versus ketamine-propofol combinations for sedation in patients after coronary artery bypass graft surgery. Ann Card Anaesth 2017; 20: 182–187.

34.

Perbet

Verdonk

Godet

, et al. Low doses of ketamine reduce delirium but not opiate consumption in mechanically ventilated and sedated ICU patients: A randomised double-blind control trial. Anaesth Crit Care Pain Med 2018; 37: 589–595.

35.

Raj

Singh

Tiwari

, et al. Safety and efficacy of ketamine-dexmedetomidine versus ketamine-propofol combination for short-term sedation in postoperative obstetric patients on mechanical ventilation: A randomised clinical trial. J Clin Diagn Res 2022; 16: 1–4.

36.

Reese

Sullivan

Boyer

, et al. A Non-Comparative prospective pilot study of ketamine for sedation in adult septic shock. Mil Med 2018; 183: e409–e13.

37.

Schmittner

Vajkoczy

Horn

, et al. Effects of fentanyl and S(+)-ketamine on cerebral hemodynamics, gastrointestinal motility, and need of vasopressors in patients with intracranial pathologies: a pilot study. J Neurosurg Anesthesiol 2007; 19: 257–262.

38.

Suleiman

Santer

Munoz-Acuna

, et al. Effects of ketamine infusion on breathing and encephalography in spontaneously breathing ICU patients. J Intensive Care Med 2023; 38: 299–306.

39.

Atchley

Tesoro

Meyer

, et al. Hemodynamic effects of ketamine compared with propofol or dexmedetomidine as continuous ICU sedation. Ann Pharmacother 2022; 56: 764–772.

40.

Buchheit

Yeh

Eikermann

, et al. Impact of Low-Dose ketamine on the usage of continuous opioid infusion for the treatment of pain in adult mechanically ventilated patients in Surgical Intensive Care Units. J Intensive Care Med 2019; 34: 646–651.

41.

Jaeger

Attridge

Neff

, et al. Safety and effectiveness of sedation with adjunctive ketamine versus nonketamine sedation in the Medical Intensive Care Unit. J Pharm Pract 2021; 34: 850–856.

42.

Pruskowski

Harbourt

Pajoumand

, et al. Impact of ketamine use on adjunctive analgesic and sedative medications in critically ill trauma patients. Pharmacotherapy 2017; 37: 1537–1544.

43.

Shurtleff

Radosevich

Patanwala

. Delirium and coma-free days in patients receiving continuous- infusion ketamine versus propofol. Crit Care Med 2018; 46: 462.

44.

Umunna

B-P

Tekwani

Barounis

, et al. Ketamine for continuous sedation of mechanically ventilated patients. J Emerg Trauma Shock 2015; 8: 11–15.

45.

Flinspach

Booke

Zacharowski

, et al. High sedation needs of critically ill COVID-19 ARDS patients-A monocentric observational study. PLoS One 2021; 16: e0253778.

46.

Garber

Droege

Carter

, et al. Continuous infusion ketamine for adjunctive analgosedation in mechanically ventilated, critically ill patients. Pharmacotherapy 2019; 39: 288–296.

47.

Garner

Patterson

Mejia

, et al. Impact of ketamine as an adjunct sedative in acute respiratory distress syndrome due to COVID-19 pneumonia. Respir Med 2021; 189: 106667.

48.

Groetzinger

Rivosecchi

Bain

, et al. Ketamine infusion for adjunct sedation in mechanically ventilated adults. Pharmacotherapy 2018; 38: 181–188.

49.

Von der Brelie

Seifert

Rot

, et al. Sedation of patients with acute aneurysmal subarachnoid hemorrhage with ketamine is safe and might influence the occurrence of cerebral infarctions associated with delayed cerebral ischemia. World Neurosurg 2017; 97: 374–382.

50.

Achar

. Efficacy of ketamine infusion in refractory asthma complicated by acute myocardial infarction. Anaesth Intensive Care 1993; 21: 115–117.

51.

Benken

Goncharenko

. The future of Intensive Care Unit Sedation: A report of continuous infusion ketamine as an alternative sedative agent. J Pharm Pract 2017; 30: 576–581.

52.

De Pinto

Jelacic

Edwards

. Very-low-dose ketamine for the management of pain and sedation in the ICU. J Opioid Manag 2008; 4: 54–56.

53.

Hamimy

Zaghloul

Abdelaal

. The application of a new regimen for short term sedation in the ICU (ketofol) – Case series. Egypt J Anaesthesia 2012; 28: 179–182.

54.

Park

Manara

Mendel

, et al. Ketamine infusion. Its use as a sedative, inotrope and bronchodilator in a critically ill patient. Anaesthesia 1987; 42: 980–983.

55.

Treu

Groth

Patel

. The use of continuous ketamine for analgesia and sedation in critically ill patients with opioid abuse: A Case Series. J Crit Care Med 2017; 3: 148–152.

Continuous infusion ketamine for sedation of mechanically ventilated adults in the intensive care unit: A scoping review