A Systematic Review of Low-Dose Ketamine for Acute Pain Management in the Emergency Department

Abstract

Acute pain is the most common presentation in the emergency department (ED), accounting for approximately 78% of visits and highlighting the need for rapid and effective pain management. Ketamine is a well-established analgesic that can serve as an alternative when opioids are contraindicated. Recently, low-dose ketamine (LDK), also known as sub-dissociative or subanesthetic ketamine, has emerged as a potential option for acute pain control in ED settings; however, evidence regarding its efficacy and safety remains inconsistent. This systematic review aimed to evaluate the effectiveness and safety of LDK for acute pain management in the emergency department. A comprehensive search of SCOPUS, PubMed, and Web of Science was conducted using predefined keywords related to ketamine, acute pain, and emergency care. Peer-reviewed randomized controlled trials (RCTs) published in English between 2015 and 2025 involving adults aged ≥ 18 years were included, and the risk of bias was assessed using the Cochrane Risk of Bias 2 (ROB-2) tool. Sixteen RCTs including 1908 adult patients met the inclusion criteria. The findings were heterogeneous: several studies demonstrated that LDK provided effective pain reduction within 30 min compared with commonly used analgesics such as morphine, ketorolac, and fentanyl, whereas others found no significant superiority, including one placebo-controlled trial. The analgesic effect appeared dose- and administration-dependent, with short intravenous infusions showing better tolerability than bolus dosing. Transient neuropsychiatric adverse effects were reported, but no serious adverse events were identified. Larger multicenter studies are needed to further clarify optimal dosing strategies and confirm the safety profile of LDK in ED pain management.

1. Introduction

Acute pain is a sudden onset physiological response triggered by external or internal stimuli that lasts for hours to less than 7 days [1,2]. It is considered the most common presentation in the emergency department (ED) with around 78% of total visits [3]. This highlights the urgent need for rapid and effective pain management in patients presenting with acute pain [4]. Recent epidemiological data indicate that pain accounts for over 130 million emergency department visits annually in the United States alone, with wait times for analgesia often exceeding 60 min. Multimodal analgesia—combining agents with different mechanisms such as ketamine, nonsteroidal anti-inflammatory drugs (NSAIDs), acetaminophen, and regional blocks—is increasingly recommended to reduce opioid reliance and improve pain outcomes. However, evidence for specific multimodal combinations in the ED remains limited. Acute pain management remains one of the most challenging aspects of emergency medicine. The goal is to provide timely and effective pain relief while minimizing adverse events and improving patient outcomes and satisfaction [5].

Traditionally, opioids were potent, readily available, rapid analgesics used for acute pain management in the ED settings [6,7]. However, the highly unpleasant and potentially serious adverse events (AEs) of opioids limit their use for analgesia. Common adverse effects associated with opioids include nausea, vomiting, bradycardia, hypotension, respiratory depression, and hypoxia. Additionally, vulnerable populations such as elderly patients, children, individuals with chronic pulmonary disease, alcohol dependence, opioid misuse disorder, or a history of addiction represent a significant proportion of ED visits. Opioid therapy may be limited in these groups due to the increased risk of respiratory depression [8,9]. Consequently, there is a critical need for effective, safe non-opioid analgesic alternatives to manage acute pain in the ED [10].

Ketamine is an N-methyl-D-aspartate (NMDA) receptor antagonist that acts within the central nervous system. It is Food and Drug Administration (FDA) approved as a general anesthetic, alone or in combination with other analgesics [11]. It is a safe option that ED physicians use for procedural sedations, intubation, and prehospital agitation [12]. Also, it is used as an analgesic for acute pain management in sub-dissociative dosing known as low-dose ketamine (LDK) [13]. It is characterized by a rapid onset of action, flexible routes of administration (intravenous [IV], intramuscular, oral, intranasal, and by rectum), and preserving the airway reflexes and stable hemodynamics [14]. Moreover, the American College of Emergency Physicians (ACEP) published a policy for reshaping acute pain management in the ED in 2017. It suggested ketamine as a non-opioid option for acute pain management in the ED, which subsequently indicates the promising use of ketamine as a suitable, convenient alternative in acute pain management [15].

Since LDK is a relatively novel analgesic in ED, its efficacy and safety profile, as well as physician and patient satisfaction, have not been well studied. Furthermore, existing evidence is heterogeneous and inconsistent. ED pain presentations vary widely, from traumatic fractures and renal colic to migraine and sickle cell crisis, each with potentially different responses to LDK. This heterogeneity complicates pooled analyses and may explain inconsistent findings across trials. This systematic review aims to evaluate the efficacy and safety of low-dose ketamine for the management of acute pain in the emergency department.

2. Materials and Methods

2.1. Search Strategy

This study followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting guidelines [16]. This review analyzed the use of low-dose ketamine (LDK) in adults for managing acute pain in the emergency department (ED). A comprehensive search was conducted in three electronic databases: PubMed, Web of Science, and Scopus. All searches were limited to peer-reviewed articles published in English between January 2015 and December 2025. The complete search strings for each database are provided below.

PubMed (n = 115): (ketamine OR “low-dose ketamine” OR “sub-dissociative ketamine” OR “subanesthetic ketamine”) AND (“acute pain” OR analgesia OR “pain management”) AND (“emergency department” OR “emergency room” OR ED OR “emergency care”).

Web of Science (n = 79): TS = (ketamine OR “low-dose ketamine” OR “sub-dissociative ketamine” OR “subanesthetic ketamine”) AND TS = (“acute pain” OR analgesia OR “pain management”) AND TS = (“emergency department” OR “emergency room” OR ED OR “emergency care”) AND TS = (randomized OR randomised OR trial OR placebo) NOT TS = (review).

Scopus (n = 162): TITLE-ABS-KEY (ketamine OR “low-dose ketamine” OR “sub-dissociative ketamine” OR “subanesthetic ketamine”) AND TITLE-ABS-KEY (“acute pain” OR analgesia OR “pain management”) AND TITLE-ABS-KEY (“emergency department” OR “emergency room” OR ED OR “emergency care”) AND TITLE-ABS-KEY (randomized OR randomised OR trial OR placebo) AND NOT TITLE-ABS-KEY (review).

2.2. Eligibility Criteria

Inclusion Criteria

• Study design: Peer-reviewed English RCTs.
• Publication period: Published from 2015 to 2025.
• Rationale: Low-dose ketamine for acute pain in the ED emerged as a distinct research focus following the 2014 American College of Emergency Physicians policy statement on multimodal analgesia.
• Limitation acknowledged: Earlier RCTs may exist.
• Population: Adults aged ≥ 18 years.
• Setting: Admitted to the ED.
• Pain type: Acute pain (defined according to the contemporary biopsychosocial framework as pain lasting < 7 days, arising from tissue injury or disease, and involving sensory, emotional, and cognitive components [1,2]).
• Pain severity: Visual Analogue Scale (VAS) or Numeric Rating Scale (NRS) score ≥ 5. The threshold of ≥5 was chosen to select patients with moderate-to-severe pain, as lower scores often do not require pharmacological intervention. However, we recognize that this may exclude studies using lower thresholds and could introduce selection bias, which we acknowledge as a limitation.
• Intervention: Intravenous LDK at <1 mg/kg, administered as either a bolus or infusion [17].
• Primary outcome: Analgesic effect or reduction in pain intensity measured by NRS or VAS.
• Secondary outcomes: Adverse events related to intervention or control, and requirement of rescue analgesics.

2.3. Exclusion Criteria

Study design: Observational studies, reviews, editorials, case reports/series, in vitro studies, animal models, protocols, qualitative studies.

Population: Pediatrics (<18 years), adults with chronic or psychiatric pain, pre- or post-operative pain.

Intervention/setting: LDK administered intramuscularly (IM) or intranasally (nebulized) in non-ED settings.

Intervention type: LDK used as adjunct therapy or combined with other medications.

2.4. Screening

Following the retrieval of potential studies, duplicates were identified and removed using Rayyan (Rayyan Systems Inc., Cambridge, MA, USA). Two authors (A.A. and A.A.) independently screened titles and abstracts against the eligibility criteria. Disagreements were resolved by discussion or by consulting a third author (N.A.). Inter-rater agreement was calculated (κ = 0.89). The full texts of potentially eligible studies were then independently assessed by the same two authors. The study selection process is illustrated in the PRISMA flow diagram (Figure 1).

2.5. Data Extraction & Synthesis

The following data were extracted from the included studies: Last name of the first author, year, country, study design, sample size, gender distribution, mean age, acute pain type, and mean baseline NRS/VAS score, ketamine dose and administration method, type/dose/route of the control. The outcomes analyzed mean postintervention VAS/NRS scores measured after 30 and 60 min, or mean change in the pain score from the baseline after 30 and 60 min, incidence of adverse events, and rescue analgesics required. Data extraction was performed independently by two authors using a standardized data collection form. Extracted data were compared, and discrepancies were resolved through discussion or referral to a third author. Study investigators were not contacted for missing or additional data.

No methods were used to impute missing summary statistics or to convert data. Only data reported explicitly in each study were extracted and presented as reported. When data were insufficient for a particular outcome, they were noted as not available. No sensitivity analyses were conducted.

2.6. Quality Assessment

Risk of bias for each trial was assessed using the Cochrane Risk of Bias 2 (ROB-2) tool for randomized controlled trials [18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33]. The tool included five domains: (1) bias arising from the randomization process, (2) bias due to deviations from intended interventions, (3) bias due to missing outcome data, (4) bias in measurement of the outcome, and (5) bias in selection of the reported result [34]. Each domain was judged as low, some concerns, or high, and the overall risk of bias for the trial was assessed accordingly. Two authors independently assessed the risk of bias for each included study using the ROB-2 tool. Any disagreements were resolved by consensus or by involving a third author.

The overall risk of bias judgment (Figure 2) was used to interpret findings: the single study with high risk of bias [20] was not excluded, but its findings were noted as less reliable in the Discussion.

Formal assessment of publication bias was not performed due to heterogeneity across studies and outcomes. The certainty of evidence was not formally assessed using GRADE or a similar tool.

A narrative synthesis was performed due to clinical heterogeneity across studies in pain types, LDK doses, and outcome timepoints. No meta-analysis was conducted. VAS and NRS were treated as equivalent for pain intensity, as both are 0–10 scales. When studies reported change-from-baseline, those values were used; when only post-intervention scores were given, absolute scores were used. Timepoints were harmonized to 30 and 60 min post-administration.

3. Results

Our systematic search identified 356 potential studies from the databases; 53 duplicates were removed. Among the remaining articles, 280 were excluded after title and abstract screening, while seven were excluded after full-text screening. A total of 16 RCTs were included in this review, as shown in the PRISMA flow diagram (Figure 1).

3.1. General Characteristics of the Included RCTs

This review involved 16 RCTs; eight were double-blinded, three were prospective double-blinded, one was prospective double-blinded double-dummy, one was parallel-group prospective blinded pragmatic, one was prospective double-blinded noninferiority, one was prospective double-blinded placebo double-dummy, and one was double-blinded placebo. The RCTs were conducted in three countries: eight in Iran, seven in the USA, and one in Saudi Arabia.

A total of 1908 adult patients of both genders with acute pain were admitted to the ED; among them, 920 (48.2%) received IV LDK for pain management. Their mean age ranged from 29.4 to 77.3 years; however, one study did not report the mean age of the patients. Only one study specifically enrolled older adults (mean age 77.3 years), which limits the generalizability of our findings to geriatric populations [26]. Trauma was the most common cause of pain, accounting for 9 out of 16 studies (56.3%). Other pain conditions included acute migraine (2 studies), renal colic (1 study), vaso-occlusive crisis (1 study), and mixed non-traumatic pain (3 studies). The mean (SD) baseline pain score of the included patients was numerically similar between the ketamine and control groups, which was above 7 in all RCTs, as shown in

Table 1. General characteristics of the included randomized controlled trials (RCTs).

Author’s Last Name, Year, Country	Study Design	Sample Size n			Gender n (%)		Mean Age (SD) (Years)	Cause of Pain	Mean (SD) Baseline VAS/NRS Score
		Total	Intervention	Control	Female	Male			Ketamine	Control
Azizikhani et al., 2025, Iran [18]	Double-blind	77	38	39	11 (14.3)	66 (85.7)	38.72 (14.13)	Trauma	NRS: Median (IQR): 8 (7–10)	Median (IQR): 8 (8–10)
Mahmoodabadi et al., 2024, Iran [19]	Double-blind	90	45	45	10 (11.1)	80 (88.9)	Median (IQR): 40.9 (37.4–44.3)	Chest trauma with/without rib fractures	NRS: Median (IQR): 8 (7–10)	Median (IQR): 7 (6–8)
Nguyen et al., 2024, the USA [20]	Prospective, double-blind	150	75	75	Ketamine: 41 (55) Control: 39 (52)	Ketamine: 34 (45) Control: 36 (48)	Ketamine:46 (13) Control:47 (16)	MS non-traumatic, MS traumatic, Abdominal, Flank, Genitourinary	NRS 8.2 (1.6)	8.2 (1.5)
Alshahrani et al., 2022, Saudi Arabia [21]	Parallel-group, prospective, blind	278	138	140	116 (41.7)	162 (58.2)	29.4 (8.1)	Vasospasm-occlusive crisis	NRS 8.6 (1.3)	8.7 (1.3)
Lovett et al., 2021, the USA [22]	Prospective, double-blind	98	49	49	LD: 30 (61) HD: 27 (55)	LD: 19 (39) HD: 22 (45)	LD: 39 (36–42) HD: 37 (34–41)	Flank, back, MS, Headache and Abdominal	NRS 8.8 (8.4–9.3)	8.2 (7.8–8.6)
Esfahani et al., 2021, Iran [23]	Double-blind	73	36	37	14 (19.2)	59 (80.8)	32.9 (10.4)	Isolated limb trauma	NRS 8.4 (1.5)	8.9 (1.3)
Maleki Verki et al., 2019, Iran [24]	Double-blind	127	65	62	62 (48.9)	65 (51.1)	Ketamine: Median (IQR): 36.3 (10.73) Fentanyl 34.5 (11.97)	Limb fracture	VAS 7.38 (2.5)	7.59 (1.8)
Sotoodehnia et al., 2019, Iran [25]	Double-blind	126	62	64	Ketamine: 18 (29) Ketorolac: 12 (18.8)	Ketamine: 44 (71) Ketorolac: 52 (81.2)	Ketamine: 34.2 (9.9) Ketorolac: 37.9 (10.6)	Renal colic	NRS 8.4 (1.5)	8.7 (1.4)
Motov et al., 2018, the USA [26]	Prospective, double-blind	60	30	30	Ketamine: 23 (76.7) Morphine: 23 (76.7)	Ketamine: 7 (23.3) Morphine: 7 (23.3)	Ketamine: 77.3 (8.4) Morphine: 77.1 (8.5)	Abdominal, Cancer, Back, MS, Fracture, and Flank	NRS 8.97 (1.5)	8.4 (1.4)
Etchison et al., 2018, the USA [27]	Double-blind	34	16	18	26 (76)	8 (24)	34.3 (11.74)	Acute migraine	NRS Median (IQR): 8.25 (7.75–10)	8 (7–9)
Jahanian et al., 2018, Iran [28]	Double-blind	156	77	79	NA	NA	NA	Upper and lower extremity long bone fractures and direct blunt trauma	NRS Ulnar: 7.75 (0.5) Tibia: 8.13 (0.5) Shaft radius: 7.78 (0.4) Proximal humerus: 8.27 (1.1) Femor: 9.18 (0.87) Distal radius: 7.96 (0.5)	NRS Ulnar: 7.5 (0.5) Tibia: 8 (0.67) Shaft radius: 7.4 (0.5) Proximal humerus: 9 (0.9) Femor: 9.37 (0.74) Distal radius: 7.9 (0.5)
Jahanian et al., 2018, Iran [29]	Double-blind	156	78	78	4 (28.8)	111 (71.2)	35.87 (3.4)	Traumatic fractures of the long bones	VAS 8.28 (1.55)	8.18 (1.63)
Mahshidfar et al., 2017, Iran [30]	Double-blind	300	150	150	Ketamine: 174 (16) Morphine: 177 (18)	Ketamine: 126 (84) Morphine: 123 (82)	Ketamine: 34.4 (7.6) Morphine: 34.1 (7.3)	Trauma	NRS 8.1 (1.1)	8.4 (0.9)
Motov et al., 2017, the USA [31]	Prospective, double-blind	48	24	24	IV push: 15 (62.5) Infusion: 12 (50)	IV push: 9 (37.5) Infusion: 12 (50)	IV push: 42.2(15.1) Infusion: 43.6 (12.3)	Abdominal, flank, and MS	Not Reported	Not Reported
Motov et al., 2015, the USA [32]	Prospective, double-blind	90	45	45	Ketamine: 30 (67) Morphine: 28 (62)	Ketamine: 15 (33) Morphine: 17 (38)	Ketamine: 35 (9.5) Morphine: 36 (10.5)	Abdominal, flank, and MS	NRS 8.6 (1.5)	8.5 (1.5)
Miller et al., 2015, the USA [33]	Prospective, double-blind	45	24	21	22 (49)	23 (51)	30 (11)	Abdominal, flank, low back, or extremity	NRS 7.13 (1.7)	7.14 (1.5)

SD: Standard deviation, n: Number, VAS: Visual analogue scale, NRS: Numerical rating scale, MS: Musculoskeletal, IQR: Interquartile range.

.

3.2. Primary Outcome- LDK Effectiveness

In general, eleven RCTs evaluated the IV LDK in comparison to other analgesics such as 30 mg IV ketorolac, 0.05 and 0.1 mg/kg IV morphine, and 4 μg/kg nebulized fentanyl. In addition, four RCTs evaluated the effect of different doses and methods of administration of LDK in pain management. Only one RCT compared the LDK with saline (placebo), which surprisingly did not show any significant difference in median reduction of 0.2 mg/kg IV ketamine and normal saline [27]. The findings of this review emphasized that five out of 16 (31.25%) RCTs showed a significant reduction in pain score after 30 min of LDK administration compared to the controls. The lowest mean pain intensity scores after 30 min ranged from 2.1 to 2.8 across studies, with Esfahani et al. [23] reporting 2.1 and Maleki Verki et al. [24] reporting 2.14 (

Table 2. Primary and secondary outcomes of acute pain management using LDK among the included RCTs.

Author’s Last Name, Country, Year	Dose/ Administration of Ketamine (mg/kg)	Dose/Route of Control (mg/kg)	Mean (SD) Post-Intervention VAS/NRS Score		Adverse Events n (%)		Additional Analgesic Request (Rescue) n (%)
			Ketamine	Control	Ketamine	Control	Ketamine	Control
Azizikhani et al., 2025, Iran [18]	Bolus of 0.15 mg/kg for 1 min, then 30 min 0.15 mg/kg ketamine infusion	Ketamine Bolus of 0.3 mg/kg for 1 min, then 30 min 0.9% saline infusion	T30: Median (IQR): 4 (2–5) T60: 4 (1–6)	T30: Median (IQR): 6 (3–7) p = 0.068 T60: 5 (5–7) p = 0.155	Feeling of unreality:49 (59.7) Agitation: 37 (48.1) Vertigo: 37 (48.1) Nausea: 26 (33.8) Hallucination: 19 (24.7) Sense of doom 18 (23.4) Dissociation: 1 (1.3)		Morphine: 19 (50)	Morphine: 25 (64.1)
Mahmoodabadi et al., 2024, Iran [19]	0.25 mg/kg, IV	Ketorolac, 30 mg, IV	T30: median [IQR] 95% CI NRS 3.0 [1.0] 2.8–3.5 T60: NRS 3.0 [2.0] 2.7–3.7	T30: 5.0 [4.5] 4.2–5.8, p  =  0.006 T60: 5.6 [1.7] 4.7–6.4, p  <  0.001	Facial flushing: 1 (2.2) Nausea: 13 (28.9) Nystagmus: 21 (46.7)	Non-pleuritic chest pain: 5 (11.1) Nausea: 4 (8.9), p = 0.015	One morphine dose: 13 (28.9) Two morphine: 2 (4.4)	One morphine: 27 (60), p = 0.012 Two morphine: 7 (15.6), p = 0.032
Nguyen et al., 2024, the USA [20]	0.3 mg/kg, IV	Ketamine, 0.75 mg/kg, nebulized	T30: 3.6 (3.3) T60: 3.3 (2.8)	T30: 3.8 (3.3) MD: 0.23 (95% CI −1.32 to 0.857) T60: 4.1 (3.4) MD: 0.79 (95% CI −1.83 to 0.245)	None		Ketorolac and morphine: 10 (13.3) → 1 at T30 4 at T60 5 at T90	Ketorolac and morphine 21 (28) → 1 at T15 2 at T30 4 at T60 8 at T90 6 at T120
Alshahrani et al., 2022, Saudi Arabia [21]	0.3 mg/kg, IV	Morphine, 0.1 mg/kg, IV	6.9 (5.27)	6.8 (4.11) p = 0.78 MD: 0.16 (−0.96–1.27)	Total 8 (6.3) → Dizziness: 5 (3.9) Nausea: 4 (3.1) Vomiting: 1 (0.8)	Total: 3 (2.2) → Dizziness: 3 (2.2) OR: 2.81 (0.65–16.74)	Morphine Mean: 0.89 (0.88) Tramadol: 6 (4.3)	Morphine 0.9 (1.44) Tramadol: 10 (7.1)
Lovett et al., 2021, the USA [22]	0.15 mg/kg, IV	0.3 mg/kg, IV	T30: 4.7, CI (3.8–5.5) T60: 5.1 (4.2–6)	T30: 5, CI (4.2–5.8) MD: −0.3, CI (−1.6–1) T60: 4.7 (3.8–5.6) MD: −0.4 (−1.6–0.9)	T15: Mood alteration: 6 (12)	T15: Mood alteration: 14 (29) Hearing changes: 6 (13) Hallucination: 6 (13)	T30: 1 (2) T60: 2 (4)	T60: 4 (8)
Esfahani et al., 2021, Iran [23]	0.1 mg/kg, IV	Morphine, 0.05 mg/kg, IV	T30: 2.1 (1.2) Mean change: −6.2, 95%CI (−5.71 to −6.69)	T30: 3 (1.3) p = 0.002 Mean change: −5.8, 95%CI (−5.15 to −6.48)	20 (55.6)	9 (24.3) p = 0.009	2 (5.6)	5 (13.5)
Maleki Verki et al., 2019, Iran [24]	0.4 mg/kg, IV with nebulized saline	Fentanyl, 4 μg/kg of 50 μg/mL, nebulized and 100 cm3 IV infusion of normal saline	T30: 2.14 (1.4) T60: 2.33 (0.84)	T30: 3.66 (2.8) p = 0.001 T60: 3.11 (1.3) p = 0.001	NA	NA	0 (0)	44 (71) p = 0.001
Sotoodehnia et al., 2019, Iran [25]	0.6 mg/kg, IV	Ketorolac, 30 mg, IV	T30: 2.8 (2.9) T60: 1.4 (2.4)	T30: 1.8 (2.3) T60: 1 (2.2)	Dizziness: 25 (40.3) Hypertension: 15 (27.8) Tachycardia: 2 (3.2) Agitation: 9 (14.5) Nausea 7 (11.3)	Hypertension: 1 (1.9) Nausea 9 (14.1)	NA	NA
Motov et al., 2018, the USA [26]	0.3 mg/kg, IV	Morphine, 0.1 mg/kg, IV	T30: 4.2 (3.4) T60: 3.9 (3.2)	T30: 4.4 (3.1) MD: −0.2 (−1.93 to 1.46) T60: 4 (2.9) MD: −0.1 (−1.68 to 1.48)	Dizziness (62.7) Feeling of unreality (26.7) Discomfort (16.7)	Dizziness (30) and fatigue (16.7)	4 → T30: 2 (7) T60: 1 (3) T90: 1 (3)	9 → T30: 1(3) T60: 2 (7) T90: 2 (7) T120: 3 (10)
Etchison et al., 2018, the USA [27]	0.2 mg/kg, IV	Saline	Median (IQR) reduction from baseline 1 (0–2.25)	2 (0–3.75) MD: −1 (−2 to1) p = 0.5035	None	None	T30: 69 (11)	78 (14)
Jahanian et al., 2018, Iran [28]	0.5 mg/kg, IV	Morphine, 0.1 mg/kg, IV	T30 Ulnar: 3.75 (0.5) Tibia: 4.53 (1.4) Shaft radius: 3.89 (0.3) Proximal humerus: 5.4 (1.8) Femor: 5.5 (2) Distal radius: 4.37 (0.9)	T30 Ulnar: 4.75 (1.58) Tibia: 4.58 (1) Shaft radius: 7.4 (0.5) Proximal humerus: 5.6 (1.9) Femor: 6.4 (1.3) Distal radius: 4.34 (0.9) p = 0.03	NA	NA	NA	NA
Jahanian et al., 2018, Iran [29]	0.5 mg/kg, IV	Morphine, 0.1 mg/kg, IV	T30: 4.63 (1.14) T60: 3.47 (0.73)	T30: 4.84 (0.95) T60: 3.48 (0.86)	Nausea & vomiting: 9 (11.5) Dyspepsia: 7(8.9) Drowsiness: 3 (3.8)	Nausea & vomiting: 12 (15.4) Dyspepsia: 8 (10.3) Drowsiness: 5 (6.4)	Fentanyl IV 1 µg/kg 6 (7.7) Ketamine 24 (30.7)	Fentanyl IV 1 µg/kg 3 (3.8) Morphine 24 (30.7)
Mahshidfar et al., 2017, Iran [30]	0.2 mg/kg, IV	Morphine, 0.1 mg/kg, IV	T30: 4.5 (3.1) T60: 4.9 (3.3)	T30: 3.8 (3) p = 0.01 T60: 3.2 (2.9) p < 0.001	Nausea: 24 (16) Dizziness: 51 (34) Mood changes: 6 (4) Reduced Oxygen sat.: 6 (4)	Nausea: 26 (17) Flushing: 54 (36) Dizziness: 48 (32) Mood changes: 4 (2) Reduced Oxygen sat: 27 (18)	51 (34)	15 (10) p = 0.001
Motov et al., 2017, the USA [31]	0.3 mg/kg, IV push for 5 min	Ketamine, 0.3 mg/kg, short infusion for 15 min	Mean change T0-T15 5.17 (3.53)	5.75 (3.48) p = 0.026	Unreality: 22 (91.7) Dizziness: 16 (66.7) Vision: 6 (25) Discomfort: 6 (25)	Unreality: 13 (54.2), p = 0.008 Dizziness: 18 (75) Vision: 9 (37.5) Discomfort: 4 (16.7)	8 (33.3)	7 (29.1)
Motov et al., 2015, the USA [32]	0.3 mg/kg, IV	Morphine, 0.1 mg/kg, IV	T30: 4.1 (3.2) T60: 4.8 (3.2)	T30: 3.9 (3.1) MD: 0.2 (−1.19 to 1.46) T60: 3.4 (3) MD:1.4 (0.13–2.75)	Dizziness: 24 (53) Disorientation: 13 (29) Mood changes: 6 (13) Nausea: 4 (9)	Dizziness: 14 (31) Disorientation: 1 (2) Mood changes: 1 (2) Nausea: 4 (9)	Fentanyl 25 (55.5)	17 (37)
Miller et al., 2015, the USA [33]	0.3 mg/kg, IV	Morphine, 0.1 mg/kg, IV	Change from T0 to T60: −3.5 (−5.4 to −1.6)	Change from T0 to T60: −4.8 (−5.8 to −3.8)	Nausea: 3 Dysphoria:4 Dizziness: 2 Hallucination:3	Nausea: 2 Dizziness: 1 Headache: 3 Drowsiness: 2	2nd dose: 13 (54) 3rd dose: 6 (25)	2nd dose: 8 (38) 3rd dose: 3 (14)

MD: Mean difference; Bold indicates statistical significance (p < 0.05).

).

3.2.1. LDK Versus Other Analgesics

Maleki Verki et al. reported that 0.4 mg/kg ketamine had a significantly lower pain score after 30 min of administration (p = 0.0001) compared to 4 μg/kg fentanyl [24]. Mahmoodabadi et al. showed that 0.25 mg/kg IV ketamine was significantly more favorable (p = 0.006) over 30 mg IV ketorolac [19], which is contrary to the findings of Sotoodehnia et al., who used 0.6 mg/kg IV ketamine versus 30 mg IV ketorolac [25].

Moreover, 0.1 mg/kg IV morphine had comparable efficacy to 0.3 mg/kg IV ketamine in four RCTs [21,26,32,33]. However, another three RCTs found that 0.1, 0.2, and 0.5 mg/kg IV ketamine had significantly higher mean change and reduction in the pain score after 30 min of administration (p = 0.002, p = 0.03, p = 0.01), respectively, compared to 0.05 and 0.1 mg/kg IV morphine [23,28,30]. Interestingly, another RCT with the same setting as Jahanian et al.’s study had contrary results [29].

However, these comparisons should be interpreted cautiously because studies varied substantially in pain type (trauma vs. renal colic vs. migraine), age (29 to 77 years), LDK dose (0.1–0.6 mg/kg), and administration method (bolus vs. infusion).

3.2.2. Effect of Route of Administration and Dose of LDK on Pain Management

Lovett et al. and Azizikhani et al. both agreed that there is no significant difference between 0.15 mg/kg bolus IV ketamine and 0.3 mg/kg bolus IV ketamine in efficacy and side effects [18,22]. Nguyen et al. found that 0.3 mg/kg IV ketamine had a similar effect to 0.75 mg/kg nebulized ketamine [20]. Furthermore, Motov et al. showed that 0.3 mg/kg IV ketamine administered as a short infusion had better outcomes in terms of efficacy and safety (p = 0.026) compared to bolus [31].

3.3. Secondary Outcome- Adverse Events (AEs) and Rescue Analgesia

(i)

Overall patterns

Eleven studies reported neuropsychiatric adverse events associated with low-dose ketamine, including feelings of unreality, nausea, vomiting, hallucination, dissociation, dizziness (vertigo), nystagmus, and dysphoria. No serious or persistent adverse events were observed in any of the included studies.

(ii)

Dose and administration comparisons

A short infusion of 0.3 mg/kg IV ketamine had a significantly lower rate of feeling of unreality than a bolus push (p = 0.008) [31]. Higher doses of ketamine (0.3 and 0.6 mg/kg IV) were associated with higher rates of adverse events compared to lower doses.

(iii)

Comparisons with control treatments: When compared to other analgesics, the adverse event profile varied by comparator.

vs. morphine: Esfahani et al. found that 0.1 mg/kg IV ketamine had higher rates of adverse events (p = 0.009) compared to 0.05 mg/kg IV morphine [23].

vs. ketorolac: Mahmoodabadi et al. reported that 30 mg IV ketorolac had a higher rate of nausea (p = 0.015) than 0.25 mg/kg IV ketamine [19].

vs. fentanyl: Maleki Verki et al. found that fentanyl 4 µg/kg had notable adverse events (p = 0.001) compared to no reported adverse events related to 0.4 mg/kg IV ketamine [24].

(iv)

Rescue analgesia requirement: Ketorolac, morphine, tramadol, and fentanyl were the most used rescue analgesics when the initial intervention provided insufficient pain relief. Mahshidfar et al. showed that 0.2 mg/kg IV ketamine required significantly more additional doses of rescue analgesia (p = 0.001) compared to 0.1 mg/kg IV morphine [30]. Additionally, Mahmoodabadi et al. demonstrated that 30 mg IV ketorolac provided insufficient analgesia, requiring one or two additional doses of morphine (p = 0.012 and p = 0.022, respectively) [19].

Regarding the rescue analgesia requirement, ketorolac, morphine, tramadol, and fentanyl were the most reported analgesics used when there was insufficient analgesic effect by ketamine or control. Also, the results showed that 0.2 mg/kg IV ketamine required significant additional doses of rescue analgesia (p = 0.001) compared to 0.1 mg/kg IV morphine [30]. In addition, Mahmoodabadi et al. demonstrated that 30 mg IV ketorolac had insufficient analgesia, requiring an extra one or two doses of morphine (p = 0.012 and p = 0.022) [19].

3.4. Risk of Bias Assessment

All 16 RCTs were assessed using the ROB-2 tool. The findings showed that 11 trials had some concerns, four had a low risk of bias [23,28,29,33], and one trial had a high risk of bias [20]. Bias from deviations from the intended interventions, missing outcome data, outcome measurement, and timing of randomization were the main concerns in most of the RCTs, as shown in Figure 2.

4. Discussion

This review offers a valuable summary of RCTs evaluating the efficacy and safety of LDK in managing acute pain in adults who visited the ED. The results were conflicting and heterogeneous, as some studies confirmed the effectiveness of LDK in pain reduction within 30 min compared to other analgesics such as morphine, ketorolac, and fentanyl, while others did not find any notable superiority. Surprisingly, one placebo-controlled study showed no significant benefit over saline. Additionally, the analgesic efficacy of LDK appeared to be dose- and administration-dependent, with short IV infusions offering a more tolerable safety profile than bolus dosing. Although transient neuropsychiatric effects were common with LDK, particularly at higher doses, no serious AEs were reported.

Across the 16 included RCTs, transient neuropsychiatric effects were the most commonly reported adverse events associated with low-dose ketamine. These included feeling of unreality (reported in 26–92% of patients across studies), dizziness (40–75%), nausea (11–34%), hallucinations (13–25%), and dissociation. No serious or persistent adverse events (e.g., respiratory depression, need for intubation, prolonged psychomimetic effects) were documented in any of the studies.

Dose-relatedness was evident; higher doses of ketamine (0.3–0.6 mg/kg) were associated with more frequent adverse events compared to lower doses (0.1–0.15 mg/kg). Additionally, the route of administration influenced the safety profile: short infusion of 0.3 mg/kg ketamine resulted in significantly lower rates of feeling of unreality compared to bolus push (p = 0.008) [31].

Compared with other analgesics, the adverse event profile varied. LDK was associated with higher rates of neuropsychiatric effects than morphine [23] but lower rates of nausea than ketorolac [19] and lower rates of overall adverse events than fentanyl [24].

While LDK appears safe in terms of serious harms, clinicians should counsel patients about the possibility of temporary dissociative symptoms, dizziness, and nausea, particularly when using bolus dosing or higher doses. The choice between LDK and other analgesics should consider the trade-off between neuropsychiatric effects (more common with LDK) and other adverse events such as nausea or respiratory depression (more common with opioids).

In fact, the heterogeneous results were attributed to the site of pain, which has a substantial influence on the effect of IV LDK. Specifically, abdominal, musculoskeletal, neuropathic, migraine, or renal colic pain may have different responses to treatments [35]. This is aligned with the results of previous systematic reviews. A systematic review and meta-analysis performed by Lee et al., including six RCTs and 438 patients, found that the effect of LDK is pain-site dependent [36].

Our findings support this, as we observed better responses in trauma patients compared to those with renal colic or migraine. Additionally, Karlow et al. conducted a meta-analysis that included three RCTs and reported a pooled estimate of the mean change in pain scores between the LDK and morphine of 0.42 (95% CI = −0.70 to 1.54), indicating no significant difference between the two analgesics [37]. Our results are consistent with this finding, as 4 out of 6 studies comparing LDK to morphine showed no statistically significant difference. Likewise, Balzer et al. included eight RCTs with a total of 1191 patients (LDK = 598, morphine = 593). The study reported a pooled estimate of the mean pain scores of LDK and morphine, between 15 and 30 min (MD = −0.03, 95% CI = −0.37 to 0.32), 30–45 min (MD = 0.40, 95% CI = −0.89 to 1.68), and 45–60 min (MD = 0.52, 95% CI = −0.03 to 1.07), which indicated no meaningful difference [38]. In contrast, Sandberg et al. found a higher reduction in pain scores in the LDK group compared to fentanyl (MD: −3.0 (95% CI −3.86 to 2.14) and morphine (MD −0.4 (95% CI −0.08 to 0.0) among 2760 patients [39].

The adverse event profile of LDK differs from that of opioids. In previous systematic reviews, LDK was consistently linked to higher rates of neurological and neuropsychiatric effects such as agitation, dissociation, and feelings of unreality, whereas opioids more commonly caused nausea, vomiting, and cardiovascular issues [36,38]. These trends match our own findings. In our review, transient neuropsychiatric adverse events including feelings of unreality, dizziness, nausea, and hallucinations, were common with LDK, but no serious or lasting adverse events were reported in any of the 16 included RCTs. Direct comparisons showed that LDK had higher rates of neuropsychiatric effects than morphine [23], but lower rates of nausea than ketorolac [19], and fewer overall adverse events than fentanyl [24]. These results imply that selecting between LDK and other pain relievers should weigh the neuropsychiatric side effects against other adverse event risks.

Two additional systematic reviews support our overall conclusion. Ghate et al. [40] found that LDK is effective with self-limited neuropsychiatric adverse events, making it a favorable alternative to opioid analgesics in the ED, though they noted that evidence remains inconsistent. Similarly, Parvizrad et al. [41] included 13 RCTs and reported that 47% of the included studies showed significant pain reduction with LDK, and 26.7% showed fewer adverse events associated with LDK compared to controls. Taken together with our findings, these reviews indicate that while LDK has promise for acute pain management in the ED, consistent superiority over conventional analgesics has not been established across all pain types and clinical contexts.

Based on our findings, LDK may be most beneficial for traumatic pain when opioids are contraindicated. However, clinicians should expect variability in response depending on pain type, dose, and administration method. The lack of benefit in the placebo-controlled migraine trial [27] suggests that LDK is not universally effective for all acute pain conditions.

This review has several methodological limitations: it was not registered, no protocol was prepared, a formal certainty assessment (GRADE) was not performed, and publication bias was not assessed. These factors may affect the confidence in the findings.

5. Conclusions

The evidence from 16 RCTs indicates that low-dose ketamine reduces acute pain in some emergency department settings, particularly for traumatic pain, with effects that appear dose- and administration-dependent. However, the overall evidence does not support LDK as a universally superior analgesic: only 31.25% of the included studies demonstrated significant superiority over comparators, one placebo-controlled trial found no benefit over saline, and marked heterogeneity across pain types, dosing regimens, and administration methods precludes definitive conclusions. Transient neuropsychiatric adverse events were common but no serious adverse events were reported. Larger, multicenter, long-term studies are needed to identify which patient subgroups and pain conditions are most likely to benefit from LDK before evidence-based recommendations can be made for clinical practice.