What is the safest rate for IVP valproic acid?

What is the safest rate for IVP valproic acid?

Comment by InpharmD Researcher

Available evidence does not establish a single safest intravenous push (IVP) rate for valproic acid. A prospective trial found undiluted valproate doses of 20 to 30 mg/kg administered at 6 or 10 mg/kg/min to be safe and well tolerated, although paresthesia was more frequent at 10 mg/kg/min. Retrospective studies reported that administration at 500 mg/min shortened time to treatment without statistically significant increases in most adverse events, while a separate cohort of 570 patients receiving undiluted valproate over ≤5 minutes reported few adverse events and none classified as probably or definitely drug related. Overall, the available data support the tolerability of IVP valproic acid at rates up to 10 mg/kg/min or 500 mg/min, but do not identify one rate as definitely safest.

Background

A recent review article on intravenous push (IVP) administration of antiseizure medications emphasizes the growing use of this approach in emergency departments. IVP offers a significant advantage by eliminating the need for pharmacy compounding and preparation, as well as the setup of infusion materials, tubing, and pumps, allowing for faster drug administration. Regarding the intravenous push of valproic acid (VPA), the authors highlight that this method could reduce delays associated with traditional infusion techniques, offering the potential for quicker intervention. Current dosing guidelines for VPA suggest a range of 15 to 45 mg/kg, with infusion times of 2.5 to 7.5 minutes. Limited research indicates that IVP administration at a rate of 6 mg/kg/min may be feasible, with low rates of adverse events, such as dizziness, thrombocytopenia, and mild hypotension, none of which were related to the infusion rate. Given the promising data from studies on undiluted rapid VPA administration, further research into IVP administration for the treatment of status epilepticus is warranted. Additionally, VPA has demonstrated efficacy in terminating benzodiazepine-resistant status epilepticus when compared to other antiseizure medications, further supporting the potential benefits of IVP administration in emergency settings. [1]

Literature Review

A search of the published medical literature revealed 5 studies investigating the researchable question:

What is the safest rate for IVP valproic acid?

Level of evidence

B - One high-quality study or multiple studies with limitations Read more→

Please see Tables 1-5 for your response.

Safety of Rapid Intravenous Loading of Valproate
Design	Open-label, prospective trial N= 40
Objective	To evaluate the safety of administration of undiluted valproic acid (VPA) sodium (20 or 30 mg/kg/min) administered intravenously at rates of 6 or 10 mg/kg/min
Study Groups	20 mg/kg (n= 20) 30 mg/kg (n= 20)
Inclusion Criteria	Age 19 years or older, taking VPA to control seizures or epilepsy indication for use
Exclusion Criteria	Allergy to VPA, liver disease, status epileptics, concurrent high-dose lamotrigine therapy
Methods	VPA sodium was prepared from 5 mL vials (equivalent to 100 mg valproic acid) as an undiluted solution to be infused via peripheral access by manual injection, using a stopwatch to monitor administration rate. IV access was confirmed with a saline flush before and after dosing. Patients were divided to receive 20 mg/kg or 30 mg/kg as loading dose, with each cohort group further divided to be infused at a rate of 6 mg/kg/min or 10 mg/kg/min. 6 mg/kg/min = 3.3 min administration time for a 20 mg/kg dose and 5 min for a 30 mg/kg dose 10 mg/kg/min = 2 min administration time for a 20 mg/kg dose and 3 min for a 30 mg/kg dose
Duration	Up to 24 hours monitoring
Outcome Measures	Local tolerance graded on a 1-10 scale (0= none, 10= worst), heart rate (HR), mean arterial pressure (MAP), changes in vital signs and level of consciousness
Baseline Characteristics		All patients (N= 40)
	Age, years	39
	Female	13
	Race White Black Hispanic Asian	28 9 2 1
		20 mg/kg (n= 20)	30 mg/kg (n= 20)
	Weight, kg (standard deviation [SD])	91.8 (27.8)	73.2 (14.2)
	Valproate dose, mg (SD)	1834.0 (556.1)	2190.6 (439.3)
Results	Endpoint	20 mg/kg (n= 20)	30 mg/kg (n= 20)	p-Value
	Local tolerance (SD) Pain/burning Paresthesia Duration	5.5 (2.5) 2.9 (3.2) 2.1 (0.5)	5.9 (2.3) 2.3 (3.3) 2.3 (0.4)	0.71 0.67 0.31
		6 mg/kg/min infusion rate (n= 20)	10 mg/kg/min infusion rate (n= 20)
	Local tolerance (SD) Pain/burning Paresthesia Duration	5.4 (2.2) 1.0 (2.5) 2.2 (0.6)	5.9 (2.7) 4.5 (2.9) 2.1 (0.2)	0.63 0.005 0.47
	The results of the repeated measures analysis showed that there were no significant changes in MAP (p= 0.7) and HR (p= 0.9) over time
Adverse Events	There were 3 complaints of sedation and 1 complaint of nausea among treated patients. No patients reported a decline in level of consciousness
Study Author Conclusions	Rapid administration of undiluted valproate is safe and well tolerated at infusion rate up to 10 mg/kg/min and doses of up to 30 mg/kg. The lack of serious cardiovascular, neurological, hepatic, or local adverse effects supports the use of VPA in emergent situations.
InpharmD Researcher Critique	Based on the weight of patients, the max infusion rate could have exceeded a 200 mg/min infusion rate. However, there is a lack of investigation for patients that would have received high infusion rates due to their weight.

Table 1 References:
[2] Limdi NA, Knowlton RK, Cofield SS, et al. Safety of rapid intravenous loading of valproate. Epilepsia. 2007;48(3):478-483. doi:10.1111/j.1528-1167.2007.00989.x

Safety of Intravenous Push Valproate Compared with Intravenous Piggyback at a Tertiary Academic Medical Center
Design	Single-center, retrospective, pre-post descriptive analysis N= 1,330
Objective	To compare the safety profile of valproate administration via intravenous piggyback (IVPB) and intravenous push (IVP)
Study Groups	IVPB (n= 890) IVP (n= 440)
Inclusion Criteria	Adult patients (≥18 years) who received intravenous valproate from March to August 2022
Exclusion Criteria	Known prior allergy to valproate, incomplete or unavailable medical record, or received both IVPB and IVP administrations
Methods	Retrospective analysis of valproate administration via IVPB from March to May 2022 and IVP from June to August 2022. IVP doses were administered at 500 mg/min. Data collected included patient demographics, concomitant medications, physiological variables, and time from pharmacist verification to administration. Safety endpoints were assessed using the Naranjo Adverse Drug Reaction Probability Scale.
Duration	March 2022 to August 2022
Outcome Measures	Primary: Incidence of infusion site reactions (infiltration or phlebitis) Secondary: Valproate-associated bradycardia, hypotension, sedation, and time from pharmacist verification to administration
Baseline Characteristics		IVPB (n= 890)	IVP (n= 440)	p-value
	Age, years	56.9 ± 16.9	63.2 ± 14.6	0.03
	Male	43 (59.7%)	35 (63.6%)	0.65
	BMI, kg/m2	27.5 ± 6.3	26.3 ± 6.9	0.31
	Location - ICU	368 (41.3%)	145 (33.0%)	<0.01
Results		IVPB (n= 890)	IVP (n= 440)	p-value
	Infusion site reactions - Phlebitis	2 (0.2%)	4 (0.9%)	0.10
	Infusion site reactions - Infiltration	2 (0.2%)	8 (1.8%)	0.01
	Bradycardia - New onset	8 (0.9%)	6 (1.4%)	0.43
	Hypotension - New onset	16 (1.8%)	13 (3.0%)	0.17
	Sedation - New onset	59 (6.6%)	23 (5.2%)	0.32
Adverse Events	Infusion site reactions were minimal and similar between groups. Bradycardia, hypotension, and sedation events were also similar and classified as possible or doubtful by the Naranjo algorithm.
Study Author Conclusions	IVP valproate administration may be considered safe, allowing for more optimal clinical and operational outcomes in the acute care setting.
Critique	The study's retrospective design and single-center nature may limit generalizability. The sample size may not have been adequately powered to detect significant adverse effects, especially at higher doses of valproate. The study did not evaluate differences in clinical efficacy between administration methods.

Table 2 References:
[3] Wang FY, McLaughlin KC, Schontz MJ, DeGrado JR, Dannemiller RE. Safety of Intravenous Push Valproate Compared with Intravenous Piggyback at a Tertiary Academic Medical Center. Clin Drug Investig. 2024;44(3):175-181. doi:10.1007/s40261-024-01349-z

Valproate protein binding following rapid intravenous administration of high doses of valproic acid in patients with epilepsy
Design	Open-label, prospective, safety and pharmacokinetic study N= 40
Objective	To characterize protein binding in patients with epilepsy who achieve transient high (>150 mg/L) total plasma concentrations following rapid valproate infusion at very high doses
Study Groups	20 mg/kg VPA (n= 20) 30 mg/kg VPA (n= 20)
Inclusion Criteria	Patients with epilepsy administered 20 or 30 mg/kg loading doses of undiluted valproate sodium injection
Exclusion Criteria	Not specified
Methods	Patients were administered 20 or 30 mg/kg loading doses (6 or 10 mg/kg/min) of undiluted valproate sodium injection. Total and unbound valproic acid concentrations were used to assess VPA binding to plasma albumin. One- and two-binding site models were explored in a nonlinear mixed effects population analysis framework. Blood samples were collected at baseline and various intervals post-infusion for determination of plasma concentrations of VPA.
Duration	Not specified
Outcome Measures	Primary: Characterization of VPA protein binding using one-binding site model Secondary: Influence of covariates on binding site association constant (K)
Baseline Characteristics		20 mg/kg (n= 20)	30 mg/kg (n= 20)
	Total dose, mg	1,796 ± 552	2,192 ± 428
	Age, years	37.2 ± 14.3	40.8 ± 15.0
	Weight, kg	89.9 ± 27.6	73.3 ± 13.9
	Male	65%	70%
	White	80%	60%
	Co-therapy with enzyme-inducing AEDs	15%	50%
Results		20 mg/kg (n= 20)	30 mg/kg (n= 20)
	Mean unbound VPA concentration , mg/L	28.1 (0.800 to 86.7)	28.1 (0.800 to 86.7)
	Mean total VPA concentration , mg/L	130 (11.2 to 235)	130 (11.2 to 235)
	Mean unbound fraction, Fu	0.201 (0.0287 to 0.489)	0.201 (0.0287 to 0.489)
Adverse Events	Not specified
Study Author Conclusions	A one-binding site model without any significant covariates for binding constants optimally described VPA protein binding. The estimated dissociation constant was within the therapeutic range for unbound VPA concentrations, and protein binding was nonlinear. Although the range of unbound fraction and VPA concentrations were much higher than previous studies, the dissociation constant was consistent with historical data in normal healthy adults and epilepsy patients receiving lower doses.
Critique	The study effectively characterized VPA protein binding at high concentrations using a one-binding site model. However, the small sample size and exclusion of certain data points may limit the generalizability of the findings. Additionally, the study did not explore potential adverse events or long-term effects of high-dose VPA administration.

Table 3 References:
[4] Dutta S, Faught E, Limdi NA. Valproate protein binding following rapid intravenous administration of high doses of valproic acid in patients with epilepsy. J Clin Pharm Ther. 2007;32(4):365-371. doi:10.1111/j.1365-2710.2007.00831.x

Speed and Safety of IV Push Versus IV Piggyback Valproate Loading Doses: A Retrospective Cohort Study
Design	Single-center, retrospective cohort study N= 131
Objective	To compare time from pharmacist verification to documented administration for intravenous push (IVP) versus intravenous piggyback (IVPB) valproate loading doses and compare adverse event frequencies
Study Groups	IV push at 500 mg/min (n= 60) IV piggyback administered over more than 60 minutes (n= 71)
Inclusion Criteria	Patients who received ≥15 mg/kg of IV valproate and were ≥18 years of age between October 2023 and May 2025
Exclusion Criteria	Initial dose not a loading dose (e.g., initial maintenance dose >15 mg/kg)
Methods	Adults admitted to a large academic medical center who received an IV valproate loading dose were identified through the institutional data warehouse, with manual chart review used to confirm eligibility and clinical information. Before August 15, 2024, valproate was prepared in a hazardous-drug compounding hood and administered as an IVPB over more than 60 minutes. Beginning August 15, 2024, institutional practice changed to bedside administration of undiluted IV valproate by IV push at 500 mg/min without a maximum dose cap. Pharmacy verification workflow, staffing, and order-priority defaults remained unchanged, although valproate vials were stocked in automated dispensing cabinets on high-use units after implementation. Patients were grouped according to administration method. Baseline characteristics, indication, loading dose, administration location, seizure characteristics, clinical outcomes, adverse events, and post-load concentrations were collected using a standardized REDCap form. Immediate adverse events were assessed within 4 hours, and valproate-associated adverse events were assessed within 24 hours.
Duration	October 2023 to May 2025
Outcome Measures	Primary: Time from pharmacist verification to loading dose administration Secondary: Order-to-administration time, adverse events, ICU admission, length of stay, mortality
Baseline Characteristics		IVP (n = 60)	IVPB (n = 71)	p-value
	Age, median (IQR)	62 (42 to 72)	63 (49 to 70)	0.65
	Male	39 (65%)	46 (65%)	1.00
	Neurologic comorbidities Seizure disorder Substance use disorder Intracranial malignancy Traumatic brain injury Ischemic stroke Neurodegenerative disorder Intracerebral hemorrhage Subarachnoid hemorrhage Other None	30 (50) 11 (18) 9 (15) 8 (13) 7 (12) 6 (10) 5 (8) 3 (5) 1 (2) 10 (17)	29 (41) 13 (18) 9 (13) 8 (11) 7 (10) 5 (7) 4 (6) 4 (6) 4 (6) 12 (17)	0.31 0.98 0.76 0.79 0.84 0.62 0.67 0.81 0.28 0.99
	Abbreviations: IQR, interquartile range.
Results		IVP (n= 60)	IVPB (n= 71)	Effect estimate, IVPB - IVP (95% CI)	p-value
	Time to administration, minutes, median (IQR)	28 (13 to 48)	63 (50 to 84)	36 (22-52)	<0.001
	Order placed STAT	23 (38%)	33 (46%)	—	0.42
	Secondary outcomes favored IVP administration for operational efficiency, with a shorter median order-to-administration time than IVPB (34.0 [IQR 19.8 to 60.0] vs 75.0 [54.0 to 98.5] minutes; p< 0.001) and fewer delayed STAT doses administered more than 60 minutes after pharmacist verification (2/23 [9%] vs 12/33 [36%]; p= 0.027). Immediate adverse events within 4 hours occurred in 23 IVP patients (38%) and 20 IVPB patients (28%; p= 0.27), including cardiovascular events in 38% versus 28% (p= 0.27) and sedation in 3% versus 0% (p= 0.20); no phlebitis or infiltration occurred in either group. Laboratory-defined hyperammonemia within 24 hours occurred in 4 IVP patients (7%) and 1 IVPB patient (1%; p= 0.18), with no thrombocytopenia or pancreatitis reported. Post-load intubation (8% vs 3%; p= 0.24), subsequent ICU admission among patients not already in the ICU (14% vs 5%; p= 0.24), ICU length of stay (median, 4 vs 8 days; p= 0.18), hospital length of stay (14 vs 22 days; p= 0.22), and in-hospital mortality (20% vs 15%; p= 0.52) did not differ significantly between groups. Median post-load total valproate concentrations were higher with IVP than IVPB (108.6 [IQR 84.6 to 128.6] vs 81.8 [65.8 to 99.8] mcg/mL; p= 0.0039).
Adverse Events	Cardiovascular events occurred in 38% of IVP patients and 28% of IVPB patients (P = 0.27). Sedation occurred in 3% of IVP patients and 0% of IVPB patients (P = 0.20). Valproate-associated adverse events, including hyperammonemia, occurred in 7% of IVP patients and 1% of IVPB patients (P = 0.18). No phlebitis or infiltration occurred.
Study Author Conclusions	Administering IV valproate loading doses as IVP versus IVPB significantly reduced the time to administration without affecting the frequencies of adverse drug events. These findings support IVP as a safe and more efficient method for valproate loading in acute care settings.
Critique	This is the largest reported cohort specifically evaluating IVP valproate loading and provides clinically useful comparative evidence that undiluted administration at 500 mg/min can substantially shorten administration time without a statistically detectable increase in adverse events. However, the study was retrospective, single-center, and underpowered for uncommon or modest safety differences; because it evaluated only one IVP rate, it supports the apparent safety of 500 mg/min in this setting but does not establish that 500 mg/min is the “safest” rate or determine whether slower or faster IVP rates are preferable.

Table 4 References:
[5] Crawford M, Johnson R, Webb AJ. Speed and safety of IV push versus IV piggyback valproate loading doses: a retrospective cohort study. Ann Pharmacother. Published online 2026. doi:10.1177/10600280261459930.

Rapid administration of undiluted IV push valproic acid
Design	Retrospective, observational, single-center cohort analysis N= 570
Objective	To evaluate the safety and tolerability of IVP administration of undiluted VPA for doses up to 4000 mg
Study Groups	All patients (N= 570)
Inclusion Criteria	Adults 18 years and older who received at least three undiluted VPA via rapid IV administration (≤ 5-min push) between January 1, 2020 and May 27, 2025
Exclusion Criteria	Pregnant, nursing, prisoners, and patients who received less than three doses of VPA during admission
Methods	Retrospective analysis of patients receiving at least 3 doses of undiluted IVP VPA. Doses administered via IVP over five minutes, with a max loading dose of 4000 mg. Data collected included dose, frequency, number of doses, and administration location (peripheral or central IV).
Duration	January 1, 2020 to May 27, 2025
Outcome Measures	Primary: Safety and tolerability of undiluted IVP VPA doses Secondary: Adverse events or medication errors related to IV VPA administration
Baseline Characteristics		All patients (N= 570)
	Age, years (IQR)	50.4 (38.6 to 62)
	Weight, kg (IQR)	80.5 (67.1 to 97.5)
	Height, cm (IQR)	167.6 (160 to 175.3)
	Hospital length of stay, days	5.8 (3.1 to 12.7)
	Male	224 (39%)
	White	481 (84.4%)
	Admission disposition - ICU	203 (36%)
	Admission disposition - Ward	367 (64%)
	History of epilepsy	87 (15.3%)
	Indication for VPA - Seizures	101 (17.7%)
	Indication for VPA - Psychiatric disorder	90 (15.8%)
	Indication for VPA - Headache/pain	370 (64.9%)
	Indication for VPA - Other	9 (1.6%)
	Resumption of home VPA	51 (8.9%)
Results		All patients (n= 570)
	Total doses administered	5,546
	Loading doses administered	55
	Most common loading dose	3,000 mg (33% of cases)
	Most common maintenance dose	500 mg (78.6%)
	Peripheral IV access	67.6%
	Adverse events identified	3
	Adverse events classified as probable or definite drug-related	0
Adverse Events	Three adverse events were identified, none classified as probable or definite drug-related reactions according to the Naranjo scale. Events included IV line infiltration, upper extremity swelling, and a diffuse drug rash, all of which were attributed to alternative causes or multiple concomitant medications.
Study Author Conclusions	Rapid administration of undiluted IVP VPA across a broad dosing range was well tolerated and associated with low rates of adverse events. This method may reduce delays in medication preparation and administration while simplifying nursing workflows in various clinical scenarios.
Critique	The study's large sample size and extensive dose range provide robust data on the safety of undiluted IVP VPA. However, its retrospective design may lead to underreporting of adverse events, and the exclusion of patients with fewer than three doses could underestimate the true adverse event rate. Additionally, conclusions regarding doses exceeding 3000 mg should be interpreted cautiously due to limited data on higher doses.

Table 5 References:
[6] Rollins S, Radosevich JJ, Sammani N, Haller JT. Rapid administration of undiluted IV push valproic acid. J Neurol. 2026;273:453. doi:10.1007/s00415-026-13983-y.