Promethazine

What it is

Promethazine is a synthetic phenothiazine drug given by mouth as a tablet or syrup, as a rectal suppository, or by injection. Chemically it sits in the same family as the older antipsychotic phenothiazines but has a branched side chain and no ring substitution, which the label credits for its much weaker dopamine-blocking activity. The raw substance is a white to faint yellow crystalline powder that is freely soluble in water. It is a prescription drug in the United States, not a nutrient or supplement.

What the research says

The evidence is thin for a drug this old. Promethazine blocks histamine H1 receptors and also has sedative, anticholinergic and weak dopamine-blocking effects, which is why the same drug is used for allergy, nausea, motion sickness and sedation. Head-to-head emergency-department trials show it works about as well as ondansetron for nausea and worse than prochlorperazine, while causing markedly more sedation in both comparisons. Its best-evidenced use is actually an off-label one: combined with haloperidol for acute agitation in psychosis, where Cochrane rates the evidence high quality. The harms are the dominant story: a boxed warning for fatal respiratory depression in children under two, a separate boxed warning on the injection for gangrene and tissue necrosis, and documented non-medical use.

Evidence grade: Limited evidence.

How it works

Drug class: Phenothiazine-derivative first-generation H1 (histamine) receptor antagonist, used as an antihistamine, antiemetic and sedative

Promethazine blocks histamine H1 receptors, which is how it damps down allergic symptoms such as itch, hives and runny nose. It also crosses into the brain, where the same blockade plus anticholinergic and weak dopamine-blocking activity produces drowsiness and an anti-sickness effect, the latter attributed to dopamine antagonism in the brain’s chemoreceptor trigger zone. Oral effects appear within about 20 minutes and usually last four to six hours. (Source 1)

Boxed warning

PROMETHAZINE HYDROCHLORIDE TABLETS, USP SHOULD NOT BE USED IN PEDIATRIC PATIENTS LESS THAN 2 YEARS OF AGE BECAUSE OF THE POTENTIAL FOR FATAL RESPIRATORY DEPRESSION.

(Source 2)

What it is used for

• Two randomised emergency-department trials put promethazine on a par with ondansetron and behind prochlorperazine for relieving nausea, with more sedation than either comparator. No placebo-controlled outcome trial was found in this search. Evidence: limited. (Source 3)
• These are the oldest approved indications and rest on the FDA label rather than on trials we could reach. We found no systematic review or randomised outcome trial of promethazine for allergic disease in this search. Evidence: unknown. (Source 4)
• A 2022 Cochrane review of antihistamines for motion sickness found moderate-certainty evidence of benefit under real travel conditions, but the trials under those conditions used cinnarizine and dimenhydrinate, not promethazine, so the result is class-level only. Evidence: limited. (Source 5)
• Sedation is an approved use and is also the drug’s most common side effect. The evidence we reached for it is the label itself; both comparative nausea trials quantified sedation as a harm rather than a benefit. Evidence: limited. (Source 4)
• In a double-blind trial of 149 hospitalised women, promethazine matched metoclopramide for vomiting episodes and well-being but produced substantially more drowsiness, dizziness and dystonia. Evidence: limited. (Source 6)
• One 170-patient randomised trial in Iran found intramuscular promethazine reduced vertigo more than intravenous ondansetron, but with more side effects and a lower relief score. Evidence: limited. (Source 7)
• This is the best-evidenced use. Cochrane found high-quality evidence that haloperidol plus promethazine calms people faster than haloperidol alone and prevents the acute dystonia that haloperidol alone caused; midazolam alone acted faster still. Evidence: established. (Source 8)

Interactions

• Alcohol (label): Alcohol adds to promethazine’s sedation and can deepen the impairment of driving and other skilled tasks. Limit: The label prints an empty cross-reference inside this sentence - '( see and )' with nothing between the brackets - because the links in the electronic label have no text. It is kept here as the label prints it, and the sentence's full stop comes after it. Nothing is missing from the warning itself: alcohol and the other depressants named are the point. (Source 9)
• Opioids, benzodiazepines, barbiturates and other central nervous system depressants (label): Promethazine deepens and prolongs their sedation. The label instructs that barbiturate doses be at least halved and narcotic doses reduced by a quarter to a half when given with it. (Source 10)
• Epinephrine (adrenaline) (label): Promethazine can reverse adrenaline’s blood-pressure-raising effect, so adrenaline must not be used to treat low blood pressure caused by promethazine overdose. (Source 10)
• Monoamine oxidase inhibitors (case reports): Taken together, MAO inhibitors and phenothiazines have been reported to cause more extrapyramidal (movement) side effects. (Source 11)
• Other drugs with anticholinergic (drying) effects (label): Adding another anticholinergic drug compounds dry mouth, blurred vision, constipation, urinary retention and confusion. (Source 10)
• Vitamins and herbal supplements generally (label): The patient labelling asks people to disclose all vitamins and herbal supplements, but names no specific supplement. We found no documented pharmacokinetic interaction between promethazine and St John’s wort, grapefruit, calcium, iron, magnesium, potassium, vitamin K, fish oil, turmeric or red yeast rice in this search. (Source 12)
• Codeine (as combined cough syrup) and alcohol or soft drinks (case reports): The promethazine-codeine syrup mixture is drunk recreationally with soft drinks or alcohol for its sedating and euphoric effect, which stacks two respiratory depressants. The review that documents this says the practice is spread unevenly and is not limited to any one type of user. (Source 13)

Stopping it

• No withdrawal syndrome or tapering schedule for promethazine appears in the FDA label or in any source we reached. The label’s only stopping instructions are reactive: stop the drug if neuroleptic malignant syndrome occurs, and because NMS has recurred with phenothiazines, restarting should be weighed carefully. (Source 14)
• Promethazine is used non-medically, most often alongside opioids, so stopping can be about misuse rather than about a therapeutic taper. In one San Francisco clinic a quarter of methadone patients had promethazine in their urine and only 15% of those had a prescription for it. (Source 15)
• For a paradoxical reaction (hyperexcitability or abnormal movements, which can follow a single dose) the label advises stopping promethazine and considering a different drug. (Source 16)

What goes wrong

In hospitalised women with hyperemesis gravidarum, promethazine was no more effective than metoclopramide but produced far more drowsiness, dizziness and dystonia. (Source 6)

• Randomized trial, Moderate certainty.
• Size: 149 women analysed (73 metoclopramide, 76 promethazine)
• Who: women at their first hospitalisation for hyperemesis gravidarum needing intravenous antiemetic therapy.
• How long: 24 hours.
• Result: Median vomiting episodes 1 (range 0-26) versus 2 (range 0-26), P=.81. Drowsiness 58.6% versus 83.6% (P=.001, NNTb 5); dizziness 34.3% versus 71.2% (P<.001, NNTb 3); dystonia 5.7% versus 19.2% (P=.02, NNTb 8); therapy stopped for adverse events 0 of 73 (0%) versus 7 of 76 (9.2%), P=.014.
• Funding: not stated.

Reported drowsiness (58.6% compared with 83.6%, P=.001, number needed to treat to benefit [NNTb] 5), dizziness (34.3% compared with 71.2%, P<.001, NNTb 3), dystonia (5.7% compared with 19.2%, P=.02, NNTb 8), and therapy curtailment owing to adverse events (0 of 73 [0%] compared with 7 of 76 [9.2%], P=.014) were encountered less frequently with metoclopramide.

Promethazine carries a boxed warning against use in children under two years of age because of fatal respiratory depression reported after marketing. (Source 2)

• Official position, Certainty not rated.
• Size: postmarketing case reports; number not stated in the label.
• Who: children under 2 years of age, and with caution those 2 years and older.
• How long: not applicable.
• Result: Contraindicated under 2 years; the label states a wide range of weight-based doses have caused respiratory depression, so no dose is considered safe in that age group.
• Funding: not stated.

POSTMARKETING CASES OF RESPIRATORY DEPRESSION, INCLUDING FATALITIES, HAVE BEEN REPORTED WITH USE OF PROMETHAZINE HYDROCHLORIDE TABLETS, USP IN PEDIATRIC PATIENTS LESS THAN 2 YEARS OF AGE. A WIDE RANGE OF WEIGHT-BASED DOSES OF PROMETHAZINE HYDROCHLORIDE TABLETS, USP HAVE RESULTED IN RESPIRATORY DEPRESSION IN THESE PATIENTS.

The injectable form carries a second, different boxed warning for severe tissue injury including gangrene, with some cases needing fasciotomy, skin graft or amputation. (Source 17)

• Official position, Certainty not rated.
• Size: postmarketing reports; number not stated in the label.
• Who: patients given promethazine by any parenteral route.
• How long: not applicable.
• Result: Intravenous injection at concentrations greater than 1 mg/mL, intra-arterial injection and subcutaneous injection are all contraindicated.
• Funding: not stated.

Severe chemical irritation and damage to tissues regardless of the route of parenteral administration has been reported in patients treated with PHENERGAN Injection including gangrene, tissue necrosis, and thrombophlebitis; and in some cases, surgical intervention including fasciotomy, skin graft, and/or amputation have been required

A narrative review records that promethazine given by mistake under the skin, into an artery or leaking out of a vein has caused vascular spasm, venous thrombosis, nerve damage, paralysis, abscess, tissue necrosis and gangrene, with case reports of fasciotomy or amputation and of long-term loss of function. (Source 18)

• Expert review, not systematic, Low certainty.
• Size: not stated; selective literature search plus regulatory communications.
• Who: emergency and perioperative patients given parenteral promethazine.
• How long: not applicable.
• Result: No incidence rate is given; the review calls the complications relatively uncommon but says their severity is disproportionate to the benefit of the drug.
• Funding: not stated.

Limit of this finding: The review is a narrative review and expert commentary, not a systematic review, and it says so. It gives no incidence figure: it calls these complications relatively uncommon and notes in its own limitations that most promethazine safety data come from case reports and postmarketing surveillance, which may underestimate how often injury happens while emphasising the worst outcomes. Read it as evidence that severe injury can happen, not as a measure of how often. The run of digits inside the quotation, '11,16-18', is the journal's own reference numbering printed in the sentence, not part of the clinical list. The FDA labelling changes of 2009 and 2023 are a different passage of the same review and are cited separately.

Reported injuries include local pain, burning, erythema, induration, and edema, as well as more serious outcomes such as vascular spasm, venous thrombosis, nerve damage, paralysis, abscess formation, tissue necrosis, and gangrene.11,16–18 Case reports further illustrate the severity of these events, including inadvertent intra-arterial injection leading to immediate vascular compromise and progressive tissue necrosis requiring surgical intervention, such as fasciotomy or amputation. Additional reports describe severe extravasation injuries resulting in long-term functional impairment.

The injection is formulated with phenol at a pH of 4 to 5.5, which the review gives as the reason parenteral promethazine damages tissue; the review adds that deep intramuscular injection into a large muscle is the preferred route, while labelling still permits the slow intravenous push that remains usual in many hospitals. (Source 19)

• Expert review, not systematic, Low certainty.
• Size: not applicable.
• Who: patients given promethazine by injection.
• How long: not applicable.
• Result: pH between 4 and 5.5; no rate of injury is given. The review notes the labelling describes promethazine as an irritant that can be highly caustic to the intima of blood vessels and surrounding tissue.
• Funding: not stated.

Promethazine injection is formulated with phenol and has a pH between 4 and 5.5, increasing the risk of tissue injury when administered parenterally, such as via intramuscular (IM) or intravenous (IV) routes. The manufacturer’s product labeling describes promethazine as an irritant drug that can be highly caustic to the intima of blood vessels and surrounding tissue. Because of these risks, deep intramuscular injection into a large muscle is the preferred route of administration.

One case report describes neuroleptic malignant syndrome in a 13-year-old girl who had been given standard 12.5 mg intramuscular promethazine doses, with the syndrome appearing during a later general anaesthetic. (Source 20)

• Case report, Very low certainty.
• Size: 1 patient.
• Who: a 13-year-old girl weighing 29 kg with a chemotherapy-treated tumour, given 12.5 mg intramuscular promethazine for vomiting and before each of eight platelet transfusions; the last promethazine dose was three days before surgery.
• How long: onset during general anaesthesia, three days after the last promethazine dose.
• Result: Hyperpyrexia, generalised rigidity and autonomic dysfunction with a normal creatine phosphokinase level; the patient recovered with cooling, airway control and supportive care.
• Funding: not stated.

Limit of this finding: This is a single case report in a child, and it cannot show that promethazine alone caused the syndrome. The last promethazine dose was three days before the operation, no promethazine was given during the anaesthetic, and the event happened under fentanyl, propofol and sevoflurane, with malignant hyperthermia considered and then set aside. The authors' point is that neuroleptic malignant syndrome can follow ordinary doses and has no specific laboratory marker, not that promethazine was the sole cause.

The patient manifested classic clinical features of NMS, including hyperpyrexia, autonomic dysfunction, and generalized rigidity. However, the patient’s serum creatine phosphokinase (CPK) levels were within normal limits, demonstrating the heterogeneous presentation of the disorder.

A quarter of methadone-maintenance patients tested positive for promethazine in urine, and most of them had no prescription for it. (Source 15)

• Survey study, Low certainty.
• Size: 334 methadone maintenance patients and 139 community-recruited injection drug users.
• Who: patients at a county methadone clinic and community-based opioid injection drug users in San Francisco.
• How long: urine samples and past-month self-report.
• Result: 26% of methadone patients had promethazine-positive urine; only 15% of those had an active prescription. 17% of community injection drug users reported past-month use, rising to 24% among those in methadone treatment.
• Funding: not stated.

The prevalence of promethazine positive urine samples among the methadone maintenance patients was 26 percent. Only 15 percent of promethazine positive patients had an active prescription for promethazine. Among IDUs reporting injection of opiates in the community-based survey, 17 percent reported having used promethazine in the past month; 24 percent of the IDUs who reported being enrolled in methadone treatment reported using promethazine in the past month.

A systematic review of seven studies found that misuse of codeine-promethazine cough syrup as "purple drank" or "lean" is spread unevenly across populations and is explicitly not confined to any one type of user, correcting the newspaper and social-media characterisation the review quotes. (Source 13)

• Systematic review, Very low certainty.
• Size: 7 papers from 138 records screened.
• Who: people misusing codeine and promethazine cough syrup; the review states the pattern is not confined to any single group.
• How long: not applicable.
• Result: No pooled effect estimate. The review reports a heterogeneous diffusion of misuse, a low level of risk of bias where assessable, and states the misuse is not exclusively linked to a specific type of user.
• Funding: not stated.

Limit of this finding: This review is the source of the widely repeated claim that purple drank is a rap-culture drug, and part of that framing is the review's own: it says the mixture has become a popular drug among rap singers who promote its tranquilizing and euphoric effects. It then qualifies that picture rather than endorsing it, reporting that the misuse is not exclusively linked to a specific type of user and attributing the narrower characterisation to American newspapers and social media. It pooled no numbers, so it supports no estimate of how common the misuse is.

Epidemiological data highlighted a heterogeneous diffusion of the misuse of this mixture, which is not exclusively linked to a specific type of user (African-American teenagers, athletes, and rappers), as previously reported in American newspapers and in the social media.

Drowsiness is the most common adverse effect of promethazine, and the label lists extrapyramidal reactions, blood-cell suppression, potentially fatal respiratory depression and apnoea, and neuroleptic malignant syndrome among reported reactions. (Source 21)

• Official position, Certainty not rated.
• Size: not stated; spontaneous reporting.
• Who: patients taking oral promethazine.
• How long: not applicable.
• Result: No rates are given in the label for any of the listed reactions.
• Funding: not stated.

Limit of this finding: The label itself is broken at two points in this passage: after 'apnea (potentially fatal)' and after 'has also been reported' it prints an empty cross-reference, '(see )', because the link in the electronic label has no text in it. Those six characters are kept here exactly as the label prints them. They point nowhere, so there is nothing further to read at either point; the reactions listed are the label's own and no rate is given for any of them.

Drowsiness is the most prominent CNS effect of this drug. Sedation, somnolence, blurred vision, dizziness; confusion, disorientation, and extrapyramidal symptoms such as oculogyric crisis, torticollis, and tongue protrusion; lassitude, tinnitus, incoordination, fatigue, euphoria, nervousness, diplopia, insomnia, tremors, convulsive seizures, excitation, catatonic-like states, hysteria. Hallucinations have also been reported.

Hyperexcitability and abnormal movements have been reported after a single dose of promethazine, a paradoxical reaction for which the label advises considering stopping the drug. (Source 16)

• Official position, Certainty not rated.
• Size: not stated.
• Who: patients given promethazine, including after one dose.
• How long: after a single administration.
• Result: No rate given; respiratory depression, nightmares, delirium and agitated behaviour have also been reported in some of these patients.
• Funding: not stated.

Hyperexcitability and abnormal movements have been reported in patients following a single administration of promethazine hydrochloride. Consideration should be given to the discontinuation of promethazine hydrochloride and to the use of other drugs if these reactions occur. Respiratory depression, nightmares, delirium, and agitated behavior have also been reported in some of these patients.

The FDA added boxed-warning language about severe tissue injury including gangrene and amputation in 2009 and required further labelling changes for the intravenous route in 2023. (Source 22)

• Expert review, not systematic, Low certainty.
• Size: not applicable.
• Who: patients given parenteral promethazine.
• How long: not applicable.
• Result: Regulatory actions in 2009 and 2023; no rates given.
• Funding: not stated.

In 2009, the FDA mandated additional boxed warning language emphasizing the risk of severe tissue injury, including gangrene and amputation, associated with parenteral administration. More recently, in 2023, the FDA required further labeling updates to reduce the risk of severe chemical irritation and tissue damage associated with intravenous administration.

In children the label warns that promethazine’s movement side effects can be mistaken for the brain signs of an undiagnosed illness such as Reye’s syndrome, and that acutely ill dehydrated children are more prone to dystonia. (Source 23)

• Official position, Certainty not rated.
• Size: not stated.
• Who: children 2 years and older given promethazine.
• How long: not applicable.
• Result: No rates given. Hallucinations and convulsions have occurred at therapeutic doses as well as in overdose in children.
• Funding: not stated.

Limit of this finding: One sentence in this passage ends with an empty cross-reference the label itself prints - 'may cause sudden death (see )' - because the link in the electronic label has no text in it. Those characters are kept exactly as the label prints them and point nowhere. The warning is unaffected: the label contraindicates these tablets below two years of age and gives no rates for any of the effects listed.

THE EXTRAPYRAMIDAL SYMPTOMS WHICH CAN OCCUR SECONDARY TO PROMETHAZINE HYDROCHLORIDE TABLETS, USP ADMINISTRATION MAY BE CONFUSED WITH THE CNS SIGNS OF UNDIAGNOSED PRIMARY DISEASE, e.g., ENCEPHALOPATHY OR REYE'S SYNDROME. THE USE OF PROMETHAZINE HYDROCHLORIDE TABLETS, USP SHOULD BE AVOIDED IN PEDIATRIC PATIENTS WHOSE SIGNS AND SYMPTOMS MAY SUGGEST REYE'S SYNDROME OR OTHER HEPATIC DISEASES. Excessively large dosages of antihistamines, including promethazine hydrochloride tablets, USP in pediatric patients may cause sudden death (see ).Hallucinations and convulsions have occurred with therapeutic doses and overdoses of promethazine hydrochloride in pediatric patients. In pediatric patients who are acutely ill associated with dehydration, there is an increased susceptibility to dystonias with the use of promethazine hydrochloride.

What the evidence supports

Cochrane found high-quality evidence that haloperidol plus promethazine tranquillises agitated people with psychosis better than haloperidol alone, and that haloperidol alone caused acute dystonia that the combination did not. (Source 8)

• Systematic review, High certainty.
• Size: 316 participants in 1 RCT for the primary outcome (6 studies, 1367 participants randomised in the whole review)
• Who: people with psychosis-induced aggression or agitation, mostly in emergency psychiatric services.
• How long: 30 minutes for the primary outcome.
• Result: Not tranquil or asleep at 30 minutes: RR 0.65, 95% CI 0.49 to 0.87. Ten occurrences of acute dystonia on haloperidol alone and none on the combination.
• Funding: independent.

the combination treatment was clearly more effective (n=316, 1 RCT, RR 0.65, 95% CI 0.49 to 0.87, high‐quality evidence). There were 10 occurrences of acute dystonia in the haloperidol alone arm and none in the combination group. The trial was stopped early as haloperidol alone was considered to be too toxic.

What the evidence does not support

Prochlorperazine relieved uncomplicated nausea and vomiting faster and more completely than promethazine, with fewer treatment failures and less sleepiness. (Source 24)

• Randomized trial, Low certainty.
• Size: 84 patients (42 per arm)
• Who: adults 18 years or older with presumed uncomplicated gastritis or gastroenteritis at two academic emergency departments.
• How long: 60 minutes.
• Result: Prochlorperazine better at 30 and 60 minutes (P=.004 and P<.001); treatment failures 9.5% versus 31% (difference 21%, 95% CI 5 to 38, P=.03); sleepiness 38% versus 71% (difference 33%, 95% CI 13 to 53, P=.002).
• Funding: not stated.

Limit of this finding: The published abstract prints the sleepiness p-value twice in the same bracket (P =.002 appears at both ends of the figure). That duplication is in the paper, not a transcription slip, and it does not change the result: 38% versus 71% sleepiness, a 33 percentage-point difference with a 95% confidence interval of 13 to 53.

There were significantly fewer treatment failures with prochlorperazine (P =.03, 9.5% versus 31%; difference 21%, 95% confidence interval 5 to 38). There was no difference in incidence of extrapyramidal effects. Prochlorperazine caused significantly fewer complaints of sleepiness (P =.002, 38% versus 71%; difference 33%, 95% confidence interval 13 to 53; P =.002).

Where the evidence is mixed

In emergency-department patients with nausea, intravenous promethazine reduced nausea no better than ondansetron but caused significantly more sedation. (Source 3)

• Randomized trial, Moderate certainty.
• Size: 120 subjects, 60 per arm.
• Who: non-pregnant adults in an urban academic emergency department with at least 40 mm of self-reported nausea on a 100 mm visual analogue scale.
• How long: 30 minutes (primary outcome)
• Result: Nausea change: ondansetron -34 mm, promethazine -36 mm; difference -2 mm (95% CI -13 to 8 mm). Sedation: ondansetron 5 mm, promethazine 19 mm; difference 14 mm (95% CI 5 to 24 mm). No akathisia with ondansetron, 2 cases with promethazine.
• Funding: not stated.

Ondansetron and promethazine reduced nausea similarly (ondansetron -34 mm, promethazine -36 mm; difference -2 mm; 95% CI = -13 to 8 mm). The reduction in anxiety was similar (ondansetron -13 mm, promethazine -14 mm; difference -1 mm; 95% CI = -10 to 10 mm). Promethazine was associated with significantly more sedation than ondansetron (ondansetron 5 mm, promethazine 19 mm; difference 14 mm; 95% CI = 5 to 24 mm).

Midazolam alone tranquillised aggressive people with psychosis faster than haloperidol plus promethazine, and one participant on the combination had a seizure. (Source 25)

• Systematic review, High certainty.
• Size: 301 participants, 1 RCT.
• Who: people aggressive because of psychosis.
• How long: 24-hour follow-up.
• Result: Not tranquillised: RR 2.90, 95% CI 1.75 to 4.8 favouring midazolam. Serious adverse event overall RR 1.01, 95% CI 0.06 to 15.95. 1 in 150 on haloperidol plus promethazine had a swiftly reversed seizure.
• Funding: independent.

there was clear evidence that midazolam is more swiftly tranquillising of an aggressive situation than haloperidol plus promethazine (n=301, 1 RCT, RR 2.90, 95% CI 1.75 to 4.8, high‐quality evidence). On its own, midazolam seems to be swift and effective in tranquillising people who are aggressive due to psychosis. There was no difference in risk of serious adverse event overall (n=301, 1 RCT, RR 1.01, 95% CI 0.06 to 15.95, high‐quality evidence). However, 1 in 150 participants allocated haloperidol plus promethazine had a swiftly reversed seizure, and 1 in 151 given midazolam had swiftly reversed respiratory arrest.

A 2022 Cochrane review found first-generation antihistamines probably prevent motion sickness under real travel conditions and may cause more sedation, with little or no difference in blurred vision or impaired cognition; the trials run under travel conditions used cinnarizine and dimenhydrinate, not promethazine, so this is class-level evidence only. (Source 5)

• Systematic review, Moderate certainty.
• Size: 240 participants across 3 studies for prevention; 190 across 2 studies for sedation (9 RCTs, 658 participants in the review)
• Who: motion-sickness-susceptible adults aged 16 to 55; no study clearly assessed children.
• How long: single journeys or single rotating-chair exposures.
• Result: Symptoms prevented 25% placebo versus 40% antihistamines, RR 1.81 (95% CI 1.23 to 2.66), 3 studies, 240 participants, moderate certainty. Sedation 44% placebo versus 66% antihistamines, RR 1.51 (95% CI 1.12 to 2.02), 2 studies, 190 participants, low certainty. Blurred vision RR 1.14 (0.53 to 2.48) and impaired cognition RR 0.89 (0.58 to 1.38), both little or no difference, low certainty.
• Funding: independent.

Limit of this finding: This is evidence about first-generation antihistamines as a class, not about promethazine. The review states that every trial that induced motion sickness naturally evaluated only cinnarizine and dimenhydrinate, so no figure here was measured with promethazine. The review also reports that no study measured impaired cognition for the scopolamine comparison, and that no study measured sedation, impaired cognition or blurred vision for the antiemetic comparison.

Antihistamines are probably more effective than placebo at preventing motion sickness symptoms under natural conditions (symptoms prevented: 25% placebo; 40% antihistamines) (risk ratio (RR) 1.81, 95% confidence interval (CI) 1.23 to 2.66; 3 studies; 240 participants) (moderate‐certainty). The evidence is very uncertain about the effect of antihistamines on preventing motion sickness under experimental conditions (standardised mean difference (SMD) 0.32, 95% CI ‐0.18 to 0.83; 2 studies; 62 participants) (very low‐certainty). No studies reported results on the resolution of existing motion sickness symptoms. Antihistamines may result in little or no difference in gastric tachyarrhythmia under experimental conditions (mean difference (MD) ‐2.2, 95% CI ‐11.71 to 7.31; 1 study; 42 participants) (low‐certainty). No studies reported results for any other physiological measures. When compared to placebo, antihistamines may be more likely to cause sedation (sedation: 44% placebo; 66% antihistamines) (RR 1.51, 95% CI 1.12 to 2.02; 2 studies; 190 participants) (low‐certainty)

The Cochrane motion-sickness review tested only nine small trials in adults aged 16 to 55, found high risk of selective reporting in most, and included no study that clearly assessed children. (Source 26)

• Systematic review, Low certainty.
• Size: 9 RCTs, 658 participants.
• Who: adults aged 16 to 55 across seven countries; motion sickness induced naturally in six studies and experimentally in four.
• How long: single exposures.
• Result: No pooled estimate in this passage; the review notes mostly low risk of bias for randomisation and allocation concealment but mostly high risk for selective reporting.
• Funding: independent.

Limit of this finding: The naturally induced trials in this review tested cinnarizine and dimenhydrinate only, so its motion-sickness results do not measure promethazine. No included study clearly assessed children.

All the naturally induced studies only evaluated first‐generation antihistamines (cinnarizine and dimenhydrinate). Risk of bias across the studies varied, with mostly low risk for random sequence generation and allocation concealment, and mostly high risk for selective reporting. Only the experimentally induced studies measured physiological parameters and only the naturally induced studies evaluated adverse effects. There were no studies that clearly assessed the paediatric population.

Where the research disagrees

What the boxed warning for promethazine says - the oral and injectable products carry different ones

• FDA-approved oral tablet label (Preferred Pharmaceuticals, effective 24 August 2026), regulatory position based on postmarketing case reports: PROMETHAZINE HYDROCHLORIDE TABLETS, USP SHOULD NOT BE USED IN PEDIATRIC PATIENTS LESS THAN 2 YEARS OF AGE BECAUSE OF THE POTENTIAL FOR FATAL RESPIRATORY DEPRESSION. (Source 2)
• FDA-approved PHENERGAN injection label (Hikma Pharmaceuticals USA, effective 20 May 2024), regulatory position based on postmarketing case reports: WARNING: RESPIRATORY DEPRESSION IN PEDIATRICS AND SEVERE TISSUE INJURY, INCLUDING GANGRENE Respiratory Depression – Pediatrics PHENERGAN injection should not be used in pediatric patients less than 2 years of age because of the potential for fatal respiratory depression. Postmarketing cases of respiratory depression, including fatalities, have been reported with use of promethazine in pediatric patients less than 2 years of age. (Source 17)

Whether injectable promethazine should still be used at all

• Ryan and colleagues, SAGE Open Medicine 2026, in the conclusion of their narrative review and expert commentary, narrative review and expert commentary with institutional formulary experience: While many healthcare systems—particularly in the United States—have reduced or eliminated its use in favor of safer alternatives, it persists in some settings due to cost considerations, formulary inertia, and clinical familiarity. Ongoing reports of tissue injury and other adverse events underscore that its risks remain clinically relevant. As such, current practice is best characterized not by universal discontinuation, but by progressive restriction—highlighting the need for continued emphasis on safer, evidence-based strategies. (Source 27)
• FDA, which continues to approve promethazine injection with instructions for safer administration, regulatory position: PHENERGAN Injection may be administered intravenously after dilution through an intravenous catheter inserted in a large vein. Preferably through a central venous catheter (see DOSAGE AND ADMINISTRATION). If pain occurs at the injection site during intravenous infusion, immediately discontinue the infusion and evaluate for possible arterial injection or perivascular extravasation, and initiate appropriate medical management. (Source 28)

How much

• Reference intake: Promethazine is prescription-only and the dose is set by the prescriber. As a position, the FDA label (Preferred Pharmaceuticals tablet label, effective 24 August 2026) opens its DOSAGE AND ADMINISTRATION section by stating that promethazine hydrochloride tablets are contraindicated for children under 2 years of age, and then gives an average effective oral dose of 25 mg for active nausea and vomiting in children or adults, repeatable in 12.5 to 25 mg doses every 4 to 6 hours. The same section's weight-based paediatric figure of 0.5 mg per pound applies only to children 2 years and older, and the label's boxed warning is about fatal respiratory depression in younger children. (Source 29)
• Upper limit: The label sets no single maximum daily dose. Its hard limits are categorical rather than numerical: contraindicated under 2 years of age, and the lowest effective dose in children 2 and over with other respiratory depressants avoided (FDA label, effective 24 August 2026). (Source 2)
• Studied: Braude 2008 gave 25 mg promethazine intravenously as a single dose, against 4 mg ondansetron. (Source 3)
• Studied: Tan 2010 gave 25 mg promethazine every 8 hours for 24 hours intravenously, against 10 mg metoclopramide. (Source 6)
• Studied: Ernst 2000 gave 25 mg promethazine intravenously, against 10 mg prochlorperazine. (Source 24)
• Studied: The NMS case report describes repeated 12.5 mg intramuscular doses in a 13-year-old weighing 29 kg. (Source 20)

A common belief, and what the research shows

The belief: Promethazine is just an old, mild antihistamine, so it is a safe choice for a child who is vomiting.

What the research shows: It is contraindicated under two years of age, and the label says in capitals that the danger is not simply a matter of getting the dose right: "RESPIRATORY DEPRESSION AND APNEA, SOMETIMES ASSOCIATED WITH DEATH, ARE STRONGLY ASSOCIATED WITH PROMETHAZINE PRODUCTS AND ARE NOT DIRECTLY RELATED TO INDIVIDUALIZED WEIGHT-BASED DOSING, WHICH MIGHT OTHERWISE PERMIT SAFE ADMINISTRATION." The same section adds that "ANTIEMETICS ARE NOT RECOMMENDED FOR TREATMENT OF UNCOMPLICATED VOMITING IN PEDIATRIC PATIENTS, AND THEIR USE SHOULD BE LIMITED TO PROLONGED VOMITING OF KNOWN ETIOLOGY." For adults, two randomised trials found it no better than ondansetron and worse than prochlorperazine, with much more sedation than either.

Questions and answers

What is it?

Promethazine is a prescription phenothiazine medicine, sold for decades as Phenergan. It comes as a tablet or syrup to swallow, a suppository, or an injection. It is a single synthetic chemical, not a vitamin or a plant extract. (Source 30)

What does it do in the body?

It blocks histamine H1 receptors, which calms allergic symptoms. Because it also enters the brain and has anticholinergic and weak dopamine-blocking activity, it makes people sleepy and settles nausea. Oral effects start within about 20 minutes and generally last four to six hours, sometimes up to twelve. (Source 1)

Is it good or bad for you?

It depends heavily on who is taking it and how. In agitated psychosis, combined with haloperidol, Cochrane rates the evidence high quality and the combination prevented the dystonia haloperidol alone caused. For nausea in adults it is no better than cheaper, less sedating alternatives and worse than prochlorperazine. In children under two it is contraindicated because of fatal respiratory depression, and the injectable form can destroy tissue. (Source 24)

How do you get more of it?

Promethazine is prescription-only in the United States; there is no food or supplement source and no over-the-counter version. A prescriber decides the form and dose. The label lists the approved reasons for prescribing it, from allergy and motion sickness to sedation and postoperative nausea. (Source 4)

If it is harmful, what reduces it?

There is no antidote. In overdose, care is supportive: airway and breathing support, activated charcoal, diazepam for seizures, intravenous fluids for low blood pressure. Two specific traps are documented: naloxone does not reverse promethazine, and adrenaline must not be used for the low blood pressure because it can lower it further. Dialysis does not help. (Source 31)

Why might someone be low in it or missing it?

This question does not apply in the way it would to a nutrient: nobody is "low in" promethazine, because the body does not need or make it. People stop having it in their system when a prescription ends, or because it is contraindicated for them - under two years of age, in a coma, with a previous reaction to a phenothiazine, or for lower respiratory symptoms including asthma. (Source 32)

Which whole foods contain it or feed it?

No whole food contains promethazine. It is a manufactured phenothiazine supplied as tablets, syrup, suppositories or injection, and nothing in the literature we searched describes a dietary source of it. (Source 30)

We searched: We searched PubMed via E-utilities and Europe PMC for promethazine together with food, dietary and nutrient terms, and read the FDA label’s description, clinical pharmacology and patient-information sections; none describes promethazine occurring in food.

What happens if you do not have it?

Nothing is lost by not taking it; there is no deficiency state. What happens if a person who was using it for nausea stops is that the nausea is untreated, and the trials suggest other drugs do that job at least as well: ondansetron matched it for nausea with far less sedation, and prochlorperazine beat it. (Source 3)

How can you test for it?

There is no routine blood test to guide promethazine treatment, and the label says vital signs need monitoring only in extreme overdose or individual sensitivity. Promethazine can, however, be detected in urine: a San Francisco study screened 334 methadone patients’ urine samples for it and found 26% positive. Urine detection shows recent exposure, not whether a dose is working. (Source 15)

References

1. DailyMed, US National Library of Medicine (Structured Product Label). Promethazine Hydrochloride Tablets USP - CLINICAL PHARMACOLOGY (FDA label). 2026. Read the source
2. DailyMed, US National Library of Medicine (Structured Product Label). Promethazine Hydrochloride Tablets USP - boxed warning (FDA label, Preferred Pharmaceuticals Inc.). 2026. Read the source
3. Academic Emergency Medicine. Ondansetron versus promethazine to treat acute undifferentiated nausea in the emergency department: a randomized, double-blind, noninferiority trial. 2008. PMID 18304050, DOI 10.1111/j.1553-2712.2008.00060.x. Read the source
4. DailyMed, US National Library of Medicine (Structured Product Label). Promethazine Hydrochloride Tablets USP - INDICATIONS AND USAGE (FDA label). 2026. Read the source
5. Cochrane Database of Systematic Reviews. Antihistamines for motion sickness. 2022. PMID 36250781, DOI 10.1002/14651858.CD012715.pub2. Read the source
6. Obstetrics and Gynecology. Promethazine compared with metoclopramide for hyperemesis gravidarum: a randomized controlled trial. 2010. PMID 20410771, DOI 10.1097/AOG.0b013e3181d99290. Read the source
7. American Journal of Otolaryngology. Ondansetron or promethazine: Which one is better for the treatment of acute peripheral vertigo?. 2019. PMID 30268346, DOI 10.1016/j.amjoto.2018.09.010. Read the source
8. Cochrane Database of Systematic Reviews. Haloperidol plus promethazine for psychosis-induced aggression. 2016. PMID 27885664, DOI 10.1002/14651858.CD005146.pub3. Read the source
9. DailyMed, US National Library of Medicine (Structured Product Label). Promethazine Hydrochloride Tablets USP - PRECAUTIONS, Information for Patients (FDA label). 2026. Read the source
10. DailyMed, US National Library of Medicine (Structured Product Label). Promethazine Hydrochloride Tablets USP - Drug Interactions (FDA label). 2026. Read the source
11. DailyMed (US National Library of Medicine), Preferred Pharmaceuticals, Inc. label. Promethazine Hydrochloride Tablets, USP - PRECAUTIONS, Drug Interactions: Monoamine Oxidase Inhibitors (MAOI). 2026. Read the source
12. DailyMed, US National Library of Medicine (Structured Product Label). Promethazine Hydrochloride Tablets USP - Can Other Medicines Or Food Affect Promethazine? (FDA label). 2026. Read the source
13. Journal of Psychoactive Drugs. "Purple Drank" (Codeine and Promethazine Cough Syrup): A Systematic Review of a Social Phenomenon with Medical Implications. 2020. PMID 32748711, DOI 10.1080/02791072.2020.1797250. Read the source
14. DailyMed, US National Library of Medicine (Structured Product Label). Promethazine Hydrochloride Tablets USP - WARNINGS, Neuroleptic Malignant Syndrome (FDA label). 2026. Read the source
15. Journal of Addiction Medicine. Promethazine misuse among methadone maintenance patients and community-based injection drug users. 2013. PMID 23385449, DOI 10.1097/ADM.0b013e31827f9b43. Read the source
16. DailyMed (US National Library of Medicine), Preferred Pharmaceuticals, Inc. label. Promethazine Hydrochloride Tablets, USP - ADVERSE REACTIONS, Paradoxical Reactions. 2026. Read the source
17. DailyMed, US National Library of Medicine (Structured Product Label). PHENERGAN (promethazine hydrochloride) Injection - boxed warning (FDA label, Hikma Pharmaceuticals USA Inc.). 2024. Read the source
18. SAGE Open Medicine. Moving beyond promethazine: Advancing precision in antiemetic therapy (tissue injury). 2026. PMID 42205501, DOI 10.1177/20503121261449562. Read the source
19. SAGE Open Medicine. Moving beyond promethazine: Advancing precision in antiemetic therapy (pharmacology and formulation). 2026. PMID 42205501, DOI 10.1177/20503121261449562. Read the source
20. Frontiers in Psychiatry. Promethazine-induced neuroleptic malignant syndrome under general anesthesia: a case report. 2026. PMID 41684798, DOI 10.3389/fpsyt.2026.1700644. Read the source
21. DailyMed, US National Library of Medicine (Structured Product Label). Promethazine Hydrochloride Tablets USP - ADVERSE REACTIONS (FDA label). 2026. Read the source
22. SAGE Open Medicine. Moving beyond promethazine: Advancing precision in antiemetic therapy (FDA labelling history). 2026. PMID 42205501, DOI 10.1177/20503121261449562. Read the source
23. DailyMed, US National Library of Medicine (Structured Product Label). Promethazine Hydrochloride Tablets USP - WARNINGS, Use in Pediatric Patients (FDA label). 2026. Read the source
24. Annals of Emergency Medicine. Prochlorperazine versus promethazine for uncomplicated nausea and vomiting in the emergency department: a randomized, double-blind clinical trial. 2000. PMID 10918098, DOI 10.1067/mem.2000.108652. Read the source
25. Cochrane Database of Systematic Reviews. Haloperidol plus promethazine for psychosis-induced aggression (midazolam comparison). 2016. PMID 27885664, DOI 10.1002/14651858.CD005146.pub3. Read the source
26. Cochrane Database of Systematic Reviews. Antihistamines for motion sickness (included studies and risk of bias). 2022. PMID 36250781, DOI 10.1002/14651858.CD012715.pub2. Read the source
27. SAGE Open Medicine. Moving beyond promethazine: Advancing precision in antiemetic therapy (current practice). 2026. PMID 42205501, DOI 10.1177/20503121261449562. Read the source
28. DailyMed, US National Library of Medicine (Structured Product Label). PHENERGAN (promethazine hydrochloride) Injection - boxed warning, routes of administration (FDA label). 2024. Read the source
29. DailyMed, US National Library of Medicine (Structured Product Label). Promethazine Hydrochloride Tablets USP - DOSAGE AND ADMINISTRATION: section opening, Allergy, Motion Sickness and Nausea and Vomiting (FDA label). 2026. Read the source
30. DailyMed, US National Library of Medicine (Structured Product Label). Promethazine Hydrochloride Tablets USP - What Is Promethazine? (FDA-approved patient labelling). 2026. Read the source
31. DailyMed, US National Library of Medicine (Structured Product Label). Promethazine Hydrochloride Tablets USP - OVERDOSAGE, Treatment (FDA label). 2026. Read the source
32. DailyMed, US National Library of Medicine (Structured Product Label). Promethazine Hydrochloride Tablets USP - CONTRAINDICATIONS (FDA label). 2026. Read the source