Caffeine Intoxication

Abstract

Background

Many case reports of acute caffeine intoxication have been published, but there have been hardly any systematic data analyses, nor have there been any prospective studies, representative epidemiologic studies, or guideline recommendations. In this review, we aim to describe the clinical entity of acute caffeine intoxication and its treatments to date on the basis of published case reports.

Methods

This review is based on pertinent case reports retrieved by a search in the PubMed, Google Scholar, and Semantic Scholar databases covering the period 1851–2023 and employing the keywords “caffeine intoxication,” “caffeine poisoning,” and “caffeine overdose.”

Results

216 published individual cases were analyzed. The median caffeine intoxication dose was 12 g (0.05–106 g), and the median serum caffeine concentration was 160 mg/L (15.6–1560 mg/L). The intoxication was deliberate with suicidal intent in 40 of cases, accidental in 22, and a treatment error in 6. 21 of the accidental intoxications were in minors. Convulsions and wide-complex tachycardias due to intoxication were associated with a worse outcome, and extracorporeal hemodialysis appears to be a safe and effective way to eliminate caffeine, with a better survival rate even after severe intoxication. Doses of 5–10 g of caffeine can be lethal.

Conclusion

This is the largest analysis of caffeine intoxications ever carried out to date, yet no definitive treatment recommendations can be derived from it. Caffeine elimination by hemodialysis can be considered if the quantity ingested is potentially lethal, or in cases with a severe clinical course.

Information on CME

This article has been certified by the North Rhine Academy for Continuing Medical Education. The questions on this article may be found (in German) at http://daebl.de/RY95 (Deutsches Ärzteblatt’s CME portal). Their English translation may be found in the PDF version of this article. The closing date for entries is September 18, 2026.

Participation is possible at cme.aerztebatt.de

Caffeine, or 1,3,7-trimethylxanthine, occurs naturally in plants (1–3) (Box). Caffeine is used for medicinal purposes (4) but also as an ingredient in beverages (5), foodstuffs (6–11), and nutritional supplements, e.g., pre-workout preparations (12). Tables 1 and 2, respectively, show the average caffeine content of various foods and the health-related aspects (13–18) of caffeine consumption. Chemically, caffeine is a moderately water-soluble organic substance with low molecular weight and low plasma protein binding (19).

Box. Examples of natural occurrence of caffeine in (crop) plants (1–3).

• Coffee tree (e.g., Coffea arabica, Coffea robusta): coffee beans

• Cacao tree (Theobroma cacao): cacao beans

• Guarana (Paullinia cupana): guarana berries or seeds

• Cola tree (Cola acuminata): cola nut

• Maté (Ilex paraguariensis): maté leaves

• Tea plant (Camellia sinensis): tea leaves

Table 1. Average caffeine content of various beverages, foodstuffs, and nutritional supplements (5–11).

Foodstuff	Caffeine content
Filter coffee (250 mL)	ca. 85 mg
Instant coffee (250 mL)	ca. 60 mg
Espresso (30 mL)	ca. 40 mg
Decaffeinated coffee (250 mL)	ca. 3 mg
Dark chocolate (50 g)	ca. 25 mg
Whole milk chocolate(50 g)	ca. 10 mg
Doppelherz Energie Plus (per sachet)1	ca. 75 mg
Energy drink (250 mL)	ca. 80 mg
Cola (360 mL)	ca. 40 mg
Green tea (250 mL)	ca. 30 mg
Black tea (250 mL)	ca. 45 mg
Vitaldin Sport Power Gummies (per portion)1	ca. 80 mg
Power bar – Power gel (67 mL)	ca. 51 mg
MySupps Caffeine Boost 300 capsules (one capsule)1	ca. 300 mg

¹ Supplements available in Germany

Table 2. Health-related aspects of coffee/caffeine consumption.

Parameters investigated, by organ system		Effect	Reference(s)
	• Mortality, morbidity, blood pressure, heart rate/variability, cholesterol	• No increased risk with caffeine up to 400 mg/d	(14)
	• Cardiovascular mortality (coronary heart disease, stroke, heart failure)	• Probable risk reduction, incrementally up to 4 cups of coffee/d	(16)
	• Stroke, heart failure	• No association to very probable risk reduction, incrementally up to 4 cups of coffee/d	(16)
	• Diabetes mellitus type 2	• Probable risk reduction, incrementally up to 4 cups or 300 mL/d coffee	(16)
	• Gallbladder stones	• Possible risk reduction, particularly with 4–5 cups of coffee/d	(16)
	• Chronic liver disease, liver cirrhosis	• Possible to very probable risk reduction, incrementally up to 4 cups of coffee/d	(16)
	• Mood (anxiety, anger, confusion, depression), headache, sleep, hazardous behavior	• No significant association with caffeine up to 400 mg/d; effects in some sensitive persons with a tendency towards anxiety/sleep disturbance	(14)
	• Depression	• Risk reduction, incrementally up to 4 cups of coffee/d	(16)
	• Alzheimer disease	• Possible to very probable risk reduction	(13, 15–17)
	• Huntington chorea	• Increased risk	(15)
	• Parkinson disease	• Probable risk reduction	(13, 15–17)
	• Glaucoma	• Risk reduction	(15)
	• Risk of fractures/falls, bone density/osteoporosis, calcium homeostasis	• No significant effect with caffeine 400 mg/d	(14)
	• Pain perception	• Reduction	(15)
	• Stamina	• Improvement with caffeine intake, in moderate doses (3–6 mg/kg [18])	(13, 18)
	• Urge incontinence	• Increased risk at high doses of caffeine	(15)
	• Fecundity, fertility, spontaneous abortion, repeated miscarriage, stillbirth, prematurity, birth defects, child behavior	• No significant concerns with caffeine up to 300 mg/d	(14)
	• Fetal growth, cancer in childhood, congenital malformations	• Signs of possible effects with caffeine up to 300 mg/d	(14)
	• Miscarriage	• Possible to probable increase in risk	(16)
	• Low birth weight, childhood cancer (leukemia)	• Possible increase in risk	(16)
	• Breast cancer, colorectal cancer, colon cancer, prostate cancer	• Probable risk reduction, incrementally up to 4 cups of coffee/d	(16)
	• Endometrial cancer	• Probable risk reduction, incrementally up to 300 mL/d	(16)
	• Liver cancer	• Possible risk reduction	(16)
	• Esophageal cancer, ovarian cancer, pancreatic cancer, kidney cancer	• No association	(16)

Intestinal absorption of caffeine takes place within about 45 minutes with almost 99 bioavailability; the plasma peak occurs after about 1 hour. Decomposition is primarily hepatic through the action of the enzyme CYP1A2, with the resulting metabolites then eliminated via the kidneys (1, 13). The metabolites paraxanthine, theobromine, and theophylline are themselves biologically active and contribute to the overall effect of caffeine (13). The half-life is 3–6 hours and is affected by numerous factors (Table 3) (1, 13, 20–22), particularly the dose of caffeine (21, 23).

Table 3. Pharmacology of caffeine.

Factors influencing caffeine metabolism (1, 13, 20–23)
Half-life ↑	• Babies, pregnancy, oral contraceptives, quinolones, antiarrhythmics, liver diseases, antimycotics, alcohol, high doses of caffeine or its metabolites (e.g., theophylline), antihistamines (cimetidine, famotidine), fluvoxamine, idrocilamide, grapefruit juice
Half-life ↕	• CYP1A2 gene polymorphisms
Half-life ↓	• Fasting, smoking, grilled meat, cabbages, psoralen, proton pump inhibitors (omeprazole, lansoprazole)
Pharmacological action depending on caffeine concentration (1, 13, 15, 24–26)
From ca. 8 mg/L	• Antagonistic action at adenosine receptors A1 and A2, for example cAMP elevation in end organs  – Adenosine A1 receptors: for example adrenal medulla: for example catecholamine secretion  – Adenosine A2 receptors: for example cerebral: for example dopaminergic stimulation
From ca. 35 mg/L	• Inhibition of acetylcholinesterases
From ca. 55 mg/L	• Antagonistic action at GABA-A receptors
From ca. 90 mg/L	• Inhibition of phosphodiesterases
From ca. 50 mg/L From ca. 1000 mg/L	• Agonism at ryanodine receptors with slight elevation of intracellular calcium • Agonism at ryanodine receptors with relevant elevation of intracellular calcium
Elevation of catecholamines	• β1-Receptor stimulation: for example cardiac: positively ino-, chrono-, bathmo-, lusi-, dromotropic (increase in contractility, heart rate, excitability, relaxation, and stimulus conduction) • β2-Receptor stimulation: for example vasodilatation, hypokalemia

CYP, Cytochrome P450; GABA, gamma-aminobutyric acid

The pharmacological action (Table 3) (1, 13, 15, 24–26) and the clinical symptoms are dose-dependent. While intake of around 250 mg of caffeine has been described as leading to euphoria, increased alertness, enhanced concentration, and improved performance, doses of 500 mg have been reported to result in nervousness, anxiety, and palpitations, for example (21, 27, 28).

Symptoms of intoxication can occur starting at about 1–2 g of caffeine or a serum concentration of >15 mg/L. Amounts of 5–10 g, concentrations of 100–200 mg/kg, or serum levels of >80 mg/L are viewed as potentially fatal (1, 14, 29, 30). The thresholds are derived predominantly from case reports.

Caffeine intoxication is defined as the occurrence, after high doses of caffeine, of specific symptoms that cannot be attributed to other physical or mental disorders (31, 32).

The frequency of caffeine intoxication is unknown. Several thousand cases were reported to poison control centers in the USA in 2019, but the actual prevalence is probably higher (33). The symptoms (1, 13, 14, 22, 34, 35, e124) and differential diagnoses of acute intoxication (1, 36) are summarized in Table 4. There are currently no (inter)national guidelines on the treatment of caffeine intoxication.

Table 4. Symptoms and differential diagnoses of acute caffeine intoxication.

Symptoms
by organ system (DSM-5: 305.90, ICD-11: 6C48.2) (1, 13, 14, 22, 34, 35, e124)
	• Hyper-/hypotension, tachy-/bradycardia, AV block, SVT, VT, VF, MI, asystole/PEA
	• Hyperventilation, ARDS in aspiration
	• Nausea, vomiting, abdominal pain, diarrhea
	• Nervousness, anxiety, insomnia, tremor, psychosis, hallucinations, epileptic seizures/convulsions
	• Weakness, rigidity, rhabdomyolysis
	• Polyuria, acute kidney injury, more rarely crush kidney
	• Hypokalemia/-natremia, metabolic acidosis, respiratory alkalosis, lactatemia, hyperglycemia
Differential diagnoses
for acute caffeine intoxication (1, 36)
	• Intoxication: amphetamine, cathinone, disulfiram, iron, carbon monoxide, cocaine, salicylate, theophylline, cyanide
	• Status epilepticus, delirium tremens, intracranial hemorrhage, bipolar disorder, sleep disorder, anxiety disorder
	• Diabetic ketoacidosis, hyperthyroidism in thyrotoxicosis
	• Septic shock

ARDS, Acute respiratory distress syndrome; AV, atrioventricular; MI, myocardial ischemia; PEA, pulseless electrical activity; SVT, supraventricular tachycardia; VF, ventricular fibrillation; VT, ventricular tachycardia

The European Food Safety Authority (EFSA) issued a caffeine consumption safety assessment in 2015 after several EU member states had expressed concerns about caffeine consumption among the general population, in certain groups, and by individual persons, e.g., physically active adults and consumers of energy drinks (11).

Caffeine in amounts of up to 400 mg/d for healthy adults, 200 mg/d for pregnant and breastfeeding women, and < 2.5 mg/kg body weight for children is considered harmless (11, 37).

Method

This narrative review is based on data from publications between 1851 and 2023 identified by retrieving case reports (search terms: “caffeine intoxication,” “caffeine poisoning,” “caffeine overdose”) from the databases PubMed.gov, Google Scholar, and Semantic Scholar (for details, see the eBox).

eBox. eMethods.

The aim of this study was to describe potential treatment options on the basis of published case reports. Between October 2021 and April 2024, we searched Pubmed.gov, Google Scholar, and Semantic Scholar for open access/institutional access case reports in the period 1851 to 2023; the earliest case report found appeared in 1883. Owing to the extremely heterogeneous nature of the data, no further selection on the basis of quality criteria took place. The 142 case reports (eTable 1, [e1–e142]) published between 1883 and 2023 described 216 cases of caffeine intoxication/overdose, 110 of them with a fatal outcome. In 76 of these 110 cases the patients were found dead, so the analyses were conducted post mortem. In six fatalities it was not clear whether the person was already dead when found. The number of cases was too low for analysis of any contribution to intoxication of other ingested substances, so solely caffeine intoxication was considered.

Results

The power of this study is significantly limited by several factors. First among these is the lack of randomized prospective data on the treatment of severe caffeine intoxication. The data are therefore derived solely from retrospective case reports and case series, which may be prone to publication bias, e.g., changing publication behavior. The level of evidence regarding treatment options is therefore definitely to be classed as very low. Moreover, the comparability of the data is restricted by the partial lack of information on intoxication, diagnosis, and the patient’s previous history. Furthermore, treatment comparisons are severely hampered by differences in the treatment concepts pursued and by the advances in intensive care medicine over time. Finally, the power of the study is also limited by the fact that the caffeine causing intoxication was ingested in various ways and forms, and the presence of further, unknown substances could not be ruled out.

Intoxication

Overall, the median amount of caffeine causing intoxication in the 142 case reports (eTable 1) was 12 g (0.05–106 g), yielding a median serum concentration of 160 mg/L (15.6–1 560 mg/L). The source of intoxication with caffeine-containing substances was injections in 4 of cases, beverages in 5, powders in 10, and tablets in 43; in the remaining 38, the source was unknown. In one case both tablets and beverages containing caffeine had been ingested. The intoxication occurred with suicidal intent in 40 of cases, accidentally in 22, as a treatment error in 6, as child abuse in 1, and deliberately without suicidal intent or involvement of other persons in 1; in 29, the cause was unknown. In view of the inhomogeneous nature of the data, in some cases it is probably uncertain exactly what was involved: accidental intoxication by an unintentional overdose, intentional overdosing with undesired intoxication, or deliberate intoxication by overdosing.

eTable 1. Data from analysis of all case reports found by the literature search described in the Methods section.

216 cases of caffeine intoxication in 142 case reports/case series
110 cases with fatal outcome after caffeine intoxication in 48 reports
28 cases of persons found alive in 19 reports
eRef.	e1, e4, e19, e24, e28, e34, e35, e37, e39, e63, e67, e70, e73, e80, e84, e85, e93, e103, e129
76 post-mortem cases in 23 reports
eRef.	e2, e5, e10, e14, e21, e27, e29, e30, e42, e45, e50, e51, e58, e65, e74, e76–e79, e82, e97, e114, e130
6 cases with imprecise data regarding time of death in 6 reports
eRef.	e12, e53, e55, e56, e91, e90
106 cases with positive outcome after caffeine intoxication in 94 reports
eRef.	e3, e6–e9, e11, e13, e15–e18, e20, e22, e23, e25, e26, e31–33, e36, e38, e40, e41, e43, e44, e46-e49, e52, e5e54, e57, e59–62, e64, e66, e68, e69, e71, e72, e75, e81, e83, e86–e89, e92, e94–96, e98–e102, e104–e113, e1e115–128, e131–e142

The published case reports were dominated by suicidal and accidental caffeine intoxication (eFigure). The rate of accidental intoxications was slightly higher in men than women (46 versus 35). The median age of this group was 23 years, and 21 were minors, including a large proportion of infants and young children.

eFigure.

Numbers of case reports of caffeine intoxication in 10-year intervals from the 1960s to the 2010s, by cause

The median age of those intending to commit suicide was 25 years, and 55 were women (eTable 2).

eTable 2. Analysis of the 216 cases in 142 case reports during the period 1883–2023 (m = 82, f = 97, n.d. = 37).

Cause	Suicidal 40.3 (n = 87)	Accidental 22.2 (n = 48)	Medical error 5.6 (n = 12)	Child abuse 1.4 (n = 3)	Deliberate 1.4 (n = 3)	n.d. 29.2 (n = 63)
Period	1985–2023	1974–2023	1883–2012	1968–1997	1994–2012	1936–2023
		Children 1997–2023	Infants 1980–2001
Median age [years] (33 minors, 15.3)	25 (7 minors, 8)	23 (10 minors, 21)	< 1 (8 infants, 67)	1 (3 minors, 100)	34 (0 minors, 0)	31 (5 minors, 8)
Sex distribution	m = 41.4 f = 55.2 n.d. = 3.4	m = 45.8 f = 35.4 n.d. = 18.8	m = 50 f = 41.7 n.d. = 8.3	m = 33.3 f = 0 n.d. = 66.7	m = 33.3 f = 66.7 n.d. = 0	m = 25.4 f = 39.7 n.d. = 34.9
Median dose [12 g] (0.05–106 g)	20 g (2 g–106 g)	5 g (0.05 g2 –30 g)	0.3 g (0.06 g–3.2 g)	n.d.	n.d.	10 g (0.13 g–60 g)
Median serum concentration [160 mg/L] (15.6–1560 mg/L)	177 mg/L (45.5–574 mg/L)	158.5 mg/L (15.6– 560 mg/L)	82.5 mg/L (26–346 mg/L)	117.3 mg/L (117–1040 mg/L)	78.5 mg/L (29–128 mg/L)	165.3 mg/L (42–567 mg/L)
Tablets (n = 93; 43)	n = 55 (63.2)	n = 18 (37.5)	n = 1 (8.3)	n = 2 (66.7)	n = 2 (66.7)	n = 15 (23.8)
Powder (n = 22; 10.2)	n = 8 (9.2)	n = 10 (20.8)	n = 1 (8.3)	n = 0 (0)	n = 1 (33.3)	n = 2 (3.2)
Beverage (n = 11; 5.1)	n = 0 (0)	n = 9 (18.8)	n = 0 (0)	n = 0 (0)	n = 0 (0)	n = 2 (3.2)
Injection (n = 8. 3.7)	n = 0 (0)	n = 0 (0)	n = 6 (50)	n = 1 (33.3)	n = 0 (0)	n = 1 (1.6)
Mixed intoxication1 (n = 1, 0.5)	n = 1 (1.1)	n = 0 (0)	n = 0 (0)	n = 0 (0)	n = 0 (0)	n = 0 (0)
n.d. (n = 81, 37.5)	n = 23 (26.4)	n = 11 (22.9)	n = 4 (33.3)	n = 0 (0)	n = 0 (0)	n = 43 (68.3)
Died (n = 110, 50.9)	n = 41 (47.1)	n = 25 (52.1)	n = 2 (16.7)	n = 3 (100)	n = 0 (0)	n = 39 (61.9)
Survived (n = 106, 49.1)	n = 46 (52.9)	n = 23 (47.9)	n = 10 (83.3)	n = 0 (0)	n = 3 (100)	n = 24 (38.1)
Patients with dialysis (survived/died)	n = 25 (24/1)	n = 5 (5/0)	n = 0	n = 0	n=0	n = 8 (7/1)
Persons initially found alive(n = 134)	n = 56 (6 minors)	n = 31 (10 minors)	n = 12 (8 minors)	n = 2 (2 minors)	n = 3 (0 minors)	n = 30 (5 minors)

216 published cases with acute caffeine intoxication worldwide during the period 1883 to 2023 (f, female; n, number of cases; m, male; n.d., no data: sex, cause of intoxication, source of intoxication)

¹ Mixed intoxication with caffeine from various sources

² Note in intoxication dose of 0.05 g: child, male, age 12 years, weight 34 kg

eTable 3 shows that caffeine intoxication is a worldwide medical problem. The reported sources of caffeine were caffeine-containing tablets/injected solutions, powders, and beverages.

eTable 3. Occurrence of caffeine intoxication over time, with causes and geographical distribution, in the 216 cases described in 142 reports worldwide in the period 1883 to 2023.

Worldwide (100)	Cases	Fatalities	Suicidal	Accidental	Treatment error	Child abuse	Deliberate	n.d.
1883–2023	216	110	87	48	12	3	3	63
1883–1959	3	1	0	0	2	0	0	1
1960–1969	2	1	0	0	0	1	0	1
1970–1979	10	4	1	3	1	0	0	5
1980–1989	34	13	12	9	6	1	0	6
1990–1999	19	11	7	5	1	1	1	4
2000–2009	34	28	20	4	1	0	0	9
2010–2019	86	48	31	21	1	0	2	31
2020–2023	28	4	16	6	0	0	0	6
North America (35.6)	Period	Cases	Fatalities	Country	Period	Cases	Fatalities
USA	1936–2023	75	33	Canada	1988–2021	2	0
Europe (34.7)	Period	Cases	Fatalities	Country	Period	Cases	Fatalities
Sweden	1993–2010	26	25	Switzerland	2012–2015	2	0
Germany	1991–2020	7	5	Spain	2013	2	0
France	1987–2021	7	2	Finland	1959	1	1
Poland	2013–2022	6	3	Slovakia	2019	1	1
Great Britain	1883–2020	6	0	Belgium	1991	1	0
Portugal	2011–2021	4	0	Austria	1994	1	0
Netherlands	1992–2023	3	0	Serbia	2012	1	0
Denmark	1999–2004	3	3	Czech Republic	2015	1	0
Italy	2014–2020	3	2
Asia (27.3)	Period	Cases	Fatalities	Country	Period	Cases	Fatalities
Japan	1993–2023	55	32	Israel	2016	1	0
Taiwan	2022	1	0	Turkey	2001	1	0
Korea	2022	1	0
Africa (0.5)	Period	Cases	Fatalities	Oceania (0.9)	Period	Cases	Fatalities
Nigeria	2023	1	0	Australia	2001–2021	2	1
Region/ Country	Period	Cases	Fatalities
n.d. (0.9)	1968–1973	2	2

n.d., Unknown

Details of cases of accidental and deliberate intoxication can be found in eTable 4. All analyses below refer explicitly to the 134 cases in which the intoxicated person was found alive.

eTable 4. Details on accidental and deliberate caffeine intoxication.

Details	Cases
No details or specific data	6
Intoxication in neonate, infant, child	9
Intoxication by bodybuilding/pre-workout supplements	2
Intoxication by energy drinks	2
Intoxication due to intended weight reduction in eating disorder	2
Intoxication due to intended increase of alertness	2
Intoxication in Münchhausen syndrome	1
Intoxication in known borderline syndrome	1
Intoxication in a bodybuilder	1
Intoxication in a candidate military officer	1
Intoxication by guarana extract	1
Intoxication in a student	1
Known history of drug consumption	1
“Antidote” after diazepam intake	1
To counter leg edema during a flight	1
Increased coffee drinking due to emotional stress	1
Desired “high”	1

Symptoms

Data on initial (pre)clinical symptoms and/or laboratory test results were present in 89 cases. The most frequent clinical presentations were agitation (24 of patients), anxiety (16), and tremor (14), while 30 reported nausea and 10 experienced palpitations. Moreover, recurrent vomiting was found in 60 of cases, tachycardia in 87, and tachypnea in 36. Advanced intoxication was accompanied by metabolic acidosis (25), lactatemia (23), hypokalemia (33), and less frequently by respiratory alkalosis (3), hypophosphatemia (8), hyperglycemia (8), and rhabdomyolysis (10). Later in the course, 15 of the 89 persons affected experienced convulsions. More than 50 of the patients with tonic–clonic seizures also had hemodynamically relevant cardiac arrhythmia, in some cases refractory. Altogether, 18 of the 134 patients had convulsions, 13 ventricular tachycardia, 14 ventricular fibrillation, 2 torsades, and 5 pulseless electrical activity.

A closer look at these severe symptoms reveals that the serum concentrations tended to be lower in narrow-complex tachycardia than in wide-complex tachycardia or convulsions. Patients with convulsions/wide-complex tachycardia had higher caffeine doses/serum concentrations than those without (eTable 5). Classification into different types of convulsions was impossible due to lacking or imprecise data.

eTable 5. Reported caffeine dose/serum concentration in cases of intoxication with cardiac arrhythmia and convulsions.

Dose	All	No arrhythmia/convulsions	Narrow-complex tachycardia		Wide-complex tachycardia		Convulsions
Cases	n = 134	n = 74	yes (n = 24)	no (n = 110)	yes (n = 35)	no (n = 99)	yes (n = 24)	no (n = 110)
Dose	n = 89	n = 47	n = 20	n = 69	n = 28	n = 61	n = 14	n = 75
Median [g]	12	6	22	10	216	6	28.1	10
Min–max [g]	0.05–106	0.06–100	0.05–106	0.1–100	2–106	0.1–100	0.5–100	0.1–106
Mean [g]	20.9	12.4	28.6	18.6	33.8	14.9	32.8	18.6
[95 CI] [g]	[15.8; 25.9]	[7.1; 17.7]	[15.8; 42.5]	[13.2; 24]	[23.1; 44.6]	[10; 20]	[18.3; 47.6]	[13.3; 23.9]
	All	No arrhythmia/convulsions	Narrow-complex tachycardia (without wide-complex tachycardia/ convulsions)		Wide-complex tachycardia (without narrow-complex tachycardia/ convulsions)		Convulsions (without narrow- or wide-complex tachycardia)
Cases	n = 134	n = 74	yes (n = 13)		yes (n = 17)		yes (n = 10)
Dose	n = 89	n = 47	n = 10		n = 12		n = 2
Median [g]	12	6	11		20		47.5
Min–max [g]	0.05–106	0.06–100	0.05–60		2–75		45–50
Mean [g]	20.9	12.4	19.9		27.5		47.5
[95 CI] [g]	[15.8; 25.9]	[7.1; 17.7]	[4.45; 35.4]		[12.6; 42.4]		[15.8; 79.3]
Concentration	All	No arrhythmia/convulsions	Narrow-complex tachycardia		Wide-complex tachycardia		Convulsions
Cases	n = 134	n = 74	yes (n = 24)	no(n = 110)	yes (n= 35)	no(n = 99)	yes (n = 24)	no (n = 110)
Concentration known	n = 91	n = 47	n = 16	n = 75	n = 24	n = 67	n = 18	n = 73
Median [mg/L]	128	101	98.6	147	248	101	205	117
Min–max [mg/L]	15.6–1040	15.6 – 1040	45.5–300	15.6–1040	19–574	15.6–1040	29–574	15.6–1040
Mean [mg/L]	177	152	143	185	242	155	247	160
[95 CI] [mg/L]	[145; 210]	[103; 202]	[96.6; 189]	[147; 223]	187–297	116–193	167–328	126–195
	All	No arrhythmia/convulsions	Narrow-complex tachycardia (without wide-complex tachycardia/ convulsions)		Wide-complex tachycardia (without narrow-complex tachycardia/ convulsions)		Convulsions (without narrow- or wide-complex tachycardia)
Cases	n = 134	n = 74	yes (n = 13)		yes (n = 17)		yes (n = 10)
Concentration known	n = 91	n = 47	n = 10		n = 12		n = 10
Median [mg/L]	128	101	94		238		145
Min–max [mg/L]	15.6–1040	15.6–1040	70–254		19–385		29–567
Mean [mg/L]	177	152	117		221		202
[95 CI] [mg/L]	[145; 210]	[103; 202]	[74.7; 160]		[149; 294]		[82.8; 321]

The ECG findings specified in the case reports (asystole, pulseless electrical activity, bigeminus, sinus tachycardia, AV node reentry tachycardia, supraventricular tachycardia, ventricular fibrillation, torsades, ventricular tachycardia, wide-complex tachycardia) were divided into the categories narrow-complex tachycardia, wide-complex tachycardia, or neither. Among the patients with narrow-complex tachycardia, six had wide-complex tachycardia and two had convulsions, while among those with wide-complex tachycardia, six had narrow-complex tachycardia and nine had convulsions. Of those with convulsions, two had narrow-complex tachycardia and nine had wide-complex tachycardia. In three cases there were narrow-/wide-complex tachycardia and convulsions (min, minimal; max, maximal).

Examination of the distribution of intoxication doses/serum concentrations among those with narrow-/wide-complex tachycardia as well as convulsions reveals that widely varying clinical manifestations can occur in persons with similar caffeine doses and serum concentrations. Cases of narrow-complex tachycardia were described with caffeine doses from 0.05 g and caffeine serum concentrations from 46 mg/L; convulsions from 0.5 g and 29 mg/L respectively; and wide-complex tachycardia from 2 g and 19 mg/L. More rarely, patients had neurological symptoms, such as decreased vigilance (11), myoclonus (7), disorientation (5), vertigo (3), hallucinations (2), and amyosthenia (1).

Clinical course

The type and duration of intoxication as well as the (pre)clinical course were extremely variable in the case reports analyzed.

Intrahospital diagnosis

Among the 134 cases, patients with tablet intoxication had a median serum concentration of 141 mg/L and mortality of 27. The figures for powder intoxication were 19 mg/L and 12. In the cases with caffeine from an unknown source, the median serum concentration was 195 mg/L and the mortality rate for the cases analyzed was 15. Data on caffeine dose were missing in 45 cases (33.6), on the highest measured serum concentration in 43 cases (32.1), and on both in 10 cases (7.5).

Treatment and prognosis

Contact with a poison control center was reported in 3.7 of cases. To prevent further caffeine resorption, activated charcoal was administered in 21.6 of cases and stomach pumping was carried out in 11.2, usually within 2 hours after ingestion. In three cases (2.2) from the 1980s, ipecacuanha syrup was given.

Persons in whom wide-complex tachycardia occurred without accompanying convulsions survived in 83 of the cases analyzed. For those who had convulsions unaccompanied by cardiac arrhythmia, the survival rate was only 44. Combined cardiac arrhythmia and convulsions were survived by a surprising 86.7 of patients; this may have been associated with more intensive treatment or may be explained by publication bias. Supraventricular cardiac arrhythmia was treated primarily with beta blockers, less often with digoxin, verapamil, adenosine, amiodarone, or benzodiazepines. Amiodarone and lidocaine were the drugs used primarily to treat ventricular cardiac arrhythmia. Some 6.7 of patients underwent extracorporeal membrane oxygenation (ECMO) due to refractory cardiac arrhythmia. The survival rate in the small number of cases with ECMO was 78.

Among the 28 persons found alive, the stated causes of death were 5 × cardiac/cardiocirculatory arrest, 3 × asystole, 2 × ventricular fibrillation, 4 × shock or cardiac/cardiocirculatory failure, 3 × unsuccessful cardiopulmonary resuscitation, 1 × pneumonia, 1 × pulmonary edema, and 1 × cerebral edema with incarceration. In eight cases the cause of death was not specified.

In six cases a 20 lipid solution was infused to stabilize the hemodynamics and heartbeat. Five of these six persons had first been given active charcoal or had their stomach pumped, and the sixth had been treated with high-dose insulin. The assumed reason for the successful outcome of lipid emulsion treatment is caffeine dose reduction in the heart and brain. The precise mechanism is still unknown, but the most likely candidates are thought to be the (scavenging) shuttle effect and the lipid sink theory (e57).

Extracorporeal blood cleansing was carried out in 28 of cases, being used in 38 of patients with convulsions, 33 of those with narrow-complex tachycardia, and 54 of those with wide-complex tachycardia. It therefore seems that wide-complex tachycardia more frequently led to intensification of treatment by means of eliminatory procedures. The stated indications were a potentially fatal dose of caffeine, acute kidney injury, rhabdomyolysis, acidosis, electrolyte imbalance, volume overload, neurological factors, and (refractory) instability of hemodynamics and cardiac rhythm. Amelioration of the symptoms and improvements in metabolism, hemodynamics, and stability of heart rhythm were described, beginning about 0.5 to 4 hours after the start of the blood cleansing procedure.

The highest survived caffeine serum concentration was 368mg/L without and 574mg/L with blood cleansing. Compared with those without blood cleansing, patients with blood cleansing had a higher median caffeine dose (29 g versus 6 g), a higher median initial serum concentration (185 mg/L versus 121 mg/L), and a higher survival rate (95 versus 73). According to the case reports, survivors with blood cleansing had higher caffeine doses/initial caffeine serum concentration than those without blood cleansing. There were no indications of potential differences with regard to intoxication and clinical manifestations between those with blood cleansing who survived and those without blood cleansing who died.

The intoxication dose was less often stated or known in those without blood cleansing who died than in survivors with/without blood cleansing (39 versus 67/86, eTable 6). There was no comparison of the various blood cleansing methods due to the low case numbers and the lack of adjustability of the data by disease severity. A synopsis of all treatment procedures can be found in eTable 7.

eTable 6. Reported intoxication dose/serum concentration in those who did and did not survive, with and without blood cleansing.

	All	Survived		Died
Cases	134	HD/HP/HDF yes (n = 36)	HD/HP/HDF no (n = 70)	HD/HP/HDFyes (n = 2)	HD/HP/HDF no (n = 26)
Dose known	89	31 (86 )	47 (67 )	1 (50 )	10 (39 )
Median [g]	12	30	5	20	19
Min–max [g]	0.05–106	4–106	0.05–70	–	3–100
Mean [g]	20.9	32.9	11.3	20	28.5
[95 CI] [g]	[15.8; 25.9]	[23.4; 42.4]	[6.4; 16.2]	–	[7.3; 49.7]
	All	Survived		Died
Cases	134	HD/HP/HDF yes (n = 36)	HD/HP/HDF no (n = 70)	HD/HP/HDFyes (n = 2)	HD/HP/HDF no (n = 26)
Serum concentration known	91	28 (78 )	41 (59 )	0	22 (85 )
Median [mg/L]	128	185	97	–	160
Min–max [mg/L]	15.6 –1040	45.5–574	15.6–368		19–1 040
Mean [mg/L]	178	205	122	–	147
[95 CI] [mg/L]	[145; 210]	[152; 257]	[93.9; 149]	–	[146; 349]

HD, Hemodialysis; HP, hemoperfusion; HDF, hemodiafiltration; max, maximal; min, minimal; 95 CI, 95 confidence interval

eTable 7. Synopsis of the symptomatic and causal measures and procedures used in the cases reported.

Symptomatic treatment of caffeine intoxication in the case reports
Organ	Symptom	Treatment goal	Medication	Procedures
	• Tachycardia, bradycardia, AV block, asystole, PEA, SVT, VT, VF • Hypertension • Hypotension, cardiogenic shock	• Rate control • Rhythm control • Circulatory stabilization	• Beta blockers (propranolol, metoprolol, esmolol, landiolol, labetolol), digoxin, atropine, amiodarone, lidocaine, procainamide, magnesium • Epinephrine, noradrenaline, terlipressin, vasopressin, dopamine, dobutamine, levosimendan	• (Invasive) monitoring • Carotid massage, electrocardioversion • Cardiopulmonary resuscitation, defibrillation • VA-ECMO, target temperature management
	• Hyperventilation, respiratory insufficiency • ARDS	• Improvement of oxygenation and ventilation • Normalization of respiratory rate	• BenzodiazepinesAntibiosis	• Intubation/tracheotomy, mechanical ventilation, ECMO
	• Nausea, vomiting • Diarrhea, abdominal pain	• Antiemetic • Reduction of gastric acid • Analgesia	• Ondansetron, trimethobenzamide • Prokinetics (metoclopramide) • Antacids (algedrate, magnesium hydroxide), proton pump inhibitors (lansoprazole, pantoprazole, omeprazole), antihistamines (cimetidine, ranitidine), sucralfate • Paracetamol	• Stomach tube
	• Nervousness, anxiety, insomnia, tremor, psychosis, myoclonus, hallucinations, epileptic seizures/convulsions	• Antipsychotic • Anticonvulsive • Sedation, anesthesia • Relaxation • Anxiolysis	• Benzodiazepines (diazepam, lorazepam, midazolam) • Sedatives (doxepin, dexmedetomidine) • Anesthetics (propofol), fentanyl • Anticonvulsives (phenytoin, valproate, phenobarbital) • Muscle relaxants (rocuronium, vecuronium) • Antipsychotics (haloperidol)
	• Weakness rigidity, rhabdomyolysis	–	–	–
	• Polyuria, acute kidney injury, • Crush kidney	• Normuria	• Volume substitution	• Balancing • Renal replacement therapy
	• Hypokalemia/-natremia, metabolic acidosis, respiratory alkalosis, lactatemia, hyperglycemia	• Normalization of electrolytes and metabolism	• Electrolyte substitution • Sodium bicarbonate • Insulin	• Blood gas analysis
Causal treatment of caffeine intoxication in the case reports
Caffeine		Treatment goal	Medication	Procedures
		• Prevention of further resorption	• Active charcoal • Laxative • Emetics (ipecacuanha syrup)	• Stomach pumping
		• Lowering of serum concentration	• Lipid emulsion	• Renal replacement therapy, exchange transfusion

For suggestions on treatment of caffeine intoxication, see also (e147, e148).

ARDS, Acute respiratory distress syndrome; AV, atrioventricular; MI, myocardial ischemia; PEA, pulseless electrical activity; SVT, supraventricular tachycardia; VA-ECMO, veno-arterial extracorporeal membrane oxygenation; VF, ventricular fibrillation, VT, ventricular tachycardia

According to the case reports, survival can be achieved by means of timely antiresorptive and eliminatory therapy and symptomatic treatment despite high intoxication doses of up to 100 g and initial serum concentrations as high as 574 mg/L. In individual cases, however, patients died in spite of these measures.

Discussion

Caffeine intoxication—especially with suicidal intent—can be an extremely challenging clinical entity, an entity for which awareness of the problem, preventive measures, as well as medical structures for diagnosis and treatment are still lacking. The latter is due in part to the lack of reliable data and the absence of recommendations and guidelines.

The published systematic reviews concern themselves primarily with pharmacokinetics and the health-related consequences of caffeine consumption (13, 14, 20–22, 37, 38); only individual case reports and case series are available on the diagnosis and treatment of acute intoxication, particularly with (supra)lethal doses of caffeine.

Overall, the available data are very limited and also extremely heterogeneous, because only multiple individual case reports have been published. However, we were able to show that the clinical symptoms are particularly relevant for estimation of the severity of intoxication. Convulsions and wide-complex tachycardia are associated with poorer outcome and can occasionally even occur at low double-digit serum concentrations. Moreover, our analysis suggests that antiresorptive measures may be associated with a survival advantage and that extracorporeal blood cleansing represents an established, safe, and efficient method for caffeine elimination, with a high survival rate even in cases of severe intoxication. This review is the largest analysis to date of case reports on acute caffeine intoxication for evaluation of treatment.

Diagnosis and treatment were often hampered by difficulty in obtaining the patient’s history and a lack of means to determine the caffeine concentration. A potential solution to this problem has been provided by Morita et al., who showed in a group of 25 previously healthy persons with known caffeine intoxication that the serum concentration of caffeine could be estimated by semiquantitative demonstration of glucose and ketone bodies on urine test strips (39).

We attempted to correlate the occurrence of cardiac arrhythmia and convulsions with the dose and serum concentration of caffeine. While narrow-complex tachycardia occurred at a median concentration of 99 mg/L, the corresponding figures for convulsions and wide-complex tachycardia were 205 mg/L and 248 mg/L respectively (eTable 5). It was not possible to analyze the extent to which errors in measurement (technique), previous illness, combined intoxications, or features of the course taken by serum concentration affected disease progression and treatment response. It must also be borne in mind that the maximum concentration measured does not necessarily represent peak serum caffeine. To date, caffeine resorption studies have not been conducted with (supra)lethal doses. The data on resorption come exclusively from case reports. It is conceivable that the intoxication itself influences further resorption, e.g., by retropulsive peristalsis.

Because convulsions and wide-complex tachycardia are associated with poorer outcome (40, e143), it is advisable to prevent the intoxication progressing to this point or, failing that, to take measures to eliminate the caffeine. Intralipid administration and blood cleansing have proved to be the most effective ways of eliminating high concentrations. With regard to the intravenous infusion of lipid emulsions, it should be noted that these can falsify laboratory test results (e144) and affect clearance via hemodialysis filters (e145).

The prevalence, incidence, and mortality of intoxication with caffeine, alone or in combination with other substances, in the general population remain to be determined. On the basis of the case reports and the characteristics of caffeine, possible risk groups are very young persons (neonates, infants, children), athletes (pre-workout supplements), persons who are under high pressure at work, need to concentrate, or suffer from lack of sleep (military personnel, students), persons with mental illness (e.g., depression), and those who attend events where there is an elevated risk of combined intoxication (e.g., clubbers). Persons who consume energy drinks, particularly those with high caffeine content (e146), may also be at risk.

Conclusion

Increasing numbers of case reports on caffeine intoxication are being published, with particular reference to suicidal and accidental intoxications and intoxication of unknown cause. To date, no evidence-based treatment guidelines exist.

As well as symptomatic treatment and antiresorptive measures, eliminatory procedures should be considered in patients with potentially fatal caffeine intoxication doses or serum concentrations, especially in those with aggravated clinical manifestations such as convulsions and clinically relevant cardiac arrhythmia.

Questions on the article from issue 19/2025:

Caffeine Intoxication

The submission deadline is 18 September 2026.

Only one answer is possible per question. Please select the answer that is most appropriate.

Question 1

Which of the following chemical descriptions of caffeine is given in the article?

1. A high-molecular-weight organic substance, completely water-soluble, low plasma protein binding

2. A high-molecular-weight anorganic substance, completely water-soluble, lipophobic

3. A low-molecular-weight anorganic substance, poorly water-soluble, high plasma protein binding

4. A low-molecular-weight anorganic substance, poorly water-soluble, highly lipophilic

5. A low-molecular-weight organic substance, moderately water-soluble, low plasma protein binding

Question 2

Which of the following statements regarding the dose-dependent action of caffeine is the most accurate?

1. At around 25 mg, caffeine has a euphoric and performance-enhancing effect.

2. At around 50 mg, caffeine has a depressive and performance-lowering effect.

3. At around 250 mg, caffeine causes palpitations and nervousness.

4. At around 250 mg, caffeine has a euphoric and performance-enhancing effect.

5. At around 500 mg, caffeine has a euphoric and performance-enhancing effect.

Question 3

From which of the following serum concentrations of caffeine can intoxication symptoms occur?

1. > 0.15 mg/L

2. > 5 mg/L

3. > 15 mg/L

4. > 80 mg/L

5. > 120 mg/L

Question 4

The article analyzes 142 cases of caffeine intoxication. Which of the following reasons is named as the most common, at 40?

1. Suicidal

2. Accidental

3. Treatment error

4. Child abuse

5. Trial of courage

Question 5

Which of the following treatments has been used in cases of caffeine intoxication to reduce uptake, eliminate caffeine, or ameliorate the symptoms??

1. Hemodialysis

2. Administration of active charcoal

3. Administration of beta blockers

4. Administration of ethylenediaminetetraacetic acid (EDTA)

5. Infusion of a lipid solution

Question 6

Which of the following factors decreases the half-life of caffeine?

1. Pregnancy

2. Alcohol

3. Grapefruit juice

4. Liver disease

5. Proton pump inhibitors

Question 7

What pharmacological action of caffeine occurs at a concentration of only around 8 mg/L?

1. An antagonistic effect at GABA-A receptors

2. Inhibition of phosphodiesterases

3. An antagonistic effect at adenosine receptors A1 and A2

4. An antagonistic effect at GABA-B receptors

5. Inhibition of β-adrenoreceptors

Question 8

What maximal daily dose of caffeine is viewed as harmless in the safety assessment issued by the European Food Safety Authority (EFSA) in 2015?

1. 750 mg/d for adults, 400 mg/d for pregnant and breastfeeding women

2. 500 mg/d for adults, 350 mg/d for pregnant and breastfeeding women

3. 400 mg/d for adults, 200 mg/d for pregnant and breastfeeding women

4. 200 mg/d for adults, 100 mg/d for pregnant and breastfeeding women

5. 100 mg/d for adults, 50 mg/d for pregnant and breastfeeding women

Question 9

In which of the following plants is caffeine not a natural component?

1. Cacao tree (Theobroma cacao)

2. Rooibos (Aspalathus linearis)

3. Guarana (Paullinia cupana)

4. Maté (Ilex paraguariensis)

5. Cola tree (Cola acuminata)

Question 10

What is the average caffeine content of a cup of filter coffee (250 mL)?

1. Approximately 15 mg

2. Approximately 40 mg

3. Approximately 85 mg

4. Approximately 250 mg

5. Approximately 330 mg