1065

Cerebral Infarction Immediately After Ingestion
of Hydrogen Peroxide Solution
Scott J. Sherman, MD, PhD; Leslie V. Boyer, MD; William A. Sibley, MD

Downloaded from http://ahajournals.org by on April 11, 2024

Background We report the clinical and neuroimaging findings of a patient who sustained multiple cerebral infarcts after
the ingestion of concentrated hydrogen peroxide solution sold
as a "health food" product.
Case Description An 84-year-old man sustained focal neurological deficits immediately after ingesting 30 mL of 35%
hydrogen peroxide solution. Physical examination disclosed a
left hemiparesis, frontal release signs, and cerebellar dysfunctions. Magnetic resonance imaging revealed multiple cerebral

and cerebellar infarctions in the anterior, middle, and posterior vascular territories.
Conclusions The likely mechanism of pathogenesis involves
cerebral oxygen gas embolization. The use of hyperbaric
therapy should be considered in treating hydrogen peroxide
poisoning. (Stroke. 1994^25:1065-1067.)

C

sensation in the mouth but had no recall of subsequent
events.
When paramedics arrived on the scene, they found
the patient with his eyes open but unresponsive to
verbal command. There was foamy material around the
mouth and nose. Initial vital signs in the field were as
follows: blood pressure was 186/90 mm Hg, pulse was 94
beats per minute and regular, and respiratory rate was
20 per minute. His color was pale but not cyanotic. He
had a vigorous gag reflex.
In the emergency department, the patient was noted
to have a decreased level of consciousness (Glasgow
Coma Scale score of 11). There was purposeful movement of the right arm. A dense left hemiparesis was
noted along with flattening of the left nasolabial fold. A
left Babinski was present.
A computed tomographic scan of the brain without
contrast showed only diffuse atrophy (Figure, panel
1A). Arterial blood gas drawn while the patient received
40% supplemental oxygen by mask showed the following: pH, 7.38; Pcoj, 38; and POj, 176. The chest
roentgenogram revealed a left lower lobe consolidation
and a small left pleural effusion, which resolved by the
next day.
On the fifth hospital day, a cranial magnetic resonance imaging scan showed multiple, bilateral areas of
infarction (Figure, panels IB and 1C). On the right
there were distinct areas of infarction in the frontal,
posterior parietal, and occipital regions. There were
multiple right cerebellar infarcts. On the left there was
a single frontal infarct. The absence of a flow void in the
left supraclinoid carotid artery and findings of a limited
magnetic resonance angiogram were consistent with left
carotid artery occlusion. This finding was confirmed by
Doppler ultrasound studies.
The patient showed initial improvement in the level of
consciousness during the first 5 days; however, severe
neurological deficits remained. These deficits showed
good clinical correlation with the magnetic resonance
findings (Figure, panels IB and 1C). At the time of
discharge on the 14th hospital day, he was able to answer

oncentrated hydrogen peroxide (35%) is a commercially available oxidant with widespread
applications in the chemical and manufacturing industries. Recently this concentrated form of
hydrogen peroxide has been sold as "Food Grade
Hydrogen Peroxide for Hyper-Oxygenation Therapy"
by merchants claiming sundry health benefits.1 Although instructions for dilution to nontoxic concentrations may accompany the sale of the product, no clear
warning of the potential for serious morbidity or mortality is given. We now report a case of extensive
neurological damage immediately subsequent to the
purposeful ingestion of 35% hydrogen peroxide taken as
a health remedy. We propose that the likely mechanism
of tissue damage in the central nervous system involves
oxygen gas embolization of the cerebral vasculature.

Case Report
An 84-year-old right-handed man ingested approximately 30 mL of 35% hydrogen peroxide after diluting it
in a small volume (100 to 300 mL) of water. The patient
purchased the hydrogen peroxide after being told by an
acquaintance that it would be a beneficial remedy for
arthritis. Although the patient did have extensive preexisting cerebrovascular disease and had undergone left
carotid endarterectomy in the remote past, he had never
had a stroke. By history, his premorbid functioning was
normal.
Within 3 minutes after the ingestion, the patient
slumped over and became unresponsive. After awakening, the patient remembered mixing the hydrogen peroxide and swallowing the solution. He recalled a "fizzy"
Received September 22, 1993; final revision received January
24, 1994; accepted January 31, 1994.
From the Departments of Neurology (SJ.S., W.A.S.) and
Surgery (L.V.B.), University of Arizona, Health Sciences Center,
Tucson, Ariz.
Correspondence to Scott J. Sherman, MD, PhD, Department of
Neurology, University of Arizona, Health Sciences Center, Tucson, AZ 85724.

Key Words • cerebral infarction • embolism • hydrogen
peroxide • hyperbaric oxygenation

1066

Stroke

Vol 25, No 5 May 1994

Axial T,-weighted (1B, 2B, 3B) and T r weighted
(1C, 2C, 3C) magnetic resonance imaging sequences were obtained on trie fifth hospital
day. These are compared with analogous images (1A, 2A, 3A) of a computed tomographlc
scan obtained on admission, which serve as a
baseline study. The initial computed tomographic scan was obtained too soon to reflect
acute ischemic changes but does effectively
rule out the possibility that the cortical infarcts
were present before the ingestion of hydrogen
peroxide. Acute Infarcts (arrows) are present in
the territories of the posterior (1B, 1C), middle
(2B, 2C), and anterior (3B, 3C) cerebral arteries. Additional areas of infarction (not shown in
these sequences) were present, as discussed
in the text.

Downloaded from http://ahajournals.org by on April 11, 2024

simple questions appropriately. The left arm and leg
continued to show a paresis consistent with an upper
motor neuron lesion. There was a prominent grasp reflex
on the right. There were signs of cerebellar dysfunction on
the right as manifested by ataxia of the arm and inability
to perform rapid alternating movements.

Discussion
This case report illustrates an unusual toxicity of a
commercially available compound. Thirty milliliters of
35% hydrogen peroxide when fully decomposed into
reaction products will produce 3.5 L of oxygen gas.2
Multiple cerebral infarctions occurring immediately after the ingestion of this solution were probably due to
oxygen gas embolism.
Several pathophysiological mechanisms could ultimately lead to cerebral damage after ingestion of hydrogen peroxide. The initial foaming reaction produced
as the hydrogen peroxide is ingested could lead to
laryngospasm, bronchospasm, and pulmonary aspiration. The ensuing hypoxemia could lead to central
nervous system damage directly or could cause cardiopulmonary arrest with subsequent cerebral anoxia. This
mechanism is likely to have occurred in at least two of
the previous case reports.3-4

Undissociated hydrogen peroxide remaining in the
stomach would be readily absorbed through the mucosa
and enter into the portal venous system. Once the
hydrogen peroxide reached the mesenteric blood, it
would begin to undergo rapid gaseous dissociation
because of the high concentration of the enzyme catalase present in erythrocytes.5 The presence of gas in the
portal venous system has been documented after hydrogen peroxide ingestion.6 Embolization could then proceed to involve the vena cava, right heart chambers, and
the pulmonary arterial tree. One clinical report demonstrated the presence of massive amounts of gas in the
right side of the heart after ingestion of hydrogen
peroxide.7
Arterial gas embolization may occur by at least three
mechanisms. Gas bubbles that have formed in the venous
system can travel from the right atrium to the left atrium
via a patent foramen ovale. Some of the hydrogen peroxide dissolved in venous blood could remain undissociated
until crossing the pulmonary capillary bed and then
undergo catalysis in the arterial circulation. Finally, pulmonary aspiration of hydrogen peroxide could lead to
direct embolization of the arterial system after absorption
from the pulmonary capillaries.
The case presented here is most consistent with
neurological damage due to gas embolization of the

Sherman et al

cerebral vasculature. The patient did not present with
initial circulatory collapse, as described in several previous reports of serious hydrogen peroxide poisoning.3.4,7 Rather, there was rapid onset of focal neurological deficits occurring in the setting of adequate arterial
pressure, pulse rate, and oxygen saturation. As shown in
the Figure, the right hemisphere sustained infarcts in
the territories of the anterior, middle, and posterior
cerebral arteries. In addition, there were left frontal and
multiple right cerebellar infarcts. This pattern of damage is most consistent with arterial gas embolization.
The relative sparing of the left hemisphere is also
consistent with gas embolization. The occluded left
carotid artery may have had a protective effect because
gas bubbles following the ascending blood flow would be
less likely to enter a circuitous collateral circulation to
the left hemisphere.
We believe that this report demonstrates that cerebral gas embolization is a potential consequence of the
ingestion of hydrogen peroxide. Although to date there
have been no reports of the use of hyperbaric therapy in
the treatment of hydrogen peroxide toxicity, it is a
well-accepted treatment for cerebral nitrogen gas em-

Cerebral Infarction From Hydrogen Peroxide

1067

bolization caused by decompression sickness. Furthermore, even delayed hyperbaric therapy has been shown
to reverse neurological deficits in cerebral air embolization.8 We therefore suggest that hyperbaric therapy be
considered in a stabilized patient snowing focal neurological deficits after the ingestion of hydrogen peroxide.

References
1. Dept of Health and Human Services. Hyper-Oxygenation Therapy.
Rockville, Md: US Food and Drug Administration, Federal-States
Relations; 1989.
2. Humberston CL, Dean BS, Krenzelok EP. Ingestion of 35%
hydrogen peroxide. Clin Toxicol. 1990;28:95-100.
3. Zecevic D, Gasparec C. Death caused by hydrogen peroxide. Z
Rechstmed. 1979;84:57-59.
4. Giusti GV. Fatal poisoning with hydrogen peroxide. Forensic Sci.
1973;2:99-100.
5. Beutler E. Red Cell Metabolism: A Manual of Biochemical Methods.
2nd ed. Orlando, Fla: Grune & Stratton, Inc; 1975:89-147.
6. Rackoff WR, Merton DF. Gas embolism after ingestion of hydrogen
peroxide. Pediatrics. 1990;85:593-594.
7. Christensen DW, Faught WE, Black RE, Woodward GA, Otwell
TD. Fatal oxygen embolization after hydrogen peroxide ingestion.
Crit Care Med. 1991;20:543-544.
8. Armon C, Deschamps C, Adkinson C, Fealey RD, Orszulak TA.
Hyperbaric treatment of cerebral air embolism sustained during an
open-heart surgical procedure. Mayo Clin Proc. 1991;66:565-571.

Downloaded from http://ahajournals.org by on April 11, 2024