Full text of DOI:10.1592/phco.21.7.764.34567

Fatal Hemolysis After High-Dose Etoposide:
Is Benzyl Alcohol to Blame?
Adrienne L. Smith, Pharm.D., Kamal Haider, M.D., Jill M. Schachtner, Pharm.D., Sharad Mathur, M.D.,
Ray VanOrden, B.S, and Teresa C. Gentile, M.D.
A 53-year-old African-American man with relapsed non-Hodgkin’s lymphoma
developed seizures and respiratory arrest 2 hours after an infusion of high-
dose etoposide in preparation for an autologous bone marrow transplant.
Laboratory tests revealed both rapid hemolysis and severe metabolic acidosis.
The patient died the following day. Based on toxicities observed, we suspect
that our patient possessed an ethnic polymorphism of the enzyme alcohol
dehydrogenase. Further research is required to determine the relationship
between the benzyl alcohol metabolic rate and toxicity and genetic
polymorphisms of alcohol dehydrogenase in African-Americans.
(Pharmacotherapy 2001;21(6):764–766)
A 53-year-old African-American man with
refractory non-Hodgkin’s lymphoma was
admitted to our hospital for an autologous bone
marrow transplant, with a conditioning regimen
of total body irradiation and high-dose
chemotherapy with etoposide and cyclophos-
phamide. The patient had a history of chronic
renal insufficiency with a baseline calculated
clearance of 60 ml/minute, and prior
chemotherapy induced tumor lysis. He had no
known drug allergies or history of seizures. He
received total body irradiation on days -8
through -5 of his regimen and tolerated it well,
with only minor complaints of fatigue and
xerostomia. Chemotherapy was started on day
-4 with etoposide 3 g/m2 (body surface area 2.02
m2) for a total dosage of 6060 mg intravenously
over 4 hours. The patient also received
approximately 9000 mg of benzyl alcohol (90
mg/kg), which is used as a preservative in
intravenous etoposide preparations. Before the
etoposide infusion, the patient’s vital signs were
stable: blood pressure 154/71 mm Hg, respiratory
rate 22 breaths/minute, temperature 37.6°C, and
oxygen saturation 99% on room air. The man
was slightly tachycardic with a pulse of 112
beats/minute. Laboratory tests revealed
hemoglobin 8.9 g/dl, hematocrit 25.8%, aspartate
aminotransferase 47 U/L, alanine amino-
transferase 25 U/L, lactate dehydrogenase 691
U/L, creatinine 1.9 mg/dl, and bilirubin 1.5
mg/dl.
Approximately 2 hours after the infusion was
completed, the patient seized, suffered
respiratory arrest, and was intubated. Three and
one-half hours after its completion, laboratory
tests revealed decreased hemoglobin and other
results suggestive of hemolysis (Figure 1).
Arterial blood gas showed pH 7.12, partial
pressure of carbon dioxide 27, bicarbonate less
than 10, and anion gap greater than 22,
implicating anion gap metabolic acidosis.
Glucose-6-phosphate dehydrogenase deficiency
was negative. The patient was pronounced dead
the following day.
From the Departments of Pharmacy (Drs. Smith and
Schachtner), Medicine (Drs. Gentile and Haider), and
Clinical Pathology (Dr. Mathur), State University of New
York; and the Onondaga County Center for Forensic
Sciences (Mr. VanOrden), Syracuse, New York.
Address reprint requests to Adrienne L. Smith, Pharm.D.,
Department of Pharmacy, University Hospital, 750 East
Adams Street, Syracuse, NY 13210.
Discussion
No cases of high-dose etoposide causing
massive hemolysis and metabolic acidosis are
reported in the literature, but we found case

FATAL HEMOLYSIS AFTER HIGH-DOSE ETOPOSIDE Smith et al
765
reports implicating benzyl alcohol, an aromatic
alcohol present in high concentrations in
intravenous etoposide.
antibacterial action appears to be much lower
than the concentration that would be toxic to
adults.1, 4
One study investigated the safety and tolerance
of an intravenous methylprednisolone sodium
succinate formulation containing 0.9% benzyl
alcohol and found low toxicity in adult humans.5
Twenty-four healthy men were divided into three
groups: one group received a single dose of
methylprednisolone sodium succinate 2 g
preserved with benzyl alcohol (~290 mg); one
group received the same methylprednisolone
dose, but preserved with parabens instead; the
third group received saline placebo. There were
no clinically significant differences in vital signs,
or electrocardiogram, electroencephalogram, or
laboratory parameters.
The most frequently reported toxicities of
benzyl alcohol involve adverse neurologic effects
with intrathecal administration in adults and
low-birthweight newborns exposed to parenteral
solutions in which benzyl alcohol is used as a
preservative or diluent. In 1982, the United
States Food and Drug Administration issued a
warning against using agents or reconstituted
solutions containing benzyl alcohol to treat
newborns.6
Several studies evaluated the relationship
between products containing benzyl alcohol and
morbidity and mortality in neonatal intensive
care units. One found fewer neurologic
impairments in low-birthweight infants, as well
as significant decreases in the frequency of
cerebral palsy, after discontinuing the use of
solutions containing benzyl alcohol.7 In another
study, benzyl alcohol toxicity was shown to be a
major cause of morbidity (intraventricular
hemorrhage and metabolic acidosis) and
mortality in newborns weighing less than 1000 g
at birth.8
Elimination of benzyl alcohol occurs through
oxidation to benzoic acid, catalyzed principally
by alcohol dehydrogenase (ADH).1, 2 Alcohol
dehydrogenase 2 (ADH2) has 3 allelic variants—
ADH2*1, ADH2*2, ADH2*3—that exhibit ethnic
polymorphism3 and different Km values (the
Michaelis-Menten constant that measures affinity
of an enzyme for a drug), leading to variable rates
of metabolism. Specifically, the ADH2*3 allelic
variant has a high Km for alcohol and is expressed
to a 20% degree in African-Americans.3
Our patient may have possessed the ADH2*3
allele, predisposing him to decreased clearance of
benzyl alcohol, resulting in increased serum
concentrations, which have been associated with
increased toxicity. The International Bone
Marrow Transplant Registry (IBMTR) listed 297
African-American patients who underwent bone
marrow transplantation with etoposide
conditioning regimens between January 1989 and
April 2000. (Data were obtained from the
statistical center of the IBMTR and the
Autologous Blood and Marrow Transplant
Registry [ABMTR]. Analysis has not been
reviewed or approved by the advisory committee
of the IBMTR / ABMTR.)
At concentrations of 0.9–2.0%, benzyl alcohol
is commonly employed as an antibacterial agent
in a variety of pharmaceutical formulations
intended for intravenous administration. It is
rapidly oxidized to benzoic acid, which is
conjugated with glycine to form hippuric acid
and excreted in the urine as glucuronide.1
Although benzyl alcohol toxicity is recognized,
the concentration that is necessary for
A fatal, multiple-organ deterioration referred to
as gasping syndrome in newborns who receive
multiple injections of bacteriostatic normal saline
and drugs containing 0.9% benzyl alcohol was
described in one study.9 After receiving an
average daily quantity of benzyl alcohol 99–234
mg/kg, the newborns followed a typical course:
gradual neurologic deterioration, severe
metabolic acidosis, the striking onset of gasping
respiration, hematologic abnormalities, skin
breakdown, hepatic and renal failure, hypo-
tension, and cardiovascular collapse. The
investigators attributed the benzyl alcohol
toxicity in these newborns to the large doses
administered relative to the size of the patients,
30
25
20
Hgb (g/dl)
15
Hct (%)
10
Bili (mg/dl)
LD x 10-2 (U/L)
5
0
Sept. 16
8 A.M.
Sept. 20
3 A.M.
Sept. 20
Etoposide
infusion
Sept. 20
5:28 P.M.
Sept. 21
3:41 A.M.
12–4 P.M.
Figure 1. Laboratory results suggestive of hemolysis. Hgb
= hemoglobin; Hct = hematocrit; Bili = bilirubin; LD =
lactate dehydrogenase.

766
PHARMACOTHERAPY Volume 21, Number 6, 2001
damaging action.2
the reduced ability of the infants’ metabolic
systems to detoxify benzyl alcohol, or both.
The oxidative processes and conjugation
reactions required to metabolize foreign
compounds such as benzyl alcohol are lower in
premature infants than in full-term infants.7 The
neurologic symptoms (hypoactivity, coma,
seizure activity) seen in these newborns may
have been due to a direct toxic effect of benzyl
alcohol on the central nervous system.
Benzyl alcohol poisoning was reported with
intravenous administration of diazepam in a 5-
year-old child with encephalitis-induced coma
and status epilepticus receiving approximately
180 mg/kg/day of benzyl alcohol.10 The patient
developed hypotension, hypernatremia, and
severe metabolic acidosis 36 hours after a
continuous infusion of diazepam was started.
The dosage was reduced, but the patient died 10
days later.
Conclusion
This is the first report that we are aware of
describing massive hemolysis after an infusion of
high-dose etoposide. We suspect our patient may
have possessed the ADH2*3 allele, leading to
benzyl alcohol-induced hemolysis and metabolic
acidosis. Further investigation is required to
determine the relationship between benzyl
alcohol metabolic rate and toxicity, and genetic
polymorphisms of alcohol dehydrogenase in the
African-American population.
Acknowledgments
We thank David Lehmann, M.D., Pharm.D., and
Timothy P. Rohrig, Ph.D., DABFT, for their valuable
advice and review of the manuscript.
Hemolytic effects caused by benzyl alcohol
have been suggested.11 A 1978 experiment on
the interaction of benzyl alcohol with human
erythrocytes found hemolysis induced by benzyl
alcohol to be time, dose, and temperature
dependent.2 At 37°C, benzyl alcohol-induced
hemolysis of human erythrocytes was relatively
steady at a benzyl alcohol concentration of 100
mM. Little or no hemolysis occurred at 4°C at
References
1. Baselt RC, Cravey RH, eds. Disposition of toxic drugs and
chemicals in man, 3rd ed. Littleton, MA: Yearbook Medical
Publishers, 1989:85–6.
2. Ohmiya Y, Nakai K. Interaction of benzyl alcohol with human
erythrocytes. Japan J Pharmacol 1978;28:367–73.
3. Chen CC, Lu RB, Chen YC, et al. Interaction between the
functional polymorphisms of the alcohol-metabolism genes in
protection against alcoholism. Am
1999;65:795–807.
4. Kimura ET, Darby TD, Krause RA, Brondyk HD. Parenteral
toxicity studies with benzyl alcohol. Toxicol Appl Pharmacol
1971;18:60–8.
J Hum Genet
the same alcohol concentration.
The
concentration required for 50% hemolysis at
37°C after 60 minutes was approximately 100
mM. Radiotracer studies revealed that, in
contrast to the rapid cellular uptake, which was
independent of temperature, the binding of
benzyl alcohol to membranes increased gradually
with time and was dependent mainly on
temperature. The critical hemolytic levels of the
alcohol bound to membranes was estimated to be
500 nmol/mg protein. The authors suggest that
hemolysis induced by benzyl alcohol depends on
the alcohol’s binding to erythrocyte membranes,
and that, since benzyl alcohol is metabolized
principally in the liver by alcohol dehydrogenase,
which is absent in erythrocytes, unchanged
benzyl alcohol must exert the membrane-
5. Novak E, Stubbs SS, Sanborn EC, Eustice RM. The tolerance
and safety of intravenously administered benzyl alcohol
methylprednisolone sodium succinate formulations in normal
human subjects. Toxicol Appl Pharmacol 1972;23:54–61.
6. Jackson D. Doxapram and potential benzyl alcohol toxicity: a
moratorium on clinical investigation? [letter] Pediatrics
1986;78:541.
7. Benda GI, Jiller JL, Reynolds JW. Benzyl alcohol toxicity:
impact on neurologic handicaps among surviving very low
birth weight infants. Pediatrics 1986;77:507–12.
8. Menon PA, Thach BT, Smith CH, et al. Benzyl alcohol toxicity
in a neonatal intensive care unit: incidence, symptomatology,
and mortality. Am J Perinatol 1984;1:288–92.
9. Gershanik J, Boecler B, Ensley H, et al. The gasping syndrome
and benzyl alcohol poisoning.
1982;307(22):1384–8.
N
Engl
J
Med
10. Lopez-Herce J. Benzyl alcohol poisoning following diazepam
intravenous infusion [letter]. Ann Pharmacother 1995;29:632.
11. McOrmond P, Guick B, Duggan HE. Hemolytic effect of benzyl
alcohol [letter]. Drug Intell Clin Pharm 1980;14:549.