Emerg Med Clin N Am

25 (2008) 715–739

Critical Care Toxicology

Christopher P. Holstege, MDa,b,c,*,
Stephen G. Dobmeier, BSNc, Laura K. Bechtel, PhDa
a
Division of Medical Toxicology, Department of Emergency Medicine, University of
Virginia School of Medicine, PO Box 800774, 1222 Jefferson Park Avenue,
4th Floor, Charlottesville, VA 22908–0774, USA
b
Department of Pediatrics, University of Virginia School of Medicine, PO Box 800774,
Charlottesville, VA 22908–0774, USA
c
Blue Ridge Poison Center, University of Virginia Health System, Room 4601,
1222 Jefferson Park Avenue, Charlottesville, VA 22903, USA

Critically poisoned patients are commonly encountered in emergency

medicine. Exposure to potential toxins can occur by either accident
(ie, occupational incidents or medication interactions) or intentionally
(ie, substance abuse or intentional overdose). The outcome following a poi-
soning depends on numerous factors, such as the type of substance, the
dose, the time from exposure to presentation to a health care facility, and
the pre-existing health status of the patient. If a poisoning is recognized
early and appropriate supportive care is initiated quickly, most patient out-
comes are favorable. In modern hospitals with access to life support equip-
ment the case fatality rate for self-poisonings is approximately 0.5%, but
this can be as high as 10% to 20% in the developing world lacking critical
care resources [1].
This article introduces the basic concepts for the initial approach to the
critically poisoned patient and the steps required for stabilization. It intro-
duces some key concepts in diagnosing the poisoning, using clinical clues
and ancillary testing (ie, laboratory, ECG, and radiology). Finally, specific
management issues are discussed.

Clinical evaluation
When evaluating a patient who has presented with a potential toxicologic
emergency, the health care practitioner should not limit the differential diag-
nosis. A comatose patient who smells of ethanol may be harboring an

* Corresponding author. PO Box 800774, Charlottesville, VA 22908–0774.

E-mail address: [email protected] (C.P. Holstege).

0733-8627/08/$ - see front matter Ó 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.emc.2008.04.003 emed.theclinics.com
716 HOLSTEGE et al

intracranial hemorrhage; an agitated patient who seems anticholinergic may

actually be encephalopathic from an infectious etiology. Patients must be
thoroughly assessed and appropriately stabilized. Rarely is there a specific
antidote for a poisoned patient; supportive care is the most important
intervention.
All patients presenting with toxicity should be aggressively managed. Poi-
soned patients may seem to be in extremis (ie, brain dead), but most fully
recover. The patient’s airway should be patent and adequate ventilation
ensured. If necessary, endotracheal intubation should be performed. Too
often physicians are lulled into a false sense of security when a patient’s ox-
ygen saturations are adequate on high-flow oxygen. If the patient has either
inadequate ventilation or poor airway protective reflexes, then the patient
may be at risk for subsequent CO2 narcosis with risk for worsening acidosis
and the potential for aspiration. The initial treatment of hypotension for all
poisonings consists of intravenous fluids. Close monitoring of the patient’s
pulmonary status should be performed to ensure that pulmonary edema
does not develop as fluids are infused. The health care providers should
place the patient on continuous cardiac monitoring with pulse oximetry
and make frequent neurologic checks. Glucose should be checked at bedside
in all patients with altered mental status. Poisoned patients should receive
a large-bore peripheral intravenous line and all symptomatic patients should
have a second line placed in either the peripheral or central venous system.
Many toxins can potentially cause seizures. In general, toxin-induced
seizures are treated in a similar fashion to other seizures. Clinicians should
ensure the patient maintains a patent airway and the blood glucose should
be measured. Most toxin-induced seizures are self-limiting. For seizures re-
quiring treatment, the first-line agent should be parenteral benzodiazepines.
If benzodiazepines are not effective at controlling seizures, a second-line
agent, such as phenobarbital, should be used. In rare poisoning cases (ie,
isoniazid) pyridoxine should be administered. In cases of toxin-induced sei-
zures, phenytoin is generally not recommended. It is usually ineffective and
may add to the underlying toxicity of some agents, such as cyclic antidepres-
sants, theophylline, cocaine, and lidocaine [2]. If a poisoned patient requires
intubation, it is important to avoid the use of long-acting paralytic agents
because these agents may mask seizures if they develop.
Rapid recognition of a toxidrome, if present, can help determine whether
a poison is involved in a patient’s condition and can help determine the class
of toxin involved. Toxidromes are the constellation of signs and symptoms
associated with a class of poisons. Table 1 lists selected toxidromes and their
characteristics. Patients may not present with every component of a toxi-
drome and toxidromes can be clouded in mixed ingestions. Certain aspects
of a toxidrome can have great significance. For example, noting dry axilla
may be the only way of differentiating an anticholinergic patient from a sym-
pathomimetic patient, and miosis may distinguish opioid toxicity from
a benzodiazepine overdose. There are several notable exceptions to the
 CRITICAL CARE TOXICOLOGY 717

Table 1
Toxidromes
Toxidrome Clinical effects
Opioid Sedation, miosis, decreased bowel sounds, decreased respirations
Anticholinergic Mydriasis, dry skin, dry mucous membranes, decreased bowel sounds,
sedation, altered mental status, hallucinations, tachycardia, urinary
retention
Sympathomimetic Agitation, mydriasis, tachycardia, hypertension, hyperthermia,
diaphoresis
Cholinergic Miosis, lacrimation, diaphoresis, bronchospasm, bronchorrhea,
vomiting, diarrhea, bradycardia
Serotonin syndrome Altered mental status, tachycardia, hypertension, hyperreflexia,
clonus, hyperthermia

recognized toxidromes. For example, several opioid agents do not cause

miosis (ie, meperidine and tramadol). In most cases, a toxidrome does not
indicate a specific poison, but rather a class of poisons.

Testing in the critically poisoned patient

When evaluating the critically ill poisoned patient, there is no substitute
for a thorough history and physical examination. Numerous television
medical shows depict a universal toxicology screen that automatically deter-
mines the agent causing a patient’s symptoms. Unfortunately, samples can-
not be simply ‘‘sent to the laboratory’’ with the correct diagnosis to a clinical
mystery returning on a computer printout. Clues from a patient’s physical
examination are generally more likely to be helpful than a ‘‘shotgun’’ labo-
ratory approach that involves indiscriminate testing of blood or urine for
multiple agents [3].
When used appropriately, diagnostic tests may be of help in the manage-
ment of the intoxicated patient. When a specific toxin or even class of toxins
is suspected, requesting qualitative or quantitative levels may be appropriate
[4]. In the suicidal patient whose history is generally unreliable or in the
unresponsive patient where no history is available, the clinician may gain
further clues as to the etiology of a poisoning by responsible diagnostic test-
ing. In the intentionally poisoned patient, an acetaminophen level should be
obtained to rule out coexisting toxicity.

Anion gap
A basic metabolic panel should be obtained in all suicidal poisoned pa-
tients. When low serum bicarbonate is discovered on a metabolic panel,
the clinician should determine if an elevated anion gap exists. The formula
most commonly used for the anion gap calculation is [5]
 þ   
Na  Cl þ HCO3
718 HOLSTEGE et al

This equation allows one to determine if serum electroneutrality is being

maintained. The primary cation (sodium) and anions (chloride and bicarbon-
ate) are represented in the equation [6]. There are other contributors to this
equation that are unmeasured [7]. Other serum cations are not commonly in-
cluded in this calculation, because either their concentrations are relatively
low (ie, potassium) or assigning a number to represent their respective contri-
bution is difficult (ie, magnesium, calcium) [7]. Similarly, there is also a mul-
titude of other serum anions (ie, sulfate, phosphate, organic anions) that are
also difficult to measure and quantify in an equation [6,7]. These unmeasured
ions represent the anion gap calculated using the previously mentioned equa-
tion. The normal range for this anion gap is accepted to be 8 to 16 mEq/L [7],
but some have recently suggested that because of changes in the technique for
measuring chloride, the range should be lowered to 6 to 14 mEq/L [6]. An
increase in the anion gap beyond an accepted normal range, accompanied
by a metabolic acidosis, represents an increase in unmeasured endogenous
(ie, lactate) or exogenous (ie, salicylates) anions [5]. A list of the more
common causes of this phenomenon is organized in the classic MUDILES
pneumonic as shown (the ‘‘P’’ has been removed from the older acronym
of MUDPILES, because paraldehyde is no longer available):
Methanol
Uremia
Diabetic ketoacidosis
Iron, inhalants (ie, carbon monoxide, cyanide, toluene), isoniazid, ibuprofen
Lactic acidosis
Ethylene glycol, ethanol ketoacidosis
Salicylates, starvation ketoacidosis, sympathomimetics
It is imperative that clinicians who evaluate poisoned patients initially
presenting with an increased anion gap metabolic acidosis investigate the
etiology of that acidosis. Symptomatic poisoned patients may have an initial
mild metabolic acidosis on presentation because of an elevation of serum
lactate that can be caused by a number of processes before stabilization
(ie, agitation, hypoxia, hypotension). With adequate supportive care includ-
ing hydration and oxygenation, the anion gap acidosis should begin to
resolve. If, despite adequate supportive care, an anion gap metabolic acido-
sis worsens in a poisoned patient, the clinician should consider toxins that
form acidic metabolites (ie, ethylene glycol, methanol) [8]; toxins that them-
selves can worsen the acidosis as absorption increases (ie, ibuprofen) [9]; or
toxins that cause lactic acidosis by interfering with aerobic energy produc-
tion (ie, cyanide or iron) [10].

Osmole gap
The serum osmole gap is a common laboratory test that may be useful
when evaluating poisoned patients. This test is most often discussed in the
 CRITICAL CARE TOXICOLOGY 719

context of evaluating the patient suspected of toxic alcohol (ie, ethylene gly-
col, methanol, and isopropanol) intoxication. Although this test may have
use in such situations, it has many pitfalls and limitations to its effectiveness.
Osmotic concentrations are themselves expressed in both terms of osmo-
lality (milliosmoles per kilogram of solution) and osmolarity (milliosmoles
per liter of solution) [11,12]. This concentration can be measured by use
of an osmometer, a tool that most often uses the technique of freezing point
depression and is expressed in osmolality (OsmM) [13]. A calculated serum
osmolarity (OsmC) may be obtained by any of a number of equations
[14], involving the patient’s glucose, sodium, and urea that contribute to
almost all of the normally measured osmolality [15]. One of the most com-
monly used of these calculations is expressed below:
 
OsmC ¼ 2 Naþ þ ½BUN=2:8 þ ½glucose=18

The correction factors in the equation are based on the relative osmotic
activity of the substance in question [11]. Assuming serum neutrality,
sodium as the predominant serum cation is doubled to account for the cor-
responding anions. Finding the osmolarity contribution of any other osmot-
ically active substances that is reported in milligrams per deciliter (like
blood, urea, nitrogen [BUN] and glucose) is accomplished by dividing by
one tenth its molecular weight in daltons [11]. For BUN this conversion
factor is 2.8 and for glucose it is 18. Similar conversion factors may be added
to this equation to account for ethanol and the various toxic alcohols as
shown below:
 
OsmC ¼ 2 Naþ þ ½BUN=2:8 þ ½glucose=18 þ ½ethanol=4:6
þ ½methanol=3:2 þ ½ethylene glycol=6:2 þ ½isopropanol=6:0

The difference between the measured (OsmM) and calculated (OsmC) is

the osmole gap (OG) and is depicted by the equation below [11]:

OG ¼ OsmM  OsmC

If a significant osmole gap is discovered, the difference in the two values

may represent presence of foreign substances in the blood [13]. Possible
causes of an elevated osmole gap are listed as follows:
Acetone
Ethanol
Ethylene glycol
Isopropanol
Methanol
Propylene glycol
720 HOLSTEGE et al

Unfortunately, what constitutes a normal osmole gap is widely debated.

Traditionally, a normal gap has been defined as less than or equal to
10 mOsm/kg. The original source of this value is an article from Smithline
and Gardner [16] that declares this number as pure convention. Further
clinical study has not shown this assumption to be correct. Glasser and
colleagues [17] studied 56 healthy adults and reported that they found the
normal osmole gap to range from 9 to 5þ mOsm/kg. A study examining
a pediatric emergency department population (N ¼ 192) found a range
from 13.5 to 8.9 [18]. Another study by Aabakken and colleagues [19]
looked at the osmole gaps of 177 patients admitted to their emergency
department and reported their range to be from 10 to 20 mOsm/kg. A vital
point brought forth by the authors of this study is that the day-to-day co-
efficient of variance for their laboratory in regards to sodium was 1%.
They believed this variance translated to a calculated analytic standard
deviation of 9.1 mOsm in regards to osmole gap. This analytic variance
alone may account for the variation found in patient’s osmole gaps. This
concern that even small errors in the measurement of sodium can result in
large variations of the osmole gap has been voiced by other researchers
[18,20]. Overall, the clinician should recognize that there is likely a wide
range of variability in a patient’s baseline osmole gap.
There are several concerns in regard to using the osmole gap as a screening
tool in the evaluation of the potentially toxic-alcohol poisoned patient. The
lack of a well-established normal range is particularly problematic. For ex-
ample, a patient may present with an osmole gap of 9 mOsm, a value consid-
ered normal by the traditionally accepted normal maximum gap value of
10 mOsm. If this patient had an osmole gap of 5 just before ingestion
of a toxic alcohol, the patient’s osmole gap must have been increased by
14 mOsm to reach the new gap of 9 mOsm. If this increase was caused
by ethylene glycol, it corresponds to a toxic level of 86.8 mg/dL [21]. In ad-
dition, if a patient’s ingestion of a toxic alcohol occurred at a time distant
from the actual blood sampling, the osmotically active parent compound
has been metabolized to the acidic metabolites. The subsequent metabolites
have no osmotic activity of their own and hence no osmole gap is detected
[14,22]. It is possible that a patient may present at a point after ingestion
with only a moderate rise in their osmole gap and anion gap. Steinhart [23]
reported a patient with ethylene glycol toxicity who presented with an osmole
gap of 7.2 mOsm caused by delay in presentation. Darchy and colleagues [20]
presented two other cases of significant ethylene glycol toxicity with osmole
gaps of 4 and 7, respectively. The lack of an abnormal osmole gap in these
cases was speculated either to be caused by metabolism of the parent alcohol
or a low baseline osmole gap that masked the toxin’s presence.
The osmole gap should be used with caution as an adjunct to clinical
decision making and not as a primary determinant to rule out toxic alcohol
ingestion. If the osmole gap obtained is particularly large, it suggests an
agent from the previous list may be present. A normal osmole should be
 CRITICAL CARE TOXICOLOGY 721

interpreted with caution; a negative study may not rule out the presence of
such an ingestion; and the test result must be interpreted within the context
of the clinical presentation. If such a poisoning is suspected, appropriate
therapy should be initiated presumptively (ie, ethanol infusion, 4-methyl-
pyrazole, hemodialysis, and so forth) while confirmation from serum levels
of the suspected toxin are pending.

Urine drug screening

Many clinicians regularly obtain urine drug screening on altered patients
or on those suspected of ingestion. Such routine urine drug testing is of
questionable benefit. Kellermann and colleagues [24] found little impact of
urine drug screening on patient management in an urban emergency setting,
and Mahoney and colleagues [25] similarly conclude that toxic screening
added little to treatment or disposition of overdose patients in their emer-
gency department. In a study of over 200 overdose patients, Brett [26]
showed that although unsuspected drugs were routinely detected, the results
rarely led to changes in management and likely never affected outcome. In
a similar large study of trauma patients, Bast and colleagues [27] noted that
a positive drug screen had minimal impact on patient treatment.
Some authors do argue in favor of routine testing. Fabbri and colleagues
[28] countered that comprehensive screening may aid decisions on patient dis-
position, resulting in fewer admissions to the hospital and less demand on crit-
ical care units. The screen used in their retrospective study tested for over
900 drugs and is not available to most clinicians. Milzman and colleagues
[29] argued in favor of screening trauma victims, stating that the prognosis
of intoxicated patients is unduly poor secondary to low Glasgow Coma Scores,
although patient treatment and disposition did not seem to be affected [29].
The effect of such routine screening on patient management is low because
most of the therapy is supportive and directed at the clinical scenario (ie, men-
tal status, cardiovascular function, respiratory condition). Interpretation of
the results can be difficult even when the objective for ordering a comprehen-
sive urine screen is adequately defined. Most assays rely on the antibody iden-
tification of drug metabolites, with some drugs remaining positive days after
use, and may not be related to the patient’s current clinical picture. The
positive identification of drug metabolites is likewise influenced by chronicity
of ingestion, fat solubility, and coingestions. In one such example, Perrone
and colleagues [30] showed a cocaine retention time of 72 hours following
its use. Conversely, many drugs of abuse are not detected on most urine
drug screens, including gamma hydroxybutyrate, fentanyl, and ketamine.
Interpretation is further confounded by false-positive and false-negative
results. George and Braithwaite [31] evaluated five popular rapid urine
screening kits and found all lacked significant sensitivity and specificity.
The monoclonal antibodies used in these immunoassays may detect epitopes
from multiple drug classes. For example, a relatively new antidepressant,
722 HOLSTEGE et al

venlafaxine, produced false-positive results by cross-reactivity with the

antiphencyclidine antibodies used in a urine test device [32]. False-positive
benzodiazepine results were found in patients receiving the nonsteroidal
anti-inflammatory drug oxaprozin who were screened using urine immuno-
assays [33]. Conversely, antibodies used in the immunoassays may not detect
all drugs classified within a specific drug class. For example, one urine
immunoassay does not detect physiologic doses of methadone. This assay
detects codeine and its metabolites: morphine and morphine-3-glucuronide.
Additionally, cross-reactivity of both prescription and over-the-counter
medications used in therapeutic amounts for true illness may elicit positive
screens. Diphenhydramine has been documented to interfere with one urine
immunoassay for propoxyphene [34].
The use of ordering urine drug screens is fraught with significant testing
limitations, including false-positive and false-negative results. Many authors
have shown that the test results rarely affect management decisions. Routine
drug screening of those with altered mental status, abnormal vital signs,
or suspected ingestion is not warranted and rarely guides patient treatment
or disposition.

Radiographic studies
The role of radiologic testing specifically in the diagnosis and manage-
ment of the critically poisoned patient is limited. Radiologic testing is com-
monly used to diagnose complications associated with poisonings, such as
aspiration pneumonitis and anoxic brain injury.
In some circumstances, plain radiography can assist in the diagnosis and
management of poisonings if the substance in question is radiopaque. The
primary use of radiography in the critically ill poisoned patient is in the
detection and management of iron poisoning (Fig. 1) [35–37]. Attempts to
use the presence or absence of radiopacities consistently to predict severe
iron toxicity have not been successful [38,39]. Not all iron products are
equally radiopaque; whereas ferrous sulfate and ferrous fumurate are
typically radiopaque, other preparations may not be radiopaque [40]. For
example, chewable iron supplements are unlikely to be seen on abdominal
radiographs [37] secondary to both the low elemental content in these chew-
able products and as a result of their quick dissolution in the gastrointestinal
tract [37]. In using abdominal radiographs in diagnosing iron poisonings,
time from ingestion is important; as time passes visualization becomes
more difficult [35].
Another situation where plain radiography may prove useful is with body
packers. These couriers, also known as ‘‘mules,’’ swallow multiple packages
of illicit drugs for the purposes of transporting without detection. The con-
tainer of choice is often condoms, latex, or cellophane formed into balls or
ovals that are 2 to 4 cm in size [41]. Besides the illegal nature of this occupa-
tion, there is a serious health risk to these patients who may suffer from
 CRITICAL CARE TOXICOLOGY 723

Fig. 1. A radiograph demonstrating a grouping of iron tablets in the stomach (arrows) follow-
ing a suicidal ingestion of ferrous fumarate.

intestinal obstruction or from the direct effects of the illicit drugs themselves
if the packages leak [42]. One study by Karhunen and colleagues [43] in
Finland looked at a total of 82 patients admitted for abdominal radiographs
because of suspected body packing. Twelve of these were read as positive,
and nine of these proved to be true positives (75%). The three false-positives
(3.6%) were thought secondary to constipation involving compact feces with
increased radiodensity mimicking narcotic packages. There were 70 films
that were read as negative with only one false-negative (1.2%) that was at-
tributed to the inexperience of the radiologist. In a recent review of the liter-
ature, plain abdominal radiography was identified as the radiologic method
of choice for finding these packets, as opposed to ultrasound and CT [41].
The authors based this decision on ease of use, availability, patient tolerance,
and the relatively high sensitivity and specificity shown by the Karhunen
study.
Besides these established examples, plain radiography of the abdomen
has also been studied to identify other pills that may be radiopaque in acute
overdose. Multiple studies regarding the radiopacity of ingested pharmaceu-
ticals have not consistently supported the use of radiography in manage-
ment of these patients. O’Brien and colleagues [44] studied the detect
ability of 459 different tablets and capsules using plain radiography. The in-
vestigators used a ferrous sulfate tablet as a control grading the other
tablets’ radiopacity. These pills were then placed in the middle of a plastic
container that contained 20 cm of water to simulate the human body den-
sity. Overall, of the wide variety of pills tested, only 29 drugs (6.3%) were
graded as having the same or greater radiopacity as ferrous sulfate; 136 pills
(29.6%) were regarded as having at least moderate opacity; and the largest
724 HOLSTEGE et al

remaining portion of pills, 294 (64%), were regarded as no more than min-
imally detectable. The authors concluded that indiscriminate use of plain
abdominal radiographs was not justified and that a negative film could
not be relied on to rule out potential toxic pill ingestions, especially if there
is time to allow the pills to dissolve.

Electrocardiogram
The interpretation of ECG in the poisoned patient can challenge even the
most experienced clinician. There are numerous drugs that can cause ECG
changes. The incidence of ECG changes in the poisoned patient is unclear
and the significance of various changes may be difficult to define. Despite
the fact that drugs have widely varying indications for therapeutic use,
many unrelated drugs share common cardiac electrocardiographic effects
if taken in overdose. Potential toxins can be placed into broad classes based
on their cardiac effects. Two such classes, agents that block the cardiac
potassium efflux channels and agents that block cardiac fast sodium chan-
nels, can lead to characteristic changes in cardiac indices consisting
of QRS prolongation and QT prolongation, respectively. The recognition
of specific ECG changes associated with other clinical data (toxidromes)
can be potentially life saving [45].
Studies suggest that approximately 3% of all noncardiac prescriptions
are associated with the potential for QT prolongation [46]. Myocardial re-
polarization is driven predominantly by outward movement of potassium
ions [47]. Blockade of the outward potassium currents by drugs prolongs
the action potential, resulting in QT interval prolongation and the poten-
tial emergence of T- or U-wave abnormalities on the ECG [48,49]. The
prolongation of repolarization causes the myocardial cell to have less
charge difference across its membrane, which may result in the activation
of the inward depolarization current (early afterdepolarization) and pro-
mote triggered activity. These changes may lead to reentry and subsequent
polymorphic ventricular tachycardia, most often as the torsades de pointes
variant of polymorphic ventricular tachycardia [50]. The QT interval is
simply measured from the beginning of the QRS complex to the end of
the T wave. Within any ECG tracing, there is lead-to-lead variation of
the QT interval. In general, the longest measurable QT interval on an
ECG is regarded as determining the overall QT interval for a given tracing
[51]. The QT interval is influenced by the patient’s heart rate. Several for-
mulas have been developed to correct the QT interval for the effect of
heart rate (QTc) using the RR interval (RR), with Bazett’s formula
(QTc ¼ QT/RR1/2) being the most commonly used. QT prolongation is
considered to occur when the QTc interval is greater than 440 milliseconds
in men and 460 milliseconds in women, with arrhythmias most commonly
associated with values greater than 500 milliseconds (Fig. 2). The potential
for an arrhythmia for a given QT interval varies from drug to drug and
 CRITICAL CARE TOXICOLOGY 725

Fig. 2. ECG demonstrating sinus bradycardia with marked QT prolongation (660 ms) follow-
ing an overdose of sotalol.

patient to patient [47]. Drugs associated with QT prolongation are listed in

Box 1 [52].
Other etiologies involved in possible prolongation of the QT interval
include congenital long QT syndrome, mitral valve prolapse, hypokalemia,
hypocalcemia, hypomagnesemia, hypothermia, myocardial ischemia, neuro-
logic catastrophes, and hypothyroidism [53].
The ability of drugs to induce cardiac Naþ channel blockade and prolong
the QRS complex has been well described in numerous literature reports
(Fig. 3) [54]. This Naþ channel blockade activity has been described as
a membrane stabilizing effect, a local anesthetic effect, or a quinidine-like
effect. Cardiac voltage-gated sodium channels reside in the cell membrane
and open in conjunction with cell depolarization. Sodium channel blockers
bind to the transmembrane Naþ channels and decrease the number available
for depolarization. This creates a delay of Naþ entry into the cardiac myo-
cyte during phase 0 of depolarization. As a result, the upslope of depolariza-
tion is slowed and the QRS complex widens [55]. In some cases, the QRS
complex may take the pattern of recognized bundle branch blocks [56,57].
In the most severe cases, the QRS prolongation becomes so profound that
it is difficult to distinguish between ventricular and supraventricular rhythms
[58,59]. Continued prolongation of the QRS may result in a sine wave pat-
tern and eventual asystole (Fig. 4). It has been theorized that the Naþ chan-
nel blockers can cause slowed intraventricular conduction, unidirectional
block, the development of a reentrant circuit, and a resulting ventricular
tachycardia [60]. This can then degenerate into ventricular fibrillation. Dif-
ferentiating a prolongation of the QRS complex because of Naþ channel
blockade in the poisoned patient versus other nontoxic etiologies can be dif-
ficult. Rightward axis deviation of the terminal 40 milliseconds of the QRS
axis has been associated with tricyclic antidepressant poisoning [61,62]. The
occurrence of this finding in other Naþ channel blocking agents is unknown.
726 HOLSTEGE et al

Box 1. K+ efflux channel blocking drugs

Antihistamines
Astemizole
Clarithromycin
Diphenhydramine
Loratidine
Terfenadine
Antipsychotics
Chlorpromazine
Droperidol
Haloperidol
Mesoridazine
Pimozide
Quetiapine
Risperidone
Thioridazine
Ziprasidone
Arsenic trioxide
Bepridil
Chloroquine
Cisapride
Citalopram
Clarithromycin
Class IA antiarrhythmics
Disopyramide
Quinidine
Procainamide
Class IC antiarrhythmics
Encainide
Flecainide
Moricizine
Propafenone
Class III antiarrhythmics
Amiodarone
Dofetilide
Ibutilide
Sotalol
Cyclic antidepressants
Amitriptyline
Amoxapine
Desipramine
Doxepin
 CRITICAL CARE TOXICOLOGY 727

Imipramine
Nortriptyline
Maprotiline
Erythromycin
Fluoroquinolones
Ciprofloxacin
Gatifloxacin
Levofloxacin
Moxifloxacin
Sparfloxacin
Hydroxychloroquine
Levomethadyl
Methadone
Pentamidine
Quinine
Tacrolimus
Venlafaxine

Myocardial Naþ channel blocking drugs comprise a diverse group of phar-

maceutical agents (Box 2).
Patients poisoned with these agents have a variety of clinical presenta-
tions. For example, sodium channel blocking medications, such as diphen-
hydramine, propoxyphene, and cocaine, may also develop anticholinergic,
opioid, and sympathomimetic syndromes, respectively [63–65]. In addition,
specific drugs may affect not only the myocardial Naþ channels, but also
calcium influx and potassium efflux channels [66,67]. This may result in
ECG changes and rhythm disturbances not related entirely to the drug’s
Naþ channel blocking activity. All the agents listed in Box 2 are similar in
that they may induce myocardial Naþ channel blockade and may respond
to therapy with hypertonic saline or sodium bicarbonate [59,64,65]. It is

Fig. 3. A rhythm strip demonstrating marked QRS prolongation following a propoxyphene

overdose.
728 HOLSTEGE et al

Fig. 4. An ECG demonstrating a sinusoidal wave pattern secondary to the sodium channel
blockade induced by an acute overdose of hydroxychloroquine.

reasonable to treat poisoned patients with a prolonged QRS interval, partic-

ularly those with hemodynamic instability, empirically with 1 to 2 mEq/kg
of sodium bicarbonate. A shortening of the QRS can confirm the presence
of a sodium channel blocking agent. It can also improve inotropy and help
prevent arrhythmias [54].
There are multiple agents that can result in cardiotoxicity and subsequent
ECG changes from the changes noted previously to other alterations, such
as bradycardia and tachycardia. Physicians managing patients who have
taken overdoses on medications should be aware of the various electrocar-
diographic changes that can potentially occur in the overdose setting.

Management
After initial evaluation and stabilization of the critically poisoned patient
as described previously, the physician can consider whether there is a need
for the administration of specific therapies. Decontamination should be con-
sidered. Several poisons have specific antidotes that if used in a timely and
appropriate manner can be of great benefit. Finally, the safe disposition of
the patient must be determined (ie, monitored floor bed or ICU).
Approximately 80% of all poisonings occur by ingestion and the most
common type of decontamination performed is gastrointestinal decontami-
nation using a variety of techniques including emesis, gastric lavage, acti-
vated charcoal, cathartics, and whole-bowel irrigation. Poisonings may
also occur by dermal and ocular routes, which necessitate external decon-
tamination. Significant controversy exists concerning the need for routine
gastric emptying in the poisoned patient. Current available evidence
dissuades from the routine use of gastric decontamination. Gastric decon-
tamination may be considered in select cases and specific scenarios. Before
 CRITICAL CARE TOXICOLOGY 729

Box 2. Na+ channel blocking drugs

Amantadine
Carbamazepine
Chloroquine
Class IA antiarrhythmics
Disopyramide
Quinidine
Procainamide
Class IC antiarrhythmics
Encainide
Flecainide
Propafenone
Citalopram
Cocaine
Cyclic antidepressants
Diltiazem
Diphenhydramine
Hydroxychloroquine
Loxapine
Orphenadrine
Phenothiazines
Medoridazine
Thioridazine
Propranolol
Propoxyphene
Quinine
Verapamil

performing gastrointestinal decontamination techniques, the clinician

responsible for the care of the poisoned patient must clearly understand
whether the benefit of decontamination outweighs any potential harm.
The number of pharmacologic antagonists or antidotes available to treat
the critically poisoned patient is quite limited (Table 2). There are few anti-
dotes that rapidly reverse toxic effects and restore a patient to a previously
healthy baseline state. Administering some pharmacologic antagonists may
worsen patient outcome compared with simply optimizing basic supportive
care. Antidotes should be used cautiously, with a clear understanding of
indications and contraindications.

Atropine
Atropine is the initial drug of choice in symptomatic patients poisoned
with organophosphates or carbamates. Atropine acts as a muscarinic
730 HOLSTEGE et al

Table 2
Antidotes
Agent or clinical finding Antidote
Acetaminophen N-acetylcysteine
Benzodiazepines Flumazenil
b-blockers Glucagon
Cardiac glycosides Digoxin immune Fab
Crotalid envenomation Crotalidae polyvalent immune Fab
Cyanide Hydroxocobalamin
Ethylene glycol Fomepizole
Iron Deferoxamine
Isoniazid Pyridoxine
Methanol Fomepizole
Methemoglobinemia Methylene blue
Opioids Naloxone
Organophosphates Atropine
Pralidoxime
Sulfonylureas Octreotide

receptor antagonist and blocks neuroeffector sites on smooth muscle, car-

diac muscle, secretory gland cells, and the central nervous system. Atropine
is useful in alleviating bronchoconstriction, bronchorrhea, tenesmus, ab-
dominal cramps, nausea, vomiting, bradydysrythmias, and seizure activity.
Atropine can be administered by the intravenous, intraosseous, intramuscu-
lar, or endotracheal route. The dose varies with the type of exposure, requir-
ing a few milligrams in mild cases and hundreds of milligrams in extreme
cases [68]. For the mildly and moderately symptomatic patient, 2 mg for
adults and 0.02 mg/kg for children (minimum of 0.1 mg) is administered
every 5 minutes. In the severely poisoned patient, dosages may need to be
increased and given more rapidly [69]. Tachycardia is not a contraindication
to atropine administration in these patients. Drying of the respiratory secre-
tions and resolution of bronchoconstriction are the therapeutic end points
used to determine the appropriate dose of atropine. This is clinically appar-
ent as the patient’s work of breathing improves [68]. Atropine has no effect
on the nicotinic receptors and has no effect on autonomic ganglia and neu-
romuscular junction [69]. Muscle weakness, fasciculations, tremors, and
paralysis are not indications for further atropine dosing. It does have a par-
tial effect on the central nervous system and is helpful in resolving or
preventing seizures [70]. It is most effective in preventing seizures if given
within 5 minutes of organophosphate exposure [71]. After 5 to 10 minutes
anticholinergic treatment alone is not effective at terminating seizures and
benzodiazepines must be added to treat seizures effectively [71,72].

Deferoxamine
Deferoxamine is an effective chelator of iron. Deferoxamine chelates iron
and converts it to a water-soluble complex, ferrioxamine, which is
 CRITICAL CARE TOXICOLOGY 731

eliminated readily by the urine. Indications for deferoxamine infusion in-

clude significant clinical signs of iron toxicity, metabolic acidosis, shock,
profound lethargy, coma, serum iron levels greater than 500 mg/dL, or
a radiograph positive for multiple pills [73]. Deferoxamine should be infused
intravenously at a starting rate of 15 mg/kg/h, not to exceed 1 g/h, over a to-
tal of 6 hours, followed by re-evaluation. Deferoxamine-induced hypoten-
sion may occur at fast rates, and adequate hydration should be ensured
before infusion initiation [73]. As iron is chelated and excreted, urine may
develop a characteristic rusty-red (vine rose) appearance.

Crotalidae antivenin
Use of antivenin in the appropriate doses can control local swelling and
serious systemic effects (ie, coagulopathy) that occur in patients who have
been envenomated [74]. Antivenin should not be used prophylactically
because a significant number of snake bites are dry bites. There are
numerous dosage regimens that vary with the degree of systemic toxicity
and regional treatment preferences. Consultation with a poison center or
a clinical toxicologist is advised for the most contemporary treatment
recommendations.

Digoxin immune Fab

Digoxin-specific Fab fragments are antibody fragments produced by
enzymatic cleavage of sheep IgG antibodies to digoxin. Fab fragments
can reverse digitalis-induced dysrhythmias, conduction disturbances, myo-
cardial depression, and hyperkalemia in severely poisoned patients [75].
Patients can have reversal of ventricular arrhythmia within 2 minutes
and most patients have settling of toxic dysrhythmias within 30 minutes
of Fab administration [76]. Within 6 hours 90% of patients have complete
or partial response [75]. Animal studies and case reports have demon-
strated the efficacy of Fab fragments to the cardiac glycoside contained
in plants [77–79]. Digoxin-specific Fab fragment therapy should be admin-
istered in a digoxin poisoned patient for the following indications: (1) po-
tassium greater than 5 mEq/L following acute ingestion, (2) hemodynamic
instability, and (3) patients with potentially life-threatening dysrhythmias
[76].
Although serum digoxin levels should not be the sole factor in determin-
ing the need to administer Fab, dosage calculations for Fab are based on the
serum digoxin level, or estimated body load of digoxin. It is assumed that
equimolar doses of antibody fragments are required to achieve neutraliza-
tion. Forty milligrams of Fab (one vial) bind 0.6 mg of digoxin. When pre-
sented with a severely poisoned patient in whom the quantity ingested
acutely is unknown, an empiric dose of 5 to 10 vials at a time should be
given and the clinical response observed. If cardiac arrest is imminent or
has occurred, the dose can be given as a bolus, but it should be infused
732 HOLSTEGE et al

over 30 minutes in stable patients. For patients with chronic therapeutic

overdoses the digoxin levels are often only mildly elevated and one to two
vials of Fab may be sufficient [75]. The recommended dose for a given
patient can be determined using the tables in the package insert or by con-
tacting a regional poison center or toxicology consultant.

Flumazenil
Benzodiazepines are involved in many intentional overdoses. Although
these overdoses are rarely fatal when a benzodiazepine is the sole ingestant,
they often complicate overdoses with other central nervous system depres-
sants (eg, ethanol, opioids, and other sedatives) because of their synergistic
activity. Flumazenil finds its greatest use in the reversal of benzodiazepine-
induced sedation from minor surgical procedures. The initial flumazenil
dose is 0.2 mg and should be administered intravenously over 30 seconds.
If no response occurs after an additional 30 seconds, a second dose is recom-
mended. Additional incremental doses of 0.5 mg may be administered at
1-minute intervals until the desired response is noted or until a total of
3 mg has been administered. It is important to note that resedation may occur
[80], and patients should be observed with close monitoring after requiring re-
versal. Flumazenil should not be administered as a nonspecific coma-reversal
drug and should be used with extreme caution after intentional benzodiaze-
pine overdose because it has the potential to precipitate withdrawal in benzo-
diazepine-dependent individuals or induce seizures in those at risk [80].

Fomepizole
Fomepizole (4-methyl-pyrazole) is an alcohol dehydrogenase inhibitor. It
is administered in cases of suspected or confirmed ingestion and intoxication
with ethylene glycol or methanol [81]. Fomepizole should be administered in-
travenously as a loading dose of 15 mg/kg, followed by doses of 10 mg/kg ev-
ery 12 hours for four doses (48 hours) then 15 mg/kg every 12 hours thereafter
[8,82]; all doses should be administered as a slow intravenous infusion over
30 minutes. During hemodialysis, the frequency of dosing should be increased
to every 4 hours. Therapy should be continued until ethylene glycol or meth-
anol concentrations are less than 20 mg/dL and the patient is asymptomatic.

Hydroxocobalamin
Hydroxocobalamin is a safe and effective treatment of cyanide toxicity
that has recently been approved in the United States. The reaction of hy-
droxocobalamin with cyanide results in the displacement of a hydroxyl
group by a cyano group to form cyanocobalamin (vitamin B12), which is
then excreted in the urine [83]. The usual adult dose of hydroxocobalamin
is 5 g, which may be repeated in cases of massive cyanide poisoning
[84,85]. The pediatric dose is 70 mg/kg up to 5 g [86]. Virtually every patient
 CRITICAL CARE TOXICOLOGY 733

receiving this antidote develops orange-red discoloration of the skin,

mucous membranes, and urine. This resolves within 24 to 48 hours [87].

N-Acetylcysteine
Significant acetaminophen overdoses may need to be treated with
N-acetylcysteine if the patient has a toxic serum acetaminophen concentration
or has indicators of hepatocellular damage [88]. N-acetylcysteine increases
glutathione levels and serves as a glutathione surrogate. An acetaminophen
overdose may deplete glutathione, permitting the toxic metabolite to destroy
hepatocytes. N-acetylcysteine is most effective if administered within 8 hours
of the acetaminophen ingestion; however, it can still be effective days after
the ingestion when patients are already in hepatic failure and acetamino-
phen levels are no longer detectable [88].
N-acetylcysteine can be given by both oral and intravenous administra-
tion [88]. Oral is 140 mg/kg loading dose followed by 70 mg/kg every 4 hours
for 17 doses. Intravenous is 150 mg/kg loading dose followed by 50 mg/kg
over 4 hours followed by 100 mg/kg infused over 16 hours.
Parenteral administration of N-acetylcysteine eliminates compliance
problems associated with oral therapy (adverse taste and odor caused by
the sulfhydryl groups) and circumvents the problems associated with acet-
aminophen-induced vomiting.

Naloxone
Opioid poisoning from the abuse of morphine derivatives or synthetic
narcotic agents may be reversed with the opioid antagonist naloxone [89].
Naloxone is commonly used in comatose patients as a therapeutic and
diagnostic agent. The standard dosage regimen is to administer from 0.4
to 2 mg slowly, preferably intravenously. The intravenous dose should be
readministered at 5-minute intervals until the desired end point is achieved:
restoration of respiratory function, ability to protect the airway, and an im-
proved level of consciousness [90]. If the intravenous route of administration
is not viable, alternative routes include intramuscular and intraosseous [90].
A patient may not respond to naloxone administration for a variety of
reasons: insufficient dose of naloxone, the absence of an opioid exposure,
a mixed overdose with other central nervous and respiratory system depres-
sants, or medical or traumatic reasons.
Naloxone can precipitate profound withdrawal symptoms in opioid-
dependant patients. Symptoms of withdrawal include agitation, vomiting,
diarrhea, piloerection, diaphoresis, and yawning [90]. Care should be taken
to administer this agent as necessary only to restore adequate respiration
and airway protection. Naloxone’s clinical efficacy can last for as little as
45 minutes [89]. Patients are at risk for recurrence of narcotic effect. This
is particularly true for patients exposed to methadone or sustained-release
734 HOLSTEGE et al

opioid products. In addition, renal insufficiency increases naloxone’s elimi-

nation half-life, placing the patient at risk for resedation hours after the
initial dose. Patients should be observed in a monitored setting for reseda-
tion for at least 4 hours after reversal with naloxone. If a patient does rese-
date, it is reasonable to administer naloxone as an infusion. An infusion of
two thirds the effective initial bolus per hour is usually effective [90].

Pralidoxime chloride
Pralidoxime chloride reactivates acetylcholinesterase by exerting a nucle-
ophilic attack on the phosphorus resulting in an oxime-phosphate bond that
splits from the acetylcholinesterase leaving the regenerated enzyme. This
reactivation is clinically most apparent at skeletal neuromuscular junctions,
with less activity at muscarinic sites [91]. Pralidoxime must be administered
concurrently with adequate atropine doses. The process of aging prevents
pralidoxime from regenerating the acetylcholinesterase-active site, and is
ineffective after aging has occurred. The sooner pralidoxime is administered,
the greater the clinical effect. The recommended dose of pralidoxime is 1 to 2 g
for adults or 20 to 50 mg/kg for children by intravenous route. Slow
administration over 15 to 30 minutes has been advocated to minimize
side effects [68,92]. These side effects include hypertension, headache,
blurred vision, epigastric discomfort, nausea, and vomiting. Rapid admin-
istration can result in laryngospasm, muscle rigidity, and transient impair-
ment of respiration [91].
Pralidoxime is rapidly excreted by the kidney with a half-life of approx-
imately 90 minutes [93]. A continuous infusion is often recommended after
the loading dose to maintain therapeutic levels [94–97]. Currently, the
World Health Organization recommends a bolus of greater than 30 mg/kg
followed by an infusion of greater than 8 mg/kg/h [98].

Pyridoxine
Isoniazid, hydrazine, and the Gyrometria species of mushrooms can
decrease the brain concentrations of g-aminobutyric acid by inhibiting pyr-
idoxal-5-phosphate activity, resulting in the development of severe seizure
activity [99,100]. The administration of pyridoxine (vitamin B6) can prevent
or actively treat the central nervous system toxicity associated with these
toxins [101]. Pyridoxine is administered on a gram-for-gram basis with iso-
niazid (ie, the amount of pyridoxine should equal the amount of isoniazid).
If the ingested amount of agent is unknown, the dose of pyridoxine should
be 5 g administered intravenously [101]. This dose can be repeated.

Summary
The emergency physician often is required to care for critically poisoned
patients. Prompt action must be taken for patients who present with serious
 CRITICAL CARE TOXICOLOGY 735

toxic effects or after potentially fatal ingestions. Because many poisons have
no true antidote and the poison involved may initially be unknown, the first
step is simply focused on supportive care. Identifying the causative poison,
through a detailed history, recognizing a toxidrome, or laboratory analysis
may help direct care. There are several antidotes available that can be life
saving, and the clinician should promptly identify those patients who may
benefit from these agents.

References
[1] Eddleston M, Haggalla S, Reginald K, et al. The hazards of gastric lavage for intentional
self-poisoning in a resource poor location. Clin Toxicol 2007;45(2):136–43.
[2] Wills B, Erickson T. Drug- and toxin-associated seizures. Med Clin North Am 2005;89(6):
1297–321.
[3] Brett AS. Implications of discordance between clinical impression and toxicology analysis
in drug overdose. Arch Intern Med 1988;148(2):437–41.
[4] Wu AH, McKay C, Broussard LA, et al. National Academy of Clinical Biochemistry
laboratory medicine practice guidelines: recommendations for the use of laboratory tests
to support poisoned patients who present to the emergency department. Clin Chem
2003;49(3):357–79.
[5] Chabali R. Diagnostic use of anion and osmoleal gaps in pediatric emergency medicine.
Pediatr Emerg Care 1997;13(3):204–10.
[6] Ishihara K, Szerlip HM. Anion gap acidosis. Semin Nephrol 1998;18(1):83–97.
[7] Gabow PA. Disorders associated with an altered anion gap. Kidney Int 1985;27(2):472–83.
[8] Mégarbane B, Borron SW, Baud FJ. Current recommendations for treatment of severe
toxic alcohol poisonings. Intensive Care Med 2005;31(2):189–95.
[9] Marciniak K, Thomas I, Brogan T, et al. Massive ibuprofen overdose requiring extracor-
poreal membrane oxygenation for cardiovascular support. Pediatr Crit Care Med 2007;
8(2):180–2 [Report].
[10] Judge BS. Metabolic acidosis: differentiating the causes in the poisoned patient. Med Clin
North Am 2005;89(6):1107–24.
[11] Suchard JR. Osmoleal gap. In: Dart RC, editor. Medical toxicology. 3rd edition. Philadel-
phia: Lippincott Williams & Wilkins; 2004. p. 106–9.
[12] Kruse JA, Cadnapaphornchai P. The serum osmol gap. J Crit Care 1994;9(3):185–97.
[13] Erstad BL. Osmoleality and osmolearity: narrowing the terminology gap. Pharmacother-
apy 2003;23(9):1085–6.
[14] Glaser DS. Utility of the serum osmole gap in the diagnosis of methanol or ethylene glycol
ingestion. Ann Emerg Med 1996;27(3):343–6.
[15] Worthley LI, Guerin M, Pain RW. For calculating osmoleality, the simplest formula is the
best. Anaesth Intensive Care 1987;15(2):199–202.
[16] Smithline N, Gardner KD Jr. Gaps–anionic and osmoleal. JAMA 1976;236(14):1594–7.
[17] Glasser L, Sternglanz PD, Combie J, et al. Serum osmoleality and its applicability to drug
overdose. Am J Clin Pathol 1973;60(5):695–9.
[18] McQuillen KK, Anderson AC. Osmole gaps in the pediatric population. Acad Emerg Med
1999;6(1):27–30.
[19] Aabakken L, Johansen KS, Rydningen EB, et al. Osmolal and anion gaps in patients
admitted to an emergency medical department. Hum Exp Toxicol 1994;13(2):131–4.
[20] Darchy B, Abruzzese L, Pitiot O, et al. Delayed admission for ethylene glycol poisoning:
lack of elevated serum osmole gap. Intensive Care Med 1999;25(8):859–61.
[21] Hoffman RS, Smilkstein MJ, Howland MA, et al. Osmole gaps revisited: normal values and
limitations. J Toxicol Clin Toxicol 1993;31(1):81–93.
736 HOLSTEGE et al

[22] Eder AF, McGrath CM, Dowdy YG, et al. Ethylene glycol poisoning: toxicokinetic and
analytical factors affecting laboratory diagnosis. Clin Chem 1998;44(1):168–77.
[23] Steinhart B. Case report: severe ethylene glycol intoxication with normal osmoleal gap–
a chilling thought. J Emerg Med 1990;8(5):583–5.
[24] Kellermann AL, Fihn SD, LoGerfo JP, et al. Impact of drug screening in suspected
overdose. Ann Emerg Med 1987;16(11):1206–16.
[25] Mahoney JD, Gross PL, Stern TA, et al. Quantitative serum toxic screening in the manage-
ment of suspected drug overdose. Am J Emerg Med 1990;8(1):16–22.
[26] Brett A. Toxicologic analysis in patients with drug overdose. Arch Intern Med 1988;148(9):
2077.
[27] Bast RP, Helmer SD, Henson SR, et al. Limited utility of routine drug screening in trauma
patients. South Med J 2000;93(4):397–9.
[28] Fabbri A, Marchesini G, Morselli-Labate AM, et al. Comprehensive drug screening in
decision making of patients attending the emergency department for suspected drug
overdose. Emerg Med J 2003;20(1):25–8.
[29] Milzman DP, Boulanger BR, Rodriguez A, et al. Pre-existing disease in trauma patients:
a predictor of fate independent of age and injury severity score. J Trauma 1992;32(2):
236–43.
[30] Perrone J, De Roos F, Jayaraman S, et al. Drug screening versus history in detection of
substance use in ED psychiatric patients. Am J Emerg Med 2001;19(1):49–51.
[31] George S, Braithwaite RA. A preliminary evaluation of five rapid detection kits for on site
drugs of abuse screening. Addiction 1995;90(2):227–32.
[32] Sena SF, Kazimi S, Wu AH. False-positive phencyclidine immunoassay results caused by
venlafaxine and O-desmethylvenlafaxine. Clin Chem 2002;48(4):676–7.
[33] Camara PD, Audette L, Velletri K, et al. False-positive immunoassay results for
urine benzodiazepine in patients receiving oxaprozin (Daypro). Clin Chem 1995;41(1):
115–6.
[34] Schneider S, Wennig R. Interference of diphenhydramine with the EMIT II immunoassay
for propoxyphene. J Anal Toxicol 1999;23(7):637–8.
[35] Ng RC, Perry K, Martin DJ. Iron poisoning: assessment of radiography in diagnosis and
management. Clin Pediatr (Phila) 1979;18(10):614–6.
[36] Kaczorowski JM, Wax PM. Five days of whole-bowel irrigation in a case of pediatric iron
ingestion. Ann Emerg Med 1996;27(2):258–63.
[37] Everson GW, Oudjhane K, Young LW, et al. Effectiveness of abdominal radiographs in
visualizing chewable iron supplements following overdose. Am J Emerg Med 1989;7(5):
459–63.
[38] Chyka PA, Butler AY. Assessment of acute iron poisoning by laboratory and clinical
observations. Am J Emerg Med 1993;11(2):99–103.
[39] Palatnick W, Tenenbein M. Leukocytosis, hyperglycemia, vomiting, and positive X-rays
are not indicators of severity of iron overdose in adults. Am J Emerg Med 1996;14(5):
454–5.
[40] Savitt DL, Hawkins HH, Roberts JR. The radiopacity of ingested medications. Ann Emerg
Med 1987;16(3):331–9.
[41] Hergan K, Kofler K, Oser W. Drug smuggling by body packing: what radiologists should
know about it. Eur Radiol 2004;14(4):736–42.
[42] McCleave NR. Drug smuggling by body packers: detection and removal of internally
concealed drugs. Med J Aust 1993;159(11–12):750–4.
[43] Karhunen PJ, Suoranta H, Penttila A, et al. Pitfalls in the diagnosis of drug smuggler’s
abdomen. J Forensic Sci 1991;36(2):397–402.
[44] O’Brien RP, McGeehan PA, Helmeczi AW, et al. Detectability of drug tablets and capsules
by plain radiography. Am J Emerg Med 1986;4(4):302–12.
[45] Holstege C, Baer A, Brady WJ. The electrocardiographic toxidrome: the ECG presentation
of hydrofluoric acid ingestion. Am J Emerg Med 2005;23(2):171–6.
 CRITICAL CARE TOXICOLOGY 737

[46] De Ponti F, Poluzzi E, Montanaro N. QT-interval prolongation by non-cardiac drugs:

lessons to be learned from recent experience. Eur J Clin Pharmacol 2000;56(1):1–18.
[47] Yap YG, Camm AJ. Drug induced QT prolongation and torsades de pointes. Heart 2003;
89(11):1363–72.
[48] Anderson ME, Al-Khatib SM, Roden DM, et al. Cardiac repolarization: current
knowledge, critical gaps, and new approaches to drug development and patient manage-
ment. Am Heart J 2002;144(5):769–81.
[49] Sides GD. QT interval prolongation as a biomarker for torsades de pointes and sudden
death in drug development. Dis Markers 2002;18(2):57–62.
[50] Nelson LS. Toxicologic myocardial sensitization. J Toxicol Clin Toxicol 2002;40(7):
867–79.
[51] Chan T, Brady W, Harrigan R, et al, editors. ECG in emergency medicine and acute care.
Philadelphia: Elsevier-Mosby; 2005.
[52] De Ponti F, Poluzzi E, Cavalli A, et al. Safety of non-antiarrhythmic drugs that prolong the
QT interval or induce torsades de pointes: an overview. Drug Saf 2002;25(4):263–86.
[53] Priori SG, Cantu F, Schwartz PJ. The long QT syndrome: new diagnostic and therapeutic
approach in the era of molecular biology. Schweiz Med Wochenschr 1996;126(41):1727–31.
[54] Kolecki PF, Curry SC. Poisoning by sodium channel blocking agents. Crit Care Clin 1997;
13(4):829–48.
[55] Harrigan RA, Brady WJ. ECG abnormalities in tricyclic antidepressant ingestion. Am
J Emerg Med 1999;17(4):387–93.
[56] Heaney RM. Left bundle branch block associated with propoxyphene hydrochloride
poisoning. Ann Emerg Med 1983;12(12):780–2.
[57] Fernandez-Quero L, Riesgo MJ, Agusti S, et al. Left anterior hemiblock, complete right
bundle branch block and sinus tachycardia in maprotiline poisoning. Intensive Care Med
1985;11(4):220–2.
[58] Brady WJ, Skiles J. Wide QRS complex tachycardia: ECG differential diagnosis. Am
J Emerg Med 1999;17(4):376–81.
[59] Clark RF, Vance MV. Massive diphenhydramine poisoning resulting in a wide-complex
tachycardia: successful treatment with sodium bicarbonate. Ann Emerg Med 1992;21(3):
318–21.
[60] Joshi AK, Sljapic T, Borghei H, et al. Case of polymorphic ventricular tachycardia in
diphenhydramine poisoning. J Cardiovasc Electrophysiol 2004;15(5):591–3.
[61] Wolfe TR, Caravati EM, Rollins DE. Terminal 40-ms frontal plane QRS axis as a marker
for tricyclic antidepressant overdose. Ann Emerg Med 1989;18(4):348–51.
[62] Berkovitch M, Matsui D, Fogelman R, et al. Assessment of the terminal 40-millisecond
QRS vector in children with a history of tricyclic antidepressant ingestion. Pediatr Emerg
Care 1995;11(2):75–7.
[63] Zareba W, Moss AJ, Rosero SZ, et al. Electrocardiographic findings in patients with
diphenhydramine overdose. Am J Cardiol 1997;80(9):1168–73.
[64] Stork CM, Redd JT, Fine K, et al. Propoxyphene-induced wide QRS complex dysrhy-
thmia responsive to sodium bicarbonate–a case report. J Toxicol Clin Toxicol 1995;
33(2):179–83.
[65] Kerns W II, Garvey L, Owens J. Cocaine-induced wide complex dysrhythmia. J Emerg Med
1997;15(3):321–9.
[66] Bania TC, Blaufeux B, Hughes S, et al. Calcium and digoxin vs. calcium alone for severe
verapamil toxicity. Acad Emerg Med 2000;7(10):1089–96.
[67] Dorsey ST, Biblo LA. Prolonged QT interval and torsades de pointes caused by the
combination of fluconazole and amitriptyline. Am J Emerg Med 2000;18(2):227–9.
[68] Newmark J. Nerve agents. Neurologist 2007;13(1):20–32.
[69] Eddleston M, Dawson A, Karalliedde L, et al. Early management after self-poisoning with
an organophosphorus or carbamate pesticide: a treatment protocol for junior doctors. Crit
Care 2004;8(6):R391–7.
738 HOLSTEGE et al

[70] Shih TM, Rowland TC, McDonough JH. Anticonvulsants for nerve agent-induced
seizures: the influence of the therapeutic dose of atropine. J Pharmacol Exp Ther 2007;
320(1):154–61.
[71] McDonough JH, Shih T-M. Neuropharmacological mechanisms of nerve agent-induced
seizure and neuropathology. Neurosci Biobehav Rev 1997;21(5):559–79.
[72] Myhrer T, Enger S, Aas P. Pharmacological therapies against soman-induced seizures in
rats 30 min following onset and anticonvulsant impact. Eur J Pharmacol 2006;548(1–3):
83–9.
[73] Madiwale T, Liebelt E. Iron: not a benign therapeutic drug. Curr Opin Pediatr 2006;18(2):
174–9.
[74] Singletary EM, Rochman AS, Bodmer JC, et al. Envenomations. Med Clin North Am
2005;89(6):1195–224.
[75] Kirk M, Judge B. Digitalis poisoning. In: Irwin J, editor. Intensive care medicine. 5th
edition. Baltimore (MD): Lippincott Williams & Wilkins; 2003. p. 1551–5.
[76] Flanagan RJ, Jones AL. Fab antibody fragments: some applications in clinical toxicology.
Drug Saf 2004;27(14):1115–33.
[77] Shumaik GM, Wu AW, Ping AC. Oleander poisoning: treatment with digoxin-specific Fab
antibody fragments. Ann Emerg Med 1988;17(7):732–5.
[78] Camphausen C, Haas N, Mattke A. Successful treatment of oleander intoxication (cardiac
glycosides) with digoxin-specific Fab antibody fragments in a 7-year-old child. Z Kardiol
2005;94(12):817–23.
[79] Clark RF, Selden BS, Curry SC. Digoxin-specific Fab fragments in the treatment of
oleander toxicity in a canine model. Ann Emerg Med 1991;20(10):1073–7.
[80] Seger DL. Flumazenil–treatment or toxin. J Toxicol Clin Toxicol 2004;42(2):209–16.
[81] Brent J, McMartin K, Phillips S, et al, The Methylpyrazole for Toxic Alcohols Study Group.
Fomepizole for the treatment of methanol poisoning. N Engl J Med 2001;344(6):424–9.
[82] Lushine KA, Harris CR, Holger JS. Methanol ingestion: prevention of toxic sequelae after
massive ingestion. J Emerg Med 2003;24(4):433–6.
[83] Hall AH, Dart R, Bogdan G. Sodium thiosulfate or hydroxocobalamin for the empiric
treatment of cyanide poisoning? Ann Emerg Med 2007;49(6):806–13.
[84] Borron SW, Baud FJ, Barriot P, et al. Prospective study of hydroxocobalamin for acute cy-
anide poisoning in smoke inhalation. Ann Emerg Med 2007;49(6):794–801, e2.
[85] Megarbane B, Delahaye A, Goldgran-Toledano D, et al. Antidotal treatment of cyanide
poisoning. J Chin Med Assoc 2003;66(4):193–203.
[86] Geller RJ, Barthold C, Saiers JA, et al. Pediatric cyanide poisoning: causes, manifestations,
management, and unmet needs. Pediatrics 2006;118(5):2146–58.
[87] DesLauriers CA, Burda AM, Wahl M. Hydroxocobalamin as a cyanide antidote. Am
J Ther 2006;13(2):161–5.
[88] Rowden AK, Norvell J, Eldridge DL, et al. Updates on acetaminophen toxicity. Med Clin
North Am 2005;89(6):1145–59.
[89] Chamberlain JM, Klein BL. A comprehensive review of naloxone for the emergency
physician. Am J Emerg Med 1994;12(6):650–60.
[90] Clarke SF, Dargan PI, Jones AL. Naloxone in opioid poisoning: walking the tightrope.
Emerg Med J 2005;22(9):612–6.
[91] Eyer P. The role of oximes in the management of organophosphorus pesticide poisoning.
Toxicol Rev 2003;22(3):165–90.
[92] Rotenberg JS, Newmark J. Nerve agent attacks on children: diagnosis and management.
Pediatrics 2003;112(3):648–58.
[93] Sidell F. Nerve agents. In: Sidell FR, Franz DR, editors. Medical aspects of chemical and
biological warfare. Washington, DC: Office of the Surgeon General at TMM publications;
1997. p. 129–80.
[94] Tush GM, Anstead MI. Pralidoxime continuous infusion in the treatment of organophos-
phate poisoning. Ann Pharmacother 1997;31(4):441–4.
 CRITICAL CARE TOXICOLOGY 739

[95] Pawar KS, Bhoite RR, Pillay CP, et al. Continuous pralidoxime infusion versus repeated
bolus injection to treat organophosphorus pesticide poisoning: a randomised controlled
trial. Lancet 2006;368(9553):2136–41.
[96] Farrar HC, Wells TG, Kearns GL. Use of continuous infusion of pralidoxime for treatment
of organophosphate poisoning in children. J Pediatr 1990;116(4):658–61.
[97] Holstege CP, Kirk M, Sidell FR. Chemical warfare: nerve agent poisoning. Crit Care Clin
1997;13(4):923–42.
[98] Bawaskar HS, Joshi SR. Organophosphorus poisoning in agricultural India: status in 2005
[comment]. J Assoc Physicians India 2005;53:422–4.
[99] Karlson-Stiber C, Persson H. Cytotoxic fungi: an overview. Toxicon 2003;42(4):339–49.
[100] Knapp JF, Johnson T, Alander S. Seizures in a 13-year-old girl. Pediatr Emerg Care 2003;
19(1):38–40.
[101] Lheureux P, Penaloza A, Gris M. Pyridoxine in clinical toxicology: a review. Eur J Emerg
Med 2005;12(2):78–85 [Review].