Toxic exposures present a unique set of issues to clinicians caring for children. Children commonly ingest household products and sometimes are exposed to medications that are taken by others living in the household. The clinician must be quick to recognize potentially dangerous exposures in order to intervene and prevent or mitigate adverse outcomes. Fortunately, the majority of pediatric ingestions involve unintentional exposure to small doses of nontoxic or minimally toxic substances. However, it is those few truly toxic exposures that demonstrate the invaluable role played by a skilled emergency physician. This review will primarily focus on pediatric poison exposures in children ages six years and under.
At 6:15 pm, nearing the end of your shift, the triage nurse carries in a 12-monthold boy whose panic-stricken mother reports he ingested some "hair relaxer" about 30 minutes earlier. The child is screaming and is noted to be drooling with some swelling of the lower lip. The mother had tried to rinse the child's mouth with water prior to arrival in the ED. You realize that the rest of this child's life is in your hands: the right actions may be lifesaving; the wrong actions may be devastating. ..
Pediatric poisoning has been the focus of a great deal of recently-published literature. This includes studies of varying complexityand quality as well as practice guidelines and topic reviews. Research in the area of pediatric poisoning is continuously published, but, unfortunately, the strength of many studies in this field is severely limited due to the study designs and inherent limitations of data in the databases referenced. The preponderance of such studies are produced by retrospectively mining data without laboratory corroboration of poison exposure and with follow up limited to the passive surveillance method of relying in clinicians to voluntarily report clinical outcomes. Several practice guidelines addressing the management of poisoned patients have been created by organizations whose members have expertise in the areas of pediatrics and medical toxicology.1-8 These are summarized in Table 1.
Herein, we review the most relevant and current literature and evidence concerning the diagnosis and management of pediatric poisonings and appraise diagnostic methodologies and therapeutic strategies. The literature search involved searching Medline for articles publishedfrom 1966 to July 2006 with the earch strategy: abstract word=poison, poisoning, toxin, toxic, intoxication, or toxicology.
The search strategy also included manually searching the table of contents and abstracts of journals published between 1996 and July 2006 for articles deemed relevant to the topic of pediatric poisoning. Additionally, the abstracts of annual U.S. national academic meetings between 2000 and October 2006 were searched for abstracts relevant to the topic.
Due to the widespread availability of databases that can be readily accessed and searched, there is broad publication of observational studies describing the effects or outcomes of poison exposures in children. Unfortunately, the value of most such studies is very limited due to their inherent design shortcomings. Very few involve laboratory confirmation of exposure to a substance by blood, urine, or other test. Since, in most cases, no clinical exposure to the substance occurred, there is understandably no adverse toxicologic event or clinical symptom.
Most studies of poison exposure are retrospective and are conducted using databases of local poison control centers, national Toxic Exposure Surveillance System (TESS) data, or other data sources. They rely on reports of exposure and reports of symptoms or clinical effects presumed to be the result of the exposure. "Exposure," as defined by the American Association of Poison Control Centers, is presumed contact with the substance in question by the following routes: dermal, ingestion, inhalation, parenteral, ocular, otic, aspiration, vaginal, rectal, other, and unknown.
An example of the prevalence of descriptive studies is the following: at the 2005 North American Congress of Clinical Toxicology, which is the annual congress of the American Academy of Clinical Toxicology, 19 published studies of clinical effects or outcomes after poison exposure including children in the study populations were presented9 and the 2006 congress had 24 such studies.10
Each of these studies involved substances for which laboratory testing of blood or urine could have definitively corroborated or confirmed exposure. In 2005, only three of these 19 reports, of tricyclic antidepressants,11 flecanide,12 and aspirin13 respectively, included laboratory confirmation of poison exposure. In 2006, only two of the 24 studies included lab confirmation of exposure, these involving lead 14 and loratidine.15 Each year, only one study with laboratory confirmation of exposure was prospective. This information serves to illustrate that there are a plethora of pediatric poisoning reports, including outcome studies, but that adequate scientific rigor is lacking in the preponderance.
With most pediatric exposures, it is unclear if any true contact with the substance occurred. An example would be a toddler found in a bathroom with a spilled bottle of medication on the floor. The only manner in which this is reported to a poison control center is as an "exposure" though, in most cases, no ingestion occurred.
Observational studies that do not involve laboratory corroboration of poison exposure are the most common type of manuscript publication addressing pediatric poison exposures. Numerous evaluations of dangerous and potentially lethal poisons, such as clonidine,16 calcium channel blockers,17 buproprion,18 sulfonylurea hypoglycemic tablets,19 benzocaine,20 beta blockers, 21and many other substances, have been published. It is unclear if any patient in these series had a true ingestion of the poison in question, making extrapolation of the results of these studies questionable, if not dangerous.
The Toxic Exposure Surveillance System is a database of reported poisoning information from 63 poison centers across 40 states. Summary data from exposures reported to TESS is the most comprehensive and largest collected in the world. While TESS data generated from poison centers are quite comprehensive, there are some weaknesses. Most importantly, the exposures are collected by a passive surveillance system. All reports are voluntary and can be made anonymously. These exposures are not confirmed in order to be entered into the TESS database. This system of collecting data might overestimate or underestimate the true scope of poisoning exposures as well as fail to detect adverse outcomes, including severe morbidity or mortality.
According to a study done in Spain, pediatric presentations due to poison exposures make up about 0.28% of all visits to emergency departments.22 Similar findings have been reported in the United States. 23 Annual summary statistics from TESS recorded 2.4 million toxic exposures in 2004 and 2.4 million in 2005. In 2004, 51% of all exposures occurred in children younger than six years of age, with only 27 fatalities in this group; in 2005, 50% of all exposures occurred in children younger than six years of age, with 24 fatalities in this group, see Table 2.
Most poison exposures reported in any given year involve younger children, and, fortunately, these are rarely fatal. This is likely due to a combination of no intent of self-harm, a small or single dose ingested with no more ingested after an initial adverse taste, and poison prevention efforts that have improved the safety of medication and poison storage in the home.
Exposures in older children, adolescents in particular, are similar to exposures of adults.24 These often involve intentional ingestion of medications and illicit drugs and frequently involve more severe toxicity. Older children and adolescents exhibit adult patterns of ingestion with suicidal ideation and the use of alcohol, resulting in much higher rates of morbidity and mortality.25 About 7% and 6.1% of all reported intentional poison exposures occurred in the 13-19 age group, with 90 and 70 known fatalities recorded in 200426 and 2005,27 respectively. Although younger children comprise the largest total number of children who have poison exposures, they result in fatality much less frequently. Unless otherwise specified, this review will not address intentional adolescent poison exposures, which are essentially no different from adult poison exposures.
Advances in childproof safety packaging, increased poison prevention efforts with resulting increased awareness and education, as well as the use of safer pharmaceutical agents have been responsible for the declining poisoning morbidity and mortality rates in children.28 Notably, fatalities in children were more often due to therapeutic errors, environmental exposures and envenomations, adverse drug reactions, and malicious exposures rather than unintentional household exposures. However, there are poisons that can be dangerous to a child when taken in only a single or small dose, see Table 3.
Neonatal poisoning secondary to breast milk exposure has been described for a variety of agents, including aspirin, acebutolol, atenolol, cocaine, and lithium.29 Neonates and young infants have ignificant differences in drug metabolism pathways.
Toxin-induced symptoms can vary widely from mild symptoms of nausea or headache to respiratory or cardiac arrest. The most common symptoms of poison exposure are nonspecific and minor. Less commonly encountered problems associated with severe poisoning include altered mental status, seizures, or cardiovascular compromise. Decreased level of consciousness is a serious complication of poisoning. Distinguishing poisoning from other medical diagnoses, such as electrolyte and glucose derangements, trauma, or infection, requires clinical acumen. The basic management strategies for patients with severe illnesses resulting from poisoning are usually the same as management strategies for severe illnesses of other etiology. Ensuring patency of airway, adequacy of breathing, and cardiovascular competency is fundamental.
When present, specific signs and symptoms may guide the physician to a clinical diagnosis. Recognition of several distinct syndromes associated with specific classes of poisons often help guide the treatment of the poisoned patient. The classic four toxidromes cholinergic, anticholinergic, sympathomimetic, and opioid may be present in children. These findings are most reliable if a single drug has been ingested. Exposure to multiple agents can decrease the utility of trying to identify a toxidrome because the combination of agents may cause conflicting patterns of clinical findings. Additionally, certain agents within a particular group may produce atypical findings. Examples might include propoxyphene, which is a synthetic opioid that may cause coma and respiratory depression but is known to cause pupillary dilation in overdose rather than miosis typically associated with the opioid toxidrome. Additionally, propoxyphene may cause cardiac dysrhythmia and seizures, serious toxicities not associated with the opioid toxidrome. Tramadolinduced seizures and cyclical depression of mental status are other examples of clinical effects of opioids that lay outside of the classic opioid toxidrome.
Transportation by ambulance may result in more rapid initiation of gastrointestinal decontamination compared to other means of transportation.30 In a retrospective study of 167 patients, the median interval from arrival to gastrointestinal decontamination (lavage or charcoal) for patients transported by ambulance was shorter than for patients who arrived by other means at 55 and 73 minutes, respectively. This difference was largely related to more rapid gastric lavage, and there was no difference in admission rates between the two groups. It is reasonable to extrapolate these results and infer that therapeutic interventions likely to be used in cases of poisoning, such as orogastric lavage, nasogastric aspiration, use of activated charcoal, or antidote use, will be applied sooner if the patient is transported by ambulance. Certainly, for any ingestion that carries a risk of cardiorespiratory compromise, it is advised to transport the patient to the hospital by ambulance rather than by laypersons. If prehospital personnel are involved in a case of poisoning, they might initiate treatments that otherwise would be delayed until hospital contact, particularly activated charcoal administration. Essential lifesaving treatment and ensuring competency of airway, breathing, and circulation, will also be provided. In addition, hypoglycemia may be diagnosed and treated in the field.
Prehospital personnel should examine the scene thoroughly for any potential medications or household toxins that the child may have ingested and bring the container to the hospital.
Current American Academy of Pediatric (AAP) guidelines emphasize that prevention strategies be communicated to new parents at prenatal and well-infant visits. Since prevention of all toxic exposures is unlikely, parents should also be instructed what to do in case of such an event. If the victim has collapsed or has stopped breathing, 911 should be called immediately. In other circumstances, the caregiver should contact a poison control center for help by telephoning 800-222-1222.
Ipecac is an extremely effective emetic agent that decreases absorption of ingested drugs.31 However, due to the lack of evidence of improved clinical outcomes with ipecac use and recognition of its risks and adverse effects, the use of ipecac has steadily declined.32 A recent study examined the use of ipecac at home and concluded that it does not improve patient outcome nor does it decrease resource utilization.33 In 2003, the American Academy of Pediatrics recommended that ipecac no longer be routinely used in pediatric poisonings.
The clinician must first determine whether the patient is critically ill and needs immediate life saving intervention. While the majority of exposed children will have minimal or no symptoms and normal vital signs, it is imperative to rapidly stabilize any patient who is unconscious, has abnormal vitals, or is seizing. Immediate treatment of symptomatic hypoglycemia includes the administration of oral glucose if possible, as well as a bolus of 1 gm/kg of glucose. This may be 10 mL/kg of 10% glucose, 4 mL/kg of 25% glucose solution, or 2 mL/kg of 50% glucose. It is recommended that 25% or 50% glucose solutions only be Administered via a central venous line. Diluting these solutions is preferable to obtaining central venous access simply for the purpose of administering glucose.
A seldom needed but critical initial interventionis external decontamination. The list of poisons for which decontamination should occur prior to entrance to the ED includes organophosphate agents, such as pesticides, liquid chlorine, caustics, and other agents. The reason for decontamination outside the ED is twofold. Decontamination is indicated to prevent or limit toxicity of the exposed patient and to prevent contamination of the ED.34
Historical data is usually obtained from the caregiver and occasionally from prehospital personnel. ideally, the following would be known: the actual poison or container from which the poison was held, the name of the medication or toxin, the time of exposure, the amount taken, and the names of other medications or toxins available to the child, as well as symptomatology and progression of signs and symptoms. When calculating a possible dose involved, it is prudent to assume the maximal dose consistent with the circumstances of the exposure; remember that, with certain poisons, even a single pill or small dose could be dangerous or lethal to a child,35 see Table 6.
For certain toxins, there are specific questions about symptoms that may be of particular help with diagnosis or prognosis. For example, coughing or vomiting after hydrocarbon exposure, vomiting or drooling after caustic exposure, or loss of consciousness after carbon monoxide exposure would all potentially alter subsequent management.
Two situations merit special mention because they require particular attention and care with regard to history taking: Munchausen's Syndrome by Proxy and pediatric body packing.
This is a form of child abuse where the caretaker deliberately injures the child surreptitiously, in this circumstance by poisoning, for some form of secondary gain. Suspect this form of child abuse when childhood poisoning is associated with bizarre or inexplicable presentations, repeated exposures, unusual poisons, or the reported circumstances of poisoning are not consistent with the child's motor skills. Munchausen's Syndrome by Proxy involving poisoning is a situation in which laboratory testing and the involvement of child protection and law enforcement authorities must be recruited.36
Involvement of a clinical toxicologist in such cases is highly recommended, both because they are likely to contribute to a carefully considered plan that will both yield information relevant to the clinical management of the poisoning as well as forensic information useful for legal protection of the child and action against the abuser. Cases of malicious poisoning in children involve a vast array of poisons administered by varied routes, making diagnosis even more difficult than it typically would be. Examples of malicious poisoning involve iron,37 diphenhydramine,38 ipecac,39-41 and rodenticide,42 but this phenomenon could occur with any type of poison.
Though the practice of "body packing" has been known for many years, only recently has this occurrence in children been recognized and documented.43 Body packers, also known as "mules," ingest large numbers of well-sealed packets commonly containing cocaine or heroin for the purpose of drug trafficking. The packets may cause GI obstruction, and if they perforate in the GI tract, the leaking contents may result in severe systemic toxicity. In one report, two children ages 12 and 16 were used to smuggle heroin. One patient had one of the packets rupture and required naloxone infusion. Both patients were treated with aggressive GI decontamination and had an uneventful recovery. This first report of pediatric body packing prompts a high index of suspicion under certain circumstances in patients with a history of recent international travel, particularly from a location of drug origin, such as Latin America, Africa, or Southeast Asia.
In the emergency setting, it is occasionally necessary to initiate critical treatments before a full examination can be completed. In these cases, full examination will occur after initial stabilization and critical therapy. Serial examination is often necessary to evaluate the effectiveness of treatment, to essess for progression of illness, and to ultimately determine the patient's disposition. A focused physical examination in a poisoned child includes assessment of vital signs, mental status, and neurologic examination as well as examination of the eyes, mucus membranes, skin, bowel, and bladder for identification of specific toxidromes.
Frequently, the vital signs are extremely helpful in determining the toxidrome involved. Additionally, maintenance of vital signs within acceptable limits is a critical aspect of supportive care of the poisoned child. Table 7 summarizes vital sign abnormalities typical with certain types of poisons. A caveat to use of this information is that severe stress as well as decompensation may result in vital signs opposite of those typically associated with the poison in question. For example, exposure to cocaine or other sympathomimetic agents is typically associated with hypertension and/or tachycardia, but with decompensation, hypotension and bradycardia may result. It is helpful to remember that, when in extremis, the expected findings in the table might not be present.
The clinician should note whether alteration of mental status is present. First and foremost, any alteration of sensorium or neurologic status that might be consistent with hypoglycemia should prompt immediate bedside assessment of the serum glucose level. The physical examination may reveal specific findings, such as hallucinations, formications, altered muscle tone, tremor, fasciculations, twitching, dystonic posturing, impaired coordination, weakness, hyperreflexia, impaired coordination, cranial nerve deficit, or other neurologic abnormality. Focal neurologic findings or decerebrate or decorticate posturing should prompt evaluation for intracranial hemorrhage or lesion. It is critical to differentiate alterations of behavior or psychiatric findings from neurologic findings. Neurologic findings are much more concerning for CNS injury or lesion, and may warrant intensive imaging and management techniques that require ICU level care, whereas psychiatric effects are most often treated by providing a safe environment for the patient to stay in while the drug effects abate.
Substance withdrawal is not otherwise reviewed here; it is critical to note that alteration of mental status is common with benzodiazepine, barbiturate, and other sedative-hypnotic substance withdrawal. These withdrawal syndromes can be life threatening. Neonates are the age group most likely to present with substance withdrawal symptoms. In a neonate, severe changes in behavior may result from withdrawal from a spectrum of substances, including amphetamines, barbiturates, caffeine, cocaine, opioids, and selective serotonin reuptake inhibitors.
Recent publications have advocated the use of the Glasgow Coma Scale (GCS)44 or AVPU (alert / verbal / painful / unresponsive) scale in circumstances of poisoning.45 Evidence of the utility of the GCS for prognosis or management of poisoned patients is inadequate to support such a practice.46,47
Dilated pupils may suggest a sympathomimetic or anticholinergic toxidrome, whereas constricted pupils suggest exposure to opiates, cholinergic agents, clonidine, and imidazolines. Although, theoretically, sympathomimetic stimulants would result in a dilated reactive pupil and anticholinergic substances would result in a dilated nonreactive or sluggish pupil, we find that reactivity of the pupil does not reliably exclude anticholinergic toxicity.
Rotatory as well as horizontal nystagmus is commonly associated with toxicity from the anticonvulsants phenytoin and carbamazepine, as well as the related drugs ketamine, dextromethorphan, and phencyclidine (PCP), although other drugs may cause this effect. Absence of the finding does reliably exclude toxicity from the aforementioned agents, though the presence of nystagmus is helpful in raising he index of suspicion for them.
If there is suspicion of caustic exposure, the eyes should be examined for corneal abrasions with luoroscein stain after thorough irrigation and assessment of the pH. Twenty minutes after completion of irrigation, the pH should be rechecked to assure that all of the caustic substance is removed; this is particularly important with alkali exposures as they penetrate more rapidly and are not as readily neutralized by proteins of eye surface as acids. Irrigation by Morgan lens is recommended, as other methods, such as dripping fluid onto the eye by gravitydependent flow from IV tubing, are inadequate.
The patient should be undressed and carefully examined, noting the skin color, hydration status, and temperature. Table 8 summarizes findings that mayprovide a key to a final diagnosis.
When caring for the poisoned patient, it is helpful to conduct a dextrose assessment. Of the deleterious effects of poisons, hypoglycemia is the most easily detected and most readily remedied. Any patient with alteration of mental status or exposure to a poison that might cause hypoglycemia should have a bedside bloodglucose determination.
The mnemonic for initial stabilization of patients with poison exposure is "A,B,C,D,E," with "E" signifying ECG. An electrocardiogram should be obtained in any patient with altered mental status, abnormal heart rate or blood pressure, or exposure to a poison capable of causing cardiac dysrhythmia. Agents that could produce cardiac abnormalities include newer or atypical antipsychotics, phenothiazines, tricyclic antidepressants, digitalis, beta blockers, calcium channel blockers, clonidine, opioids, antidysrhythmics, alpha blockers, theophylline, cocaine and other sympathomimetics, and sodium-channel blockers.
In adults, exposure to tricyclic antidepressants is confirmed by finding R' in the terminal 40 ms in lead AVR, S' in AVL, and I.48 In cases when these terminal 40 ms changes and tachycardia are present, lengthening of the QRS interval beyond 100 ms is indicative of TCA toxicity. Since there is significant right axis deviation in children, it is unsafe to presume that terminal 40 ms changes have the same prognostic value in children as in adults.
Other agents that can cause widened QRS include TCAs, antidysrhythmics, phenothiazines, cocaine, diphenhydramine, propoxyphene, ethanol, and quinine. Bradycardia may suggest ingestion of digoxin, beta-blockers, calcium channel blockers, lithium, opioids, clonidine, or sedative-hypnotic drugs.
A discussion of the management of poisoned patients would be incomplete without addressing the issue of "toxicology screening." Discussing toxicology screening (TS) is difficult because there is no standard definition of what constitutes TS. Generally speaking, a "tox screen" involves qualitative detection of specific drugs or metabolites in urine or serum. The term is most frequently used to denote assay detection of commonly abused drugs, most frequently amphetamines, cocaine, marijuana, opioids, and phencyclidine (PCP), though many other abused drugs are detectable as well. The drugs of abuse mostly widely available in multiple substance assays are commonly called the "NIDA 5," an abbreviation indicating the National Institute of Drug Abuse 5-panel drug screen; these five drugs are amphetamines, cannabanoids (marijuana), cocaine, opioids, and phencyclidine. Benzodiazepines and sometimes barbiturates may be included in drug of abuse screening assays as well.
There are numerous limitations of drug of abuse screening. An example of how the traditional five drug panel is not adequate for contemporary use is the amphetamine screening assay. Amphetamines, typically pharmaceutical amphetamines, were widely abused at the time the drug screening panel was developed. However, screening for amphetamines will not reliably detect today's most widely used amphetamines, such as MDMA (ecstacy), methamphetamine, or methylphenidate.49,50 Unless a healthcare facility has taken the strides to update their drug of abuse screening assay, they will not detect the most widely abused amphetamines and instead will be screening for a relatively rarely abused amphetamine.
For a comprehensive review of drugs of abuse, see the May 2005 Emergency Medicine Practice article,
"Drugs of Abuse: Providing The Best In Evidence Based Care To "Self-Medicated" Patients" or the October 2006 Pediatric Emergency Medicine Practice article, "When Kids Do Drugs: Evaluation And Treatment In The Emergency Department."
Most importantly, use of drug of abuse screening assays has repeatedly been demonstrated to lack sefulness in the management of poisoned patients. In one retrospective study, 463 cases of TS in pediatric patients were reviewed.51 In this study, 50% of patients were under 11 years of age, and 16% were under five years of age. A TS assay testing for over 550 toxins produced 234 positive screening tests of which seven (3%) were positive without clinical suspicion of exposure. None of the positive results affected the medical management of the patients, while the total cost of the tests added up to $88,433. In another prospective study, treating physicians documented whether TS affected their management of suspected pediatric drug ingestions.52This study concluded that, while quantitative assays for certain drugs, such as acetaminophen, anticonvulsants and salicylates, may assist in patient management, qualitative screening of over 500 other toxins has not demonstrated a similar effect. Another retrospective review of TS in pediatric patients in an emergency department reached similar conclusions.53 Use of a "tox screen" does not improve patient outcome or change the care provided, but they incur significant cost. Although intuition may suggest that toxicology screening will aid in diagnosis and medical management of poisoned patients, this has never been demonstrated despite numerous studies evaluating TS in the emergency department setting.
A critical caveat to this is the use of urine toxicology screening for forensic purposes. In any circumstance in which it is suspected that a child has nonvolitionally been exposed to illicit drugs, screening for such drugs is warranted. This is distinct from the use of such screens for medical management, but may play a key role in the ultimate disposition of the patient and action taken by child welfare authorities. Forensic testing requires confirmation of a positive result with a second assay by a different methodology. Usually, this confirmation is by gas chromatography mass spectroscopy (GC-MS). Since this confirmation is not readily available and takes one to several days, the results of forensic testing are generally of no use in the disposition of ED patients.
There are some toxins for which quantitative assays can help guide clinical management, see Table 9.
With the exception of acetaminophen levels, these tests should not be used routinely, but ordered selectively when particular drug exposures are suspected. Since acetaminophen is a common coingestant and might not produce symptoms early after ingestion, it is prudent to routinely measure it in all patients who have been exposed to acetaminophen or in any patient who has ingested any substance with intent of self harm.
There is limited use for urine testing in the evaluation of a pediatric patient with a toxic exposure. Ferricchloride testing may help confirm the presence of salicylate in urine. A positive result occurs when ferric chloride in combination with urine turns purple. This test does not give any information about toxicity, and a quantitative serum salicylate level is indicated.
Urine microscopy is occasionally used to corroborate exposure to ethylene glycol. Ethylene glycol precipitates calcium oxalate crystals in the kidney and these may be seen on a routine microscopy. However, this finding may be absent early after ingestion and, if proper therapy is instituted, it may be not detected at all. Urinary fluorescence has been advocated as a method of screening for exposure to ethylene glycol, but such testing is unreliable as an indicator of ethylene glycol exposure in children.54
Urine that is heme-positive with no visible erythrocytes should alert the clinician to suspect hemolysis or myoglobinuria from rhabdomyolysis.
These should be obtained whenever the clinician suspects a toxic exposure that can produce an anion gap metabolic acidosis. Evaluation of the anion gap can help to rapidly narrow down the diagnosis. Anion gap is calculated easily from serum electrolytes.
Any value over 12 indicates the presence of elevated unmeasured anions. There are several causes of increased gap: ketones, uremia, lactate, and toxins. The first three categories can be easily ruled out with routine tests, and, at this point, the presence of an unexplained anion gap should prompt the clinician to consider toxins. In these patients, the osmol gap should be calculated. If a specific diagnosis is not made, measurement of the osmolal gap should be performed to investigate for toxic alcohol exposure.
An osmol gap greater than 10 indicates the presence of unmeasured osmoles, such as methanol, ethylene glycol, or isopropanol. The clinician can estimate the serum level of these substances by using the formula. While an increased gap is helpful, an osmol gap less than 10 does not rule out the possibility of these substances being present because different individuals have a wide variation at baseline and, depending on the time of presentation, the parent compound may have been largely metabolized, leaving little parent compound in the serum.
The decision of whether to order radiological studies depends on the particular clinical scenario and the suspected toxin. Most ingested agents are radiolucent and do not appear on a routine radiograph. Notable exceptions to this rule can be remembered by the mnemonic COINS: Chloral hydrate, Calcium, Opiate (or other drug) packets, Iron and other metals, Neuroleptic agents, and Sustained-release and enteric coated preparations. Radiographs can also help diagnose other radio-opaque ingested or spirated foreign bodies. Radiological studies may be helpful in diagnosing serious sequellae of toxic exposures, such as chemical or aspiration pneumonitis.
Once the ABCDE's are addressed, the next most important determination is whether gastric econtamination is indicated and, if so, what mode of decontamination should be used. Most pediatric ingestions involve a non-toxic substance. These low risk patients are managed without decontamination. If gastric decontamination is considered appropriate, the emergency physician has several choices: syrup of ipecac, orogastric lavage, nasogastric lavage, adsorption with activated charcoal, and whole bowel irrigation.
1997, after critically appraising the literature on ipecac, the American Academy of Clinical Toxicology (AACT) and the European Association of Poison Centres and Clinical Toxicologists (EAPCCT) issued an unequivocal position statement recommending that ipecac not be administered routinely in the anagement of poisoned patients, but acknowledged that there is insufficient data to support or exclude its use soon after poison ingestion.
Syrup of ipecac may be helpful in few select cases where the toxin may cause severe toxicity or fatality, is not well adsorbed to charcoal, and the pills are too big to pass through a gastric lavage tube. 55 examples of such poisons might include lithium or iron tablets
Syrup of ipecac is absolutely contraindicated in any patient with depressed mental status or who could potentially become unstable within the hour. Although generally considered safe, ipecac induced emesis can cause serious adverse effects, such as gastric rupture, traumatic diaphragmatic hernia, and aspiration pneumonia.
Orogastric lavage (OGL) is another method of gastric emptying. It should not be employed routinely in patients with mild or moderate risk ingestions, but rather for patients who present within the first hour after a potentially life threatening overdose. Examples of poisons for which OGL might be indicated include cyanide, calcium-channel blockers, colchicine, chloroquine, tricyclic antidepressants, or salicylates.
Concomitant with the recognition that the apparent benefit of OGL is limited to an extremely small number of patients, the procedure has fallen out of favor and is uncommonly practiced. In the practice of pediatric emergency medicine, use of orogastric lavage is exceedingly rare and it is essentially a lost art.56 OGL carries the risk of significant complications, such as aspiration, esophageal perforation, hypothermia, and hyponatremia. OGL is generally contraindicated for hydrocarbon ingestions as most hydrocarbons do not cause serious toxicity once they have passed the stomach, and can cause severe morbidity if aspirated. The exceptions to this rule include highly toxic substances contained in a hydrocarbon carrier, such as pesticides, paraquat, halogenated and aromatic hydrocarbons, camphor and heavy metals, that require aggressive GI decontamination. Orogastric or nasogastric lavage may be appropriate in these patients.
OGL is absolutely contraindicated for ingestion of caustics because of the risk of esophageal perforation as well as the risk of creating a more serious burn of the esophagus or airway. Perforation may occur due to a weakened esophageal or gastric wall that has been damaged by the caustic. Regurgitation of stomach contents with gastric lavage is a known complication of the procedure. If the regurgitated material contacts the esophagus or airway, it may result in more severe damage than was present at the time of presentation.
Only a physician experienced with OGL should perform the procedure in the limited circumstances when it is indicated. Patients with decreased level of consciousness or decreased ability to protect their airway should be endotracheally intubated prior to the procedure, but it can be preformed in alert and awake patients without endotracheal intubation. Patients should be positioned in left lateral decubitus with the stretcher tilted 20 degrees head down. 24 Fr orogastric tubes are recommended for toddlers and 36 Fr for adolescents.57
Nasogastric lavage for removal of liquid poisons does not carry the same risk of OGL . It has been emonstrated to be efficacious in reducing the amount of drug absorbed in adult human volunteers,58 but it is unclear if this technique has any utility in children who have ingested a liquid poison. It is our opinion that this method of nasogastric lavage be reserved for pediatric ingestion of a liquid poison capable of causing severe morbidity or mortality.
Because of its perceived safety and ease of administration, activated charcoal (AC) has become the most popular modality of gastrointestinal decontamination. It works by adsorbing the toxin to its surface, thus preventing systemic absorption. A higher ratio of AC to toxin is considered more effective for decontamination. A ratio of 10:1 of AC to toxin is considered to be the minimal dose necessary. 1 g/kg is the standard pediatric dose of AC. If additional doses are used, these should be given without cathartics. The use of cathartics is not routinely recommended with the first dose of AC, but if the AC formulation available is pre-mixed with a cathartic, this is acceptable for one-time administration.
The binding ability of AC diminishes progressively as time since ingestion increases, and it is for this reason that, if it is to be used, AC administration is recommended within the first hour after ingestion. When the decision is made to administer activated charcoal beyond one hour after exposure, every effort must be made to decrease the "door to charcoal window."59
Several compounds are amenable to "gut dialysis" with AC.60 Gut dialysis is the phenomenon whereby drug contained in the blood compartment diffuses across the gut into AC that has been orally administered. For such substances, repeat dosing of AC, called multiple-dose activated charcoal, may be helpful. Multiple-dose activated charcoal use is safe but does carry an additional risk of bowel bstruction if given excessively or in the presence of ileus.61 Multiple-dose activated charcoal is given 1g/kg every four hours for four doses; further dosing should be done only under the direction of a toxicologist.
The main complications of AC administration are vomiting and, rarely, aspiration with consequent charcoal laryngitis62 or pneumonitis.63 In a prospective study of 275 children receiving AC in a pediatric ED, there was one case of pulmonary aspiration.64
Those who vomit prior to AC administration or who have nasogastric tubes have the highest risk of vomiting; the use of sorbitol does not seem to increase this risk.65 Because of this real and potentially serious adverse affect, AC should be given only when it is certainly indicated, and its routine administration in non-toxic or minimally-toxic ingestions should be avoided. When given by NG tube, verification of position in the stomach is critical to ensure that iatrogenic pulmonary damage does not occur.66 Commercial preparations combine charcoal with a cathartic, such as sorbitol, to make it more palatable and promote GI motility. Cathartics by themselves are not an acceptable method of GI decontamination and should never be used alone.67 The recommended dose of sorbitol is 1-2 mg/kg; it is not indicated in children less then one year of age. The addition of cathartics has not been shown to increase the efficacy of AC, decrease bioavailability of ingested substances, or improve outcomes. Because of the potential risks and lack of benefit, these agents should not be used. The addition of cola to AC to make the mixture more palatable is an acceptable practice.68
The objective of this method of GI decontamination is propelling the toxic agent through the entire GI tract before it is fully absorbed. It should be considered in patients who ingested enteric-coated pills or substances poorly adsorbed to charcoal, such as lithium, iron, or other metals. It is also used in body packers and body stuffers to accelerate the passage of drug filled packets.69
Whole bowel irrigation is accomplished by continuous administration of polyethylene glycol solution (Go-Lytely) orally or via the NG tube until the rectal effluent is clear. The pediatric dose is 25 cc/kg/hour up to 500 mL/hour in school-aged children and 1liter/hour in adolescents. When preformed correctly, the procedure usually lasts at least four to six hours.70
The dose-limiting adverse effects associated with this procedure include vomiting, abdominal cramps,and bloating. Antiemetics, including metoclopramide or ondansetron, may be necessary. It should be noted that successful WBI is typically accompanied by no discernible clinical effect, so it may be easily mistaken as being unnecessary or unhelpful.
In addition to the previously mentioned elimination techniques, many toxins exposures have specific antidotes, see Table 14
Acetaminophen is one of the most commonly used medications in children and remains a leading cause of poisoning deaths.71 In overdose, it can cause hepatic damage and fulminant hepatic failure. Children, however, may have a decreased risk for hepatotoxicity following a moderate overdose due to their greater ability to metabolize acetaminophen via hepatic oxidation and sulfation.72,73,74 It is thought that, since the oxidative pathways are much more active in children, detoxification might be much more efficient in children.75,76
N-acetylcysteine has been shown to restore glutathione levels as well as acting as an alternate substrate to conjugate and thus detoxify NAPQI. Oral N-acetylcysteine has historically been the mainstay of US treatment of acetaminophen toxicity.77 Recent experience with the intravenous administration of NAC has highlighted the ease of administration, efficacy, and greater drug tolerance with fewer GI effects.
In a study by Yip et al, IV dosing was found as effective as oral dosing in adult patients.78 Randomized controlled trials evaluating the safety and efficacy of IV NAC in pediatric patients are lacking. A clinical trial of 25 pediatric patients has shown it to be just as effective as the oral formulation. 79 In another chart review study of 89 pediatric patients receiving either oral NAC or intravenous NAC, the use of IV NAC was associated with shorter hospitalizations and a trend toward fewer side effects, although overall cost of either treatment was the same.80 Other side effects listed in adult retrospective chart studies were mostly benign cutaneous reactions that were treated with antihistamines.81
The intravenous formulation of N-acetylcysteine (Acetadote) was approved by the FDA and made available in March 2004 for the treatment of acetaminophen overdose. The recommended protocol calls for 20 hours of treatment with an initial loading dose of 150 mg/kg in 200 mL D5W over 15 minutes. This is followed by 50 mg/kg in 500 mL D5W over four hours then 100 mg/kg in 1 liter D5W over the next 16 hours. The use of varying concentrations of the IV formulation have been associated with calculation errors made by healthcare providers administering the antidote.82,83
The manufacturer has not yet submitted the safety data for administration of this agent to patients under the age of 16. Serious concerns have been raised about dosing in children as the large volume administered has the risk of causing volume overload, hyponatremia, and subsequent seizures.84 The manufacturer has specific recommendations for limiting the quantity of fluid administered to patients weighing less than 40 kg as well as patients who require fluid restriction. The convention is to dilute 20% NAC to a final concentration of 40 mg/mL. The final milligram per kilogram dosing is the same; the free water is less than in the adult schedule.85 The website www.acetadote.com contains a tool for determining the quantity of fluid to dilute the antidote for patients of various weights.
Exposure to the toxic alcohols ethylene glycol and methanol may be managed using the alcohol dehydrogenase inhibitor fomepizole. Ethylene glycol is commonly found in antifreeze and other de-icing solutions; it may be ingested by children due to its sweet taste. The metabolic pathway leading to formation of toxic metabolites begins with the enzyme alcohol dehydrogenase. Methanol is a component found in windshield washer fluid. Acommon solvent, its primary metabolite via alcohol dehydrogenase is formic acid. The metabolites of ethylene glycol and methanol can result in anion-gap metabolic acidosis and fatality
Current treatment options for toxic alcohol poisoning include fomepizole versus the historically-used ethanol. Ethanol is given as a 10% solution with an IV loading dose of 10 mL/kg, followed by 1 to 2 mL/kg/h. An ethanol level of 100 mg/dL is sufficient to block toxic alcohol metabolism. Serum ethanol levels and glucose levels must be frequently checked under this treatment. Recent studies of ethanol use in children point to the greater safety in its use with lower than predicted rates of adverse outcomes.86 Fomepizole is a newer antidote that acts as a competitive inhibitor of alcohol dehydrogenase. It is currently approved by the FDA for the treatment of both methanol and ethylene alycol ingestions in adults. Though these toxicities are uncommon in children, a number of case reports have demonstrated fomepizole effectiveness in children for both ethylene glycol and methanol toxicity.87,88 The advantages of using fomepizole are that there are no levels to monitor and it does not cause hypoglycemia or inebriation, which are known side-effects of ethanol. Fomepizole should be administered as soon as possible after suspected ethylene glycol or methanol ingestion. A case series of 11 patients with methanol poisoning reported a loading intravenous dose of fomepizole of 15 mg/kg followed by doses of 10 mg/kg every 12 hours for the first four doses which is then increased to 15 mg/kg every 12 hours.89 Dosing is continued until ethylene glycol/methanol levels are below 20 mg/dL and the patient is asymptomatic with a normal pH.
Many centers will undertake a combination approach to the treatment of ethylene glycol/methanol toxicity and initiate hemodialysis in order to eliminate the toxic metabolites that have already been formed or because the half-life of methanol is too long to wait for natural clearance in patients treated with fomepizole. Hemodialysis should be considered for patients with significant metabolic acidosis (pH less than 7.25-7.30), development of change in vision, deteriorating vital signs despite intensive supportive care, renal failure, electrolyte imbalance unresponsive to conventional therapy, or a serum ethylene glycol/methanol concentration greater than 50 mg/dL.
Although evidence is still limited, a recent retrospective study of six pediatric patients who underwent either ethanol and/or fomepizole therapy without the use of hemodialysis for severe ethylene glycol toxicity demonstrated successful treatment without any long-term adverse effects at the time of discharge.90 Table 15 proposes the indications for fomepizole and its administration.91
Octreotide has been used since 1993 for the treatment of hypoglycemia caused by sulfonylureas.92 mall children who ingest even one pill are at risk for profound hypoglycemia that can ultimately result in permanent neurological disability or even death.93,94 However, there are sparse prospective data to truly ascertain the effect of these agents on children, and treatment guidelines are largely based on case reports, retrospective reviews, and poison centerdata. Table 16 provides a description of the protocol for octreotide therapy.
When caring for children who have ingested sulfonylurea hypoglycemic agents, it is critical to be vigilant to detect late-onset hypoglycemia. Depending on the agent in question, hypoglycemia may occur after many hours of compensated euglycemia.95 Normal blood sugar in the first hours after ingestion should provide no assurance that hypoglycemia will not develop. For this reason, all children with suspected ingestions who are not treated with octreotide should have hourly glucose assessment for a minimum of 24 hours after the time of ingestion.
Calcium channel-blocking (CCB) agents are widely used for the treatment of high blood pressure and angina. Toxicity from this class of drug includes bradycardia, hypotension, cardiac conduction delay, hypoinsulinemia, hyperglycemia, and elevation of serum lactate.96 Commonly-used CCB agents include conduction-modulating drugs, such as verapamil and diltiazem, and newer dihydropyridine drugs, such as nifedipine and amlodipine, that have more peripheral cardiovascular activity. Several CCB agents are available in sustained-release formulations.
Traditionally, treatment of CCB overdose and toxicity has involved the administration of intravenous fluids (atropine, glucagon, and parenteral calcium) and pressors (dopamine, dobutamine, and norepinephrine).97,98 These therapies are largely ineffective in severely poisoned patients.99 Calcium channel blockers are known to be potentially deadly to children, even if taken in a single or small dose. 100
Recent animal data and a growing body of clinical experience in humans has led to the support and advocacy of high-dose insulin/euglycemia (HIE) therapy for CCB toxicity.101,102 A comprehensive case series demonstrated survival without sequelae of highest recorded verapamil concentrations for which HIE was administered when standard therapies failed.103 Other small case series corroborate these findings.104
Hypoinsulinemia, such as that seen in CCB toxicity, compromises the utilization of glucose and worsens myocardial depression, leading to further hemodynamic instability. Administration of HIE counteracts this downhill cascade of events, improving inotropy and peripheral vascular resistance and has been anecdotally shown to improve cardiac ejection fraction.105
In cases of CCB toxicity, HIE therapy results in improved arterial blood pressure and pH. Bradycardia, heart block, and conduction delay might not respond to HIE therapy or may respond after a delay of 30-60 minutes subsequent to initiation of HIE therapy. Administration of insulin in HIE therapy must be accompanied by maintenance of euglycemia. Interestingly, not all patients who receive an insulin infusion for calcium channel blocker toxicity require supplemental glucose. Despite this, it is absolutely necessary to frequently monitor blood glucose levels. Complications of HIE therapy include hypokalemia during therapy and hypotension after the discontinuation of HIE therapy. Due to limited literature describing the benefits of HIE therapy, there is considerable variation in the recommended protocols for HIE therapy by toxicologists. However, most do agree that HIE should be considered in the treatment of CCB toxicity that does not respond to other therapies. Further studies will elucidate whether it should be advanced to initial therapy. Table 17 summarizes a protocol for HIE treatment.106
In December 2006, the United States FDA approved hydroxocobalamin, a vitamin B12 precursor, for use as antidotal therapy for cyanide poisoning. Based on the 2005 AAPCC TESS data, cyanide poisoning resulting in mortality occurs infrequently in the pediatric population. There were 4716 exposures to cyanide and cyanogenic compounds. There were 3825 pediatric exposures (less than 19 years of age), accounting for 81% of the total exposures. Four of six cyanide related deaths were due to adult intentional suicidal behavior. An adult and a seven-year-old child died from environmental exposure. Cyanide poisoning is a complication known to occur in patients who have been in closed-space fires. It is likely that cyanide exposures in closed-space fires often go unrecognized or misdiagnosed as carbon monoxide poisoning.
Cyanide is a cellular poison that inhibits cytochrome oxidase aa3, resulting in cellular hypoxia. Manifestations of cyanide poisoning in adults and children occur in multiple organ systems and include headache, dizziness, confusion, seizures, coma, hypotension, dysrhythmias, cardiovascular collapse, tachypnea or apnea, venous hyperoxemia, nausea, vomiting, abdominal pain, cherry-red skin color, and bitter almond breath. The pediatric population may be more susceptible to inhalational cyanide poisoning with their higher respiratory rate, lower body mass, and immature metabolic mechanisms.107 Traditionally in the U.S., oxygen therapy and the cyanide antidote kit have been used to treat cyanide poisoning. The cyanide antidote kit that is currently available, often called a "Lilly" kit due to the fact that they were manufactured by Lilly Company, has shortcomings. First and foremost is the creation of a methemoglobinemic state when used properly. Use of amyl nitrate and sodium nitrite results in ethemoglobinemia. The purpose of this is to bind cyanide as cyanomethemoglobin. Understanding the implications of methemoglobinemia in patients who suffer from cyanide exposure in a closed-space fire has led many to avoid using the nitrate/nitrite portion of the kit. Inhalation of fire smoke can lead to carbon monoxide poisoning and methemoglobinemia as well as cyanide poisoning.
In the circumstance of possible or confirmed carbon monoxide toxicity with possible cyanide toxicity, we recommend using only the sodium thiosulfate portion of the cyanide antidote kit. Sodium thiosulfate serves as the sulfur donor in the rhodanesemediated transformation of cyanide to thiocyanate, an irreversible reaction that creates a renally eliminated harmless compound, except in the setting of acute renal failure. Antidotal hydroxocobalamin allows clinicians to circumvent all the shortcomings of the traditional cyanide antidote kit.
Hydroxocobalamin, a vitamin B12 precursor, has been used as antidotal therapy for cyanide toxicity in France since 1996. The antidote contains a cobalt atom that chelates cyanide forming cyanocobalamin, a natural form of vitamin B12, which is excreted by the kidneys. The standard dose of hydroxocobalamin is 5 grams. This dose can be repeated for severe cyanide toxicity for a total of 10 grams. The licensed pediatric dose in France is 70 mg/kg. There are a few side effects associated with hydroxocobalamin, such as allergic reactions and erythema of the skin, mucous membranes, and urine, which are due to the reddish color of the hydroxocobalamin molecule. There are no reported hemodynamic or ardiovascular adverse effects, except for mild increases in blood pressure which may be related to hydroxocobalamin's ability to scavenge nitric oxide.108
As this agent is expected to first be available in the US market in spring 2007, it is impossible to predict how widespread its use will be or how its use will be recommended by authoritative sources. It is likely that cost may be a factor in dictating how widespread its use is and whether clinicians shift away from sodium thiosulfate or the traditional nitrate/nitrite/thiosulfate cyanide antidote kit. This antidote is eagerly awaited and it is anticipated that hydroxocobalamin will assume a prominent role in the treatment of cyanide toxicity.
An improved understanding of the physiologic etiology of toxin-induced seizures as well as the availability of effective anticonvulsants warrants review of this subject.109 Anticonvulsant therapy of a toxin-induced seizure best begins with the use of a benzodiazepine in the same manner as other seizures. Following a benzodiazepine, management of a toxin-induced seizure deviates from management of seizures from other causes. The first difference is that, in cases of unknown toxin, an order for pyridoxine should be made to the pharmacy. Pyridoxine is a GABA precursor, and its deficiency after exposure to hydrazines, such as isoniazid, Gyrometra mushrooms, or hydrazines in rocket fuel, are causes of status epilepticus, including nonconvulsive status. Pyridoxine should be considered as empiric therapy for status epilepticus of unknown etiology in any patient, and pyridoxine administration is the first-line therapy for seizures resulting from hydrazine poisons.
Since the dose required is massive, 70 mk/kg IV or dosing on a gram-per-gram equivalent with the quantity of hydrazine ingested, the order for the drug should be placed to allow time for it to be retrieved from the pharmacy, as well as time to draw the drug out of its vials. Empiric dosing of 5 gm in an adult would require drawing the contents of 50 unit-dose vials into a syringe for administration. In cases of empiric therapy for status epilepticus not known to be the result of hydrazine poisoning, it may be worth considering bringing the pyridoxine to the ED before taking the effort to draw up the contents of the vials. In the event the seizure terminates and the pyridoxine is not used for administration, it is of such a quantity that it may easily exhaust the entire hospital supply.
In addition to the deviation of placing an order for pyridoxine, another deviation from routine seizure management is total avoidance of phenytoin or fosphenytoin. Phenytoin and fosphenytoin are contraindicated in the treatment of toxin-induced seizures. When phenytoin has been evaluated for use in the management of toxin-induced seizures or toxins that induce seizures, phenytoin has been elated to increased morbidity and mortality.110,111 After a benzodiazepine, barbiturates are the second-line drug for managing toxin-induced seizures. The most commonly used barbiturates are phenobarbital and pentobarbital. Each is widely used in the pediatric emergency setting. The onset of action as well as maximum clinical effect of phenobarbital is evident later than pentobarbital, though the half-life of phenobarbital is much longer. It is unclear if the longer half-life is more beneficial than detrimental, as the longer half-life results in longer adverse effects, such as hypotension or respiratory depression, as well as longer desired anticonvulsant effect. The choice to use phenobarbital versus pentobarbital is probably best made on the basis of which drug the clinician and ED staff is more familiar with. As with management of seizures from other causes, use of a barbiturate after administration of a benzodiazepine should be expected to cause a degree of respiratory depression, possibly necessitating endotracheal intubation.
A drug relatively new in the ED setting, propofol, has been recognized as a effective anticonvulsant112 and is an excellent choice in the management of toxin-induced seizure. In a retrospective chart review, propofol was found to be of equal efficacy to midazolam infusion113 and more effective than hiopental114 with regard to ability to terminate status epilepticus. Propofol for toxin-induced seizures can be administered in dosing similar to that for endotracheal intubation and sedation: a 1 mg/kg bolus followed by infusion. Propfol use should involve vigilance to detect respiratory depression or airway compromise, including anticipation of the possible need for endotracheal intubation.
The general management of the acutely poisoned child rarely involves the use of active elimination techniques. Careful consultation with a medical toxicologist and/or nephrologist should accompany employment of these methods. Considerations for the use of enhanced elimination include situations where prolonged exposure to high concentrations of a toxin is a known hazard. For example, in methanol ingestion, hemodialysis is often employed to enhance the clearance of its toxic metabolite, formic acid, while fomepizole is used to block its production. Patients who show hemodynamic compromise and end-organ failure despite supportive measures also should be considered for enhanced elimination techniques.
Forced diuresis was once recommended in the treatment of salicylate and barbiturate overdose. This technique has been abandoned due to the high risk of fluid overload. Saline diuresis for lithium intoxication continues to be recommended.115
The unionized form of a drug is filtered and reabsorbed. Urinary alkalinization with intravenous sodium bicarbonate may increase the proportion of ionized drug in the tubule, essentially trapping it within the tubules where it is excreted in theurine. Aspirin is the best example of a substance that undergoes significantly enhanced excretion with urinary alkalinization. In children, the recommended protocol for urinary alkalinization is listed in on Table 18.
Decreasing urinary pH can be used to enhance the excretion of weakly alkaline drugs. Urinary acidification using ammonium chloride has previously been used to enhance excretion of amphetamine, strychnine, quinine, quinidine, and phencyclidine. However, the dangers of acidosis, hyperammonemia, and rhabdomyolysis outweigh the benefits of this technique and urine acidification is no longer recommended.
Extracorporeal elimination, including dialysis, hemoperfusion, and hemofiltration, is usually reserved for specific toxins that are life threatening and amenable to removal by this method. Clinical efficacy of these measures may be difficult to differentiate from the concomitant effects of enterohepatic circulation, hepatic metabolism, or urinary excretion.
Hemodialysis is effective when the toxin in question has a small volume of distribution and low molecular weight and is highly water-soluble and poorly protein-bound. Hemodialysis is an important consideration when there are coinciding acid-base or electrolyte disturbances. MARS is an xtracorporeal device that combines conventional hemodialysis with a secondary system containing an albumin impregnated dialysis membrane (with an albumin containing dialysate) in addition to anion-exchange and charcoal columns.116 It has been used in the management of acute liver failure in adults and, because it is able to remove protein bound substances, could theoretically be used in the management of severe poisoning with highly protein bound drugs, such as salicylates, in the future, but it remains a research tool in the few centers in which it is available for now. This technique has yet to be studied in children.
Charcoal hemoperfusion involves placing a charcoal filter into the hemodialysis machine. Substances effectively eliminated by this method must have a high affinity for charcoal and a small volume of distribution. In contrast to hemodialysis, hemofiltration can remove compounds with a high molecular weight. See Table 19 for a list of agents amenable to extracorporeal elimination.
Exchange transfusion can be employed for small children who cannot receive hemodialysis because of technical limitations. It is a potentially life-saving procedure used to counteract the effects of serious jaundice, blood abnormalities, or drug toxicities. The procedure involves the incremental removal of the patient's blood and replacement with fresh donor plasma. Case reports of benefits from this method have been demonstrated with theophylline,117 phenobarbital, 118 chloramphenicol,119 lithium,120 and aniline.121 Exchange transfusion has also been used in isolated cases of toxicity from chloral hydrate,122 salicylate,123 quinine,124 and to treat severe methemoglobinemia.125
Technical advances in pediatric life support and critical care have undoubtedly contributed to the declining mortality of children with serious poisonings. Additional advances in immunotherapy treatments have stemmed from the use of digoxin-specific antibodies, snake and black widow antivenins, and therapies for tricyclic antidepressants and colchicine. Currently, the use of surfactant therapy and extracorporeal membrane oxygenation for preventing and treating the progression of lung injury due to hydrocarbons is under study.
Acetaminophen is another poison for which database analysis has resulted in management recommendations.126 A prospective study127 evaluating calls to a regional poison control center over a 25-month period concluded that pediatric patients younger than seven years of age with acute acetaminophen exposures of up to 200 mg/kg treated with home monitoring alone do not develop signs or symptoms of hepatic injury and therefore do not require intervention. In this study, actual serum levels of acetaminophen were never obtained. Serum levels were not assessed but were projected based on the quantity of acetaminophen missing from the bottle divided by the child's weight. Due to the use of such estimates, it is unknown how many patients actually ingested any quantity of acetaminophen, much less a toxic dose. In addition, since hepatic function tests were not universally obtained and parental recognition of signs of hepatotoxicity may have been inadequate, mild elevations of liver function tests and subclinical signs of hepatotoxicity might have been missed. Although the cases in this study that were followed fully had good outcomes, cases lost to follow-up may have involved severe morbidity or even death. This is only one study representative of several evaluations of acetaminophen ingestions in children that recommend lesser or no routine evaluation of children after acetaminophen ingestion. Currently, published studies on the subject lack scientific rigor on which to base such recommendations.
The value of recommendations of expected outcome, clinical course, or management strategies based on studies that do not involve clinical or laboratory corroboration of exposure is highly questionable. Clearly, prospective studies that include laboratory corroboration or confirmation of exposure are needed to make credible recommendations about poison exposure management, but they are lacking. Until there is clear evidence demonstrating that assessment of serum acetaminophen levels after exposure is not necessary and can safely be abandoned, we recommend that clinicians continue with routine measurement of acetaminophen levels in children acutely exposed to acetaminophen in quantities greater than 200 mg/kg.
Exposure to anticoagulant rodenticides can result in morbidity or mortality as a result of hemorrhage due to anticoagulant effects that may last months.128 Two recent descriptive studies129,130 of children exposed to long-acting anticoagulant rodenticides failed to find any adverse events in the study subjects and suggestthat these patients can be safely managed at home. However, both studies suffer from significant methodologic flaws which preclude making any definitive management recommendations at this time.
It is a common practice to observe asymptomatic patients in the ED for several hours after exposure. Six hours has been a typical amount of time allotted to observation prior to discharge. Prospective study evaluating treating physicians' determination as to appropriateness of discharge from ED at two hours and four hours prior to actual discharge at six hours has yielded useful results.131 In this study, where 42% of the study population were less than 18 years of age and 14% were less than two years of age, no patient who was believed to be safe for medical discharge at either the two or four hour mark had a complication within the six hour period. The implication of this is that the EM physicians were capable of appropriately determining which patients could be safely discharged sooner than six hours based on substance ingested and clinical findings. A caveat of discharging patients earlier than the six hour observation window is to note whether or not the poison involved is capable of causing symptoms after an initial asymptomatic period. This includes regular preparations that might not result in symptoms immediately as well as sustained release or delayed release preparations. Drugs capable of causing delayed toxicity are listed in Table 20.
In many asymptomatic children with benign unintentional ingestions, six hour observation periods are foregone, and the patient is discharged home after less than a six hour observation.
Children with symptomatic toxicity resulting in abnormal vital signs, electrolytes, or ECG, or resulting from poisons with the potential to cause severe morbidity or mortality are commonly managed in an intensive care unit. If there is any doubt as to whether the patient requires intensive care or monitoring,we recommend that the involved clinicians select on the side of caution and choose the more closely monitored or managed setting if it is available. Any patient with intent of self-harm should receive psychiatric evaluation prior to discharge as well as possible admission to a psychiatric unit if they are actively suicidal.
Disappointingly, poison prevention education is lacking in many pediatric poison exposures managed in emergency departments. An ED visit is an excellent opportunity for intervention and poison revention education, which is underutilized.132
Consultation with a poison center or medical toxicologist is a useful adjunct to facilitating complex care, particularly when the emergency physician is unfamiliar with the poison involved or inexperienced with the course of treatment for a particular ingestion. Consultation with a poison center should occur in all poisonings, even those for which the management is straightforward and the ED physician is experienced. The poison control center collects valuable epidemiologic data that serves a vital public health purpose, and the poison center will continue to follow the case and may make management recommendations. Although the treating ED physician may not need any management advice, it is best to not presume this about other healthcare providers who subsequently care for the child after hospital admission or after referral of the patient back to their regular healthcare provider. Consultation with other specialties should be considered when appropriate.
Though increasingly less common as a result of poison prevention initiatives, pediatric poison exposures continue to be an ongoing cause of pediatric morbidity and mortality. Though most pediatric poison exposures result in minimal or no symptoms, management of pediatric poisonings is a critical aspect of emergency medicine. Rational management based on symtomatology, useful lab assays, and investigations, as well as selective use of GI decontamination and antidotes is essential in ensuring optimal outcomes in such patients.
1. Initial evaluation of patients should include A,B,C,D,E- beyond airway, breathing, and circulation, dextrose and ECG should be evaluated immediately when indicated.
2. Consultation with a poison control center is indicated for all exposures.
3. Never underestimate the gravity of a "one-pill kill" exposure or exposure to poisons with potential for delayed onset of symptoms.
4. Usually the physical findings, particularly toxidromes, in addition to vital signs guide management.
5. Practice congruently with current guidelines, keeping in mind that:
6. Practice safely.
7. Obtain laboratory assays in a rational manner.
8. Antidotes are uncommonly indicated.
9. Several newer antidotes are available.
10. Consider evidence from pediatric exposure studies cautiously.
1. Use fomepizole rather than ethanol to treat toxic alcohol exposure.
Though it is not yet considered standard of care, near ubiquitous use of fomepizole is beneficial. It is easier to administer, and although more expensive than alcohol for the actual antidote, overall therapy is less intensive and less expensive due to decrease monitoring and critical care stay.
2. Use IV n-acetylcysteine rather than oral n-acetylcysteine to treat acetaminophen toxicity.
Although the IV formulation is more expensive, its use is only 24 hours rather than 72 hours, decreasing length of stay.
3. Use octreotide to treat sulfonylurea-induced hypoglycemia.
This antidote in inexpensive and its use prevents precipitous rises and dips of serum glucose, and avoids the need for repeat glucose infusions.
4. Use early dialysis.
Dialysis is expensive, but preventing sequelae from poisoning, such as with salicylate, or diminishing the overall duration of poisoning, such as with methanol, decreases morbidity and critical care stays.
5. Avoid wasted costs on unnecessary lab tests.
Though a battery of lab assays are often obtained for patients with poison exposure, few are ever needed. The commonly indicated assays are bedside glucose measurement, serum electrolytes, and acetaminophen level, as well as serum drug levels for specific poisons. Drug of abuse screening, PT/PTT, CBC, and other ancillary tests should not be routinely obtained.
6. Discharge after two to four hour observation.
Traditional observation of poison exposed patients has been six hours. This is based on dogma rather than evidence. Certain poisons warrant observation for a particular period of time, but, as a general rule, a two to four hour observation period appears to be appropriate for most patients. If you have any doubt, contact a poison center or toxicologist and specifically ask if the poison in question mandates prolonged observation or admission.
1. Underestimation of the potential morbidity or mortality of a "one-pill kill" ingestion.
The list of agents that can do this is well described.Any toxin capable of disrupting normal respiratory orcardiac function is potentially deadly. Even with lackof corroborating symptoms, the potential for deteriorationmandates observation, often for a full 24 hours.
2. Waiting until a patient becomes clinically ill or unstable before administering appropriate therapy.
It is highly preferable to prevent illness or injury rather than treat it after it has developed. Early treatment may be lifesaving.
3. Failure to obtain expert advice from a poison center or medical toxicologist.
Even if you are familiar with or capable of managing the poisoning, it is possible that others caring for the patient after you may not be. Poison center involvement will result in follow-up and maintained involvement of a specialist.
4. Failure to maintain vital signs within acceptable safe limits.
This is a cornerstone of supportive care, and prompts appropriate respiratory and cardiovascular therapy.
5. Failure to control status epilepticus.
Certain poisons result in difficult to control or refractory seizures. These must be managed aggressively to prevent neurologic sequelae that may be permanent. Avoid using phenytoin or fosphenytoin for toxininduced or withdrawal seizures.
6. Failure to diagnose cyanide poisoning.
Cyanide poisoning is a well described poisoning that may result in persons who are in a closed space fire. Altered mental status, syncope, coma, or lactic acidosis in a person who was in a closed-space fire warrant consideration of immediate empiric therapy. Delay waiting for serum lactate results may prevent administration in a timely manner and an otherwise preventable death.
7. Failure to diagnose carbon monoxide poisoning.
The most common cause of poisoning fatality might not be considered in patients who have not been in a fire. The nonspecific symptoms of headache and malaise occurring in the winter are frequently wrongly attributed to a viral cause.
8. Failure to diagnose Munchausen's Syndrome By Proxy or child abuse.
Circumstances of poisoning not consistent with motor or cognitive development, unusual poison exposures, or even a "gut feeling" about the history or caregivers actions should prompt inquiry into possible abuse or neglect.
9. Overly aggressive treatment for minor ingestions.
Use of aggressive management, such as ipecac, lavage, or even charcoal, administered by NG tube for a benign ingestion may have disastrous consequences. The greatest advances in management of poisoning are arguably the paradigm shift to supportive care and away from aggressive therapies that were wellintentioned but did not actually result in improved clinical outcome. The benefit of any therapy should clearly outweigh the risk.
10. Clinical use of drug of abuse screening results.
Though the results may appear simple as reported, drug of abuse screening is quite sophisticated and has many nuances. Additionally, since results are not quantitative, it is impossible to know if a positive result is consistent with current toxicity or a previous exposure with detectable but clinically irrelevant drug levels.
The 12-month-old male patient ingested hair relaxer which usually contains sodium, calcium, or lithium hydroxide. These products are often advertised as containing no lye but, in fact, they contain alkaline caustics.133 In the ED, the patient had no respiratory symptoms and his chest xray was normal. Initial fiberoptic laryngoscopy was normal, but persistent drooling for more than 12 hours led to an esophagoscopy. Two noncircumferental second degree burns measuring about 1-2 cm in size were found. The patient was treated with IV antibiotics and continued to improve clinically. He was discharged three days later.
Evidence-based medicine requires a critical appraisal of the literature based upon study methodology and number of subjects. Not all references are equally robust. The findings of a large, prospective, randomized, and blinded trial should carry more weight than a case report.
To help the reader judge the strength of each reference, pertinent information about the study, such as the type of study and the number of patients in the study, will be included in bold type following the reference, where available.
Robert J. Hoffman; Alexander Grinshpun; Deepti T. Paulose; In-Hei Hahn
May 1, 2007