Blaine E. Benson

Prospective Multicenter Evaluation of Tramadol Exposure

BACKGROUND Tramadol is a novel analgesic possessing both opiate and noradrenergic effects. Its low potential for abuse suggests increasing use, but there are limited data on the toxicity in overdose. METHODS Multicenter prospective case series. All exposures from October 1995 through August 1996 reported to seven Poison Centers were evaluated. RESULTS There were 126 cases of which 87 were tramadol alone. Of the tramadol alone cases, 51 were female (59%). Age ranged from 1 to 86 y with a mean and median of 26.8 y (SD 17.2) and 25 y, respectively. There were 15 cases of children less than 6 years old. Symptoms reported with overdose were: lethargy 26 (30%), nausea 12 (14%), tachycardia 11 (13%), agitation 9 (10%), seizures 7 (8%), 4 each (5%) of coma and hypertension, and respiratory depression 2 (2%). All seizures were brief. Naloxone reversed sedation and apnea in 4 of 8 patients. One patient experienced a seizure immediately after administration of naloxone. Other treatments were: diazepam (3 patients), and phenytoin, lorazepam and nifedipine (1 patient each). Tramadol 500 mg was the lowest dose associated with seizure, tachycardia, hypertension or agitation while 800 mg was the lowest dose associated with coma and respiratory depression. Urine drug screens performed on 19 patients were negative for opiates. All symptomatic cases exhibited effects within 4 h of ingestion. Mean hospital stay was 15.2 h (range 2-96 h, SD 15.8). Nineteen patients were admitted to an intensive care unit with a mean stay of 25 h (SD 20). DISCUSSION Much of the toxicity in tramadol overdose appears to be attributable to the monoamine uptake inhibition rather than its opioid effects. Agitation, tachycardia, confusion and hypertension suggest a possible mild serotonin syndrome. No arrhythmias beyond tachycardia were seen. CONCLUSION This study suggests significant neurologic toxicity from tramadol overdose. Serious cardiovascular toxicity was not seen.

An analysis of energy-drink toxicity in the National Poison Data System

Context. Small studies have associated energy drinks—beverages that typically contain high concentrations of caffeine and other stimulants—with serious adverse health events. Objective. To assess the incidence and outcomes of toxic exposures to caffeine-containing energy drinks, including caffeinated alcoholic energy drinks, and to evaluate the effect of regulatory actions and educational initiatives on the rates of energy drink exposures. Methods. We analyzed all unique cases of energy drink exposures reported to the US National Poison Data System (NPDS) between October 1, 2010 and September 30, 2011. We analyzed only exposures to caffeine-containing energy drinks consumed as a single product ingestion and categorized them as caffeine-containing non-alcoholic, alcoholic, or “unknown” for those with unknown formulations. Non-alcoholic energy drinks were further classified as those containing caffeine from a single source and those containing multiple stimulant additives, such as guarana or yerba mate. The data were analyzed for the demographics and outcomes of exposures (unknown data were not included in the denominator for percentages). The rates of change of energy drink-related calls to poison centers were analyzed before and after major regulatory events. Results. Of 2.3 million calls to the NPDS, 4854 (0.2%) were energy drink-related. The 3192 (65.8%) cases involving energy drinks with unknown additives were excluded. Of 1480 non-alcoholic energy drink cases, 50.7% were children < 6 years old; 76.7% were unintentional; and 60.8% were males. The incidence of moderate to major adverse effects of energy drink-related toxicity was 15.2% and 39.3% for non-alcoholic and alcoholic energy drinks, respectively. Major adverse effects consisted of three cases of seizure, two of non-ventricular dysrhythmia, one ventricular dysrhythmia, and one tachypnea. Of the 182 caffeinated alcoholic energy drink cases, 68.2% were < 20 years old; 76.7% were referred to a health care facility. Educational and legislative initiatives to enhance understanding of the health consequences of energy drink consumption were significantly associated with a decreased rate of energy drink-related cases (p = 0.036). Conclusions. About half the cases of energy drink-related toxicity involved unintentional exposures by children < 6 years old. Educational campaigns and legal restrictions on the sale of energy drinks were associated with decreasing calls to poison centers for energy drink toxicity and are encouraged.

Initial symptoms as predictors of esophageal injury in alkaline corrosive ingestions

The predictive value of initial clinical evaluation in the management of alkaline corrosive ingestion remains unclear. This multicenter study was designed to determine if specific clinical signs and symptoms following ingestion of alkaline corrosives could predict significant esophageal injury. Alkaline corrosives were defined by a pH greater than or equal to 12. Signs and symptoms previously suggested as predictive of significant esophageal injury were documented on a standardized data form. Esophagoscopy reports were reviewed blinded to initial symptoms. Three hundred thirty-six alkaline-corrosive ingestions were analyzed. The mean number of symptoms reported in patients who did not have esophagoscopy was 1.2, in patients who had esophagoscopy was 3.0, and in patients that had visualized second or third degree esophageal burns was 4.8. Of 88 patients who had esophagoscopy, 63 (72%) had both the esophagoscopy report and initial symptom assessment available. Esophagoscopy was positive, defined as second or third degree esophageal burns, in 18 of 63 cases (29%). All patients with significant burns were symptomatic. No single or group of initially reported signs and symptoms could identify all patients with potentially serious esophageal burns.

AAPCC database characterization of native U.S. venomous snake exposures, 2001–2005

Background. Differences in victim demographics, clinical effects, managements, and outcomes among native viperid (rattlesnake, copperhead, and cottonmouth) and elapid (coral snake) species have not been systematically characterized. Methods. The database of the American Association of Poison Control Centers from 2001 through 2005 was analyzed. Results. Between 2001 and 2005, there were 23,676 human exposures (average = 4,735/year) to native venomous snakes in the United States reported to U.S. poison centers in all states except Hawaii: 98% were to viperid snakes and 2% to elapids. Overall, 77% of victims were male, 70% were adults >20 years, and 12% were aged less than 10 years. Sixty-five cases involved pregnant women, with rattlesnake bites resulting in moderate or greater effects in over 70%. The overall hospital admission rate was 53%. Outcomes were generally more severe with rattlesnake and copperhead envenomations and in children <6 years of age. The fatality rate of reported cases was 0.06%. Conclusions. Native U.S. venomous snakebite results in considerable morbidity and mortality. Rattlesnake and copperhead envenomations, and those in children <6 years of age, produce the most severe outcomes, but coral snakebites result in similar hospital admission rates.

Methanol Poisoning Exposures in the United States: 1993–1998

Objectives: We investigated U.S. methanol-poisoning exposures since little recent information is available about the frequency, sources of methanol, or outcomes of individuals who consume methanol. Methods: We reviewed human methanol exposures reported to the American Association of Poison Control Centers Toxic Exposure Surveillance System 1993–1998. Results: The mean number of cases per year was 2254. Each year 167 cases had an outcome of moderate effect, major effect, or death. One death occurred in every 183 exposures to methanol. Symptomatic cases increased abruptly from 1.8 to 2.5% for infants and children, from 14.1 to 12.3% for adolescents and adults, while the intentional exposures increased from <1 to 21% for adolescents, and 11% for adults. Cases occurred in every decade of life but toddlers had the highest number of exposures. In 68 methanol fatalities, confusion developed followed by coma, hypotension, respiratory depression, and cerebral edema. Blood methanol levels were above 100 mg/dL (70%), metabolic acidoses (62%), and anion gaps >25 (100%). Methanol products were recorded, showing windshield wiper fluids to be 60.8% of exposures. Other automotive sources were 23.7%. Commercial nonautomotive products were 12.2% and pure methanol products were 2.3%. Unintentional exposures were reported in 90.3% of all cases, while 8.3% were due to intentional exposures, and 1.4% was for unknown or mixed reasons. Intentional exposures resulted from suspected suicides (51.2%) and from abuse and misuse (38.8%). Conclusions: Methanol poisonings continue to occur in the United States with toddlers at the highest risk for exposure, but adolescents and adults at the highest risk for life-threatening intoxications. Over half of product-identified cases were due to consumption of windshield wiper fluid. Efforts should be undertaken to minimize methanol exposures.

Determinants of U.S. poison center utilization

Context. High poison center utilization has been associated with decreased emergency department usage and hospitalization rates. However, utilization requires awareness of the poison center. Penetrance, defined as the number of human poison exposures reported to a poison center per 1,000 population, has been used as a marker of poison center awareness. Objectives. To identify factors that influence poison center penetrance to optimize the life- and cost-saving benefits of poison control centers. Methods. Human poison exposures that were reported to the National Poison Data System in 2001 were analyzed to identify and rank factors affecting poison center penetrance. Results. Overall penetrance correlated with pediatric penetrance (R2  =  0.75, p < 0.01). As pediatric penetrance increased, there was a significant decline in the percent of children reported to a poison center that were already in or en route to a healthcare facility at the time of the call to the poison center (R2  =  0.41, p < 0.01). Larger poison center service populations were associated with lower penetrance (R2  =  0.23, p < 0.01). Inverse predictors of penetrance included inability to speak English well, Black/African American race, and distance from the poison center (multiple regression). Positive predictors included the percentage of the population younger than 5 years, the percentage of the adult population with a bachelors degree, poison center certification, poison center educator FTEs (full time equivalents), Asian population percentage, and population density. Discussion. The inverse correlation between pediatric penetrance and healthcare facility utilization supports prior observations of excessive healthcare utilization when a poison center is not called. Since race, language and distance are barriers to poison center utilization, and since healthcare utilization increases when poison center penetrance declines, low penetrance suggests a lack of awareness of the poison center rather than a low incidence of poisonings. Conclusion. Strategies to raise penetrance should be informed by an understanding of the barriers to utilization – language, Black/African American race, distance from the poison center, poverty, and lower education levels.

The Impact of Bittering Agents on Pediatric Ingestions of Antifreeze

Background. Legislation requiring bittering of antifreeze enables assessment of the impact on frequency, volume, and severity of pediatric antifreeze ingestions. Methods. US poison control data for antifreeze ingestions in children younger than 5 years were analyzed comparing 232 ingestions occurring in states after enactment of bittering requirements with 6218 cases occurring in states (or at times) where bittering was not required. Results. The frequency of pediatric antifreeze ingestions was unchanged after implementation of bittering in Oregon and California. The medical outcome distribution, median volume ingested, and observed clinical effects were no different in bittered compared with nonbittered groups. Likewise, the rates of hospital admission, critical care treatment, and use of alkalinization, hemodialysis, or intubation showed no differences with bittering. Conclusion. Despite the appealing logic of limiting the ingested volume and thereby the severity of poisonings by adding aversive agents, and despite promising results in volunteer studies, bittering agents do not decrease the frequency or severity of pediatric antifreeze poisonings. The addition of bittering agents to household products cannot be justified based on actual poisoning data.

Comparison of citalopram and other selective serotonin reuptake inhibitor ingestions in children

Abstract Context. In adults, citalopram is more likely to cause seizures and ECG changes than other selective serotonin reuptake inhibitors (SSRIs). Data in children are lacking, yet the 2007 American Association of Poison Control Centers out-of-hospital citalopram consensus guideline mirrors the guideline for other SSRIs. Objective. To compare the clinical effects and hazard index of citalopram with other SSRIs in pediatric ingestions. Methods. An 11-year retrospective analysis of national poison center data was conducted. Acute, known-type SSRI ingestions in children younger than 6 years with known outcome were included. Clinical effects and hazard index (number of major or fatal outcomes/1000 SSRI ingestions) were compared. Citalopram dose-response was evaluated. Results. The 35 296 included cases by SSRI type were citalopram (3747), escitalopram (4815), fluoxetine (5946), fluvoxamine (273), paroxetine (7157), and sertraline (13 358). The overall hazard index was 0.340. The hazard index for citalopram (0.801) was 2.8-fold higher than for non-citalopram SSRIs (0.285). Comparing seizures (single or multiple discrete) and cardiac effects (conduction disturbances, other ECG changes or other dysrhythmia) of citalopram with the other SSRIs, pediatric citalopram ingestions were more likely to develop seizures (5 of 3747 [0.13%] vs. 10 of 31 549 [0.03%], OR = 4.2; 1.4–12.3) and cardiac toxicity (9 of 3747 [0.24%] vs. 25 of 31 549 [0.08%], OR = 3.0; 1.4–6.5). Clinical effects occurring more frequently with other SSRIs included tachycardia (p = 0.0236), oral irritation (p = 0.0412), vomiting (p = 0.0036), agitation/irritability (p = 0.0104), and hyperthermia (p = 0.0314). There was a dose response only for single or multiple discrete seizures, mydriasis and clinically significant responses (a predetermined subset of CNS and cardiopulmonary clinical effects). Meaningful triage thresholds for citalopram could not be determined due to the low frequency of significant clinical effects. Conclusion. Children develop minimal toxicity with SSRI ingestions. Seizures and ECG changes, while uncommon, occur more frequently with citalopram. Doses associated with significant outcomes suggest that the triage guideline for citalopram does not need to be modified.

Prospective observational multi-poison center study of ziprasidone exposures

Background. Ziprasidone is an atypical antipsychotic associated with QTc prolongation during therapeutic use. We characterized the clinical manifestations associated with ziprasidone overdoses, in particular the incidence and severity of QTc prolongation. Methods. Four regional poison centers prospectively collected ziprasidone overdose data from August 1, 2003 to October 1, 2005. Cases were included if they were followed to known medical outcome and comprised single-substance ziprasidone exposures or with co-ingestants not associated with prolongation of the QTc interval. Results. Fifty-six ziprasidone exposures met inclusion criteria. The most common clinical effects were drowsiness (N=38, 67.9%) and tachycardia (N=19, 33.9%). QTc prolongation (>0.500 second) occurred in only one patient. Seven patients had QTc intervals of 0.450 to 0.500 second. Medical outcomes were coded as no effect (13, 23.2%), minor effect (21, 35.5%), moderate effect (20, 35.7%), or major effect (2, 3.4%). Conclusion. Common clinical effects following ziprasidone overdose are drowsiness and tachycardia. Clinically significant QTc prolongation occurs infrequently.

Toxicity from a Clonidine Suspension

BackgroundClonidine is frequently prescribed to children. Clonidine overdose in children has resulted in major clinical effects and deaths.Case ReportA 3.5-year-old male with a history of a seizure disorder and night terrors presented following difficulty walking, excessive sleeping, agitation when awake, and possible seizure activity. Chronic medications were valproic acid (VPA) and clonidine. On presentation, he alternated between poor responsiveness and agitation, with initial vitals: blood pressure, BP 144/76 mmHg; heart rate, 65 bpm; respiratory rate, 18 bpm; temperature 99.5°F; and pulse oximetry 96% on room air. VPA level was 35 μg/mL. A toxicology consult the next day noted a dry mouth, 2-mm pupils, intermittent gasping, and central nervous system (CNS) depression, with a diagnostic impression of clonidine overdose. The caregiver had been giving 1 mL (0.1 mg) qd of a pharmacy-compounded clonidine suspension by a provided syringe. The pharmacy procedure record agreed with the physicians order. The amount dispensed was a 30-day supply but the bottle was empty on day 19, leading us to suspect a possible accelerated dosing error. The concentration in the bottle thus could not be confirmed. The child slowly returned to his baseline state over 48 hours. A serum clonidine level drawn approximately 18 hours after his last dose later returned at 300 ng/mL (reference range = 0.5–4.5 ng/mL).Case DiscussionCompounding and liquid dosing errors are common in children and may result in massive overdoses. There was an accelerated dosing error, but whether a compounding or suspension error or even an acute overdose occurred as well is unknown.ConclusionParticular care should be taken with medications that have low therapeutic indices, that are extemporaneously compounded, and are prepared as liquids, where medication errors are more likely.