Clinical Toxicology (2014), 52, 519–524 Copyright © 2014 Informa Healthcare USA, Inc. ISSN: 1556-3650 print / 1556-9519 online DOI: 10.3109/15563650.2014.909601
CRITICAL CARE
Poisonings requiring admission to the pediatric intensive care unit: A 5-year review K. M. EVEN,1,2 C. C. ARMSBY,1 and S. T. BATEMAN1 1University
Clinical Toxicology Downloaded from informahealthcare.com by Dokuz Eylul Univ. on 11/06/14 For personal use only.
2Children’s
of Massachusetts Medical School, Worcester, MA, USA Hospital of Pittsburgh of UPMC, Pittsburgh, PA, USA
Background. Poisonings represent a significant number of preventable admissions to the pediatric intensive care unit (PICU), but data about poisonings requiring PICU-level care are limited. Objectives. To identify the demographics of patients admitted with poisonings and characterize their clinical courses related to their poisoning. Methods. All poisonings over a 5-year period (2008–2012) at an academic medical center in New England were retrospectively reviewed using electronic medical records in an observational case series. Poisonings were identified using key search terms within an admissions database. Results. There were 273 admissions for poisonings, which represent 8% of total PICU admissions over this time period. The poisonings were unintentional in 148 (54%) cases and intentional in 125 (46%). The vast majority of poisonings occurred in patients either 3 years or below (N ⫽ 121, 44%) or 13 years or above (N ⫽ 124, 45%). Most (96%) admissions were for less than 48 h and 41% were for less than 24 h. Mean PICU length of stay was 1.2 ⫾ 0.7 days. A total of 468 substances were ingested in 54 different drug classes, with analgesics and antidepressants being the most common. Eighty-five (31%) poisonings were polypharmaceutical. The most commonly used therapies were naloxone, activated charcoal, and benzodiazepines. Twenty-seven patients (10%) received mechanical ventilation. There was one fatality, an adolescent with a polypharmacy overdose in a suicide attempt. Conclusion. Pediatric poisonings are a significant percentage of admissions to the PICU. The majority of poisonings are non-fatal, require supportive care, close monitoring, and some specific treatment. Drug classes causing poisonings have changed to a higher percentage of opioids in younger patients and atypical antidepressants in adolescents. Keywords Poisoning; Drug overdose; Pediatric intensive care unit; Suicide; Polypharmacy
A study conducted by Fazen et al.4 in 1986 reviewed 90 admissions for poisonings over a 2-year period; of these, 58 (64%) were admitted to the ICU. Totally 42% (38 patients) poisonings were intentional and 45% of these poisonings were polypharmaceutical. Substances most commonly ingested were barbiturates, TCAs, theophylline, aspirin, and benzodiazepines. Totally 10% of patients received specific antidotal treatment and 8.9% (8 patients) required mechanical ventilation, and there was one fatality due to diphenoxylate overdose. The goals of this retrospective study were to (1) describe the demographics of patients admitted to the PICU for suspected and known poisonings; (2) determine the type and characteristics of poisonings that require PICU care; (3) describe the types of treatments provided to patients admitted for poisonings; (4) describe the outcomes of patients admitted for poisonings; and (5) identify trends related to poisonings that may help raise public awareness and prevent future poisonings.
Introduction Poisonings (unintentional and intentional) represent a significant number of preventable admissions to the pediatric intensive care unit (PICU), but data about poisonings requiring PICU-level care are limited. Poisonings continue to be a common cause of injury in children,1 and studies have suggested that poisonings account for 3–4.5% of PICU admissions,2 although these studies are more than 15 years old. In a study conducted in 1989, Lacroix et al.3 reviewed 105 PICU admissions over a 3-year period and found that poisoning accounted for 3.1% of all PICU admissions. Substances most commonly ingested were: tricyclic antidepressants (TCAs) (22%), benzodiazepines (15%), theophylline (10%), ethanol (10%), hallucinogens (8%), salicylates (8%), narcotics (8%), antihistamines (7%), and carbamazepine (5%). Treatment was usually nonspecific. Two patients required hemodialysis and one patient required an intestinal resection, and there were no fatalities. Two percent of patients had PICU stays more than 2 days. Received 27 January 2014; accepted 24 March 2014.
Methods
Address correspondence to Scot T. Bateman, MD, Department of Pediatrics, H5-524, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655, USA. Tel: ⫹ 774-4422164. Fax: ⫹ 774-443-2062. E-mail: [email protected]
In this observational case series, all poisonings over a 5-year period (January 1, 2008 to December 31, 2012) admitted to an 11-bed medical/surgical PICU at an academic medical 519
Clinical Toxicology Downloaded from informahealthcare.com by Dokuz Eylul Univ. on 11/06/14 For personal use only.
520
K.M. Even et al.
center in New England were reviewed using electronic medical records. This PICU has a catchment area of just under one million. Decision to admit to the PICU was made by the pediatric emergency department attending and PICU attending in consultation with our toxicology service. Criteria for PICU admission was respiratory failure or significant risk of respiratory failure, depressed mental status, seizures, and cardiovascular dysfunction or significant risk of arrhythmias. Patients could be admitted to the PICU as “intensive care status” if they need critical care support, hourly monitoring, or “intermediate care status”, if they could be monitored every 2 h. Cases were identified using an admissions database with search terms “overdose,” “poisoning,” “ingestion,” “intoxication”, “suicide attempt,” and “altered mental status.” Data extracted from the medical record included: type of poisoning; age, sex, race of patient, length of stay (LOS); disposition; pediatric risk of mortality III score (calculated at the time of admission and already in the record)5; single or multiple substance(s) ingested; source of substance; substance ingested (per report); amount of substance ingested (if known); drug level (if determined); toxicology screen results (if performed); duration of mechanical ventilation if provided; and antidote/treatment received. Poisonings were categorized as unintentional or intentional based on information available in the medical record by the authors (KE, SB, and CA), who were not blinded to the study objective. When polypharmacy or combination medication poisonings were present, each substance or active ingredient was counted separately; for example, if a patient ingested acetaminophen and ibuprofen, this was counted as one acetaminophen poisoning and one ibuprofen poisoning. When analyzing data, the comprehensive urine toxicology screen results were used unless a known drug was ingested that was not screened for by the toxicology screen. In cases where no toxicology screen was done or it was negative, the toxidrome and clinical story was used for identifying the substance (i.e., candle
oil ingestion and warfarin). The study was approved by Institutional Review Board of the hospital. Comparisons between unintentional and intentional overdoses were made using T-test, Chi-square test, and Mann–Whitney U test where appropriate. P values less than 0.05 were considered statistically significant. All data were analyzed with SAS 9.2 statistical software (CAS Institute, Cary, NC).
Results There were a total of 3282 PICU admissions over this 5-year time period; 273 (8%) of these were for poisonings. The majority of admissions were from pediatric emergency room of the institution, and 20% of patients were transfers from outlying hospitals. Totally 148 (54%) poisonings were unintentional and 125 (46%) were intentional. Among them 91 patients (33%) intentionally ingested substances in attempted suicide. The age distribution of the patients was bimodal with 121 (44%), 3 years or below and 124 (45%), 13 years or above (Fig. 1). There were no intentional poisonings in patients below 10 years. One hundred fifty-one (55%) patients were female. Demographic data for the admissions are shown in Table 1. The distribution of race in this population is consistent with the county’s demographics with the exception of Hispanics/Latinos, who make up 9% of the county’s population but accounted for 17% of poisoning admissions.6 Severity of illness scores were higher among patients admitted for intentional overdoses (Table 2) and larger proportion of these patients received more intensive level of care compared to patients admitted following unintentional ingestions (56 [45%] vs. 33 [22%]; P ⫽ 0.0001). Most (96%) admissions were for less than 48 h and 41% were for less than 24 h. Mean PICU LOS was 1.2 ⫾ 0.7 days. For patients receiving mechanical ventilation, the mean PICU LOS was 1.7 ⫾ 0.9 days.
Fig. 1. Age distribution of PICU admissions for poisoning. Clinical Toxicology vol. 52 no. 5 2014
Poisonings requiring admission to the pediatric intensive care unit 521 Table 1. Demographic data. Total N ⫽ 273 Age (years, mean ⫹ SD) Male sex (N, %) Race (N, %) White Hispanic/Latino Black/African-American Asian Multi-racial American Indian/Native Alaskan Native Hawaiian Not specified
Unintentional Intentional N ⫽ 148 N ⫽ 125
8.6 ⫾ 6.8 122 (45%)
2.7 ⫾ 2.5 79 (53%)
15.6 ⫾ 1.6 43 (34%)
198 (73%) 46 (17%) 18 (7%) 5 (2%) 2 (0.7%) 1 (0.4%)
102 (69%) 28 (19%) 12 (8%) 2 (1%) 1 (0.7%) 1 (0.7%)
96 (77%) 18 (14%) 6 (5%) 3 (2%) 1 (0.8%) 0
1 (0.4%) 2 (0.7%)
1 (0.7%) 1 (0.7%)
0 1 (0.8%)
Clinical Toxicology Downloaded from informahealthcare.com by Dokuz Eylul Univ. on 11/06/14 For personal use only.
N, number of patients; SD, standard deviation
Most patients received supportive care (close monitoring including telemetry, frequent neurological and vital sign checks, and symptomatic management) without specific intervention for their ingestion. Specific therapies that were given to patients are summarized in Table 2. Patients
acquired medication and substances from a variety of places and people (Table 3). Younger patient were more likely to have obtained the medication from a parent or older relative while adolescent patients were more likely to take an overdose of their own medication, use over the counter medications or obtain drug(s) from a friend. Most patients (69%) ingested a single substance (Table 4). A total of 468 substances in 54 different drug classes7 were ingested by 273 patients. Comprehensive toxicology screens were sent for 234 (86%) patients. Ingested substances were determined by history and toxicology screen. The most commonly ingested substances were antidepressants (accounting for 13% of poisonings), analgesics (12%), opioid dependency agents (buprenorphine/naloxone and methadone) (9%), cardiac medications (8%), and alpha-2 adrenergic agonists (clonidine and guanfacine) (7%) (Supplementary Appendix 1 available online at http://informahealthcare.com/doi/abs/ 10.3109/15563650.2014.909601). Twenty-seven patients (10%) required mechanical ventilation for hypoventilation (Supplementary Appendix 2 available online at http://informahealthcare.com/doi/abs/
Table 2. Severity of illness, length of stay, and therapies provided. Total N ⫽ 273 Level of care (N, %): Intensive care Intermediate care PRISM III score (median, range) Length of stay, days (mean ⫾ SD) Disposition (N, %) Home Inpatient ward Inpatient psychiatric facility Deceased Therapies provided (N, %): Intubation Duration of MV (mean ⫾ SD) Oro/naso-gastric tube placement Activated charcoal Gastric lavage Bowel irrigation/Go-LYTELY Dialysis Reversal agents/Antidotes and other medications provided: Naloxone Benzodiazepines N-acetylcysteine Sodium bicarbonate Vasopressor Magnesium Physostigmine Antiepileptic Antipsychotic Calcium gluconate Glucagon Flumazenil Octreotide Vitamin K Diphenhydramine Atropine Pyridoxine
Unintentional N ⫽ 148
Intentional N ⫽ 125
89 (33%) 184 (67%) 0 (0–18) 1.2 ⫾ 0.7
33 (22%) 115 (78%) 0 (0–10) 1.1 ⫾ 0.6
56 (45%) 69 (55%) 0 (0–18) 1.3 ⫾ 0.8
153 (56%) 45 (16%) 74 (27%) 1 (0.4%)
127 (85%) 21 (14%) 0 0
26 (21%) 24 (19%) 74 (59%) 1 (0.8%)
27 (10%) 0.82 ⫾ 0.83 32 (12%) 41 (15%) 1 (0.4%) 3 (1%) 2 (0.7%)
7 (5%) 0.69 ⫾ 0.27 8 (5%) 22 (15%) 0 0 1 (0.7%)
20 (16%) 0.87 ⫾ 0.95 24 (19%) 19 (15%) 1 (0.8%) 3 (2%) 1 (0.8%)
44 (16%) 33 (12%) 15 (5%) 16 (6%) 6 (2%) 5 (2%) 4 (1%) 3 (1%) 3 (1%) 3 (1%) 3 (1%) 2 (0.7%) 2 (0.7%) 2 (0.7%) 2 (0.7%) 1 (0.4%) 1 (0.4%)
25 (17%) 8 (5%) 1 (0.7%) 2 (1%) 1 (0.7%) 0 0 2 (1%) 0 2 (1%) 1 (0.7%) 1 (0.7%) 2 (1%) 1 (0.7%) 1 (0.7%) 1 (0.7%) 0
19 (15%) 25 (20%) 14 (11%) 14 (11%) 5 (4%) 5 (4%) 4 (3%) 1 (0.8%) 3 (2%) 1 (0.8%) 2 (2%) 1 (0.8%) 0 1 (0.8%) 1 (0.8%) 0 1 (0.8%)
N, number of patients; SD, standard deviation; PRISM III, Pediatric Risk of Mortality score6; MV, Mechanical ventilation. Copyright © Informa Healthcare USA, Inc. 2014
522
K.M. Even et al.
Table 3. Source of substance(s) ingested.
Patient’s own medication Parent Grandparent/older relative Older sibling Friend/boyfriend/girlfriend Other (person known) OTC drug or household item available in home for use More than one source Unknown
Total N ⫽ 273
Unintentional N ⫽ 148
Intentional N ⫽ 125
61 (22%) 58 (21%) 45 (16%) 15 (5%) 16 (6%) 21 (8%) 25 (9%)
8 (5%) 48 (32%) 44 (30%) 14 (9%) 0 19 (13%) 6 (4%)
53 (42%) 10 (8%) 1 (0.8%) 1 (0.8%) 16 (13%) 2 (2%) 19 (15%)
7 (3%) 25 (9%)
0 9 (6%)
7 (6%) 16 (13%)
Clinical Toxicology Downloaded from informahealthcare.com by Dokuz Eylul Univ. on 11/06/14 For personal use only.
Data presented as number of patients (percent). OTC, Over the counter.
10.3109/15563650.2014.909601). The mean duration of mechanical ventilation was 0.82 ⫾ 0.83 days. Totally 93 patients (34%) were monitored with serial electrocardiograms (ECGs), 33 (12%) had some degree of QTc prolongation more than 440 ms, and 9 (3%) had widening of the QRS complex more than 110 ms. Unintentional poisonings The mean age among patients admitted for unintentional poisonings was 2.7 ⫾ 2.5 years. There was a greater proportion of male patients admitted for unintentional poisonings compared to intentional overdoses (N ⫽ 79 [53%] vs. N ⫽ 43 [34%]; P ⫽ 0.002). The racial composition was similar in the two groups (Table 1). Patients admitted for unintentional poisonings had a statistically shorter LOS compared to those admitted for intentional overdoses (1.1 ⫾ 0.6 days vs. 1.3 ⫾ 0.8 days, P ⫽ 0.02) though not a clinically significant difference. Nearly all patients were admitted for 2 days or less (N ⫽ 146, 99%) and many (N ⫽ 66, 45%) were admitted for less than 24 h. Most patients (N ⫽ 127, 85%) were discharged directly to home from the PICU. Patients frequently took medication belonging to a parent (N ⫽ 48, 32%) or older relative (N ⫽ 44, 30%) (Table 3), and most patients ingested a single substance (N ⫽ 124, 84%; Table 4). Patients ingested a total of 201 substances in 43 different drug classes (Supplementary Appendix 1 available online at http://informahealthcare.com/doi/abs/10.3109/1556365 0.2014.909601). The most commonly ingested substances were buprenorphine/naloxone (accounting for 17% of poisonings), cardiac medications (16%), and alpha-2 adrenergic agonists (10%). Overdoses of antidepressants and analgesics Table 4. Characteristics of ingestion. Total N ⫽ 273 Single substance ingested Multiple substances ingested Amount ingested known Initial drug levels measured
Unintentional Intentional N ⫽ 148 N ⫽ 125
188 (69%) 85 (31%)
124 (84%) 24 (16%)
64 (51%) 61 (49%)
139 (50%) 51 (19%)
70 (47%) 11 (7%)
69 (55%) 40 (32%)
were significantly less likely in unintentional poisonings compared to those of intentional overdoses (16 [8%] vs. 43 [16%]; P ⫽ 0.009 and 12 [6%] vs. 44 [16%]; P ⫽ 0.0005, respectively). Seven patients (5%) were intubated for respiratory failure, six of whom required mechanical ventilation for less than 24 h (Supplementary Appendix 2 available online at http:// informahealthcare.com/doi/abs/10.3109/15563650.2014. 909601), and there were no fatalities. This group includes one patient intentionally injected with heroin, one patient exposed to cocaine in his home, three patients unintentionally given their siblings’ medication, and two patients who were unintentionally given overdoses of their own prescribed medications. Intentional poisonings Of the 125 patients admitted for intentional overdoses, 91 (73%) had suicidal intent or intent to harm, 31 (25%) were not suicidal (recreational drug use), and 3 (2%) had unclear intent. Most patients (N ⫽ 115, 92%) were admitted for 2 days or less and 45 (36%) were admitted for less than 24 h. Most patients (N ⫽ 74, 59%) were discharged to a psychiatric facility. Patients frequently ingested their own medication (N ⫽ 53, 42%), over-the-counter medication that was readily available in the home (N ⫽ 19, 15%) or substances they obtained from a friend (N ⫽ 16, 13%; Table 3). About half of patients (N ⫽ 64, 51%) ingested a single substance (Table 4). The amount of drug ingested was known for 69 patients (55%), and initial drug levels (specifically for acetaminophen, aspirin, ethanol, lithium, anticonvulsants, and antipsychotics) were measured for 40 patients (32%). Patients intentionally ingested a total of 267 substances in 43 different drug classes (Supplementary Appendix 1 available online at http://informahealthcare.com/doi/ abs/10.3109/15563650.2014.909601). The most commonly ingested substances were analgesics (accounting for 16% of overdoses), antidepressants (16%), and alcohol and illicit drugs (11%). Patients admitted following intentional overdose were more than three times more likely to require intubation for respiratory depression/failure compared to patients admitted for unintentional poisoning (N ⫽ 20 [16%] vs. N ⫽ 7 [5%]; P ⫽ 0.002; Supplementary Appendix 2 available online at http://informahealthcare.com/doi/abs/10.3109/15563650. 2014.909601). Half of the intubated patients had a polysubstance ingestion. Several patients had multiple PICU admissions for overdoses during this period. One adolescent female was admitted thrice for heroin overdoses, and three young women were admitted twice each for polypharmacy overdoses. One adolescent patient did not survive after an intentional poisoning of bupropion, clonidine, and methylphenidate with suicidal intention. He developed cardiac arrest following a period of prolonged uncontrollable seizure activity and the family subsequently withdrew life-sustaining therapies. Two patients received hemodialysis for lithium toxicity. A 15-year-old female developed significant liver dysfunction and was briefly considered for a liver transplant Clinical Toxicology vol. 52 no. 5 2014
Poisonings requiring admission to the pediatric intensive care unit 523 following an acetaminophen overdose, but subsequently recovered.
Clinical Toxicology Downloaded from informahealthcare.com by Dokuz Eylul Univ. on 11/06/14 For personal use only.
Discussion This study is a PICU-specific report on the demographics of significant pediatric poisonings. It was undertaken to describe the current burden of poisonings in the PICU and to determine the patient population and the drugs used as well as their clinical course. The burden of hospitalization and treatment for poisonings remains relatively high and some important differences were found compared to historical reports. Poisonings represented 8.0% of all PICU admissions at our institution over this 5-year period. This is about twice the rate documented by other studies.2 A bimodal age distribution seen in this study is consistent with the PICU data presented by Lacroix et al.,3 although our patient cohort had a higher percentage of patients 13 years or above, suicidal patients, and polypharmaceutical poisonings. The drugs ingested by both young children and adolescents have changed with time. Compared with these earlier studies, a rise in opioid and clonidine unintentional ingestions was particularly noted as well as a change in adolescent drug poisonings toward antipsychotic medication combinations and away from tricyclic medications. These trends have also been shown in broader pediatric poisoning reports of large databases of poisonings, both inpatient and outpatient, as described by Finkelstein et al.8 Our study demonstrates that younger children admitted with unintentional poisonings tended to be less sick, got less intensive therapy, and were more likely to be discharged directly home compared with adolescent patients admitted with intentional overdoses. Only seven children were admitted after ingesting household items, which is far less than reported in the US and other countries,9–12 suggesting many of these poisonings do not require intensive care. Buprenorphine/naloxone, an opioid dependency agent, was the most common unintentionally ingested substance in young children. Although none of these 35 children required mechanical ventilation, 16 required naloxone, and 5 had some degree of respiratory depression. This medication is used in the northeastern United States but may not be frequently prescribed in other parts of the country; other reports have also noted an alarming increase in New England of the incidence of children ingesting this medication.13,14 This may be related to a combination of factors, including its increased use and availability to patients, its sublingual absorption so the child does not need to swallow the pill to get the toxic effects, and the orange color and candy-like appearance of the tablet. The rate of unintentional pediatric poisoning is about eight times higher for buprenorphine/ naloxone tablets versus film.14,15 Based on this data, Reckitt Benckiser Pharmaceuticals agreed to voluntarily discontinue the production of the tablet form as on March 18, 2013; it will continue to manufacture the sublingual film, which is individually wrapped.16 All 35 children in our study ingested the tablet form. Copyright © Informa Healthcare USA, Inc. 2014
The public health and safety implications of these findings include the importance of discussing medication safety, including the use of lock boxes and childproof containers, with parents and caregivers. Prescribers of buprenorphine/ naloxone should be wary of their patients living with children under 4 and consider prescribing the safer sublingual version of the drug.16 The availability of drugs in the home can also be decreased by removing old medications. The National Prescription Drug Take-Back Day, organized by the Federal Drug Enforcement Agency, aimed at removing prescription drugs from circulation, which is an example of efforts to help avoid unintentional poisonings.17 Almost all our teenage ingestions were intentional (defined as purposefully taking substances with intent to harm, intent to get high, or a combination of both). Older children admitted with intentional poisonings were sicker and tended to require a higher level of care. Most of these patients were suicidal. Suicide is the third leading cause of death in pediatric patients in the United States, with 1922 children and adolescents committing suicide in 2009.18 For every completed suicide, there are many more attempts made, with a lifetime prevalence of attempting suicide of approximately 4%.19 Because so many intentional poisonings were polypharmacy overdoses, patients on multiple antipsychotic medication should be monitored closely. Mental health services, including availability of counseling services at schools, may help reduce suicidality and, in turn, reducing intentional poisonings in adolescents. In our cohort, most poisonings received only supportive care, and complications were short-lived and self-limiting. Naloxone, activated charcoal and benzodiazepines were the treatments most commonly provided. Patients were monitored with ICU monitors, telemetry, and often serial ECGs. In order to care for poisoning patients, a PICU needs to be prepared to provide emergent intubation, cardiac monitoring, arrhythmia management, seizure control, and acute psychiatric services. The most serious complications were seen in adolescents that were attempting suicide, usually with multiple medications. There was one fatality, and our 0.4% fatality rate is consistent with the low overall mortality rate of pediatric poisonings.9 This retrospective chart review has potential limitations that impact the generalizability of this study. This study examined a specific subset of pediatric patients that required PICU-level care at one institution, so it did not capture the overall picture of all pediatric poisonings. Patients were admitted to this PICU for advanced telemetry monitoring that may be done on a cardiac inpatient ward at other institutions. The high rate of buprenorphine/naloxone poisonings may be reflective of prescribing practices among addiction specialists in our area and may not be seen in other parts of the country.
Conclusion Unintentional and intentional poisonings in childhood continue to be a public health concern with many children requiring PICU-level care. This study identified an increased
524
K.M. Even et al.
rate of PICU admissions for poisoning compared 25 years ago with more suicidal patients ingesting multiple medications and more patients requiring mechanical ventilation. There were changes in the types of substances ingested compared with earlier studies, particularly a higher rate of buprenorphine/naloxone poisoning. Although the majority of poisonings are non-fatal and the patients received mostly supportive care, the burden of hospitalization on the patients, their families, and health care resources cannot be ignored. Additional studies are needed to identify public health interventions that may help to decrease the poisoning rate.
Take Home Points: Clinical Toxicology Downloaded from informahealthcare.com by Dokuz Eylul Univ. on 11/06/14 For personal use only.
• • • •
Poisonings requiring Pediatric ICU care are common and the types of drugs have changed over time. Unintentional poisonings by younger children usually involve a parent/caregiver’s drugs. Intentional poisonings by adolescents involve multiple drugs, including newer antipsychotics. A significant rise in buprenorphine/naloxone ingestions in young children was noted.
Funding Source No external funding was secured for this study.
Financial Disclosure The authors have no financial relationships relevant to this article to disclose.
Declaration of interest The authors report no declarations of interest. The authors alone are responsible for the content and writing of the paper.
References 1. Bond GR, Woodward RW, Ho M. The growing impact of pediatric pharmaceutical poisoning. J. Pediatr 2012; 160:265–270.e1. 2. Vernon DD, Gleich MC. Poisoning and drug overdose. Crit Care Clin 1997; 13:647–667. 3. Lacroix J, Gaudreault P, Gauthier M. Admission to a pediatric intensive care unit for poisoning: a review of 105 cases. Crit Care Med 1989; 17:748–750.
4. Fazen LE III, Lovejoy FH Jr, Crone RK. Acute poisoning in a children’s hospital: a 2-year experience. Pediatrics 1986; 77:144–151. 5. Pollack MM, Patel KM, Ruttimann UE. PRISM III: an updated Pediatric Risk of Mortality score. Crit Care Med 1996; 24:743–752. 6. U.S. Department of Commerce, United States Census Bureau, 2010 Census Interactive Population Search. Available at: http://www.census. gov/2010census/popmap/ipmtext.php. Accessed August 14, 2013. 7. Micromedex® Healthcare Series [Internet database]. Greenwood Village, Colo: Thomson Healthcare. Updated periodically. 2013. 8. Finkelstein Y, Hutson JR, Wax PM, Brent J, on behalf of the Toxicology Investigators Consortium (ToxIC) Case Registry. Toxico-surveillance of infant and toddler poisonings in the United States. J Med Toxicol 2012; 8:263–266. 9. Bronstein AC, Spyker DA, Cantilena LR Jr, Rumack BH, Dart RC. 2011 Annual report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 29th Annual Report. Clin Toxicol (Phila) 2012; 50:911–1164. 10. Jayashree M, Singhi S. Changing trends and predictors of outcome in patients with acute poisoning admitted to the intensive care. J Trop Pediatr 2011; 57:340–346. 11. Kohli U, Kuttiat VS, Lodha R, Kabra SK. Profile of childhood poisoning at a tertiary care centre in North India. Indian J Pediatr 2008; 75:791–794. 12. Andiran N, Sarikayalar F. Pattern of acute poisonings in childhood in Ankara: what has changed in twenty years? Turk J Pediatr 2004; 46:147–152. 13. Pedapati EV, Bateman ST. Toddlers requiring pediatric intensive care unit admission following at-home exposure to buprenorphine/naloxone. Pediatr Crit Care Med 2011; 12:e102–e107. 14. Martin TC, Rocque MA. Accidental and non-accidental ingestion of methadone and buprenorphine in childhood: a single center experience, 1999–2009. Curr Drug Saf 2011; 6:12–16. 15. Lavonas EJ, Banner W, Bradt P, Bucher-Bartelson B, Brown KR, Rajan P, et al. Root causes, clinical effects, and outcomes of unintentional exposures to buprenorphine by young children. J Pediatr 2013; 163:1377–83.e1–3. doi:10.1016/j.jpeds.2013.06.058. Epub 2013 Aug 29. 16. RB Press Release, Reckitt Benckiser Pharmaceuticals Inc. Submits Citizen Petition to US FDA Requesting Action to Mitigate Risk of Pediatric Exposure with Opioid Dependence Treatment; Company Voluntarily Discontinues the Supply of Suboxone® Tablets (buprenorphine and naloxone sublingual tablets [CIII]) in the United States. Available at: http://www.rb.com/site/rkbr/templates/mediainvestorsgeneral2.aspx?pageid ⫽ 1332 & cc ⫽ gb. Accessed April 9, 2013. 17. National Take-Back initiative. Available at: http://www.deadiversion. usdoj.gov/drug_disposal/takeback. Accessed April 28, 2013. 18. Kochanek KD, Kirmeyer SE, Martin JA, Strobino DM, Guyer B. Annual summary of vital statistics: 2009. Pediatrics 2012; 129: 338–348. 19. Nock MK, Green JG, Hwang I, McLaughlin KA, Sampson NA, Zaslavsky AM, Kessler RC. Prevalence, correlates, and treatment of lifetime suicidal behavior among adolescents: results from the National Comorbidity Survey Replication Adolescent Supplement. JAMA Psychiatry 2013; 70:300–310.
Supplementary material available online Supplementary Appendix Tables 1 and 2.
Clinical Toxicology vol. 52 no. 5 2014
CRITICAL CARE
Poisonings requiring admission to the pediatric intensive care unit: A 5-year review K. M. EVEN,1,2 C. C. ARMSBY,1 and S. T. BATEMAN1 1University
Clinical Toxicology Downloaded from informahealthcare.com by Dokuz Eylul Univ. on 11/06/14 For personal use only.
2Children’s
of Massachusetts Medical School, Worcester, MA, USA Hospital of Pittsburgh of UPMC, Pittsburgh, PA, USA
Background. Poisonings represent a significant number of preventable admissions to the pediatric intensive care unit (PICU), but data about poisonings requiring PICU-level care are limited. Objectives. To identify the demographics of patients admitted with poisonings and characterize their clinical courses related to their poisoning. Methods. All poisonings over a 5-year period (2008–2012) at an academic medical center in New England were retrospectively reviewed using electronic medical records in an observational case series. Poisonings were identified using key search terms within an admissions database. Results. There were 273 admissions for poisonings, which represent 8% of total PICU admissions over this time period. The poisonings were unintentional in 148 (54%) cases and intentional in 125 (46%). The vast majority of poisonings occurred in patients either 3 years or below (N ⫽ 121, 44%) or 13 years or above (N ⫽ 124, 45%). Most (96%) admissions were for less than 48 h and 41% were for less than 24 h. Mean PICU length of stay was 1.2 ⫾ 0.7 days. A total of 468 substances were ingested in 54 different drug classes, with analgesics and antidepressants being the most common. Eighty-five (31%) poisonings were polypharmaceutical. The most commonly used therapies were naloxone, activated charcoal, and benzodiazepines. Twenty-seven patients (10%) received mechanical ventilation. There was one fatality, an adolescent with a polypharmacy overdose in a suicide attempt. Conclusion. Pediatric poisonings are a significant percentage of admissions to the PICU. The majority of poisonings are non-fatal, require supportive care, close monitoring, and some specific treatment. Drug classes causing poisonings have changed to a higher percentage of opioids in younger patients and atypical antidepressants in adolescents. Keywords Poisoning; Drug overdose; Pediatric intensive care unit; Suicide; Polypharmacy
A study conducted by Fazen et al.4 in 1986 reviewed 90 admissions for poisonings over a 2-year period; of these, 58 (64%) were admitted to the ICU. Totally 42% (38 patients) poisonings were intentional and 45% of these poisonings were polypharmaceutical. Substances most commonly ingested were barbiturates, TCAs, theophylline, aspirin, and benzodiazepines. Totally 10% of patients received specific antidotal treatment and 8.9% (8 patients) required mechanical ventilation, and there was one fatality due to diphenoxylate overdose. The goals of this retrospective study were to (1) describe the demographics of patients admitted to the PICU for suspected and known poisonings; (2) determine the type and characteristics of poisonings that require PICU care; (3) describe the types of treatments provided to patients admitted for poisonings; (4) describe the outcomes of patients admitted for poisonings; and (5) identify trends related to poisonings that may help raise public awareness and prevent future poisonings.
Introduction Poisonings (unintentional and intentional) represent a significant number of preventable admissions to the pediatric intensive care unit (PICU), but data about poisonings requiring PICU-level care are limited. Poisonings continue to be a common cause of injury in children,1 and studies have suggested that poisonings account for 3–4.5% of PICU admissions,2 although these studies are more than 15 years old. In a study conducted in 1989, Lacroix et al.3 reviewed 105 PICU admissions over a 3-year period and found that poisoning accounted for 3.1% of all PICU admissions. Substances most commonly ingested were: tricyclic antidepressants (TCAs) (22%), benzodiazepines (15%), theophylline (10%), ethanol (10%), hallucinogens (8%), salicylates (8%), narcotics (8%), antihistamines (7%), and carbamazepine (5%). Treatment was usually nonspecific. Two patients required hemodialysis and one patient required an intestinal resection, and there were no fatalities. Two percent of patients had PICU stays more than 2 days. Received 27 January 2014; accepted 24 March 2014.
Methods
Address correspondence to Scot T. Bateman, MD, Department of Pediatrics, H5-524, University of Massachusetts Medical School, 55 Lake Avenue North, Worcester, MA 01655, USA. Tel: ⫹ 774-4422164. Fax: ⫹ 774-443-2062. E-mail: [email protected]
In this observational case series, all poisonings over a 5-year period (January 1, 2008 to December 31, 2012) admitted to an 11-bed medical/surgical PICU at an academic medical 519
Clinical Toxicology Downloaded from informahealthcare.com by Dokuz Eylul Univ. on 11/06/14 For personal use only.
520
K.M. Even et al.
center in New England were reviewed using electronic medical records. This PICU has a catchment area of just under one million. Decision to admit to the PICU was made by the pediatric emergency department attending and PICU attending in consultation with our toxicology service. Criteria for PICU admission was respiratory failure or significant risk of respiratory failure, depressed mental status, seizures, and cardiovascular dysfunction or significant risk of arrhythmias. Patients could be admitted to the PICU as “intensive care status” if they need critical care support, hourly monitoring, or “intermediate care status”, if they could be monitored every 2 h. Cases were identified using an admissions database with search terms “overdose,” “poisoning,” “ingestion,” “intoxication”, “suicide attempt,” and “altered mental status.” Data extracted from the medical record included: type of poisoning; age, sex, race of patient, length of stay (LOS); disposition; pediatric risk of mortality III score (calculated at the time of admission and already in the record)5; single or multiple substance(s) ingested; source of substance; substance ingested (per report); amount of substance ingested (if known); drug level (if determined); toxicology screen results (if performed); duration of mechanical ventilation if provided; and antidote/treatment received. Poisonings were categorized as unintentional or intentional based on information available in the medical record by the authors (KE, SB, and CA), who were not blinded to the study objective. When polypharmacy or combination medication poisonings were present, each substance or active ingredient was counted separately; for example, if a patient ingested acetaminophen and ibuprofen, this was counted as one acetaminophen poisoning and one ibuprofen poisoning. When analyzing data, the comprehensive urine toxicology screen results were used unless a known drug was ingested that was not screened for by the toxicology screen. In cases where no toxicology screen was done or it was negative, the toxidrome and clinical story was used for identifying the substance (i.e., candle
oil ingestion and warfarin). The study was approved by Institutional Review Board of the hospital. Comparisons between unintentional and intentional overdoses were made using T-test, Chi-square test, and Mann–Whitney U test where appropriate. P values less than 0.05 were considered statistically significant. All data were analyzed with SAS 9.2 statistical software (CAS Institute, Cary, NC).
Results There were a total of 3282 PICU admissions over this 5-year time period; 273 (8%) of these were for poisonings. The majority of admissions were from pediatric emergency room of the institution, and 20% of patients were transfers from outlying hospitals. Totally 148 (54%) poisonings were unintentional and 125 (46%) were intentional. Among them 91 patients (33%) intentionally ingested substances in attempted suicide. The age distribution of the patients was bimodal with 121 (44%), 3 years or below and 124 (45%), 13 years or above (Fig. 1). There were no intentional poisonings in patients below 10 years. One hundred fifty-one (55%) patients were female. Demographic data for the admissions are shown in Table 1. The distribution of race in this population is consistent with the county’s demographics with the exception of Hispanics/Latinos, who make up 9% of the county’s population but accounted for 17% of poisoning admissions.6 Severity of illness scores were higher among patients admitted for intentional overdoses (Table 2) and larger proportion of these patients received more intensive level of care compared to patients admitted following unintentional ingestions (56 [45%] vs. 33 [22%]; P ⫽ 0.0001). Most (96%) admissions were for less than 48 h and 41% were for less than 24 h. Mean PICU LOS was 1.2 ⫾ 0.7 days. For patients receiving mechanical ventilation, the mean PICU LOS was 1.7 ⫾ 0.9 days.
Fig. 1. Age distribution of PICU admissions for poisoning. Clinical Toxicology vol. 52 no. 5 2014
Poisonings requiring admission to the pediatric intensive care unit 521 Table 1. Demographic data. Total N ⫽ 273 Age (years, mean ⫹ SD) Male sex (N, %) Race (N, %) White Hispanic/Latino Black/African-American Asian Multi-racial American Indian/Native Alaskan Native Hawaiian Not specified
Unintentional Intentional N ⫽ 148 N ⫽ 125
8.6 ⫾ 6.8 122 (45%)
2.7 ⫾ 2.5 79 (53%)
15.6 ⫾ 1.6 43 (34%)
198 (73%) 46 (17%) 18 (7%) 5 (2%) 2 (0.7%) 1 (0.4%)
102 (69%) 28 (19%) 12 (8%) 2 (1%) 1 (0.7%) 1 (0.7%)
96 (77%) 18 (14%) 6 (5%) 3 (2%) 1 (0.8%) 0
1 (0.4%) 2 (0.7%)
1 (0.7%) 1 (0.7%)
0 1 (0.8%)
Clinical Toxicology Downloaded from informahealthcare.com by Dokuz Eylul Univ. on 11/06/14 For personal use only.
N, number of patients; SD, standard deviation
Most patients received supportive care (close monitoring including telemetry, frequent neurological and vital sign checks, and symptomatic management) without specific intervention for their ingestion. Specific therapies that were given to patients are summarized in Table 2. Patients
acquired medication and substances from a variety of places and people (Table 3). Younger patient were more likely to have obtained the medication from a parent or older relative while adolescent patients were more likely to take an overdose of their own medication, use over the counter medications or obtain drug(s) from a friend. Most patients (69%) ingested a single substance (Table 4). A total of 468 substances in 54 different drug classes7 were ingested by 273 patients. Comprehensive toxicology screens were sent for 234 (86%) patients. Ingested substances were determined by history and toxicology screen. The most commonly ingested substances were antidepressants (accounting for 13% of poisonings), analgesics (12%), opioid dependency agents (buprenorphine/naloxone and methadone) (9%), cardiac medications (8%), and alpha-2 adrenergic agonists (clonidine and guanfacine) (7%) (Supplementary Appendix 1 available online at http://informahealthcare.com/doi/abs/ 10.3109/15563650.2014.909601). Twenty-seven patients (10%) required mechanical ventilation for hypoventilation (Supplementary Appendix 2 available online at http://informahealthcare.com/doi/abs/
Table 2. Severity of illness, length of stay, and therapies provided. Total N ⫽ 273 Level of care (N, %): Intensive care Intermediate care PRISM III score (median, range) Length of stay, days (mean ⫾ SD) Disposition (N, %) Home Inpatient ward Inpatient psychiatric facility Deceased Therapies provided (N, %): Intubation Duration of MV (mean ⫾ SD) Oro/naso-gastric tube placement Activated charcoal Gastric lavage Bowel irrigation/Go-LYTELY Dialysis Reversal agents/Antidotes and other medications provided: Naloxone Benzodiazepines N-acetylcysteine Sodium bicarbonate Vasopressor Magnesium Physostigmine Antiepileptic Antipsychotic Calcium gluconate Glucagon Flumazenil Octreotide Vitamin K Diphenhydramine Atropine Pyridoxine
Unintentional N ⫽ 148
Intentional N ⫽ 125
89 (33%) 184 (67%) 0 (0–18) 1.2 ⫾ 0.7
33 (22%) 115 (78%) 0 (0–10) 1.1 ⫾ 0.6
56 (45%) 69 (55%) 0 (0–18) 1.3 ⫾ 0.8
153 (56%) 45 (16%) 74 (27%) 1 (0.4%)
127 (85%) 21 (14%) 0 0
26 (21%) 24 (19%) 74 (59%) 1 (0.8%)
27 (10%) 0.82 ⫾ 0.83 32 (12%) 41 (15%) 1 (0.4%) 3 (1%) 2 (0.7%)
7 (5%) 0.69 ⫾ 0.27 8 (5%) 22 (15%) 0 0 1 (0.7%)
20 (16%) 0.87 ⫾ 0.95 24 (19%) 19 (15%) 1 (0.8%) 3 (2%) 1 (0.8%)
44 (16%) 33 (12%) 15 (5%) 16 (6%) 6 (2%) 5 (2%) 4 (1%) 3 (1%) 3 (1%) 3 (1%) 3 (1%) 2 (0.7%) 2 (0.7%) 2 (0.7%) 2 (0.7%) 1 (0.4%) 1 (0.4%)
25 (17%) 8 (5%) 1 (0.7%) 2 (1%) 1 (0.7%) 0 0 2 (1%) 0 2 (1%) 1 (0.7%) 1 (0.7%) 2 (1%) 1 (0.7%) 1 (0.7%) 1 (0.7%) 0
19 (15%) 25 (20%) 14 (11%) 14 (11%) 5 (4%) 5 (4%) 4 (3%) 1 (0.8%) 3 (2%) 1 (0.8%) 2 (2%) 1 (0.8%) 0 1 (0.8%) 1 (0.8%) 0 1 (0.8%)
N, number of patients; SD, standard deviation; PRISM III, Pediatric Risk of Mortality score6; MV, Mechanical ventilation. Copyright © Informa Healthcare USA, Inc. 2014
522
K.M. Even et al.
Table 3. Source of substance(s) ingested.
Patient’s own medication Parent Grandparent/older relative Older sibling Friend/boyfriend/girlfriend Other (person known) OTC drug or household item available in home for use More than one source Unknown
Total N ⫽ 273
Unintentional N ⫽ 148
Intentional N ⫽ 125
61 (22%) 58 (21%) 45 (16%) 15 (5%) 16 (6%) 21 (8%) 25 (9%)
8 (5%) 48 (32%) 44 (30%) 14 (9%) 0 19 (13%) 6 (4%)
53 (42%) 10 (8%) 1 (0.8%) 1 (0.8%) 16 (13%) 2 (2%) 19 (15%)
7 (3%) 25 (9%)
0 9 (6%)
7 (6%) 16 (13%)
Clinical Toxicology Downloaded from informahealthcare.com by Dokuz Eylul Univ. on 11/06/14 For personal use only.
Data presented as number of patients (percent). OTC, Over the counter.
10.3109/15563650.2014.909601). The mean duration of mechanical ventilation was 0.82 ⫾ 0.83 days. Totally 93 patients (34%) were monitored with serial electrocardiograms (ECGs), 33 (12%) had some degree of QTc prolongation more than 440 ms, and 9 (3%) had widening of the QRS complex more than 110 ms. Unintentional poisonings The mean age among patients admitted for unintentional poisonings was 2.7 ⫾ 2.5 years. There was a greater proportion of male patients admitted for unintentional poisonings compared to intentional overdoses (N ⫽ 79 [53%] vs. N ⫽ 43 [34%]; P ⫽ 0.002). The racial composition was similar in the two groups (Table 1). Patients admitted for unintentional poisonings had a statistically shorter LOS compared to those admitted for intentional overdoses (1.1 ⫾ 0.6 days vs. 1.3 ⫾ 0.8 days, P ⫽ 0.02) though not a clinically significant difference. Nearly all patients were admitted for 2 days or less (N ⫽ 146, 99%) and many (N ⫽ 66, 45%) were admitted for less than 24 h. Most patients (N ⫽ 127, 85%) were discharged directly to home from the PICU. Patients frequently took medication belonging to a parent (N ⫽ 48, 32%) or older relative (N ⫽ 44, 30%) (Table 3), and most patients ingested a single substance (N ⫽ 124, 84%; Table 4). Patients ingested a total of 201 substances in 43 different drug classes (Supplementary Appendix 1 available online at http://informahealthcare.com/doi/abs/10.3109/1556365 0.2014.909601). The most commonly ingested substances were buprenorphine/naloxone (accounting for 17% of poisonings), cardiac medications (16%), and alpha-2 adrenergic agonists (10%). Overdoses of antidepressants and analgesics Table 4. Characteristics of ingestion. Total N ⫽ 273 Single substance ingested Multiple substances ingested Amount ingested known Initial drug levels measured
Unintentional Intentional N ⫽ 148 N ⫽ 125
188 (69%) 85 (31%)
124 (84%) 24 (16%)
64 (51%) 61 (49%)
139 (50%) 51 (19%)
70 (47%) 11 (7%)
69 (55%) 40 (32%)
were significantly less likely in unintentional poisonings compared to those of intentional overdoses (16 [8%] vs. 43 [16%]; P ⫽ 0.009 and 12 [6%] vs. 44 [16%]; P ⫽ 0.0005, respectively). Seven patients (5%) were intubated for respiratory failure, six of whom required mechanical ventilation for less than 24 h (Supplementary Appendix 2 available online at http:// informahealthcare.com/doi/abs/10.3109/15563650.2014. 909601), and there were no fatalities. This group includes one patient intentionally injected with heroin, one patient exposed to cocaine in his home, three patients unintentionally given their siblings’ medication, and two patients who were unintentionally given overdoses of their own prescribed medications. Intentional poisonings Of the 125 patients admitted for intentional overdoses, 91 (73%) had suicidal intent or intent to harm, 31 (25%) were not suicidal (recreational drug use), and 3 (2%) had unclear intent. Most patients (N ⫽ 115, 92%) were admitted for 2 days or less and 45 (36%) were admitted for less than 24 h. Most patients (N ⫽ 74, 59%) were discharged to a psychiatric facility. Patients frequently ingested their own medication (N ⫽ 53, 42%), over-the-counter medication that was readily available in the home (N ⫽ 19, 15%) or substances they obtained from a friend (N ⫽ 16, 13%; Table 3). About half of patients (N ⫽ 64, 51%) ingested a single substance (Table 4). The amount of drug ingested was known for 69 patients (55%), and initial drug levels (specifically for acetaminophen, aspirin, ethanol, lithium, anticonvulsants, and antipsychotics) were measured for 40 patients (32%). Patients intentionally ingested a total of 267 substances in 43 different drug classes (Supplementary Appendix 1 available online at http://informahealthcare.com/doi/ abs/10.3109/15563650.2014.909601). The most commonly ingested substances were analgesics (accounting for 16% of overdoses), antidepressants (16%), and alcohol and illicit drugs (11%). Patients admitted following intentional overdose were more than three times more likely to require intubation for respiratory depression/failure compared to patients admitted for unintentional poisoning (N ⫽ 20 [16%] vs. N ⫽ 7 [5%]; P ⫽ 0.002; Supplementary Appendix 2 available online at http://informahealthcare.com/doi/abs/10.3109/15563650. 2014.909601). Half of the intubated patients had a polysubstance ingestion. Several patients had multiple PICU admissions for overdoses during this period. One adolescent female was admitted thrice for heroin overdoses, and three young women were admitted twice each for polypharmacy overdoses. One adolescent patient did not survive after an intentional poisoning of bupropion, clonidine, and methylphenidate with suicidal intention. He developed cardiac arrest following a period of prolonged uncontrollable seizure activity and the family subsequently withdrew life-sustaining therapies. Two patients received hemodialysis for lithium toxicity. A 15-year-old female developed significant liver dysfunction and was briefly considered for a liver transplant Clinical Toxicology vol. 52 no. 5 2014
Poisonings requiring admission to the pediatric intensive care unit 523 following an acetaminophen overdose, but subsequently recovered.
Clinical Toxicology Downloaded from informahealthcare.com by Dokuz Eylul Univ. on 11/06/14 For personal use only.
Discussion This study is a PICU-specific report on the demographics of significant pediatric poisonings. It was undertaken to describe the current burden of poisonings in the PICU and to determine the patient population and the drugs used as well as their clinical course. The burden of hospitalization and treatment for poisonings remains relatively high and some important differences were found compared to historical reports. Poisonings represented 8.0% of all PICU admissions at our institution over this 5-year period. This is about twice the rate documented by other studies.2 A bimodal age distribution seen in this study is consistent with the PICU data presented by Lacroix et al.,3 although our patient cohort had a higher percentage of patients 13 years or above, suicidal patients, and polypharmaceutical poisonings. The drugs ingested by both young children and adolescents have changed with time. Compared with these earlier studies, a rise in opioid and clonidine unintentional ingestions was particularly noted as well as a change in adolescent drug poisonings toward antipsychotic medication combinations and away from tricyclic medications. These trends have also been shown in broader pediatric poisoning reports of large databases of poisonings, both inpatient and outpatient, as described by Finkelstein et al.8 Our study demonstrates that younger children admitted with unintentional poisonings tended to be less sick, got less intensive therapy, and were more likely to be discharged directly home compared with adolescent patients admitted with intentional overdoses. Only seven children were admitted after ingesting household items, which is far less than reported in the US and other countries,9–12 suggesting many of these poisonings do not require intensive care. Buprenorphine/naloxone, an opioid dependency agent, was the most common unintentionally ingested substance in young children. Although none of these 35 children required mechanical ventilation, 16 required naloxone, and 5 had some degree of respiratory depression. This medication is used in the northeastern United States but may not be frequently prescribed in other parts of the country; other reports have also noted an alarming increase in New England of the incidence of children ingesting this medication.13,14 This may be related to a combination of factors, including its increased use and availability to patients, its sublingual absorption so the child does not need to swallow the pill to get the toxic effects, and the orange color and candy-like appearance of the tablet. The rate of unintentional pediatric poisoning is about eight times higher for buprenorphine/ naloxone tablets versus film.14,15 Based on this data, Reckitt Benckiser Pharmaceuticals agreed to voluntarily discontinue the production of the tablet form as on March 18, 2013; it will continue to manufacture the sublingual film, which is individually wrapped.16 All 35 children in our study ingested the tablet form. Copyright © Informa Healthcare USA, Inc. 2014
The public health and safety implications of these findings include the importance of discussing medication safety, including the use of lock boxes and childproof containers, with parents and caregivers. Prescribers of buprenorphine/ naloxone should be wary of their patients living with children under 4 and consider prescribing the safer sublingual version of the drug.16 The availability of drugs in the home can also be decreased by removing old medications. The National Prescription Drug Take-Back Day, organized by the Federal Drug Enforcement Agency, aimed at removing prescription drugs from circulation, which is an example of efforts to help avoid unintentional poisonings.17 Almost all our teenage ingestions were intentional (defined as purposefully taking substances with intent to harm, intent to get high, or a combination of both). Older children admitted with intentional poisonings were sicker and tended to require a higher level of care. Most of these patients were suicidal. Suicide is the third leading cause of death in pediatric patients in the United States, with 1922 children and adolescents committing suicide in 2009.18 For every completed suicide, there are many more attempts made, with a lifetime prevalence of attempting suicide of approximately 4%.19 Because so many intentional poisonings were polypharmacy overdoses, patients on multiple antipsychotic medication should be monitored closely. Mental health services, including availability of counseling services at schools, may help reduce suicidality and, in turn, reducing intentional poisonings in adolescents. In our cohort, most poisonings received only supportive care, and complications were short-lived and self-limiting. Naloxone, activated charcoal and benzodiazepines were the treatments most commonly provided. Patients were monitored with ICU monitors, telemetry, and often serial ECGs. In order to care for poisoning patients, a PICU needs to be prepared to provide emergent intubation, cardiac monitoring, arrhythmia management, seizure control, and acute psychiatric services. The most serious complications were seen in adolescents that were attempting suicide, usually with multiple medications. There was one fatality, and our 0.4% fatality rate is consistent with the low overall mortality rate of pediatric poisonings.9 This retrospective chart review has potential limitations that impact the generalizability of this study. This study examined a specific subset of pediatric patients that required PICU-level care at one institution, so it did not capture the overall picture of all pediatric poisonings. Patients were admitted to this PICU for advanced telemetry monitoring that may be done on a cardiac inpatient ward at other institutions. The high rate of buprenorphine/naloxone poisonings may be reflective of prescribing practices among addiction specialists in our area and may not be seen in other parts of the country.
Conclusion Unintentional and intentional poisonings in childhood continue to be a public health concern with many children requiring PICU-level care. This study identified an increased
524
K.M. Even et al.
rate of PICU admissions for poisoning compared 25 years ago with more suicidal patients ingesting multiple medications and more patients requiring mechanical ventilation. There were changes in the types of substances ingested compared with earlier studies, particularly a higher rate of buprenorphine/naloxone poisoning. Although the majority of poisonings are non-fatal and the patients received mostly supportive care, the burden of hospitalization on the patients, their families, and health care resources cannot be ignored. Additional studies are needed to identify public health interventions that may help to decrease the poisoning rate.
Take Home Points: Clinical Toxicology Downloaded from informahealthcare.com by Dokuz Eylul Univ. on 11/06/14 For personal use only.
• • • •
Poisonings requiring Pediatric ICU care are common and the types of drugs have changed over time. Unintentional poisonings by younger children usually involve a parent/caregiver’s drugs. Intentional poisonings by adolescents involve multiple drugs, including newer antipsychotics. A significant rise in buprenorphine/naloxone ingestions in young children was noted.
Funding Source No external funding was secured for this study.
Financial Disclosure The authors have no financial relationships relevant to this article to disclose.
Declaration of interest The authors report no declarations of interest. The authors alone are responsible for the content and writing of the paper.
References 1. Bond GR, Woodward RW, Ho M. The growing impact of pediatric pharmaceutical poisoning. J. Pediatr 2012; 160:265–270.e1. 2. Vernon DD, Gleich MC. Poisoning and drug overdose. Crit Care Clin 1997; 13:647–667. 3. Lacroix J, Gaudreault P, Gauthier M. Admission to a pediatric intensive care unit for poisoning: a review of 105 cases. Crit Care Med 1989; 17:748–750.
4. Fazen LE III, Lovejoy FH Jr, Crone RK. Acute poisoning in a children’s hospital: a 2-year experience. Pediatrics 1986; 77:144–151. 5. Pollack MM, Patel KM, Ruttimann UE. PRISM III: an updated Pediatric Risk of Mortality score. Crit Care Med 1996; 24:743–752. 6. U.S. Department of Commerce, United States Census Bureau, 2010 Census Interactive Population Search. Available at: http://www.census. gov/2010census/popmap/ipmtext.php. Accessed August 14, 2013. 7. Micromedex® Healthcare Series [Internet database]. Greenwood Village, Colo: Thomson Healthcare. Updated periodically. 2013. 8. Finkelstein Y, Hutson JR, Wax PM, Brent J, on behalf of the Toxicology Investigators Consortium (ToxIC) Case Registry. Toxico-surveillance of infant and toddler poisonings in the United States. J Med Toxicol 2012; 8:263–266. 9. Bronstein AC, Spyker DA, Cantilena LR Jr, Rumack BH, Dart RC. 2011 Annual report of the American Association of Poison Control Centers’ National Poison Data System (NPDS): 29th Annual Report. Clin Toxicol (Phila) 2012; 50:911–1164. 10. Jayashree M, Singhi S. Changing trends and predictors of outcome in patients with acute poisoning admitted to the intensive care. J Trop Pediatr 2011; 57:340–346. 11. Kohli U, Kuttiat VS, Lodha R, Kabra SK. Profile of childhood poisoning at a tertiary care centre in North India. Indian J Pediatr 2008; 75:791–794. 12. Andiran N, Sarikayalar F. Pattern of acute poisonings in childhood in Ankara: what has changed in twenty years? Turk J Pediatr 2004; 46:147–152. 13. Pedapati EV, Bateman ST. Toddlers requiring pediatric intensive care unit admission following at-home exposure to buprenorphine/naloxone. Pediatr Crit Care Med 2011; 12:e102–e107. 14. Martin TC, Rocque MA. Accidental and non-accidental ingestion of methadone and buprenorphine in childhood: a single center experience, 1999–2009. Curr Drug Saf 2011; 6:12–16. 15. Lavonas EJ, Banner W, Bradt P, Bucher-Bartelson B, Brown KR, Rajan P, et al. Root causes, clinical effects, and outcomes of unintentional exposures to buprenorphine by young children. J Pediatr 2013; 163:1377–83.e1–3. doi:10.1016/j.jpeds.2013.06.058. Epub 2013 Aug 29. 16. RB Press Release, Reckitt Benckiser Pharmaceuticals Inc. Submits Citizen Petition to US FDA Requesting Action to Mitigate Risk of Pediatric Exposure with Opioid Dependence Treatment; Company Voluntarily Discontinues the Supply of Suboxone® Tablets (buprenorphine and naloxone sublingual tablets [CIII]) in the United States. Available at: http://www.rb.com/site/rkbr/templates/mediainvestorsgeneral2.aspx?pageid ⫽ 1332 & cc ⫽ gb. Accessed April 9, 2013. 17. National Take-Back initiative. Available at: http://www.deadiversion. usdoj.gov/drug_disposal/takeback. Accessed April 28, 2013. 18. Kochanek KD, Kirmeyer SE, Martin JA, Strobino DM, Guyer B. Annual summary of vital statistics: 2009. Pediatrics 2012; 129: 338–348. 19. Nock MK, Green JG, Hwang I, McLaughlin KA, Sampson NA, Zaslavsky AM, Kessler RC. Prevalence, correlates, and treatment of lifetime suicidal behavior among adolescents: results from the National Comorbidity Survey Replication Adolescent Supplement. JAMA Psychiatry 2013; 70:300–310.
Supplementary material available online Supplementary Appendix Tables 1 and 2.
Clinical Toxicology vol. 52 no. 5 2014
