- Email: [email protected]
Prevalence of autonomic signs and symptoms in antimuscarinic drug poisonings
The Journal of Emergency Medicine, Vol. 26, No. 1, pp. 89 –94, 2004 Copyright © 2004 Elsevier Inc. Printed in the USA. All rights reserved 0736-4679/04 $–see front matter
doi:10.1016/j.jemermed.2003.09.003
Selected Topics: Toxicology
PREVALENCE OF AUTONOMIC SIGNS AND SYMPTOMS IN ANTIMUSCARINIC DRUG POISONINGS Raj J. Patel,
MD,*
Tim Saylor,
MD,†
Saralyn R. Williams,
MD,*‡
and Richard F. Clark,
MD*‡
*Division of Medical Toxicology, Department of Emergency Medicine, University of California, San Diego Medical Center, San Diego, California, †The Department of Emergency Medicine, Kern County Medical Center, Bakersfield, California, and ‡California Poison Control System, San Diego Division, San Diego, California Reprint Address: Richard F. Clark, MD, Department of Emergency Medicine, UCSD Medical Center, 200 West Arbor Drive, San Diego, CA 92103-8676
e Abstract—Classically described antimuscarinic poisoning signs and symptoms include mydriasis, decreased secretions, ileus, urinary retention, hyperthermia, tachycardia, and altered mental status. These features may be used clinically to assist in the diagnosis of patients with unknown poisonings. We sought to analyze the prevalence of antimuscarinic physical examination findings in evaluating patients presenting with acute poisoning from antimuscarinic agents. We conducted a retrospective, medical record review at two urban tertiary care teaching hospitals. The study population consisted of patients presenting to the Emergency Department with a diagnosis of acute poisoning secondary to medications with known antimuscarinic side effects during a 78-month period between January 1994 and July 2001. Cases were excluded for incomplete medical records or unreliable histories of ingestion, and when concomitant ethanol intoxication was present on laboratory analysis. Clinical information obtained from each patient included vital signs, pupillary size, electrocardiogram abnormalities, the presence of mucous membrane and axillary secretions, initial urine output after bladder catheterization, quality of bowel sounds, mental status changes, the occurrence of seizures and coma, need for orotracheal intubation, and time required for clinical resolution. Diagnostic and therapeutic information including laboratory tests, administration of sodium bicarbonate, and usage of physostigmine was also collected. We identified a total of 345
cases, 213 of which met inclusion criteria. Of these cases, the most common documented findings included decreased secretions in 75.1%, tachycardia in 68.1%, confusion in 49.3%, drowsiness in 48.2%, and hypoactive or absent bowel sounds in 44.6%. Combining signs and symptoms to predict this toxic syndrome was not very reliable. Tachycardia, decreased oral or axillary secretions, and mydriasis proved to be the most predictive trio of clinical signs, but were found in only 28.2% of cases. At least one of these three signs was documented in 94% of our patients. The combination of tachycardia and decreased secretions was the most common pair of findings, recorded in 55.4% of cases. We conclude that the clinical presentation of antimuscarinic syndrome is variable. © 2004 Elsevier Inc. e Keywords—antimuscarinic; anticholinergic; poisoning; physical examination
INTRODUCTION Poisonings from medications with known antimuscarinic toxicity have been commonly reported and studied (1). Toxicity from antimuscarinic plants such as Datura have been described for centuries. The diagnosis of antimuscarinic toxicity is often initially based on history, clinical intuition, and physical examination. The recognition of peripheral and central signs and symptoms of antimuscarinic agents is supplemented by laboratory markers to
Presented at the 2002 North American Congress of Clinical Toxicology, Palm Springs, California.
RECEIVED: 25 November 2002; FINAL ACCEPTED: 2 September 2003
SUBMISSION RECEIVED:
31 July 2003;
89
90
R. J. Patel et al.
aid health care professionals in making accurate diagnoses. Early studies in the 1970s investigated the frequency of clinical features associated with tricyclic antidepressant (TCA) poisoning with an emphasis on electrocardiogram (EKG) abnormalities, whereas recent publications describe the symptomatology manifested by non-traditional antimuscarinic sources such as tea made from jimson weed (Datura stramonium and others) and heroin mixed with scopolamine (1– 8). Peripheral autonomic signs of muscarinic receptor blockade include dilatated and non-reactive pupils, dry and flushed skin and mucous membranes, hypoactive or absent bowel sounds, urinary retention, tachycardia, and hyperthermia (7–13). Central nervous system manifestations of antimuscarinic agents are largely dependent on the compound’s permeability through the blood-brain barrier. Toxic effects can resemble acute psychosis. Signs and symptoms include confusion, hallucinations (visual greater than tactile or auditory), disturbed speech, and most commonly, agitation (9,10). In severe cases, convulsions and coma may occur. Drugs and compounds that block muscarinic cholinergic receptors have been reported to produce the “anticholinergic toxidrome.” This phrase can be somewhat confusing because truly nonspecific anticholinergic agents would inhibit both muscarinic and nicotinic cholinergic receptors, including the neuromuscular junction. The overall incidence of these clinical features can be variable and without uniformity, complicating the ability to correctly formulate an early diagnosis. We therefore conducted this study to investigate the prevalence of antimuscarinic signs and symptoms in patients presenting to the Emergency Department (ED) with acute poisoning from agents with antimuscarinic effects.
MATERIALS AND METHODS We retrospectively reviewed the charts of all consecutive patients evaluated by our medical toxicology service (MTS) between January 1994 and July 2001. Cases were identified using ICD-9 codes for poisoning due to agents with potential for antimuscarinic toxicity using medical records and a registry of cases kept by the MTS. The study sites were two urban tertiary care referral centers located in close proximity with comparable patient populations. Inclusion criteria for cases consisted of 1) a history of acute poisoning or overdose with a primary agent known to possess significant antimuscarinic activity (TCA, phenothiazines, antihistamines, antispasmodic agents, antiparkinsonian agents, belladonna alkaloids, and Datura), 2) antimuscarinic drug presence with qualitative urine toxicology screening (confirmed by gas chromatography/
mass spectroscopy), and 3) complete medical records. Histories were verified as “reliable” in the chart either when confirmed by a knowledgeable family member or friend, or as related by the patient after resolution of toxicity. Patients were excluded if their medical records were incomplete or if their primary poisoning consisted of any other type of ingestion based on history, physical examination, or laboratory data. Patients with unreliable or uncertain histories of poisoning as documented in the medical record were also excluded, as were those with blood or breath ethanol determinations greater than 100 mg/dL. Cases in which benzodiazepines or other sedative hypnotic agents were found by history or urine screening were included except in the sub-analysis of mental status. These patients were included in the analysis of confusion. Patients with histories or urine screens with opioids present were included because 2 mg or more of naloxone was administered in every case where mental status was not normal before examination by the MTS (patients with symptoms or complaints of opioid withdrawal after naloxone were excluded). Complete medical records were defined as those consisting of identifiable physical examinations, history, laboratory values, and progress notes during toxicity and after the resolution of symptoms. Patients with positive urine toxicology screens for sympathomimetic compounds were included if the antimuscarinic agent was noted in the medical record to be the primary poison, and there was no indication in the chart that an overdose or recent (less than 24 h before arrival) use of a sympathomimetic agent had occurred. Background demographic data obtained from medical records included date and time of initial presentation, age, gender, primary intoxicant, co-ingestants (if any), and time until the patient was clinically cleared from the poisoning as determined by the MTS. Clinical information extracted included vital signs (pulse, blood pressure, temperature, and respiratory rate), pupillary size and reactivity, presence or absence of oral secretions, presence or absence of axillary perspiration, urine output on initial urinary bladder catheterization (if performed), quality of bowel sounds, mental status (Glascow Coma Score [GCS] and the presence of confusion, hallucinations, agitation, drowsiness, dysarthria, convulsions, and coma as documented in the medical record), and the occurrence of hypotension at any point during the patient’s hospital course. All findings recorded in the medical record were compared among the various participants in each patient’s medical care to check the consistency of each finding. When obvious differences in a recorded finding were noted in the record, the most commonly reported finding was included in the analysis. If no clear majority of opinion was present in the record about a sign or symptom, the physical examination of the most senior physician examiner was used for analysis.
Antimuscarinic Poisonings
For the purposes of analysis, “tachycardia” was defined as heart rate greater than 100 beats per minute (bpm). “Urinary retention” was defined as presence of greater than 200 mL of urine upon initial bladder catheterization. “Mydriasis” was defined as pupillary size greater than 4 mm in room lighting. “Hypotension” was defined as systolic blood pressure less than 100 mm Hg. “Agitation,” “drowsiness” and “dysarthria” were noted as subjectively recorded by the examiner in the medical record. “Confusion” was assessed for the entire group and defined as disorientation to person, place or date, combined with agitation; patients with drowsiness or coma were not included in this category. The presence or absence of oral secretions and axillary perspiration were subjective, and taken from documentation in the medical records as routinely recorded in notes by the MTS. All the above definitions were applied to and recorded from only the notes of the initial presentation findings on physical examination in the ED. Laboratory data collected consisted of EKG abnormalities, including QRS duration and corrected QT duration, and comprehensive urine toxicology screening. Therapeutic interventions recorded included the use of sodium bicarbonate, intubation, and physostigmine when applicable. Subjective time to resolution of all symptoms as recorded in the nursing and physician records was also determined. Statistical analysis was applied using Analysis of Variance (ANOVA). A probability value of ⬍ 0.05 was accepted as statistically significant. The study was reviewed and approved by the investigational review boards of both institutions.
RESULTS We identified a total of 345 cases of poisoning involving potentially antimuscarinic agents at two medical centers during the study period. Of these, 132 were excluded based on incomplete medical records, another primary poisoning, inability to confirm the poisoning agent through laboratory testing, or ethanol level greater than 100 mg/dL. No cases of opioid withdrawal required exclusion. A total of 213 cases were identified that met all criteria for study. Patient demographic information and admission vital signs are presented in Table 1. The vast majority of cases involved four main types of potentially antimuscarinic compounds: tricyclic antidepressants (TCAs), antihistamines, antiparkinsonian agents, and antipsychotics (Table 2). Antihistamines included all sedating medicinal antihistamine products, such as diphenhydramine, hydroxyzine, and chlorpheniramine. Other common agents encountered in our group included carbamazepine (n ⫽
91 Table 1. Demographics and Vital Signs of the 213 Patients Included in this Series Age (years) Gender (male:female) Vital signs Pulse (beats/min) SBP (mm Hg) DBP (mm Hg) RR (per minute) Temperature ⬎ 38°C (100.4°F), patients (%)
36 ⫾ 11.8 134:79 111 ⫾ 25.3 129 ⫾ 22.7 77 ⫾ 16.1 19 ⫾ 5.1 10 (4.7)
Averages are listed with standard deviations. SBP ⫽ systolic blood pressure; DBP ⫽ diastolic blood pressure; RR ⫽ respiratory rate.
10), belladonna alkaloids including scopolamine and atropine (n ⫽ 5), Datura sp. plant ingestion (n ⫽ 5), and multiple-drug antimuscarinic ingestions (n ⫽ 7). The average age, respiratory rate, and temperature measurements were comparable and statistically insignificant among the four main categories of agents when compared to the entire group. The mean pulse and systolic blood pressure (119 ⫾ 27.2 beats per minute [bpm] and 137 ⫾ 17.2 mm Hg, respectively) of the antihistamine group were statistically higher than the total means (111
Table 2. Classes and Types of Agents Included in Analysis Class of drug
Number of patients
Antihistamines Diphenhydramine Hydroxyzine Chlorpheniramine Doxylamine Tricyclic antidepressants Amitriptyline Doxepin Imipramine Nortriptyline Desipramine Unknown TCA Antipsychotics Thioridizine Chlorpromazine Perphenazine Olanzapine Thiothixene Mesoridazine Prochlorperazine Loxapine Unknown phenothiazine Antiparkinsonian agents Benztropine Trihexyphenidyl Other Carbamazepine Multiple antimuscarinics Datura spp. plants Atropine Scopolamine Cyclobenzaprine
66 60 3 2 1 73 38 11 10 7 6 1 25 8 6 3 2 2 1 1 1 1 20 16 4 29 10 7 5 4 1 2
92
R. J. Patel et al.
Table 3. Physical Examination Findings Related to Type of Exposure Antihistamines (n ⫽ 66)
TCA (n ⫽ 73)
Benztropine/ Antipsychotics trihexyphenidyl (n ⫽ 25) (n ⫽ 20)
Other (n ⫽ 29)
Totals (n ⫽ 213)
Pupil size 3.9 ⫾ 1.2 mm 3.6 ⫾ 1.2 mm 3.2 ⫾ 1.2 mm 4.4 ⫾ 1.2 mm 4.0 ⫾ 1.0 mm 3.8 ⫾ 1.3 mm Decreased secretions (%) 45 (68) 59 (87) 18 (72) 16 (80) 22 (76) 160 (75.1) Urine output (n ⫽ 83) 659 ⫾ 586 mL 573 ⫾ 527 mL 1055 ⫾ 720 mL 968 ⫾ 478 mL 589 ⫾ 566 mL 700.3 ⫾ 584.3 mL Bowel sounds Hypoactive bowel sounds (%) 23 (34.8) 32 (43.8) 15 (60) 10 (50) 15 (51.7) 95 (44.6) Absent bowel sounds (%) 3 (4.5) 5 (6.8) 1 (4) 1 (5) 3 (10.3) 13 (6.1) Hypotension, SBP ⬍ 100 2 (3.0) 9 (12.3) 4 (16) 2 (10) 0 17 (7.9) mm Hg Mental status (n ⫽ 168) (n ⫽ 54) (n ⫽ 59) (n ⫽ 19) (n ⫽ 15) (n ⫽21) Average GCS 11.9 ⫾ 3.3 10.5 ⫾ 4.3 12.6 ⫾ 3.3 11.0 ⫾ 3.3 11.4 ⫾ 3.4 11.7 ⫾ 3.8 Confusion (%)* 33 (50) 28 (38.4) 14 (56) 15 (75) 15 (51.7) 105 (49.3) Hallucinations (%) 15 (27.8) 11 (18.6) 8 (42.1) 6 (40) 4 (19) 44 (26.2) Agitation (%) 15 (27.8) 17 (28.8) 7 (36.8) 4 (26.7) 6 (28.6) 49 (29.2) Drowsy (%) 26 (48.1) 25 (42.4) 12 (63.2) 7 (46.7) 11 (52.4) 81 (48.2) Dysarthria (%) 16 (29.6) 15 (25.4) 5 (26.3) 8 (53.3) 12 (57.1) 56 (33.3) Seizure (%) 10 (18.5) 3 (5.1) 0 0 2 (9.5) 15 (8.9) Coma (%) 5 (9.3) 21 (38.9) 0 2 (13.3) 2 (9.5) 30 (17.9) Time to clear (hours) 24.6 ⫾ 15.7 40.1 ⫾ 28.8 26.6 ⫾ 18.4 45.6 ⫾ 41.6 31.8 ⫾ 24.1 32.6 ⫾ 26.0 Physostigmine use 16/1.14 5/0.9 6/1.5 9/0.94 6/1.25 50/1.1 (cases/average in mg) Sodium bicarbonate 1 (1.5) 14 (19.2) 0 1 (5.0) 2 (6.9) 18 (8.4) administered Intubated (%) 9 (13.6) 38 (52.1) 3 (12) 5 (25) 5 (17.2) 60 (27.9) QRS ⬎ 120ms (%) 2 (3.0) 17 (23.3) 2 (8) 1 (5) 0 22 (10.2) Data represent number of patients with individual findings present on initial examination, except where stated. The subset analyzed for altered mental status consisted of 168 cases, excluding those with history or laboratory evidence of sedative hypnotic overdose. TCA ⫽ tricyclic antidepressants; GCS ⫽ Glasgow Coma Scale. * “Confusion” was analyzed for the entire group of 213 patients.
⫾ 29.3 bpm and 129 ⫾ 22.7 mm Hg, respectively, p ⬍ 0.05 and p ⬍ 0.01, respectively). No other statistically significant differences existed in comparisons of vital signs among groups or compared to the total means. Physical examination findings and laboratory measurements are listed in Table 3. Mean pupillary size was 3.8 ⫾ 1.3 mm. A total of 160 cases (75%) had documented decreased skin or mucous membrane moisture in the medical record. The mean urine output on initial bladder catheterization was 700 ⫾ 584cc from 83 cases requiring catheter placement. Hypoactive or absent bowel sounds were noted in 44.6% of cases, whereas the absence of bowel sounds was recorded in 6.1% of all cases. Sinus tachycardia was documented in 145 cases (68%). Electocardiographic widening of the QRS complex (as defined by duration greater than 120 ms) was present in 17 of the 22 patients in the TCA subgroup. Ventricular tachycardia was documented in only one case (TCA poisoning), and there was one death in another patient ingesting TCAs. Hypotension was present in 7.9% of the study patients. Analysis of the central nervous system effects of antimuscarinic poisoning yielded a mean Glasgow Coma Scale (GCS) score of 11.7 ⫾ 3.8. A total of 45 cases
were excluded in this sub-analysis due to urine screening positive for benzodiazepines or other sedatives, or a history of concomitant overdose of these agents, leaving 168 cases in this subgroup. Drowsiness was observed in 48.2% of cases, confusion in 49.3%, dysarthria in 33.3%, agitation in 29.2%, and hallucinations in 26.2%. The majority of hallucinations were recorded as being visual and often accompanied by “picking” behavior. Convulsions were noted in 8.9% of all subgroup patients, with antihistamines accounting for 10 of the 16 cases. Coma was reported in 30 patients (17.9%) in the subgroup, 21 diagnosed with TCA ingestions. Intubation secondary to respiratory depression or airway protection was necessary in 38 (27.9%) patients in the TCA subgroup, accounting for 63% of all intubations in the study (60 total). Physostigmine was administered to 23.3% (49) of all patients with a mean dose of 1.1 mg per case administered (0.5–2.0 mg). Two of these cases were tricyclic antidepressant poisonings without QRS duration abnormalities. All patients were reported to have improvements in mental status, and no significant adverse reactions were noted with the use of this agent. Sodium bicarbonate was administered to 8.4% of all study patients, with TCA poisonings consisting of 14 of the total
Antimuscarinic Poisonings
93
Table 4. Combined Frequencies of Physical Examination Findings Among 213 Patients in this Series Physical examination finding
Number of cases (%)
Tachycardia Mydriasis ⬎ 4 mm Mydriasis ⬎ 5 mm Confusion Decreased secretions Tachycardia and mydriasis ⬎ 4 mm Tachycardia and decreased secretions Decreased secretions and confusion Tachycardia and confusion Confusion and mydriasis ⬎ 4 mm Confusion and mydriasis ⬎ 5 mm Decreased secretions and mydriasis ⬎ 4 mm Tachycardia and decreased secretions and mydriasis ⬎ 4 mm Tachycardia and confusion and mydriasis ⬎ 4 mm Tachycardia and decreased secretions and confusion Tachycardia or confusion or mydriasis ⬎ 4 mm Tachycardia or confusion or mydriasis ⬎ 5 mm Tachycardia or decreased secretions or confusion Tachycardia or decreased secretions or mydriasis ⬎ 4 mm
145 (68.1) 114 (53.5) 61 (28.6) 105 (49.3) 160 (75.1) 73 (35.3) 118 (55.4) 88 (41.3) 70 (32.9) 64 (30.0) 40 (18.8) 88 (41.3) 60 (28.2) 39 (18.3) 58 (27.2) 198 (93.0) 185 (86.9) 202 (94.8) 201 (94.4)
Decreased secretions ⫽ either decreased oral saliva or axillary perspiration.
18 cases. In terms of hospital course, the mean time to “medical clearance” by the MTS from time of initial patient presentation to the ED was 32.6 h. Sixty-nine cases (32%) received a single dose of activated charcoal during the course of their admission to the hospital. No patient had any other gastric emptying procedure performed. We also analyzed combinations of physical examination findings in our cases of antimuscarinic toxicity. When three of the most commonly encountered features of toxicity (tachycardia, decreased oral or axillary secretions, and mydriasis) were evaluated together, we found that 94% (201) of cases displayed at least one of these findings. However, only 28% (60) of all cases were described as exhibiting all three of these signs of toxicity. Tachycardia, confusion, and mydriasis together were reported in 18.3% (39), whereas at least one of these signs was present in 93% (198) of cases. The frequency with which other combinations were encountered in our study is listed in Table 4. DISCUSSION Antimuscarinic compounds include a diverse group of agents, but are predominated by medications prescribed
for their psychoactive potential. Some, such as diphenhydramine and scopolamine, may be misused or abused specifically for their ability to cause hallucinations and alterations in mental status. Antimuscarinic central nervous system effects can present in a variety of forms, but are often manifested in our practice as “passive psychosis:” patients are confused and disoriented, but lack the violent and paranoid activity of sympathomimetic poisoned individuals. Another frequent finding in our patients with antimuscarinic poisoning is “picking” activity. Affected individuals tend to pick or play curiously with intravenous tubing, restraints, bedclothes, and other unfamiliar items within reach, but seldom resort to more aggressive behavior such as pulling out intravenous lines. Our retrospective evaluation identified the frequency of occurrence of some of the most commonly described physical examination findings listed in charts of patients with antimuscarinic poisoning. Our most frequent finding was dry mucous membranes, but sinus tachycardia and anhydrosis were also repeatedly encountered. Over 90% of our cases exhibited at least one of the most commonly described physical examination signs of this syndrome. However, we were somewhat surprised at the sporadic recording of many of the signs and symptoms most commonly described in previous articles and textbooks. It may be that some agents, such as tricyclic antidepressants and phenothiazine derivatives, affect so many different receptor sites in the peripheral and central nervous system that predicting the clinical presentation of a particular individual is impossible. In addition, the dose or affinity for the receptor of the antimuscarinic drug ingested may influence the degree of blockade and therefore the observed signs and symptoms. One of the limitations of our study was that milder cases of antimuscarinic poisoning could have been missed or misdiagnosed by the examining physician. This could have subjected our results to spectrum bias. Our study has other limitations, most inherent with retrospective chart reviews. The subjectivity of observations such as pupillary size, presence of secretions, and level of confusion are obviously a factor in the analysis of data. The findings in our review were checked for consistency among the different health care providers with written notes in the medical record, but are still subject to interpretation. The use of specific definitions in our methods also helped to standardize certain assessments. In addition, we included some patients in whom specific characteristics were not mentioned in the medical records. In these cases, the specific finding was assumed to be negative. This may have biased our results, however, it is the custom with our MTS fellows and attendings who examined most of these patients to list with great detail all positive findings.
94
R. J. Patel et al.
There is also a possibility that the presence of multiple drugs in some cases could have affected physical examination findings. The most obvious examples would be the effect of sedative-hypnotic agents on mental status or sympathomimetics on vital signs. For these reasons, we attempted to exclude agents such as ethanol and sedatives when evaluating levels of consciousness, and sympathomimetic agents where a possibility of recent use or overuse had occurred. However, these agents may still have contributed to some of the symptomatology in our cases. Finally, we were unable to follow the “evolution” of antimuscarinic signs and symptoms in our cases. It is possible that some patients may have developed changes in findings throughout the hospitalization that could have affected results, even though most antimuscarinic signs will present early in the course of the typical poisoning with these agents. Our review was not designed to look for incidental descriptions of the above positive findings if recorded in the record after patients left the ED. Therefore, we are not able to say with certainty that delayed occurrence or evolution of signs and symptoms (besides resolution) did not occur. It is also possible that activated charcoal administration could have affected the occurrence of antimuscarinic toxicity in some cases.
CONCLUSION We performed a retrospective review of all consecutive cases of poisoning involving muscarinic receptor antagonists seen by our medical toxicology service for a 61⁄2-year period to observe the incidence of physical examination findings. We found a high incidence of examination findings consistent with this syndrome, but failed to find any sign, symptom, or combination of findings that could predict toxicity in all subjects. Sinus tachycardia, dry skin and mucous membranes, and the presence of confusion were the signs most commonly
found in our series. The observed frequency of our findings demonstrates that certain examination characteristics may assist in the assessment of the patient poisoned with antimuscarinic agents, but the variety of presentations and lack of consistency in combined findings, especially when multiple drugs are involved, may limit the objective predictive value of the physical examination in these cases.
REFERENCES 1. Thorstrand C. Clinical features in poisonings by tricyclic antidepressants with special reference to the ECG. Acta Med Scand 1976;199:337– 44. 2. Biggs JT, Spiker DG, Petit JM, Ziegler VE. Tricyclic antidepressant overdose, incidence of symptoms. JAMA 1977;238:135– 8. 3. Noble J, Matthew H. Acute poisoning by tricyclic antidepressants: clinical features and management in 100 patients. Clin Toxicol 1969;2:403–12. 4. Thorstrand C. Cardiovascular effects of poisoning with tricyclic antidepressants. Acta Med Scand 1974;195:505–14. 5. Williams RB, Sherter C. Cardiac complications of tricyclic antidepressant therapy. Ann Intern Med 1971;74:395– 8. 6. Centers for Disease Control. Scopolamine poisoning among heroin users—New York City, Newark, Philadelphia, and Baltimore, 1995 and 1996. MMWR Morb Mortal Wkly Rep 1996;45:457– 60. 7. Coremans P, Lambrecht G, Schepens P, Vanwelden J, Verhaegen H. Anticholinergic intoxication with commercially available thorn apple tea. J Toxicol Clin Toxicol 1994;32:589 –92. 8. Beaver KM, Gavin TJ. Treatment of acute anticholinergic poisoning with physostigmine. Am J Emerg Med 1998;16:505–7. 9. American Hospital Formulary Service Drug Information. 1999; 1208.8.1065. 10. Brown JH, Taylor P. Muscarinic receptor agonsists and antagonists. In: Hardman JG, Limbird LE, eds. Goodman & Gilman’s: the pharmacologic basis of therapeutics, 9th edn. New York: Macmillan Publishing Co.; 1996:155–74. 11. Clark RF, Williams SR, Nordt SP, Pearigen PD, Deutsch R. Resource-use analysis of a medical toxicology consultation service. Ann Emerg Med 1998;31:705–9. 12. Han L, McCusker J, Cole M, Abrahamowicz M, Primeau F, Elie M. Use of medications with anticholinergic effect predicts clinical severity of delirium symptoms in older medical inpatients. Arch Intern Med 2001;161:1099 –105. 13. Benowitz NL. Tricyclic antidepressants. In: Olson KR, ed. Poisoning & drug overdose, 3rd edn. Stamford, CT: Appleton & Lange; 1999:310 –2.
doi:10.1016/j.jemermed.2003.09.003
Selected Topics: Toxicology
PREVALENCE OF AUTONOMIC SIGNS AND SYMPTOMS IN ANTIMUSCARINIC DRUG POISONINGS Raj J. Patel,
MD,*
Tim Saylor,
MD,†
Saralyn R. Williams,
MD,*‡
and Richard F. Clark,
MD*‡
*Division of Medical Toxicology, Department of Emergency Medicine, University of California, San Diego Medical Center, San Diego, California, †The Department of Emergency Medicine, Kern County Medical Center, Bakersfield, California, and ‡California Poison Control System, San Diego Division, San Diego, California Reprint Address: Richard F. Clark, MD, Department of Emergency Medicine, UCSD Medical Center, 200 West Arbor Drive, San Diego, CA 92103-8676
e Abstract—Classically described antimuscarinic poisoning signs and symptoms include mydriasis, decreased secretions, ileus, urinary retention, hyperthermia, tachycardia, and altered mental status. These features may be used clinically to assist in the diagnosis of patients with unknown poisonings. We sought to analyze the prevalence of antimuscarinic physical examination findings in evaluating patients presenting with acute poisoning from antimuscarinic agents. We conducted a retrospective, medical record review at two urban tertiary care teaching hospitals. The study population consisted of patients presenting to the Emergency Department with a diagnosis of acute poisoning secondary to medications with known antimuscarinic side effects during a 78-month period between January 1994 and July 2001. Cases were excluded for incomplete medical records or unreliable histories of ingestion, and when concomitant ethanol intoxication was present on laboratory analysis. Clinical information obtained from each patient included vital signs, pupillary size, electrocardiogram abnormalities, the presence of mucous membrane and axillary secretions, initial urine output after bladder catheterization, quality of bowel sounds, mental status changes, the occurrence of seizures and coma, need for orotracheal intubation, and time required for clinical resolution. Diagnostic and therapeutic information including laboratory tests, administration of sodium bicarbonate, and usage of physostigmine was also collected. We identified a total of 345
cases, 213 of which met inclusion criteria. Of these cases, the most common documented findings included decreased secretions in 75.1%, tachycardia in 68.1%, confusion in 49.3%, drowsiness in 48.2%, and hypoactive or absent bowel sounds in 44.6%. Combining signs and symptoms to predict this toxic syndrome was not very reliable. Tachycardia, decreased oral or axillary secretions, and mydriasis proved to be the most predictive trio of clinical signs, but were found in only 28.2% of cases. At least one of these three signs was documented in 94% of our patients. The combination of tachycardia and decreased secretions was the most common pair of findings, recorded in 55.4% of cases. We conclude that the clinical presentation of antimuscarinic syndrome is variable. © 2004 Elsevier Inc. e Keywords—antimuscarinic; anticholinergic; poisoning; physical examination
INTRODUCTION Poisonings from medications with known antimuscarinic toxicity have been commonly reported and studied (1). Toxicity from antimuscarinic plants such as Datura have been described for centuries. The diagnosis of antimuscarinic toxicity is often initially based on history, clinical intuition, and physical examination. The recognition of peripheral and central signs and symptoms of antimuscarinic agents is supplemented by laboratory markers to
Presented at the 2002 North American Congress of Clinical Toxicology, Palm Springs, California.
RECEIVED: 25 November 2002; FINAL ACCEPTED: 2 September 2003
SUBMISSION RECEIVED:
31 July 2003;
89
90
R. J. Patel et al.
aid health care professionals in making accurate diagnoses. Early studies in the 1970s investigated the frequency of clinical features associated with tricyclic antidepressant (TCA) poisoning with an emphasis on electrocardiogram (EKG) abnormalities, whereas recent publications describe the symptomatology manifested by non-traditional antimuscarinic sources such as tea made from jimson weed (Datura stramonium and others) and heroin mixed with scopolamine (1– 8). Peripheral autonomic signs of muscarinic receptor blockade include dilatated and non-reactive pupils, dry and flushed skin and mucous membranes, hypoactive or absent bowel sounds, urinary retention, tachycardia, and hyperthermia (7–13). Central nervous system manifestations of antimuscarinic agents are largely dependent on the compound’s permeability through the blood-brain barrier. Toxic effects can resemble acute psychosis. Signs and symptoms include confusion, hallucinations (visual greater than tactile or auditory), disturbed speech, and most commonly, agitation (9,10). In severe cases, convulsions and coma may occur. Drugs and compounds that block muscarinic cholinergic receptors have been reported to produce the “anticholinergic toxidrome.” This phrase can be somewhat confusing because truly nonspecific anticholinergic agents would inhibit both muscarinic and nicotinic cholinergic receptors, including the neuromuscular junction. The overall incidence of these clinical features can be variable and without uniformity, complicating the ability to correctly formulate an early diagnosis. We therefore conducted this study to investigate the prevalence of antimuscarinic signs and symptoms in patients presenting to the Emergency Department (ED) with acute poisoning from agents with antimuscarinic effects.
MATERIALS AND METHODS We retrospectively reviewed the charts of all consecutive patients evaluated by our medical toxicology service (MTS) between January 1994 and July 2001. Cases were identified using ICD-9 codes for poisoning due to agents with potential for antimuscarinic toxicity using medical records and a registry of cases kept by the MTS. The study sites were two urban tertiary care referral centers located in close proximity with comparable patient populations. Inclusion criteria for cases consisted of 1) a history of acute poisoning or overdose with a primary agent known to possess significant antimuscarinic activity (TCA, phenothiazines, antihistamines, antispasmodic agents, antiparkinsonian agents, belladonna alkaloids, and Datura), 2) antimuscarinic drug presence with qualitative urine toxicology screening (confirmed by gas chromatography/
mass spectroscopy), and 3) complete medical records. Histories were verified as “reliable” in the chart either when confirmed by a knowledgeable family member or friend, or as related by the patient after resolution of toxicity. Patients were excluded if their medical records were incomplete or if their primary poisoning consisted of any other type of ingestion based on history, physical examination, or laboratory data. Patients with unreliable or uncertain histories of poisoning as documented in the medical record were also excluded, as were those with blood or breath ethanol determinations greater than 100 mg/dL. Cases in which benzodiazepines or other sedative hypnotic agents were found by history or urine screening were included except in the sub-analysis of mental status. These patients were included in the analysis of confusion. Patients with histories or urine screens with opioids present were included because 2 mg or more of naloxone was administered in every case where mental status was not normal before examination by the MTS (patients with symptoms or complaints of opioid withdrawal after naloxone were excluded). Complete medical records were defined as those consisting of identifiable physical examinations, history, laboratory values, and progress notes during toxicity and after the resolution of symptoms. Patients with positive urine toxicology screens for sympathomimetic compounds were included if the antimuscarinic agent was noted in the medical record to be the primary poison, and there was no indication in the chart that an overdose or recent (less than 24 h before arrival) use of a sympathomimetic agent had occurred. Background demographic data obtained from medical records included date and time of initial presentation, age, gender, primary intoxicant, co-ingestants (if any), and time until the patient was clinically cleared from the poisoning as determined by the MTS. Clinical information extracted included vital signs (pulse, blood pressure, temperature, and respiratory rate), pupillary size and reactivity, presence or absence of oral secretions, presence or absence of axillary perspiration, urine output on initial urinary bladder catheterization (if performed), quality of bowel sounds, mental status (Glascow Coma Score [GCS] and the presence of confusion, hallucinations, agitation, drowsiness, dysarthria, convulsions, and coma as documented in the medical record), and the occurrence of hypotension at any point during the patient’s hospital course. All findings recorded in the medical record were compared among the various participants in each patient’s medical care to check the consistency of each finding. When obvious differences in a recorded finding were noted in the record, the most commonly reported finding was included in the analysis. If no clear majority of opinion was present in the record about a sign or symptom, the physical examination of the most senior physician examiner was used for analysis.
Antimuscarinic Poisonings
For the purposes of analysis, “tachycardia” was defined as heart rate greater than 100 beats per minute (bpm). “Urinary retention” was defined as presence of greater than 200 mL of urine upon initial bladder catheterization. “Mydriasis” was defined as pupillary size greater than 4 mm in room lighting. “Hypotension” was defined as systolic blood pressure less than 100 mm Hg. “Agitation,” “drowsiness” and “dysarthria” were noted as subjectively recorded by the examiner in the medical record. “Confusion” was assessed for the entire group and defined as disorientation to person, place or date, combined with agitation; patients with drowsiness or coma were not included in this category. The presence or absence of oral secretions and axillary perspiration were subjective, and taken from documentation in the medical records as routinely recorded in notes by the MTS. All the above definitions were applied to and recorded from only the notes of the initial presentation findings on physical examination in the ED. Laboratory data collected consisted of EKG abnormalities, including QRS duration and corrected QT duration, and comprehensive urine toxicology screening. Therapeutic interventions recorded included the use of sodium bicarbonate, intubation, and physostigmine when applicable. Subjective time to resolution of all symptoms as recorded in the nursing and physician records was also determined. Statistical analysis was applied using Analysis of Variance (ANOVA). A probability value of ⬍ 0.05 was accepted as statistically significant. The study was reviewed and approved by the investigational review boards of both institutions.
RESULTS We identified a total of 345 cases of poisoning involving potentially antimuscarinic agents at two medical centers during the study period. Of these, 132 were excluded based on incomplete medical records, another primary poisoning, inability to confirm the poisoning agent through laboratory testing, or ethanol level greater than 100 mg/dL. No cases of opioid withdrawal required exclusion. A total of 213 cases were identified that met all criteria for study. Patient demographic information and admission vital signs are presented in Table 1. The vast majority of cases involved four main types of potentially antimuscarinic compounds: tricyclic antidepressants (TCAs), antihistamines, antiparkinsonian agents, and antipsychotics (Table 2). Antihistamines included all sedating medicinal antihistamine products, such as diphenhydramine, hydroxyzine, and chlorpheniramine. Other common agents encountered in our group included carbamazepine (n ⫽
91 Table 1. Demographics and Vital Signs of the 213 Patients Included in this Series Age (years) Gender (male:female) Vital signs Pulse (beats/min) SBP (mm Hg) DBP (mm Hg) RR (per minute) Temperature ⬎ 38°C (100.4°F), patients (%)
36 ⫾ 11.8 134:79 111 ⫾ 25.3 129 ⫾ 22.7 77 ⫾ 16.1 19 ⫾ 5.1 10 (4.7)
Averages are listed with standard deviations. SBP ⫽ systolic blood pressure; DBP ⫽ diastolic blood pressure; RR ⫽ respiratory rate.
10), belladonna alkaloids including scopolamine and atropine (n ⫽ 5), Datura sp. plant ingestion (n ⫽ 5), and multiple-drug antimuscarinic ingestions (n ⫽ 7). The average age, respiratory rate, and temperature measurements were comparable and statistically insignificant among the four main categories of agents when compared to the entire group. The mean pulse and systolic blood pressure (119 ⫾ 27.2 beats per minute [bpm] and 137 ⫾ 17.2 mm Hg, respectively) of the antihistamine group were statistically higher than the total means (111
Table 2. Classes and Types of Agents Included in Analysis Class of drug
Number of patients
Antihistamines Diphenhydramine Hydroxyzine Chlorpheniramine Doxylamine Tricyclic antidepressants Amitriptyline Doxepin Imipramine Nortriptyline Desipramine Unknown TCA Antipsychotics Thioridizine Chlorpromazine Perphenazine Olanzapine Thiothixene Mesoridazine Prochlorperazine Loxapine Unknown phenothiazine Antiparkinsonian agents Benztropine Trihexyphenidyl Other Carbamazepine Multiple antimuscarinics Datura spp. plants Atropine Scopolamine Cyclobenzaprine
66 60 3 2 1 73 38 11 10 7 6 1 25 8 6 3 2 2 1 1 1 1 20 16 4 29 10 7 5 4 1 2
92
R. J. Patel et al.
Table 3. Physical Examination Findings Related to Type of Exposure Antihistamines (n ⫽ 66)
TCA (n ⫽ 73)
Benztropine/ Antipsychotics trihexyphenidyl (n ⫽ 25) (n ⫽ 20)
Other (n ⫽ 29)
Totals (n ⫽ 213)
Pupil size 3.9 ⫾ 1.2 mm 3.6 ⫾ 1.2 mm 3.2 ⫾ 1.2 mm 4.4 ⫾ 1.2 mm 4.0 ⫾ 1.0 mm 3.8 ⫾ 1.3 mm Decreased secretions (%) 45 (68) 59 (87) 18 (72) 16 (80) 22 (76) 160 (75.1) Urine output (n ⫽ 83) 659 ⫾ 586 mL 573 ⫾ 527 mL 1055 ⫾ 720 mL 968 ⫾ 478 mL 589 ⫾ 566 mL 700.3 ⫾ 584.3 mL Bowel sounds Hypoactive bowel sounds (%) 23 (34.8) 32 (43.8) 15 (60) 10 (50) 15 (51.7) 95 (44.6) Absent bowel sounds (%) 3 (4.5) 5 (6.8) 1 (4) 1 (5) 3 (10.3) 13 (6.1) Hypotension, SBP ⬍ 100 2 (3.0) 9 (12.3) 4 (16) 2 (10) 0 17 (7.9) mm Hg Mental status (n ⫽ 168) (n ⫽ 54) (n ⫽ 59) (n ⫽ 19) (n ⫽ 15) (n ⫽21) Average GCS 11.9 ⫾ 3.3 10.5 ⫾ 4.3 12.6 ⫾ 3.3 11.0 ⫾ 3.3 11.4 ⫾ 3.4 11.7 ⫾ 3.8 Confusion (%)* 33 (50) 28 (38.4) 14 (56) 15 (75) 15 (51.7) 105 (49.3) Hallucinations (%) 15 (27.8) 11 (18.6) 8 (42.1) 6 (40) 4 (19) 44 (26.2) Agitation (%) 15 (27.8) 17 (28.8) 7 (36.8) 4 (26.7) 6 (28.6) 49 (29.2) Drowsy (%) 26 (48.1) 25 (42.4) 12 (63.2) 7 (46.7) 11 (52.4) 81 (48.2) Dysarthria (%) 16 (29.6) 15 (25.4) 5 (26.3) 8 (53.3) 12 (57.1) 56 (33.3) Seizure (%) 10 (18.5) 3 (5.1) 0 0 2 (9.5) 15 (8.9) Coma (%) 5 (9.3) 21 (38.9) 0 2 (13.3) 2 (9.5) 30 (17.9) Time to clear (hours) 24.6 ⫾ 15.7 40.1 ⫾ 28.8 26.6 ⫾ 18.4 45.6 ⫾ 41.6 31.8 ⫾ 24.1 32.6 ⫾ 26.0 Physostigmine use 16/1.14 5/0.9 6/1.5 9/0.94 6/1.25 50/1.1 (cases/average in mg) Sodium bicarbonate 1 (1.5) 14 (19.2) 0 1 (5.0) 2 (6.9) 18 (8.4) administered Intubated (%) 9 (13.6) 38 (52.1) 3 (12) 5 (25) 5 (17.2) 60 (27.9) QRS ⬎ 120ms (%) 2 (3.0) 17 (23.3) 2 (8) 1 (5) 0 22 (10.2) Data represent number of patients with individual findings present on initial examination, except where stated. The subset analyzed for altered mental status consisted of 168 cases, excluding those with history or laboratory evidence of sedative hypnotic overdose. TCA ⫽ tricyclic antidepressants; GCS ⫽ Glasgow Coma Scale. * “Confusion” was analyzed for the entire group of 213 patients.
⫾ 29.3 bpm and 129 ⫾ 22.7 mm Hg, respectively, p ⬍ 0.05 and p ⬍ 0.01, respectively). No other statistically significant differences existed in comparisons of vital signs among groups or compared to the total means. Physical examination findings and laboratory measurements are listed in Table 3. Mean pupillary size was 3.8 ⫾ 1.3 mm. A total of 160 cases (75%) had documented decreased skin or mucous membrane moisture in the medical record. The mean urine output on initial bladder catheterization was 700 ⫾ 584cc from 83 cases requiring catheter placement. Hypoactive or absent bowel sounds were noted in 44.6% of cases, whereas the absence of bowel sounds was recorded in 6.1% of all cases. Sinus tachycardia was documented in 145 cases (68%). Electocardiographic widening of the QRS complex (as defined by duration greater than 120 ms) was present in 17 of the 22 patients in the TCA subgroup. Ventricular tachycardia was documented in only one case (TCA poisoning), and there was one death in another patient ingesting TCAs. Hypotension was present in 7.9% of the study patients. Analysis of the central nervous system effects of antimuscarinic poisoning yielded a mean Glasgow Coma Scale (GCS) score of 11.7 ⫾ 3.8. A total of 45 cases
were excluded in this sub-analysis due to urine screening positive for benzodiazepines or other sedatives, or a history of concomitant overdose of these agents, leaving 168 cases in this subgroup. Drowsiness was observed in 48.2% of cases, confusion in 49.3%, dysarthria in 33.3%, agitation in 29.2%, and hallucinations in 26.2%. The majority of hallucinations were recorded as being visual and often accompanied by “picking” behavior. Convulsions were noted in 8.9% of all subgroup patients, with antihistamines accounting for 10 of the 16 cases. Coma was reported in 30 patients (17.9%) in the subgroup, 21 diagnosed with TCA ingestions. Intubation secondary to respiratory depression or airway protection was necessary in 38 (27.9%) patients in the TCA subgroup, accounting for 63% of all intubations in the study (60 total). Physostigmine was administered to 23.3% (49) of all patients with a mean dose of 1.1 mg per case administered (0.5–2.0 mg). Two of these cases were tricyclic antidepressant poisonings without QRS duration abnormalities. All patients were reported to have improvements in mental status, and no significant adverse reactions were noted with the use of this agent. Sodium bicarbonate was administered to 8.4% of all study patients, with TCA poisonings consisting of 14 of the total
Antimuscarinic Poisonings
93
Table 4. Combined Frequencies of Physical Examination Findings Among 213 Patients in this Series Physical examination finding
Number of cases (%)
Tachycardia Mydriasis ⬎ 4 mm Mydriasis ⬎ 5 mm Confusion Decreased secretions Tachycardia and mydriasis ⬎ 4 mm Tachycardia and decreased secretions Decreased secretions and confusion Tachycardia and confusion Confusion and mydriasis ⬎ 4 mm Confusion and mydriasis ⬎ 5 mm Decreased secretions and mydriasis ⬎ 4 mm Tachycardia and decreased secretions and mydriasis ⬎ 4 mm Tachycardia and confusion and mydriasis ⬎ 4 mm Tachycardia and decreased secretions and confusion Tachycardia or confusion or mydriasis ⬎ 4 mm Tachycardia or confusion or mydriasis ⬎ 5 mm Tachycardia or decreased secretions or confusion Tachycardia or decreased secretions or mydriasis ⬎ 4 mm
145 (68.1) 114 (53.5) 61 (28.6) 105 (49.3) 160 (75.1) 73 (35.3) 118 (55.4) 88 (41.3) 70 (32.9) 64 (30.0) 40 (18.8) 88 (41.3) 60 (28.2) 39 (18.3) 58 (27.2) 198 (93.0) 185 (86.9) 202 (94.8) 201 (94.4)
Decreased secretions ⫽ either decreased oral saliva or axillary perspiration.
18 cases. In terms of hospital course, the mean time to “medical clearance” by the MTS from time of initial patient presentation to the ED was 32.6 h. Sixty-nine cases (32%) received a single dose of activated charcoal during the course of their admission to the hospital. No patient had any other gastric emptying procedure performed. We also analyzed combinations of physical examination findings in our cases of antimuscarinic toxicity. When three of the most commonly encountered features of toxicity (tachycardia, decreased oral or axillary secretions, and mydriasis) were evaluated together, we found that 94% (201) of cases displayed at least one of these findings. However, only 28% (60) of all cases were described as exhibiting all three of these signs of toxicity. Tachycardia, confusion, and mydriasis together were reported in 18.3% (39), whereas at least one of these signs was present in 93% (198) of cases. The frequency with which other combinations were encountered in our study is listed in Table 4. DISCUSSION Antimuscarinic compounds include a diverse group of agents, but are predominated by medications prescribed
for their psychoactive potential. Some, such as diphenhydramine and scopolamine, may be misused or abused specifically for their ability to cause hallucinations and alterations in mental status. Antimuscarinic central nervous system effects can present in a variety of forms, but are often manifested in our practice as “passive psychosis:” patients are confused and disoriented, but lack the violent and paranoid activity of sympathomimetic poisoned individuals. Another frequent finding in our patients with antimuscarinic poisoning is “picking” activity. Affected individuals tend to pick or play curiously with intravenous tubing, restraints, bedclothes, and other unfamiliar items within reach, but seldom resort to more aggressive behavior such as pulling out intravenous lines. Our retrospective evaluation identified the frequency of occurrence of some of the most commonly described physical examination findings listed in charts of patients with antimuscarinic poisoning. Our most frequent finding was dry mucous membranes, but sinus tachycardia and anhydrosis were also repeatedly encountered. Over 90% of our cases exhibited at least one of the most commonly described physical examination signs of this syndrome. However, we were somewhat surprised at the sporadic recording of many of the signs and symptoms most commonly described in previous articles and textbooks. It may be that some agents, such as tricyclic antidepressants and phenothiazine derivatives, affect so many different receptor sites in the peripheral and central nervous system that predicting the clinical presentation of a particular individual is impossible. In addition, the dose or affinity for the receptor of the antimuscarinic drug ingested may influence the degree of blockade and therefore the observed signs and symptoms. One of the limitations of our study was that milder cases of antimuscarinic poisoning could have been missed or misdiagnosed by the examining physician. This could have subjected our results to spectrum bias. Our study has other limitations, most inherent with retrospective chart reviews. The subjectivity of observations such as pupillary size, presence of secretions, and level of confusion are obviously a factor in the analysis of data. The findings in our review were checked for consistency among the different health care providers with written notes in the medical record, but are still subject to interpretation. The use of specific definitions in our methods also helped to standardize certain assessments. In addition, we included some patients in whom specific characteristics were not mentioned in the medical records. In these cases, the specific finding was assumed to be negative. This may have biased our results, however, it is the custom with our MTS fellows and attendings who examined most of these patients to list with great detail all positive findings.
94
R. J. Patel et al.
There is also a possibility that the presence of multiple drugs in some cases could have affected physical examination findings. The most obvious examples would be the effect of sedative-hypnotic agents on mental status or sympathomimetics on vital signs. For these reasons, we attempted to exclude agents such as ethanol and sedatives when evaluating levels of consciousness, and sympathomimetic agents where a possibility of recent use or overuse had occurred. However, these agents may still have contributed to some of the symptomatology in our cases. Finally, we were unable to follow the “evolution” of antimuscarinic signs and symptoms in our cases. It is possible that some patients may have developed changes in findings throughout the hospitalization that could have affected results, even though most antimuscarinic signs will present early in the course of the typical poisoning with these agents. Our review was not designed to look for incidental descriptions of the above positive findings if recorded in the record after patients left the ED. Therefore, we are not able to say with certainty that delayed occurrence or evolution of signs and symptoms (besides resolution) did not occur. It is also possible that activated charcoal administration could have affected the occurrence of antimuscarinic toxicity in some cases.
CONCLUSION We performed a retrospective review of all consecutive cases of poisoning involving muscarinic receptor antagonists seen by our medical toxicology service for a 61⁄2-year period to observe the incidence of physical examination findings. We found a high incidence of examination findings consistent with this syndrome, but failed to find any sign, symptom, or combination of findings that could predict toxicity in all subjects. Sinus tachycardia, dry skin and mucous membranes, and the presence of confusion were the signs most commonly
found in our series. The observed frequency of our findings demonstrates that certain examination characteristics may assist in the assessment of the patient poisoned with antimuscarinic agents, but the variety of presentations and lack of consistency in combined findings, especially when multiple drugs are involved, may limit the objective predictive value of the physical examination in these cases.
REFERENCES 1. Thorstrand C. Clinical features in poisonings by tricyclic antidepressants with special reference to the ECG. Acta Med Scand 1976;199:337– 44. 2. Biggs JT, Spiker DG, Petit JM, Ziegler VE. Tricyclic antidepressant overdose, incidence of symptoms. JAMA 1977;238:135– 8. 3. Noble J, Matthew H. Acute poisoning by tricyclic antidepressants: clinical features and management in 100 patients. Clin Toxicol 1969;2:403–12. 4. Thorstrand C. Cardiovascular effects of poisoning with tricyclic antidepressants. Acta Med Scand 1974;195:505–14. 5. Williams RB, Sherter C. Cardiac complications of tricyclic antidepressant therapy. Ann Intern Med 1971;74:395– 8. 6. Centers for Disease Control. Scopolamine poisoning among heroin users—New York City, Newark, Philadelphia, and Baltimore, 1995 and 1996. MMWR Morb Mortal Wkly Rep 1996;45:457– 60. 7. Coremans P, Lambrecht G, Schepens P, Vanwelden J, Verhaegen H. Anticholinergic intoxication with commercially available thorn apple tea. J Toxicol Clin Toxicol 1994;32:589 –92. 8. Beaver KM, Gavin TJ. Treatment of acute anticholinergic poisoning with physostigmine. Am J Emerg Med 1998;16:505–7. 9. American Hospital Formulary Service Drug Information. 1999; 1208.8.1065. 10. Brown JH, Taylor P. Muscarinic receptor agonsists and antagonists. In: Hardman JG, Limbird LE, eds. Goodman & Gilman’s: the pharmacologic basis of therapeutics, 9th edn. New York: Macmillan Publishing Co.; 1996:155–74. 11. Clark RF, Williams SR, Nordt SP, Pearigen PD, Deutsch R. Resource-use analysis of a medical toxicology consultation service. Ann Emerg Med 1998;31:705–9. 12. Han L, McCusker J, Cole M, Abrahamowicz M, Primeau F, Elie M. Use of medications with anticholinergic effect predicts clinical severity of delirium symptoms in older medical inpatients. Arch Intern Med 2001;161:1099 –105. 13. Benowitz NL. Tricyclic antidepressants. In: Olson KR, ed. Poisoning & drug overdose, 3rd edn. Stamford, CT: Appleton & Lange; 1999:310 –2.