Article Text
PDF
Abstract
The Guidelines in Emergency Medicine Network (GEMNet) has been created to promote best medical practice in a range of conditions presenting to emergency departments (EDs) in the UK.
This guideline presents a summary of the best available evidence to guide the management of adult patients who present to the ED following an overdose of tricyclic antidepressant agents (TCA).
The document has been developed following discussion among emergency physicians to decide which topics would benefit from the development of clinical guidelines.
The document is intended as a guideline for use in the ED by emergency physicians and is based on the review of the best existing evidence for the diagnostic tools and treatments used in this setting.
The document is summarised as a clinical decision support guideline that has been presented as an easy to follow algorithm.
The intention is for each guideline to be updated and reviewed as further evidence becomes available. The formal revision date has been set at 5 years from publication, although the guideline is subject to continuous informal review.
- Poisoning
- tricyclic antidepressants
- overdose
- guideline
- clinical management
- mental health
- prehospital care
- toxicology
- management, emergency department management
- mental health, overdose
- prehospital care, clinical management
Statistics from Altmetric.com
- Poisoning
- tricyclic antidepressants
- overdose
- guideline
- clinical management
- mental health
- prehospital care
- toxicology
- management, emergency department management
- mental health, overdose
- prehospital care, clinical management
Introduction
Responsibility for development
This document has been developed in response to a perceived need to improve clinical effectiveness for care in this field. The emergency department (ED) at the Manchester Royal Infirmary has been undertaking primary and secondary research for a number of years to achieve this aim. The intention is to distil this information into practical advice for clinicians working in the department. The information is presented in the form of clinical decision support guidelines, available on the ‘shop floor’ in the form of a Clinical Decision Support Manual and on individual A4 sized forms.
Departmental consultants have considered clinical conditions that may benefit from evidence-based guidelines and, following discussion with other clinical staff, have compiled a list of topics that included tricyclic antidepressant (TCA) overdose.
The Guideline Working Group
A Guideline Working Group met to discuss this condition and decide on the clinical questions, consider the evidence available and develop the recommendations. The group process ensured that the Working Group had access to the relevant information and the required resources in order to develop in a constructive manner.
The guideline has been developed in accordance with the principles described by the National Institute for Health and Clinical Excellence guideline development methods.1
Topic introduction
TCAs are prescribed in the UK for problems including depression, anxiety and chronic pain. Recent recommendations have meant that prescribing practices are changing and the availability of TCA is declining.2 Despite this, TCA overdose still accounts for up to 18% of all poisoning deaths in the UK.3 The toxicity of the TCA coupled with the high-risk patient group who have access to TCAs means that self-poisoning episodes are more likely to be fatal.4 In the UK in 2005 there were 272 deaths related to TCA overdose.5
Patients presenting to the ED with significant overdose pose difficult management issues. TCAs block α-adrenergic receptors and have anticholinergic effects. This may lead to cardiovascular effects including sinus tachycardia, cardiac conduction abnormalities, vasodilation, arrhythmias, hypotension and asystole.6–11 The anticholinergic effects of TCAs may also lead to dry mouth, blurred vision, dilated pupils, hyperthermia and delayed gastric emptying.12 13 Intestinal obstruction and perforation have been reported,14 15 as has pancreatitis.16 Finally, TCAs exert a number of effects on the central nervous system which may lead to drowsiness, coma, respiratory depression, seizures and delirium.17–20 Opthalmoplegia has also been reported.21 22 Many patients require intensive care support or hospital admission.19 20
To date, the Toxbase database provided by the National Poisons Information Service has been the initial portal for treatment advice in TCA overdose.23 This guideline does not aim to replace previous advice but to present a complementary structured guideline and evidence-based flowchart to aid the decision-making process for these patients within the ED. The document is presented as a series of clinical questions which have been answered using the previously described Best BETs methodology.24
The aim of the guideline is to summarise the evidence supporting the various therapeutic options that have been advocated in the management of TCA overdose within the ED. It is hoped that this will help to optimise and standardise the standard of care that may be delivered to this patient group.
Scope of the guideline
This guideline encompasses adult patients (>16 years of age) presenting to the ED with suspected lone TCA overdose. The key aspects this is designed to include are initial assessment, decontamination, active management and disposition of the patient from the ED. The initial assessment and management recommendations can be followed using resources available in any UK ED. Disposition may vary depending on local resources, but the guideline may be adapted as appropriate.
This document does not provide guidance regarding patients aged <16 years, patients with multiple drug overdose and those patients who present in cardiac arrest. The use of experimental or limited availability treatments such as extracorporeal mandatory oxygenation (ECMO) is also excluded because of limited availability throughout the country.
Methodology
This guideline was developed using a novel methodology that has recently been used in cardiothoracic surgery.25 Many guidelines perform a single systematic review of the literature in order to answer all of the relevant clinical questions. To maximise sensitivity, we performed a separate short-cut systematic review of the literature for each clinical question identified.
Guideline development was structured into several stages. Initially the two lead guideline developers (TB and RB) met to discuss the scope of the guideline and to identify all clinical questions that may have been relevant. In order to answer the clinical questions identified, we performed a series of structured short-cut systematic reviews (Best Evidence Topic Summaries, BETs), the principles of which have been previously described.26 Where relevant BETs had already been created, the search strategies were checked and updated when necessary. Literature searching was standardised for each short-cut systematic review by using custom-designed filters for each element of the search (Appendix 2). Relevant papers are summarised in tabulated format in Appendix 1.
Having gathered and collated the evidence for each clinical question, the principal guideline developers met to create a series of guideline recommendations which were used to create an evidence-based flowchart (Figure 1). Following consultation with the senior author (KMJ), modifications were made before the final guideline was agreed upon.
Evidence-based flowchart. RSI, rapid sequence induction; TCA, tricyclic antidepressant.
Levels of evidence and grading of recommendations
Studies included in this guideline were graded for level of evidence according to previously accepted definitions.27 In summary, level 1 evidence comes from well-designed randomised controlled trials (RCTs), level 2 evidence from large cohort studies or poorly designed RCTs, level 3 evidence from small cohort studies or case–control studies and level 4 evidence from experimental studies, case series or case studies. The suffix ‘a’ implies that evidence at this level is from systematic review or meta-analysis, whereas the suffix ‘b’ implies that the evidence is from original research.
The recommendations that have been made were graded according to the level of evidence upon which they were based:
Grade A: Based upon multiple level 1a or 1b papers.
Grade B: Based upon individual level 1a or 1b papers or multiple level 2a or 2b papers.
Grade C: Based upon individual level 2a or 2b papers or multiple level 3a or 3b papers.
Grade D: Based upon individual level 3a or 3b papers or level 4 papers.
Grade E: Based on consensus guidelines or studies of expert opinion.
Definition of TCA overdose
For the purposes of this guideline, TCA overdose is defined as suspected deliberate or accidental ingestion of TCA at above the recommended therapeutic dose.
Summary of recommendations
Airway protection
Patients with Glasgow Coma Score (GCS) ≤8 should undergo rapid sequence induction at the earliest opportunity (Grade C).
Some patients with GCS >8 may also need intubation, particularly in the presence of airway compromise, hypoventilation or refractory seizures (Grade C).
Benzodiazepines may be considered to control agitation following TCA overdose (Grade E).
Gastric decontamination
Activated charcoal may be considered for use within 1 h of TCA ingestion, but only in patients with an intact or secured airway. The potential risk of aspiration should be strongly considered before use (Grade D).
Multiple dose activated charcoal should not be considered (Grade D).
Gastric lavage may be considered for potentially life-threatening TCA overdoses only when it can be delivered within 1 h of ingestion and the airway is protected (Grade D).
Initial assessment
An ECG should be recorded at presentation to the ED following TCA overdose (Grade B).
The ECG should be used to risk stratify patients with TCA overdose and to guide subsequent therapy (Grade B).
Serial ECG recordings should be examined for the presence of QRS prolongation (>100 ms), QTc prolongation (>430 ms) and R/S ratio >0.7 in lead aVR. These changes identify patients at high risk of developing complications following TCA overdose (Grade B).
Blood pH for risk stratification
Blood gas analysis is an important part of the initial assessment and monitoring of patients who have taken a TCA overdose (Grade E).
Venous sampling for blood gas analysis is an acceptable alternative to arterial sampling unless hypoxia or hypoventilation are suspected (Grade D).
Treatment of haemodynamic instability
A bolus of intravenous fluids should be considered as first-line therapy to treat hypotension induced by TCA overdose (Grade D).
Sodium bicarbonate is indicated for the treatment of dysrhythmias or hypotension associated with TCA overdose (Grade C).
Sodium bicarbonate may be considered for the treatment of QRS prolongation (>100 ms) associated with TCA overdose (Grade E).
The treatment of dysrhythmias or hypotension should include alkalinisation to a serum pH of 7.45–7.55 (Grade E).
Vasopressors should be used for hypotension following TCA overdose that has not responded to initial treatment (including sodium bicarbonate and intravenous fluids) (Grade D).
Epinephrine may be superior to norepinephrine for treating refractory hypotension and preventing arrhythmias (Grade D).
It is not unreasonable to administer 10 mg intravenous glucagon to treat life-threatening hypotension or arrhythmias refractory to other measures (Grade D).
Magnesium sulphate may be considered for the treatment of TCA-induced dysrhythmias when other treatments have been unsuccessful (Grade D).
Lipid emulsion may be considered for treatment of life-threatening toxicity following TCA overdose that is refractory to other measures (Grade D).
Management of seizures
Phenytoin should be avoided in patients with TCA overdose (Grade D).
Benzodiazepines should be used to control seizures following TCA overdose (Grade E).
Observation of asymptomatic patients
Following TCA overdose, asymptomatic stable patients with no significant ECG abnormalities 6 h after ingestion may be safely discharged (Grade B).
Evidence for recommendations
Below are summaries of the short-cut systematic reviews used to establish the recommendations for this guideline. The three-part question and search details are presented with comments and clinical bottom line. The search strategies are summarised and can be found in full in the appendix.
Airway protection
Assessing the need for intubation following TCA overdose in patients with reduced level of consciousness.
Can sedation be safely used in agitated patients with TCA overdose?
Assessing the need for intubation following TCA overdose in patients with reduced level of consciousness
Three-part question
In (adult patients who present to the ED following a psychotropic drug overdose with a reduced level of consciousness) does (endotracheal intubation vs standard treatment alone) lead to (fewer respiratory complications, reduced mortality and reduced length of hospital stay)?
Search strategy
Ovid Medline 1950–2008 May week 2
Ovid Embase 1980–2008 week 21
(Overdose filter) AND (Intubation filter) AND (Unconsciousness filter) limit to humans and English language.
Search outcome
Sixty-two papers were identified in Medline and 159 in Embase. Six were relevant to the three-part question (table AI).
Comments
In total we identified five retrospective analyses of patients who had been admitted following psychotropic drug overdoses and one prospective diagnostic cohort study that investigated the association between Matthew-Lawson coma grade and serious complications following TCA overdose. Although the studies have significant weaknesses, a strong correlation has consistently been shown between level of consciousness and the development of serious complications including death, hypoventilation and aspiration pneumonia following drug overdose.
Of interest, both Hulten et al28 and Emerman et al29 showed that TCA drug levels are of little use for predicting complications, especially when coma grade and QRS width were taken into account. Furthermore, it seems that level of consciousness is a stronger independent predictor of complications than QRS width. The evidence strongly suggests that patients with GCS ≤8 should undergo intubation at an early stage in the ED. Results from the retrospective study by Liisanantti et al30 suggest that intubation at the earliest possible opportunity may reduce complication rates. Furthermore, in the study by Emerman et al, GCS ≤8 was only 86.5% sensitive for prediction of hypoventilation or loss of protective airway reflexes.29 Thus, intubation may still be necessary for some patients with GCS >8 from a pragmatic patient safety viewpoint.
Clinical bottom line
Patients who present to the ED following psychotropic drug overdose with GCS ≤8 should undergo intubation at the earliest opportunity. Some patients with GCS >8 may also need intubation.
Recommendations
Patients with GCS ≤8 should undergo rapid sequence induction at the earliest opportunity (Grade C).
Some patients with GCS >8 may also need intubation, particularly in the presence of airway compromise, hypoventilation or refractory seizures (Grade C).
Can sedation be safely used in agitated patients with TCA overdose?
Three-part question
In (agitated adult patients who present to the ED after an overdose of TCA drugs) does (the use of sedative agents) lead to (an acceptably low rate of pulmonary aspiration)?
Search strategy
Ovid MEDLINE 1950–2008 June week 1
Ovid EMBASE 1980–2008 week 24
(TCA filter) AND (Overdose filter) AND ((Benzodiazepine filter) OR ((Sedation filter) AND (Aspiration filter))) LIMIT to humans and English language.
Search outcome
One thousand seven hundred and eighty-seven papers were identified (194 in Medline and 1593 in Embase). None were relevant to the three-part question.
Comments
There is no evidence of harm when intravenous sedation is administered in agitated patients who have taken an overdose of TCAs. The National Poisons Information Service recommends the use of benzodiazepines to control delirium in this situation.31
Because TCAs are known to delay gastric emptying and many patients who have taken an overdose have also consumed a large amount of alcohol, it would be advisable to exercise caution when sedating these patients. When there is doubt regarding a patient's protective airway reflexes, endotracheal intubation may be necessary. However, there is no evidence to suggest that sedation should not be attempted in these patients.
Clinical bottom line
There is no evidence of harm when sedating agitated patients following TCA overdose. National Poisons Information Service guidance advocates the use of benzodiazepines to control delirium in this situation. Caution should be exercised in view of the potential risks of pulmonary aspiration.
Recommendation
Benzodiazepines may be considered to control agitation following TCA overdose (Grade E).
Gut decontamination
Activated charcoal.
Multiple dose activated charcoal.
Gastric lavage.
Activated charcoal
Three-part question
In (adults who have taken a TCA overdose) is (activated charcoal) effective at (reducing drug absorption and reducing complication rates)?
A short-cut systematic review to answer this three-part question has been documented within the literature.32 This was updated.
Search strategy
Ovid Medline 1950–2008 May week 3
Ovid Embase 1980–2008 week 22
(TCA filter) AND (Overdose filter) AND (Charcoal filter) limit to humans and English language.
Search outcome
Sixty-seven papers were found in Medline and 125 in Embase. Eight papers were relevant to the three-part question (table AII).
Comments
Experimental volunteer studies have consistently shown that administration of activated charcoal to patients who have ingested TCA within 1 h leads to a reduction in TCA absorption and bioavailability. However, it is not possible to extrapolate these results to the clinical situation of patients with TCA overdose. Larger doses of TCA may lead to delayed gastric emptying, which may alter the observed effects of activated charcoal. Further, the risk of pulmonary aspiration may be increased.
One small observational study showed that time to charcoal administration was directly correlated with estimated plasma TCA half-life.33 However, the study involved small numbers and had significant weaknesses, meaning that it is difficult to interpret the results. Three randomised controlled trials of charcoal have been reported.33–35 None of these trials was able to demonstrate a significant improvement in clinical outcome following charcoal administration. Furthermore, in one study of 51 patients, 15.7% of patients aspirated.34
As pulmonary aspiration is a significant risk in patients with TCA overdose and a well-described complication of activated charcoal administration, caution should be exercised before prescribing activated charcoal in this patient group.36–41
Clinical bottom line
There is no clinical evidence that activated charcoal is of benefit in patients with TCA overdose. Experimental data suggest that drug absorption may be reduced. Activated charcoal may be considered within 1 h of significant drug overdose, but the potential for pulmonary aspiration should be strongly considered before use.
Recommendation
Activated charcoal may be considered for use within 1 h of TCA ingestion but only in patients with an intact or secured airway. The potential risk of aspiration should be strongly considered before use (Grade D).
Multiple dose activated charcoal
Three-part question
In (TCA overdose) is (Multiple dose Activated charcoal better than single dose Activated charcoal) at (reducing toxicity and improving clinical outcome)?
Search strategy
Ovid Medline 1950—2008 May week 3
Ovid Embase 1980—2008 week 22
(TCA filter) AND (Overdose filter) AND (Charcoal filter) LIMIT to Humans and English language.
Search outcome
Sixty-seven papers were found in Medline and 125 in Embase. Six were relevant to the three-part question. One study was excluded due to insufficient quality (table AIII).
Comment(s)
Multiple dose charcoal appears to increase elimination. However, the level of evidence is poor due to the use of volunteer studies. These studies are difficult to apply to the clinical setting of the ED as the patients did not receive the overdose amount and were treated more quickly than in the clinical setting.
The effect of multiple dose charcoal on clinical outcomes and complications such as arrhythmias and hypotension have not been studied, therefore the effect of multiple dose charcoal in the clinical setting cannot truly be assessed as the measurements are not clinically relevant. Studies used in the clinical setting have small numbers of patients.
There is a need for larger studies in the clinical setting.
Clinical bottom line
There is no convincing clinical evidence that multiple dose activated charcoal reduces toxicity and improves clinical outcome.
Recommendation
Multiple dose activated charcoal should not be considered (Grade D).
Gastric lavage
Three-part question
In (TCA overdose) which (method of gastric decontamination) is better at (reducing toxicity and improving clinical outcome)?
Search strategy
Medline 1950–2008 June week 1
Embase 1980–2008 week 23
(TCA filter) AND (Overdose filter) AND (Lavage filter) LIMIT to Humans and English language.
Search outcome
Fifty-eight papers were identified in Medline and 141 in Embase. Two papers were directly relevant to the three-part question (table AIV).
Comment(s)
There seems to be no significant difference between gastric lavage and activated charcoal. Kulig et al42 showed that gastric lavage improved clinical outcomes after drug overdose (not specifically TCA overdose) when performed within 1 h compared with no treatment. The European toxicologists' consensus statement is, at least in part, based upon this.43 One small study of 13 consecutive patients who presented to the ED with evidence of antidepressant overdose and underwent gastric lavage showed that, where estimated time of ingestion was available, none of the patients received gastric lavage within 1 h of ingestion. The mean time to delivery of gastric lavage was 6 h. Furthermore, a mean of only 8.7% of the estimated dose ingested was recovered.44
Clinical bottom line
There is no clinical evidence for the benefit of gastric lavage in TCA overdose. In a clinical setting, gastric lavage is unlikely to recover a clinically significant amount of antidepressant. Its use should only be considered in the context of a potentially life-threatening overdose with a protected airway where lavage can be delivered within 1 h of ingestion. Activated charcoal is less invasive and may be a preferable alternative in conscious patients.
Recommendation
Gastric lavage may be considered for potentially life-threatening TCA overdoses only when it can be delivered within 1 h of ingestion and the airway is protected (Grade D).
Electrocardiography (ECG)
ECG versus serum drug level as a predictive tool.
ECG changes as predictors of severity of overdose.
ECG versus serum drug level as a predictive tool
Three-part question
In (TCA overdose) is the (ECG a greater predictor than serum drug level) at predicting (seizures and arrhythmias)?
Search strategy
Ovid Medline 2008 June week 1
Ovid Embase 2008 week 23
(TCA filter) AND (ECG filter) AND (Overdose filter).
Search outcome
Three hundred and eighty-eighty studies were found including one systematic review that incorporated a meta-analysis of all other relevant studies that had been identified (table AV).
Comment(s)
The meta-analysis by Bailey et al shows that QRS duration and serum drug levels are roughly equivalent for predicting complications including death, seizures and ventricular arrhythmias. The use of the ECG allows rapid and repeated measurement in the emergency setting. It may be immediately examined for multiple abnormalities, each of which may aid in the prediction of complications. While the serum drug level provides comparable predictive value, it has the disadvantages of being more invasive, taking longer to obtain results and being less widely available in the ED.
Future research may concentrate on multivariate analysis to determine which variables are independent predictors of complications, ideally with a view to deriving a clinical decision rule to guide management and disposition of patients who have taken a TCA overdose.
Clinical bottom line
The ECG is preferable to serum drug level for the prediction of complications following TCA overdose.
Recommendations
An ECG should be recorded at presentation to the ED following TCA overdose (Grade B).
The ECG should be used to risk stratify patients with TCA overdose and to guide subsequent therapy (Grade B).
ECG changes as predictors of severity of overdose
Three-part question
In (TCA overdose) which (ECG abnormalities) are (predictive of death, seizures and arrhythmias)?
Search strategy
Ovid Medline 2008 June week 1
Ovid Embase 2008 week 23
(TCA filter) AND (ECG filter) AND (Overdose filter).
Search outcome
Three hundred and eighty-eighty papers were found (143 in Medline and 245 in Embase) including one systematic review that incorporated a meta-analysis of all other relevant studies that had been identified (table AV).
Comments
The ECG has long been used to aid in the risk stratification and management of patients who have taken a TCA overdose. The meta-analysis by Bailey et al demonstrates that ECG abnormalities are fairly good predictors of serious complications including death, seizures and ventricular arrhythmias. A QRS width >0.1 s would appear to be the strongest predictor of complications. Indeed, the wider the QRS complex the greater is the apparent risk of arrhythmias, with one group reporting a 50% incidence of arrhythmias when the QRS complex is >0.16 s in duration.45 However, the results of one study also suggest that QTc >430 ms predicts ventricular arrhythmias with reasonable sensitivity (78%) but lower specificity (56%) than QRS prolongation. Furthermore, one study demonstrated that R/S ratio >0.7 in lead aVR has a high positive predictive value (positive likelihood ratio 15.7) for predicting ventricular arrhythmias.
Importantly, it is recognised that the timing of ECG recording is important and serial recordings should be considered.
Clinical bottom line
QRS width >100 ms is a good predictor of complications following TCA overdose. QTc >430 ms and R/S ratio >0.7 in lead aVR may be useful for predicting complications.
Recommendation
Serial ECG recordings should be examined for the presence of QRS prolongation (>100 ms), QTc prolongation (>430 ms) and R/S ratio >0.7 in lead aVR. These changes identify patients at high risk of developing complications following TCA overdose (Grade B).
Blood pH for risk stratification
pH versus ECG for risk stratification
Arterial or venous pH in conscious patients with TCA overdose
Arterial pH versus ECG for risk stratification
Three-part question
In (TCA overdose) is (ECG or blood PH) superior for (predicting seizures, reduced cardiovascular function and death)?
Search strategy
Ovid Medline 2008 June week 1
Ovid Embase 2008 week 23
(TCA filter) AND (ECG filter) AND (Overdose filter).
Search outcome
Three hundred and eighty-eighty studies were identified (143 in Medline and 245 in Embase), none of which were relevant to the three-part question.
Comment(s)
There is no evidence that can assist in answering this question. The use of ECG as a predictor of complications in TCA overdose has been proved, but this has never been compared with the pH. More research is required in this area.
Clinical bottom line
Local advice should be followed.
Arterial or venous blood gas estimation for monitoring and risk stratification following TCA overdose
Three-part question
In (patients who have taken an overdose of TCAs) does (measurement of arterial or venous blood gases) lead to (superior risk stratification and monitoring of blood pH)?
Search strategy
Ovid Medline 1950–2008 June week 1
Ovid Embase 1980–2008 week 23
(TCA filter) AND (Blood gas filter) limit to humans and English language.
Search outcome
A total of 81 papers were identified (65 in Medline, 18 in Embase). One paper was directly relevant to the three-part question (table AVI).
Comments
Assessment of acid-base balance is an essential part of the initial assessment and monitoring of patients who have taken a significant overdose of TCAs. An important part of the management of these patients is alkalinisation, which has been reported to result in profound alkalaemia and high mortality.46 However, arterial blood sampling is often painful. In alert patients who do not have suspected hypoventilation, venous blood gas analysis would be preferable if it could be shown to be equivalent for risk stratification and monitoring.
The only relevant paper did seek to answer this question directly and had an appropriate sample size. Although statistically significant differences were detected in all relevant parameters between arterial and venous blood gas analysis, the clinical effects of the differences in bicarbonate and pH (in particular) are questionable. Furthermore, a fairly strong linear relationship was demonstrated between arterial and venous pH measurements.
The study did not attempt to determine which sampling method enabled superior prediction of complications. However, the evidence is sufficient to recommend that venous blood gas analysis is likely to be acceptable for the initial assessment and subsequent monitoring of these patients so long as hypoxia or hypoventilation are not suspected.
Clinical bottom line
Venous blood gas analysis is an acceptable alternative to arterial blood gas analysis following TCA overdose unless hypoxia or hypoventilation are suspected.
Recommendations
Blood gas analysis is an important part of the initial assessment and monitoring of patients who have taken a TCA overdose (Grade E).
Venous sampling for blood gas analysis is an acceptable alternative to arterial sampling unless hypoxia or hypoventilation are suspected (Grade D).
Adjunctive therapies
Intravenous fluids
Sodium bicarbonate
Vasopressors
Glucagon
Magnesium sulphate
Lipid emulsion
Intravenous fluids
Three-part question
In (patients who have taken an overdose of TCAs and have developed hypotension) does (the administration of normal saline, colloid or no intravenous fluid) lead to (superior success in treating hypotension, quicker resolution of hypotension, fewer arrhythmias and quicker recovery)?
Search strategy
Ovid MEDLINE 1950–2008 June week 1
Ovid EMBASE 1980–2008 week 23
(TCA filter) AND (Hypotension filter) AND (Intravenous fluids filter) limit to English language.
Search outcome
One hundred and fifty-eight papers were identified (118 in Embase and 40 in Medline). None was relevant to the three-part question.
Comment(s)
There is no direct evidence for the use of intravenous fluids to treat hypotension in TCA overdose. However, the absence of evidence does not equate to evidence of absence.
TCA-induced hypotension is likely to result from a combination of myocardial depression and reduced systemic vascular resistance. While intravenous fluids will not counter either of these effects, they may optimise cardiac preload thus improving the chances that a sufficient cardiac output will be achieved.
It is unlikely that a cautious fluid bolus will cause harm in this situation. Where concern exists about potential volume overload, invasive haemodynamic monitoring may be prudent.
The age-old argument of colloid versus crystalloid cannot be answered even for this well-defined situation. Colloid is believed to remain in the intravascular compartment for longer than crystalloid. Of note, however, there is some evidence that sodium loading may be important in reversing TCA toxicity,47 which may lead the undecided clinician to favour saline infusion.
Clinical bottom line
There is no evidence within the literature that intravenous fluids counter TCA-induced hypotension. As there is a sound physiological rationale for their use, they may still be considered as a useful first-line treatment.
Recommendation
A bolus of intravenous fluids should be considered as first-line therapy to treat hypotension induced by TCA overdose (Grade D).
Use of sodium bicarbonate for arrhythmias and hypotension
Three-part question
In (TCA overdose) does (sodium bicarbonate) improve (arrhythmias and hypotension)?
Search strategy
Ovid Medline 1950–2008 June week 1
Ovid Embase 1980–2008 week 23
(TCA filter) AND (Bicarbonate filter) limit to humans and English language.
Search outcome
Three hundred and fifty-seven papers were found (86 in Medline, 271 in Embase). One systematic review was relevant to the three-part question.48 While this incorporated all other relevant papers, the data were not suitable for meta-analysis. Four relevant papers are therefore tabulated (table AVII). Individual case reports are discussed but not tabulated. One survey of expert opinion is discussed.
Comment(s)
The use of sodium bicarbonate to treat the complications of TCA overdose is so well established in everyday clinical practice that it is perhaps surprising to discover that its use is not based upon high-level evidence. The evidence to supports its use is of a low level including only experimental animal studies, case reports and retrospective analyses.
In addition to the tabulated papers, the meta-analysis by Blackman et al cites a total of eight case reports where bicarbonate therapy has reportedly led to beneficial effects including resolution of QRS prolongation, recovery of hypotension, successful treatment of arrhythmias and spontaneous return of circulation following cardiac arrest.48 Furthermore, they cite a case series of 10 patients with QRS prolongation following TCA overdose in whom the QRS duration normalised during periods of hypocapnoea and worsened during periods of normocapnoea.49
Given the available evidence, it would be prudent to use sodium bicarbonate to treat major toxicity following TCA overdose, including arrhythmias and refractory hypotension. Furthermore, as QRS prolongation is associated with a high risk of arrhythmias, the use of sodium bicarbonate would also be reasonable in this situation.
Most of the relevant studies provide few details regarding the target pH for successful alkalinisation therapy. However, in the largest published study the recommended regime was alkalinisation to a pH of 7.50–7.55.50 It would appear that the absence of acidosis need not preclude the use of sodium bicarbonate in this situation. The successful use of bicarbonate to treat TCA-induced arrhythmias has been reported in a patient with alkalosis.51 Notably, however, a case series of two patients reported the aggressive use of bicarbonate and hyperventilation in two patients with QRS prolongation and ventricular arrhythmias resulting in profound alkalosis (peak pH of 7.83 and 7.66, respectively) and death.46
A 2003 survey asked 58 medical directors of United States Poisons Centres to specify the clinical situations in which they would recommend the use of sodium bicarbonate. 100% recommended sodium bicarbonate to treat QRS prolongation, 62% to treat hypotension, 53% to treat seizures, 31% to treat tachycardia, 16% to treat ventricular dysrhythmias and 3% to treat acidosis. 53% would use a QRS width threshold of 100 ms to recommend bicarbonate. Finally, 62% believed that the minimum target pH for alkalinisation should be 7.45 and 66% considered 7.55 to be the maximum pH target for alkalinisation therapy.52
Current practice in many centres is to use 50–100 ml 8.4% (50 mmol) sodium bicarbonate; however, in stable patients the use of 500 ml 1.26% (75 mmol) sodium bicarbonate is safer in the event of extravasation.
Clinical bottom line
Sodium bicarbonate may be used to treat arrhythmias, hypotension and significant ECG abnormalities to a pH of 7.45–7.55 in TCA overdose even in the absence of initial acidosis.
Recommendations
Sodium bicarbonate is indicated for the treatment of dysrhythmias or hypotension associated with TCA overdose (Grade C).
Sodium bicarbonate may be considered for the treatment of QRS prolongation (>100 ms) associated with TCA overdose (Grade E).
The treatment of dysrhythmias or hypotension should include alkalinisation to a serum pH of 7.45–7.55 (Grade E).
Vasopressors
Three-part question
In (TCA overdose with refractory hypotension) does the use of (catecholamines) improve (hypotension and survival)?
Search strategy
Ovid Medline 1950–2008 June week 1
Ovid Embase 1980–2008 week 23
(TCA filter) AND (Overdose filter) AND (Vasopressor filter) limit to English language.
Search outcome
Eight hundred and ten papers were identified (699 in Embase and 111 in Medline). Five were relevant to the three-part question (table AVIII).
Comment(s)
There is no published evidence of the effectiveness of catecholamines to treat refractory hypotension following TCA overdose. Perhaps importantly, however, there were no reports of harmful or potential pro-arrhythmic effects of catecholamines in this situation. Experimental studies in animals suggest that epinephrine may be more effective than norepinephrine, with epinephrine potentially reducing some of the cardiotoxic effects of TCAs.
Clinical bottom line
There is no published evidence of benefit or harm with intravenous catecholamines following TCA overdose. They may be a useful adjunct in the treatment of refractory hypotension in this situation. Animal evidence suggests that epinephrine may be preferable to norepinephrine.
Recommendations
Vasopressors should be used for hypotension following TCA overdose that has not responded to initial treatment (including sodium bicarbonate and intravenous fluids) (Grade D).
Epinephrine may be superior to norepinephrine for treating refractory hypotension and preventing arrhythmias (Grade D).
Glucagon
Three-part question
In (overdose with TCAs) does (the addition of glucagon to standard treatments) improve (clinical outcome)?
Search strategy
Ovid Medline 1950–2008 June week 1
Ovid Embase 1980–2008 week 23
Search details
(TTCA filter) AND (Glucagon filter) limit to human and English language.
Search outcome
Eighty-four papers were identified (71 in Embase, 13 in Medline). Three papers were relevant to the three-part question (table AIX).
Comment(s)
There have been three case reports of the successful use of glucagon to treat refractory hypotension and arrhythmias and correct QRS prolongation following TCA overdose. In each of these cases the patient had received several other treatments, although the authors state that the improvement in clinical condition was temporally related to glucagon administration. If it is effective, a 10 mg intravenous bolus may be necessary to elicit clinical improvement.
No reports of failure to respond to glucagon therapy were identified in the literature, although this is most probably attributable to reporting bias. Further research is necessary.
Clinical bottom line
There is not enough evidence currently available to support the routine use of glucagon in TCA overdose. It is not unreasonable to administer 10 mg intravenous glucagon to treat life-threatening hypotension or arrhythmias refractory to other measures.
Recommendation
It is not unreasonable to administer 10 mg intravenous glucagon to treat life-threatening hypotension or arrhythmias refractory to other measures (Grade D).
Magnesium sulphate
Three-part question
In (patients who have taken an overdose of TCAs and develop dysrhythmias) does (magnesium sulphate or standard treatment) lead to (improved rates of cardioversion to sinus rhythm and haemodynamic stability)?
Search strategy
Ovid Medline 1950–2008 June week 1
Ovid Embase 1980–2008 week 23
(TCA filter) AND (Magnesium filter) AND (Dysrhythmias filter) limit to humans and English language.
Search outcome
One hundred and eleven papers were identified (102 in Embase and 9 in Medline). No relevant comparative trials were identified. Three case studies were identified and have been tabulated (table AX). An experimental animal study was identified and is discussed.
Comment(s)
There are three reports of the successful use of magnesium sulphate for dysrhythmias associated with TCA use, two of which are from Turkey and have striking similarities. The effects have not been scientifically validated.
Knudsen and Abrahamsson53 reported that magnesium sulphate was superior to lidocaine for the successful cardioversion of amitriptyline-induced ventricular tachycardia in rats. There are no reports of potential adverse effects of magnesium sulphate in this context.
Clinical bottom line
It is reasonable to consider the use of magnesium sulphate for refractory dysrhythmias causing haemodynamic instability in the context of TCA overdose.
Recommendation
Magnesium sulphate may be considered for the treatment of TCA-induced dysrhythmias when other treatments have been unsuccessful (Grade D).
Lipid emulsion
Three-part question
In (patients who have hypotension or circulatory collapse following TCA overdose) does (lipid emulsion or standard therapy alone) lead to (lower mortality and fewer complications)?
Search strategy
Ovid Medline 1950–2010 January week 1
Ovid Embase 1980–2010 week 2
(TCA filter) AND (Lipid Emulsion filter).
Search outcome
Twenty-three papers were identified including three relevant animal studies and one study in healthy volunteers (table AXI). There were no randomised controlled trials, cohort studies, case series or case reports of the use of lipid emulsion in the clinical environment.
Comment(s)
Lipid emulsion is a promising treatment for TCA overdose. TCAs are lipid soluble. Some have postulated that lipid emulsion may reduce toxicity by creating an intravascular lipid compartment into which lipid-soluble drugs may be sequestered.54 55 Alternatively, lipid emulsion may work by enhancing free fatty acid metabolism.56 The only human data come from a volunteer study which demonstrated that lipid emulsion did not significantly alter TCA blood levels.
Our search also identified three animal studies. Harvey et al showed that lipid emulsion was superior to saline infusion for treatment of TCA-induced hypotension.57 The same group has also shown that lipid emulsion is superior to sodium bicarbonate for treatment of TCA-induced hypotension in rabbits.58 Finally, Yoav et al demonstrated that infusion of lipid emulsion resulted in lower mortality than saline infusion in TCA-intoxicated rats.
The Lipid Rescue website also contains one informal case report of the successful use of lipid emulsion to treat refractory QRS prolongation in an intubated patient following TCA overdose.59 The ECG changes resolved after 4 h and the patient recovered. In itself, this represents weak evidence for the efficacy of lipid emulsion in this situation. Furthermore, the patient developed haematuria after 4–5 h and haemoglobin could not be estimated because of the lipaemic sample. Interference with laboratory assays and hypertriglyceridaemia are important side effects of lipid emulsion.
However, given the evidence presented and in the absence of stronger evidence, it would be reasonable to administer lipid emulsion to a patient with serious life-threatening cardiotoxicity secondary to TCA overdose that is refractory to other measures.
Clinical bottom line
There is no evidence that lipid emulsion is of benefit as a standard treatment for TCA overdose in humans. However, given the results of three animal studies and a plausible physiological mechanism, lipid emulsion should be considered for life-threatening cardiotoxicity that is refractory to other measures following TCA overdose.
Recommendation
Lipid emulsion may be considered for treatment of life-threatening toxicity following TCA overdose that is refractory to other measures (Grade D).
Management of seizures
Phenytoin
Benzodiazepines
Phenytoin
Three-part question
In (patients with TCA overdose who develop prolonged seizures) does (phenytoin or benzodiazepines) lead to (quicker and more reliable termination of seizures with fewer complications)?
Search strategy
Ovid MEDLINE 1966–2007 June week 1
Ovid EMBASE 1980–2007 week 24
(TCA filter) AND (Overdose filter) AND (Phenytoin filter) limit to human and English language.
Search outcome
Seven hundred and ten papers were identified (293 in Medline and 417 in Embase). None directly answered the three-part question. Several papers discussed the use of phenytoin in TCA overdose. These are discussed.
Comment(s)
Intravenous phenytoin is licensed for use in status epilepticus. Its use in the context of TCA overdose is controversial. There have been sporadic case reports of the successful use of intravenous phenytoin for the treatment of patients with severe TCA overdose who have developed cardiac conduction abnormalities.60 61 It is proposed that this may result from its class Ia antiarrhythmic action. However, there is evidence of interaction between the two drugs, with TCAs increasing phenytoin levels.62 63 Owing to the narrow therapeutic window of phenytoin, this interaction is of concern. Furthermore, in an animal model phenytoin was found to increase the likelihood of ventricular arrhythmias when TCAs were also infused.64 In light of this potential interaction, guidelines from the National Poisons Information Service state that phenytoin should be avoided in patients who have taken a TCA overdose.65
Clinical bottom line
Phenytoin has not been compared with benzodiazepines in patients with TCA overdose. Evidence for the benefit of phenytoin in TCA overdose comes only from sporadic case studies. As there are doubts regarding the safety of phenytoin in these patients, it should be avoided.
Recommendation
Phenytoin should be avoided in patients with TCA overdose (Grade D).
Benzodiazepines
Three-part question
In (adult patients who develop seizures following TCA overdose) does the use of (benzodiazepines) lead to (safe and effective termination of seizures)?
Search strategy
Ovid MEDLINE 1950–May 2008 week 2
Ovid EMBASE 1980–2008 week 21
(TCA filter) AND (Overdose filter) AND (Benzodiazepine filter) LIMIT to English language.
Search outcome
One thousand seven hundred and forty-three papers were identified (186 in Medline and 1557 in Embase). None was relevant to the three-part question.
Comment(s)
There were no studies found that were relevant to the three-part question. Notably, there have been no reports of harmful interactions when benzodiazepines are used in TCA overdose. The National Poisons Information Service recommends the use of intravenous benzodiazepines to control seizures associated with TCA overdose.66
Clinical bottom line
There is no evidence of benefit or harm when benzodiazepines are used to control seizures associated with TCA overdose. As there is no evidence of harm, the National Poisons Information Service guidance, which advocates the use of benzodiazepines in this situation, ought to be followed.
Recommendation
Benzodiazepines should be used to control seizures following TCA overdose (Grade E).
Observation of asymptomatic patients
Three-part question
In (a clinically stable patient following TCA overdose) what (period of observation) enables (safe discharge)?
Search strategy
Ovid Medline 1950–2008 June week 1
Ovid Embase 1980–2008 week 24
(TCA filter) AND (Overdose filter) AND (Observation filter) limit to Humans and English language.
Search outcome
Five hundred and ninety-two papers were identified (156 in Medline, 436 in Embase). Seven were relevant to the three-part question (table AXII).
Comment(s)
Late complications including cardiac arrhythmias have been reported to occur as long as several days after TCA overdose.67–70 However, in all of these cases there were significant signs of toxicity at a much earlier stage. There are no reports of late complications occurring in clinically stable patients who are alert, normotensive and have had no ECG abnormalities after 6 h of observation.
Clinical bottom line
Stable patients with TCA overdose who show no sign of toxicity and have had no significant ECG abnormalities (including QRS<0.10 s) for 6 h can safely be discharged.
Recommendation
Following TCA overdose, asymptomatic stable patients who have had no significant ECG abnormalities 6 h after ingestion may be safely discharged (Grade B).
Therapy notes
Indications for rapid sequence induction (RSI)
TCA overdose delays gastric emptying and may cause vomiting, increasing aspiration risk, particularly in patients with reduced level of consciousness. A low threshold for early intubation should be adopted and the need should be continually reassessed. It is imperative to ensure the availability of adequate expertise during rapid sequence induction.
Gastric decontamination
Activated charcoal may be considered for use within 1 h of TCA ingestion but only in patients with an intact or secured airway. The potential risk of aspiration should be strongly considered before use. Gastric lavage may be considered for potentially life-threatening TCA overdoses only when it can be delivered within 1 h of ingestion and the airway is protected.
Hypotension
TCA overdose causes hypotension by reducing preload and afterload as well as direct effects on the myocardium. Optimising the preload may reverse hypotension. This may be achieved by head-down tilt and bolus of intravenous fluid. Sodium bicarbonate may reverse hypotension even in the absence of acidosis and is indicated if hypotension is persistent. If hypotension still persists, vasopressors/inotropes should be used. There is some evidence that epinephrine may be preferable to norepinephrine in this situation.
Arrhythmias
Administration of sodium bicarbonate, even in the patient without acidosis, may reverse TCA-induced arrhythmias. If arrhythmias are persistent, magnesium sulphate may be given, although there is limited available evidence for its efficacy.
ECG abnormalities
QRS prolongation (>0.10 s) and right axis deviation are associated with increased risk of cardiac arrhythmias. The use of sodium bicarbonate should be strongly considered in this situation.
Sodium bicarbonate
For life-threatening toxicity use 50–100 ml 8.4% sodium bicarbonate. The dose can be repeated with blood gas monitoring to a target pH of 7.45–7.55. For more stable patients, 500 ml 1.26% sodium bicarbonate carries less risk of skin necrosis in the event of extravasation.
Refractory haemodynamic instability
Use of glucagon, magnesium sulphate and lipid emulsion may be considered. Animal studies suggest that lipid emulsion (eg, Intralipid 20% 1.5 ml/kg over 1 min) may be a particularly promising therapy for the future, although evidence in humans is lacking.
Seizures
Prolonged seizures should be treated initially with benzodiazepines. Phenytoin should be avoided because of a possible interaction with TCAs. If there is no response to benzodiazepines, RSI should be considered.
ECG monitoring
ECG monitoring is essential for all patients at moderate/high risk. Serial 12-lead ECG recording is recommended in all patients to monitor for changes in QRS duration.
Appendix 1
Relevant papers
Assessing the need for intubation in semiconscious patients presenting to the ED following psychotropic drug overdose
Single dose activated charcoal in tricyclic antidepressant (TCA) overdose
Multiple dose activated charcoal following tricyclic antidepressant (TCA) overdose
Gastric lavage following tricyclic antidepressant (TCA) overdose
ECG and serum drug levels for predicting complications
Venous versus arterial blood gas sampling
Sodium bicarbonate following tricyclic antidepressant (TCA) overdose
Intravenous catecholamines to treat hypotension following tricyclic antidepressant (TCA) overdose
Glucagon to treat haemodynamic instability after tricyclic antidepressant (TCA) overdose
Magnesium sulphate to treat dysrhythmias following TCA overdose
Lipid emulsion for treatment of TCA toxicity
Observation of asymptomatic patients following tricyclic antidepressant (TCA) overdose
Appendix 2
Search filters
TCA filter
(exp Antidepressive Agents, Tricyclic/OR tricyclic.mp. OR amitriptyline.mp. OR exp Amitriptyline/OR desipramine.mp. OR exp Desipramine/OR clomipramine.mp. OR exp Clomipramine/OR doxepin.mp. OR exp Doxepin/ OR dothiepin.mp. OR exp Dothiepin/OR imipramine.mp. OR exp Imipramine/OR lofepramine.mp. OR exp Lofepramine/OR nortriptyline.mp. OR exp Nortriptyline/OR trimipramine.mp. OR exp Trimipramine/).
Charcoal filter
(exp Charcoal/OR charcoal.mp.)
Lavage filter
(gastric lavage.mp. OR exp Gastric Lavage/OR irrigation.mp. OR exp Irrigation/OR lavage.mp. OR exp Decontamination/OR gastric decontamination.mp. OR washout.mp. OR gut decontamination.mp OR exp Stomach Emptying/OR exp Stomach Lavage/).
Overdose filter
(exp Overdose/OR exp Poisoning/OR overdose.mp. OR exp Drug Overdose/).
ECG Filter
(ECG.mp. OR exp Electrocardiography/ OR electrocardiogram.mp. OR EKG.mp.)
Vasopressor filter
(exp Catecholamines/OR exp Epinephrine/OR exp Norepinephrine/OR exp Dopamine/OR (catecholamine OR epinephrine OR norepinephrine OR dopamine OR epinephrine OR norepinephrine).mp.)
Bicarbonate filter
(exp Sodium Bicarbonate/OR exp Bicarbonates/OR (sodium bicarbonate OR bicarbonates).mp.)
Observation filter
(exp Monitoring, Physiologic/OR exp Patient Admission/OR (admission OR monitoring).mp.)
Benzodiazepine filter
(exp Benzodiazepines/OR exp Diazepam/OR exp Clonazepam/OR exp Midazolam/ OR exp Temazepam/OR exp Nitrazepam/OR (benzodiazepin$ OR diazepam OR clonazepam OR nitrazepam OR clonazepam OR midazolam OR temazepam).mp.
Phenytoin filter
(exp Phenytoin OR phenytoin.mp. OR epilim.mp.)
Seizure filter
(exp Seizure/OR (seizur$ OR convuls$ OR fitting OR fit OR fits).mp.)
Intubation filter
(exp Intubation, Intratracheal/OR (rapid sequence induction).mp OR rsi.mp OR intubation.mp OR (crash induction).mp OR airway management.mp).
Sedation filter
exp ‘Hypnotics and Sedatives’/OR sedation.mp. OR sedat$.mp. OR hypnotic$.mp.
pH filter
(exp Hydrogen Ion Concentration/OR pH.mp.)
Blood gas filter
(Exp Blood Gas Analysis/OR exp Blood Gas/ OR blood gas$.mp.)
Unconsciousness filter
(Glasgow Coma Scale.mp. OR exp Coma/ OR exp Glasgow Coma Scale/ OR exp Unconsciousness/OR (unconscious$ or semiconscious$ or obtund$ or unresponsive$).mp.)
Hypotension filter
(exp Hypotension/OR (hypotension OR hypotensive).mp.)
Intravenous fluids filter
(exp Infusion/OR exp Infusion Fluid/ OR exp Colloid/ OR exp Polygeline/OR exp Gelatin Succinate/ OR exp Sodium Chloride/OR (infusion OR colloid OR gelofusine OR haemaccel OR saline).mp.)
Magnesium filter
(exp Magnesium/OR exp Magnesium Sulfate/OR magnesium.mp.)
Dysrhythmias filter
(exp Heart Ventricle Tachycardia/ OR exp Heart Arrhythmia/ OR exp Arrhythmias, Cardiac/ OR (dysrhythmias$ OR arrhythmia$).mp.)
Glucagon filter
(exp Glucagon/OR glucagon.mp.)
Lipid emulsion filter
(intralipid.mp. OR exp Fat Emulsions/OR exp Fat Emulsions, Intravenous/)
References
Footnotes
Funding Funding for the development of this guideline was received from the College of Emergency Medicine, London, UK.
Provenance and peer review Not commissioned; not externally peer reviewed.