Abstract
To enable consistency of investigation and the establishment of best practice standards, consensus guidelines were formulated previously by the UK National Poisons Information Service and the Association for Clinical Biochemistry. These joint guidelines have now been updated to reflect current best practice. The types of laboratory investigation required for poisoned patients are categorized as either (a) essential common laboratory investigations or (b) specific toxicological assays, and also as either (i) common or (ii) specialist or infrequent. Tests in categories (a) and (bi) should be available 24 hours per day, with a maximum turnaround time of 2 h. For the specialist assays, i.e. category (bii), availability and turnaround times have been specified individually. The basis for selection of these times has been clinical utility. The adoption of these guidelines, along with the use of the National Poisons Information Service (0844 8920111) and its online poisons information resource TOXBASE® (www.toxbase.org) enable the poisoned patient to receive appropriate, ‘best practice’ investigations according to their clinical needs and will avoid unnecessary investigations.
Introduction
Acute poisoning is a common reason for presentation and admission to hospital 1 . The great majority of such patients do not require any specific treatment and recover completely without serious complications. Although the in-hospital mortality rate from acute poisoning in the United Kingdom remains low, a small minority of patients have life-threatening poisoning. Laboratory assays for toxins and/or their metabolites are a very important part of the management of patients with potentially serious poisoning. However, there is anecdotal evidence that laboratory toxicological investigations are overused by medical staff. There is also evidence that the availability of these investigations varies between hospitals, particularly when required outside normal working hours. This may present problems in management, particularly as many poisoned patients present during the night or at weekends.
To enable consistency of investigation and the establishment of best practice standards, consensus guidelines were formulated previously by the UK National Poisons Information Service (NPIS) and the Association for Clinical Biochemistry 2 . These joint guidelines have now been updated to current best practice.
The purpose of this document is to set out the laboratory assays that should be available for the management of poisoned patients and to define, where possible, the circumstances under which each should be used. The document concentrates on the measurement of specific drugs and toxins in blood and urine. These measurement assays are divided into two groups: those that should be readily available in all acute hospitals and those for which alternative arrangements should apply. Guidance is also included on the appropriate availability of other supportive investigations for patients with suspected poisoning. Distinction is made between assays required urgently, for which 24-h availability is necessary, and those that can be performed routinely, i.e. during the next normal working day.
The document is not intended to be a comprehensive review of therapeutic drug monitoring or the forensic use of blood samples. It is intended to give guidance to acute hospitals on the current view regarding appropriate standards for the availability and use of laboratory assays for poisoned patients. Up-to-date guidance on toxicology tests and their urgency, together with advice on the clinical management of patients, is available from the NPIS (0844 8920111) and via its online database TOXBASE® (www.toxbase.org). The NPIS should be contacted for advice in complicated cases.
General principles
The indications for laboratory assays can be summarized as follows:
To confirm the diagnosis of poisoning when this is in doubt. To influence patient management, e.g. the need for:
○ further investigations; ○ antidotes; ○ haemodialysis or other extracorporeal methods; ○ cessation of treatment. To plan the re-institution of chronic therapy. In the diagnosis of brain death and in assessing the suitability of potential organ donors. For medicolegal or forensic reasons.
The use of laboratory investigations out-of-hours should be restricted to those instances when an urgent result is needed to inform immediate patient management. It may also be appropriate to obtain samples to be stored for later analysis.
Reporting units
The consensus meeting held by the Association for Clinical Biochemistry 3 agreed that concentrations for drugs should be reported in mass per litre with the exception of iron, lithium, methotrexate and thyroxine. It is recommended that, with the exception of these four agents, laboratories that report in molar units should also provide the result in mass units and should be encouraged for patient safety reasons to adopt the recommended use of mass units. (For the purposes of this paper, equivalent units have been stated and these have been rounded.)
Sample collection
For most investigations relevant to poisoned patients, serum or plasma is suitable for analysis and these terms can be used interchangeably. Clinicians should check with their local laboratory for specific requirements as gel separation tubes may affect some assays. Exceptions are ethanol (plasma with fluoride/oxalate [or fluoride/EDTA] anticoagulant), carboxyhaemoglobin and methaemoglobin (heparinized whole blood), and red cell cholinesterase, lead and mercury (whole blood anticoagulated with EDTA).
Target (therapeutic) and toxic concentrations
This document contains details of concentrations of toxins in blood and urine that may be associated with toxicity. It should be recognized that target and toxic ranges for these substances are approximate. Clinicians should take this into account when making management decisions. More detailed information about the interpretation of laboratory assay results and on the clinical features and management of poisoning can be found via the NPIS online information resource TOXBASE®, which is available free of charge to registered health professionals (www.toxbase.org). Advice is also available via the NPIS telephone information service (0844 8920111).
Supportive investigations
The following laboratory investigations should be available on a 24-h basis to all hospitals where patients with acute poisoning are admitted and should be available at least as quickly as Group 1 specific assays. Samples should be taken using appropriate containers and, when necessary, anticoagulants. Information on the interpretation of some of these investigations in the poisoned patient is provided in Tables 1 and 2 in Appendix 1.
Full blood count Sodium, potassium, urea, creatinine Glucose Calcium, albumin, magnesium International normalized ratio (INR) Liver function tests (transaminases) Bilirubin Anion gap (chloride and bicarbonate) Plasma osmolality (freezing point depression method) and osmolar gap Arterial blood gases Creatine kinase
Specific assays
Specific assays have been divided into two groups. The first group should be available on a 24-h basis in all hospitals that admit patients with acute poisoning. The second group are assays that are important in patient management but which are infrequently needed. For these, arrangements need to be in place so that the assays can be obtained from specialist laboratories if they are not available on site.
Group 1: assays that should be available on a 24-h basis in all acute hospitals
Carboxyhaemoglobin Digoxin Ethanol Iron Lithium Methaemoglobin Paracetamol Paraquat (qualitative urine test) Salicylate Theophylline Valproate
Results should normally be available within a maximum of 2 h of presentation (or sooner if possible) unless otherwise stated. Their use is summarized in Table 3 in Appendix 1.
Carboxyhaemoglobin
Carboxyhaemoglobin should be measured urgently in all patients with suspected carbon monoxide poisoning (including those with suspected smoke inhalation). It may also be elevated in patients who have ingested methylene chloride. A carboxyhaemoglobin percentage of >30% total haemoglobin indicates severe poisoning. However, concentrations less than this do not exclude significant exposure, and the relationship between carboxyhaemoglobin and severity of poisoning and/or clinical outcome is poor especially when sampling occurs late after exposure. Management should be determined by the clinical condition of the patient rather than the carboxyhaemoglobin concentration. High flow oxygen therapy should be administered pending the results of carboxyhaemoglobin measurement.
Digoxin
Plasma or serum digoxin concentrations correlate poorly with the severity of poisoning, especially soon after acute overdose. The usual target range for digoxin is 1.0–2.0 µg/L (1.3–2.6 nmol/L). Severe toxicity is usually (but not invariably) associated with concentrations >4 mg/L (>5.2 nmol/L) but much higher concentrations may be observed without severe clinical features within a few hours of acute overdose. Hypokalaemia enhances digoxin toxicity. In acute life-threatening poisoning, hyperkalaemia is usually present. Urgent measurement of plasma/serum digoxin concentration is essential if digoxin-specific antibodies are to be used. The digoxin concentration is useful in determining an appropriate dose of digoxin-specific antibodies, as well as confirming the diagnosis. In patients with life-threatening arrhythmias due to digoxin toxicity, treatment with digoxin-specific antibodies should not be delayed pending the results of plasma digoxin concentrations. Samples should always be taken before antibody administration as plasma digoxin concentrations cannot be interpreted once these have been given. Samples taken to investigate possible chronic digoxin intoxication should be taken at least 6 h after dosing and do not usually need to be analysed urgently, unless life-threatening features are present and use of digoxin-specific antibodies is being contemplated. Repeat samples, analysed routinely, may help determine when to re-institute chronic therapy after acute overdose. However, these are not of value for several days following the administration of digoxin-specific antibodies, as the elimination half-life of the complex under normal conditions is 16–20 h. Assays routinely used in the United Kingdom are not ideal for accurate quantification of digitoxin or plant glycosides, although they may provide qualitative supportive evidence of exposure.
Ethanol
Plasma ethanol concentrations are usually not required in patients who have ingested ethanol, unless severe poisoning is suspected or if there is doubt about the diagnosis. Ethanol concentrations should be measured urgently:
in patients with undiagnosed coma; in patients with a widened osmolar gap; in patients with suspected severe ethanol poisoning; in children with unexplained acidosis. The use of different units for reporting plasma ethanol concentrations is confusing and potentially dangerous. The recommended units for reporting ethanol concentration are mg/L. Concentrations >1800 mg/L (>1.8 g/L, >180 mg/dL, >39 mmol/L) are associated with disorientation. In the absence of other toxins, ethanol concentrations in excess of 3500 mg/L (3.5 g/L, 350 mg/dL, 76 mmol/L) are usually required to produce coma. Fatal poisoning is usually associated with concentrations >4500 mg/L (>450 mg/dL, >98 mmol/L). Ethanol toxicity is enhanced in the presence of other sedative agents (and vice versa). Plasma ethanol concentrations performed urgently are essential for monitoring the use of ethanol as an antidote for poisoning with ethylene glycol or methanol, particularly if dialysis is also being used. Therapeutic plasma ethanol concentrations should be monitored every 1–2 h initially until a concentration of 1000–1500 mg/L (1.0–1.5 g/L; 100–150 mg/dL; 21–33 mmol/L) is reached and every 2–4 h thereafter. Ethanol concentration should be repeated 1 h after any change of administration rate. For conscious patients, breath alcohol measurement may be used for monitoring ethanol therapy, if facilities are available locally, aiming for target concentrations of 440–660 µg/L (44–66 µg/100 mL) breath. (This is approximately equivalent to 1000–1500 mg/L blood, if a blood-breath partition coefficient value of 2300 is used.)
Iron
Plasma or serum iron concentrations help to determine prognosis and the need for antidotal treatment with desferrioxamine in patients with suspected iron poisoning. These concentrations should be measured urgently in:
○ asymptomatic patients who have ingested >20 mg/kg elemental iron within 6 h; ○ patients with symptoms (including transient symptoms) suspected to be due to iron intoxication. In patients with suspected severe poisoning, a sample should be taken immediately to confirm the history. In all other patients, the sample should be taken after at least 4 h have elapsed since iron ingestion and an additional sample after a further 2 h is desirable to determine whether the concentration is still rising. After 6 h, the peak serum iron concentration is likely to have passed and interpretation is more difficult. It is important that the sample is not haemolysed. However, iron may itself cause intravascular haemolysis at high concentrations so a haemolysed sample should be an index for concern. Severe toxicity is unlikely if symptoms have not developed within 6 h of ingestion. A sample should be taken urgently although interpretation is more difficult. A repeat serum iron concentration within 2 h may confirm that it is falling. However, if the serum iron concentration is >5 mg/L (90 µmol/L) in a symptomatic patient, then consider treatment with desferrioxamine. Peak plasma iron concentrations following ingestion are interpreted as follows:
<55 µmol/L (<3.0 mg/L) mild poisoning; 55–90 µmol/L (3.0–5.0 mg/L) moderate poisoning; >90 µmol/L (>5.0 mg/L) severe poisoning. Antidotal therapy with desferrioxamine is indicated without waiting for the plasma iron concentration in patients with severe clinical features (e.g. unconscious, fitting, shocked, metabolic acidosis, gastrointestinal bleeding or haemolysis). Antidotal treatment is indicated for symptomatic patients with iron concentrations >90 µmol/L (>5 mg/L) and also for those with iron concentrations 55–90 µmol/L (3.0–5.0 mg/L), particularly if these are rising or are associated with features of moderate as well as severe poisoning. Further advice on individual cases can be obtained from the NPIS. All colorimetric iron assays are unreliable in the presence of desferrioxamine. Measurement of iron-binding capacity has no role in the management of iron poisoning.
Lithium
The lithium concentration should be measured immediately and 6 h later in patients with acute lithium intoxication associated with relevant symptoms and in patients with suspected acute-on-chronic or chronic toxicity. In acute lithium overdose where there are no relevant symptoms, the serum concentration should be measured approximately 6 h after ingestion and the assay requested urgently. If severe poisoning is confirmed, or if a sustained release preparation may have been taken, the plasma lithium concentration should be repeated 6–12 hourly until the concentration is falling. Lithium heparin tubes should not be used for the sample. The principal value of urgent lithium measurement is to determine the need for haemodialysis in severe poisoning. Low thresholds should be considered for haemodialysis in the presence of neurological or cardiac features, particularly if concentrations are increasing. Advice on the interpretation of lithium concentrations and on the appropriate use of haemodialysis can be obtained from the NPIS. There is a risk of rebound increases in lithium concentration after haemodialysis, and therefore lithium concentrations should be measured 6 h after haemodialysis is discontinued. Repeated measurements of lithium concentration, performed routinely, are helpful in timing the appropriate re-institution of chronic therapy following an episode of toxicity. The usual target range is approximately 0.4–1.0 mmol/L 12 h after dose.
Methaemoglobin
A range of poisons may cause methaemoglobinaemia. Examples are:
○ Drugs: nitrites, local anaesthetics (e.g. benzocaine, prilocaine), dapsone, sulphonamides and primaquine. ○ Other chemicals: aniline dyes, nitrobenzenes, naphthalene and chlorates. Measurement of methaemoglobin concentration is required to confirm diagnosis and assess severity. It is essential for it to be measured urgently when administration of methylthioninium chloride (methylene blue) is contemplated. Methaemoglobin concentrations of less than 20% are not usually associated with symptoms and require no treatment to decrease the methaemoglobin concentration. Concentrations >20% may require treatment with methylthioninium chloride (methylene blue). Further advice on individual cases can be obtained from the NPIS.
Paracetamol
Measurement of plasma paracetamol concentration is essential for determining the need for antidotal treatment following acute overdose and should be performed urgently in all patients with known or suspected paracetamol overdose. For acute overdoses, the concentration obtained may then be compared to a recognized treatment nomogram. The assays should have a lower limit of quantification for paracetamol of 10 mg/L (0.07 mmol/L) or less. This is particularly important when assessing patients who have taken drug overdoses more than 12 h prior to blood sampling. Plasma paracetamol concentrations should be measured urgently in all patients when there is a clinical suspicion that paracetamol poisoning may be present. Examples would be:
○ patient reporting paracetamol overdose ○ drug overdose patients, when the history appears unreliable. ○ patients with undiagnosed coma where there is a clinical suspicion of drug overdose. Measurement of paracetamol concentrations in alert patients who deny taking paracetamol, when there is no clinical suspicion, rarely provides evidence of significant paracetamol toxicity. The sample must be taken 4 h after ingestion, or immediately if the patient presents after 4 h. The INR should be measured together with electrolytes, urea, creatinine and liver function tests (transaminases and gamma glutamyl transferase) as these may indicate the presence of or predisposition to hepatic toxicity. These should be repeated in patients at risk of, or developing, hepatotoxicty. Plasma paracetamol concentration measurement is occasionally helpful in patients with unexplained hepatotoxicty, although a negative result does not exclude paracetamol as a cause if they present late.
Paraquat (urinary qualitative assay)
There is no specific treatment of proven value for paraquat poisoning. Investigations are directed at confirming exposure and determining prognosis. Although paraquat has now been withdrawn as an herbicide in the United Kingdom, it is expected that cases will continue to present for some time due to storage of previously sold products. A qualitative urine test (dithionite spot test) should be performed urgently in all patients presenting with suspected paraquat poisoning to confirm exposure. This is a simple procedure that can be performed by any laboratory, although reagents must be freshly prepared to avoid misleading results. Instructions for performing this test can be found on TOXBASE®. In patients with a positive spot test, a blood sample should be taken for routine analysis in a specialist laboratory, as this provides valuable prognostic information [see section ‘Paraquat (quantitative plasma assay)’ below]. If after 4 h a urine test remains negative, there is no indication for further investigation or treatment of paraquat poisoning.
Salicylate
There is no need to measure salicylate concentrations in conscious overdose patients who deny taking salicylate-containing preparations and who have no features suggesting salicylate toxicity. Plasma salicylate concentration should be measured urgently for patients who are thought to have ingested >125 mg/kg of aspirin (acetyl salicylate) as well as those who have taken methylsalicylate (oil of wintergreen) or salicylamide. The sample should be taken at least 2 h (symptomatic patients) or 4 h (asymptomatic patients) following ingestion, since it may take several hours for peak plasma concentrations to occur and up to 12 h for enteric coated preparations. In cases where significant toxicity is suspected, a repeat sample should be taken after a further 2 h because of the possibility of continued absorption. Under these circumstances, measurements should be repeated every 3 h until concentrations are falling. Salicylate concentration should also be measured in patients with unidentified poisoning or those with undiagnosed clinical features consistent with salicylate poisoning, e.g. coma, metabolic acidosis, respiratory alkalosis, tinnitus. The severity of poisoning cannot be assessed from plasma salicylate concentrations alone, and clinical and biochemical features should be taken into account. However, salicylate intoxication is usually associated with plasma concentrations >350 mg/L (>2.5 mmol/L). Patients with moderate salicylate poisoning may require urine alkalinization, while those with severe poisoning may need treatment with haemodialysis. Advice on the interpretation of plasma salicylate concentrations and the need for urinary alkalinization and haemodialysis can be obtained from the NPIS. Plasma salicylate concentrations should be repeated after dialysis.
Theophylline
Patients with suspected theophylline poisoning should have the severity graded according to simple clinical indicators, including the plasma potassium and arterial blood gases. Plasma theophylline concentration measurement should be performed immediately for patients with any clinical features suggesting theophylline toxicity, including convulsions, tachycardia, hypokalaemia or acidosis. Patients who have a history of theophylline poisoning but no clinical features on presentation should have their concentration determined 4 h after exposure. In patients with severe poisoning or theophylline concentration >60 mg/L (333 µmol/L), the theophylline concentration should be repeated every 2–4 h, until peak concentrations have passed. This is particularly important if a slow release preparation has been taken. Advice on the interpretation of plasma theophylline concentrations and the need for multiple-dose activated charcoal can be obtained from the NPIS and TOXBASE®.
Valproate
The therapeutic range for valproate is not well validated but is around 50–100 mg/L (350–700 µmol/L). Serum valproate concentrations correlate with clinical features of poisoning. Concentrations above 850 mg/L (5900 µmol/L) are more likely to be associated with coma, respiratory depression, hypotension or metabolic acidosis. It is generally unnecessary to measure valproate concentrations urgently. The principal value of urgent valproate measurement is to determine the need for haemodialysis in severe poisoning or when the diagnosis is in doubt in a mixed overdose. Haemodialysis may be considered in severely poisoned patients with serum valproate concentrations >850 mg/L (5900 µmol/L).
Group 2: specialist or infrequent assays
It is not necessary for the following assays to be available directly from all acute hospital laboratories. However, arrangements should be in place so that these assays can be accessed urgently when necessary. This may involve an arrangement with a supra-regional specialist toxicology laboratory or a subregional centre. It is the responsibility of each individual hospital to ensure that appropriate arrangements are in place and that staff can follow these arrangements when the need arises, including outside normal working hours. Laboratory staff should have contact details readily available for specialist laboratories providing these assays, together with information on how samples should be collected and transported. Clinical staff should discuss the use of the following assays with a local clinical biochemist; advice is also available from the NPIS when required.
Arsenic Carbamazepine Cholinesterase (plasma and erythrocyte) Cyanide Ethylene glycol Lead Mercury Methanol Methotrexate Paraquat (quantitative plasma assay) Phenobarbital Phenytoin Thallium Thyroxine Toxicology screen*
*(The scope of the toxicology screen may vary according to local needs.)
For poisoning by cyanide, results from laboratory analysis are very unlikely to be available rapidly enough to influence the acute management of an individual patient. However, analysis of samples may inform management of further cases in the event of a release involving multiple casualties and also for medicolegal and forensic purposes. For these reasons only, a cyanide assay is not included in these guidelines for the acute management of individual patients.
Carbamazepine
The acute management of carbamazepine poisoning, including the need for multiple doses of oral activated charcoal, is determined by the clinical picture. There is no need for an urgent carbamazepine assay in the great majority of patients who have taken an overdose. Urgent measurement of plasma carbamazepine concentrations is only required when multiple-dose activated charcoal is being considered or when there is doubt about the diagnosis, for example in patients with:
○ coma; ○ respiratory depression; ○ arrhythmias. When indicated, optimal clinical management requires that an urgent result from this assay should be available within 2 h. Serious complications are unusual at plasma concentrations less than 25 mg/L (105 µmol/L). Most patients with life-threatening toxicity have plasma carbamazepine concentrations in excess of 40 mg/L (170 µmol/L). Non-urgent measurement is helpful in determining when to restart chronic carbamazepine therapy.
Cholinesterase
Measurement of plasma or red cell cholinesterase is useful to confirm exposure to substances that inhibit cholinesterase, e.g. organophosphorus or carbamate compounds. This may occur as a result of exposure to pesticides (farm workers, treatment of pests) or nerve agents. Optimal clinical management requires rapid access to an assay result which should ideally be available within 3 h, including journey time. Plasma cholinesterase activity is simpler to measure and is inhibited more rapidly in poisoning. However, reduced activity is less specific and may also occur in genetic pseudocholinesterase deficiency (suxamethonium apnoea), early pregnancy, liver disease, malignancy and hypoalbuminaemia. Measurement of the red cell cholinesterase activity is more specific and may be used to confirm severe organophosphate or carbamate poisoning when the diagnosis is in doubt. For optimum clinical management, a red cell cholinesterase activity result should be available within 6 h, allowing for journey times. Antidotal treatment, initially with atropine and subsequently (in selected cases) with an oxime (e.g. pralidoxime, obidoxime), should not be delayed pending the result of assays, when severe poisoning is suspected. Red cell cholinesterase activity is usually reduced to <50% with clinically significant poisoning and to <10% in severe cases. Plasma cholinesterase falls and recovers more rapidly after exposure than red blood cell (RBC) cholinesterase. It may take 90–120 days for RBC cholinesterase to recover to normal values. Advice on the interpretation of cholinesterase assays and on the use of antidotes can be obtained from the NPIS.
Ethylene glycol(1,2 ethanediol) and methanol
Urgent measurement of plasma or serum ethylene glycol or methanol concentration is essential for optimum management. These assays are not readily available and their use should be discussed with a senior local biochemist and the NPIS. When an assay is indicated, optimal clinical management requires that a result should be available within a maximum of 4 h. It is reasonable to restrict use of these assays to:
○ patients who give a history of substantial ingestion; ○ patients with suspected severe toxicity as evidenced by metabolic acidosis, especially pH <7.2 (hydrogen ion >60 nmol/L) in the presence of an increased anion gap, with or without an increased osmolar gap, when a reliable history is unavailable; In these cases, antidotal treatment should be commenced pending the result of the assay. A raised osmolar gap may be present in the early stages of poisoning. Due to the molecular weight of ethylene glycol, and the normal variability of osmolar gaps between patients, toxin concentrations of ethylene glycol may be present without the osmolar gap being greater than normal. An increased osmolar gap is consistent with poisoning but a normal gap does not exclude it. It may take several hours for toxic organic acids to be formed from ethylene glycol and methanol. Therefore, the anion gap may be normal in the early stages of methanol and ethylene glycol poisoning, even when life-threatening amounts have been ingested. Antidotal treatment with fomepizole (4-methylpyrazole) or ethanol is indicated for patients with evidence of severe poisoning. Advice on the criteria for antidotal treatment, haemodialysis or haemofiltration can be obtained from the NPIS and it may be necessary to measure concentrations daily. Antidotal treatment should be continued until the plasma ethylene glycol or methanol concentration is below the lower limit of quantitation (e.g. 50 mg/L) and the patient is asymptomatic with a normal pH. Haemodialysis or haemofiltration should be continued until the plasma ethylene glycol or methanol concentration is below the lower limit of quantitation and acidosis and signs of systemic toxicity have resolved. Daily measurement of ethylene glycol or methanol concentrations is helpful in determining the appropriate time to discontinue dialysis and/or antidotal treatment. Plasma concentrations should be repeated a few hours after haemodialysis is discontinued as rebound increases in concentrations have been reported. Ethylene glycol is metabolized to oxalic acid which forms complexes with calcium, so close monitoring of the plasma calcium concentration is required. Similarly, creatinine and electrolytes should be checked daily as severe renal impairment may occur with ethylene glycol poisoning.
Heavy metals
Acute poisoning with heavy metals is rare. Measurement of plasma and/or urinary heavy metal concentrations is useful in confirming the diagnosis. Urgent assays are rarely indicated. Urgent measurement is, however, occasionally justified for patients with suspected severe poisoning with lead or mercuric salts. Urgent use of these assays must be discussed with a senior local biochemist and the NPIS. In such cases, these results should be available for urgent cases within 2 h. Samples should be taken into a metal-free container. A whole blood specimen is essential for lead and mercury analysis, as these metals are bound predominantly to red blood cells. Advice on appropriate specimen collection and use of anticoagulants should be obtained from the laboratory performing the analysis. Advice on the management of heavy metal poisoning, including the use and interpretation of assays and the utility of chelation therapy, can be obtained from the NPIS.
Arsenic
Arsenic poisoning may occur after ingestion of wood preservatives or in the production of glass, alloys, rodenticides, pesticides, marine paints and semiconductors. Clinical features of acute arsenic poisoning include abdominal pain, vomiting, a garlic odour on the breath, diarrhoea (sometimes with blood), myocardial depression, vasodilation, electrolyte disturbances, acute renal failure, cerebral oedema, coma, convulsions and ventricular fibrillation. Chronic ingestion may cause anorexia, weight loss, diarrhoea, peripheral neuropathy or skin rashes, including palmar keratosis. Blood arsenic concentrations should be measured in cases of suspected acute arsenic toxicity to confirm the diagnosis, particularly if chelation therapy is being contemplated. However, concentrations correlate poorly with the severity of poisoning. Following acute toxicity, blood arsenic concentrations decline rapidly into the normal range in spite of continuing evidence of clinical toxicity, and may be undetectable more than 4 h after ingestion of potentially fatal amounts. Toxicity is usually associated with blood concentrations >100 µg/L (>1.3 µmol/L), while concentrations >200 µg/L (>2.7 µmol/L) indicate significant acute exposure. Measurement of urinary arsenic concentrations is useful for late presentations of acute poisoning and for assessing chronic toxicity.
Lead
Lead poisoning may occur following exposure to lead paint, leaded petrol, contaminated drinking water and ethnic remedies or cosmetics. Occupational exposure may occur in plumbers, lead miners, shipbuilders, construction workers, pottery manufacturers and demolition workers. Clinical features of acute and chronic lead poisoning are described on TOXBASE®. Non-urgent measurement of whole blood lead concentration is required in patients with suspected lead exposure or toxicity to confirm the diagnosis. This is particularly important if chelation therapy is being contemplated. However, urgent measurement of plasma lead concentration is necessary in children with suspected lead encephalopathy. Whole blood lead concentrations are normally <20 µg/L (<0.1 µmol/L) in the general population. Patients with concentrations higher than this should be discussed with the NPIS. Concentrations in excess of 600 µg/L (3 µmol/L) [children] or 800 µg/L (4 µmol/L) [adults] are usually associated with severe toxicity.
Mercury
Measurement of whole blood mercury concentration is helpful in patients with suggested mercury intoxication from acute exposure to mercuric salts or mercury vapour. The clinical features are described on TOXBASE®. Whole blood mercury should always be measured prior to administration of antidotes such as DMPS and the use of antidotes should be discussed with the NPIS. Urgent measurement is only indicated in patients with suspected severe acute poisoning from mercuric salts. Whole blood mercury concentrations are usually less than 25 µg/L (125 nmol/L). Patients with concentrations higher than this should be discussed with the NPIS. Chelation therapy may be considered in patients with blood mercury concentrations >100 µg/L (>500 nmol/L). Urinary mercury should be measured in patients with suspected chronic mercury intoxication.
Thallium
Thallium poisoning may occur following ingestion of rodenticides or occupational exposure during manufacture of optical and electrical equipment. Measurement of thallium concentration in blood or urine is helpful in patients with suspected thallium intoxication. The clinical features are described on TOXBASE®. The thallium concentration should always be determined prior to administration of antidotes such as Prussian Blue (potassium ferric hexacyanoferrate). Use of antidotes should be discussed with the NPIS. Urgent measurement is only indicated in patients with suspected severe acute thallium poisoning. Chelation therapy with Prussian Blue should be considered in patients with blood thallium concentrations >10 µg/L (>50 nmol/L) or urine thallium concentrations >20 µg/L (>100 nmol/L). These patients should be discussed with the NPIS.
Methotrexate
Acute methotrexate overdose is rare but chronic, or acute on chronic exposure is more common. Measurement of the plasma methotrexate concentration at least 4–6 h after the most recent ingestion is helpful. To be clinically useful, a turnaround time of <24 h is necessary. Methotrexate drug concentrations help determine an appropriate initial dose and the timing of discontinuation of folinic acid. However, treatment should not be delayed pending a result. Treatment with folinic acid, with or without urine alkalinization, should continue until the plasma drug concentration is <0.05 µmol/L (<50 nmol/L).
Paraquat (quantitative plasma assay)
In patients with a positive urine test (dithionite spot test) [see section ‘Paraquat (qualitative urine test)’ above], a single plasma sample should be taken for analysis in a specialist laboratory. Relating the paraquat concentration to the time since ingestion provides valuable prognostic information, when a plasma sample is available within 24 h of exposure. Concentrations at various time points may be interpreted using information provided on TOXBASE®. In patients presenting later than this, there is less information available.
Phenobarbital (phenobarbitone)
Most patients who have taken overdoses of phenobarbital do not require measurement of plasma concentrations. This applies even if poisoning is severe, provided the diagnosis is not in doubt. It is sometimes appropriate to obtain phenobarbital concentrations urgently in patients with severe toxicity when there is doubt about the diagnosis. Examples would include:
○ Patients with unexplained unconsciousness, where phenobarbital poisoning is suspected. ○ Patients with clinical features consistent with phenobarbital toxicity being considered for multiple-dose activated charcoal, when the diagnosis is in doubt and in the absence of a clear history of phenobarbital exposure. Urgent measurement of plasma phenobarbital concentration is also occasionally required to monitor anti-epileptic treatment in neonates. Use of assays for therapeutic drug monitoring is outside the scope of this guidance. The usual target range for phenobarbital is 10–20 mg/L (43–86 µmol/L). Serious toxicity is usually associated with concentrations in excess of 75 mg/ L (325 µmol/L), while concentrations >150 mg/L (>645 µmol/L) suggest life-threatening toxicity. Optimal clinical management requires that the results of urgent phenobarbital assays should be available within 2–4 h. Routine measurements may be useful to monitor anti-epileptic therapy or to time the re-institution of chronic therapy after overdose.
Phenytoin
Most patients with acute phenytoin overdose do not require measurement of plasma phenytoin concentration. An urgent phenytoin concentration is helpful (but not essential) if multiple-dose activated charcoal is being contemplated, particularly if the diagnosis is in doubt, e.g. in patients with coma, respiratory depression or arrhythmias. However, the clinical value of this elimination method for phenytoin intoxication is unproven. Rarely, urgent measurement of the phenytoin concentration may help to differentiate between convulsions due to phenytoin toxicity and those resulting from inadequate anticonvulsant concentrations. Optimal clinical management requires that, when indicated, the results of urgent phenytoin assays should be available within 2 h. Patients with suspected chronic phenytoin toxicity as a result of therapeutic dosing should have their plasma phenytoin concentration measured, but there is no need for this to be done urgently. The usual target range for phenytoin is 8–15 mg/L (32–60 µmol/L). Symptomatic toxicity is usually associated with concentrations in excess of 20 mg/L (80 µmol/L), while concentrations >40 mg/L (>160 µmol/L) suggest serious toxicity. Routine measurements may be useful to monitor anti-epileptic therapy or to time the re-institution of chronic therapy after overdose.
Thyroxine
Plasma free thyroxine should be measured in patients presenting with thyroxine overdose greater than 0.1 mg/kg. There is no need for this to be done urgently. Patients with high free thyroxine concentrations 6–12 h after thyroxine ingestion should have outpatient review 3–6 days after ingestion to detect and manage delayed-onset hyperthyroidism.
Toxicology screen
Various types of urine screening tests are available, either as immunoassays or as commercial or in-house chromatography systems. Immunoassays are targeted towards specific drugs or groups of drugs, e.g. methadone, benzodiazepines and opiates, whereas chromatographic systems detect a much wider range of substances. However, a negative result does not rule out poisoning, even when both techniques are applied, as there is a host of compounds that they will not detect. A positive result does not prove that the substance identified is the cause of the patient’s clinical features. Use of these screening tests without an understanding of their potential limitations is hazardous. Toxicology screens are useful under some specific circumstances, for example:
○ for determining previous ingestion of illicit drugs; ○ in the management of ‘body packers’; ○ in the diagnosis of unexplained coma; ○ as an aid to the confirmation of brain death and for assessing suitability for organ donation; ○ for forensic reasons. Close cooperation is required between local laboratory and clinical staff to choose tests that reflect local needs and to ensure that clinical staff obtain optimum use of these assays and understand their limitations.
Footnotes
Acknowledgements
The authors wish to thank all those who contributed to the previous guidelines (DNB, RB, NSB, GSC, CD, PD, NE, MH, DRJ, DJ, AJ, EM, KNM, SM, MAP, PAR, JPT, SHLT, JPT, JAV, GV, IDW and BW).