Postgrad Med J: first published as 10.1136/pgmj.69.811.337 on 1 May 1993. Downloaded from Postgrad Med J (1993) 69, 337 - 339 i) The Fellowship of Postgraduate Medicine, 1993 Review Article Diagnosis and treatment ofdigoxin toxicity Gregory Y.H. Lip, Malcolm J. Metcalfe and Francis G. Dunn Department ofCardiology, Stobhill General Hospital, Glasgow G21 3UW, UK Introduction Cardiac glycosides are unusual in having a narrow important to emphasize that the clinical diagnosis therapeutic range, which is idiosyncratic to the of toxicity is of fundamental importance and individual. In view of this it is perhaps not surpris- should not be discarded because of 'normal' ing that toxicity is a common occurrence, being plasma digoxin concentrations. reported in up to 35% of digitalized patients.' There are several mechanisms which can lead to Use ofplasma concentration measurements this problem. Firstly, digoxin is excreted mainly by the kidneys, and therefore, any impairment ofrenal In an attempt to improve digoxin therapy, it is function may lead to higher than expected plasma frequently advocated that the plasma digoxin con- concentrations. Congestive cardiac failure, renal centration should be measured. Trough plasma failure and advanced age can also cause toxicity by concentrations below 0.8 ng/ml (1.0 nmol/l) are reducing the volume of distribution of the drug. considered sub-therapeutic and levels greater than Concomitant electrolyte imbalance, notably 2.0 ng/ml (2.56 nmol/l) toxic. Unfortunately, there hypokalaemia, hypomagnesaemia and hypercal- is a marked overlap of measured plasma levels copyright. caemia can potentiate digoxin toxicity. Approx- between groups of patients with and without imately 30% of digoxin is plasma protein bound evidence of toxicity.3'4 For example, one patient and thus certain other drugs such as amiodarone may exhibit evidence of toxicity at a measured and calcium antagonists can lead to higher than plasma drug level ofonly 0.8 ng/ml, whilst another expected plasma concentrations. Lastly, several may be symptom free at a level of 3.0 ng/ml. clinical conditions such as hypothyroidism, It can thus be seen that plasma level concentra- chronic lung disease and cardiac amyloid are tion estimation is only of limited clinical value. associated with an abnormally high myocardial Plasma level measurements may be useful to http://pmj.bmj.com/ sensitivity to digoxin. Despite all of this, however, confirm clinical signs of toxicity or to see whether there is often still no clear relationship between or not there is any scope for reducing the dose and these factors and manifest toxicity. still achieving a therapeutic effect. Plasma levels are also useful to monitor compliance, to allow adjust- ment of dosage in the presence of certain con- Diagnosis oftoxicity comitant therapy and (in certain circumstances) to monitor treatment of overdosage. There is, never- on September 24, 2021 by guest. Protected Clinicalfeatures theless, no place for performing 'routine' drug level assessments in clinical practice. Features of cardiac glycoside toxicity are usually non-specific and can be conveniently divided into cardiac and non-cardiac effects.2'3 Non-cardiac General principles ofmanagement symptoms such as anorexia, nausea, vomiting, lethargy, headaches, confusion and more rarely When a patient is diagnosed as suffering from visual disturbances are the most frequent present- digoxin toxicity, the initial step is to assess the ing features. With increasing toxicity, the more severity of the problems. The reason for the serious cardiac features become inevitable, development of toxicity should always be although occasionally they can be the presenting thoroughly sought: for example, use of an inap- problem. Almost any permutation of heart block, propriate high dose, the addition of concomitant brady- and tachydysrhythmias are possible.2 It is medication or the development of a reduced volume of distribution arising owing, for example, to progression of heart failure or development of Correspondence: G.Y.H. Lip, M.R.C.P. renal impairment. The question ofeither accidental Accepted: 4 December 1992 or deliberate overdosage should also be considered. 338 G.Y.H. LIP et al. Postgrad Med J: first published as 10.1136/pgmj.69.811.337 on 1 May 1993. Downloaded from If the evidence of toxicity is relatively minor Specific aspects ofmanagement with, for example, symptoms of nausea, with- drawal of the drug is often the only specific Activated charcoal treatment required. More serious evidence of tox- icity, particularly with cardiac involvement, One of the most effective agents for the treatment requires admission to hospital. ofdigoxin toxicity is activated charcoal. This agent If there is no evidence of serious cardiac prob- is especially useful for patients who have taken a lems, such as heart block or significant dysrhyth- recent overdosage of digoxin, for patients without mias, then following the withdrawal of the drug serious evidence of toxicity but otherwise sympto- and the correction of electrolyte imbalance, symp- matic, or in situations of serious toxicity when tomatic measures will usually suffice. The presence digoxin-specific antibodies are unavailable. In of major dysrhythmias is clearly potentially life- order to achieve an adequate therapeutic response, threatening and carries a high mortality. In these multiple oral doses of activated charcoal are often circumstances, a rapid and more aggressive man- required to maintain an effective concentration agement stratagem must therefore be followed, gradient. A suggested regimen is a loading dose of ideally within the coronary care unit with 50-100 g, followed by either 10 g hourly, 10-20 g experienced staff available. 2 hourly or 40 g 4 hourly.'0-12 Immediate management centres around treat- Activated charcoal has been shown to increase ment of heart block with atropine (2-3 mg) and the clearance of digoxin by two mechanisms.""3 temporary pacing. Malignant ventricular dys- Firstly, and of greatest importance, is its effect in rhythmia should be treated with beta-blockers, moderating 'gastrointestinal dialysis' using the lignocaine or phenytoin. Phenytoin (100 mg intra- gastrointestinal wall as a dialysis membrane.""2 venously, repeated after 5 minutes if required) is a Secondly, it interferes with the enterohepatic cir- useful antidysrhythmic as it opposes digitalis bind- culation of digoxin by binding to the drug in the ing and may improve atrioventricular conduction intestinal lumen, resulting in increased elimination by its anticholinergic properties.' Beta-adreno- ofthe recycled drug.'2 Other binding agents such as ceptor blocking agents can terminate glycoside- colestipol and cholestyramine have also been induced ventricular dysrhythmias but are more shown to be effective.""3 In contrast, haemo-copyright. likely to lead to heart block.5-7 Therefore an dialysis and haemoperfusion are clinically ineffec- ultra-short acting intravenous beta-blocker tive.14"15 preparation such as esmolol (half-life, tt= 9 The potential benefits ofsuch an effective, simple minutes) may be useful in these circumstances as its and inexpensive mode of treatment have to be effects may be rapidly reversed.8 Correction of weighed, however, against any resultant morbidity. hypokalaemia is important, generally requiring The major problems are unpalatability of the intravenous supplementation of no more than activated charcoal and the risk of aspiration in http://pmj.bmj.com/ 20 mmol/h of potassium, to reduce digitalis bind- patients who may well be experiencing severe ing to Na/K ATPase."9 More rapid infusion of nausea and vomiting. The treatment of these latter potassium may lead to asystole and concomitant symptoms also poses problems, as the majority of infusion of dextrose may aggravate any hypo- antiemetic agents have concomitant anticholiner- kalaemia.S gic properties, decreasing gastrointestinal motility Cardioversion should be avoided wherever pos- and consequently the concentration gradient sible, due to the risk of precipitating asystole, and across the bowel. Ifthe patient is in a state ofshock when necessary should be attempted using the then reduced splanchnic blood flow will also lead to on September 24, 2021 by guest. Protected lowest energy possible. Elimination of the cardiac reduced gastrointestinal motility. glycoside should then be performed as expeditiously as possible using digoxin-specific Digoxin-specific antibodyfragments antibody fragments. In accidental or deliberate overdosage, general Digoxin-specific fab antibody fragments (Digi- principles oftreatment remain the same. Ifthe drug bind, Wellcome) are possibly the most effective has been ingested within 4 hours the patient should treatment available. However, this therapy is undergo immediate gastric lavage leaving 100 g of expensive and therefore should be reserved for activated charcoal within the stomach. If ingested treatment of serious toxicity, especially in the after 4 hours then the patient should be treated only presence of cardiac dysrhythmias. with activated charcoal. There is evidence that this These antibodies have a high affinity and specifi- therapy is as effective as ipecacuanha-induced city for cardiac glycosides and have been shown to vomiting in preventing drug absorption.'0 reverse digoxin toxicity and reduce the risk of death.'6-18 In several large studies, approximately 80% of patients had complete resolution of all evidence of toxicity, 10% improved whilst 10% Postgrad Med J: first published