Rational use of digitalis requires adequate knowledge of the drug, the disease and the ... tion (e.g. deslanoside) with similar therapeutic half-lives. (to facilitate ...
31 Januarie 1976
Digitalis in Clinical Practice P. H. JOUBERT SUMMARY Rational use of digitalis requires adequate knowledge of the drug, the disease and the patient, and an ability to integrate these factors. The pharmacology of digitalis is reviewed and the pathophysiology of cardiac failure and supraventricular arrhythmias, as well as aspects of toxicity and the use of serum levels, is discussed. Emphasis is placed on individualisation of therapy and on careful clinical observation.
S. Air. med. J., 50, 146 (1976).
The rational therapeutic use of the cardiac glycosides (collectively referred to as digitalis) requires adequate knowledge of the individual characteristics of the patient and his disease, and of the pharmacological propertie of digitalis. Insight into the interaction between patient, disease and drug is required for an intelligent evaluation of the patient's response to therapy and for rational prognosis.
PHARMACOLOGY Chemistry The cardiac glycosides are characterised by a steroid nucleus to which various sugars, an unsaturated lactone ring and various smaller radicals are attached. The steroid nature of digitalis is important with regard to sideeffects and interference with radio-immunoassay techniques. Although chemical differences exist between the commonly used cardiac glycosides, their basic pharmacological properties are similar. The main differences are of a pharmacokinetic nature (absorption, metabolism, excretion, duration of action, etc.). In my opinion it is sound practice to be thoroughly familiar with a single oral preparation (e.g. digoxin) and a single parenteral preparation (e.g. deslanoside) with similar therapeutic half-lives (to facilitate interchangeable use), rather than to have an inadequate knowledge of several preparations.
Pharmacokinetics The pharmacokinetic differences between some of the cardiac glycosides are shown in Table 1. '-3 Any drug Department of Phannacology, University of the Orange Free Slate, Bloemfontein P. H. JO BERT, B.Se., 1\I.B. CH.B., F.C.P. ( .A.), Senior Lecturer Date received:
given at regular intervals will take approximately 5 halflives to reach a 'steady state' (e.g. ± 1 week for digoxin and ± 25 days for digitoxin) if its excretion is not impaired. There appears to be a close correlation between serum half-life and pharmacological half-life (measured by systolic time intervals) for the cardiac glycosides.... It is, however, possible to demonstrate pharmacological effects on the heart by electrocardiography when digitalis cannot be detected in the serum" The rationale for a 'digitalising' dose is therefore to achieve an optimal 'steady state' rapidly by an initial loading dose, thereby circumventing the 5 half-lives. If one employs a loading dose of digitalis, the amount given depends largely on the lean body mass of the patient, whereas maintenance will be determined by the rate of elimination. This is determined mainly by renal function for digoxin and deslanoside.
PharmacodyDanncs Considerable uncertainty still exists regarding the exact mechanism of action of digitalis and how this is related to its classical effects, positive inotropism and negative chronotropism (caused by vagal stimulation, direct sinus node depression, reversal of the compensatory tachycardia of cardiac failure and decreased ventricular response owing to diminished atrioventricular conduction in supraventricular tachyarrhythmias). It is now well established that digitalis inhibits sodium potassium ATPase at a subcellular lever-' and that this correlates with pharmacological effects."1O The net effect of this inhibition is a potentiation of potassium ion loss from the cell and entry of sodium into the cell. In addition, it has been shown that digitalis increases the availability of calcium ions intracellularly. 1-',11 Since calcium ions are essential for exitation-contraction coupling,'3 the increased availability of these ions presumably forms the basis of the positive inotropic action of digitalis. It is not clear whether sodium potassium ATPase inhibition and positive inotropism are related. After the removal of digitalis from an isolated preparation, ATPase inhibition persists, but positive inotropism is lost. ~," It does, however, appear reasonable to attribute cardiac toxicity to abnormal sodium potassium exchange across the myocardial cell membrane. It is imperative to appreciate the narrow therapeutic index of digitalis, i.e. the small difference between 'therapeutic' and 'toxic' dosages. Furthermore, individual variation and factors such as electrolyte balance and tbe nature of the myocardial disease may influence the response to a particular dosage. These considerations are demonstrated by the overlap between therapeutic and toxic serum levels 15- 11 (Fig. 1).
31 January 1976
TABLE I. PHARMACOKINETIC PROPERTIES OF DIGOXIN, DESLANOSIDE AND DIGITOXIN Absorption after oral c:dministration (%)
Digoxin (Lanoxin) Deslanoside (Cedilanid) Digitoxin
80 (tablets) 100 (elixir) 30 - 40 100
Serum half-life (days)
Excreted unchanged in urine per day ('1'~ )
Hepatic metabolism per day ('1'0)
Side-Effects and Toxicity 9 E '-
D i He rence
Cardiac toxicity. The cardiac manifestations of digitalis toxicity can basically be related to impaired atrioventricular conduction and increased automaticity. Con equently, a wide variety of arrhythmias may occur, e.g. supraventricular, ventricular and various degrees of atrioventricular block. Severe overdosage can result in cardiac arrest which does not respond to electrical pacing," and death is associated with severe extracellular hyperkalaemia," probably a direct result of sodium potas ium ATPase inhibition. A characteristic manifestation of cardiac toxicity is ventricular extrasystoles with a tendency to coupling (Fig. 3).
p .- D. 01
4 3 2
o GOOD CONTROL OF CARD IAC FAILURE WITH NO TOXIC ITV
CLINICAL DIGITALIS TOXICITY
Fig. 1. Serum digoxin levels of patients with and without features of digitalis toxicity.
The main electrocardiographically demonstrable effects of digitalis in therapeutic dosages are PR-interval prolongation, QT shortening, ST' depression and T-wave depression. 1S,19 The so-called 'digitalis effect' on the ST segment ('hockey-stick' or 'inverted correction mark') can be quite characteristic, especially in the absence of pre-existing T-wave abnormalities (Fig. 2), and is best seen when leads with the largest net positive QRS deflection are used.
Fig. 2. Electrocardiogram showing the typical 'hockey-stick' appearance of the ST segment in a patient receiving digitalis.
Fig. 3. Coupled ventricuiar extrasystoles in a patient with digitalis toxicity; a = normal QRS complex; b = ventricular extrasystole.
It is important to realise that hypokalaemia, hypomagnesaemia, hypercalcaemia, and disease states such as myocarditis, hypothyroidism and myocardial infarction, may predispose patients to digitalis-induced cardiac arrhythmias at dosages of digitali and serum levels which would normally be regarded as therapeutic. Extracardiac manifestations of toxicity. Extracardiac signs and symptoms of digitalis toxicity are largely limited to the gastro-intestinal tract and the central nervous system. Anorexia, nausea, vomiting and diarrhoea are common toxic manifestations. Visual complaints, headache and mental impairment (particularly in the elderly) may be toxic signs which could easily be overlooked. Allergic reactions to digitalis are rare. An uncommon side-effect is gynaecomastia,"'" which occurs especially when digitalis i used in conjunction with another teroidal drug such as spironolactone (Aldactone). Thi is probably caused by increased levels of circulating oestrogen in patients on long-term digitalis therapy."
CARDIAC FAILURE Digitalis is used clinically for its positive inotropic effect in conge tive cardiac failure and for it effect on atrioventricular conduction in upraventricular tachyarrhythmias (particularly atrial fibrillation) to decrease the rate of ventricular response.
Pathophysiology Cardiac failure can re ult from a multitude of factors uch a hypertension, myocardial i chaemia, the cardiomyopathy een in Black and rheumatic heart disease. It is a serious condition with a poor prognosis" and warrants meticulous attention. The clinical picture is due to the interplay of several factors: the decrease in cardiac output and tissue perfusion, the inability to cope with venous return and several compensatory mechanisms (Fig. 4).
flUID RETENT I ON
ItKREASED REN I N
L - ALO~STERONE
Fig. 4. Schematic representation of the heart as a pump and the various mechanisms operating in cardiac failure.
31 lanuarie 1976
The net effect of compen atory mechanisms is therefore tachycardia, positive inotropi m and fluid retention. As cardiac failure progre es these mechani ms become le and le effective in maintaining the circulation and the feature of venou congestion become more prominent. The situation is eventually war ened by secondary hyperaldosteronism," which is caused by greatly diminished renal perfusion. Hyperaldosteronism causes increased sodium and fluid retention (Fig. 4). As cardiac output fall , the clinical feature of low cardiac output develop, indicating a serious tage of cardiac failure.
Effect of Digitalis Since digitalis is a positive inotropic drug, it will increase cardiac output, unless gross mechanical obstruction exists. The Starling curve is moved upwards, i.e. the ability of the heart to respond to diastolic volume overload by positive inotropism is enhanced. In addition, the oxygen consumption per unit work performed is decreased."' As cardiac output improves, the sign and symptoms of hypervolaemia and venous congestion diminish. Tisue and, pecificaliy, renal perfusion improve and a secondary diuresis occurs. The compensatory increase in sympathetic tone is progressively lost" and there is a consequent decrease in heart rate and peripheral vasoconstriction. Although it is the usual practice to combine digitalis with diuretics, digitalis can be given without concomitant diuretic therapy·' However, resolution of oedema will be slow and incomplete in severe cases. Diuretics should be used responsibly because of the risks of electrolyte disturbances which may complicate the safe and effective use of digitalis. There are some who advocate the use of diuretics in congestive cardiac failure without the use of digitalis. 33 The logic of this in terms of pathophysiology is not clear. It should also be borne in mind that hypervolaemia is sometimes necessary for the maintenance of an adequate cardiac output and that the use of diuretics in the presence of severe pump failure may cause or potentiate the signs and symptoms of low output failure.
Compensatory Mechanisms and Decompensation
Apart from muscular hypertrophy, two basic compensatory mechanisms operate to maintain cardiac output (Fig. 4). The first is an increase in sympathetic nervous system tone,'· which occurs in response to a decreased cardiac output. The second mechanism i a consequence of the increa ed end-dia tolic volume, which is caused by a decreased cardiac output. Increased end-diastolic volume causes stretch of myocardial fibres which results in an in reased force of contraction (the Starling principle)."'" The myocardial response to increases in enddia tolic volume deteriorate as cardiac failure worsens.
Atrial fibrillation may be caused by several conditions, such as myocardial ischaemia, rheumatic heart disease and hyperthyroidism. The exact mechanism is not clear, but re-entry mechanisms in the presence of alterations in the refractory period seem a plausible explanation. If the ventricular rate of response is too rapid, haemodynamic problems occur because of an inadequate and varying blood volume presented to the ventricles. As a compen atory mechani m, various degrees of atrioventricular block may develop and may prevent a large
31 January 1976
percentage of the atrial impulses from reaching the ventricles. This is usually insufficient to prevent haemodynamic deficits. The logical approach to such a situation would be to restore sinus rhythm or to protect the ventricle from excessive stimuli by decreasing atrioventricular conduction.
Effect of Digitalis
The efficacy of digitalis in atrial fibrillation is largely owing to its depressant effect on atrioventricular conduction, the slower ventricular rate allowing for better diastolic filling and improved cardiac output. Digitalis is therefore of particular value in cases of atrial fibrillation of acute onset with haemodynamic impediment where electrical conversion is not available or hazardous, or when atrial fibrillation which has existed for a long time relapses after cardioversion. Although sinus rhythm may be restored during digitalisation, quinidine and electrical conversion are the most effective means of achieving this goal. Since quinidine initially enhances atrioventricular conduction, digitalisation is essenti,.l beforehand. Digitalis increases the risk of arrhythmias occurring after electrical cardioversion. This risk can be minimised by using lower energy levels than normal."'
OTHER ARRHYTHMIAS Digitalis can be effective in the control of paroxysmal supraventricular tachycardia. It should, however, be kept in mind that this condition may be a manifestation of digitalis cardiotoxicity. In atrial flutter, digitalis may increase the degree of atrioventricular block and thus reduce ventricular response. It may also convert flutter to fibrillation, which is a more stable and manageable rhythm.
Fig. 5. Serum digoxin lenls in a normal volunteer laking 0,25 mg of digo,xin per day from point A and 0,50 mg per day from point B. Note Ihat it takes ± 1 week to reach a stead}' state.
A digilalising dose in the case of a supraventricular arrhythmia wilh haemodynamic impairment seems entirely reasonable. In the emergency cardiac failure situation its value is less clear, and in acute pulmonary oedema it seems that therapy with morphine, oxygen, diuretics and theophylline is of greater importance. In emergency situations intravenous digitalisation is the only predictable route of administration, although it is not without hazards. The peak after oral absorption seems to be delayed in patients with le er degree of cardiac failure (Fig. 6), probably owing to inactivity, mucosal congestion and venous stasis, and it seems reasonable to assume that this is also the case in mOTe serious cardiac failure situations. By dividing the digitalising dose and giving repeated small intravenous doses every 2 - 4 hours while monitoring the clinical re ponse, the incidence of seriolls toxicity should be minimal.
3.0 A E
---. (0". c:
'" o 4 TIME
Fig. 6. Serum profiles in an ambulatol'}' normal volunteer (A), a normal volunteer I}'ing down (B) and a patient with congestive cardiac failure (C), after a similar single oral dose of digoxin. Tote the time of peak concentration.
If a loading dose of digitalis is indicated in a patient with cardiac failure, a rea onable approach is to use a loading dose of 0,01 mg/kg, e.g. 0,75 mg for a patient
weighing 75 kg. The maintenance do e, when renal function i normal, would be one-third of the loading do e, i.e. 0,25 mg per day in the ituation de cri bed. This regimen wit! re ult in an acceptable erum level in most instances and a low incidence of toxicity.""' Formulas to correct the maintenance do e for impaired renal function do exist. The one suggested by lellife'" i : creatinine clearance Maintenance do e = (14 + ) % 5 of loading dose. omograms to calculate creatmlOe clearance from serum creatinine, age and body weight have been devised." Tbe best initial dose for the average patient appears to be 0,25 mg of digoxin per day. With doses of 0,25 mg, 0,50 mg and 0,75 mg of digoxin (Lanoxin) per day no statistically significant differences are found in the number of patients who achieve 'therapeutic' serum levels (Fig. 7). However, a significant progressive increase in the number of patients with 'toxic' serum levels is seen as the doses increase. Except with severe renal impairment, the incidence of 'toxic' serum levels with a daily dose of 0,25 mg is negligible.
• D;ff~r~nc:u (.1) p 0.05
o D.S-l,e "q/_I
0.2S ~"l O",o· i " d.lily
Oigo.. ;1'\ d.lily
d .. ill'
Fig. 7. Incidence of 'subtherapeutic' « 0,8 ng/ml) 'therapeutic' (0,8 -1,8 ng/ml) and 'toxic' (> 1,8 ng/ml) serum levels in patients on 3 different dosage regimens.
If adequate control is not achieved with this regimen, the dosage can be increased. However, patients with 'refractory cardiac failure' often appear to be receiving too mucb rather than too little digoxin." The atrial fibrillation situation is often overlooked in discussions of this nature, and precise data are difficult to obtain from the literature. It is, however, clear that patients with a rapid ventricular rate need higher dosages of digoxin than do patients with cardiac failure. Factors such as initial ventricular rate and the degree of response to digitalis will determine the dose needed. Although a marked degree of individual variation exists, there is a relationship between dose or serum level and ventricular rate in individual patients studied serially: In emergency situations, small doses of intravenous digitalis could therefore be titrated again t ventricular response.
31 lanuarie 1976
It i therefore clear that rigid do age chedules do not constitute a replacement for careful clinical assessment and judiciou monitoring of patient re pon e. They are guides to effective therapy, but not the be-all and end-all of good therapeutics.
SERUM LEVELS Red cell Rb uptake" and radio-immunoassay'·'" technique have made it possible to determine cardiac glycoside concentrations in body fluids and tissues with relative ease. Of these two techniques, the latter is the most widely used, and kits for these determinations are commercially available. The literature now abounds with studies of serum digoxin levels, and estimations are done routinely by laboratories in many centres. However, the correlation between serum levels and effects is not clear. It is difficult to quantitate the therapeutic effects of digitalis in heart failure because of extraneous factors which may influence the response and because noninvasive endpoints (such as oedema, third heart sound, hepatomegaly, etc.) are often vague and subjective. Furthermore, the disease itself may not be static, and bed rest and other drugs also have marked effects. Toxicity is easier to evaluate because definite electrocardiographic criteria exist" which are useful in retrospective analyses. In prospective situations it becomes more difficult to evaluate. For example, only 49,5% of patents who present with fixed coupled ventricular extrasystoles have been taking digitalis:" It is therefore pos ible that many patients who develop bigeminy while receiving digitalis may have developed it for reasons unrelated to the drug. Extracardiac signs of toxicity can also be caused by factors other than digitalis. We have noted that patients who were diagnosed as suffering from digitalis poisoning on the basis of nausea and vomiting had a higher incidence of raised blood urea levels than those dianosed for otber reasons"· Gastro-intestinal sympto~s may, therefore, have been caused by uraemia. It is apparent that the diagnosis of digitalis toxicity and the interpretation of therapeutic response are confo:mded by many factors and that correlations with serum le~els are therefore difficult. In our experience, patients With good control of cardiac failure without features of toxicity usually have serum levels between 08 and 1,8 ngj ml. Patients suffering from digitalis t~xicity have higher blood levels, but considerable overlap exists." This has also been shown by most other workers,...·•• although one investigator·' could not find any difference between serum levels of 'toxic' and those of 'non-toxic' patients. Serum levels reach a peak within I - 2 hours after oral administration, and if blood is drawn during this period, serum levels might be within the toxic range, although the steady- tate values are much lower. We therefore recommend that blood should be drawn at least 6 hours, preferably even longer, after the last administration of digoxin.
Serum digoxin estimation is meaningless without in(ormation about the clinical state of the patient, the [Iature of the disease, renal function, electrolyte balance, dosage, the duration of therapy, the time lapse ince the l.ast administration, and other drugs administered. With the advent of laboratory and technological medicine, special investigations have become a crutch for the weakening legs of clinical acumen. Serum digoxin l;tstimation is no exception. Injudicious and indiscriminate serum digoxin estimations are a waste of time, energy and money, but if they are used with insight and in the correct situation they can be an invaluable adjuvant to good therapeutics. The most important determinant of serum digoxin levels appears to be the reliability of the patient in taking his medication." If a baseline level is estimated in outpatients of doubtful reliability, non-compliance can easily be detected and the unnecessary cycle of an ever-growing prescription can be avoided. In compliant patients who are difficult to control, serum digoxin estimations can be invaluable, e.g. the serum half-life can be determined in patients with impaired renal function and the dosage can be tailored accordingly. Serum level estimations will prevent unnecessary toxicity in cases of 'refractory' cardiac failure. Serum digoxin estimation may be indicated as an emergency procedure when a clinician wants to exclude digitalis toxicity as a cause of arrhythmia in a patient who is suffering from serious supraventricular arrhythmias, and for whom digitalis is the treatment of choice. It should be realised that, as with any other special investigation, the possibility of human or machine error should not be disregarded. The radio-immunoassay has other pitfalls. The presence of other sources of radioactivity (e.g. after isotope studies), jaundice or haemolysis may lead to erroneous results. The affinity of the antibody for steroid structures may give rise to problems by binding to other steroidal substances:' and we have from time to time encountered false high values due to the presence of spironolactone (Aldactone).