7.8.5 Atrial fibrillation
7.8.5 Atrial fibrillation

Atrial fibrillation is the most common dysrhythmia affecting exercise tolerance. It is important to learn to recognise this dysrhythmia from clinical examination. It is particularly important to be able to distinguish AF from 20AVB because long diastolic pauses can occur in both arrhythmias. With careful auscultation this should not be difficult; however, mistakes are sometimes made.

Poor athletic performance is the most common presenting sign in horses with AF. Less frequently, epistaxis, prolonged recovery following exercise and tachypnoea are seen and occasionally CHF, collapse, ataxia, myopathies and colic are associated with AF. In a significant number of cases, particularly in non-athletic horses, AF is an incidental finding.

In the USA, AF has been reported most often in Standardbred horses and is more common in animals of racing age. In the UK, it has been reported to affect heavy-hunter type animals and draught horses more frequently than other breeds, and is most often seen in horses aged 5-15 years. These reports may reflect the hospital populations at different institutions. In the author's experithere are three general categories of animal which are most frequently affected by the condition. The first are young and middle aged racehorses (2-10 years) with no evidence of underlying heart disease that present with a sudden deterioration in performance. The second are middle-aged (8-15 years) large Thoroughbred or Thoroughbred-cross horses with no evidence of underlying heart disease which have a more vague history of exercise intolerance. The third group is animals with underlying heart disease, particularly MR. AF is rarely found in ponies.

The pathophysiology of atrial fibrillation

In most species, AF is associated with significant underlying heart disease which results in atrial enlargement, usually left atrial enlargement. For example, in dogs it is most commonly seen in animals with dilated cardiomyopathy or severe MR; in humans it is often associated with mitral stenosis. AF is usually initiated by an APC which may or may not be related to myocardial disease. Experimentally, rapid atrial stimulation can set up AF. The arrhythmia will only be maintained if there is inhomogeneity of the state of refractoriness of the atrial myocardium. The premature beat starts a circus movement of depolarisation around the atria spreading from one area of excitable tissue to another. This requires a large mass of atrial tissue and suitable refractory and excitable cells. Surface mapping of atrial electrical activity in the presence of AF demonstrates a number of wave-fronts, with some areas of the atria being depolarised, some in a refractory state and others in an excitable state. The impulse can pass from the depolarising area to adjoining excitable areas, but cannot depolarise refractory areas. However, these refractory areas then become excitable, allowing the wavefront to come back and cause depolarisation. In sinus rhythm, once a wavefront has been conducted, it is surrounded by refractory tissue and the impulse cannot continue but in AF, the wavefront always has an adjacent excitable area which can be stimulated so the wavefront is perpetuated. Marked atrial enlargement increases the chances of circus movement of wavefronts developing because the activation pathways are increased in length.

AF is particularly common in horses for two reasons. In normal horses, of approximately 15 hands or more in size, the atria are sufficiently large for AF to persist once it has been set up. In animals with atrial enlargement it is even more likely that the condition will persist rather than revert to sinus rhythm. An additional factor which is responsible for the persistence of AF in horses is the high vagal tone found in this species. The release of acetylcholine shortens the refractory period to differing degrees in different cells within the atria, resulting in

an increased inhomogeneity of refractoriness. The inhomogeneity increases the chance of circus movements being set up. The effects of increased vagal tone are demonstrated in other species, for example, cattle with intestinal conditions which result in increased vagal tone may develop AF, but is abolished if the primary problem is reversed. Experimentally, dog atria have been maintained in AF by bathing them in acetylcholine (the neurotransmitter released by the vagus) once AF has been induced by rapid electrical pacing.

The cardiovascular effects of atrial fibrillation

Effective atrial contraction is important in animals with cardiac disease, because a stiff ventricle may not fill normally during the passive phase of early diastolic filling. Atrial contraction is required to force blood into the ventricles in this situation, so if AF develops, there may be a deterioration in clinical signs. However, in the absence of underlying heart disease, at normal resting heart rates, co-ordinated atrial contraction has little effect on ventricular filling. Conmany animals with AF show no clinical signs at rest.

Atrial contraction is of much greater importance during exercise when, at higher heart rates, it contributes a significant volume of blood to ventricular filling. AF may therefore limit ventricular filling during exercise, reducing stroke volume and affecting athletic performance. In addition, the priming effect of atrial contraction, which increases the force of ventricular contraction via the Frank-Starling mechanism, is lost when AF is present. AF results in higher heart rates during exercise at a lower level of work than would be found in the same animal if it was in sinus rhythm. Maximal heart rates may reach 240-260 bpm in these animals rather than 220-240 bpm. Cardiac output at peak exercise is reduced compared with normal animals because stroke volume falls. Horses with AF are unlikely to perform with success in athletic pursuits such as racing, three-day eventing, long distance riding, carriage riding, hunting or polo, although there are some exceptions to this rule. However, some animals, which are used for less arduous work such as show-jumping or dressage, may perform satiswhile in AF, at least in the short- to medium-term.

Clinical examination and electrocardiography

Clinical examination will reveal an irregular heart rhythm. The heart rate may be normal (as low as 24 bpm) or elevated. This contrasts with dogs, where AF is almost always accompanied by a tachycardia. There may be long pauses of up to 8 seconds or so, sometimes followed by flurries of beats. Sometimes the flurries will come in a cyclical fashion. The S1 and S2 sounds may vary in intensity due to the variable position of the atrioventricular valves at the beginning of systole. The most characteristic finding is the absence of the A sound (84). This corresponds to the ECG finding of an absence of P waves in any lead, due to the lack of any co-ordinated atrial activity. The P waves are replaced by f (fibrillation) waves. The R-R interval will be irregular; this may be more obvious at normal heart rates than in tachycardias (Figures 7.8 and 7.9). The pulse quality will depend on the diastolic interval, and on the presence of any underlying heart disease. Pulse deficits are common with very short diastolic intervals.

As a general principle, when a long diastolic interval is heard, it is important to try and identify an A sound because if A sounds are present then the cause of the arrhythmia is not AF. In 20AVB the pauses are usually multiples of the normal R-R interval and the arrhythmia has a regularly irregular, predictable nature. It may be helpful to use one's foot as a metronome to get used to the underlying rhythm. With careful auscultation, if 20AVB block is present, an A sound is likely to be heard during the pause. In AF, the rhythm can be described as irregularly irreTwo relatively uncommon situations may confuse the issue. Long diastolic intervals with no atrial contraction may be due to SA block rather than AF. In this case, SA block is usually found intermittently, with normal sinus rhythm being the predominant rhythm. A second condition which is rarely encountered is a split S1, which can be confused for an A sound and S1. In this situation AF is suson the grounds of the irregularly irregular rhythm and the absence of an A sound during the longer diastolic intervals. Where AF is suspected, it is wise to perform an ECG to confirm the diagnosis and to be able to detect any QRS complexes which are ventricular rather than junctional in origin.

Diagnostic features of AF are summarised in Table 7.3.

Management of horses with atrial fibrillation

An essential part of the management of cases with AF is a careful examination for underlying heart disease which might predispose to the condition, and an assessment of historical details. Horses with valvular or myocardial disease and those with a long-standing AF (typically more than 4 months) are less likely to be successfully treated than those without evidence of underlying disease or a recent onset of the arrhythmia. These horses are also more likely to suffer a recurrence of the condition at a later date even if treatment is initially successful.

A thorough clinical evaluation should always be made prior to treatment. Pathological murmurs should be identified (Chapters 3 and 6). Measurement of an accurate resting heart rate is also helpful. An elevated heart rate is more common in horses with underlying heart disease. Electrolyte abnormalities particularly hypokalaemia, may predispose to AF and therefore it may be beneficial to measure plasma levels in selected cases. Further tests such as a fractional excretion test can also be helpful (see section 2.3.6).

If echocardiography is available, it is the best method of assessing the severity of valvular or myocardial disease. The most common predisposing factor to AF is MR. Structural mitral valve abnormalities may not be seen echocardiobut measurement of left atrial size is very helpful. Pulsed-wave or colour-flow Doppler echocardiography can be used to map the area of the regurgitant jet and its origin, in order to provide a semi-quantitative measureof the severity of MR. Myocardial contractility can also be assessed, (although fractional shortening should be measured over several cardiac cycles because widely differing R-R intervals affect measurements) (see section 4.2.6 and section 6.7.1). Aortic regurgitation is less commonly associated with AF, but if significant AR is detected on clinical examination or echocardiography, the prognosis for return to sinus rhythm is reduced. MR is usually present as well as AR in these cases. TR may result in dilation of the right atrium and predispose to AF. However, it should be remembered that TR is a frequent incidental finding in fit, large racehorses and may be an incidental finding in animals with AF also (see section 6.3). Echocardiography is a very useful aid to assessment of severity of TR. In each case, particularly with MR, if echocardiography shows atrial or ventricular dilation the prognosis for conversion to sinus rhythm is worsened in comparison to animals with no volume overload. If severe myocardial disease is present, treatment should not be undertaken. In horses with a recent respiratory infection and reduced fractional shortening on echocardiography, treatment should be postponed for approximately 1-2 months, before re-evaluation.

In horses in which valvular disease is significant, or in which the arrhythmia may have been present for more than approximately four months, the risks of failure of treatment, potential complications during treatment or a reversion to AF later, and financial considerations, need careful discussion with the owner or trainer. These horses may not return to previous performance levels even if treatment is successful. In animals which have high financial or sentimental value, treatment can be pursued, but both the veterinary surgeon and the owner should be aware of these increased risks.

Horses with AF may have a better long-term prognosis if they are successfully treated rather than left untreated so, even in pet animals, treatment is usually beneficial in suitable cases. However, there are some circumstances when the prognosis for successful treatment is thought to be poor because of a prolonged history, or because treatment has been unsuccessful, or because AF has recurwhen the horses can continue in work despite the presence of AF. When AF is likely to have been present for some time and the horse is performing satiswith a normal resting heart rate and no signs of significant underlying heart disease, the potential risks of treatment may not be justified. Treatment is not risk free, and if the horse's performance is normal it may not be essential. In the author's opinion, AF does not in itself increase the risk of collapse at lower levels of exercise; it is when the condition is accompanied by significant heart disease and the horse is asked to work hard when it is tired that a risk to the rider develops. The first sign of compromise due to the condition and/or the underdisease is likely to be tiring or prolonged recovery following exercise. Owners should be made aware of this and advised to pull up horses once they feel them beginning to tire. The horse may be unsuitable for riders who are too inexperienced to detect this or when the individual horse is the type which 'soldiers' on regardless.

Breeding animals with AF usually benefit from treatment to convert them to sinus rhythm unless there are signs of marked underlying heart disease. Howboth mares and stallions may continue to perform satisfactorily when AF is refractory to treatment. If signs of CHF failure develop, these are best managed with diuretics and digoxin.

As a general rule, animals with AF should not be used for arduous competition and the level of work for which they are used should not be increased from that of which they were previously known to be capable without tiring, unless treatment is successful.

Horses with AF and CHF have a grave prognosis. Usually volume overload has resulted in dilation of the left atrium and myocardial changes which make a return of sinus rhythm unlikely.

Treatment of horses with atrial fibrillation

The most important consideration in treatment of AF is selecting suitable cases.

Animals with evidence of CHF or a resting heart rate of >55-60 bpm are not suitable for treatment. However, treatment of cases without evidence of underlying heart disease frequently results in a permanent return to sinus rhythm and subsequent normal athletic performance. Some animals can be treated repeatedly and perform well during the periods in which they are in sinus rhythm.

Quinidine sulphate is the drug of choice for treatment of AF. This is insoluble and must be given orally in a suspension There have been some reports of its use intravenously, but side-effects were common and this route of administration is not recommended under any circumstances. Quinidine gluconate is the soluble form of the drug, and has been used intravenously for treatment of AF with success. However, oral treatment is more reliable, particularly in cases which have been in AF for more than a few days. Quinidine gluconate is not available in the UK at the present time. Other drugs such as procainamide have been used for treatment of AF but do not appear to be as reliable as quinidine.

Oral treatment with quinidine sulphate The standard treatment for AF is quinidine sulphate, administered by stomach tube. Some clinicians recommend a test dose of 5 g per horse on the day before treatment is planned, to test for anaphylactic reactions. This is a rare problem and the author does not use a test dose. The dosage rate is 20 mg/kg (i.e. 10 g for a 500 kg horse), administered by stomach tube. The rate of absorption and half-life dictate the therapeutic regime. The aim of treatment is to titrate the drug to a plasma concentration within the therapeutic range (approximately 2.5-5.0 mg/i). Peak plasma levels are usually reached approximately 2 hours after administration. The dose is therefore repeated every 2 hours until the horse converts to normal sinus rhythm, or until toxic side-effects are recognised (see below). Increased dosage resulting in plasma levels above 5~0 mg/i is likely to result in side-effects and will not increase the chances of reversion to sinus rhythm because conversion normally only occurs with plasma concentrations in the therapeutic range. Ideally, on site quinidine assays can be used for monitoring plasma levels, but this is seldom possible in practice.

If the therapeutic range is exceeded, animals may convert to sinus rhythm when the plasma level falls back into the therapeutic range, after treatment has been stopped. Many horses revert to sinus rhythm after a dose of approximately 30-60 g. Side-effects are much more common after total doses in the range of

50-80 g have been administered. Continued dosage beyond this point may be unwise unless the clinician is confident that the toxic range has not been reached. Very few 500kg horses have drug levels below the therapeutic range after 60 g of quinidine given at two hourly intervals.

The approach to treating horses with AF is summarised in Table 7.4.

If initial treatment is unsuccessful A number of steps can be taken if initial treatment is unsuccessful. One approach is to take a blood sample for a quinidine assay to see if therapeutic levels have been reached. Treatment can be stopped while the results are obtained. If absorption has been poor, the plasma levels may be below the therapeutic range. The treatment can be repeated at a later date, with a higher total dose, if necessary.

Traditionally, if treatment was unsuccessful after the initial titration, it was resumed the following day. However, evidence suggests that the longer plasma levels are kept in the therapeutic range, the greater the chances of successful treatment. Since the half-life of quinidine is around 6 hours, doses can be repeated every 6 hours to try to maintain the drug level in the therapeutic range until conversion is achieved. If on-site assays are possible, it is desirable to repeat the assay 2 hours after administration of a dose and, if necessary, just before treatment, to obtain precise drug levels. The 6-hourly treatment can be repeated until side-effects become a problem or until the patience of the clinician is exhausted. Six-hourly administration of quinidine has been successful in some animals as long as three days after the beginning of treatment.

Digoxin is routinely used in addition to quinidine by some clinicians; however, there does not appear to be any particular advantage in the use of digoxin and, in the author's view, the additional drug only complicates evaluation of the patient. Interactions between the two drugs mean that administration of digoxin may increase plasma levels of quinidine, which may not be desirable. Under some circumstances the use of digoxin may be indicated. If, after 24 hours, treatment is still unsuccessful, intravenous administration of digoxin at a dose of 0.0025 mg/ kg, followed by oral administration at a dose of 0~015 mg/kg twice daily, may result in a return to sinus rhythm. If treatment with the standard regime is unsuccessful, using an oral maintenance dose of digoxin for three to five days prior to repeating the administration of quinidine may be worthwhile.

Horses which fail to convert with the traditional initial titration regime, but convert after 6-hourly treatments or treatment on another day, are more likely to revert to AF at a later date than animals which respond to the initial titration. The reversion rate in these animals is approximately 50%. Most horses which stay in sinus rhythm for more than a year after treatment are unlikely to revert to AF. Some horses can be treated repeatedly and turn in useful performances in the interim period. The suitability of a case for repeated treatment depends on the value of the horse's performance or sentimental factors. Owners should be warned of the increased risk of failure or later reversion in animals which need repeated treatment.

Management of side-effects and toxicity Many of the side-effects which have been reported to be associated with the use of quinidine are likely to have resulted from drug levels well into the toxic range and out of the therapeutic range. With the treatment regime recommended above, drug levels in the toxic range are less likely to be reached than with some regimes used previously. The use of on-site assays of plasma levels is the ideal method of making sure that animals stay in the therapeutic rather than the toxic range. However even with this facility, side-effects do occur, and it is important to recognise those which are to be expected in the course of normal successful treatment, those which are a cause of concern and indicate that further treatment should be stopped, and those which are life4hreatening and require intensive treatment.

Side-effects which are frequently seen during successful treatment include depression, mild colic and mild diarrhoea. More serious signs include marked tachycardia (heart rate > 100-120 bpm), weakness, severe colic and nasal oedema. If these signs are encountered, the drug levels are likely to be above the therapeutic range and treatment should be restricted to a six-hour dosage regime or stopped if they do not resolve. Colic may be sufficiently severe to require the use of analgesics and even cessation of treatment. Occasionally, ataxia, diarlaminitis, hypotension and collapse are reported. These situations indithat no further treatment should be given. Monitoring the ECG is helpful because it allows significant arrhythmias to be detected in addition to identifying the return of sinus rhythm. It may also be useful to monitor heart rate, which can be difficult to measure by palpation or auscultation at fast heart rates. Radiomonitoring of an ECG allows significant arrhythmias to be detected early, and corrective treatment can be instituted as soon as possible. A number of arrhythmias may develop during the course of treatment. Once AF has been terminated, a rapid supraventricular tachycardia with a higher heart rate than was present with AF can develop, and may be associated with a deterioration of clinical signs. Prolongation of the QRS complex by more than 25% has been reported to be an indication of toxicity.

The most significant cardiovascular side-effect of quinidine is hypotension. Therefore, where side-effects are a cause for concern, fluid therapy with a balanced electrolyte solution should be instituted. Ideally, an intravenous catheter should already be in place so that emergency treatment can be given if necesIf the side-effects are severe, intravenous sodium bicarbonate (1 mEq/kg) should be administered because it will increase the protein binding of quinidine. Phenylephrine has been used occasionally in cases where hypotension is severe (dose rate 5-10 mg in 500 ml saline per 500 kg horse to effect). Digoxin can be used if the heart rate becomes very high (>120) and should be administered if an atrial tachycardia develops. Digoxin reduces the rate of conduction through the AV node; however, it may not affect the heart rate as much as might be hoped because it will not reverse the effects of high sympathetic tone. It is important that animals which require intensive treatment can be treated without moving them to another box. Absolute rest is advisable because moving a hypotensive animal may cause it to collapse.

In a very limited number of cases a horse may appear particularly sensitive to the effects of the drug and sudden death during treatment is a potential hazard. Although this is very uncommon, it is wise to warn owners of the potential dangers of treatment.

Stomach tube problems Passage of a stomach tube may be difficult in some horses, particularly when repeated dosing is required, or if nasal oedema or haemorrhage develop. The stomach tube should always be cleaned thoroughly because the taste of quinidine will make the horse resent its use even more. Sedation should be avoided if at all possible. Phenothiazines (e.g. acepromazine) will reduce blood pressure and may make side-effects more severe. Alpha2 agonists (e.g. detomidine) will increase vagal tone, reduce myocardial conand may also decrease cardiac output in an animal in which it is already compromised. If necessary, the animal can be sedated to allow a tube to be passed and the tube left in place for treatment to commence once the effects of the sedative have worn off. The tube should be plugged to avoid aerophagia resulting in bloat; however, where possible the use of an indwelling tube should be avoided. It may be possible to administer a mixture of quinidine and molasses as a paste in unco-operative animals, but oral ulceration can occur because of the direct effect of the drug on the mucous membranes.

Contraindications for treatment It should be remembered that quinidine treatment is contraindicated in cases of CHF. The hypotensive and negatively inotropic effects of quinidine can be fatal to horses with CHF. Where there is a resting heart rate above 60 bpm, it is imperative to lower the heart rate with digoxin before quinidine is used. Few animals with a resting heart rate of >60 bpm are likely to be suitable for treatment even if the rate can be reduced. Because the long-term outlook for these animals is almost invariably poor, treatment is unlikely to be worthwhile. The exception to this rule would be animals with myocardial disease which may resolve. Even in this situation, a wiser course of action would be to control signs with the judicious dosage of digoxin and frusemide rather than risk the use of quinidine.

Intravenous use of quinidine gluconate for treatment of AF Where quinidine gluconate is available, it can be used to convert animals with AF to sinus rhythm. It appears to be most successful in horses which have been in AF for only a few days. The ease and speed of administration mean that it has some practical advantages compared to the use of quinidine sulphate. Side-effects are reported to be uncommon, provided that appropriate cases are selected.

Intravenous boluses of quinidine gluconate (0.5-1.0 mg/kg) are given at ten-minute intervals until conversion, or until a total dose of 10 mg/kg is reached.

Prevention of reversion to atrial fibrillation Most horses that have no underlying heart disease and which return to sinus rhythm during the initial titration period are unlikely to relapse into AF. However, there are steps which can be taken to determine which animals are in danger of reversion and which may need further managemental or therapeutic procedures.

It is important to consider conditions which might predispose to the develof AF at the time of treatment. The most common predisposing abnormality is thought to be the presence of an irritant focus in the atrial myowhich results in APCs. APCs may trigger AF if they occur at a critical time during repolarisation. They may be more common in animals that have a recent history of respiratory disease, possibly due to a myocarditis. APCs may be detected by auscultation or standard ECG analysis after successful treatment of AF; however, they are more likely to be found by monitoring cardiac rhythm using a Holter monitor at rest and by radiotelemetry during exercise. Animals with APCs should be rested for at least a month and may need treatment with digoxin and/or corticosteroids (see above). After this period the ECGs should be repeated to see if the frequency of the APCs has decreased to a level where a return to work is advisable. If Holter monitoring is not available, a history of recent respiratory disease may be sufficient indication to rest the horse for 1-2 months after treatment in order to decrease the risk of reversion to AF. Animals without evidence of APCs may be returned to light work approximately 3 days after treatment and to full training after 7-10 days.

Electrolyte imbalance, particularly hypokalaemia, may also predispose anito the development of AF. In animals in which AF recurs, electrolytes levels should be measured prior to repeat treatment and if necessary at intervals following treatment. Electrolyte analysis is also worthwhile in animals known or suspected to suffer from paroxysmal AF. Plasma or serum levels can be meahowever, there is some controversy about their value. Measurement of urinary fractional excretion of potassium can also be used to detect underlying potassium depletion (see section 2.2.6). Deficiencies can be treated by suppleof the diet with potassium.