Triple tachycardia☆☆☆

The patient was a 70-year-old woman with a history of congestive heart failure (left ventricular ejection fraction = 16%) and chronic obstructive pulmonary disease (on home oxygen). Her medication included 0.125 mg of digoxin daily and 150 mg roxithromycin twice a day from an episode of left lower lobe pneumonia 2 weeks earlier. On day of admission she was found weak and obtunded at home; the ambulance staff recorded a systolic blood pressure 95 mmHg and pulse rate 120 to 140 beats per minute. There were no focal neurologic signs. She responded to intravenous fluids and oxygen by mask and became more alert by the time she reached the emergency room. She remained disorientated but denied any cardiorespiratory symptoms. She was afebrile 36.3°C. Physical examination showed pulse rate 158 beats per minute and the blood pressure 144/100 mmHg; scattered rhonchi where audible over both lungs except for the left base, which was silent. Pulse oxymetry showed 98% saturation on 2L per minute oxygen flowing through nasal prongs. A chest radiograph showed left basal collapse noted one month previously. She was thin, weighing 40 kg (88 lbs); the musculoskeletal system was normal apart from moderate generalized weakness. The initial 12-lead electrocardiogram is shown in Fig 1.

A few minutes later, the tachycardia changed spontaneously (Fig 2).

Serum potassium was 3.2 mmol/L, blood urea nitrogen was 26.9 mmol/L, and creatinine level was 159 mmol/L (normal range = 3.5-5.0, 2.5-7.5, and 40-90, respectively). Serum digoxin was 3.4 nmol/L (therapeutic range = 1.0-2.6). The cardiac enzymes and routine blood tests were unremarkable. A coronary angiogram (for possible angioplasty in view of the marked anteroseptal ST-segment elevation) was virtually normal. She was transferred to the coronary care unit, treated with supplemental potassium, and recovered.

Fig 1 shows classic bidirectional ventricular tachycardia with alternating marked right and left axis deviation in the standard leads and right bundle branch-block-like complexes in the precordial leads. The QRS duration is 0.10 to 0.12 seconds, denoting its fascicular provenance. As is commonly the case, there is a slight but definite alternation of the cycle length (type II bidirectional tachycardia in the Scherf and Kisch1 classification). It has been strongly associated with digitalis toxicity since its original description in 1922.2 Variant forms, not necessarily related to digoxin therapy, have been described since then.3, 4

Atrial activity is best seen in lead V1 as sharp positive monophasic spikes; the rate is almost identical to that of the ventricular tachycardia even though the 2 are, at times, dissociated. At the beginning of the recording no atrial waves are visible until after the sixth QRS; this is because they are hidden by the ventricular complexes. The latter vary in shape and duration; a QS complex in V1 is quite narrow at 0.08 seconds. This is a sign of fusion (a proof, inter alia, of the ventricular origin of bidirectional tachycardia). It is also a sign of possible conduction of the atrial tachycardia to the ventricles.

Fig 2 shows the extent of that conduction, which is probably none. Some cycles are shorter than others, indicating possible captures; the longer cycles could involve a blocked P wave. The putative would be block is Wenckebach in type, involving “over the top” conduction.5 However, even then there is no true Wenckebach periodicity; some of the PR intervals actually become shorter before the supposed block. The simplest explanation is a completely dissociated accelerated junctional rhythm of 83 beats per minute, which is dissociated because of complete AV nodal block.

The second QRS complex is a fusion beat, but it does not result from a transmitted impulse of atrial tachycardia. Its supraventricular component is a complex from the accelerated junctional rhythm. It can now be seen why some of the tachycardia complexes are so narrow: a right bundle-branch block-like complex is fusing with an incomplete left bundle-branch block complex. As Schamroth states—two wrongs making a right. Whatever the ventricular mechanism, the atrial activity constitutes an atrial tachycardia with block, the “PAT with block” of old—another classic manifestation of digoxin toxicity.6

The original rhythm is double tachycardia, also first described in 1922.7 This, too, is most commonly seen in digoxin toxicity. Roxithromycin, like other macrolide antibiotics, increases digoxin absorption from the gut. Cases similar to this one (PAT with block combined with bidirectional tachycardia) are described in the comprehensive review by Castellanos et al8 in 1960. This example is even more complex because it is a form of triple tachycardia. The junctional rhythm, through the fusion beats, contributes to it.