Arrhythmias occur in the neuro-ICU in patients with known cardiac disease as well as in those without apparent cardiac risk factors. Again, many neurocritical care patients have cardiac disease, and some of these disturbances are attributable to coronary artery insufficiency, whereas others have conduction disturbances that result from the neurological illness. In addition, a fair number of patients manifest atrial arrhythmias (mainly atrial fibrillation) while under observation in the first few days after stroke. The implication of this finding is generally held to be that there was a similar previous arrhythmia that created the substrate for the (embolic) stroke, but it may be that some rhythm disturbances arise as a result of infarction of brain areas that modulate cardiac conduction (see Tachyarrhythmias).
Animal models of cardiac changes in relation to acute brain injury have demonstrated some cerebral localization of these effects (
1). Acute focal injuries to brain tissue, ischemic stroke, hemorrhage, and trauma lead to cardiac effects, as well as electrical activation of brain tissue as seen in epilepsy. Tachycardia and pressor responses are more common after stimulation of the right insular cortex and stimulation of the left vagus, which innervates the atrioventricular node and cardiac conduction system. Bradycardia seems to be more common after stimulation of the left insular cortex or the right vagus nerve, which innervates the sinoatrial node (
2). Tachycardias are most commonly seen with seizure and epilepsy, although bradycardias and sinus arrest have been reported (
3).
Bradycardias
Excepting the effects of β-adrenergic blocking drugs, a heart rate below 60 beats per minute (BPM) is usually the result of sinus node dysfunction or an atrioventricular conduction disturbance. Very acute cerebral diseases can also produce a vasovagal response that is pronounced enough to cause the heart rate to drop and blood pressure to fall to the point of causing syncope. Closed head injury is one setting that leads to bradycardia and conduction block, likely through a vagal mechanism (
5). Whether similar mechanisms pertain in SAH and epilepsy is not known, although similar bradycardia is seen. Vagal influences inhibit sinus nodal activity but a sympathetic discharge of medullary origin probably accounts for the concurrent vasodepressor effect. When the vasodepressor component is prominent, even cardiac pacing may not eliminate hypotension but vasopressor drugs may reverse it. Atropine can be helpful in the reversing the bradycardic component in acute setting but, in extreme and protracted cases, external pacing may be necessary for short periods of time.
Some patients have similar responses after carotid angioplasty and stenting procedures (
6). Stimulation of the carotid sinus by the angioplasty balloon can lead to profound bradycardia, including complete heart block. Pacing may be required for a brief period, although pretreatment with an anticholinergic usually is adequate. The stenting procedure may cause a more sustained stimulation of the carotid sinus, in which case hypotension and bradycardia may persist for 24 to 48 hours. Some of these patients require vasopressor therapy to maintain adequate perfusion (
7). The vasovagal response is obviated during and after carotid endarterectomy by regional blockade of the carotid sinus, but other problems arise in the postoperative period, mainly hypertension.
Finally, bradycardia in a neuro-ICU raises concern as a possible indication of increased intracranial pressure. The Cushing reflex (bradycardia, hypertension, and respiratory depression) (
Chapter 2) results from acutely increased intracranial pressure and diminished cerebral perfusion. Animal models demonstrate increases in circulating catecholamines as the probable cause of hypertension, possibly as a compensatory effect to increase cerebral perfusion (
8,
9). The bradycardia appears to be the result of pressure that is transmitted to mechanically sensitive centers in the floor of the fourth ventricle. However, it should be pointed out that in clinical circumstances tachycardia is as often observed as bradycardia with episodes of ICP elevation.
Tachycardias
Supraventricular tachycardias are common in all critical care settings. Sinus tachycardia is defined here as a heart rate exceeding 100 BPM, and usually represents a physiologic response to pain, stress, hypotension, congestive heart failure, or excessive catecholamine drive. It is generally not treated as a primary dysrhythmia, but rather the precipitating conditions are sought and corrected.
Paroxysmal supraventricular tachycardias (PSVT) related to a reentry or similar mechanism at the atrioventricular node are also common. Treatment with vagal stimulation maneuvers, especially carotid sinus massage, are useful. Low doses of verapamil, 2.5 to 10 mg, by intravenous (i.v.) injection or of adenosine, 6 to 12 mg, also can be used to “break” the tachycardia. Adenosine has a very short half-life, and in our experience is quite effective in this setting. β-Blockade reduces the risk of recurrent episodes of PSVT but there are a number of alternative approaches.
Atrial fibrillation with a rapid ventricular response is also very common in relation to neurological disease, particularly in older age groups. In the neuro-ICU, pain, infections, intravascular volume overload, and neurological injuries such as SAH are the most common precipitants. The rapid ventricular response is usually the most urgent issue, because the high rate may precipitate coronary ischemia or compromise cardiac function. Short-acting
β-adrenergic blockers such as labetalol and esmolol can be used by i.v. bolus or continuous infusion. Diltiazem also is quite effective as an infusion that can be titrated to achieve rate control and is the approach we have preferred. The major limitation of β-blockade and calcium channel antagonists in the neurological patients relates to the induced hypotension that often accompanies doses adequate to slow rate. There is also some concern that certain calcium antagonists may increase intracranial pressure; however, the clinical effects have been minor in our experience. When hypotension must be avoided in order to maintain tenuous cerebral perfusion (e.g., soon after an ischemic stroke), treatment with digoxin may be preferable. The effects of digoxin, of course, are slower, requiring several hours, and may not always be effective, but the lack of associated hypotension makes it valuable in care of the neurological patient.
Ventricular tachycardias are less frequent than are atrial ones in the neuro-ICU, but their implications and potential for severe hypotension underscores their importance. Isolated premature ventricular contractions (PVCs) are seen commonly in many monitored patients, and rarely require treatment. The use of dopamine as a vasopressor agent may be associated with more frequent PVCs and at some point the ventricular irritability mandates the use of alternate agents. The Lown classification (
Table 5.2) has long been used to grade ventricular ectopy in terms of potential morbidity and risk of cardiac sudden death. Although much more sophisticated information can now be gathered by electrophysiologic studies, this classification still provides a simple and useful guide (
10,
11). Low-grade ectopy (grades I and II), usually are benign; treatment usually is considered in grade III and IV arrhythmias.
Ventricular flutter and fibrillation are most commonly seen in patients with underlying ischemic heart disease. Prolonged Q-T intervals are a risk for ventricular ectopy. This circumstance arises with hypokalemia and in relation to some medications, including phenothiazines and tricyclic antidepressants. The ventricular tachycardias are emergent problems for which management protocols are essential.
A specific type of ventricular tachycardia, torsade de pointes, is also related to Q-T prolongation and deserves special mention because it has been reported in relation to acute neurological injuries such as SAH and cranial trauma (
12,
13). Repletion of potassium and magnesium is critical and low-dose β-blockade may be helpful. Careful monitoring and support are essential until the arrhythmia ceases and the Q-T interval shortens.