|Home | About | Journals | Submit | Contact Us | Français|
The insular cortex is known to be involved in central sympathovagal regulation. Both right and left insular lesions have been associated with cardiac derangements.
Previous stroke studies suggest that right insular lesions may cause reduced inhibition of sympathetic activity resulting in tachyarrhythmias, ECG changes, elevation of troponin T (TnT) and serum catecholamine levels, decreased heart rate variability, higher occurrence of sudden death and overall death rate within 3 months.
In contrast, recent data indicate a significant role of the left insula in mediating the parasympathetic tone, and therefore lesions in this region may be associated with a shifted balance towards sympathetic activity.1 In patients surgically treated for epilepsy, direct stimulation of the right insular cortex mainly resulted in sympathetic cardiovascular responses whereas stimulation of the left insula mainly caused parasympathetic reactions (bradycardia).2 This is in agreement with the observation that barbiturate infusion into the human left carotid artery produced tachycardia whereas infusion into the right carotid artery generated bradycardia.3
Left insular lesions may be associated with decreased parasympathetic tone, shifting the balance towards enhanced sympathetic outflow and decreased heart rate variability.1 Left insular strokes have been associated with adverse cardiac outcome, suggesting this mechanism.
Nevertheless, it is possible that both types of responses can be evoked from either insula. In the rat, prolonged left insular stimulation results in both cardiac sympathetic and parasympathetic activation.4 Cortical lateralisation of cardiac autonomic regulation has been suggested but the data are contradictory.
We report the case of a 51‐year‐old man with right insular haemorrhage who suffered from repetitive asystoles during his stay in our neurological intensive care unit (NICU). The patient had a history of treated arterial hypertension and presented with sudden onset left sided hemiplegia and somnolence. Cranial CT showed a right insular haemorrhage with intraventricular blood (approximate volume 22 ml). CT angiography did not show any arteriovenous malformation or aneurysm. The patient was treated within the 4 h window with activated recombinant factor VIIa (rFVIIa) and phenytoin because of left sided partial seizures. Because of the developing obstructive hydrocephalus, the patient was intubated, mechanically ventilated and an external ventricular drainage was inserted. The ventilator associated pneumonia was treated with ciprofloxacin and piperacillin with tazobactam, which were later changed to meropenem. Within the first 9 days, phenytoin, a beta blocker and propofol were administrated without any complications for 8, 6 and 4 days, respectively.
After reduction of analgosedation on day 9, the patient suddenly developed asystole requiring cardiopulmonary resuscitation. Episodes of sudden asystoles without preceding bradycardia recurred up to five times with a duration of 10–50 s. At the time of the first asystole, the patient was on a low dose of beta blocker, phenytoin and propofol. These drugs were discontinued and phenytoin was substituted for levetiracetam. All asystole events were managed successfully with 0.5–1 mg atropine. A transthoracic cardiac pacemaker was placed temporarily. The observation was made that the asystoles were triggered by vagal stimulation (eg, the patient breathing against the ventilator, manipulation of the endotracheal tube or suctioning). The next day, the bradycardiac episodes ceased by minimal handling, deep analgosedation and after placement of a tracheotomy. No temporal relation of asystoles with any newly started medication was observed. Subsequent cardiologic workup, including repeat ECGs, serum TnT and creatine kinase‐MB levels and transthoracic echocardiography, revealed no cardiac disease. On day 12, transient depression of the ST segment as well as elevation of cardiac TnT to 0.04 μl/l was noted. By day 13, the ST segment and cardiac TnT level had normalised.
To our knowledge a case of asystole in a patient with right insular haemorrhage has not yet been reported. We have found one case reporting asystole in acute right middle cerebral artery (MCA) stroke involving the territory of the right insular cortex.5 We hypothesise that the asystole resulted from the right insular haemorrhage which induced a shift in autonomic balance with disinhibition of the left insular parasympathetic influence. Triggered by vagal stimulation, the parasympathetic outflow imbalance may have been responsible for the cardioinhibitory effects and asystole.
Another phenomenon observed in this patient was ST depression and TnT elevation. Observations made in physiological animal models suggest that stimulation of the left insula may result in ST depression, QT prolongation, pronounced bradycardia and rhythms ending in asystole.4 Interestingly, these changes were associated with elevated noradrenaline levels suggesting that left insular activation may result in both parasympathetic and sympathetic responses.4 Similarly, we assume that our patient had a complex autonomic imbalance favouring left insular effects, which may be interpreted as both parasympathetic and sympathetic dysregulation. Interestingly, an important delay between stroke onset and the development of asystole and/or ST depression and TnT elevation was observed. This agrees with the observations of Pasquini6 that autonomic changes associated with insular damage in acute stroke may have a delayed appearance.
It could be argued that we cannot completely exclude the possible prothrombotic effect of rFVIIa. However, the dynamics of ST segment changes and TnT levels were not characteristic of a possible cardiac origin. Furthermore, the drug is unlikely to have been the cause of the ST and TnT changes because of the short half life of rFVIIa. Phenytoin, beta‐blockers and propofol have the potential to cause bradyarrhythmias and in combination may have contributed to generating the asystoles. However, there was no temporal association between the introduction of medications and the asystole events. Moreover, these drugs were administered continuously for several days without any complications. There was no electrolyte imbalance or any other condition favouring cardiac rhythm disturbances. Intracranial pressure remained stable (<10 mm Hg) for the whole observation period.
In the context of previous data, the coexistence of the insular cortex lesion and the cardiac derangements is highly suggestive of a causal association.
Our case and the results of previous studies demonstrate the ambiguity of central autonomic disturbances mediated by lesions affecting the central autonomic network, in particular the insular cortex and connections between the left and right insula. However, the close association of insular lesions with cardiac derangements is strong, and intensive monitoring of patients with this type of lesion is necessary.
Competing interests: None.