NEUROREPORT

AUTONOMIC NERVOUS SYSTEM

Lateralization in autononic dysfunction in ischemic
stroke involving the insular cortex
S. Meyer, M. Strittmatter,1,CA C. Fischer,1 T. Georg2 and B. Schmitz3,4
Department of Neurology and Pediatrics; 2Institute for Medical Biometrics, Epidemiology, and Medical Informatics; 3Department of Neuroradiology,
University of the Saarland; 1Department of Neurology and Stroke Unit, SHG Kliniken Merzig, von-Fellenberg Stift, Trierer Strasse148, 66663 Merzig;
4
Department of Diagnostic Radiology, University Hospitals Ulm, Germany
CA

Corresponding Author: m.strittmatter@mzg.shg-kliniken.de
Received 9 October 2003; accepted14 October 2003
DOI: 10.1097/01.wnr.0000103755.04757.52

Autonomic nervous system dysfunction is a common complication
of ischemic stroke. Clinical and experimental data indicate hemispheric lateralization in the control of autonomic activity.The insular cortex has also been shown to play a crucial role in the central
autonomic network. The aim of this study was to assess cardioautonomic dysfunction in patients with ischemic insular versus
non-insular cortex infarction, and to demonstrate a possible lateralization in autonomic activity mediated by the insular cortex.
Sympathetic function was prospectively assessed by determining
plasma norepinephrine and epinephrine in 15 patients with lefthemisphere (LH; four insular infarction), and 14 with right-hemisphere (RH) middle cerebral artery (MCA) stroke (¢ve insular

infarction). Systolic and diastolic blood pressure and heart rate
were recorded during the ¢rst 5 days after stroke. Sympathetic
activity was signi¢cantly higher in insular than in non-insular infarction ( p o 0.05) with concomitantly elevated cardiovascular
parameters in insular stroke patients. The pathological activation
of the sympathetic nervous system was most excessive in
RH-stroke involving the insular cortex (p o 0.05).Our data indicate
a hemispheric lateralization in autonomic activity which is
mediated by the right-sided insular cortex. Patients with RH stroke
involving the insular cortex are most susceptible to develop cardioc 2004 Lippinautonomic dysfunction. NeuroReport 15:357^361
cott Williams & Wilkins.

Key words: Autonomic nervous system; Hemispheric lateralization; Insular cortex; Plasma catecholamines; Stroke

INTRODUCTION
There is a growing body of evidence from animal and
clinical studies that cerebrovascular disease alters cardiovascular and autonomic function [1]. Early mortality in
stroke patients is associated with a high incidence of cardiac
arrhythmias and myocardial damage [2,3]. Distinct cerebral
regions have been implicated in the genesis of the
pathological activation of the sympathetic nervous system
[4,5]. Evidence exists for cortical lateralization in the
regulation of cardiovascular functions, indicating that right
hemisphere (RH) ischemia has sympathetic consequences
greater than left hemisphere (LH) ischemia [6]. Unilateral
cerebral inactivation and stimulation causes hemisphericdependent changes in cardiovascular parameters suggesting
a suprabulbar right–left asymmetry in autonomic control
[7,8]. The location, i.e. insular infarction, and to a lesser
extent the size of stroke, may cause different autonomic
abnormalities in stroke patients [9]. Experimental data also
indicate that the insula, the amygdala, and lateral hypothalamus play a crucial role in the autonomic control of the
heart [10]. Of these, the insular cortex within the middle
cerebral artery (MCA) territory is the most important
cortical area engaged in cardiovascular regulation [11].
However, there are few data on the different effects of
LH vs RH insular cortex infarction on cardio-autonomic

c Lippincott Williams & Wilkins
0959- 4965

function. The aim of this study was to demonstrate the
differential effects of the stroke localization (insular (I) vs
non-insular (NI) infarction in LH and RH stroke) on the
autonomic nervous system, and on cardiovascular parameters in the acute post-ischemic phase.

SUBJECTS AND METHODS
Study population and entry criteria: This prospective
study was conducted according to the principles established
in the Helsinki declaration and after approval by our
institutional review board (IRB). After informed consent, 29
patients with ischemic stroke were included within 24 h of
the onset of symptoms. Fifteen patients (age 60.9 7 10.8
years) suffered from territorial LH stroke, 14 patients (age
64.0 7 10.8 years) from territorial RH stroke affecting the
MCA. In the LH and RH stroke patients, four and five
patients, respectively, suffered complete insular cortex
infarction. Cerebral infarction and localization were verified
by MRI (Siemens, Vision, 1.5 T, Erlangen, Germany) including diffusion-weighted sequences or cerebral computed
tomography (Twin Flash, Elscint, Haifa, Israel) including a
follow-up cerebral CT within 1 week. All brain scans were
reviewed by an experienced neuroradiologist blinded to the
other data.

Vol 15 No 2 9 February 2004

357

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Admission Admission + 4 h Day 2 Day 3 Day 4 Day 5 NI BPsys BPdia HR 163.0715. Data are given as means 7 s.4 93. The median time interval from symptom onset until inclusion into our study was comparable between the study groups (non-insular stroke 8.674.072.7 71.772. Unauthorized reproduction of this article is prohibited. Patients were instructed to keep their arm as motionless as possible.8 81. 1.6 76.474.2 87. No significant differences existed in the number of patients with known hypertension and/or hypertensive medication between non-insular and insular stroke patients prior to admission.0712.6712.273. 22.05 was considered significant.30 and 17.2 152.976.1a 79. patients with regular use of medications that alter autonomic function.3 71.776. Erlangen. Plasma NE and E concentrations were measured using high performance liquid chromatography (HPLC) with electrochemical detection as described previously [6].6727. Statistical comparison between the study groups was performed using the Kruskal–Wallis test with post-hoc Bonferroni testing.4 80. MEYER ETAL.6 167.6 163. 11.179.9 79. Due to the size of the study population.3 I BPsys BPdia HR 170.775.3 146.0713.) and right-hemispheric insular (RH-I) cortex stroke. NE declined significantly (p o 0.2 169.7 151.6 164.075.00 h.5 83. intracranial hemorrhage or symptoms of increased intracranial pressure.05).479.05. range 2.0723.30 h.174. Only in patients with non-insular cortex infarction there was a substantial drop in both BPsys and BPdia (p o 0.9 91. Germany).9 84. No significant differences were seen in E concentration in non-insular stroke affecting RH or LH.4 157.0 81. severe ischemic heart disease.679.8 80. 358 Vol 15 No 2 9 February 2004 Copyright © Lippincott Williams & Wilkins.9a 71. did not reach the level of significance.0711. Simultaneously to plasma catecholamine measurements.m.7 149. Siemens.05.7 80.0 77.5 82.172.8 155. pain.4–22. Table 2. fear and hunger can raise plasma catecholamine levels rapidly.975. 2). NE and E concentrations are given in ng/l.m.8 85.0 153. 4 h after admission. position.774. 2. cardiovascular parameters were recorded at 07.074. HR: Heart rate. or congestive heart failure (NYHA III-IV) were excluded.7 162.7 85. Fig.05. No significant differences in neurologic deficit according to the Scandinavian Stroke Scale Score (SSSS) existed between patients with non-insular (median SSSS 42) vs insular stroke (median SSSS 41) on admission.2711.1 h. In non-insular stroke.872.2711. p o 0. Since environmental conditions such as light.05.3 h. All study patients were synchronized to the same day.7 145. b signi¢cantly elevated heart rate in right-hemispheric insular stroke compared to left-hemispheric insular stroke.7 80.479. autonomic and cardiovascular parameters within each group were compared with the nonparametric Wilcoxon signed rank test. and Fig. and heart rate in patients with non-insular (NI) vs insular (I) cortex stroke (mean7s.8 159. a Di¡erences between BPsys and BPdia on admission vs BPsys and BPdia on day 5.5717. The results are summarized in Table 1. insularstroke 8. The most excessive sympathetic activation was seen in patients with RH stroke involving the insular cortex.5–21.NEUROREPORT S.2 79. p o 0. p o 0. Methods: Venous blood samples were drawn from an indwelling catheter for the determination of norepinephrine (NE) and epinephrine (E) on admission to the hospital.and night-time schedule (meals at 07. with the most prominent pathological activation in RH stroke with involvement of the insular cortex (p o 0.e.3 72.3 73.0737.772.3 176. and on days 2–5 at 07.3 RH-I BPsys BPdia HR 171.2713. Systolic and diastolic blood pressures (BPsys/BPdia) were determined with an automatic non-invasive BP meter (Sirecust 401. Cardiovascular parameters: Systolic (BPsys) and diastolic (BPdia) blood pressure were initially elevated in all study groups. Table 1).072.30.6717.3 77.672. with a past medical history of endocrine disease.0 h. Differences in BPsys and BPdia at the different time points.8 74.05. Table 2). BP cuffs were positioned on the unaffected arm ipsilateral to the infarct. sleep from approx.274.e.172.2b BPsys/BPdia: Blood pressure systolic/diastolic (mm Hg). Compared with non-insular cortex stroke. The highest BPsys and BPdia were seen in patients with RH-I stroke (Table 1).872. NE concentration was comparable between RH and LH stroke (Table 2). plasma E levels were also significantly higher in insular cortex stroke (Fig.172.7 79.2714.475.879. noise. In NI infarction. Patients with transient ischemic attacks.1 82. however. range 2. and non-insular cortex stroke.4 78.5 77.473. .772.4 80.9 79.7 79.9 78.972.9 h). blood sampling for the catecholamine assay was done under strictly standardized conditions with patients in bed in the supine position 30 min prior to blood sampling.00 h) thus further minimizing extraneous factors that might influence autonomic function.00 h until 06.6718. Plasma NE concentration was significantly higher in RHinsular stroke compared to LH-insular stroke and noninsular stroke patients (p o 0.0713.1 78. Systolic/diastolic blood pressure.175.00 h. Table 2). 1).174.276. RESULTS Autonomic parameters: NE and E: Plasma NE concentrations were significantly higher in patients with stroke involving the insular cortex compared to NI-cortex stroke (Fig.279.179. Table 1.472.0 74. in I-cortex infarction a sustained pathological augmentation in NE was seen.174.

1400.14]. Adm. while HR moderately increased during the 5-day study period in patients with insular stroke (Table 1).0 60.978. Di¡erences between the two groups. our study demonstrates a sustained autonomic dysfunction which prevailed over the first 5 days after stroke. myocardial necrosis and even sudden death [1–3].13]. + 4 h: 4 h after admission to the hospital.00 180.0 Non-insular stroke Insular stroke 1200.: Admission to the hospital.7 * Signi¢cantly elevated plasma NE and E levels in right-hemispheric insular stroke compared to left-hemispheric insular stroke.4 37.179.7 NI-RH NE E 422.0770.77209.576.6 1206.01). 2. insular cortex stroke.0 Adm.1 55. Fig. 1. and to a lesser extent the hemispheric laterality (right vs left) and size of stroke.17138.7759.05. Admission Admission + 4 h Day 2 Day 3 Day 4 Day 5 NI-LH NE E 467.8724.57180.7788.8746.57491.9726.8 285. Adm.1 32.m.00 600.0 00.7757.67313.5712. Adm. Unauthorized reproduction of this article is prohibited.0 306.576.3712.01.9 42.075.27223.0 80.2796.05.+4h Day 2 Day 3 Day 4 Day 5 0.8749. 2).4 171.8 997.6 322.6 318. wwp o 0. Contrary to the results from Sander and Klingelhofer [13]. and non insular cortex stroke.0 32. phasic Vol 15 No 2 9 February 2004 3 59 Copyright © Lippincott Williams & Wilkins.05.13].4 943. These complications are attributed to damage of the central autonomic network.7 425.05.2 37.6 40.7 27. Stimulation of the left and right insular cortices elicits specific cardiovascular responses in epileptic patients. wp o 0.3 132. p o 0.8710. may cause autonomic abnormalities [9.). Heart rate was significantly higher in RH-I stroke than in LH-I stroke and patients with non-insular infarction (p o 0.4 262.3 40. Plasma norepinephrine concentration (NE) in non-insular vs insular cortex stroke.1 355.1 47.9711.6732.3* 94.477.97318.77116.e.00 † 100. 1.00 20. p o 0.6* 99.1 284.0 † †† 120.5727.5 110.00 Adm. Cerebral inactivation in patients with epilepsy also suggests lateralization in control of cardiac and autonomic function with a right-sided cerebral predominance for sympathetic function [7]. stimulation of the insula entrained to the cardiac cycle is capable of inducing severe arrhythmias in rats [8.2711.3 418.00 * 400.3798.0720.2 36.4711.NEUROREPORT AUTONOMIC LATERALIZATION IN INSULAR STROKE Table 2.00 †† Non-insular stroke Insular stroke 160.00 E (ng/l) NE (ng/l) 1000.77210.0 40. DISCUSSION Several investigations in humans have shown that stroke induces changes in cardiac autonomic control which might cause hypertension.6 890.6 I-RH NE E 1295.4 41.5 361.2 47.0717. Plasma epinephrine concentration (E) in non-insular vs.5 585.7747.0733.97382. Adm + 4 h: 4 h after admission to the hospital.87106.6 I-LH NE E 498.6 45.05.0 200. We demonstrated distinct alterations in plasma catecholamine levels following cerebral ischemic infarction.6745.776.6720.5713.6 255.1 38.1 38. Plasma norepinephrine (NE) and epinephrine (E) in non-insular (NI) and insular (I) stroke dependent on left-hemispheric (LH) and right-hemispheric (RH) localization (mean7s.8 278.: Admission to the hospital. Table 1). wwp o 0. most strikingly in RH insular cortex stroke. Adm.8 74. Clinical trials have demonstrated that insular infarction. Di¡erences between the two groups: wp o 0. with a pronounced increase in sympathetic autonomic function in stroke patients with involvment of the insular cortex (Fig.3* 951.9738. arrhythmias.0 †† 140.676. severe and excessive disturbances in autonomic function were seen in RH insular cortex infarction while in LH insular cortex infarction only a moderate elevation in sympathetic autonomic function was noted in our study population.0 800.0 368.01.+4h Day 2 Day 3 Day 4 Day 5 Fig.3747.9729.1 175. . Fig.1 324. *Di¡erences between NE on admission vs NE on day 5 in noninsular stroke.7754. Adm. In patients with non-insular stroke a decrease in HR was seen (p o 0. In addition to the data presented by Sander and Klingelhofer [9.7 42.

Topcuoglu MA. 9. 697–702 (1992). most notably in RH-I stroke (Table 1). either a direct increase in sympathetic drive or a decrease in parasympathetic function with a reciprocal rise in sympathetic tone. Oppenheimer SM. Thus the possible adverse effects of sympathovagal shifts as a possible progenitor of cardiac arrhythmogenesis reported by many observers [2] may be offset by an increase in penumbra viability.25]. Up-regulation in autonomic function was paralleled by sustained elevation of cardiovascular parameters in insular cortex stroke patients. 66–72 (1990). Morillo CA. Sander D and Klingelhofer J. 741–744 (1997). Cerebral hemispheric lateralization in cardiac autonomic control. 183–200. Stroke 30. 16. Oppenheimer S and Norris JW. Subcortical sites mediating sympathetic responses from insular cortex in rats. REFERENCES 1. 12. Saribas O. Int J Cardiol 58. NY: Churchill Livingstone. Gelb A. Brain Res 569. Klingelhoefer J and Sander D. 5. Oppenheimer SM. New York. indicate that elevated plasma norepinephrine 360 S. while atropine did not increase HR in bradycardia [15]. Effects of stroke localization on cardiac autonomic balance and sudden death. the consistent data over a 5-day period with six measurements of autonomic function strongly indicate that the right insular cortex plays a crucial role in the initiation and promotion of sympathetic dysfunction in ischemic stroke. 6. Thus ischemic damage to this region would result in unleashing sympathetic autonomic tone with a concomitant increase in cardiovascular parameters. Identification of a cortical site of stress-induced cardiovascular dysfunction. despite the relatively small number of patients included in this study. Batur MK. J Neurol 242. Hachinski C. 1307–1311 (1999). 2. Girvin JP and Hachinski VC. The precise underlying pathophysiological mechanism. most apparently in RH insular stroke. 381–385 (1995). Cardiac manifestations of acute neurogical lesions. Up-regulation of cardio-autonomic function is either mediated directly via modulation of the sympathetic activity or by reduced parasympathetic activity with a reciprocal shift in sympathetic function. a reflex mainly under parasympathetic control. The vital role of the rightsided insular cortex in autonomic control is further corroborated by the fact that very similar physiological patterns in both NE and E concentration were seen in noninsular RH and LH stroke. This hypothesis is corroborated by the fact that RH stroke in humans is associated with a reduced respiratory HR variability. 3. This might be of prognostic significance since positron emission tomography studies have shown a reduced hemodynamic reserve in hypertensive patients which renders the penumbra vulnerable even to a small reduction in BP [22]. Oppenheimer and co-workers demonstrated that bradycardia elicited from stimulation of the caudo-posterior area of the insula was abolished by atropine [16]. 30–38 (2003). Changes of circadian blood pressure patterns and cardiovascular parameters indicate lateralization of sympathetic activation following hemispheric brain infarction. The cardiac chronotropic organization of the insular cortex. Cardiovascular effects of human insular cortex stimulation. 2nd edn. Two possible pathophysiological models could explain our results: RH insular cortex primarily controls and inhibits sympathetic function. Oppenheimer SM. We showed different patterns in BPsys and BPdia and heart rate with sustained hypertensive values in insular cortex infarction compared to non-insular infarction. Our data are not conclusive as to whether disturbances in cardio-autonomic tone are caused by a direct activation of the sympathetic nervous system or by reduced parasympathetic function. Wang TD. cerebral perfusion in acute ischemic states largely depends on systemic BP [19]. 133–142 (1992).). Cechetto DF and Chen SJ. Ballie`res Clin Neurol 6. Guiraudon C and Cechetto DF. Stroke-associated pathological sympathetic activation related to size of infarction and extent of insular damage. Saleh T and Cechetto DF. Wilson JX. . The differential activation of the autonomic nervous system cannot be attributed to different insular stroke size since both in RH and LH insular stroke patients the complete insular cortex was involved. Oppenheimer and Cechetto showed that the increase in HR in cerebral ischemia was abolished by the b1-antagonist atenolol. since initially high BP can prevent the early progresssion of stroke. Vol 15 No 2 9 February 2004 Copyright © Lippincott Williams & Wilkins. Neurology and General Medicine: The Neurological Aspects of Medical Disorder. MEYER ETAL. concentration is an independent predictor of poor longterm outcome in thromboembolic stroke [23]. 309–335 (1997). Cechetto DF. Unauthorized reproduction of this article is prohibited. 7. We conclude that right-sided involvement of the insular cortex should be taken into account in the management of stroke because of an excessive pathological activation of the sympathetic nervous system. Yoon BW. 15.NEUROREPORT Thus. 115–121 (1991). 308–311 (1997). 1995. Kes S and Oto A. Saleh TM. An endogenous up-regulation should have a positive effect on the clinical course of stroke. Since cerebrovascular autoregulation is impaired in stroke. remains to be elucidated. Eur Neurol 50. 14. In a second study. 11. Sander D and Klingelhofer J. Lateral hypothalamic area neurotransmission and neuromodulation of the specific cardiac effects of insular cortex stimulation. Am J Physiol 258. 8. Alternatively. 1727–1732 (1992). Guiraudon C and Cechetto D. Tokgozoglu SL. Integr Physiol Behav Sci 29. Locationdependent patterns in cardio-autonomic dysfunction in ischemic stroke. 313–318 (1995). however. Wu CC and Lee YT. Myocardial stunning after cerebral infarction. Arch Neurol 54. Wilson JX and Cechetto DF. 362–373 (1994).21]. The rise in sympathetic function. Cardiovascular consequences of clinical stroke. Plasma and organ catecholamine levels following stimulation of the rat insular cortex. might increase penumbra viability. Asymmetry of sympathetic consequences of experimental stroke. Brain Res 533. Oppenheimer SM. Fischer C. Neurology 42. 13. Georg T and Schmitz B. 4. CONCLUSION Our investigation shows that RH insular cortex stroke causes an excessive and sustained increase in cardiosympathetic tone. R245–R255 (1990). the role of plasma catecholamines on neurological outcome in stroke remains controversial [24. Arch Neurol 49. Insular cortex stimulation produces lethal cardiac arrhythmias: a mechanism of sudden death? Brain Res 550. Wilson JX. a corresponding asymmetry with respect to sympathetic cardiovascular reflexes could not be found [18]. Oppenheimer SM and Cechetto DF. whereas the lowering of BP to normal values worsens the clinical outcome [20. RH could have a greater role in cerebral regulation of cardiac function by virtue of the modification of parasympathetic effects as suggested by Ahern and co-workers [17]. In: Aminoff MJ (ed. pp. Brain Res 581. Recent data. 10. Strittmatter M. via augmented cardiovascular parameters. Experimental studies have so far yielded contradictory data. However. Cechetto DF and Hachinski V. Oppenheimer SM. This is indicative of hemispheric lateralization of the central autonomic nervous system mediated by the right insular cortex. Cerebrovasc Dis 5. Meyer S. 221–228 (1992).

460–464 (1991). Wieloch TW and Siesjoe BK. J Neurosurg 73. Stroke 21. Should hypertension be treated after acute stroke? A randomized controlled trial using photon emission computed tomography. Munich. Ichiya Y and Fujishima M. Sadoshima S. 156–159 (1994). Fujii K. Stroke 27. Anesth Analg 73. Jorgensen HS. Raaschou HO and Olsen TS. Lane RD. Naver HK. 20. 18. Ahern GL. Baughman VL and Albrecht RF. Sander D. 21. Bouma GJ and Muizelaar JP.Germany for his critical review of the manuscript. 283– 290 (1990). 368–374 (1990). 855–862 (1993). 23. Clonidine decreases plasma catecholamines and improves outcome from incomplete ischemia in the rat. 559–565 (1986). Vol 15 No 2 9 February 2004 3 61 Copyright © Lippincott Williams & Wilkins. Tran HD. 833–838 (2001). Prognostic relevance of pathological sympathetic activation after acute thromboembolic stroke. Kuwabara Y. Blomstrand C and Wallin G. Arch Neurol 50. . 24. 912–921 (2001). J Cerebr Blood Flow Metab 6. Herring AM. Klinikum Grosshadern. Cheng MA. Winbeck K. 25. Thomas C. Unauthorized reproduction of this article is prohibited.AUTONOMIC LATERALIZATION IN INSULAR STROKE 17. NEUROREPORT 22. Neurology 11. Epilepsia 42. Labiner DM. Etgen T and Conrad B. Grotta JC. Koide T. Cerebral blood flow and metabolism in normotensive and hypertensive patients with transient neurologic deficits. Weinand ME. Taylor JW. Lancet 344. Reduced heart rate variability after right-sided stroke. 247–251 (1996). Circulating catecholamines modulate ischemic brain damage. Effect of blood pressure and diabetes on stroke in progression. Nakayama H. Yao H. Okada Y. Heart rate and heart rate variability changes in the intracarotid sodium amobarbital test. 19. Sollers JL.F. Relationship between cardiac output and cerebral blood flow in patients with intact and impaired autoregulation. Klingelhofer J. Hoffman WE. Hutzler R and Thayer JF. Hamann. Acknowledgements: We thank Professor Dr G. Molony DA and Barron BJ. Ludwig-Maximilian University. Lisk DR.