- Email: [email protected]
Focal seizure and cerebral contrast retention after cardiac catheterization
Focal Seizure and Cerebral Contrast Retention After Cardiac Catheterization Richard E. Frye, MD, PhD*, Jane W. Newburger, MD†, Alan Nugent, MBBS†, and Mustafa Sahin, MD, PhD* Although rare, ionic and nonionic contrast-induced seizures occur as a complication of enhanced cranial computed tomography for both adults and children. However, contrast-induced seizures after cardiac catheterization has only been reported in adults. This report describes an 18-month-old male who developed a new-onset focal seizure 12 hours after cardiac catheterization. Unenhanced cranial computed tomography 1 hour after the seizure demonstrated general cerebral edema and unilateral focal cerebral contrast retention with sparing of the area supplied by the middle cerebral artery. The contrast was reabsorbed from the subarachnoid space over a 48-hour period, the cerebral edema resolved over several days, and the child returned to his baseline state 4 days after the seizure episode. This study documents the evolution of computed tomographic findings after contrast-induced seizures in a child. Contrast toxicity should be considered in any case of a new-onset neurologic deficit arising after angiography or enhanced computed tomography. © 2005 by Elsevier Inc. All rights reserved. Frye RE, Newburger JW, Nugent A, Sahin M. Focal seizure and cerebral contrast retention after cardiac catheterization. Pediatr Neurol 2005;32:213-216.
From the Departments of *Neurology and †Cardiology, Children’s Hospital Boston, Boston, Massachusetts.
© 2005 by Elsevier Inc. All rights reserved. doi:10.1016/j.pediatrneurol.2004.07.012 ● 0887-8994/05/$—see front matter
Introduction Radiologic contrast is commonly used to visualize vessel and organ lumens. Allergic reaction is, by far, the most common complication associated with contrast media, although the incidence has decreased with the transition from ionic to nonionic contrast media. Seizures are the most common neurologic complication of radiologic contrast media. In the largest case series, seizures complicating contrast-enhanced cranial computerized tomography were more common in patients with a seizure history or intracranial lesion [1]. Brief, self-resolving, generalized clonictonic seizures were most common regardless of seizure history or lesion focality [1]. However, convulsive and nonconvulsive status epilepticus has been reported in adults with intracranial lesions after contrast-enhanced cranial computed tomography [2,3]. Neurologic complications may occur in up to 1% of children after cardiac catheterization although a recent estimate suggests that this complication rate has improved [4,5]. The most common reversible neurologic complication after cardiac catheterization is global amnesia or cortical blindness for adults [6,7] and seizure or hemiplegia for children [4,5]. Although three cases of seizures were associated with cerebral contrast retention in adults [8-10], such an association has not been described in children. This report describes a child with complex congenital heart disease who developed partial seizures, cerebral contrast retention, and cortical edema after cardiac angiography with arterial coil occlusion. The series of cranial computed tomographic scans presented help define the temporal dynamics of subarachnoid and intraparenchymal cerebral contrast clearance and resolution of cerebral edema. Case Report An 18-month-old male with complex congenital heart disease with heterotaxy and single right ventricle underwent 3 hours of cardiac catheterization (Fig 1). At 5 months of age, the child had undergone right Blalock-Taussig shunt placement, central pulmonary artery reconstitution for discontinuous pulmonary arteries, and total anomalous pulmonary venous return repair. During the current procedure, 7 cc/kg of ioversol, a low-osmolar, nonionic contrast agent, was used for evaluation of the cardiopulmonary anatomy. Coil occlusion of the right internal mammary artery and an additional aortopulmonary collateral from the right subclavian artery was also performed. Although injection of
Communications should be addressed to: Dr. Sahin; Department of Neurology; Children’s Hospital Boston; 300 Longwood Avenue; Enders 250; Boston, MA 02115. Received January 12, 2004; accepted July 26, 2004.
Frye et al: Focal Contrast Retention 213
Figure 1. Diagram depicting the corrected cardiopulmonary anatomy of the patient. Patient had heterotaxy and single right ventricle and a common atrioventricular valve. There was a modified Blalock-Taussig shunt placed between the right innominate artery (RIA) to pulmonary artery and an interposition graft placed between the left and right pulmonary arteries. During the angiogram, contrast was injected into the right subclavian artery, pulmonary veins/left atrium, interposition shunt, and pulmonary arteries. SVC ⫽ superior vena cava; LPA ⫽ left pulmonary artery; RPA ⫽ right pulmonary artery; LCCA ⫽ left common carotid artery; RCCA ⫽ right common carotid artery; LSCA ⫽ left subclavian artery; RSCA ⫽ right subclavian artery; APC ⫽ aortopulmonary collateral. AV ⫽ atrioventricular.
contrast into the pulmonary vasculature and cardiac chambers did not differentially fill the right or left carotid circulation, the right common carotid artery was undoubtedly selectively filled with contrast when the Blalock-Taussig shunt and collaterals arising from the right subclavian artery were selectively studied. Vital signs remained stable (blood pressure: 76/51 to 132/77 mm Hg; heart rate: 95 to 152 beats/min; oxygen saturation: 66-80%, baseline oxygen saturation: middle 70s). Precatheterization hematocrit was 52.7, and 10 cc/kg of packed red blood cells was transfused during the procedure because of a 5 cc/kg estimated blood loss. Heparin was infused during the procedure and continued postprocedure. After the procedure, the child was holding and drinking from his own cup, interacting with his parents using his baseline language skills, and moving all extremities equally. At approximately 12 hours after the catheterization, two brief (each less than 3 minutes) episodes of left clonic arm movements, followed by lethargy, were observed. Intravenous lorazepam stopped the second episode. The heparin infusion was discontinued, electrolytes and blood glucose were measured as normal, and complete blood count revealed an 11-point drop in hematocrit. Skull and chest x-rays verified stable occlusion coil positions. An unenhanced cranial computed tomography obtained 1 hour after the seizure demonstrated right holohemispheric parenchymal and subarachnoid hyperdensity with relative sparing of the parenchyma in the distribution of the middle cerebral artery, and mild generalized bilateral diffuse cerebral edema (Fig 2a).
214
PEDIATRIC NEUROLOGY
Vol. 32 No. 3
Upon transfer to the intensive care unit, clonic movements of the left leg developed and responded to another bolus of lorazepam. Phenobarbital 20 mg/kg was administered as an intravenous bolus, and a maintenance dose (3 mg/kg/day) was begun for seizure control. Intravenous fluid rate was increased to expedite contrast agent clearance. No further seizures were observed throughout the hospital stay. The neurology team’s examination 3 hours after the first seizure revealed an irritable and somnolent child who did not follow commands. Funduscopic examination was normal. Cranial nerve examination indicated left nasal labial fold depression. Motor examination demonstrated decreased proximal movements and elbow flexion of the left arm. Deep tendon reflexes were slightly brisk in the upper extremities (2-3⫹). Patellar reflex demonstrated spread that was more marked on the left side. Bilateral ankle clonus was observed, and plantar responses were flexor. An electroencephalogram demonstrated continuous right occipital slowing and rare left central and frontal independent spikes. A repeat unenhanced cranial computed tomogram 10 hours after the first seizure documented minimal clearing of the contrast material and minimal reduction in the cerebral edema (Fig 2b). The next morning, the child regained full strength of his limbs but still remained irritable. He was holding his bottle with both hands and using his baseline language directed to his mother. Bilateral brisk deep tendon reflexes, ankle clonus, and flexor plantar responses remained. An unenhanced cranial computed tomogram 48 hours after the first seizure demonstrated resolution of the parenchymal and subarachnoid hyperdensity. Cerebral edema with decreased gray-white differentiation persisted in the right hemisphere but improved in the left hemisphere (Fig 2c). Four days after the seizure, the child’s neurologic examination returned to baseline. Unenhanced cranial computed tomography returned to normal 6 days after the first seizure episode (Fig 2d). A repeat electroencephalogram 2 weeks after the initial seizure was normal, and phenobarbital was discontinued. Approximately 1 month after the described events, the child underwent a Glenn procedure and cardiac catheterization without incident. At 10 months follow-up, the child did not have any further seizures and was developing normally.
Discussion To our knowledge, this is the first report documenting focal contrast retention associated with a subsequent cardiac catheterization in a child. Using serial unenhanced cranial computed tomography scans and neurologic examinations, this case report provides important insights into the natural history of contrast-induced neurotoxicity in a child with no predisposing neurologic abnormalities. The precise mechanism of contrast agent neurotoxicity is yet to be defined; however, both osmotic and chemical factors have been implicated. It is believed that contrast agents cause osmotic disruption of the blood-brain barrier [11]. An osmotic mechanism would also explain the associated cortical edema that is reported in almost every case of cerebral contrast retention and the elevated risk of increased intracranial pressure with preexisting intracranial tumors [12,13]. Increased blood contrast concentration and reduced contrast clearance in the setting of renal insufficiency are believed to be risk factors for blood-brain barrier disruption and cerebral contrast retention [8,9,10,13]. Once the contrast agent has entered the brain, it most likely results in seizures or cortical dysfunction by increasing neuronal excitability and excitotoxicity [11]. This phenomenon would be consistent with the in-
Figure 2. Axial unenhanced cranial computed tomographic images through the two levels of the basal ganglia and thalamus for four time points after the child’s first focal seizure. (A) One hour after the seizure, contrast is observed in the parenchyma and subarachnoid space of right hemisphere with sparing of the parenchyma supplied by the middle cerebral artery. Mild generalized diffuse cerebral edema is observed in both hemispheres. (B) Ten hours after the seizure there is minimal reduction in swelling and partial resolution of the contrast. (C) The contrast has cleared 48 hours after the seizure but cerebral edema persists in the right hemisphere. (D) The cranial computed tomography returned to baseline 6 days after the seizure.
creased risk of contrast-induced seizures in patients with preexisting compromise of the blood-brain barrier [1,14]. The distribution of contrast retention after coronary angiography may correlate with focality. Indeed, bilateral contrast retention was associated with generalized seizures [10], whereas unilateral contrast retention was associated with contralateral seizures [8,9]. However, in adults, cortical dysfunction such as transient visual loss after cardiac catheterization is not consistently associated with contrast retention or edema [6]. In our patient, the focality of the seizure and the electroencephalographic abnormality correlated well with the location of the retained contrast. Weissman et al. [4] reported hypodensities in a middle cerebral artery distribution in children with focal seizures followed by hemiplegia, but the presence of contrast was not mentioned. Bilateral contrast retention with unilateral hypodensity in the middle cerebral artery distribution was reported in two neonates with seizures followed by hemiplegia after balloon angioplasty [15]. Magnetic resonance imaging has demonstrated a focal infarction in children with hemiplegia after cardiac catheterization [5]. Because serial cranial computed tomograms have rarely been performed in children and cranial computed tomography and magnetic resonance imaging has never been performed on
the same child, the evolution of cranial computed tomography hypodensities in children with contrast retention is unclear. In this report, we have proposed that nonionic contrast agent penetrated the blood-brain barrier and subsequently resulted in seizure activity by a direct neurotoxicity. Alternatively, a seizure or focal ischemia may have initiated the breach of the blood-brain barrier. Because computed tomography was not performed between the angiography and the seizure, these possibilities cannot be absolutely excluded. However, we think these possibilities are unlikely for a number of reasons. First, because the patient’s motor function and mental status returned to normal within a time frame concordant with resolution of the computed tomographic findings and because no new computed tomographic findings emerged, an infarction is rather unlikely. Second, although computed tomographic changes have been rarely reported after seizures (e.g., [16]), such seizures are generalized or prolonged and neuroimaging findings are focal. Thus the bilateral persistent hemispheral edema would be extremely unlikely after a short focal seizure or small area of ischemia. Furthermore, the patient had no history of seizures, a normal neurologic examination 4 days after the procedure, and a normal electroencephalogram 2 weeks after the proce-
Frye et al: Focal Contrast Retention 215
dure. Taken together, these findings suggest that the seizure was more likely provoked by the contrastinduced edema and toxicity. The cerebral edema in this patient resolved without sequela, and the patient’s neurologic status returned to normal. In children with intracranial tumors, contrast toxicity can lead to marked increase in intracranial pressure resulting in irreversible devastating consequences [12,13]. Follow-up cranial computed tomograms were not reported for any of these children, but cerebral edema would be presumed as the cause of the increased intracranial pressure symptoms. Thus, although seizures may occur as a complication of contrast agents, cerebral edema is the more concerning complication in children. Therefore children with intracranial mass lesions and contrast-induced seizures should be observed closely and treated immediately for signs of increased intracranial pressure. In the absence of an intracranial lesion, cerebral edema induced by a contrast agent appears to be self-limited, and the prognosis for neurologic recovery is good. Although several retrospective studies have estimated the frequency of and risk factors for neurologic complications [4,5], variation in the factors investigated makes comparison across these different studies difficult. Furthermore, few studies use cranial computed tomographic scans to exclude contrast retention. Such information would be helpful to study the relationship between contrast retention, cortical edema, and neurologic dysfunction. A prospective study would be necessary to estimate the frequency of and risk factors for contrast agent entry into the central nervous system after cardiac catheterization.
The authors would like to thank Drs. James J. Riviello, Jr. and James E. Lock for their thoughtful comments on this manuscript.
216
PEDIATRIC NEUROLOGY
Vol. 32 No. 3
References [1] Nelson M, Bartlett RJ, Lamb JT. Seizures after intravenous contrast media for cranial computed tomography. J Neurol Neurosurg Psychiatry 1989;52:1170-5. [2] Drake TG, Kerr HD. Status epilepticus after cranial CT with contrast media. WMJ 1997;96:46-8. [3] Lukovits TG, Fadul CE, Pipas JM, Williamson PD. Nonconvulsive status epilepticus after intravenous contrast medium administration. Epilepsia 1996;37:1117-20. [4] Weissman BM, Aram DM, Levinsohn MW, Ben-Shachar G. Neurologic sequelae of cardiac catheterization. Cathet Cardiovasc Diagn 1985;11:577-83. [5] Liu XY, Wong V, Leung M. Neurologic complications due to catheterization. Pediatr Neurol 2001;24:270-5. [6] Zwicker JC, Sila CA. MRI findings in a case of transient cortical blindness after cardiac catheterization. Catheter Cardiovasc Interv 2002; 57:47-9. [7] Koehler PJ, Endtz LJ, den Bakker PB. Transient global amnesia after coronary angiography. Clin Cardiol 1986;9:170-1. [8] Muruve DA, Steinman TI. Contrast-induced encephalopathy and seizures in a patient with chronic renal insufficiency. Clin Nephrol 1996;45:406-9. [9] Kuhn MJ, Burk TJ, Powell FC. Unilateral cerebral cortical and basal ganglia enhancement following overdosage of nonionic contrast media. Comput Med Imaging Graph 1995;19:307-11. [10] May EF, Ling GS, Geyer CA, Jabbari B. Contrast agent overdose causing brain retention of contrast, seizures and parkinsonism. Neurology 1993;43:836-8. [11] Torvik A, Walday P. Neurotoxicity of water-soluble contrast media. Acta Radiol Suppl 1995;399:221-9. [12] Haslam RH, Cochrane DD, Amundson GM, Johns RD. Neurotoxic complications of contrast computed tomography in children. J Pediatr 1987;111:837-40. [13] Park SC, Neches WH. The neurologic complications of congenital heart disease. Neurol Clin 1993;11:441-62. [14] Neuwelt EA, Specht HD, Howieson J, et al. Osmotic bloodbrain barrier modification: Clinical documentation by enhanced CT scanning and/or radionuclide brain scanning. AJR Am J Roentgenol 1983;141:829-35. [15] Treacy EP, Duncan WJ, Tyrell MJ, Lowry NJ. Neurological complications of balloon angioplasty in children. Pediatr Cardiol 1991; 12:98-101. [16] Kramer RE, Luders H, Lesser RP, et al. Transient focal abnormalities of neuroimaging studies during focal status epilepticus. Epilepsia 1987;28:528-32.
From the Departments of *Neurology and †Cardiology, Children’s Hospital Boston, Boston, Massachusetts.
© 2005 by Elsevier Inc. All rights reserved. doi:10.1016/j.pediatrneurol.2004.07.012 ● 0887-8994/05/$—see front matter
Introduction Radiologic contrast is commonly used to visualize vessel and organ lumens. Allergic reaction is, by far, the most common complication associated with contrast media, although the incidence has decreased with the transition from ionic to nonionic contrast media. Seizures are the most common neurologic complication of radiologic contrast media. In the largest case series, seizures complicating contrast-enhanced cranial computerized tomography were more common in patients with a seizure history or intracranial lesion [1]. Brief, self-resolving, generalized clonictonic seizures were most common regardless of seizure history or lesion focality [1]. However, convulsive and nonconvulsive status epilepticus has been reported in adults with intracranial lesions after contrast-enhanced cranial computed tomography [2,3]. Neurologic complications may occur in up to 1% of children after cardiac catheterization although a recent estimate suggests that this complication rate has improved [4,5]. The most common reversible neurologic complication after cardiac catheterization is global amnesia or cortical blindness for adults [6,7] and seizure or hemiplegia for children [4,5]. Although three cases of seizures were associated with cerebral contrast retention in adults [8-10], such an association has not been described in children. This report describes a child with complex congenital heart disease who developed partial seizures, cerebral contrast retention, and cortical edema after cardiac angiography with arterial coil occlusion. The series of cranial computed tomographic scans presented help define the temporal dynamics of subarachnoid and intraparenchymal cerebral contrast clearance and resolution of cerebral edema. Case Report An 18-month-old male with complex congenital heart disease with heterotaxy and single right ventricle underwent 3 hours of cardiac catheterization (Fig 1). At 5 months of age, the child had undergone right Blalock-Taussig shunt placement, central pulmonary artery reconstitution for discontinuous pulmonary arteries, and total anomalous pulmonary venous return repair. During the current procedure, 7 cc/kg of ioversol, a low-osmolar, nonionic contrast agent, was used for evaluation of the cardiopulmonary anatomy. Coil occlusion of the right internal mammary artery and an additional aortopulmonary collateral from the right subclavian artery was also performed. Although injection of
Communications should be addressed to: Dr. Sahin; Department of Neurology; Children’s Hospital Boston; 300 Longwood Avenue; Enders 250; Boston, MA 02115. Received January 12, 2004; accepted July 26, 2004.
Frye et al: Focal Contrast Retention 213
Figure 1. Diagram depicting the corrected cardiopulmonary anatomy of the patient. Patient had heterotaxy and single right ventricle and a common atrioventricular valve. There was a modified Blalock-Taussig shunt placed between the right innominate artery (RIA) to pulmonary artery and an interposition graft placed between the left and right pulmonary arteries. During the angiogram, contrast was injected into the right subclavian artery, pulmonary veins/left atrium, interposition shunt, and pulmonary arteries. SVC ⫽ superior vena cava; LPA ⫽ left pulmonary artery; RPA ⫽ right pulmonary artery; LCCA ⫽ left common carotid artery; RCCA ⫽ right common carotid artery; LSCA ⫽ left subclavian artery; RSCA ⫽ right subclavian artery; APC ⫽ aortopulmonary collateral. AV ⫽ atrioventricular.
contrast into the pulmonary vasculature and cardiac chambers did not differentially fill the right or left carotid circulation, the right common carotid artery was undoubtedly selectively filled with contrast when the Blalock-Taussig shunt and collaterals arising from the right subclavian artery were selectively studied. Vital signs remained stable (blood pressure: 76/51 to 132/77 mm Hg; heart rate: 95 to 152 beats/min; oxygen saturation: 66-80%, baseline oxygen saturation: middle 70s). Precatheterization hematocrit was 52.7, and 10 cc/kg of packed red blood cells was transfused during the procedure because of a 5 cc/kg estimated blood loss. Heparin was infused during the procedure and continued postprocedure. After the procedure, the child was holding and drinking from his own cup, interacting with his parents using his baseline language skills, and moving all extremities equally. At approximately 12 hours after the catheterization, two brief (each less than 3 minutes) episodes of left clonic arm movements, followed by lethargy, were observed. Intravenous lorazepam stopped the second episode. The heparin infusion was discontinued, electrolytes and blood glucose were measured as normal, and complete blood count revealed an 11-point drop in hematocrit. Skull and chest x-rays verified stable occlusion coil positions. An unenhanced cranial computed tomography obtained 1 hour after the seizure demonstrated right holohemispheric parenchymal and subarachnoid hyperdensity with relative sparing of the parenchyma in the distribution of the middle cerebral artery, and mild generalized bilateral diffuse cerebral edema (Fig 2a).
214
PEDIATRIC NEUROLOGY
Vol. 32 No. 3
Upon transfer to the intensive care unit, clonic movements of the left leg developed and responded to another bolus of lorazepam. Phenobarbital 20 mg/kg was administered as an intravenous bolus, and a maintenance dose (3 mg/kg/day) was begun for seizure control. Intravenous fluid rate was increased to expedite contrast agent clearance. No further seizures were observed throughout the hospital stay. The neurology team’s examination 3 hours after the first seizure revealed an irritable and somnolent child who did not follow commands. Funduscopic examination was normal. Cranial nerve examination indicated left nasal labial fold depression. Motor examination demonstrated decreased proximal movements and elbow flexion of the left arm. Deep tendon reflexes were slightly brisk in the upper extremities (2-3⫹). Patellar reflex demonstrated spread that was more marked on the left side. Bilateral ankle clonus was observed, and plantar responses were flexor. An electroencephalogram demonstrated continuous right occipital slowing and rare left central and frontal independent spikes. A repeat unenhanced cranial computed tomogram 10 hours after the first seizure documented minimal clearing of the contrast material and minimal reduction in the cerebral edema (Fig 2b). The next morning, the child regained full strength of his limbs but still remained irritable. He was holding his bottle with both hands and using his baseline language directed to his mother. Bilateral brisk deep tendon reflexes, ankle clonus, and flexor plantar responses remained. An unenhanced cranial computed tomogram 48 hours after the first seizure demonstrated resolution of the parenchymal and subarachnoid hyperdensity. Cerebral edema with decreased gray-white differentiation persisted in the right hemisphere but improved in the left hemisphere (Fig 2c). Four days after the seizure, the child’s neurologic examination returned to baseline. Unenhanced cranial computed tomography returned to normal 6 days after the first seizure episode (Fig 2d). A repeat electroencephalogram 2 weeks after the initial seizure was normal, and phenobarbital was discontinued. Approximately 1 month after the described events, the child underwent a Glenn procedure and cardiac catheterization without incident. At 10 months follow-up, the child did not have any further seizures and was developing normally.
Discussion To our knowledge, this is the first report documenting focal contrast retention associated with a subsequent cardiac catheterization in a child. Using serial unenhanced cranial computed tomography scans and neurologic examinations, this case report provides important insights into the natural history of contrast-induced neurotoxicity in a child with no predisposing neurologic abnormalities. The precise mechanism of contrast agent neurotoxicity is yet to be defined; however, both osmotic and chemical factors have been implicated. It is believed that contrast agents cause osmotic disruption of the blood-brain barrier [11]. An osmotic mechanism would also explain the associated cortical edema that is reported in almost every case of cerebral contrast retention and the elevated risk of increased intracranial pressure with preexisting intracranial tumors [12,13]. Increased blood contrast concentration and reduced contrast clearance in the setting of renal insufficiency are believed to be risk factors for blood-brain barrier disruption and cerebral contrast retention [8,9,10,13]. Once the contrast agent has entered the brain, it most likely results in seizures or cortical dysfunction by increasing neuronal excitability and excitotoxicity [11]. This phenomenon would be consistent with the in-
Figure 2. Axial unenhanced cranial computed tomographic images through the two levels of the basal ganglia and thalamus for four time points after the child’s first focal seizure. (A) One hour after the seizure, contrast is observed in the parenchyma and subarachnoid space of right hemisphere with sparing of the parenchyma supplied by the middle cerebral artery. Mild generalized diffuse cerebral edema is observed in both hemispheres. (B) Ten hours after the seizure there is minimal reduction in swelling and partial resolution of the contrast. (C) The contrast has cleared 48 hours after the seizure but cerebral edema persists in the right hemisphere. (D) The cranial computed tomography returned to baseline 6 days after the seizure.
creased risk of contrast-induced seizures in patients with preexisting compromise of the blood-brain barrier [1,14]. The distribution of contrast retention after coronary angiography may correlate with focality. Indeed, bilateral contrast retention was associated with generalized seizures [10], whereas unilateral contrast retention was associated with contralateral seizures [8,9]. However, in adults, cortical dysfunction such as transient visual loss after cardiac catheterization is not consistently associated with contrast retention or edema [6]. In our patient, the focality of the seizure and the electroencephalographic abnormality correlated well with the location of the retained contrast. Weissman et al. [4] reported hypodensities in a middle cerebral artery distribution in children with focal seizures followed by hemiplegia, but the presence of contrast was not mentioned. Bilateral contrast retention with unilateral hypodensity in the middle cerebral artery distribution was reported in two neonates with seizures followed by hemiplegia after balloon angioplasty [15]. Magnetic resonance imaging has demonstrated a focal infarction in children with hemiplegia after cardiac catheterization [5]. Because serial cranial computed tomograms have rarely been performed in children and cranial computed tomography and magnetic resonance imaging has never been performed on
the same child, the evolution of cranial computed tomography hypodensities in children with contrast retention is unclear. In this report, we have proposed that nonionic contrast agent penetrated the blood-brain barrier and subsequently resulted in seizure activity by a direct neurotoxicity. Alternatively, a seizure or focal ischemia may have initiated the breach of the blood-brain barrier. Because computed tomography was not performed between the angiography and the seizure, these possibilities cannot be absolutely excluded. However, we think these possibilities are unlikely for a number of reasons. First, because the patient’s motor function and mental status returned to normal within a time frame concordant with resolution of the computed tomographic findings and because no new computed tomographic findings emerged, an infarction is rather unlikely. Second, although computed tomographic changes have been rarely reported after seizures (e.g., [16]), such seizures are generalized or prolonged and neuroimaging findings are focal. Thus the bilateral persistent hemispheral edema would be extremely unlikely after a short focal seizure or small area of ischemia. Furthermore, the patient had no history of seizures, a normal neurologic examination 4 days after the procedure, and a normal electroencephalogram 2 weeks after the proce-
Frye et al: Focal Contrast Retention 215
dure. Taken together, these findings suggest that the seizure was more likely provoked by the contrastinduced edema and toxicity. The cerebral edema in this patient resolved without sequela, and the patient’s neurologic status returned to normal. In children with intracranial tumors, contrast toxicity can lead to marked increase in intracranial pressure resulting in irreversible devastating consequences [12,13]. Follow-up cranial computed tomograms were not reported for any of these children, but cerebral edema would be presumed as the cause of the increased intracranial pressure symptoms. Thus, although seizures may occur as a complication of contrast agents, cerebral edema is the more concerning complication in children. Therefore children with intracranial mass lesions and contrast-induced seizures should be observed closely and treated immediately for signs of increased intracranial pressure. In the absence of an intracranial lesion, cerebral edema induced by a contrast agent appears to be self-limited, and the prognosis for neurologic recovery is good. Although several retrospective studies have estimated the frequency of and risk factors for neurologic complications [4,5], variation in the factors investigated makes comparison across these different studies difficult. Furthermore, few studies use cranial computed tomographic scans to exclude contrast retention. Such information would be helpful to study the relationship between contrast retention, cortical edema, and neurologic dysfunction. A prospective study would be necessary to estimate the frequency of and risk factors for contrast agent entry into the central nervous system after cardiac catheterization.
The authors would like to thank Drs. James J. Riviello, Jr. and James E. Lock for their thoughtful comments on this manuscript.
216
PEDIATRIC NEUROLOGY
Vol. 32 No. 3
References [1] Nelson M, Bartlett RJ, Lamb JT. Seizures after intravenous contrast media for cranial computed tomography. J Neurol Neurosurg Psychiatry 1989;52:1170-5. [2] Drake TG, Kerr HD. Status epilepticus after cranial CT with contrast media. WMJ 1997;96:46-8. [3] Lukovits TG, Fadul CE, Pipas JM, Williamson PD. Nonconvulsive status epilepticus after intravenous contrast medium administration. Epilepsia 1996;37:1117-20. [4] Weissman BM, Aram DM, Levinsohn MW, Ben-Shachar G. Neurologic sequelae of cardiac catheterization. Cathet Cardiovasc Diagn 1985;11:577-83. [5] Liu XY, Wong V, Leung M. Neurologic complications due to catheterization. Pediatr Neurol 2001;24:270-5. [6] Zwicker JC, Sila CA. MRI findings in a case of transient cortical blindness after cardiac catheterization. Catheter Cardiovasc Interv 2002; 57:47-9. [7] Koehler PJ, Endtz LJ, den Bakker PB. Transient global amnesia after coronary angiography. Clin Cardiol 1986;9:170-1. [8] Muruve DA, Steinman TI. Contrast-induced encephalopathy and seizures in a patient with chronic renal insufficiency. Clin Nephrol 1996;45:406-9. [9] Kuhn MJ, Burk TJ, Powell FC. Unilateral cerebral cortical and basal ganglia enhancement following overdosage of nonionic contrast media. Comput Med Imaging Graph 1995;19:307-11. [10] May EF, Ling GS, Geyer CA, Jabbari B. Contrast agent overdose causing brain retention of contrast, seizures and parkinsonism. Neurology 1993;43:836-8. [11] Torvik A, Walday P. Neurotoxicity of water-soluble contrast media. Acta Radiol Suppl 1995;399:221-9. [12] Haslam RH, Cochrane DD, Amundson GM, Johns RD. Neurotoxic complications of contrast computed tomography in children. J Pediatr 1987;111:837-40. [13] Park SC, Neches WH. The neurologic complications of congenital heart disease. Neurol Clin 1993;11:441-62. [14] Neuwelt EA, Specht HD, Howieson J, et al. Osmotic bloodbrain barrier modification: Clinical documentation by enhanced CT scanning and/or radionuclide brain scanning. AJR Am J Roentgenol 1983;141:829-35. [15] Treacy EP, Duncan WJ, Tyrell MJ, Lowry NJ. Neurological complications of balloon angioplasty in children. Pediatr Cardiol 1991; 12:98-101. [16] Kramer RE, Luders H, Lesser RP, et al. Transient focal abnormalities of neuroimaging studies during focal status epilepticus. Epilepsia 1987;28:528-32.