'Idiopathic' Neonatal Cerebral Infarction John F. Mantovani, MD, Gregory J. Gerber, MD \s=b\ Neonatal cerebral infarction is infre- quently reported in the living newborn and is usually associated with perinatal risk factors, trauma, and asphyxia. This report describes two cases of full-term newborns with no predisposing problems who had unequivocal left cerebral infarctions associated with seizures in the first 24 hours of life. Radiological studies suggested an embolic cause in one case. (AJDC 1984;138:359-362) is infrequently Cerebral infarction in the newborn. reported living Nonetheless, a series of newborns from an intensive care unit who had undergone autopsies indicated that 17% of full-term infants had suffered this insult.1 Thromboemboli from pla¬ cental and umbilical vessels or the ductus arteriosus and sepsis with dis¬ seminated intravascular coagulation have been emphasized in pathophysiological formulations.1,2 To our knowl¬ edge, these phenomena have not been demonstrated during life in newborn infants. Recent investigators have de¬ scribed an association with seizures1,3 and a variety of predisposing clinical conditions, such as traumatic deliv¬ ery,3,4 birth asphyxia,3,5 prematurity,1,3 meningitis,6'7 polycythemia,8 and vas¬ cular catheterization.910 We have cared for two full-term new¬ born infants with cerebral infarctions and seizures, but without apparent predisposing conditions. We have des¬ ignated these cases as "idiopathic" to emphasize the absence of known causes. Although the proximate cause of the infarct in one case was an apparFrom the Department of Neurosciences, Dean Medical Center, Madison, Wis (Dr Mantovani), and the Geneva Family Practice Clinic, Lake Geneva, Wis (Dr Gerber). Reprint requests to Department of Neurosciences, Dean Medical Center, 1313 Fish Hatchery Rd, Madison, WI 53715 (Dr Mantovani). ent intracranial embolus, the source of the embolus is not known. We describe these infants to call attention to the unexpected occurrence of stroke in the unstressed full-term newborn. REPORT OF CASES Case 1.—A 4,200-g full-term female in¬ fant was born to a 29-year-old primigrávida mother. The prenatal course was compli¬ cated by maternal mixed-cell thyroid car¬ cinoma; the thyroid was surgically removed at 14 weeks' gestation. The mother subse¬ quently received thyroid replacement ther¬ apy for the duration of the pregnancy and had no further difficulties. Labor began spontaneously, proceeded without compli¬ cation, and led to vaginal delivery without instrumentation six hours later. Apgar scores were 9 and 9 at one and five minutes, respectively, and the infant was trans¬ ferred to the well-baby nursery. She appeared to be well until 8 hours of age, when she had an apneic episode fol¬ lowed by repeated episodes of horizontal deviation of the eyes to the right and clonic jerking of the right arm. She was then transferred to the infant intensive care unit. On initial examination, she was a normal-appearing infant who was alert, crying, and moving all extremities spon¬ taneously. Her head circumference was 36 cm, and the anterior fontanel was flat. Intermittent sucking and facial jerking, as well as clonic jerking of the right arm, were observed several times during the ex¬ amination. There were no abnormalities of extraocular movements, fundi, or facial, palatal, or tongue movement. Her muscle tone, deep-tendon reflexes, and spon¬ taneous movements of the limbs were sym¬ metrical. Suck, Moro, and grasp reflexes were present. On admission, laboratory evaluation in¬ cluded a normal CBC count and concentra¬ tions of serum electrolytes, glucose, cal¬ cium, magnesium, and arterial blood gases. A lumbar puncture disclosed slightly xanthochromic spinal fluid with 1,200 RBCs and two mononuclear WBCs per cubic millimeter. The protein concentration in the CSF was 104 mg/dL, and the concen¬ tration of glucose was 53 mg/dL. An EEG obtained at 36 hours of age demonstrated paroxysmal abnormalities in the left hemi¬ sphere (Fig 1, top). Frequent clonic seizures involving the right arm and face, and subtle seizures, including eye jerking and repetitive suck¬ ing movements, continued. The infant was additionally treated with 20 mg/kg of phenytoin (Dilantin) and ultimately 0.3 mL/kg of rectal paraldehyde before the seizures became infrequent. Seizures did not recur after 72 hours of age. A computed tomographic (CT) scan per¬ formed without contrast on the third day of life showed an area of mixed increased and decreased density in the left cerebral hemi¬ sphere (Fig 2, top left). Administration of contrast medium confirmed a left middle cerebral artery (MCA) distribution abnor¬ mality (Fig 2, top right). This was in¬ terpreted as a left MCA infarction with hemorrhage. A technetium Tc 99m brain scan performed on the next day showed a wedge-shaped area of increased uptake in the distribution of the left MCA (Fig 2, bottom left). Because of the unusual nature of these findings, a cerebral arteriogram was performed on the tenth day of life; it showed a narrowed segment of the proxi¬ mal left MCA consistent with a resolving embolie occlusion (Fig 3). During hospitalization, the child had per¬ sistent mild lethargy with diffuse hypotonia and reduced tone of the right arm compared to the left. At 12 months of age, she had a spastic right hemiparesis but normal overall development as judged by the Bayley Scales of Infant Development. A CT scan performed at that time showed atrophy and porencephaly in the left hemi¬ sphere (Fig 2, bottom right). Case 2.—A 4,252-g full-term male infant was born to a 25-year-old primigrávida mother. Prenatal course was uncompli¬ cated, except for a maternal urinary tract infection requiring parenteral treatment with ampicillin sodium for ten days, six weeks before delivery. Labor began spon¬ taneously, proceeded without complica¬ tion, and led to a vaginal delivery without Downloaded From: http://archpedi.jamanetwork.com/ by a University of Michigan User on 06/14/2015 any instrumentation within 12 hours. Apgar scores were 9 and 9 at one and five minutes, respectively; the child appeared to be well and was transferred to the wellbaby nursery. Within eight hours, the in¬ fant was noted to be irritable with a re¬ duced suck. Cultures of the blood, urine, and CSF were done, and the infant was given ampicillin sodium, 200 mg/kg/day, and gentamicin sulfate, 5 mg/kg/day. At 24 hours of age, he had several brief episodes of clonic jerking of the right upper and lower extremity and was transferred to the infant intensive care unit of this hospital for further treatment. On initial physical examination, repeated episodes of staring, chewing, and clonic j erking of the right arm and leg were noted. There was mild hypertonia of all ex¬ tremities. Extraocular movements, funduscopic examination, and facial, palatal, and tongue movements were normal. Deeptendon reflexes were brisk, greater in the lower extremities than in the upper ex¬ tremities. Admission laboratory studies included a normal CBC count and serum concentra¬ tions of electrolytes, glucose, calcium, and magnesium, and normal arterial blood gas levels. Lumbar puncture produced slightly xanthochromic CSF, with 54,000 RBCs and 39 WBCs (70% polymorphonuclear) per cubic millimeter, a protein concentration of 112 mg/dL, and a glucose level of 56 mg/dL. An EEG on the second day of life showed multifocal paroxysmal abnormalities, max¬ imal in the central portion of the left cere¬ bral hemisphere (Fig 1, bottom). A CT scan without contrast medium on the second day of life suggested an area of hypodensity in the left MCA distribution (Fig 4, top left), and a CT scan with con¬ trast medium on the seventh day of life was consistent with an extensive cerebral in¬ farction in this distribution (Fig 4, top right). Technetium Tc 99m brain scan the following day showed increased uptake in the left MCA distribution (Fig 4, bottom). The seizures decreased after a single 20-mg/kg dose of phénobarbital. The child had only occasional mild j erking of the right arm until 96 hours of age, when all seizure activity ended. Subsequent examination disclosed diffuse hypotonia with poor head control, but symmetrical spontaneous and reactive movements, normal deep-tendon reflexes, and no apparent focal abnormali¬ ties. The infant was discharged on the tenth day of life. Follow-up at 6 months of age demon¬ strated a severe right spastic hemiparesis with a right visual field defect and reduced appreciation of tactile stimulation of the right upper and lower extremities as well as the trunk. Overall development was within Fp4-C4 c4-o2 4- 2 . FP3-C3 C -O " -T^-150^ a ' J Eye ' ' Eyes Jerking ^V^V^^^^v^—^^^^—v^A— ', [ j , j ¡ ( [ {j J ¡ (· j j ¡ | . " Chewing Fn-C3 Eye _150 µ ECG Fig 1.—Top, Case 1. Electroencephalogram shows recurrent spike activity in left temporoparieto-occipital regions coincident with rhythmical eye jerking to right. Bottom, Case 2. Electroencephalogram shows spike, slow-wave, and spindle activity in left temporoparietooccipital area with contralateral spreading and coincident with involuntary chewing. the normal range on the Bayley Scales of Infant Development. COMMENT Little is known about the patho¬ genesis and clinical evolution of cere¬ bral infarctions in newborns. None of the previously described risk factors for neonatal stroke was present in our infants. Both infants probably suf¬ fered an insult of unknown cause Downloaded From: http://archpedi.jamanetwork.com/ by a University of Michigan User on 06/14/2015 around the time of birth, given the initial asymptomatic period and the onset of seizures from eight to 24 hours after birth. The evolution of cerebral changes with maximal radiological ab¬ normalities seven to ten days after birth is also consistent with this view.11 It is also likely that, at least in patient 1, the infarction was due to an embo¬ lus. The angiographie findings ten days after birth support the diagnosis Fig 2.—Case 1. Top left, Computed tomographic (CT) scan without contrast medium shows left middle cerebral artery (MCA) distribution hemorrhagic infarction (arrowheads). Top right, After injection of contrast medium, CT scan shows left MCA distribution infarction. Bottom left, Posteroanterior view of nucleotide scan shows left MCA distribution abnormality. Bottom right, At 12 months of age, CT scan shows left hemisphere volume loss and porencephaly. of a resolving embolie occlusion with narrowing of the proximal left MCA. Congenital stenosis cannot be entirely excluded, but is less likely, given the focality and location of the lesion and the normal appearance of all other vessels. The distribution and hemor¬ rhagic nature of the infarcì are also consistent with the usual postmortem findings in newborns with arterial oc¬ clusion.12 To our knowledge, this is the first demonstration of intracranial embolie infarction during neonatal life. The source of such an embolus is specula¬ tive, and the cause of the infarction in the second patient is unknown. Both infarcts were single and involved the MCA distribution, the most common location in other studies.1"3 Unfor¬ tunately, the placenta and umbilical vessels were not examined in detail in either case, but no gross placental abnormalities were noted at the time of delivery. The other report of surviving new¬ borns with strokes differs in the pres¬ ence of major perinatal risk factors, Fig 3.—Case 1. Top, Lateral view of left carotid angiogram shows stenosis of left middle cerebral artery (MCA) (large arrow) at its bifurcation with pericallosal artery (broken line). Main branches of MCA (smaller arrows) are of normal caliber. Cen¬ ter, Anteroposterior view of left carotid an¬ giogram shows same area of stenosis (ar¬ row) in proximal left MCA. Bottom, Lateral view of normal right carotid angiogram. Area of proximal right MCA corresponding to stenotic region on left is indicated for comparison. No pericallosal vessels origi¬ nate from right side. Downloaded From: http://archpedi.jamanetwork.com/ by a University of Michigan User on 06/14/2015 sound may also identify focal ischemie lesions in newborns, these studies performed in our patients.13 An angiogram was performed on our first patient because of concern that a vascular anomaly was possibly causing the hemorrhagic lesion. With in¬ creased recognition of cerebral infarc¬ tions in newborns, however, such inva¬ sive studies should not be necessary in were not most cases. The presentation of these two in¬ fants within 12 months to our nursery suggests that neonatal cerebral infarc¬ tion may occur more frequently than previously suggested. Examination of all infants with seizures, using CT or nuclear brain scans between the fourth and tenth days of life, may lead to increased recognition of this disorder and more information about its patho¬ Fig 4.—Case 2. Top left, Computed to¬ mographic (CT) scan without contrast me¬ dium shows large left middle cerebral artery (MCA) distribution hypodensity (arrow¬ heads) consistent with infarction. Top right, After injection of contrast medium, CT scan genesis and prophylaxis. The medical and nursing staffs of the Infant Intensive Care Unit at St Mary's Hospital Medi¬ cal Center, R. Jaucian, MD, and M. Reale, MD, assisted with this study. R. M. Colburn, MD, gave advice on the roentgenographic findings. Sheila Prensky and Terri Dunham provided tech¬ nical assistance. shows enhancement of left MCA distribu¬ tion lesion. Bottom, Posteroanterior view of nucleotide scan shows left MCA distribution abnormality. including mechanical trauma and hy¬ poxic-ischemic insults.3 The single ex¬ ception in that series was a 30-week premature infant with severe respira¬ tory distress who had a focal left oc¬ cipital hemorrhage between days 3 and 6 of life. Two appended cases to that report may be similar to our pa¬ tients, but complete data are not avail¬ able. Although both of our patients had complications of their gestations, we are not aware of any reported asso¬ ciation between maternal thyroid car¬ cinoma or urinary tract infection and neonatal cerebrovascular disease. Since coagulation studies were not performed, maternal or infant hypercoagulability cannot be excluded as possible pathogenetic factors. Noteworthy clinical features in our patients include the prominence of sei¬ zures and the abnormal CSF. Al¬ though seizures similar to those in our patients have been described in new¬ borns with stroke,1,3 cerebral infarc¬ tion without overt trauma or asphyxia References is not mentioned among the usual causes of neonatal seizures.5 The sei¬ zures in our infants were subtle and focal-clonic in type, and reminiscent of those seen after hypoxic-ischemic in¬ sults. The CSF in both cases was hem¬ orrhagic, which was not believed to be due to traumatic lumbar punctures. These findings may indicate associated primary subarachnoid hemorrhage, or extravasation of blood from the site of cerebral infarction itself. Bloody CSF may prove to be a useful bedside clue for the presence of neonatal cerebral infarction, if confirmed by further study. These infants with seizures and bloody CSF in the first 24 hours of life raised the questions of isolated subarachnoid hemorrhage, intracerebral hematoma, or inapparent hypoxic- ischemic insult.12 Without CT and nu¬ cleotide scans, both infarctions would probably have been overlooked in the neonatal period. Although a recent report has shown that cranial ultra- Downloaded From: http://archpedi.jamanetwork.com/ by a University of Michigan User on 06/14/2015 1. Barmada MA, Moossy J, Schuman RM: Cerebral infarcts with arterial occlusion in neonates. Ann Neurol 1979;6:495-502. 2. Friede RL: Arterial occlusive disease in infancy, in Developmental Neuropathology. New York, Springer Publishing Co, 1975, pp 122-134. 3. Mannino FL, Trauner DA: Stroke in neonates. J Pediatr 1983;102:605-609. 4. Roessmann U, Miller RT: Thrombosis of the middle cerebral artery associated with brain trauma. Neurology 1980;30:889-892. 5. Volpe JJ: Neurology of the Newborn. Philadelphia, WB Saunders Co, 1981, 119-124. 6. Friede RL: Cerebral infarction complicating neonatal leptomeningitis. Acta Neuropathol 1973;23:245-253. 7. Brown LW, Zimmerman RA, Bilaniuk LT: Polycystic brain disease complicating neonatal meningitis. J Pediatr 1979;94:757-759. 8. Amit M, Camfield PR: Neonatal polycythemia causing multiple cerebral infarcts. Arch Neurol 1980;37:109-110. 9. Bull MJ, Schreiner RL, Garg BP, et al: Neurologic complications following temporal artery catheterization. J Pediatr 1980;96:1071-1073. 10. Ruff RL, Shaw CM, Beckwith JB, et al: Cerebral infarction complicating umbilical vein catheterization. Ann Neurol 1980;7:85. 11. Inoue Y, Takemoto K, Miyamoto T, et al: Sequential computed tomography scans in acute cerebral infarction. Radiology 1980;135:655-659. 12. Chaplin ER, Goldstein GW, Norman D: Neonatal seizures, intracerebral hematoma and subarachnoid hemorrhage in fullterm infants. Pediatrics 1979;63:812-813. 13. Hill A, Martin DJ, Daneman A, et al: Focal ischemic cerebral injury in the newborn: Diagnosis by ultrasound and correlation with computed tomographic scan. Pediatrics 1983;71: 790-792.