ADOLESCENT idiopathic scoliosis (AIS) is the most common form of scoliosis in the United States. 1Surgical correction is required in some children. 1,2Neurologic complications after scoliosis repair are infrequent but may be devastating and include complete or partial paraplegia, visual disturbances, and cranial and peripheral nerve injuries. 3We present the case of a 13-yr-old girl with idiopathic scoliosis in whom a posterior inferior cerebellar artery infarction developed after posterior spinal fusion with instrumentation. Neuroimaging studies obtained postoperatively revealed a previously undiagnosed Chiari I malformation, which we speculate was an essential factor in the etiology of her cerebrovascular event.

Case Report

A 13-yr-old girl (weight, 45 kg; American Society of Anesthesiologists physical status I) with AIS (a right thoracic curve of 57° and a left lumbar curve of 70°) without neurologic symptoms and normal neurologic examination results underwent posterior spinal arthrodesis from T4 to L3 using dual rod fixation with hooks, sublaminar cables, and pedicle screws. General anesthesia was maintained with propofol (100 –230 μg · kg−1· min−1) and remifentanil (0.1–0.9 μg · kg−1· min−1). Anesthesia and surgery times were 7 h, 9 min and 5 h, 53 min, respectively. Somatosensory and motor evoked potentials were monitored by a neurophysiologist. The operation proceeded uneventfully, with maintenance of mean arterial pressure between 50 and 80 mmHg, hemoglobin concentration of 8.7–12.8 g/dl, and stable somatosensory and motor evoked potentials. The patient’s platelet count decreased from 435,000/mm3preoperatively to 231,000/mm3postoperatively. During the procedure, her estimated blood loss was 1,500 ml, and she received 3,800 ml lactated Ringer’s solution, 500 ml cell saver blood, and 217 ml autologous packed erythrocytes. The patient was extubated while awake in the operating room. She was alert and comfortable, followed commands, and moved all four extremities in the recovery room. Postoperative analgesia was managed using patient-controlled analgesia with morphine (demand dose of 0.9 mg, lockout interval of 8 min, continuous infusion 0.9 mg/h, and a 1-h limit of 5.4 mg). On the first postoperative day, the continuous morphine infusion was discontinued and 4 mg ondansetron administered intravenously every 8 h was begun because the patient reported nausea and vomiting. On the second postoperative day, the patient reported occipital headache, generalized pruritus, and severe nausea and vomiting. The patient-controlled analgesia was changed from morphine to hydromorphone (demand dose of 0.2 mg, lockout interval of 8 min, and an hourly maximum of 1 mg) because these reports were attributed to the morphine. Approximately 7 h later, the patient became obtunded, requiring supplemental oxygen to maintain oxyhemoglobin saturation greater than 93%. An anesthesiology resident, believing the patient to be narcotized, administered 400 μg intravenous naloxone, and the demand-only hydromorphone patient-controlled analgesia was reprogrammed (demand dose of 0.1 mg, lockout interval of 8 min, and hourly maximum of 0.5 mg). The patient’s mental status improved transiently. She had received 1.3 mg hydromorphone in the previous 7 h. Two hours after receiving naloxone, the patient was noted to be unresponsive, with bilaterally dilated and sluggishly reactive pupils and an absent gag reflex. Arterial blood gas analysis showed a pH of 7.45, a partial pressure of carbon dioxide (P co2) of 40 mmHg, and a partial pressure of oxygen (P o2) of 81 mmHg in room air. Additional lab values included 129 mMNa+, 156 g/dl glucose, a prothrombin time of 13.3 s, a partial thromboplastin time of 32.4 s, a platelet count of 231,000/ mm3, and 10 g/dl hemoglobin. The level of consciousness improved again with naloxone. Analysis of the patient-controlled analgesia hydromorphone syringe and pump revealed no concentration errors or malfunction in delivery. An emergency computerized tomography scan of the head revealed effacement of basal cisterns and hypodensity in the upper brainstem, consistent with edema and focal ischemia (fig. 1). The patient was transferred to the intensive care unit and treated with diuretics and corticosteroids. Magnetic resonance imaging of her brain (fig. 2) revealed an infarct in the left anterior inferior cerebellum in the distribution of the posterior inferior cerebellar artery. A 10-mm cerebellar tonsillar ectopia was also noted, consistent with Chiari I malformation (fig. 3). Vertebral artery dissection was ruled out by magnetic resonance arteriography. Neurophthalmologic examination revealed left sixth cranial nerve palsy and left hypotropia. The patient’s condition improved rapidly over 24 h. Her only complaint at the time of discharge on the sixth postoperative day was diplopia. She is now more than 18 months out from her surgery and completely recovered without neurologic sequelae. Her thoracic and lumbar curves are being maintained at 23° and 12°, respectively.

Stroke has not been previously reported after AIS repair. This patient had severe occipital headache, nausea, and vomiting, which we initially attributed to morphine. In retrospect, we believe these symptoms were signs of a left posterior inferior cerebellar artery infarct. This constellation of symptoms in a child after AIS repair warrants thorough neurologic evaluation.

Chiari I malformations can occur in association with a dural tear and cerebrospinal fluid leak. 4,5However, we believe this condition preexisted in our patient because there was no dural tear in the operating room. Also, scoliosis is associated with spinal cord and posterior fossa abnormalities, including Chiari I malformation. 5–10When Chiari I malformation and scoliosis coexist, some recommend the former be repaired by posterior fossa decompression first 6because scoliosis may stabilize or improve. The indications for imaging the spinal cord and posterior fossa in children with scoliosis vary but generally include patients with abnormal neurologic signs, left-sided curves, high thoracic or cervical curves, painful curves, rapidly progressive curves, and curves that develop before the age of 10 yr. 9This patient did not have any preoperative indications for magnetic resonance imaging.

The rate of ischemic stroke in the pediatric population is 7.8 in 100,000 children/yr. 11The etiology of an ischemic stroke includes vertebral artery dissection, emboli, and coagulation abnormalities. This patient did not have clinical or laboratory evidence for a coagulopathy at any time during the perioperative period. Ischemic cerebellar stroke has been reported in a patient with Chiari I malformation. 12Vertebral artery dissection was ruled out by magnetic resonance arteriography. Both fat and venous air pulmonary emboli have been reported in adolescents undergoing scoliosis surgery, 13–15and cerebral microemboli were identified by transcranial Doppler in 11 of 13 children aged 13–17 yr who were undergoing surgical correction for scoliosis or kyphosis. 16However, microembolization was not associated with neurologic sequelae in this report. 16Although a basilar artery thromboembolic event cannot be ruled out, we speculate that mechanical factors in the perioperative period exacerbated the patient’s Chiari I malformation with compression of her left posterior inferior cerebellar artery resulting in infarction. We believe that her symptoms of nausea, vomiting, and headache, which were present soon after surgery, were related to her infarction. Progressive somnolence, unconsciousness, and her ocular findings developed 48 h postoperatively as edema surrounding the infarction enlarged. The delay in diagnosis of her cerebellar infarction was due to her symptoms being the same as those that are commonly attributed to opioid analgesics (nausea, vomiting, and somnolence) and ondansetron (headache).

We present this case to highlight the association of occult posterior fossa abnormalities in children with AIS, to raise awareness of the rare possibility of exacerbating Chiari I malformations during AIS repair and the potential for posterior inferior cerebellar artery compression with cerebellar infarction, to raise awareness of the potential for significant thromboembolic events during scoliosis repair, and to stress the importance of a comprehensive neurologic evaluation in patients with severe occipital headache, nausea, and vomiting after AIS repair.

Roach JW: Adolescent idiopathic scoliosis. Orthop Clin North Am 1999; 30:353–65
Brooks HL, Azen SP, Gerberg E, Brooks R, Chan L: Scoliosis: A prospective epidemiological study. J Bone Joint Surg Am 1975; 57:968 –72
MacEwen GD, Bunnell WP, Sriram K: Acute neurological complications in the treatment of scoliosis: A report of the Scoliosis Research Society. J Bone Joint Surg 1975; 57:404 –8
Samii C, Mobius E, Weber W, Heienbrok HW, Berlit P: Pseudo Chiari type I malformation secondary to cerebrospinal fluid leakage. J Neurol 1999; 246:162–4
Atkinson JLD, Weinshenker BG, Miller GM, Piepgras DG, Mokri B: Acquired Chiari I malformation secondary to spontaneous spinal cerebrospinal fluid leakage and chronic intracranial hypotension syndrome in seven cases. J Neurosurg 1988; 88:237–42
Eule JM, Erickson MA, O’Brien MF, Handler M: Chiari I malformation associated with syringomyelia and scoliosis: A twenty-year review of surgical and nonsurgical treatment in a pediatric population. Spine 2002; 27:1451–5
Shen WJ, McDowell GS, Burke SW, Levine DB, Chutorian AM: Routine preoperative MRI and SEP studies in adolescent idiopathic scoliosis. J Pediatr Orthop 1996; 16:350 –3
Emery E, Redondo A, Rey A: Syringomyelia and Arnold Chiari in scoliosis initially classified as idiopathic: Experience with 25 patients. Eur Spine J 1997; 6:158 –62
Maiocco B, Deeney VF, Coulon R, Parks PF: Adolescent idiopathic scoliosis and the presence of spinal cord abnormalities: Preoperative magnetic resonance imaging analysis. Spine 1997; 22:2537–41
Dure LS, Percy AK, Cheek WR, Laurent JP: Chiari type I malformation in children. J Pediatr 1989; 115:573–6
Lynch JK, Hirtz DG, DeVeber G, Nelson KB: Report of the National Institute of Neurological Disorders and Stroke workshop on perinatal and childhood stroke. Pediatrics 2002; 109:116 –23
Petit H, Jomin M, and Rousseaux M: Pseudo-tumoral cerebellar infarction and Arnold-Chiari malformation (in French). Rev Neurol (Paris) 1980; 136:473–9
Gittman JE, Buchanan TA, Fisher BJ, Bergeson PS, Palmer PE: Fatal fat embolism after spinal fusion for scoliosis. JAMA 1983; 249:779 –81
Frankel AS, Holzman RS: Air embolism during posterior spinal fusion. Can J Anaesth 1988; 35:511–4
Lang SA, Duncan PG, Dupuis PR: Fatal air embolism in an adolescent with Duchenne muscular dystrophy during Harrington instrumentation. Anesth Analg 1989; 69:132–4
Rodriguez RA, Letts M, Jarvis J, Clarke WN, Murto K: Cerebral microembolization during pediatric scoliosis surgery: A transcranial Doppler study. J Pediatr Orthop 2001; 21:532–6