CONGENITAL MULTIPLE CRANIAL NEUROPATHIES: RELEVANCE OF OROFACIAL ELECTROMYOGRAPHY IN INFANTS FRANCIS RENAULT, MD,1 ROBERTO FLORES-GUEVARA, MD, PhD,1,2 JEAN-JACQUES BAUDON, MD,3 and MARIE-PAULE VAZQUEZ, MD3,4 1
Clinical Neurophysiology Unit, H^ opital Armand-Trousseau, Assistance Publique-H^ opitaux de Paris, 28 avenue Arnold-Netter, 75571 Paris 12, France 2 Facultad de Medicina, Universidad Nacional Mayor de San Marcos, Lima, Peru 3 Faculte de Medecine Ren e-Descartes, Universit e Paris 5, Paris, France 4 Department of Maxillofacial Surgery, H^ opital Necker-Enfants Malades, Assistance Publique-H^ opitaux de Paris, Paris, France Accepted 27 February 2015 ABSTRACT: Introduction: The aim of this study was to assess diagnoses and outcomes of infants with 2 or more cranial neuropathies identified using orofacial electromyography (EMG). Methods: This retrospective study involved 90 patients. Diagnoses took into account clinical, radiological, and genetic data. EMG examined the orbicularis oculi, genioglossus, and levator veli palatini muscles, and blink responses. To evaluate outcome, neurological disability, respiratory complications, and feeding difficulties were recorded. Results: The patients had malformation syndromes (59), encephalopathies (29), or no underlying disorders (2). Neurogenic EMG signs were detected in a mean of 4 muscles, reflecting a mean of 3 affected nerves. EMG identified a higher number of neuropathies than clinical examination alone (82 vs. 31, facial; 56 vs. 2, pharyngeal; 25 vs. 3, hypoglossal). Poor outcome and death were more frequent when EMG identified 4 affected nerves (P 5 0.02). Conclusion: EMG highlights multiple cranial neuropathies that can be clinically silent in infants with malformation syndromes or encephalopathies. Muscle Nerve 52: 754–758, 2015
Electrodiagnostic
examination of paired cranial nerves and orofacial muscles is a method for assessing congenital facial weakness, orofacial malformations, and dysphagia. Because underlying disorders have numerous causes and varying degrees of severity, the diagnoses and prognoses of bulbar dysfunction are not yet clearly defined.1 Orofacial sensorimotor function is not easy to assess clinically in newborn and young infants, who have small brainstem structures and cranial nerves that are challenging to study using MRI. Electrodiagnostic studies investigate paired cranial nerves using blink responses (BRs) and needle electromyography (EMG) of muscles of the face, tongue, and soft palate.. These methods are used to identify axonal loss and demyelination, and to define their extent and severity. We have previously reported the results of orofacial EMG in patients with Pierre Robin sequence (PRS)2,3 and in Abbreviations: ANOVA, analysis of variance; BR, blink response; CHARGE, colobomata, heart disease, atresia of the choanae, retarded growth and development, genital hypoplasia, and ear anomalies or deafness; CT, computed tomography; EMG, electromyography; PRS, Pierre Robin sequence Key words: blink responses; brainstem; child; cranial nerves; electromyography Correspondence to: F. Renault; e-mail: [email protected] C 2015 Wiley Periodicals, Inc. V
Published online 3 March 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/mus.24636
754
Congenital Cranial Neuropathies
patients with other orofacial malformation syndromes.4 In these series, the patients with cranial nerve involvement had more frequent respiratory complications and required longer periods of enteral feeding. The aim of the present study was to retrospectively analyze final diagnoses and outcomes in infants with multiple cranial nerve involvement identified using early orofacial EMG. METHODS Study Design. The patients were investigated between birth and age 6 months because of congenital dysphagia or orofacial malformations. They were referred to our pediatric neurophysiology unit between July 1, 1995 and December 31, 2010. For each patient, we collected clinical, radiological, and genetic data contributing to the diagnosis. To evaluate outcome, we recorded the presence or absence of neurological disabilities (mental or sensorimotor handicap) and the need for respiratory assistance (oxygen, mechanical ventilation, tracheostomy), gastrostomy, and prolonged enteral feeding. Given the retrospective nature of the study and the unequal durations of follow-up, a standardized neurodevelopmental evaluation was not possible. This study was undertaken in accordance with French regulations. Approval by the ethics committee of our university was waived. Informed parental or guardian consent was obtained for every investigation. The data were de-identified before analysis. Patients. During the 15-year period of the study, 997 children were referred to our clinical neurophysiology unit for investigation of orofacial functions using EMG. Multiple cranial nerve involvement was diagnosed in 109 patients with neurogenic EMG signs in the territories of 2 or more cranial nerves. Because 19 patients were lost to follow-up, the series finally included 90 patients (42 boys, 48 girls). Phenotypes were established after clinical examination by a neonatologist, a maxillofacial surgeon, and a clinical geneticist. Ocular, skeletal, renal, and cardiac malformations were identified by ophthalmologic examination, X-ray, and ultrasound scans. The brain was investigated by ultrasound scan, computed tomography MUSCLE & NERVE
November 2015
decreased by at least 30%) (Fig. 1). Blink responses were recorded in the orbicularis oculi muscle in response to electrical stimulation (square pulse: duration 0.2 ms, intensity 10 mA) of the supraorbital nerve. We recorded 6 successive responses to confirm the presence or absence of R1 and R2 components. We considered an asymmetry in R1 latency of up to 3 ms to be normal.
FIGURE 1. Needle EMG recordings of the depressor anguli oris muscle in patients of the series: (A) normal pattern; (B) low-amplitude, reflecting muscle hypoplasia; and (C) neurogenic reduced recruitment.
(CT), or MRI. Upper airways were examined using laryngoscopy. A karyotype was obtained for every patient; deletion at 22q11 was found in 78 patients. Electrodiagnostic Studies. All electrodiagnostic studies were performed by the same examiner (F.R.) using the same methods and were part of every clinical evaluation. EMG was not carried out before gestational age 37 weeks. Conventional EMG was used to study muscles of the face, tongue, and soft palate, at rest and with voluntary contraction during crying. Muscle activity was detected using small disposable concentric needle electrodes (25 mm 3 30G; Viasys Healthcare, Madison, Wisconsin). The needle insertion point was anesthetized using lidocaine–prilocaine cream applied to the skin, and by touching the mucosa with a 2% solution of lidocaine. Bilateral EMG examination systematically included the orbicularis oculi, genioglossus, and levator veli palatini muscles. The orbiculis oris and depressor anguli oris muscles were also investigated in 39 patients. After fasting the patient for up to 4 hours, crying and sucking movements created bursts of EMG activity that could be recorded from oral and facial muscles. Analyses focused on the recruitment pattern and BRs. Markers were placed manually on the trace for measuring amplitudes and latencies. The maximum amplitude was measured from the highest negative peak to the lowest positive peak of the largest burst. Traces were classified as: normal interference pattern; neurogenic single or reduced interference pattern; or low-amplitude full interference pattern (maximum amplitude Congenital Cranial Neuropathies
Statistical Analysis. Numeric variables are presented as mean and range; categorical variables are presented as rates. Comparison of 2 independent groups was performed using the Mann–Whitney Utest for numeric variables and the Fisher exact test for categorical variables. Comparison of >2 groups was performed by 1-way analysis of variance (ANOVA) test for numeric variables and by chisquare test for categorical variables. All hypotheses were constructed as 2-tailed, with P < 0.05 considered significant. RESULTS
Dysphagia and orofacial malformations were the most frequent presenting symptoms leading to referral for electrodiagnostic studies (Table 1). We studied the EMG recordings of 609 muscles, reflecting the function of 180 facial nerves, 172 pharyngeal plexi, and 124 hypoglossal nerves. Median age at EMG examination was 41 days (range 2–172). Neurogenic EMG signs were detected in a mean of 4 muscles, reflecting a mean of 3 affected nerves. The facial nerve was involved in 82 of 90 patients, bilaterally in 64; BRs were altered in 40 of 63 patients, bilaterally absent (25) or asymmetrical (15). The pharyngeal plexus was affected in 56 of Table 1. Clinical features in 90 infants with multiple cranial neuropathies. Clinical signs and course
Number of patients (%)
Presenting symptoms Sucking or swallowing disorders Orofacial malformations Amimia, hypomimia Glossoptosis Cranial nerve manifestations Facial weakness (asymmetry or diplegia) Oculomotor palsy Tongue asymmetry or atrophy Soft palate hypomobility or asymmetry Complications Aspiration, choking episodes Need for gastrostomy Outcome* Respiratory assistance Enteral feeding during >6 months Neurological disabilities Death
56 38 28 18
of of of of
90 90 90 90
(62) (42) (31) (20)
31 18 3 2
of of of of
90 90 90 90
(34) (20) (3) (2)
42 of 90 (47) 37 of 90 (41) 17 60 45 19
of of of of
84 84 84 90
(20) (71) (54) (21)
*Excluding the 6 patients who died before age 6 months.
MUSCLE & NERVE
November 2015
755
Table 2. Diagnoses in 90 patients with multiple cranial neuropathies.
Table 3. Correlation between the extent or localization of cranial nerve involvement and outcome. n
Malformation syndromes Genetic anomalies* Moebius syndrome CHARGE association Other recognizable syndromes† Unidentified Encephalopathies Preterm (25–34 weeks) with vascular brain injury Localized brainstem ischemia Birth asphyxia at full term Born to addicted mother (alcohol, drugs) Progressive encephalopathy‡ No apparent underlying disorder
59 14 8 7 15 15 29 7 6 7 4 5 2
CHARGE association: colobomata, heart disease, atresia of the choanae, retarded growth and development, genital hypoplasia, and ear anomalies or deafness. *[47XXX]; [46XY, ins (12;11) (q14;q13.5q21)]; [46XX t(7;17) (q21.2;q22)]; [del 18q21]; [46XY, inv(2).ish 22q(tuple 132)]; [46XY, rec(9) del p(p22pter), dup q(q32qter)]; [dupl 4ter]; [del 18pter, tri 1qter]; [tri 2q, t(9p) (2q)] ; [45X, t(X;1) (p22.3;p34)]; [46XY/46fra (Xq)Y]; [del 22q11]; [del 22q11]; SOX2 mutation. † Including: acro-reno-ocular, Cornelia de Lange, ectodermic-anhydroticdysplasia, Fryns, Goldenhar, hydranencephaly, Larsen, Noonan, orofaciodigital I, Rubinstein-Taybi, Stickler, Townes-Brocks, Treacher-Collins (2 patients), and Wiedemann-Beckwith syndromes. ‡
Including 2 with mitochondrial disorder.
89 patients, bilaterally in 32. The hypoglossal nerve was involved in 25 of 90 patients, bilaterally in 11. The lower cranial nerves (IX-X, XII) were not involved in 25 of 90 patients. Five patients showed ipsilateral multiple cranial nerve involvement. Low-amplitude interference patterns were additionally recorded in 1 or more muscles of the face, tongue, or soft palate in 19 patients, including 17 with malformation syndromes. Brain MRI showed 20 normal and 29 abnormal imaging patterns, including malformations (13), focal vascular insults (5), and diffuse white matter (8) or gray matter (2) injury. At the brainstem level, cerebellar hypoplasia (4), Chiari malformation (1), peduncular lipoma (1), and cranial nerve abnormalities (1) were identified. Based on phenotype, family history, clinical course, and complementary examinations, the series included 59 patients with congenital malformation syndromes, 29 with diverse encephalopathies, and 2 with no apparent underlying disorder (Table 2). Except for 6 patients who died before age 6 months, patients were followed from age 6 months 1 day to age 2 years (16), from age 2 to 5 years (27), or until age >5 years (41). The patients often required mechanical ventilation or oxygen, long-term (6 months) tube feeding or gastrostomy, and home-care services or 756
Congenital Cranial Neuropathies
Outcome Orofacial EMG results
A
B
P*
2 affected nerves 3 affected nerves 4 or more affected nerves Involvement of lower nerves (IX–X or XII) Abnormal blink responses
13 10 3 14 11
26 15 23 50 29
NS NS 0.02 NS NS
EMG, electromyography; A, no handicap; B, neurological disabilities, respiratory assistance, prolonged enteral feeding, or death; NS, not statistically significant. *Fisher exact test.
specialized long-term care units. More than half of the patients had neurological disabilities. Clinical conditions were severe and worsened to death for 19 (21%) patients. The number of patients with poor outcome (neurological disabilities, need for respiratory assistance or prolonged enteral feeding) or death was significantly higher (P 5 0.02) when EMG identified a high number (4) of affected cranial nerves. However, there was no association of lower cranial nerve involvement with poor outcome (Table 3).
DISCUSSION
These findings suggest that congenital cranial neuropathies can be better characterized by combining clinical and electrodiagnostic data. EMG identified 3 times more facial palsies than clinical examination alone and revealed a high frequency of bilateral involvement. Moreover, EMG detected pharyngeal and lingual palsies that were difficult to diagnose clinically. This series brings information on the most frequent underlying disorders in young infants with multiple cranial neuropathies. Table 2 displays 2 main diagnostic groups, emphasizing that malformation syndromes may include brain dysgenesis and, conversely, that encephalopathies with prenatal onset can induce brain malformations. Diverse chromosomal aberrations and other genetic anomalies were found, without any preferential localization. Beside various and singular recognizable syndromes, Moebius syndrome and CHARGE (colobomata, heart disease, atresia of the choanae, retarded growth and development, genital hypoplasia, and ear anomalies or deafness) association constitute the only countable subset of patients for whom EMG changes help make a specific diagnosis. Moebius syndrome is characterized by bilateral facial and abducens palsies, frequently associated with additional cranial nerve palsies,5 and multiple MUSCLE & NERVE
November 2015
cranial neuropathies are typically reported in patients with CHARGE association.6,7 More than the two-thirds of patients with encephalopathies had pre- or perinatal vascular brain injuries. Although the brainstem and spinal cord are less vulnerable to ischemia than the cerebral cortex, pathologic studies have shown a pattern of brainstem damage after acute birth asphyxia,8 or prenatal ischemia.9 Moreover, the diencephalon, pons, and medulla oblongata are structural targets of neuronal loss in cases of severe birth asphyxia of the “cardiac arrest” type.10 Histological features compatible with fetal asphyxia have been described in the brainstem in premature newborns with respiratory failure and multiple cranial nerve palsies.11 Hemorrhage within the brainstem and cervical spinal cord has been reported in a term newborn resulting in absence of sucking, lingual atrophy, diaphragm paralysis, and EMG signs of denervation of the trapezius and sternocleidomastoid muscles.12 An autopsy of a child with facial diplegia, velopharyngeal incoordination, and ocular motor apraxia demonstrated neuronal depletion in cranial nerve nuclei and intact cerebral hemispheres.13 Orofacial dysfunction is frequent in children with cerebral palsy14,15 and is presumed to be of suprabulbar origin. In our series, neurogenic EMG signs were detected in muscles of the face, tongue, or soft palate in 14 patients with orofacial dysfunction, which was presumed to be due to suprabulbar vascular insult at preterm or term birth. In the same way, hypoglossal nerve involvement has been detected by EMG of the genioglossus muscle in children with periventricular leukomalacia or hemorrhagic infarction.16 In a brain MRI study of infants with perinatal hypoxicischemic events, oral motor dysfunction was strongly associated with brainstem tegmental lesions.17 The term “dorsal brainstem syndrome” has been introduced to summarize clinical conditions including cranial nerve palsies due to hypoxic-ischemic lesions at the brainstem level.18 Pathogenesis may relate to the watershed zone in the brainstem tegmentum between the terminal perfusion zones of the paramedian penetrating and long circumferential arteries.19 The other patients with encephalopathies had definite metabolic, toxic, or genetic diseases. Mitochondrial disorders are known to induce polyneuropathies, although their clinical and neurophysiological characteristics have not been adequately described, especially with regard to cranial nerve involvement.20 Prenatal exposure to alcohol may induce central nervous system dysgenesis, but cranial neuropathies have not been reported, except for optic atrophy. Brainstem dysgenesis with multiple cranial nerve involvement Congenital Cranial Neuropathies
has been reported in a child born to a cocaineaddicted mother.21 Regarding outcome, our results emphasize the poor prognosis of neonatal conditions with multiple cranial neuropathies. The extent of cranial nerve involvement had predictive value, with a larger number of nerves being associated with a poorer prognosis. Surprisingly, lower cranial nerves were not more frequently involved in patients with poor outcome or death, even though lingual or palatal palsies induce a risk for airway obstruction and aspiration. In conclusion, orofacial EMG is a useful complementary examination to highlight multiple cranial nerve involvement that can be clinically inapparent in newborns and young infants. Due to the close anatomic and functional relationships of the cranial nerves within the brainstem, their electrodiagnostic examination should not be limited to a single nerve or pathway. By demonstrating multiple cranial neuropathies, early orofacial EMG examination contributes to a better understanding of the neurologic origin of congenital facial weakness and dysphagia, and emphasizes the importance of malformation syndromes and neonatal encephalopathies as underlying etiologies. These study findings were presented at the annual meeting American Association of Neuromuscular and Electrodiagnostic Medicine, October–November 2014, Savannah, Georgia, USA. REFERENCES 1. Suresh PA, Deepa C. Congenital suprabulbar palsy: a distinct clinical syndrome of heterogeneous aetiology. Dev Med Child Neurol 2004; 46:617–625. 2. Renault F, Flores-Guevara R, Soupre V, Vazquez MP, Baudon JJ. Neurophysiological brainstem investigations in isolated Pierre Robin sequence. Early Hum Dev 2000;58:141–152. 3. Renault F, Baudon JJ, Galliani E, Flores-Guevara R, Marlin S, Garabedian EN, et al. Facial, lingual, and pharyngeal electromyography in infants with Pierre Robin sequence. Muscle Nerve 2011;43: 866–871. 4. Baudon JJ, Renault F, Goutet JM, Biran-Mucignat V, Morgant G, Garabedian EN, et al. Assessment of dysphagia in infants with facial malformations. Eur J Pediatr 2009;168:187–193. 5. Verzijl HT, van der Zwaag B, Cruysberg JR, Padberg GW. M€ obius syndrome redefined: a syndrome of rhombencephalic maldevelopment. Neurology 2003;61:327–333. 6. Roger G, Morisseau-Durand MP, van den Abbeele T, Nicollas R, Triglia JM, Narcy P, et al. The CHARGE association: the role of tracheotomy. Arch Otolaryngol Head Neck Surg 1999;125:33–38. 7. Blake KD, Hartshorne TS, Lawand C, Dailor AN, Thelin JW. Cranial nerve manifestations in CHARGE syndrome. Am J Med Genet Part A 2008;146A:585–592. 8. Leech RW, Alvord EC. Anoxic-ischemic encephalopathy in the human neonatal period. The significance of brainstem involvement. Arch Neurol 1977;34:109–113. 9. Arvold EC, Shaw CM. Congenital difficulties with swallowing and breathing associated with maternal polyhydramnios: neurocristopathy or medullary infarction? J Child Neurol 1989;4:299–306. 10. Govaert P. Cranial haemorrhage in the term newborn infant. Clinics in developmental medicine. Vol. 129. London: Mac Keith Press; 1993. 223 p. 11. Wilson ER, Mirra SS, Schwartz JF. Congenital diencephalic and brainstem damage: neuropathologic study of three cases. Acta Neuropathol 1982;57:70–74. 12. Blazer S, Hemli JA, Sujov PO, Braun J. Neonatal bilateral diaphragmatic paralysis caused by brainstem haemorrhage. Arch Dis Child 1989;64:50–52. 13. Roig M, Gratacos M, Vazquez E, del Toro M, Foguet A, Ferrer I, et al. Brainstem dysgenesis: report of five patients with congenital
MUSCLE & NERVE
November 2015
757
14. 15. 16.
17.
hypotonia, multiple cranial nerve involvement, and ocular motor apraxia. Dev Med Child Neurol 2003;45:489–493. Reilly S, Skuse D, Poblete X. Prevalence of feeding problems and oral motor dysfunction in children with cerebral palsy: a community survey. J Pediatr 1996;129:877–882. Parkes J, Hill N, Platt MJ, Donnelly C. Oromotor dysfunction and communication impairments in children with cerebral palsy: a register study. Dev Med Child Neurol 2010;52:1113–1119. Vijayakumar K, Rockett J, Ryan M, Harris R, Pitt M, Devile C. Experience of using electromyography of the genioglossus in the investigation of paediatric dysphagia. Dev Med Child Neurol 2012;54:1127– 1132. Quattrocchi CC, Longo D, Delfino LN, Cilio MR, Piersigilli F, Capua MD, et al. Dorsal brainstem syndrome: MR imaging location of brain-
758
Congenital Cranial Neuropathies
18. 19. 20. 21.
stem tegmental lesions in neonates with oral motor dysfunction. Am J Neuroradiol 2010;31:1438–1442. Hiyane M, Saito Y, Saito T, Komaki H, Nakagawa E, Sugai K, et al. A case of bulbar type cerebral palsy: representative symptoms of dorsal brainstem syndrome. Brain Dev 2012;34:787–791. Sarnat HB. Watershed infarcts in the fetal and neonatal brainstem. An aetiology of central hypoventilation, dysphagia, Moebius syndrome and micrognathia. Eur J Pediatr Neurol 2004;8:71–87. Menezes MP, Ouvrier RA. Peripheral neuropathy associated with mitochondrial disease in children. Dev Med Child Neurol 2012;54: 407–414. Boix H, Ortega-Aznar A, Vazquez E, Salcedo S, Roig-Quilis M. Brainstem dysgenesis in an infant prenatally exposed to cocaine. Pediatr Neurol 2010;42:295–297.
MUSCLE & NERVE
November 2015
Clinical Neurophysiology Unit, H^ opital Armand-Trousseau, Assistance Publique-H^ opitaux de Paris, 28 avenue Arnold-Netter, 75571 Paris 12, France 2 Facultad de Medicina, Universidad Nacional Mayor de San Marcos, Lima, Peru 3 Faculte de Medecine Ren e-Descartes, Universit e Paris 5, Paris, France 4 Department of Maxillofacial Surgery, H^ opital Necker-Enfants Malades, Assistance Publique-H^ opitaux de Paris, Paris, France Accepted 27 February 2015 ABSTRACT: Introduction: The aim of this study was to assess diagnoses and outcomes of infants with 2 or more cranial neuropathies identified using orofacial electromyography (EMG). Methods: This retrospective study involved 90 patients. Diagnoses took into account clinical, radiological, and genetic data. EMG examined the orbicularis oculi, genioglossus, and levator veli palatini muscles, and blink responses. To evaluate outcome, neurological disability, respiratory complications, and feeding difficulties were recorded. Results: The patients had malformation syndromes (59), encephalopathies (29), or no underlying disorders (2). Neurogenic EMG signs were detected in a mean of 4 muscles, reflecting a mean of 3 affected nerves. EMG identified a higher number of neuropathies than clinical examination alone (82 vs. 31, facial; 56 vs. 2, pharyngeal; 25 vs. 3, hypoglossal). Poor outcome and death were more frequent when EMG identified 4 affected nerves (P 5 0.02). Conclusion: EMG highlights multiple cranial neuropathies that can be clinically silent in infants with malformation syndromes or encephalopathies. Muscle Nerve 52: 754–758, 2015
Electrodiagnostic
examination of paired cranial nerves and orofacial muscles is a method for assessing congenital facial weakness, orofacial malformations, and dysphagia. Because underlying disorders have numerous causes and varying degrees of severity, the diagnoses and prognoses of bulbar dysfunction are not yet clearly defined.1 Orofacial sensorimotor function is not easy to assess clinically in newborn and young infants, who have small brainstem structures and cranial nerves that are challenging to study using MRI. Electrodiagnostic studies investigate paired cranial nerves using blink responses (BRs) and needle electromyography (EMG) of muscles of the face, tongue, and soft palate.. These methods are used to identify axonal loss and demyelination, and to define their extent and severity. We have previously reported the results of orofacial EMG in patients with Pierre Robin sequence (PRS)2,3 and in Abbreviations: ANOVA, analysis of variance; BR, blink response; CHARGE, colobomata, heart disease, atresia of the choanae, retarded growth and development, genital hypoplasia, and ear anomalies or deafness; CT, computed tomography; EMG, electromyography; PRS, Pierre Robin sequence Key words: blink responses; brainstem; child; cranial nerves; electromyography Correspondence to: F. Renault; e-mail: [email protected] C 2015 Wiley Periodicals, Inc. V
Published online 3 March 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/mus.24636
754
Congenital Cranial Neuropathies
patients with other orofacial malformation syndromes.4 In these series, the patients with cranial nerve involvement had more frequent respiratory complications and required longer periods of enteral feeding. The aim of the present study was to retrospectively analyze final diagnoses and outcomes in infants with multiple cranial nerve involvement identified using early orofacial EMG. METHODS Study Design. The patients were investigated between birth and age 6 months because of congenital dysphagia or orofacial malformations. They were referred to our pediatric neurophysiology unit between July 1, 1995 and December 31, 2010. For each patient, we collected clinical, radiological, and genetic data contributing to the diagnosis. To evaluate outcome, we recorded the presence or absence of neurological disabilities (mental or sensorimotor handicap) and the need for respiratory assistance (oxygen, mechanical ventilation, tracheostomy), gastrostomy, and prolonged enteral feeding. Given the retrospective nature of the study and the unequal durations of follow-up, a standardized neurodevelopmental evaluation was not possible. This study was undertaken in accordance with French regulations. Approval by the ethics committee of our university was waived. Informed parental or guardian consent was obtained for every investigation. The data were de-identified before analysis. Patients. During the 15-year period of the study, 997 children were referred to our clinical neurophysiology unit for investigation of orofacial functions using EMG. Multiple cranial nerve involvement was diagnosed in 109 patients with neurogenic EMG signs in the territories of 2 or more cranial nerves. Because 19 patients were lost to follow-up, the series finally included 90 patients (42 boys, 48 girls). Phenotypes were established after clinical examination by a neonatologist, a maxillofacial surgeon, and a clinical geneticist. Ocular, skeletal, renal, and cardiac malformations were identified by ophthalmologic examination, X-ray, and ultrasound scans. The brain was investigated by ultrasound scan, computed tomography MUSCLE & NERVE
November 2015
decreased by at least 30%) (Fig. 1). Blink responses were recorded in the orbicularis oculi muscle in response to electrical stimulation (square pulse: duration 0.2 ms, intensity 10 mA) of the supraorbital nerve. We recorded 6 successive responses to confirm the presence or absence of R1 and R2 components. We considered an asymmetry in R1 latency of up to 3 ms to be normal.
FIGURE 1. Needle EMG recordings of the depressor anguli oris muscle in patients of the series: (A) normal pattern; (B) low-amplitude, reflecting muscle hypoplasia; and (C) neurogenic reduced recruitment.
(CT), or MRI. Upper airways were examined using laryngoscopy. A karyotype was obtained for every patient; deletion at 22q11 was found in 78 patients. Electrodiagnostic Studies. All electrodiagnostic studies were performed by the same examiner (F.R.) using the same methods and were part of every clinical evaluation. EMG was not carried out before gestational age 37 weeks. Conventional EMG was used to study muscles of the face, tongue, and soft palate, at rest and with voluntary contraction during crying. Muscle activity was detected using small disposable concentric needle electrodes (25 mm 3 30G; Viasys Healthcare, Madison, Wisconsin). The needle insertion point was anesthetized using lidocaine–prilocaine cream applied to the skin, and by touching the mucosa with a 2% solution of lidocaine. Bilateral EMG examination systematically included the orbicularis oculi, genioglossus, and levator veli palatini muscles. The orbiculis oris and depressor anguli oris muscles were also investigated in 39 patients. After fasting the patient for up to 4 hours, crying and sucking movements created bursts of EMG activity that could be recorded from oral and facial muscles. Analyses focused on the recruitment pattern and BRs. Markers were placed manually on the trace for measuring amplitudes and latencies. The maximum amplitude was measured from the highest negative peak to the lowest positive peak of the largest burst. Traces were classified as: normal interference pattern; neurogenic single or reduced interference pattern; or low-amplitude full interference pattern (maximum amplitude Congenital Cranial Neuropathies
Statistical Analysis. Numeric variables are presented as mean and range; categorical variables are presented as rates. Comparison of 2 independent groups was performed using the Mann–Whitney Utest for numeric variables and the Fisher exact test for categorical variables. Comparison of >2 groups was performed by 1-way analysis of variance (ANOVA) test for numeric variables and by chisquare test for categorical variables. All hypotheses were constructed as 2-tailed, with P < 0.05 considered significant. RESULTS
Dysphagia and orofacial malformations were the most frequent presenting symptoms leading to referral for electrodiagnostic studies (Table 1). We studied the EMG recordings of 609 muscles, reflecting the function of 180 facial nerves, 172 pharyngeal plexi, and 124 hypoglossal nerves. Median age at EMG examination was 41 days (range 2–172). Neurogenic EMG signs were detected in a mean of 4 muscles, reflecting a mean of 3 affected nerves. The facial nerve was involved in 82 of 90 patients, bilaterally in 64; BRs were altered in 40 of 63 patients, bilaterally absent (25) or asymmetrical (15). The pharyngeal plexus was affected in 56 of Table 1. Clinical features in 90 infants with multiple cranial neuropathies. Clinical signs and course
Number of patients (%)
Presenting symptoms Sucking or swallowing disorders Orofacial malformations Amimia, hypomimia Glossoptosis Cranial nerve manifestations Facial weakness (asymmetry or diplegia) Oculomotor palsy Tongue asymmetry or atrophy Soft palate hypomobility or asymmetry Complications Aspiration, choking episodes Need for gastrostomy Outcome* Respiratory assistance Enteral feeding during >6 months Neurological disabilities Death
56 38 28 18
of of of of
90 90 90 90
(62) (42) (31) (20)
31 18 3 2
of of of of
90 90 90 90
(34) (20) (3) (2)
42 of 90 (47) 37 of 90 (41) 17 60 45 19
of of of of
84 84 84 90
(20) (71) (54) (21)
*Excluding the 6 patients who died before age 6 months.
MUSCLE & NERVE
November 2015
755
Table 2. Diagnoses in 90 patients with multiple cranial neuropathies.
Table 3. Correlation between the extent or localization of cranial nerve involvement and outcome. n
Malformation syndromes Genetic anomalies* Moebius syndrome CHARGE association Other recognizable syndromes† Unidentified Encephalopathies Preterm (25–34 weeks) with vascular brain injury Localized brainstem ischemia Birth asphyxia at full term Born to addicted mother (alcohol, drugs) Progressive encephalopathy‡ No apparent underlying disorder
59 14 8 7 15 15 29 7 6 7 4 5 2
CHARGE association: colobomata, heart disease, atresia of the choanae, retarded growth and development, genital hypoplasia, and ear anomalies or deafness. *[47XXX]; [46XY, ins (12;11) (q14;q13.5q21)]; [46XX t(7;17) (q21.2;q22)]; [del 18q21]; [46XY, inv(2).ish 22q(tuple 132)]; [46XY, rec(9) del p(p22pter), dup q(q32qter)]; [dupl 4ter]; [del 18pter, tri 1qter]; [tri 2q, t(9p) (2q)] ; [45X, t(X;1) (p22.3;p34)]; [46XY/46fra (Xq)Y]; [del 22q11]; [del 22q11]; SOX2 mutation. † Including: acro-reno-ocular, Cornelia de Lange, ectodermic-anhydroticdysplasia, Fryns, Goldenhar, hydranencephaly, Larsen, Noonan, orofaciodigital I, Rubinstein-Taybi, Stickler, Townes-Brocks, Treacher-Collins (2 patients), and Wiedemann-Beckwith syndromes. ‡
Including 2 with mitochondrial disorder.
89 patients, bilaterally in 32. The hypoglossal nerve was involved in 25 of 90 patients, bilaterally in 11. The lower cranial nerves (IX-X, XII) were not involved in 25 of 90 patients. Five patients showed ipsilateral multiple cranial nerve involvement. Low-amplitude interference patterns were additionally recorded in 1 or more muscles of the face, tongue, or soft palate in 19 patients, including 17 with malformation syndromes. Brain MRI showed 20 normal and 29 abnormal imaging patterns, including malformations (13), focal vascular insults (5), and diffuse white matter (8) or gray matter (2) injury. At the brainstem level, cerebellar hypoplasia (4), Chiari malformation (1), peduncular lipoma (1), and cranial nerve abnormalities (1) were identified. Based on phenotype, family history, clinical course, and complementary examinations, the series included 59 patients with congenital malformation syndromes, 29 with diverse encephalopathies, and 2 with no apparent underlying disorder (Table 2). Except for 6 patients who died before age 6 months, patients were followed from age 6 months 1 day to age 2 years (16), from age 2 to 5 years (27), or until age >5 years (41). The patients often required mechanical ventilation or oxygen, long-term (6 months) tube feeding or gastrostomy, and home-care services or 756
Congenital Cranial Neuropathies
Outcome Orofacial EMG results
A
B
P*
2 affected nerves 3 affected nerves 4 or more affected nerves Involvement of lower nerves (IX–X or XII) Abnormal blink responses
13 10 3 14 11
26 15 23 50 29
NS NS 0.02 NS NS
EMG, electromyography; A, no handicap; B, neurological disabilities, respiratory assistance, prolonged enteral feeding, or death; NS, not statistically significant. *Fisher exact test.
specialized long-term care units. More than half of the patients had neurological disabilities. Clinical conditions were severe and worsened to death for 19 (21%) patients. The number of patients with poor outcome (neurological disabilities, need for respiratory assistance or prolonged enteral feeding) or death was significantly higher (P 5 0.02) when EMG identified a high number (4) of affected cranial nerves. However, there was no association of lower cranial nerve involvement with poor outcome (Table 3).
DISCUSSION
These findings suggest that congenital cranial neuropathies can be better characterized by combining clinical and electrodiagnostic data. EMG identified 3 times more facial palsies than clinical examination alone and revealed a high frequency of bilateral involvement. Moreover, EMG detected pharyngeal and lingual palsies that were difficult to diagnose clinically. This series brings information on the most frequent underlying disorders in young infants with multiple cranial neuropathies. Table 2 displays 2 main diagnostic groups, emphasizing that malformation syndromes may include brain dysgenesis and, conversely, that encephalopathies with prenatal onset can induce brain malformations. Diverse chromosomal aberrations and other genetic anomalies were found, without any preferential localization. Beside various and singular recognizable syndromes, Moebius syndrome and CHARGE (colobomata, heart disease, atresia of the choanae, retarded growth and development, genital hypoplasia, and ear anomalies or deafness) association constitute the only countable subset of patients for whom EMG changes help make a specific diagnosis. Moebius syndrome is characterized by bilateral facial and abducens palsies, frequently associated with additional cranial nerve palsies,5 and multiple MUSCLE & NERVE
November 2015
cranial neuropathies are typically reported in patients with CHARGE association.6,7 More than the two-thirds of patients with encephalopathies had pre- or perinatal vascular brain injuries. Although the brainstem and spinal cord are less vulnerable to ischemia than the cerebral cortex, pathologic studies have shown a pattern of brainstem damage after acute birth asphyxia,8 or prenatal ischemia.9 Moreover, the diencephalon, pons, and medulla oblongata are structural targets of neuronal loss in cases of severe birth asphyxia of the “cardiac arrest” type.10 Histological features compatible with fetal asphyxia have been described in the brainstem in premature newborns with respiratory failure and multiple cranial nerve palsies.11 Hemorrhage within the brainstem and cervical spinal cord has been reported in a term newborn resulting in absence of sucking, lingual atrophy, diaphragm paralysis, and EMG signs of denervation of the trapezius and sternocleidomastoid muscles.12 An autopsy of a child with facial diplegia, velopharyngeal incoordination, and ocular motor apraxia demonstrated neuronal depletion in cranial nerve nuclei and intact cerebral hemispheres.13 Orofacial dysfunction is frequent in children with cerebral palsy14,15 and is presumed to be of suprabulbar origin. In our series, neurogenic EMG signs were detected in muscles of the face, tongue, or soft palate in 14 patients with orofacial dysfunction, which was presumed to be due to suprabulbar vascular insult at preterm or term birth. In the same way, hypoglossal nerve involvement has been detected by EMG of the genioglossus muscle in children with periventricular leukomalacia or hemorrhagic infarction.16 In a brain MRI study of infants with perinatal hypoxicischemic events, oral motor dysfunction was strongly associated with brainstem tegmental lesions.17 The term “dorsal brainstem syndrome” has been introduced to summarize clinical conditions including cranial nerve palsies due to hypoxic-ischemic lesions at the brainstem level.18 Pathogenesis may relate to the watershed zone in the brainstem tegmentum between the terminal perfusion zones of the paramedian penetrating and long circumferential arteries.19 The other patients with encephalopathies had definite metabolic, toxic, or genetic diseases. Mitochondrial disorders are known to induce polyneuropathies, although their clinical and neurophysiological characteristics have not been adequately described, especially with regard to cranial nerve involvement.20 Prenatal exposure to alcohol may induce central nervous system dysgenesis, but cranial neuropathies have not been reported, except for optic atrophy. Brainstem dysgenesis with multiple cranial nerve involvement Congenital Cranial Neuropathies
has been reported in a child born to a cocaineaddicted mother.21 Regarding outcome, our results emphasize the poor prognosis of neonatal conditions with multiple cranial neuropathies. The extent of cranial nerve involvement had predictive value, with a larger number of nerves being associated with a poorer prognosis. Surprisingly, lower cranial nerves were not more frequently involved in patients with poor outcome or death, even though lingual or palatal palsies induce a risk for airway obstruction and aspiration. In conclusion, orofacial EMG is a useful complementary examination to highlight multiple cranial nerve involvement that can be clinically inapparent in newborns and young infants. Due to the close anatomic and functional relationships of the cranial nerves within the brainstem, their electrodiagnostic examination should not be limited to a single nerve or pathway. By demonstrating multiple cranial neuropathies, early orofacial EMG examination contributes to a better understanding of the neurologic origin of congenital facial weakness and dysphagia, and emphasizes the importance of malformation syndromes and neonatal encephalopathies as underlying etiologies. These study findings were presented at the annual meeting American Association of Neuromuscular and Electrodiagnostic Medicine, October–November 2014, Savannah, Georgia, USA. REFERENCES 1. Suresh PA, Deepa C. Congenital suprabulbar palsy: a distinct clinical syndrome of heterogeneous aetiology. Dev Med Child Neurol 2004; 46:617–625. 2. Renault F, Flores-Guevara R, Soupre V, Vazquez MP, Baudon JJ. Neurophysiological brainstem investigations in isolated Pierre Robin sequence. Early Hum Dev 2000;58:141–152. 3. Renault F, Baudon JJ, Galliani E, Flores-Guevara R, Marlin S, Garabedian EN, et al. Facial, lingual, and pharyngeal electromyography in infants with Pierre Robin sequence. Muscle Nerve 2011;43: 866–871. 4. Baudon JJ, Renault F, Goutet JM, Biran-Mucignat V, Morgant G, Garabedian EN, et al. Assessment of dysphagia in infants with facial malformations. Eur J Pediatr 2009;168:187–193. 5. Verzijl HT, van der Zwaag B, Cruysberg JR, Padberg GW. M€ obius syndrome redefined: a syndrome of rhombencephalic maldevelopment. Neurology 2003;61:327–333. 6. Roger G, Morisseau-Durand MP, van den Abbeele T, Nicollas R, Triglia JM, Narcy P, et al. The CHARGE association: the role of tracheotomy. Arch Otolaryngol Head Neck Surg 1999;125:33–38. 7. Blake KD, Hartshorne TS, Lawand C, Dailor AN, Thelin JW. Cranial nerve manifestations in CHARGE syndrome. Am J Med Genet Part A 2008;146A:585–592. 8. Leech RW, Alvord EC. Anoxic-ischemic encephalopathy in the human neonatal period. The significance of brainstem involvement. Arch Neurol 1977;34:109–113. 9. Arvold EC, Shaw CM. Congenital difficulties with swallowing and breathing associated with maternal polyhydramnios: neurocristopathy or medullary infarction? J Child Neurol 1989;4:299–306. 10. Govaert P. Cranial haemorrhage in the term newborn infant. Clinics in developmental medicine. Vol. 129. London: Mac Keith Press; 1993. 223 p. 11. Wilson ER, Mirra SS, Schwartz JF. Congenital diencephalic and brainstem damage: neuropathologic study of three cases. Acta Neuropathol 1982;57:70–74. 12. Blazer S, Hemli JA, Sujov PO, Braun J. Neonatal bilateral diaphragmatic paralysis caused by brainstem haemorrhage. Arch Dis Child 1989;64:50–52. 13. Roig M, Gratacos M, Vazquez E, del Toro M, Foguet A, Ferrer I, et al. Brainstem dysgenesis: report of five patients with congenital
MUSCLE & NERVE
November 2015
757
14. 15. 16.
17.
hypotonia, multiple cranial nerve involvement, and ocular motor apraxia. Dev Med Child Neurol 2003;45:489–493. Reilly S, Skuse D, Poblete X. Prevalence of feeding problems and oral motor dysfunction in children with cerebral palsy: a community survey. J Pediatr 1996;129:877–882. Parkes J, Hill N, Platt MJ, Donnelly C. Oromotor dysfunction and communication impairments in children with cerebral palsy: a register study. Dev Med Child Neurol 2010;52:1113–1119. Vijayakumar K, Rockett J, Ryan M, Harris R, Pitt M, Devile C. Experience of using electromyography of the genioglossus in the investigation of paediatric dysphagia. Dev Med Child Neurol 2012;54:1127– 1132. Quattrocchi CC, Longo D, Delfino LN, Cilio MR, Piersigilli F, Capua MD, et al. Dorsal brainstem syndrome: MR imaging location of brain-
758
Congenital Cranial Neuropathies
18. 19. 20. 21.
stem tegmental lesions in neonates with oral motor dysfunction. Am J Neuroradiol 2010;31:1438–1442. Hiyane M, Saito Y, Saito T, Komaki H, Nakagawa E, Sugai K, et al. A case of bulbar type cerebral palsy: representative symptoms of dorsal brainstem syndrome. Brain Dev 2012;34:787–791. Sarnat HB. Watershed infarcts in the fetal and neonatal brainstem. An aetiology of central hypoventilation, dysphagia, Moebius syndrome and micrognathia. Eur J Pediatr Neurol 2004;8:71–87. Menezes MP, Ouvrier RA. Peripheral neuropathy associated with mitochondrial disease in children. Dev Med Child Neurol 2012;54: 407–414. Boix H, Ortega-Aznar A, Vazquez E, Salcedo S, Roig-Quilis M. Brainstem dysgenesis in an infant prenatally exposed to cocaine. Pediatr Neurol 2010;42:295–297.
MUSCLE & NERVE
November 2015
