British Journal of Anaesthesia 83 (6): 951-5 (1999) BJACASE REPORTS

Perinatal management of a neonate with airway obstructioncaused by rhabdomyosarcoma of the tongue

V. A. Skelton and A. Goodwin

Department of Anaesthesia, King's College Hospital, Denmark Hill, London SE5 9RS, UK

Intra-oral masses in neonates can seriously compromise the airway, potentially causing hypoxiaand death if not recognized and managed appropriately. We report a case in which an intra-oral mass was diagnosed on antenatal ultrasound scan. Preparation for delivery involved amultidisciplinary team approach, with a strategy for management at delivery. The child wasdelivered by elective Caesarean section and had a patent airway. A tracheostomy was performedimmediately after delivery. The infant underwent a debulking procedure 3 weeks afterbirth. A histological diagnosis of embryonal rhabdomyosarcoma was made and a course ofchemotherapy commenced. The child had a partial response to treatment with considerableshrinkage of the tongue mass. We discuss the management options in neonates with intra-oralmasses to provide an adequate airway and maintain fetal oxygenation. The differential diagnosisof fetal oral masses is reviewed.

BrJAnaesth 1999; 83: 951-5

Keywords: neonates; airway, obstruction; complications, rhabdomyosarcoma

Accepted for publication: June 25, 1999

Advances in prenatal ultrasound examination have led to aconsiderable increase in the number of diagnoses of congen-ital abnormalities before birth. Before this advance, unex-pected airway problems relating to head or neck massesin the neonate led to potentially life-threatening airwayobstruction and hypoxia. We describe a case in which anintra-oral mass was diagnosed on antenatal ultrasound scan.We discuss the subsequent management for delivery, andreview the options available.

Rhabdomyosarcoma (RMS) have a relative predilectionfor the head and neck regions, but involvement of thetongue is particularly rare. In the large series of theIntergroup Rhabdomyosarcoma Studies (IRS) I, II and III,only seven cases of RMS of the tongue were documented,with none in the neonatal period.1

Case reportA 29-yr-old, gravida 3, para 2 woman presented to herlocal hospital at 38 weeks' gestation with possible rupturedmembranes. Her pregnancy had been uneventful until thattime. She was discharged after a normal cardiotocographictrace and vaginal examination, but was readmitted 24 hlater because of worsening symptoms. She was started onantibiotics because of a tentative diagnosis of pyelonephritis.An ultrasound scan was performed to confirm fetal well-

being which showed a mass measuring approximately60X49X51 mm on the fetal tongue (Fig. 1). The motherwas transferred to the fetal assessment unit at our hospitalwhere a further scan confirmed the presence of a large massat the base of the tongue. This was pushing the tongueforward and was considered to be potentially obstructingthe airway. Amniotic fluid volume was normal. Lungdevelopment and tracheal diameter were also considerednormal, indicating that if airway obstruction was present,it was not complete.

A multidisciplinary team was convened comprisingobstetricians, neonatologists, paediatric anaesthetists andpaediatric ENT surgeons and an action plan formed beforedelivery of the baby by elective Caesarean section. As theprenatal ultrasound indicated that airway obstruction wasunlikely to be complete, the decision was made to deliverthe baby completely, with division of the umbilical cord,before inspection of the airway by a paediatric anaesthetistusing direct laryngoscopy. (The alternative was to performlaryngoscopy or tracheostomy with the placental circulationintact.) The paediatric ENT surgeons were to be readyto operate with all the equipment necessary to performimmediate tracheostomy if the airway was compromised.The neonatologists were to assist in resuscitation of thebaby and provision of vascular access.

In the event, an otherwise healthy male was delivered

© British Journal of Anaesthesia

Skelton and Goodwin

Fig 1 The antenatal ultrasound scan of the fetus. The mass on the tongue is indicated by A.

and cried immediately, indicating a patent airway, beforethe umbilical cord was clamped. The cord was sub-sequently divided and the baby transferred to theneonatologists with Apgar scores of 8 and 10 at 1 and5 min. His tongue was hugely enlarged with a non-compressible, non-pulsatile soft tissue mass protrudingfrom his mouth (Fig. 2a, b). It was felt that he mayencounter respiratory difficulties in the neonatal period,especially if his tongue enlarged further, and the decisionwas made to perform a tracheostomy immediately.Intravenous and intra-arterial access were secured via theumbilical vessels, atropine 100 |ig was given andlaryngoscopy performed. Intubation proved easy, with agrade I view at laryngoscopy. Anaesthesia was inducedwith an oxygen-nitrous oxide-isoflurane mixture and abolus dose of fentanyl. A tracheostomy was performedand a 3.5-mm paediatric Shiley tracheostomy tube inserted.

Postoperative cranial and abdominal ultrasound scanswere normal and an MRI scan confirmed a heterogeneoussoft tissue mass in the mid and anterior part of thetongue. Surgery to debulk the tongue was performedwithout complications 3 weeks later. Subsequent histologyreported a well differentiated embryonal rhabdomyosar-coma (RMS) of keratomuscular origin. Further investi-gation defined the tumour as stage I with no spread, and

a course of chemotherapy was commenced. After 8months of chemotherapy there was a partial response totreatment with considerable shrinkage of the tongue mass.Complete excision of the tumour will be considered afterfurther chemotherapy.

DiscussionCongenital masses of the head and neck in the neonate'can seriously compromise the airway. A review ofneonatal airway obstruction from teratomas reported 20%mortality, with most cases occurring before the eraof prenatal ultrasound diagnosis.2 Increasingly refinedultrasound imaging can alert clinicians to the potentialproblems, and may give an indication of the degree ofobstruction. Airway obstruction is more likely if anultrasound diagnosis of a head or neck mass is associatedwith polyhydramnios, probably because of the inabilityof the fetus to swallow because of oesophageal andtracheal compression.3 4 Extension of the fetal head inutero may also indicate airway compromise, the masslimiting head flexion. In complete airway obstruction,caused mainly by laryngeal stenosis, prenatal ultrasounddemonstrates a dilated trachea, large echogenic lungscaused by filling with water, flattened or inverted

952

Perinatal management of rhabdomyosarcoma of the tongue

Fig

I2 The neonate after tracheostomy to secure the airway. The large tongue tumour protrudes form the oral cavity.

diaphragms and ascites. This has been termed thecongenital high airway obstruction (CHAOS) syndrome.5 6

Our case emphasizes the importance of a multidisciplin-ary approach to the management of these patients witha clear action plan for each team. The options availableto us were:

(1) Partial (head and shoulders) delivery of the baby,maintaining the fetal-placental circulation with uterinerelaxation to prevent separation of the placenta, andperforming direct laryngoscopy or tracheostomy tosecure the airway before delivery of the baby's torso.

(2) Complete delivery of the baby, but maintaining

953

Skelton and Goodwin

uterine relaxation to prevent separation of the placentaand inspection of the airway before division of theumbilical cord.

(3) Complete delivery of the baby with division of theumbilical cord and subsequent laryngoscopy orimmediate tracheostomy if the airway was comprom-ised.

The first option was considered as fetal oxygenation viathe placenta can be maintained while the airway is secured.It was rejected on the basis that prenatal ultrasound did notindicate complete airway obstruction or obstruction of thetrachea. The method is not without hazards to the motherand performance of laryngoscopy or tracheostomy withinthe confines of the mother's abdominal incision can provetechnically difficult. The technique of placental supportshould be considered however when the antenatal ultrasoundindicates complete airway obstruction or lesions that willimpede surgical access for rapid tracheostomy, such asneck masses.

The second option of complete delivery of the baby,but maintaining the maternal-fetal circulation by uterinerelaxation was similarly rejected. There is greater risk ofpremature placental separation with complete rather thanpartial delivery, and transfusion of blood from the fetus tothe placenta may occur, as may excessive maternal bloodloss. The fetus should be maintained level with the placentaafter this type of delivery, but positioning may be limitedby the length of the umbilical cord.7

The third option was considered safest for both themother and baby. It was technically easier to access theairway with the baby fully delivered, and with the supportof ENT surgeons skilled in performing emergency trach-eostomy.

We were fortunate in this case that the baby cried ondelivery before division of the umbilical cord, allowingtime to secure the airway with the reassurance that oxygena-tion was maintained. The majority of the mass was in theanterior part of the tongue pulling it forward and probablyassisting airway patency. The decision to perform an awakelaryngoscopy was made on the basis that inhalation induc-tion, the safer option for a potentially obstructed airway,would prove technically difficult because of the inability tohold a mask over the protruding tongue. I.v. induction wasrejected because of the possible difficulty of maintainingthe airway and oxygenation if intubation proved impossible.We were fortunate that intubation proved surprisingly easy.

The ex utero intrapartum treatment (the EXIT procedure),8

also termed the OOPS procedure (operation on placentalsupport)9 was first described in animals,10 but has sincebeen used successfully to secure the airway of neonateswith head and neck masses.7 n"14 It allows up to 60 minof cord perfusion before utero-placental gas exchangebegins to deteriorate.15 During progress from the second tothe third stage of labour, delivery of the fetus stimulatesuterine contractions initiating placental separation. With theneonate's first breath, systemic vascular resistance and

pulmonary blood flow increase markedly, the ductus art-eriosus closes and neonatal circulation begins. With theEXIT or OOPS procedure, Caesarean section is performedunder general anaesthesia using a halogenated inhalationagent to relax the uterus. The fetus is either fully or partiallydelivered, but the maternal-fetal circulation is maintainedwithout division of the umbilical cord until the airway hasbeen secured. The uterus can be relaxed pharmacologicallywith either glyceryl trinitrate or ritodrine to prevent prema-ture placental separation and thus maintain fetoplacentalflow and oxygen delivery. Skarsgard and colleagues give aclear account of the procedure.9

The differential diagnosis of head and neck massesin the neonate includes cystic hygromas, haemangiomas,congenital goitres, choristomas, epignathus, neuroblas-tomas, granular cell tumours, retinoblastomas, mucoceles,cervical teratoma, branchial cleft cysts, laryngoceles,rhabdomyomas and rhabdomyosarcomas (RMS).13 A raremultigenic disorder, the Beck with-Wiedemann syndrome,associated with dysregulation of the expression of genesinvolving growth and the cell cycle, can result in macro-glossia associated with exomphalos, gigantism andhypoglycaemia.16"18 In this syndrome, polyhydramnios withaccelerated growth between 25 and 36 weeks' gestation isseen commonly on antenatal ultrasound scans, but fetaltongue protrusion may not be identified.19

RMS of the tongue in the neonate appears to be particu-larly rare. The intergroup rhabdomyosarcoma studies (IRS)I, II and III reported just seven cases of RMS of the tongue(0.34% of cases), with an age range of 2 months to 18 yrat diagnosis and all with the tumour at the base of thetongue.1 A study in Johannesburg over 25 yr identifiedeight cases of oral RMS, but none was of the tongue andthe mean age of patients was 14 (range 1-^6) yr. A literature'review found 113 cases of RMS involving the oral andpara-oral regions of which 16 involved the tongue.20 RMScan be categorized histologically into alveolar, embryonaland pleomorphic subtypes, with approximately 70% beingembryonal. Histology in our patient identified RMS of theembryonal subtype. The prognosis of embryonal RMS is-good; 89 children with localized RMS of orofacial andlaryngopharyngeal sites had a 3-yr actuarial survival of83%.21 Use of multidisciplinary therapy involving surgery,radiotherapy and chemotherapy has significantly improvedlong-term survival with better cosmetic and functionalresults. However, chemotherapy and radiotherapy in chil-dren may interfere with the development of the dentitionor the jaws, resulting in microdontia, hypoplastic enamel,underdeveloped roots, radiation caries and micrognathia.20

Prenatal magnetic resonance imaging (MRI) can be usefulto provide information on the anatomy of a mass in relationto the airway and may have been useful in our patient.15

Excellent images can be obtained with fast scanning tech-niques which combat motion artefact from the fetus moving,and minimize the necessity of sedating the mother. Forslower scans, which require a stationary patient, either

954

Perinatal management of rhabdomyosarcoma of the tongue

maternal sedation or direct injection of neuromuscularblocking agents to the fetus have been used safely.7 22 MRIscans also provide more accurate diagnoses of the aetiologyof fetal masses than ultrasound.7 15 22 23 24

Our report emphasizes the importance of a multidisciplin-ary team approach to the delivery of a neonate withsuspected airway obstruction, and careful consideration ofthe options available to maintain oxygenation while securingthe airway.

References1 Kodet R, Fajstavr J, Kabelka Z, Koutecky J, Eckschlager T,

Newton WA jr. Is fetal cellular rhabdomyosarcoma an entity ora differentiated rhabdomyosarcoma? A study of patients withrhabdomyoma of the tongue and sarcoma of the tongue enrolledin the Intergroup Rhabdomyosarcoma Studies I, II and III. Cancer1991; 67: 2907-13

2 Zerella JT, Finberg FJ. Obstruction of the neonatal airway fromteratomas. Surg Gynaecol Obstet 1990; 170: 126-31

3 Shipp TD, Bromley B, Benacerraf B. The ultrasonographicappearance and outcome for fetuses with masses distorting thefetal face. J Ultrasound Med 1995; 14: 673-8

4 Rosenfeld CR, Coin CD, Duenhoelter JH. Fetal cervical teratomaas a cause of polyhydramnios. Pediatrics 1979; 64: 176-8

' 5 Hedrick MH, Ferro MM, Filly RA, Flake AW, Harrison MR, AdzickNS. Congenital high airway obstruction syndrome (CHAOS): apotential for perinatal intervention. J Pediatr Surg 1994; 29: 271-4

6 Richards DS, Yancey MK, Duff P, Stieg FH. The perinatalmanagement of severe laryngeal stenosis. Obstet Gynecol 1992; 80:537-40

7 Catalano PJ, Urken ML, Alvarez M, et at. New approach to themanagement of airway obstruction in 'high risk' neonates. ArchOtolaryngol Head Neck Surg 1992; I 18: 306-9

8 Mychaliska GB, Bealer JF, Graft JL, Adzick NS, Harrison MR.Operating on placental support: The ex utero intrapartumtreatment (EXIT) procedure. J Pediatr Surg 1997; 32: 227-30

9 Skarsgard ED, Chitkara U, Krane EJ, Riley ET, Halamek LP, DedoHH. The OOPS procedure (operation on placental support): inutero airway management of the fetus with prenatally diagnosedtracheal obstruction. J Pediatr Surg 1996; 31: 826-8

10 Harrison MR, Anderson J, Rosen MA, Ross NA, Hendrick AG.Fetal surgery on the primate I: anaesthetic, surgical and tocolyticmanagement to maximise fetal-neonatal survival. J Pediatr Surg1982; 2: 115-22

I I Langer JC, Tabb T, Thompson P, Paes BA, Caco CC. Managementof prenatally diagnosed tracheal obstruction: access to the airwayin utero prior to delivery. Fetal Diagn Ther 1992; 7: 12-6

12 Chen MK, Gross E, Lobe TE. Perinatal management of entericduplication cysts of the tongue. Am J Perinatol 1997; 14: 161-3

13 Stocks RM, Egerman RS, Woodson GE, Bower CM, ThompsonJW, Wiet GJ. Airway management of neonates with antenatallydetected head and neck anomalies. Arch Otolaryngol Head NeckSurg 1997; 123: 641-5

14 Longaker MT, Golbus MS, Filly RA, Rosen MA, Chang SW,Harrison MR. Maternal outcome after open fetal surgery: a reviewof the first human cases. JAMA 1991; 265: 737-41

15 Hubbard AM, Crombleholme TM, Adzick NS. Prenatal MRIevaluation of giant neck masses in preparation for the fetal exitprocedure. Am J Perinatol 1998; 15: 253-7

16 Lazarus L, Young JD, Friend JC. E.M.G. syndrome and carbohydratemetabolism. Lancet 1968; 2: 1347-8

17 Wiedemann HR. E.M.G. syndrome. Lancet 1973; 2: 626-718 Li M, Squire JA, Weksberg R. Molecular genetics of Beckwith-

Wiedemann syndrome. Curr Opin Pediatr 1997; 9: 623-919 Ransini AC, Day-Salvatore D, Turner T, Smulian JC, Vintzileos

AM. Intrauterine growth and ultrasound findings in fetuses withBeckwith-Wiedemann syndrome. Obstet Gynecol 1997; 89: 538-42

20 Peters E, Cohen M, Altini M, Murray J. Rhabdomyosarcoma ofthe oral and paraoral region. Cancer 1989; 63: 963-6

21 Doval DC, Kannan V, Acharya RS, Mukherjee G, Shenoy AM,Bapsy PP. Rhabdomyosarcoma of the tongue. E>r J Oral MaxillofacSurg 1994; 32: 183-6

22 Moise KJ, Carpenter RJ, Deter RL, Kirshon B, Diaz SF. The useof fetal neuromuscular blockade during intrauterine procedures.Am J Obstet Gynecol 1987; 157: 874-79

23 Baker PM, Johnson IR, Harvey PR, Gowland PA, Manfield P. Athree-year follow-up of children imaged in utero with echo-planarmagnetic resonance. Am J Obstet Gynecol 1993; 170: 32-3

24 Williams LJ, YankowitzJ, Robinson RA, BendaJA. Nasopharyngealteratoma. Report of a case with prenatal diagnosis and postnatalmanagement J Reprod Med 1997; 42: 587-92

955