Obstructive Sleep Apnea in Children with Achondroplasia

Abstract

OBJECTIVE: To evaluate the prevalence of obstructive sleep apnea in a large population of children with achondroplasia and to evaluate the effectiveness of adenoidectomy and/or tonsillectomy as treatment.

METHODS: Retrospective review of 95 children with achondroplasia.

RESULTS: Thirty-six patients (38%) had clinical evidence of obstructive sleep apnea. Thirty-four patients underwent surgery, with more than 1 procedure required in 10 children (29%). Adenotonsil-lectomy was the initial procedure for 22 of 34 patients, and further therapy was required in only 18% of this group. Adenoidectomy was the initial procedure for 10 of 34, with 90% requiring further surgery for recurrent obstructive sleep apnea. Tonsillectomy alone was performed in 2 patients: 1 was effectively treated and 1 later required adenoidectomy. Endotracheal intubation was accomplished in all patients without complication; 53% required a smaller endotracheal tube than would be predicted by their age. Eight postoperative complications were recorded.

CONCLUSIONS: Obstructive sleep apnea is very common in children with achondroplasia. Surgery is effective, but recurrent symptoms are common, particularly when the initial procedure is adenoidectomy. The complication rate is higher than that observed in a general pediatric population but is readily managed with standard therapy. Anesthesia can be given safely to these patients with special consideration for limited neck extension and appropriate endotracheal tube size.

Achondroplasia is the most common skeletal dysplasia. It is inherited in an autosomal dominant pattern, but at least 80% of cases are new mutations. 1 Achondro-plasia is usually caused by an arginine-for-glycine substitution at amino acid 380 in fibroblast growth factor receptor 3. 2 , 3 This amino acid substitution causes constitutive activation of the receptor that, in turn, results in structural abnormalities of the axial skeleton and those portions of the skull that form by endochondral ossification. 4 Skeletal components of membranous origin develop normally. 5 The clinical features of achondroplasia are well known and include skull enlargement with narrowing of the skull base and foramen magnum, 6 midfacial hypoplasia, flattening of the nasal base, 1 , 7 , 8 shortening of the proximal limbs (rhizomelia), chest wall constriction and deformity, 9 , 10 and generalized hypotonia including hypotonia of the respiratory muscles. 7 , 9

Sleep-disordered breathing has been extensively described in patients with achondroplasia. 9 , 11 – 17 Several mechanisms for respiratory difficulty exist. Airway obstruction may be caused by midfacial hypoplasia, a small upper airway, and hypotonia of airway muscles. 9 , 12 , 17 – 19 Central apnea is caused by brain stem compression from foramen magnum and cervical spinal canal stenosis. 11 , 13 – 17 Thoracic deformity in some of these patients, including decreased anteroposterior diameter, pectus excavatum, thoracic kyphosis, and lumbar lordosis, also contributes to respiratory symptoms. 10 , 20

The reported frequency of obstructive sleep apnea (OSA) in this population varies widely, from 10% to 75%. 9 , 13 , 14 , 17 , 19 It is generally believed that OSA in patients with achondroplasia is caused by their anomalous orofacial structure compounded by factors seen in persons of average stature, such as adenotonsillar hypertrophy and obesity. Although adenotonsillectomy is effective in treating OSA in the general pediatric population, 21 , 22 studies of its effectiveness in achondroplasia have been limited.

This study was undertaken to better determine the prevalence of OSA in a population of children with achondroplasia and to assess both the short-term and long-term effectiveness of adenotonsillectomy as treatment. Additionally, perioperative and anesthetic evaluations and precautions were reviewed.

METHODS

The 95 children with achondroplasia seen regularly between 1981 and 1996 at the Midwest Regional Bone Dysplasia Clinic of the Clinical Genetics Center, University of Wisconsin–Madison were assessed. These patients are seen annually for comprehensive medical evaluation and diagnostic testing. A database search revealed 58 of the 95 children had a clinical history of apnea, polysomnographic evidence of apnea, or a history of previous tonsillectomy and/or adenoidectomy.

Comprehensive chart review of these 58 patients was performed, and pertinent information about medical history, physical examination, diagnostic studies, and surgical procedures was obtained. Historic information included parental or caregiver reports of snoring, glottal stops, apnea, compensatory sighs, neck hyperextension in sleep, nocturnal self-awakenings, nocturnal emesis, and daytime symptoms of excessive irritability or somnolence. Physical examination features abstracted included qualitative estimates of oropharyngeal and tonsillar size, presence of redundant oropharyngeal tissue, palatal and uvular anatomy, and observations of neck hyper-extension and respiratory distress while awake. All diagnostic studies, including polysomnography (PSG), home apnea studies, echocardiography, soft tissue lateral radiographs, and imaging studies of the head and craniocervical junction, were recorded. Surgical procedures, postoperative complications, and duration of hospital stay were noted. Anesthesia records were reviewed for ease of mask ventilation, method of intubation, and size of endotracheal tube used. Finally, if a patient received any nonsurgical treatments, such as nocturnal oxygen supplementation or continuous positive airway pressure (CPAP), the outcomes were abstracted.

Information obtained from PSG included the number of apneic and hypopneic episodes, an apnea-hypopnea index, the lowest oxygen saturation, the longest apnea event, and the general interpretation of the study by the sleep laboratory director. For this report, studies were then graded as “no OSA,” “mild OSA,” or “severe OSA,” on the basis of the severity of obstruction-precipitated oxygen desaturation and the number of obstructive events. A study with no desaturation below 90% was labeled “no OSA,” desaturations associated with hypopnea and obstruction that fell between 80% and 90% were labeled “mild OSA,” and desaturations lower than 80% were deemed “severe OSA.” Children whose caregivers observed sleep-disordered breathing, as evidenced by any combination of snoring, glottal stops, apneic events, gasping respirations, neck hyperextension, nocturnal self-awakenings, and daytime symptoms of excessive irritability or somnolence, were considered to have OSA and were included in this analysis regardless of the degree of oxygen desaturation measured by PSG.

The prevalence of apnea and the age at diagnosis were determined for 2 different diagnostic criteria: PSG and a physician's recognition of apnea episodes. For those patients treated for apnea, the age for their first treatment was determined. Mean ages were calculated for those treated by adenoidectomy, tonsillectomy, and adenotonsillectomy. The average time until recurrence was determined for those patients with recurrent clinical or polysomnographic evidence for apnea after treatment. The percentage of patients with symptoms before treatment and during their first visits within 12 months after treatment were compared to determine whether treatment helped to improve patients' symptoms. Fisher's exact test was used to compare the outcomes of the different surgical interventions. The association between OSA, as diagnosed by PSG, and patient symptoms at the time of PSG were evaluated with Fisher's exact test to determine the predictive value of specific symptoms for OSA.

RESULTS

Clinical Assessments

Patient ages at first evaluation in the Midwest Regional Bone Dysplasia Clinic ranged from 1 day to 14 years but were heavily skewed toward younger ages at initial assessment. The mean age at first evaluation was 19 months, but the median age was 8 months. Follow-up ranged from 1 to 13 years (mean 4.9 years, median 4.2 years). Fifty-one of the 58 were still being actively observed, 5 had moved or were lost to follow-up, and 2 had died.

Of the 58 patients initially selected for review, 22 were eliminated from analysis because they did not have OSA by either of the diagnostic criteria. Nine had insufficient data to support the diagnosis of sleep apnea, 6 had exclusively central rather than obstructive apnea, 6 had undergone tonsillectomy and/or adenoidectomy for reasons other than OSA (middle ear dysfunction or chronic tonsillitis), and 1 died of brain stem compression. 15

The remaining 36 patients met the criteria for the diagnosis of OSA. Twenty-eight children underwent PSG. Twenty-three had polysomnographic confirmation of OSA. Sixteen of these children also had apneic events witnessed by their caregivers, whereas 7 did not report apnea during sleep at home. Five individuals had negative PSG test results and no witnessed apneic events, but had other clinical features of sleep-disordered breathing judged to be sufficient to warrant the diagnosis of OSA. An additional eight children had witnessed apneic events at home, and did not undergo PSG.

Fig 1

. Outcome of adenotonsillectomy as initial procedure for OSA.

Table 1

Symptoms as predictors of positive PSG

Preoperative symptoms	PSG-confirmed OSA	Normal PSG	P value
Snoring	10/10	4/12	0.002
Reported apnea episodes	7/10	1/12	0.006
Glottal stops	4/10	0/12	0.029
Neck hyperextension	6/10	2/12	0.074
Self-awakenings	4/10	1/12	0.135
Daytime symptoms	1/10	0/12	0.455
Nocturnal emesis	1/10	0/12	0.455
Compensatory sighs	3/10	3/12	1.00
Awake respiratory distress	2/10	2/12	1.00

Table 2

Physical examination as predictor of positive PSG

Finding	PSG-confirmed OSA	Normal PSG	P value
Small oropharyngeal size	5/10	4/4	0.505
4+ tonsil size	2/4	0/2	0.600
Redundant oropharyngeal tissue	2/10	2/12	1.000

Children whose OSA was diagnosed by PSG had an average age at diagnosis of 3.7 years, and those whose OSA was diagnosed by caregiver-reported apnea had an average age of 4.0 years at diagnosis. Overall, the prevalence of OSA in this population of children with achondroplasia is 38% (36 of 95).

Several signs and symptoms were predictive of OSA defined by abnormal PSG (Table 1). In particular, snoring, observed apneic episodes, and glottal stops were all reliably associated with PSG confirmation of OSA. Correlation of physical examination findings with OSA was difficult because of incomplete data. Although such features as small oropharyngeal size, 4+ tonsillar hypertrophy (pharyngeal tonsil tissue meeting in the midline), and redundant oropharyngeal tissue were common in those with OSA (Table 2), none discriminated between those with and those without OSA by PSG.

Surgical Intervention and Outcome

Thirty-four of the 36 patients had at least 1 surgical procedure for their upper airway obstruction.

Twenty-two patients had adenotonsillectomy as the initial procedure (Fig. 1), at a mean age of 3.8 years (median 3 years). Of those, 14 had marked improvement or resolution of their symptoms and have required no further intervention, with a follow-up ranging from 3 to 60 months (mean 23 months). Six of these 14 had PSG after surgical intervention, which in every instance showed improvement in sleep architecture. In 2 children who had required nocturnal oxygen supplementation since infancy, oxygen use could be discontinued after adenotonsillectomy.

Five of these 22 patients had continued sleep disturbance after adenotonsillectomy. One was successfully treated with prone positioning in sleep, and 1 was treated with nocturnal CPAP. In one child, OSA resolved, but persistent central apnea was documented by PSG. Two children continued to snore, but postsurgical symptoms were judged to be mild enough that no further intervention has been undertaken.

Fig 2

. Outcome of adenoidectomy as initial procedure for OSA.

Additional surgery was required in 3 children (14%) after initial adenotonsillectomy. In 1 child, OSA symptoms recurred within 3 months but were successfully treated with revision adenoidectomy. The second child had improvement for 5 years before obstruction recurred; uvulectomy and repeat adenoidectomy resulted in marked improvement. One year later OSA was again documented by clinical signs and PSG. Only after a third adenoidectomy and use of CPAP has her breathing in sleep been normalized. The third child, whose initial surgery was at age 3.3 years, had multiple recurrences of clinically and PSG-confirmed OSA requiring a series of surgical interventions: repeat adenoidectomy and uvulectomy (age 3.9 years), uvulopalatopharyngo-plasty (age 4.4 years), and repeat adenoidectomy (age 5.4 years). After each procedure obstruction was temporarily relieved, and postoperative PSG documented relief of the OSA, only to recur after very short periods. However, he has been healthy and symptom free since the last of these procedures, with follow-up since then of 4.3 years.

Ten children underwent adenoidectomy alone as the initial procedure for OSA (Fig. 2), at an average age of 3.6 years (median age 2.2 years). All but 1 had recurrence of OSA, on average at 18 months after surgery (median 15 months) despite uniform initial apparent improvement. Of those who had postoperative PSG, 4 of 5 had unchanged or worsened obstruction compared with preoperative studies. Additional surgery (tonsillectomy in 3 and tonsillectomy and repeat adenoidectomy in 2) has resulted in resolution of symptoms in 5 of these children, with an average follow-up of 1.5 years. Another child who had repeat adenoidectomy requires nocturnal CPAP. In 3 others additional surgical intervention is pending. In only 1 child has adenoidectomy alone resulted in lasting improvement.

Two children underwent tonsillectomy alone as the initial procedure for OSA at an average age of 3.5 years. In 1, adenoidectomy was deferred because preoperative studies suggested that it might result in velopharyngeal insufficiency. OSA transiently improved but recurred, necessitating adenoidectomy 1.5 years later. Her OSA resolved after the adenoidectomy, and no velopharyngeal insufficiency developed. The second child has shown improvement with tonsillectomy alone, but follow-up has been short (6 months).

Finally, 2 individuals, who were both symptomatic and had OSA by PSG evaluation, had resolution of OSA at 6 months and at 2 years after initial diagnosis without any surgical intervention.

When adenotonsillectomy is compared with adenoidectomy for relief of symptoms with surgery, adenotonsillectomy was reliably superior in relieving symptoms of OSA (P < 0.01).

Of the 34 children treated surgically, 28 had at least 1 follow-up visit to the Midwest Regional Bone Dysplasia Clinic within 1 year of the initial surgery. At initial follow-up, the frequency of symptoms had declined (Table 3). Snoring decreased from 93% before surgery to 59% after, glottal stops from 35% to 12%, compensatory sighs from 30% to 4%, self-awakening from 26% to 11%, neck hyperextension from 52% to 33%, daytime symptoms from 7% to 4%, and awake respiratory distress from 15% to 0%.

Table 3

Postoperative symptom improvement (n = 28)

Symptom	Before surgery (%)	After surgery (%)	% Improved
Awake respiratory distress	15	0	100
Compensatory sighs	30	4	88
Glottal stops	35	12	67
Daytime symptoms	7	4	67
Self-awakenings	26	11	57
Snoring	93	59	33
Neck hyperextension	52	33	31

Perioperative Management and Postoperative Complications

From a total of 47 procedures completed in 34 patients, 29 had sufficiently complete records that peri-operative and postoperative care could be assessed.

Of the 20 children undergoing tonsillectomy with or without concomitant adenoidectomy or palatopharyn-goplasty, 14 (70%) were hospitalized, and 5 (25%) had the procedure completed as outpatients. In 1 patient records were not complete enough to determine whether the procedure was done on an inpatient or outpatient basis. The average length of hospitalization for those admitted was 1.5 days. The mean age for those admitted was 2.8 years compared with a mean age of 4.8 years in those who had outpatient surgery. Complications arose in 6: fever with atelectasis or pneumonia in 2, transient postoperative respiratory distress in 2, pulmonary edema in 1, and posttonsillectomy bleeding in 1. Of these, all resolved within 24 hours except in 1 child who was rehospitalized for 5 days because of pneumonia.

Nine children undergoing primary or revision adenoidectomy alone had complete perioperative records available. Five of 9 (56%) were cared for as outpatients. In those admitted, the average length of stay was 1.5 days. One postoperative fever in an outpatient and 1 instance of delayed extubation in an inpatient were the only complications.

Anesthesia Management

Anesthesia records regarding 17 procedures (in 11 patients) were available. Information regarding ease of mask ventilation, intubation, and size of endotracheal tube used was tabulated.

In all but 1 case, mask ventilation was noted to be “easy.” The child with a “difficult” mask airway during an adenoidectomy had uvulopalatopharyngoplasty 6 months later and had an “easy” airway at the second procedure.

In all instances, care was taken to limit neck extension during intubation and during the surgical procedure to reduce risks associated with uncontrolled head movement around a constricted foramen magnum. Intubation was completed by direct laryngoscopy in all but 1 patient. In this child general anesthesia was induced for adenoidectomy by mask ventilation, and the airway was described as “easy.” An uncomplicated fiberoptic intubation was performed. This child had an uncomplicated direct laryngeal visualization and intubation for tonsillectomy 1 year later.

Endotracheal tube size in children can be calculated on the basis of either the child's weight or age, with age providing better correlation. The endotracheal tube size used in these patients was compared with the size used in persons of normal stature. 23 By both age and weight, in 8 of 17 patients undergoing anesthesia (47%) the expected size of endotracheal tube was used. In the remaining 9 patients (53%) a smaller tube than predicted by the child's age was needed. In 6 the tube was 1/2; size smaller, in 2 it was 1 whole size smaller, and in 1 it was 1/2; sizes smaller.

DISCUSSION

In this cohort of children with achondroplasia, the prevalence of OSA was 38%. Although a considerable range (from 10% to 75%) has been reported previously, 9 , 13 , 14 , 17 , 19 , 24 all studies have concluded that OSA is common in this population.

Estimating the prevalence of OSA in achondroplasia is difficult because many studies are small, are biased by referral, or have limited follow-up time. Furthermore, different studies have used different, nonstandardized criteria for grading of sleep studies in children with achondroplasia.

This study represents the largest cohort of children with achondroplasia reviewed to date. The individuals were followed up with well-documented serial clinical assessments by a single examiner for their diagnosis of achondroplasia rather than because of symptoms related to disturbances of breathing. Most had extensive follow-up, averaging nearly 5 years. Children with central apnea were not included, a distinction that has not always been made in previous reports. This study accounts for several sources of variation and may better reflect the occurrence of OSA in children with achondroplasia as a whole.

Clinical history was an effective discriminator of those with OSA. In particular the occurrence of loud snoring, glottal stops, and observed episodes of obstructive apnea were each predictive of confirmation of OSA by PSG (P < 0.05). Although common, other signs and symptoms including compensatory sighs, neck hyperextension in sleep, self-awakenings, and daytime irritability or somnolence did not effectively predict which children would have abnormal polysomnograms. This is, in part, because even within this large cohort the number of instances of polysomnographic assessment was relatively small. Because each of these symptoms is common in OSA, all should be sought by careful clinical history taking.

The children with normal sleep study results despite sleep-disordered breathing fall into the category of upper airway resistance syndrome, characterized by symptoms of OSA and frequent awakenings with normal PSG studies. 24 , 25 Children with achondroplasia have been reported to sometimes have atypical OSA in which no alteration of blood gases or lengthy apneas are observed. 17

Guilleminault et al 26 observed that both craniofacial features and tonsillar hypertrophy correlate with the likelihood for OSA in children of average stature. The craniofacial characteristics of children with achondroplasia, including midfacial hypoplasia, depressed nasal base, small nasal airways, and diminished oropharyngeal depth, predispose them to development of OSA. 9 , 24 With age-dependent physiologic hypertrophy of Waldeyer's ring 27 and obesity, symptomatic OSA develops in many.

The use of adenotonsillectomy in OSA has been previously studied in the general pediatric population. It remains the most used treatment for OSA and is reported to be effective in as many as 90% to 100% of children without craniofacial abnormalities. 21 , 28 Its effectiveness in children with achondroplasia is less certain. One might expect that the value of adenotonsillectomy would be inversely correlated with the severity of other airway abnormalities. Given the structural differences in children with achondroplasia, a more modest success rate might be predicted. Waters et al 19 have reported rather limited success in treating OSA with adenotonsillectomy in children with achondroplasia, with 60% of treated children requiring ongoing CPAP therapy because of persisting airway obstruction. In contrast, 64% (14 of 22) of those treated with adenotonsillectomy in this series had complete resolution of sleep-disordered breathing and did not require further intervention. An additional 4 children still had mild symptoms, resulting in improvement in 82% overall. Although the success rate is lower than in the general pediatric population, the long-term follow-up of most of these individuals suggests that a majority of children with achondroplasia and OSA can be effectively treated with adenotonsillectomy.

In contrast, adenoidectomy alone resulted in a surprisingly high rate of recurrence of OSA, with 9 of 10 children needing additional intervention. Five of these children have subsequently had either tonsillectomy or tonsillectomy with revision adenoidectomy, and these procedures have been successful to date. Our analyses led us to conclude that combined adenotonsillectomy should be the primary procedure of choice in these children.

General pediatric patients with OSA who undergo adenotonsillectomy are at significant risk for postoperative respiratory compromise, which has been reported in up to 27% of patients. 29 Overnight inpatient observation has been recommended for any child with OSA undergoing adenotonsillectomy, 29 but at particular risk are the young (age < 3 years), those with severe obstructive symptoms, and those with craniofacial anomalies. 29 – 31 Complication rates of adenoidectomy and tonsillectomy in this series were comparable (8 of 29, or 28%) and were treated with standard therapy. Respiratory distress caused by pulmonary edema developed in at least 1 of these patients, presumably because of acute relief of longstanding obstruction. 32 Therefore, particularly in patients with severe preoperative symptoms, inpatient hospitalization is prudent.

Considerable risk can attend anesthesia in individuals with achondroplasia, 7 , 8 , 33 including difficulty with laryngeal visualization; problems with airway management because of facial anomalies, small caliber of the airway, short neck, and upper airway muscle hypotonia; risk of cervicomedullary compression with neck hyper-extension or uncontrolled head movement; difficulty with ventilation because of restrictive lung disease; and difficulty with selection of an appropriate endotracheal tube size. In this series we did not encounter any of these difficulties when using mask ventilation in the 17 procedures that were reviewable. Likewise, endotracheal intubation was accomplished with direct laryngeal visualization and without difficulty. The need for a smaller endotracheal tube size than would be expected for the patient's age should be anticipated. Special care should be taken in manipulating the neck because several prior reports have noted that cervicomedullary compression can result in apnea 11 , 14 – 16 , 18 as well as cervical myelopathy and consequent paralysis 15 , 33 , 34 and death. 15 , 16

Nonsurgical alternatives also may be of great benefit in children with achondroplasia and OSA but were not specifically studied here. Positive airway pressure is effective both in the general pediatric population 21 and in children with achondroplasia in whom surgical therapy has failed. 17 , 19 However, because CPAP requires patient and family motivation and is a long-term therapy, acceptance and compliance are difficult to achieve.

SUMMARY

OSA is common in children with achondroplasia, arising in 38% and often in early childhood. It can be recognized by careful attention to clinical history and by comprehensive clinical assessment and can be confirmed with polysomnographic evaluation. When the diagnosis of OSA is confirmed, adenotonsillectomy, but not adenoidectomy alone, is usually an effective treatment. Some children will require multiple surgical procedures. Despite previously recognized concerns about anesthesia in individuals with achondroplasia, with appropriate care taken related to manipulation of the neck and airway management, general anesthesia and otolaryngologic surgery can be successfully carried out in children with achondroplasia without significant danger. A higher than usual rate of transient postoperative respiratory complications supports the recommendation that inpatient observation after adenotonsillectomy be customary in these children.

References

Hall

JG.

The natural history of achondroplasia. Basic Life Sci 1988;48:3–9.

Shiang

Thompson

Zhu

YA.

Mutations in the transmem-brane domain of FGFR3 cause the most common genetic form of dwarfism, achondroplasia. Cell 1994;78:335–42.

Bellus

Hefferon

Ortiz de Luna

Achondroplasia is defined by recurrent G380R mutations of FGFR3. Am J Hum Genet 1995;56:368–73.

Naski

Wang

Jingsong

Graded activation of fibroblast growth factor receptor 3 mutations causing achondroplasia and thanatophoric dysplasia. Nat Genet 1996;13:233–7.

Rimoin

Hughes

Kaufman

Endochondral ossification in achondroplastic dwarfism. N Engl J Med 1970;283:728–35.

Hecht

Horton

Reid

Growth of the foramen magnum in achondroplasia. Am J Med Genet 1989;32:528–35.

Berkowitz

Raja

Bender

Dwarfs: pathophysiology and anesthetic implications. Anesthesiology 1990;73:739–59.

Lauri

Marri

Galli

Anaesthesia in achondroplastic dwarves. Basic Life Sci 1988;48:167–74.

Stokes

Phillips

Leonard

Respiratory complications of achondroplasia. J Pediatr 1983;102:534–41.

10.

Stokes

Wohl

MEB

Wise

The lungs and airways in achondroplasia. Chest 1990;98:145–52.

11.

Fremion

Garg

Kalsbeck

Apnea as the sole manifestation of cord compression in achondroplasia. J Pediatr 1984;104: 398–401.

12.

Larsen

Snyder

Matsuo

Achondroplasia associated with obstructive sleep apnea [letter]. Arch Neurol 1983;40:769.

13.

Mador

Tobin

MJ.

Apneustic breathing: a characteristic feature of brainstem compression in achondroplasia?

Chest 1990;97:877–83.

14.

Nelson

Hecht

Horton

Neurological basis of respiratory complications in achondroplasia. Ann Neurol 1988;24:89–93.

15.

Pauli

Horton

Glinski

Prospective assessment of risks for cervicomedullary-junction compression in infants with achondroplasia. Am J Hum Genet 1995;56:732–44.

16.

Pauli

Scott

Wassman

Apnea and sudden unexpected death in children with achondroplasia. J Pediatr 1984;104:342–8.

17.

Waters

Everett

Sillence

Breathing abnormalities in sleep in achondroplasia. Arch Dis Child 1993;69:191–6.

18.

Reid

Pyeritz

Kopits

Cervicomedullary compression in young patients with achondroplasia: value of comprehensive neurologic and respiratory evaluation. J Pediatr 1987;110:522–30.

19.

Waters

Everett

Sillence

Treatment of obstructive sleep apnea in achondroplasia. Am J Med Genet 1995;59:460–6.

20.

Stokes

Pyeritz

Wise

Spirometry and chest wall dimensions in achondroplasia. Chest 1988;93:364–9.

21.

Brooks

LJ.

Treatment of otherwise normal children with obstructive sleep apnea. Ear Nose Throat J 1993;72:77–9.

22.

Marcus

Carroll

Koerner

Determinants of growth in children with obstructive sleep apnea syndrome. J Pediatr 1994;125:556–62.

23.

Smith

RM.

Anesthesia for infants and children. 5th ed. St. Louis: Mosby; 1980. p. 275.

24.

Zucconi

Weber

Castronovo

Sleep and upper airway obstruction in children with achondroplasia. J Pediatr 1996;129:743–9.

25.

Guilleminault

Philip

Polygraphic investigation of respiration during sleep in infants and children. J Clin Neurophysiol 1992;9:48–55.

26.

Guilleminault

Pelayo

Leger

Recognition of sleep-disordered breathing in children. Pediatrics 1996;948:871–82.

27.

Tomaski

SM.

Embryology and anatomy of the mouth, pharynx and esophagus. In: Bluestone

Stool

Kenna

, editors. Pediatric otolaryngology. 3rd ed. Philadelphia: WB Saunders; 1996. p. 931–2.

28.

Frank

Kravath

Pollak

Obstructive sleep apnea and its therapy: clinical and polysomnographic manifestations. Pediatrics 1983;71:737–42.

29.

Rosen

Muckle

Maliowald

Postoperative respiratory compromise in children with obstructive sleep apnea syndrome: can it be anticipated?

Pediatrics 1994;93:784–8.

30.

McColley

April

Carroll

Respiratory compromise after adenotonsillectomy in children with obstructive sleep apnea. Arch Otolaryngol Head Neck Surg 1992;118:940–3.

31.

Price

Hawkins

Kahlstrom

EJ.

Tonsil and adenoid surgery for airway obstruction: perioperative respiratory morbidity. Ear Nose Throat J 1993;72:526–30.

32.

Galvis

Stool

Bluestone

CS.

Pulmonary edema following relief of acute upper airway obstruction. Ann Otol Rhinol Laryngol 1980;89(2 pt 1):124–8.

33.

Dvorak

Rusnak

Morcos

JJ.

Multiple trauma in the achondroplastic dwarf: an emergency medicine physician perspective case report and literature review. Am J Emerg Med 1993;11:390–5.

34.

Colamaria

Mazza

Beltramello

Irreversible respiratory failure in an achondroplastic child: the importance of an early cervicomedullary decompression and a review of the literature. Brain Dev 1991;13:270–9.