Children With 22.Q.11.2 Deletion Syndrome: Sleep-Disordered Breathing and Management
Funding: The authors received no specific funding for this work.
ABSTRACT
Patients with 22q11.2 deletion syndrome (22q11DS) are predisposed to obstructive sleep apnea (OSA) due to an abnormal craniofacial anatomy with pharyngeal hypotonia, retrognathia, micrognathia, and glossoptosis. The aim of the study was to describe the prevalence and management of OSA in a cohort of children with 22q11DS. All patients with 22q11DS seen at the national reference center of craniofacial anomalies at Necker-Enfants malades hospital (Paris, France) between April 2014 and April 2024 had a systematic respiratory polygraphy (PG) in room air. Clinical data, PGs, and subsequent OSA management were retrospectively analyzed. The data of 52 patients were analyzed. Associated disorders were common, with 79% of the patients having an upper airway anomaly, 58% a cardiopathy, and 30% a pulmonary disease. Mean age at baseline PG was 6.6 ± 4.6 (0.1–18) years. Twelve (23%) patients had an adenoidectomy and/or tonsillectomy, and 10 (19%) patients a posterior flap pharyngoplasty prior to baseline PG. Four patients were treated with continuous positive airway pressure (CPAP) and 2 patients with a cardiopathy were treated with long-term oxygen therapy prior to baseline PG. Mean AHI was 4.0 ± 9.1 (0–43) events/h, with 24 (46%) patients having OSA, with 15 (29%) having mild OSA, 5 (9%) moderate OSA, and 4 (8%) severe OSA. A young age (p = 0.003), an immune deficiency (p = 0.018) and a pulmonary disease (p = 0.028) were more common in patients with OSA as compared to those without OSA. On follow-up, OSA improved after upper airway surgery in 4 patients or spontaneously, with only 2 patients requiring CPAP for persistent moderate OSA. In conclusion, the prevalence of OSA in children with 22q11DS is high. OSA severity is mainly mild except in infants aged < 1 year with an immune deficiency and a pulmonary disease being more common in patients with OSA as compared to those without OSA.
1 Introduction
22q11.2 deletion syndrome (22q11DS) is the most frequent chromosomal microdeletion syndrome, with a prevalence between 1 per 3000 to 1 per 6000 live births (McDonald-McGinn et al. 2015). The syndrome results from a ~ 1.5 to 3 Megabase hemizygous deletion (~ 46 protein-coding genes) in the locus 22q11.2. It is a multisystem disorder including physical, cognitive, and behavioral issues of variable severity (McDonald-McGinn 2020).
Before the genetic discovery, several clinical syndromes were described as separate entities, such as DiGeorge syndrome, conotruncal anomaly facial syndrome, velocardiofacial syndrome, and Cayler cardiofacial syndrome. These syndromes have many similar phenotypic features; a common cause was suggested. After the 1990s, the fluorescent in situ hybridization (FISH) technique made it possible to understand that these previously described clinical pictures (with similar characteristics) were a unique clinical condition with a common etiology: the microdeletion of band 11.2 on the long arm of the autosomal chromosome 22. Clinically, it is possible to observe a heterogeneous presentation that can be associated with multi-organ dysfunction. Congenital malformations, including cardiac (primarily conotruncal heart malformations) and upper airway anomalies (such as velopharyngeal insufficiency (VPI) or cleft palate), are common in 22q11DS. Other symptoms include immune deficiency (due to thymic aplasia/hypoplasia) and endocrine disorder (hypocalcemia) (Óskarsdóttir et al. 2023).
Patients with 22q11DS are predisposed to obstructive sleep apnea (OSA) due to an abnormal craniofacial anatomy with pharyngeal hypotonia, retrognathia, micrognathia, and glossoptosis. The genetic defect is reflected in embryogenetic malformations of the larynx, predisposing to structural abnormalities of the upper airway wall associated with dynamic instability and OSA (Verheij et al. 2018). All these factors explain the high prevalence of OSA, with ranges from 20% to 58% (Arganbright et al. 2020; Kennedy et al. 2014), compared to 2% to 10% in the general pediatric population (Magnusdottir and Hill 2024).
Adenotonsillectomy represents the first-line treatment of OSA. However, because of the anatomical facial and upper airway anomalies, this surgical approach is not always conclusive (Arganbright et al. 2022). Moreover, due to pharyngeal hypotonia, VPI surgery may be indicated, with the risk that OSA may develop or worsen after VPI surgery (Crockett et al. 2014; Trabelsi et al. 2022).
The aim of the study was to describe the prevalence and management of OSA in a cohort of children with 22q11DS treated at a national reference center.
2 Materials and Methods
2.1 Patients
All patients with a genetic diagnosis of 22q11DS seen at the national reference center of craniofacial anomalies at Necker-Enfants malades hospital (Paris, France) between April 2014 and April 2024 had a systematic respiratory polygraphy (PG). The PGs and subsequent OSA management were retrospectively analyzed, and clinical data such as comorbidities, surgical history, and treatments were gathered. Follow-up data were collected for patients who underwent upper airway surgery and/or who were treated with continuous positive airway pressure (CPAP).
The study was performed in agreement with the French regulation and all the parents and patients > 6 years old gave their informed consent. The study received appropriate legal and ethical approval from the ethical committee (CPP Ile de France II, protocol 2013-A00374-41).
2.2 Sleep Studies
- Obstructive apnea (OA) was defined as a decrease in nasal airflow of at least 90% with continued abdominal and chest wall movements for at least two breaths.
- Hypopnea was defined as a decrease in nasal airflow of at least 30% with a corresponding decrease in SpO2 of at least 3%.
- Central apnea (CA) was defined as a decrease in nasal airflow of at least 90% without inspiratory effort, with a duration of at least two breaths and associated with a 3% oxygen desaturation (OD), or of a duration of at least 20 s.
- Mixed apnea was defined as an apnea that begins as central and ends as obstructive, or vice versa.
The apnea-hypopnea index (AHI) was calculated as the sum of apneas and hypopneas per hour of time in bed. An AHI ≤ 1 event/h was considered normal, mild OSA was defined as 1 < AHI < 5 events/h, moderate OSA as 5 ≤ AHI < 10 events/h, and severe OSA as AHI ≥ 10 events/h (Benedek et al. 2020). The central apnea index (CAI) was defined as the number of central apneas per hour, with a CAI < 5 events/h being considered normal (Felix et al. 2016).
Mean and minimal SpO2, and the percentage of recording time with a SpO2 < 90% were calculated. The oxygen desaturation index (ODI), defined as the number of desaturations ≥ 3% per hour of recording time, was calculated and was considered abnormal when > 5 events/h. Mean and maximal PtcCO2, and the percentage of recording time with a PtcCO2 > 50 mmHg were calculated.
2.3 Therapeutic Interventions
All the patients were managed by a pediatric multidisciplinary team comprising sleep specialists, pulmonologists, clinical geneticists, endocrinologists, neurologists, radiologists, and surgeons (neurosurgery, ENT, maxillofacial surgery and orthodontics, and orthopedics). All medical, surgical, and/or respiratory interventions (such as CPAP, adenotonsillectomy or VPI surgery) were decided after collegial discussions, taking into account the patients' symptoms, age, and the results of the various investigations. The treatment was then proposed in a shared-decision process between the multidisciplinary team, the parents, and the child when possible.
2.4 Statistical Analysis
Data are expressed as mean ± standard deviation, or median (range). Comparisons between quantitative variables were done using the Student t-test, and comparisons between qualitative variables were done using the χ2 test. A p value < 0.05 was considered statistically significant.
3 Results
3.1 Patients
The data of 52 patients were analyzed (Table 1). Associated disorders were common, with 79% of the patients having an upper airway anomaly, 58% a cardiopathy, and 30% a pulmonary disease. Mean age at baseline PG was 6.6 ± 4.6 (0.1–18) years. Twelve (23%) patients had a prior adenoidectomy and/or tonsillectomy, and 10 (19%) patients had a prior posterior flap pharyngoplasty (PFP). Four patients were treated with CPAP, and 2 patients with a cardiopathy were treated with long-term oxygen therapy prior to baseline PG.
| Total population, n = 52 | |
|---|---|
| Age at first study (years), mean ± SD | 6.6 ± 4.6 |
| Male/female (number) | 24/28 |
| Body mass index a , mean ± SD (percentile) | 48 ± 35 |
| Obesity b (n, %) | 3 (6%) |
| Associated disorder (n, %) | |
| Cardiopathy | 30 (58%) |
| Intra-ventricular communication | 12 (23%) |
| Intra-auricular communication | 8 (15%) |
| Aortic outflow anomaly c | 9 (17%) |
| Pulmonary outflow anomaly d | 5 (10%) |
| Upper airway anomaly | 41 (79%) |
| Velopharyngeal insufficiency | 22 (42%) |
| Cleft palate | 12 (23%) |
| Laryngeal anomaly e | 11 (21%) |
| Glossoptosis and/or retrognathia and/or micrognathia and/or Pierre-Robin sequence | 7 (14%) |
| Adeno/tonsillar hypertrophy | 7 (14%) |
| Hypoacousia | 4 (8%) |
| Immune deficiency | 8 (15%) |
| Gastrointestinal disease f | 6 (12%) |
| Pulmonary disease | 15 (30%) |
| Asthma (n, %) | 10 (19%) |
| Tracheo/bronchomalacia | 3 (6%) |
| Pulmonary hypertension | 2 (4%) |
| Endocrinological disease | 9 (18%) |
| Hypocalcemia | 6 (12%) |
| Hypothyroidism | 3 (6%) |
| Scoliosis (n, %) | 5 (10%) |
| Prior upper airway surgery (n, %) | |
| Adenoidectomy and/or tonsillectomy | 12 (23%) |
| Posterior flap pharyngoplasty | 10 (19%) |
| Continuous positive airway pressure | 4 (8%) |
| Long term oxygen therapy | 2 (4%) |
- a Body mass index (BMI) was calculated in patients > 2 years old.
- b Obesity was defined as BMI > 95th percentile for age and sex in patients > 2 years old.
- c Abnormal development of the aortic valve or ascending aorta or aortic arch.
- d Abnormal development of the pulmonary valve or pulmonary artery.
- e Laryngomalacia, laryngeal stenosis, laryngeal web, vocal cord paralysis.
- f Gastro-esophageal reflux, treatment for constipation, celiac disease or hemorrhagic rectocolic disease.
3.2 Baseline Polygraphic Findings
PG was performed in spontaneous breathing with room air, except for the 2 patients on oxygen therapy. Mean AHI was 4.0 ± 9.1 (0–43) events/h, with 24/52 (46%) patients having OSA, with 29% having mild OSA, 9% moderate OSA, and 8% severe OSA (Table 2). Mean SpO2 was 96% ± 2% with a mean minimal SpO2 of 88% ± 9%. Seven (14%) patients spent > 2% of recording time with a SpO2 < 90% and only one patient spent > 2% of recording time with a PtcCO2 > 50 mmHg.
| Respiratory events | |
| Mean AHI (mean ± SD, median [min; max]) | 4.0 ± 9, 0 [0.0; 43] |
| AHI ≤ 1 events/h, n (%) | 28 (54%) |
| AHI > 1 and ≤ 5 events/h, n (%) | 15 (29%) |
| AHI > 5 and ≤ 10 events/h, n (%) | 5 (9%) |
| AHI > 10 events/h, n (%) | 4 (8%) |
| Mean obstructive AI (mean ± SD, median [min; max]) | 1.3 ± 4.4, 0 [0; 28] |
| Obstructive AI > 5 events/h, n (%) | 0 |
| Obstructive AI > 10 events/h, n (%) | 3 (7%) |
| Mean central AI (mean ± SD, median [min; max]) | 0.6 ± 1, 0.2 [0.1; 6] |
| Central AI > 5 events/h, n (%) | 1 (2%) |
| Nocturnal gas exchange a | |
| Mean SpO2 (%) mean ± SD, median [min; max] | 96 ± 2, 97 [90; 99] |
| Minimal SpO2 (%), mean ± SD, median [min; max] | 88 ± 9, 90 [42; 95] |
| % of time with SpO2 < 90% (%), mean ± SD, median [min; max] | 2 ± 7, 0 [0; 44] |
| Number of patients with > 2% of time with SpO2 < 90%, n (%) | 7 (14%) |
| Mean 3% ODI (events/h), mean ± SD, median [min; max] | 10 ± 19, 2.3 [0; 81] |
| Number of patients with an ODI > 10 events/h, n (%) | 10 (19%) |
| Mean PtcCO2 (mmHg), mean ± SD, median [min; max] | 41 ± 5, 42 [31; 51] |
| Maximal PtcCO2 (mmHg), mean ± SD, median [min; max] | 44 ± 4, 45 [35; 56] |
| % of time with PtcCO2 > 50 mmHg (%) mean ± SD, median [min; max] | 2 ± 12, 0 [0; 79] |
| Number of patients with > 2% of time with PtcCO2 > 50 mmHg, n (%) | 1 (2%) |
- Abbreviations: 3% ODI: 3% oxygen desaturation index, AHI: apnea-hypopnea index, AI: apnea index, PtcCO2: transcutaneous carbon dioxide pressure, SpO2: pulse oximetry.
- a 2 patients performed the baseline polygraphy under oxygen therapy.
Of the 12 patients who had prior adenotonsillectomy, 3 and 2 patients had mild and moderate OSA, respectively, with the remaining patients having no OSA. Of the 10 patients who had prior PFP, 2 had mild OSA. Of the 4 patients with prior CPAP, only 2 had OSA (one mild and one severe).
Table 3 compares the characteristics and PG data of the patients with and without OSA. Patients with OSA were younger compared to those without OSA. Immune deficiency and pulmonary disease were more common in patients with OSA compared to patients without OSA. SpO2 parameters but not PtcCO2 parameters were significantly different between the 2 groups. No other differences were observed between the patients with and without OSA. One patient with a cardiopathy and pharyngomalacia had a prior treatment with CPAP, but at baseline PG, 1 month after CPAP weaning, he had moderate isolated hypercapnia without OSA, with a mean PtcCO2 of 51 mmHg and a maximal PtcCO2 of 53 mmHg, which was not observed on a control PG 3 months later.
| No OSA (n = 28) | OSA (n = 24) | p c | Mild (n = 15) | OSA (n = 24) | Severe (n = 4) | |
|---|---|---|---|---|---|---|
| Moderate (n = 5) | ||||||
| Characteristics of the patients | ||||||
| Age at baseline (years), median (min; max) | 7.5 (0.8; 18) | 3.4 (5.1; 17.6) | 0.003 | 5.3 (0.3; 17.6) | 0.8 (0.4; 15.5) | 0.3 (0.1; 0.7) |
| Male (n, %) | 12 (43%) | 12 (50%) | 0.813 | 8 (47%) | 3 (60%) | 1 (25%) |
| Body mass index, mean ± SD (kg/m2) | 17.4 ± 4.8 | 17 ± 3.5 | 1.000 | 16.9 ± 3.2 | 18.4 ± 4.6 | 15.2 ± 2.9 |
| Obesity (n, %) | 0 | 2 (8%) | 0.208 | 1 (7%) | 1 (20%) | 0 |
| Associated disorder (n, %) | ||||||
| Cardiopathy | 16 (57%) | 14 (58%) | 0.845 | 6 (38%) | 5 (100%) | 3 (75%) |
| Upper airway disease a | 25 (86%) | 16 (67%) | 0.099 | 10 (67%) | 2 (40%) | 4 (100%) |
| Immune deficiency | 1 (4%) | 7 (29%) | 0.018 | 3 (20%) | 2 (40%) | 2 (50%) |
| Gastrointestinal disease b | 4 (14%) | 2 (8%) | 0.674 | 1 (7%) | 0 | 1 (25%) |
| Pulmonary disease | 4 (14%) | 11 (46%) | 0.028 | 4 (27%) | 5 (100%) | 2 (50%) |
| Endocrinological disease | 2 (7%) | 7 (29%) | 0.064 | 2 (13%) | 3 (60%) | 2 (50%) |
| Scoliosis | 2 (7%) | 3 (13%) | 0.652 | 1 (7%) | 2 (40%) | 0 |
| Prior upper airway interventions (n, %) | ||||||
| Adenoidectomy and/or tonsillectomy | 7 (25%) | 5 (21%) | 0.980 | 3 (20%) | 2 (40%) | 0 |
| Velopharyngeal insufficiency surgery | 8 (29%) | 2 (8%) | 0.056 | 2 (13%) | 0 | 0 |
| Laryngeal surgery | 4 (14%) | 0 | 0.115 | 0 | 0 | 0 |
| Prior continuous positive airway pressure (n, %) | 2 (7%) | 2 (8%) | 1.000 | 1 (7%) | 0 | 1 (25%) |
| Respiratory events, mean ± SD, median [min; max] | ||||||
| Mean AHI (events/h) | 0.4 ± 0.3, 0.4 [0; 1] | 8 ± 12, 3 [1.2; 43] | < 0.001 | 2 ± 0.7, 2 [1; 3] | 7 ± 1, 6 [6; 9] | 32 ± 12, 33 [21; 43] |
| Mean obstructive AI (events/h) | 0.1 ± 0.2, 0.0 [0; 0.5] | 3 ± 6, 0.6 [0; 28] | < 0.001 | 0.4 ± 0.5, 0.2 [0; 1] | 2 ± 2, 2 [0.4; 4] | 12 ± 12, 11 [0; 28] |
| Mean central AI (events/h) | 0.2 ± 0.2, 0.1 [0; 1] | 1 ± 1, 0.5 [0; 6] | < 0.001 | 0.6 ± 0.6, 0.5 [0; 2] | 1.3 ± 1.4, 1 [0; 3] | 2.4 ± 2.5, 2 [0.5; 6] |
| Nocturnal gas exchange, mean ± SD, median [min; max] | ||||||
| Mean SpO2 (%) | 97 ± 1, 97 [93; 98] | 95 ± 3, 96 [90; 99] | 0.023 | 96 ± 2, 97 [92; 99] | 93 ± 3, 92 [90; 97] | 95 ± 3, 95 [92; 99] |
| Minimal SpO2 (%) | 91 ± 3, 92 [83; 95] | 84 ± 11, 87 [42; 94] | < 0.001 | 89 ± 4, 89 [78; 94] | 81 ± 4, 83 [75; 85] | 67 ± 17, 72 [42; 81] |
| % of time with SpO2 < 90% (%) | 0 ± 0.2, 0 [0; 1] | 5 ± 10, 0 [0; 44] | 0.009 | 1 ± 3, 0 [0; 10] | 14 ± 19, 3.7 [0; 44] | 8 ± 6, 7.6 [0; 15] |
| Patients with > 2% of time with SpO2 < 90%, n | 0 | 7 (29%) | 0.003 | 1 | 3 | 3 |
| Mean 3% ODI (events/h) | 1.5 ± 1.3, 1 [0; 5] | 20 ± 25, 9 [0.9; 81] | < 0.001 | 5 ± 5, 3.1 [1; 20] | 38 ± 29, 33 [10; 73] | 52 ± 19, 44 [40; 81] |
| 3% ODI > 10 events/h, n | 0 | 10 (42%) | < 0.001 | 1 | 5 | 4 |
| Mean PtcCO2 (mmHg) | 42 ± 5, 43 [31; 51] | 40 ± 4, 41 [32; 48] | 0.351 | 40 ± 4, 41 [32; 45] | 43 ± 5, 44 [38; 48] | 40 ± 4, 40 [35; 43] |
| Maximal PtcCO2 (mmHg) | 45 ± 4, 45 [35; 53] | 44 ± 5, 43 [38; 56] | 0.480 | 43 ± 3, 43 [38; 47] | 48 ± 8, 49 [40; 56] | 43 ± 4, 42 [39; 47] |
| % of time with PtcCO2 > 50 mmHg (%) | 4 ± 17, 0 [0; 79] | 0 | 0.352 | 0 | 0 | 0 |
| Patients with > 2% of time with PtcCO2 > 50 mmHg, n | 1 | 0 | 1.000 | 0 | 0 | 0 |
- Note: Bold numbers (p) represent the significant p values.
- Abbreviations: 3% ODI: 3% oxygen desaturation index, AHI: apnea-hypopnea index, AI: apnea index, PtcCO2: transcutaneous carbon dioxide pressure, SpO2: pulse oximetry.
- a Velopharyngeal insufficiency, cleft palate, and/or laryngeal anomaly, glossoptosis, retrognathia, micrognathia, or Pierre-Robin sequence, adenotonsillar hypertrophy.
- b Gastro-esophageal reflux, treatment for constipation, celiac disease, or hemorrhagic rectocolic disease.
- c Comparisons between patients without and with OSA.
3.3 Therapeutic Management of OSA
Of the 15 patients with mild OSA, the majority did not have any airway intervention (Table S1). Only 2 patients (#4 and #13) had upper airway surgery after the baseline PG. CPAP was resumed in one patient (patient #14) after an increase in the AHI at 15.7 events/h 1 year later. The patient could then be weaned from CPAP after a tonsillectomy with normal follow-up PGs up to age 7 years. Three patients (#1, #6 and #9) had a subsequent PFP, with patient #1 having a follow-up PG that was normal.
Of the 5 patients with moderate OSA, one patient with a congenital cardiopathy (#16), already treated with long-term oxygen therapy, continued oxygen therapy at a lower flow rate, and 2 patients (#19 and #20) required CPAP with persistent moderate OSA at follow-up PG.
The 4 patients with severe OSA were all less than 1 year of age. They all had a very high AHI ranging from 20.5 to 42.3 events/h (Table S1). One infant (#21) was treated with CPAP for 1 year with normal PGs at follow-up due to a spontaneous correction of OSA. Two patients (#22 and #24) had laryngeal surgery, with one patient requiring CPAP until age 1.6 years (patient #24). Patient #22 had a PFP at the age of 7 years after a normal PG at age 6.6 years and no follow-up PG after the PFP. Only one patient (#23) had no OSA treatment because of a rapid spontaneous improvement of OSA at age 0.8 years, with an AHI at 23.4 events/h at age 0.3 years and at 11.5 events/h at age 0.8 years.
3.4 Follow-Up of Patients Without OSA at Baseline
Of the 28 patients without OSA, 6 (21%) had a follow-up PG (Table S2). For 4 patients, the follow-up PG was performed after a PFP. Three patients (#3, #4, #6) did not develop OSA after the PFP, while one patient (#2) developed severe OSA, which was treated with CPAP for 8 months, with a spontaneous improvement thereafter. The 2 last patients (#1 and #5) had a normal PG 2 and 3 years later, respectively, with one patient having an adenotonsillectomy 2 months after his baseline PG.
4 Discussion
Our study is one of the first that assesses the prevalence and risk factors of OSA in a consecutive, non-selected cohort of children followed at a national reference center. Our results show that the prevalence of OSA is high, reaching almost 46%, but with a majority of patients (15/24, 63%) having mild OSA. A younger age, and in particular an age less than 1 year, but also an associated immune deficiency or a pulmonary disease, were associated with a higher prevalence of OSA.
The high prevalence of OSA has been reported in previous studies, but most patients were selected based on sleep-disordered breathing symptoms or pre-operative planning (Kennedy et al. 2014). Screening for OSA by means of validated sleep questionnaires has indeed shown that patients with 22q11DS are at highest risk for positive screens (Silvestre et al. 2014), but, as in all children with complex OSA related to craniofacial anomalies, no significant correlations were observed between sleep questionnaires and OSA presence or severity (Ingram et al. 2024). This underlines the importance of systematic screening for OSA in children with 22q11DS, as recommended recently by the 22q11.2 Society (Óskarsdóttir et al. 2023). OSA figures among the list of features and risks in children with 22q11DS, but a sleep evaluation is only recommended between ages 1–5 and 6–12 years. Our results show that OSA is more common and more severe in infants aged under 1 year, as compared to older children. As children with 22q11DS are at risk of cognitive dysfunction and learning and behavioral problems, the treatment of OSA and the maintenance of a good sleep quality are of paramount importance, namely during the first months and years of life (Óskarsdóttir et al. 2023). Indeed, as we have shown in infants with Down syndrome, who share with children with 22q11DS facial anomalies and neurocognitive dysfunction, the risk of OSA culminates during the first months of life (Fauroux et al. 2024). Most importantly, during this crucial period of rapid neuronal growth, the diagnosis and treatment of OSA may be associated with an improvement in neurocognitive function and behavior (Fauroux et al. 2024). Our results also confirmed that a majority of patients with 22q11DS who had OSA had mild OSA (Kennedy et al. 2014). Regarding the follow-up, none of the patients with OSA developed severe OSA, except patient #14 who had an AHI of 15.7 events/h at the age of 3.8 years and patient #23 who had an AHI of 11.5 events/h at the age of 0.8 months.
The mechanisms by which an immune deficiency or a pulmonary disease, such as asthma, trachea-bronchomalacia, or pulmonary hypertension, are associated with an increased risk of OSA are unclear. An immune deficiency may be associated with an increased risk of upper airway infection and a consequent hypertrophy of the adenoids and/or the tonsils. But an adeno-tonsillar hypertrophy was observed in only 7 (14%) of the patients. An immune deficiency or a pulmonary disease may favor upper airway inflammation, which may precipitate or aggravate OSA in young children. Importantly, a prior upper airway intervention aiming at improving upper airway patency, such as an adenoidectomy and/or tonsillectomy or a laryngeal surgery, or even interventions associated with a risk of OSA, such as VPI surgery, did not affect the risk of OSA.
VPI surgery, which includes posterior pharyngeal augmentation, palatoplasty, sphincter pharyngoplasty, PFP, and a combination of these procedures, narrows the velopharyngeal port and may predispose patients to OSA. In a series of 21 patients, 4 (19%) developed OSA after VPI surgery, but all were in the mild to moderate range (Crockett et al. 2014). Another study reported a prevalence of OSA after VPI surgery (PFP, n = 16, sphincteroplasty n = 1) of 9/17 (53%) patients, with all patients having mild to severe OSA (median AHI 3 events/h, range 1.9–15) (Kennedy et al. 2014). A more recent study reported that mean obstructive AHI did not change significantly after VPI surgery (1.1 vs. 2.1 events/h) (Lee et al. 2020). Only 2/40 patients had significant OSA before and after surgery, with both having severe OSA requiring CPAP therapy. But this potential risk of OSA seems to be low in children with a normal pre-PFP PG and no SDB symptoms after VPI surgery (Crockett et al. 2014; Trabelsi et al. 2022). Indeed, in another recent study, only 2 out of 18 (11%) patients with VPI who had a normal preoperative PG developed new-onset OSA after PFP (Trabelsi et al. 2022). The type of VPI surgery may impact the risk of OSA, with PFP being associated with a very low risk of OSA (Trabelsi et al. 2022). This is shown in the present study, with only one patient (patient #2) developing a transient OSA after PFP.
Adeno-tonsillectomy represents the first-line treatment for OSA, but only a few patients with OSA had upper airway surgery in the present study. Of the 15 patients with mild OSA, one patient (patient #4) had a tonsillectomy with a pre-operative AHI of 1.4 events/h, another patient (patient #13) had an adenoidectomy and a turbinoplasty with a pre-operative AHI of 3 events/h, and the last patient (patient #14) had a tonsillectomy after the worsening of his OSA at the age of 4 years. None of the patients with moderate OSA had any upper airway surgery. Concerning the 4 young patients with severe OSA, 2 (patients #22 and #24) had prior laryngeal surgery at the age of 2 and 5 months, with a tonsillectomy at the age 6 years in patient #22. Of note, one patient (#5) with no OSA also had an adenotonsillectomy following baseline PG. Our findings show that hypertrophy of the adenoids and/or the tonsils is not the main cause of OSA in children with 22q11DS, as in other disorders at risk for complex OSA. The low percentage of adenoidectomies in the present series may also be explained by the potential risk of VPI after this procedure in patients with 22q11DS (Perkins et al. 2000). But no cases occurred in our sample. Six (12%) patients required a transient treatment with CPAP, with no patient requiring a long-lasting CPAP treatment. This shows that upper airway surgery is not always able to cure or improve OSA in children with 22q11DS, or that CPAP may be an effective therapy following OSA after VPI surgery. Three patients required CPAP for a short period during the first year of life, underlining the possibility of a spontaneous improvement of OSA with age.
Our study has several limitations. This cohort comprises patients with 22q11DS seen at a national reference center. PG were performed at different ages with no systematic PG at follow-up. Only 21% of the patients without OSA at baseline PG who had a follow-up PG, which may underestimate the real prevalence of OSA at an older age. Moreover, PG were not systematically performed prior to surgery. The patients had a PG and not a polysomnography, which may underestimate the hypopnea index due to the impossibility to score hypopneas associated with an arousal. The patients had multiple comorbidities, which may affect sleep disordered breathing and the risk of OSA. We did not use an age-specific definition of OSA in infants aged < 1 year, as they may have a higher physiologic AHI than older children. However, all the 4 infants with severe OSA had very high AHI ranging from 20.5 to 42.3 events/h (Table S1).
In conclusion, the present study highlights the high prevalence of OSA in children with 22q11DS. OSA severity is mainly mild except in infants aged < 1 year, which may consider the extension of the systematic screening for OSA during the first year of age (Óskarsdóttir et al. 2023). An immune deficiency and a pulmonary disease are more common in patients with OSA as compared to patients without OSA. Follow-up shows that OSA is rare at an older age, underlining the efficacy of a multidisciplinary management.
Author Contributions
All authors read and approved the final manuscript. B.F. conceptualization, methodology, supervision, data analysis, validation, and writing – review and editing. D.P.L.R. writing – original draft, data curation, and formal analysis. L.G. and C.P. recruitment of patients, data analysis, and final writing. S.K. formal analysis and writing – review and editing. F.M. and R.N. final writing.
Conflicts of Interest
The authors declare no conflicts of interest.
Open Research
Data Availability Statement
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.
