Low incidence of pulmonary hypertension in children with suspected obstructive sleep apnea: A prospective observational study.

Objectives: Pulmonary hypertension (PH) secondary to obstructive sleep apnea (OSA) is an uncommon but serious perioperative risk factor in children undergoing surgery for adenotonsillar hypertrophy. Routine pre-operative echocardiography is commonly requested if severe OSA is suspected. We investigated the incidence of PH in children with suspected OSA and explored the association between PH and OSA severity.

Methods: A prospective study of children aged 1-13 years with suspected OSA admitted for overnight oximetry (OO) and echocardiography at a pediatric referral hospital in Cape Town, South Africa from 2018 to 2019. OSA severity was defined by McGill Oximetry Score (MOS): MOS 1-2 (mild-moderate) and MOS 3-4 (severe). PH was defined as mean pulmonary arterial pressure (mPAP) ≥20 mmHg estimated on echocardiographic criteria. Children with congenital heart disease, underlying cardio-respiratory or genetic disorders, and severe obesity were excluded.

Results: One hundred and seventy children median age 3.8 years (IQR 2.7-6.4) were enrolled and 103 (60%) were female. Twenty-two (14%) had a BMIz >1.0 and 99 (59%) had tonsillar enlargement grade 3/4. One hundred and twenty-two (71%) and 48 (28%) children had mild-moderate and severe OSA, respectively. Echocardiographic assessment for PH was successful in 160 (94%) children of which eight (5%) had PH with mPAP 20.8 mmHg (SD 0.9): six with mild-moderate OSA and two with severe OSA. No significant difference in mPAP and other echocardiographic indices was observed in children with mild-moderate (16.1 mmHg; SD 2.4) and severe OSA (15.7 mmHg; SD 2.1). Similarly, no clinical and OSA severity differences were observed in children with and without PH.

Conclusion: PH is uncommon in children with uncomplicated OSA and there is no association of PH with severity of OSA measured by OO. Routine echocardiographic screening for PH in children with clinical symptoms of OSA without co-morbidity is unwarranted.