Background: Complications in sickle cell anemia (SCA, HbSS) include neurocognitive difficulties in attention and processing speed associated with low daytime and night-time oxygen saturation. These effects can be compounded by obstructive sleep apnea (OSA).However, there is concern that oxygen supplementation in SCA may lead to bone marrow suppression. Continuous Positive Airways Pressure (CPAP) is an accepted treatment for OSA in the general population and prevents dips in oxygen saturation.. The aim of this single-blind, randomised, controlled phase II trial is to compare Auto-adjusting CPAP (APAP) with standard care to standard care alone in subjects with HbSS to determine whether the intervention is safe and improves Cancellation, a measure of selective attention and processing speed.

Methods: Eligibility criteria included ability to provide informed consent, age >8 and <16 years, diagnosis of HbSS and mean overnight saturation of <90% for <30% of the night. Key exclusion criteria were overnight respiratory support, respiratory or decompensated cardiac failure, chronic transfusion or contra-indications to APAP therapy or MRI.

Minimisation/stratification factors were age group (8-11, 12-15 years), silent infarction on MRI, minimum overnight oxygen saturation >90% or <90%, and hydroxyurea (HU) use.

APAP adherence was defined as using APAP for an average of 4 hours a night for >50% of the time and was recorded using software documenting hours of use each night. Participant support in terms of appropriate facemask and facilitating adherence were provided by an unblinded sleep physiologist.

Full blood counts were obtained at baseline, 2 weeks, 3 months and 6 months.

Data were analysed by intention-to-treat. The primary outcome is change in the Cancellation subtest from the Wechsler scales, and secondary outcomes include general cognitive functioning, assessed at baseline and after 6 months of treatment by assessors blind to treatment assignment. Analysis of Covariance (ANCOVA) models, adjusted for minimisation factors, were used to calculate least-square mean changes from baseline to 6 months.

Results: 30 children (18 boys; median age 12.5; range 7.9-16 years) with SCA were randomised to standard care + APAP (n=15) or standard care alone (n=15) for 6 months. One child in the standard care alone arm withdrew after 6 weeks, and 8 children in the APAP arm were not adherent to treatment. Increase in cancellation score was numerically greater in the APAP arm (mean 1.46, SE 0.59 vs 1.01, SE 0.61) but this was not significant (mean difference 0.44, 95% CI: -1.42; 2.31; p=0.626). Increase in Cancellation score was greater (mean 2.63; 95%CI: 0.95, 4.30) in those whose adherence was in the highest quartile (> 2.4 hours/night) compared with the other 3 quartiles (mean 0.71, 95%CI: -0.32, 1.75; mean difference 1.91, 95%CI: -0.06, 3.88; p=0.057). In subjects assigned to APAP, cancellation scores were significantly higher with hydroxyurea use (p=0.01). There was no evidence of decline in haemoglobin in either group; hydroxyurea use was associated with an increase in haemoglobin (p=0.01). There were 7 subjects with serious adverse events in the standard care alone arm, compared to 3 in the APAP + standard care arm, all related to hospital admission for pain.

Discussion: APAP for 6 months is feasible and safe in children with SCA. Alhough >50% were not adherent by the pre-defined definition, those who were compliant appeared to have more benefit in terms of improvement in attention/processing speed. There appears to be an interaction with HU use, consistent with the importance of oxygen supply and carriage for brain function. If delivery of the intervention can be improved, avoidance of oxygen desaturation with overnight respiratory support alongside HU use may play a role in improving cognition in SCD.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

Sign in via your Institution