Sleep in Prader-Willi Syndrome: Sleepiness and Sleep Disordered Breathing- Relationship to Genotype, Growth Hormone Therapy and Body Composition
Korwyn Williams 1, Ann Scheimann 1,2, Vernon Sutton 1, Betsy Hayslett 3, Dan Glaze 1
Departments of Neurology, Medical Genetics and Pediatrics, Baylor College of
Medicine, Houston, TX.
Prader-Willi syndrome (PWS) is a genetic disorder, characterized by early hypotonia followed by onset of obesity, short stature, behavioral problems during childhood and underlying growth hormone (GH) deficiency. Excessive daytime sleepiness is a common symptom in PWS. Individuals with PWS also have primary abnormal ventilatory responses to hypoxia and hypercapnia; these abnormalities may be exacerbated by obesity.
Retrospective chart review of all patients followed in the Prader-Willi Syndrome clinic with sleep studies performed at Texas Children's Hospital from January 2000 to the present. Patients who underwent sleep studies outside of BCM/Methodist Hospital were excluded from the study. Clinical data collected for analyses included: age at the time of the sleep study, sex, ethnicity, genotype (uniparental disomy/deletion/imprinting mutation), anthropometry, body composition via Total Body Electrical Conductivity (TOBEC), and growth hormone dosage. BMI values were adjusted for age, sex and ethnicity using the BMI z-score calculator available through Ken Ellis, Ph.D. Sleep study data gathered included: sleep efficiency, apnea/hypopnea index, maximum carbon dioxide level, minimum oxygen saturation, arousal index, and objective tests of daytime sleepiness (mean sleep latency tests). Statistical analyses performed included standard population statistics from Windows Excel software. This study was approved by the IRB at Baylor College of Medicine.
A total of 32 patients (50% male; 68% Caucasian, 9% African-American, 28% Hispanic) followed in the PWS clinic at Texas Children’s Hospital underwent sleep studies. Genetic records revealed 18 (56%) deletions /13 (41%) UPD/1 (3%) imprinting; records on 3 patients (7%) were not available on initial review. The mean age at the time of the study was 8.2 (+/- 5.9) years. Mean BMI z-score was 1.97 ( +/- 1.36). Mean % body fat via TOBEC was 34.5% (+/- 8.6). 15 patients (47%) were on GH replacement at the time of the sleep study; the mean GH dose was 0.221 (+/- 0.055) mg/kg/wk. Mean sleep efficiency was 86 +/- 11%. Mean apnea-hypopnea index was 15.1 +/- 25 events/hour. Minimum oxygen saturation was 80 +/- 12%. Maximum carbon dioxide level was 56 +/- 15 mm Hg. There were 11 +/- 9 arousals/hour. Of the fourteen patients who underwent multiple sleep latency tests (MSLTs), the average mean latency to sleep onset was 9.6 + 4.4 minutes; six patients experienced at least one episode of REM onset sleep with the MSLT. While the BMI z-scores were significantly different between growth hormone treated and untreated patients (two-tailed t= 0.02), there was no significant difference regarding apnea/hypopnea indices, minimum oxygen saturation, or MSLTs (p>0.2 for all).
Prader-Willi patients are at increased risk for sleep-disorder breathing and for excessive daytime sleepiness. Other than lower BMI z-scores, growth hormone treated patients did not differ from non-treated patients in respect to parameters of sleep disordered breathing or daytime sleepiness. The presence of significant variability in some of the sleep parameters suggests that other factors may play a larger role in sleep-disordered breathing and daytime sleepiness than obesity; further subset analysis and correlation is planned.