Maislin et al developed a score for predicting sleep apnea in patients. This can be used at sleep centers as part of the initial evaluation and to help identify patients who may require further studies. The authors are from sleep centers at the University of Pennsylvania, Johns Hopkins Hospital and in Pittsburgh.
Endpoint: prediction of respiratory disturbance index (RDI) >= 10
respiratory disturbance index (RDI) =
= (number of apneas per hour sleep) + (number of hypopneas per hour sleep)
Parameters for predictive score:
(1) symptoms of sleep apnea
(2) age
(3) gender
(4) body mass index (BMI)
Symptoms:
(1) snorting or gasping
(2) loud snoring
(3) breathing stops, choke or struggle for breath
|
Symptom Frequency
|
Points
|
|
never
|
0
|
|
rarely
|
1
|
|
1-2 times per week
|
2
|
|
3-4 times per week
|
3
|
|
5-7 times per week
|
4
|
where:
• I assume that the times per week indicate affected nights per week.
symptom score =
= SUM(points for symptoms known to be present)
index 1 = apnea symptom index =
= (symptom score) / (number of symptoms present)
where:
• Other indices evaluated in the study included difficulty sleeping, excessive daytime sleepiness and narcolepsy symptoms.
• Index 1 is the average score for the "nonmissing" symptoms.
X =
= (1.299 * (Index 1)) + (0.163 * (BMI)) – (0.028 * (index 1) * (BMI)) + (0.032 * (age)) + (1.278 * (points for gender)) – 8.160
probability of (RDI >= 10) =
= 1 / (1 + EXP((-1) * X))
where:
body mass index =
= (body weight in kilograms) / ((height in meters) ^2)
|
Gender
|
Points
|
|
female
|
0
|
|
male
|
1
|
Predicting RDI >= 10 from Index 1 alone (based on patients from Philadelphia)
|
Index 1
|
Percent with RDI >= 10
|
|
0 to 0.99
|
20%
|
|
1.00 to 1.99
|
46%
|
|
2.00 to 2.99
|
62%
|
|
3.00
|
71%
|
|
3.33 to 3.67
|
73%
|
|
4.00
|
74%
|
from Table 6, page 162
where:
• Using Index 1 alone was helpful for patients who were NOT extremely obese (BMI < 40).
