[parts of this thread were copied from our old forum]
ATLANTA-The apnea-hypopnea index (AHI) is useless for measuring the severity of sleep-disordered breathing (SDB), says Colin E. Sullivan, MD, PhD. He presented arguments in support of that statement in a pro/con debate at the recent annual meeting of the American Thoracic Society in Atlanta. Offering the opposing view was David M. Rapoport, MD, who maintains that the AHI does have its place in clinical practice.
"There is not much association between the AHI and anything else-sleepiness, muscle dysfunction," or other markers for SDB, claimed Dr. Sullivan, who heads the Sleep Disorders Unit at the University of Sydney in Australia. The management of SDB should hinge on the history, examination, and clinical judgment, he asserted.
LIMITATIONS OF THE AHI
The lack of a standard definition for hypopnea is another limitation of the AHI. Furthermore, measuring hypopnea is difficult because of the inaccuracy of the devices currently available to monitor airflow during sleep.
For example, thermistors do not actually detect airflow but the passage of hot air, Dr. Sullivan explained. Even pressure transducers, which do measure airflow, have only limited ability to detect changes in breathing, he noted.
The AHI can mislead physicians about the severity of SDB, Dr. Sullivan added. At certain points in the menstrual cycle, for example, women with SDB may respond to apnea with a large rise in blood pressure (BP) rather than in the AHI. Catastrophic BP elevations with no change in the AHI were even observed during sleep apnea in a woman with preeclampsia.
In infants and children, apnea usually manifests as partial upper airway obstruction; breathing is loaded even though the AHI is very low. In these groups, the AHI is "unequivocally the wrong metric" of SDB, Dr. Sullivan stated.
What alternatives are there to the AHI? "Hypertension is a good start," said Dr. Sullivan. It has long been known that BP rises in obstructive sleep apnea, he related.
Researchers are also evaluating the usefulness of measuring fibrinogen concentrations, which are often elevated in the morning in patients with SDB. Others are looking at sleep apnea-induced changes in the levels of circulating and cellular mediators and in cellular adhesion molecules.
Probably the best indicator of SDB, however, is simply the response to continuous positive airway pressure (CPAP) treatment. "It really is a no-brainer," Dr. Sullivan remarked, pointing out that CPAP administration is especially easy with the newer devices that automatically set the appropriate amount of positive pressure.
IMPERFECT BUT USEFUL
The AHI may not be perfect, but it is useful, argued Dr. Rapoport. No one has devised anything better for distinguishing people with and without obstructive sleep apnea, pointed out Dr. Rapoport, Medical Director of the Sleep Disorders Center at New York University School of Medicine in New York City.
Furthermore, efforts have been made to improve the AHI. The American Academy of Sleep Medicine has recently attempted to standardize the index, for example, and work has begun to define its normal range.
Also, AHI values correlate with symptoms of SDB. "[The correlation] is mediocre perhaps, but it is not absent," stated Dr. Rapoport, countering Dr. Sullivan's earlier assertion. Because the AHI is imperfect, Dr. Rapoport views it as a marker for the apnea-hypopnea syndrome and not as a definitive metric. He has found the AHI most useful for detecting severe apnea-hypopnea syndrome.
He defines severe cases as those with an AHI of 30 to 50 events per hour or greater. "This is definitely bad .. and I want to treat it," he stressed. Obstructive sleep apnea symptoms are also likely to be severe enough to warrant treatment in patients with an AHI of about 20 per hour; SDB can probably be ruled out at an AHI of about 10 per hour.
Dr. Rapoport cautioned against over-interpreting the "gray zone" between AHIs of 10 and 20 per hour. "I do not quite know what that means," he acknowledged.
1. Sullivan CE, Rapoport DM. The apnea hypopnea index is a useless metric of sleep disordered breathing: pro-con. Presented at: Annual Meeting of the American Thoracic Society; May 21, 2002; Atlanta, Ga.