| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Baltimore, MD
Dr. Collop is Associate Professor of Medicine, Division of Pulmonary/Critical Care Medicine, Department of Medicine, Johns Hopkins University.
Correspondence to: Nancy A. Collop, MD, FCCP, Division of Pulmonary/Critical Care Medicine, Department of Medicine, Johns Hopkins University, 1830 East Monument St, Room 555, Baltimore, MD 21287; e-mail: ncollop{at}jhmi.edu
Obstructive sleep apnea (OSA) is a highly prevalent disease with estimates that 20% of white men and women with body mass index between 25 and 28 kg/m2 have an apnea-hypopnea index (AHI) of 
5.1
In recent years, OSA has been associated with a number of common morbid conditions, including heart disease, stroke, diabetes, and motor vehicle accidents. These links and the fact that the disease is readily treatable with nasal continuous positive airway pressure (CPAP) have accelerated the need for prompt and accurate diagnosis. The current standard clinical workup includes a history, a physical examination, and a referral to a sleep disorders laboratory for an overnight complete polysomnogram. If OSA is found, a second overnight polysomnogram with nasal CPAP is performed. Given the length of time, the costs and technical expertise required, along with the advances in computer technology, a move toward less complicated techniques in the home has gained popularity. In a recent issue of CHEST, an extensive evidence-based review of the current status of portable monitoring for the diagnosis of OSA was published.2
A subsequent publication in Sleep used this evidence review to develop clinical practice guidelines.3
The outcome of these articles suggests that portable monitoring for the diagnosis of OSA is not quite "ready for primetime." The evidence review was based on 49 articles in the peer-reviewed literature. The authors of the evidence review systematically evaluated each article with the help of an evidence practice center to determine its quality. The evidence was used to generate a practice parameter, which states that unattended portable monitoring devices (specifically level III devices, recording typically 4 to 7 parameters, and level IV devices, recording one or two parameters) should not be used routinely in clinical practice. Many would argue that the question regarding limited portable studies would have been better addressed with a consensus conference rather than a literature-based review; however, given the extreme complexity of this problem, the sponsoring organizations agreed that this was the most appropriate approach. The article by Liesching et al in this month’s issue of CHEST (see page 886) points out many of the hazards of using portable monitoring without appropriate scientific validation.
In this article, the authors retrospectively evaluated patients who had been referred to their sleep laboratory after undergoing a home sleep study with a portable device (SNAP; SNAP Laboratories; Glenview, IL). The home sleep study had been ordered by each patient’s primary care physician. The results show that the SNAP results frequently differed considerably from the polysomnography results. In approximately 65% of the patients studied, the severity of sleep apnea had been misclassified by the SNAP device relative to polysomnography! These differences included both "overdiagnoses" and "underdiagnoses," and have the potential for serious ramifications.
In all fairness, the study is not without faults, as follows: it was retrospective; it compared an AHI during a laboratory-based polysomnogram with a different denominator (ie, total sleep time) than that for the portable device (ie, total recording time), which might have led to an underestimate of sleep-disordered breathing severity by the SNAP device; the studies were not performed at the same time, and sometimes polysomnography was performed several months after the portable device study; there were a number of dropouts from the original group, with data from only 31 of the original 39 patients remaining analyzable; and polysomnography was assumed to be the "gold standard." Nevertheless, this study should raise some red flags about this type of technology. For instance, four patients (13%) with either moderate or severe OSA based on the results of polysomnography would have received a diagnosis of either normal or mild OSA based on the SNAP device. These false-negative results associated with SNAP testing would leave some patients with significant OSA untreated. On the other hand, seven patients (23%) were found to have moderate-to-severe sleep apnea based on the SNAP results, rather than normal or mild OSA based on the laboratory-based polysomnography. These patients run the risk of being unnecessarily treated with nasal CPAP or surgical procedures.
Another problem with the SNAP technology is that the software used to "score respiratory events" is proprietary, so the physicians treating the patient never see raw data. This "black box" approach is disconcerting because the scoring algorithm has not been validated. In addition to estimating AHI, the SNAP recording provides an estimate of the CPAP pressure required for treatment, which is also unvalidated.4
One of the biggest problems with level III and IV monitors is the lack of sleep monitoring. Scoring reductions in airflow alone may be overly sensitive in some patients. Devices specifically using nasal pressure are extremely sensitive to airflow changes and the sensitivity may actually be too high without monitoring for a change in sleep status or some other surrogate of arousal, hence overdiagnosing some patients.5 With this SNAP device, it is possible that flow limitation with snoring could be "scored" as an event even if it was not associated with sleep disruption. On the other hand, the addition of more leads (signals) to portable monitors such as those used to monitor sleep will increase the complexity of these studies and the chance of data loss from unattended studies.
In summary, although portable monitoring for the diagnosis of OSA may be in the offing, we must still make sure that the technology is robust and accurate. While the continued pursuit of algorithms utilizing portable monitors for diagnosing OSA is admirable, cost-benefit analysis, the stratification of patients, and the incorporation of management strategies will be required to determine how portable monitoring should be deployed as a diagnostic strategy for OSA.
References
This article has been cited by other articles:
|
|
 |
|
  T. J. Kehoe, R. P. Millman, and N. Collop SNAP Technology and Sleep Apnea Chest, April 1, 2005; 127(4): 1465 - 1467. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
