For many years, polysomnography (PSG) has been the gold standard for diagnosing sleep-disordered breathing (SDB). Traditional PSG is a labor-intensive procedure involving a continuous multichannel recording of various physiologic functions and activities during sleep. One Internet source puts the average cost for a sleep study at about $2,500 to $2,700.1 Many sleep labs have a backlog of patients and are scheduling several months in the future for patients to be tested. In addition to the cost and the delay in getting a study done, PSG can be uncomfortable for the patient due to the many wires, belts, and attachments that are required to gather the various bits of data, and it involves a large investment of capital for equipment and building. In light of these issues, researchers have been looking for and evaluating alternative ways to diagnose sleep disorders—particularly obstructive sleep apnea.


From a study published in Sleep, researchers used image analysis software to examine photographs of the face and neck and landmarks in a process called “craniofacial photographic analysis” to assess the patient for risk factors that relate to OSA.2 The risk factors that are described in this approach include skeletal restrictions, regional adiposity, and obesity. The research used the photographic evidence combined with clinical evidence to predict the occurrence of OSA. Clinical evidence included age, body mass index (BMI), witnessed apneas, comorbidities, the score from the Epworth Sleepiness Scale (ESS), the modified Mallampati class (a scoring system that examines the oropharynx), and neck and waist circumference.

Using different variations of these assessments, the researchers found that predicting OSA by photographic evidence alone was better than using clinical evidence alone. Moreover, the best prediction model found in this research project was a combination of 13 photographic measurements and two particular clinical measurements: witnessed apneas and the modified Mallampati class. This combination was compared to standard PSG results using a prospective cohort study of 180 subjects. PSG found OSA in 114 of the subjects (63.3%). The prediction model mentioned above correctly classified 79.4% of the subjects and had a sensitivity of 85.1% and a specificity of 69.7%. The authors state that this method “may potentially allow prediction of OSA with photographs taken using any camera, although further validation is required.”2

Another group of researchers used radiologic images of the skull and neck (lateral radiographs) to study the anatomy of the upper airway—described as cephalometry. This study3 produced an equation for predicting the presence of OSA using neck circumference, gender, desaturations in pulse oximetry, the Epworth Sleepiness Scale score, and the distance measured between the gonion (the point on each side of the lower jaw at the mandibular angle) and the gnathion (the lowest point in the midline of the lower jaw). The distance between the gonion and gnathion can be thought of as a measure of the length of the lower jaw from just under the ear to the anterior edge in the midline.

When comparing the prediction model to the results of a full sleep study performed and scored according to standard procedures (R&K), the sensitivity was 93.9% and the specificity was 83.3%. The authors point out that computed tomography (CT) and magnetic resonance imaging (MRI) would give much better detail as opposed to the lateral radiograph, but these are more expensive and involve higher levels of radiation.3

Both of these prediction models (one using photography and the other radiography) need to have the respective model validated in the general population to see how they perform as a tool for screening patients.


Snoring is a common issue in the general population and disturbs the sleep of both the patient as well as the bed partner. Estimates for the prevalence of snoring range from 20% to 40% of the population,4 and when examined by gender, an estimated 60% of men and 32% of women are snorers.5 In children 6 to 11 years old who have sleep-disordered breathing, one study found that the presence of snoring coupled with excessive daytime sleepiness and learning problems may have strong enough predictive value to remove the need for polysomnography.6 There has been a distinction made between those who have steady snoring and no OSA versus those who have irregular snoring associated with the resumption of breathing after an obstructive apnea event. There are acoustic differences for those with OSA compared to those with no OSA (snoring power measured in Hz shows a distinction for those who have snoring above 800 Hz versus those who have snoring power below 800 Hz). There has also been a correlation found between the loudness of snoring and sleep-disordered breathing.

Snoring is characterized by a particular pitch, and a snoring period has stops and starts in the production of the sound—these are called intra-snore-pitch-jumps (ISPJ). According to a study published in 2005, analysis of snoring and ISPJ uncovered the presence of OSA with a sensitivity of 86% to 100% and a specificity of 50% to 80%.7

In a more recent publication of the journal Laryngoscope from April 2010, researchers examined the sound intensity and sound wave frequencies of snoring in patients undergoing full-night polysomnography. They were able to differentiate between snoring patients according to the severity of their OSA based on the two parameters (intensity and frequency).5


There are new assessment tools for sleep and other related activities that are being developed and evaluated. These new tools allow for freedom of movement while the patient is being assessed. There are no wires or connections needed; these tools monitor from a distance or utilize biochemical analysis to uncover issues with sleep. Thermal infrared imaging is a method of monitoring airflow that uses an infrared camera to “see” the expiration from a patient.8 Called an expiratory plume, this image confirms the presence or absence of ventilation.

Researchers evaluating the thermal infrared imaging system found that this approach was very close in performance when compared to nasal pressure, end-tidal CO2, and thermistor monitoring of ventilation (agreement measuring 99.6%, 98%, and 99.2%, respectively). This novel tool will need more study and will have to be much lower in cost before being considered as clinically useful. Still, it is a thought-provoking approach to noninvasive monitoring of ventilation that can be done without ever touching the patient.

Published research from early 2010 involves a sleep monitoring system that is incorporated in the bed. This system uses an air mattress that is made of 20 air cells running across the bed. Two of the cells are sensor cells that monitor pressure; the remainder are support cells. The two sensor cells are connected through a pneumatic resistor (called a balancing tube) that reduces problems with recording when a patient changes position and shifts their weight.9 This system has some problems that need to be solved before going forward in consideration for accepted clinical monitoring. These problems include issues with accurately recognizing some events with specific individuals due to variations in height, body weight, and signal strength from the subjects, and problems with motion artifact.


OSA has been associated with increased sympathetic activity, and many have considered this to be the main underlying cause for hypertension in OSA patients. The increased sympathetic activity is also thought to be the link between OSA and cardiovascular diseases. Hypoxia is thought to trigger this increased activity through stimulation of the sympathetic system as well as stimulation of the adrenal gland. Increased presence of catecholamines in the urine is a measure of sympathoadrenal excitation. In a study from Europe published in 2002, the researchers found that in hypertensive males, there is an independent association between sleep disturbances and urinary catecholamines.10

A more recent publication from 2009 examined the presence of proteins in the urine of children to see if this biomarker could distinguish between those with OSA and those who had primary snoring but no OSA, and those who were healthy control subjects. Children between 2 and 9 years of age were studied, and the findings showed that this biomarker could potentially be used as a screening tool for OSA. When examining the data from four particular proteins of the 16 proteins that have been associated with OSA, sensitivity was found to be 95% and specificity reached 100% in linking to OSA when compared against polysomnography.11

As with many of these new approaches to uncovering OSA, the authors in both of these studies recommend well-designed large studies to validate the use of these tools in screening for OSA.

With the growing number of alternative and imaginative tools, there may be a combination that captures the highest level for quality, reliability, and accuracy. Research is needed to look for the right combination and to validate the tool. Accomplishing this will possibly help reduce both patient cost and clinic cost, patient wait time, and patient discomfort.

Bill Pruitt, MBA, RRT, CPFT, AE-C, is a senior instructor and director of clinical education in the Department of Cardiorespiratory Sciences, College of Allied Health Sciences, at the University of South Alabama in Mobile, and a PRN therapist at Springhill Medical Center and Mobile Infirmary Medical Center in Mobile. He can be reached at

  1. New Choice Health medical cost comparison. Accessed November 3, 2010.
  2. Lee RW, Petocz P, Prvan T, Chan AS, Grunstein RR, Cistulli PA. Prediction of obstructive sleep apnea with craniofacial photographic analysis. Sleep. 2009;32:46-52.
  3. Julià-Serdà G, Pérez-Peñate G, Saavedra-Santana P, et al. Usefulness of cephalometry in sparing polysomnography of patients with suspected obstructive sleep apnea. Sleep Breath. 2006;10:181-187.
  4. Pevernagie D, Aarts RM, De Meyer M. Physiological review: the acoustics of snoring. Sleep Med Rev. 2010;14(2):131-144.
  5. Fiz JA, Jané R, Solà-Soler J, Abad J, Garcia MA, Morera J. Continuous analysis and monitoring of snores and their relationship to the apnea-hypopnea index. Laryngoscope. 2010;120:854-862.
  6. Goodwin JL, Kaemingk KL, Mulvaney SA, Morgan WJ, Quan SF. Clinical screening of school children for polysomnography to detect sleep-disordered breathing—the Tucson Children’s Assessment of Sleep Apnea study (TuCASA). J Clin Sleep Med. 2005;1:247-254.
  7. Abeyratne UR, Wakwella AS, Hukins C. Pitch jump probability measures for the analysis of snoring sounds in apnea. Physiol Meas. 2005;26:779–798.
  8. Murthy JN, van Jaarsveld J, Fei J, et al. Thermal infrared imaging: a novel method to monitor airflow during polysomnography. Sleep. 2009;32:1521-1527.
  9. Shin JH, Chee YJ, Jeong DU, Park KS. Nonconstrained sleep monitoring system and algorithms using air-mattress with balancing tube method. IEEE Trans Inf Technol Biomed. 2010;14(1):147-156.
  10. Elmasry A, Lindberg E, Hedner J, Janson C, Boman G. Obstructive sleep apnoea and urine catecholamines in hypertensive males: a population-based study. Eur Respir J. 2002;19:511-517.
  11. Gozal D, Jortani S, Snow AB, et al. Two-dimensional differential in-gel electrophoresis proteomic approaches reveal urine candidate biomarkers in pediatric obstructive sleep apnea. Am J Respir Crit Care Med. 2009;180:1253-1261.