A sudden jump in split-night studies and number of patients found to have OSA turns out to be a case of better equipment.
When considering new or upgraded polysomnographic testing equipment for their sleep laboratories, most people will look at the major manufacturers of the sleep systems. For a majority, the main area of concern is the price of the equipment, and this usually acts as the main driving force when they consider which system to adopt. Other features that are used as marketing tools by the system manufacturers are the number of parameters the system can record, the ease of scoring/reviewing the study, the ability to re-reference the different parameters that are recorded, the incorporation of digital video, and, finally, the appearance or layout of the screen that displays the test results.
One assumes that, with this information, the patient will receive a quality study, but is that necessarily true? There are areas that should be included in the review of sleep systems to ensure that the patient will be evaluated and diagnosed properly. These areas are airflow, thoracic and abdominal effort, and relevance these two measures will have with positive airway pressure (PAP) therapy.
When purchasing or updating a system, you will meet with sales representatives who will often talk about the pros of their companies’ systems as compared to other similar products currently on the market. They may also compare the latest version of their companies’ devices to older versions of the same systems. In December 2004, the sleep laboratory I was working for updated the polysomnographic testing equipment used for its four beds from thermocouples and piezo crystal belts to nasal/oral pressure transducers and respiratory inductive plethysmography (RIP) technology. The laboratory kept data spreadsheets of patients, and, from those spreadsheets, we saw how the changing of equipment impacted different areas of patient care. We noted changes in the overall number of patients diagnosed with OSA, as well as in the number of patients who were diagnosed with no sleep disordered breathing.
One of the things that we found during the first month after the new system was installed was a dramatic increase in the number of “emergency split-night studies” the laboratory did. We had a split-night study protocol at that time that listed three reasons a patient could be changed from a diagnostic polysomnogram to a split-night study. These included:
1. Severe desaturations, with a saturation of less than 70%.
2. Cardiac abnormalities (for example pauses greater than 3 seconds in duration) that are associated with sleep disordered breathing.
3. An apnea/hypopnea index (AHI) greater than 50 events per hour of sleep.
We trended the patient volume numbers by factors such as the total numbers of polysomnograms, split-night studies, seizure studies, multiple sleep latency tests (MSLTs), PAP titrations, and bilevels. Overall, the percentages for split-night studies were relatively consistent until the first week of December, which was when the upgrade was installed. After that, there was a period of 4 to 6 months when the percentages of split-night studies appeared to be very high as compared to the total number of studies conducted each month (Table 1).
Through staff meetings, as well as by looking at the scoring of the studies, we found a few areas that raised concern. The first area was the overall quality of the “air flow.” The original equipment used before the upgrade had measured airflow using a thermocouple. The upgraded system allowed airflow to be measured with pressure transducers. The pressure transducer was integrated into the system via a DC channel for both the diagnostic polysomnogram and the PAP titration. We discussed this with the technical staff, as well as the medical director, and came to a consensus agreement that flow limitation was much easier to read via the pressure transducer. With the ability to see airflow plateau, flow limitation was easily picked up with the pressure transducers. Then there was also the upgrade to RIP belts. These belts allowed us to see using the tracings as flow volume loops. These flow volume loops made seeing respiratory events much easier.
RIP, in essence, measures the volume changes in the abdominal and thoracic cavities with bands that are placed around the cavities. There is a tiny wire within the belt that will reflect changes in the inductance of the wire as the patient breathes in and out. These changes are calibrated to detect differences in volume, thus showing alterations in respiratory effort. These changes are represented as a sinusoidal waveform on a polysomnographic study, the waveforms we are so used to.
 |
 |
The system we invested in has the ability to show this data as a flow volume loop (Image 1). The two vertical pink lines are the “breath” that we are looking at as compared to a baseline breath. The loop window has two different colored loops; the red is the baseline breath and the blue represents the breath we are looking at above (between the pink lines). The percent tidal volume change (%Vt) shows 18%; this means that this breath is only 18% of the baseline breath, or it decreased by 82%, quite a decrease that easily can be labeled a hypopnea or possibly an apnea. It is obvious in the tracing above that most people would score this as an obstructive apnea, so the flow volume loop is not needed. But as in Figure 2 (above), we can see that the breaths labeled 2 through 4 appear normal. However, with further investigation, as on a breath-by-breath analysis via the flow volume loop, we can see that there are indeed changes. A question could be, “Where does the hypopnea begin?” By breath No. 5, the tidal volume is reduced by 24% (almost meeting part of the criteria for a hypopnea). By breath No. 6, the tidal volume is reduced by 34%, and by breath No. 7, it is reduced by 50%. Breaths 5 and 6 are only 8.3 seconds total, but if you were to add in breath No. 4, there would be 12.76 seconds, thus meeting the time criteria for a hypopnea, and an overall reduction of more than 70%.
With the two new technologies in place, there was the increase in split-night studies (Table 1) from the high AHI being seen due to the increased sensitivity of the equipment. We can also corroborate this with the overall diagnoses for patients for the 3 months prior to the new equipment as well as for the 3 months after. There was an increase (overall) for the severity of the sleep disorders, as measured by the total number of events per patient. This was also reflected in the number of patients being diagnosed with moderate or severe sleep disordered breathing (Table 2).
As seen from the table, we can notice a few differences from prior to the December upgrade versus post upgrade. There is a dramatic increase in the number of patients that are diagnosed with OSA. Pre-upgrade, the percentages were 45%, 55.8%, and 50%, respectively, for the months of September, October, and November. For December and January, there was an increase of about 20% more patients with the diagnosis of OSA. Upon further examination, there is an increase in the number of patients who are diagnosed with moderate to severe OSA as compared to the months prior to December. This is reflected by the AHIs of the patients on their diagnostic polysomnograms. Most likely, this was due to the upgrades, such as pressure transducers and RIP technology aiding in the diagnosis of the patients’ sleep disorders. It can be concluded that the patients tested prior to the use of pressure transducers and RIP belts may have been underdiagnosed as to the severity of their disorders. Prior to December, the percentage of patients with no OSA was 11.1%, overall, but after the upgrade, the average was only 5.5%. It appears that 50% of the patients that were diagnosed with no OSA may have had sleep disordered breathing but were undiagnosed due to the limited ability of the equipment (thermocouples and piezo crystal belts). It can be concluded that the number of patients considered with no OSA may also be overdiagnosed.
We also noticed that the overall PAP level, which was an average of all patients placed on CPAP, remained relatively consistent. This leads to the conclusion that the thermocouples were accurate for titrating patients on PAP therapy, but may have been inaccurate in evaluating the patients for the severity of their sleep disordered breathing.
When considering the purchase or upgrade of a sleep system, consider the options that will lead to better patient care. These include the ability to re-reference, the use of digital video, scoring/staging ease, and various other features. The ability to properly detect respiratory events, using RIP technology, pressure transducers, and an intergrated flow volume loop, as with the system we chose, will greatly aid in treating the patient population. The ultimate goal is to evaluate, diagnose, and treat the patient’s sleep disorder. Having the ability to properly achieve those goals is very rewarding, and will grow the reputation of your sleep facility for conducting quality studies.
is a clinical instructor in the Respiratory Care Program at the School of Health Technology and Management at Stony Brook University, Stony Brook, NY, and a member of the Sleep Review editorial advisory board. Bernadette White, BS, RRT, RPSGT, supervisor of the John T. Mather Memorial Hospital Sleep Disorders Center in Port Jefferson, NY, also assisted in the creation of this case report.
