Abstract
Rationale: Home respiratory polygraphy (HRP) is an alternative to polysomnography (PSG) for sleep apnea–hypopnea syndrome (SAHS) diagnosis. However, therapeutic decision-making is a different process than diagnosis.
Objectives: This study aimed to determine the agreement between HRP and in-hospital PSG for therapeutic decision-making in a large sample.
Methods: Patients with an intermediate or high SAHS suspicion were included in a multicenter study (eight sleep centers) and assigned to home and hospital protocols in a random order. Therapeutic decisions (continuous positive airway pressure, no continuous positive airway pressure, or impossible decision) were made by an investigator in each center, based on using either HRP or PSG and a single set of auxiliary clinical variables. Patients and diagnostic methods (HRP and PSG) were assessed in random order using an electronic database. After a month the same therapeutic decision-making procedure was repeated with the same method.
Measurements and Main Results: Of 366 randomized patients, 348 completed the protocol. The “impossible decision” case was not observed with either PSG or HRP. Therapeutic decisions using HRP had a sensitivity of 73%, a specificity of 77%, and an agreement level (sum of true positives and negatives) of 76%. Patients with higher HRP apnea–hypopnea index (AHI) scores (≥ 30; 41% of the total sample) had a sensitivity of 94%, a specificity of 44%, and the agreement level was 91%.
Conclusions: The HRP-based therapeutic decision was adequate when AHI was high, but deficient in the large population of patients with mild to moderate AHI. Therefore, both selecting patients with a high suspicion and severity of SAHS and future prospective cost-effectiveness studies are necessary.
Home respiratory polygraphy is an alternative to polysomnography for sleep apnea-hypopnea syndrome diagnosis in patients with intermediate and high clinical probability. However, therapeutic decision-making is a different process than diagnosis and its agreement with polysomnography is practically unexplored.
The consideration of therapeutic decision-making reduces the home respiratory polygraphy utility for patients with high apnea-hypopnea index.
The prevalence of sleep apnea–hypopnea syndrome (SAHS) is about 2–5% in the adult population (1). Several studies have shown associations with arterial hypertension (2), cardiovascular mortality (3), and traffic accidents (4, 5).
The gold standard for diagnosis of SAHS is the attended polysomnogram (PSG), but it is time-consuming and expensive. The Type 3 portable monitoring device is an accepted (6) and cost-effective (7–10) alternative for SAHS diagnosis in selected patients. This portable monitoring device (11), also called respiratory polygraphy, includes sensors for airflow, respiratory effort measured with bands, and pulse oximetry recordings.
The therapeutic decision-making process is different from the diagnostic process. It consists of recommending continuous positive airway pressure (CPAP) or other treatments, such as dietary measures and less frequently surgical approaches or mandibular advance devices (12–17). Most guidelines recommend CPAP, depending on the apnea–hypopnea index (AHI) score; clinical symptoms (primarily sleepiness); and comorbidities (primarily hypertension and cardiovascular problems) (12–18). Agreement exists on recommending CPAP for patients with severe SAHS (AHI >30) (6, 12, 14–17). Most guidelines recommend CPAP in mild or moderate SAHS (AHI between 5 and 30) depending on the severity of symptoms and their repercussions (12, 14, 16, 17). Some guidelines recommend CPAP in moderate SAHS (AHI between 15 and 30) (6) or for AHI between 20 and 30 (16, 19) regardless of symptoms or comorbidities and with AHI between 5 and 15 (6) or between 10 and 20 (16, 18) when taking into account symptoms or comorbidities.
To make this therapeutic decision, it is normally assumed that the AHI values from PSG and home respiratory polygraphy (HRP) are equal or at least similar. However, there is known variability between the two methods, more frequently observed for the intermediate and high AHI values (8–10, 20-27), and a systematic underestimation of AHI when measured by HRP (28). The latter is caused by the fact that the number of apneas and hypopneas is divided by recorded time in HRP rather than sleep time, as is the case with PSG. Variability in higher values may have less relevance than in intermediate values, which can be crucial for making the CPAP treatment decision (i.e., AHI between 5 and 30).
We performed a multicenter, randomized, blinded crossover study with the objective of determining the agreement of HRP with PSG for therapeutic decision-making, in a large sample with suspected SAHS.
Some of the results of these studies have been previously reported in the form of an abstract (29) and the cost-effectiveness analysis of diagnosing SAHS was published separately (7).
We included patients between 18 and 70 years old, from December 2008 to December 2009, drawn from patients referred to eight sleep centers in Spain by pulmonary physicians for suspected SAHS because of snoring, observed apneas, sleepiness (Epworth Sleepiness Scale >10), or morning tiredness, and who had no other suspected sleep disorders. We excluded patients with severe and unstable heart disease, who were unable to set up the HRP instrument in a trial, or who refused to participate in the study. Inclusion was sequential and before starting the inclusion process each center selected specific days or weeks per month during the enrollment period to enroll all potential patients. The inclusion period finished when the total number of patients included was 366. To avoid imbalance, each center could include up to 50 patients.
The ethics committees of the eight centers approved the study. Informed consent was obtained in writing from all patients.
All patients underwent PSG and HRP in random order (Figure 1). PSG and HRP scorings were done separately and the technicians were masked to any identifying information about patients and to any results of the previous arm (PSG or HRP). Once the first test was begun, the second test was scheduled for within the next 3 days.
Figure 1. General flow-chart for patients during the study; 359 patients completed both branches of the study and 348 patients had a valid PSG and HRP. HRP = home respiratory polygraphy; PSG = polysomnography; SAHS = sleep apnea–hypopnea syndrome.
[More] [Minimize]Our HRP (BreastSC20; Breast Medical AB, Mölnlycke, Sweden) measurements included oxygen saturation with an average time of four seconds (model 8000 J; Nonin Medical, Plymouth, MN); airflow through a nasal cannula; and thoracic and abdominal movements measured by piezoelectric bands (Pro-Tech reference 1295; Respironics, Pittsburgh, PA), which also included a sensor to measure body position.
All patients were instructed on home use of the HRP device by a technician in the hospital setting before randomization. HRP instruments were moved from home to home by trained personnel from CPAP service companies in each hospital area, acting as transport companies. No additional assistance was provided by the transport services to help the patients set up the HRP devices. The raw data files were telematically transmitted from home to the hospital (7). The hospital technician scored the raw data following a manual scoring protocol.
We used the American Academy of Sleep Medicine (AASM) 2007 recommendations (6) regarding configuration, filters, and sample signal rates. The neurologic variables were electroencephalogram, electrooculogram, and electromyogram (on the chin and both legs). Flow tracing was provided by a nasal cannula and thoracoabdominal motion by piezoelectric bands. Oxygen saturation was measured with a finger-pulse oximeter (average time among centers was between 2 and 4 s). Electrocardiogram and body position were also measured. The polysomnographic studies were analyzed manually at each participating center, according to the Rechtschaffen and Kales and the American Sleep Disorders Association 1992 criteria (30, 31) for sleep periods and arousals and according to the Spanish Sleep Network rule for respiratory scoring (see below).
A valid PSG or HRP had at least 3 recorded hours. In addition, a valid HRP had at least 3 hours of flow or band and oxymetry measurements for scoring. An invalid recording could be repeated up to two times.
For the PSG, an apnea was defined as the absence of airflow (≥ 90% reduction) for greater than or equal to 10 seconds and a hypopnea as a discernible airflow or band reduction (≥ 30% and <90%) of at least 10-second duration with a greater than or equal to 3% drop in oxygen saturation or final arousal (12). For the HRP, apnea and hypopnea were defined in the same way, but without the final arousal criteria for hypopnea. The number of apneas and hypopneas was divided by valid recording time for HRP and sleep time for PSG to determine the AHI.
Although the real therapeutic decision based on PSG results was performed when each patient finished the two arms of the protocol, at the end of the study (when every patient finished the protocol), we performed a masked post hoc analysis to ascertain the agreement in therapeutic decision-making between PSG and HRP. No information about real therapeutic decision was taken into account for this post hoc analysis.
Therapeutic decisions (CPAP, no CPAP, or impossible decision) for HRP and PSG were made by a researcher in each center based on the same set of information from each patient collected at baseline using questionnaires and direct measurements: age, sex, body mass index, neck circumference, systolic and diastolic blood pressure, comorbidities, job, alcohol intake, and tobacco consumption; subjective nocturnal sleep time and napping time on working days and holidays were ascertained based on the previous 4 weeks; episodes of subjective asphyxia, nocturia, morning headache, morning tiredness, and sleepiness while driving (5) were collected in four degrees of intensity (never, sometimes, frequently, and always) based on the previous 4 weeks; Epworth Sleepiness Scale and the American Sleep Disorders Association (32) sleepiness scale measured based on the previous 4 weeks; PSG (recording time, sleep time, sleep periods, AHI [total sleep time and sleep time in supine position], arousal and desaturation indexes, and time with SatO2 <90%); and HRP (recording and valid recording time, AHI [total valid recording time and valid time in supine position], desaturation index, and time with SatO2 <90%). Patients and diagnostic methods (HRP and PSG) were assessed in a random order using an electronic database located on a website. When the researcher chose one of the three options (CPAP, no CPAP, or impossible decision) data from another patient were presented at random to repeat the process. Each patient was presented twice (with PSG or HRP information) at random and nonconsecutively (Figure 2). Participant identification numbers for patients and other data were hidden. After a month, the same therapeutic decision procedure was repeated.
Figure 2. Patient information for therapeutic decision-making. (A) With PSG information. (B) With HRP information. ASDA = American Sleep Disorders Association; BMI = body mass index; CPAP = continuous positive airway pressure; HRP = home respiratory polygraphy; PSG = polysomnogram.
[More] [Minimize]The criteria for recommending CPAP were an AHI greater than or equal to 30 or an AHI between 5 and 30 with significant symptoms or consequences, according to the Spanish Sleep Network guidelines (12) (Figure 3). We also performed a simulation based on AASM criteria (CPAP would be recommended for both PSG and HRP if the AHI score was ≥ 15 or between 5 and 15 with significant symptoms or consequences) (6).
Figure 3. Treatment algorithm from Spanish Sleep Network guidelines. Each patient must have an individualized treatment assessment, especially the elderly and women. CPAP treatment is definitive after 3 months of adequate monitoring. Adapted from reference 12. AHI = apnea–hypopnea index; CPAP = continuous positive airway pressure; ESS = Epworth Sleepiness Scale; OA = oral appliance; RDI = respiratory disturbance index; SAHS = sleep apnea–hypopnea syndrome.
[More] [Minimize]We used Bland-Altman plots to assess the agreement between AHI measurements produced by PSG and HRP.
To evaluate the validity of the therapeutic decision between HRP and PSG we used (1) sensitivity and specificity; (2) negative (1-sensitivity/specificity) and positive (sensitivity/1-specificity) likelihood ratios (LR); (3) calculated posttest probability of obtaining a true positive when the test was positive (probability of recommending CPAP in agreement with PSG) or negative (probability of recommending CPAP in disagreement with PSG), based on the pretest probability of recommending CPAP (percentage of the total cases with CPAP recommendation based on PSG) and positive and negative LRs (28); and (4) agreement level (percentage of true-positives plus true-negatives).
We used the kappa test to assess baseline reliability between the therapeutic decisions before and after a month of HRP or PSG. We performed sensitivity, specificity, LRs, posttest probability, and agreement level analysis with the two therapeutic decisions (before and after a month) based on PSG results. These were considered the “Reference” values because they express the variability of the gold standard (PSG) and the “ideal” result for HRP.
Finally, we performed a logistic regression to determine potential variables (contained in the therapeutic decision, discussed previously and in Table 1) related to the probability of obtaining an HRP AHI greater than or equal to 30 (factor). We included variables with statistical significance (P < 0.05) in the model from the univariate analysis.
| Characteristic | N = 348 |
| Male, % | 75.6 |
| Age, yr, mean (SD) | 48.7 (11.8) |
| White race, % | 100 |
| Body mass index, mean (SD) | 31 (6.6) |
| Obesity, % | 48.3 |
| Job with accident risk, % | 10.8 |
| Alcohol, g, mean (SD) | 8.9 (22.3) |
| Smokers, % | 23.9 |
| Subjective sleep time per day, hr, mean (SD) | 6.9 (1.4) |
| Subjective nap time, hr, mean (SD) | 0.5 (0.6) |
| Epworth Sleepiness Scale, mean (SD) | 11.6 (5) |
| ASDA scale, % | |
| Absence | 13.5 |
| Mild | 28.8 |
| Moderate | 43.3 |
| Severe | 14.4 |
| Sleepiness while driving, % | |
| Never | 63.1 |
| Sometimes | 24 |
| Frequently | 12.5 |
| Always | 1.3 |
| Kilometers driven, yr, mean (SD) | 17,603 (30,861) |
| Traffic accidents, 2 yr, mean (SD)* | 0.09 (0.41) |
| Snoring, % | |
| Never | 4.8 |
| Sometimes | 7.4 |
| Frequently | 21.8 |
| Always | 66 |
| Observed apneas, % | |
| Never | 25 |
| Sometimes | 30.1 |
| Frequently | 20.5 |
| Always | 24.4 |
| Subjective asphyxia episodes, % | |
| Never | 45.2 |
| Sometimes | 37.2 |
| Frequently | 13.1 |
| Always | 4.5 |
| Nocturia, % | |
| Never | 30.8 |
| Sometimes | 31.1 |
| Frequently | 16.3 |
| Always | 21.8 |
| Morning headache, % | |
| Never | 47.4 |
| Sometimes | 31.7 |
| Frequently | 12.5 |
| Always | 8.3 |
| Morning tiredness, % | |
| Never | 14.4 |
| Sometimes | 24.4 |
| Frequently | 27.2 |
| Always | 34.3 |
| Depression or anxiety, % | 23.3 |
| Hypertension, % | 30.7 |
| Hypertensive drugs, mean (SD)† | 1.38 (1.11) |
| Cardiac disease, % | 3.7 |
| Cerebrovascular disease | 1.9 |
| Systolic pressure, mm Hg mean (SD) | 130.5 (17.2) |
| Diastolic pressure, mm Hg mean (SD) | 76 (12.1) |
Initially, 377 patients were selected, of whom 11 were excluded (Figure 1). Of the 366 randomized patients, 18 (4.9%) could not produce a valid HRP or PSG and 11 (3%) patients could not produce a valid HRP after repetitions (once in 51 patients and a second time in 34 patients). Eight of the 11 without a valid HRP after repetitions were caused by a bad signal or patient difficulties in setting up the device and three because of deficient telematic procedure. The clinical and anthropometric characteristics of the 348 patients with valid PSG and HRP results are shown in Table 1.
Figure 4 shows Bland and Altman plots. As expected, lower AHI values showed better agreement than intermediate and high values. Table 2 shows data from sleep studies. AHI from HRP were lower than those from PSG, in part because of higher valid recording times from HRP than sleep times from PSG.
Figure 4. Mean AHI versus the difference in AHI between PSG and HRP (Bland-Altman plot). Central lines represent mean values and upper and lower lines represent agreement limits (two standard deviations). AHI = apnea–hypopnea index; HRP = home respiratory polygraphy; PSG = polysomnography.
[More] [Minimize]| Study | PSG Mean (SD) | HRP Mean (SD) |
| Recording time, min | 443 (43) | 428 (82) |
| Sleep time or valid time, min | 372 (64) | 404 (71) |
| Non-REM stages 1 and 2, % | 64 (22) | — |
| Non-REM stages 3 and 4, % | 17 (12) | — |
| REM stage, % | 17 (13) | — |
| Arousal index | 39 (25) | — |
| AHI | 38 (29) | 31 (24) |
| AHI ≥ 5, % | 90 | 90 |
| AHI ≥ 15, % | 75 | 70 |
| AHI ≥ 30, % | 53 | 41 |
| AHI supine | 45 (33) | 38 (28) |
| SaO2 <90% of sleep or recording time | 9.9 (19) | 6.5 (14) |
| Desaturation index | 27 (29) | 29 (24) |
The “impossible decision” scenario was not observed with either PSG or HRP.
Therapeutic decisions between the initial and second evaluations (after a month) showed similar Kappa values for PSG (0.776; P < 0.001) and HRP (0.789; P < 0.001), expressing high reliability in both.
Therapeutic decisions for HRP compared with PSG had a sensitivity of 73%, a specificity of 77%, a negative LR of 0.32, a positive LR of 3.53, and an agreement level of 76% (Table 3). Therefore, the probability of recommending CPAP in agreement with PSG remained low, increasing from 67% (pretest) to 88% (posttest), and the probability of recommending CPAP in disagreement with PSG remained high, decreasing from 67% to 40%. These values are far from the Reference values.
| Se | Sp | LR+ (95% CI) | Posttest prob. + (95% CI) | LR − (95% CI) | Posttest prob. − (95% CI) | AL (%) | |
| Total sample (n = 348; pretest probability 67%) | |||||||
| HRP | 73 | 77 | 3.53 (2.45–5.07) | 88 (84–91) | 0.32 (0.25–0.41) | 40 (34–46) | 76 |
| Ref | 93 | 89 | 8.74 (5.11–15) | 95 (91–97) | 0.08 (0.05–0.13) | 14 (9–21) | 92 |
| HRP AHI ≥ 30 (n = 144; pretest probability 94%) | |||||||
| HRP | 94 | 44 | 1.69 (0.94–3.04) | 96 (93–98) | 0.13 (0.05–0.36) | 66 (43–84) | 91 |
| Ref | 98 | 78 | 4.40 (1.30–15) | 99 (95–100) | 0.03 (0.01–0.09) | 31 (13–58) | 97 |
| HRP AHI between 5 and 30 (n = 171; pretest probability 56%) | |||||||
| HRP | 52 | 76 | 2.18 (1.39–3.41) | 73 (64–81) | 0.63 (0.50–0.81) | 44 (39–50) | 59 |
| Ref | 86 | 88 | 7.29 (3.93–14) | 90 (83–95) | 0.16 (0.09–0.26) | 17 (10–25) | 87 |
| Total sample AASM simulation (n = 348; pretest probability 77%) | |||||||
| HRP | 87 | 78 | 3.91 (2.59–5.89) | 93 (90–95) | 0.17 (0.12–0.23) | 36 (28–43) | 85 |
| Ref | 98 | 94 | 17 (6.72–45) | 98 (96–99) | 0.02 (0.01–0.05) | 7 (4–16) | 98 |
Figure 5 shows the evolution of the agreement level in therapeutic decisions between HRP and PSG and the Reference values by selecting populations with increasing severity, based on HRP AHI scores.
Figure 5. Evolution of the agreement level (sum of true-positives and true-negatives) in therapeutic decisions made between HRP and polysomnography and the Reference selecting patients by increasing severity according to HRP AHI. AHI = apnea–hypopnea index; HRP = home respiratory polygraphy; Reference = agreement between before and after a month by the polysomnography.
[More] [Minimize]Patients with a higher HRP AHI (≥ 30) had acceptable agreement level (91%) but only in a limited sample (41%) of all patients available for diagnosis by HRP (46% of patients with HRP AHI ≥ 5). In this population (Table 3), therapeutic decisions based on HRP results, compared with those based on PSG results, had a sensitivity of 94%, a specificity of 44%, a negative LR of 0.13, and a positive LR of 1.69. Accordingly, the probability of recommending CPAP in agreement with PSG remained high (96%) because the pretest probability was also high (94%) and the probability of a CPAP recommendation in disagreement with PSG decreased from 94% to 66%. These values are close to the Reference values and they can be considered acceptable for making therapeutic decisions based on HRP, mainly because most patients with an HRP AHI greater than or equal to 30 should also need CPAP based on PSG results.
In the population with an HRP AHI between 5 and 30, a therapeutic decision based on HRP compared with PSG had a sensitivity of 52%, a specificity of 76%, a negative LR of 0.63, a positive LR of 2.18, and an agreement level of 59% (Table 3). Therefore, the probability of recommending CPAP in agreement with PSG remained low, increasing from 56% (pretest) to 73% (posttest) and the probability of recommending CPAP in disagreement with PSG remained high, decreasing from 56% to 44%. These values are far from the Reference values.
The results, based on the simulation with AASM criteria, were as follows: therapeutic decisions based on HRP, compared with PSG, had a sensitivity of 87%, a specificity of 78%, a negative LR of 0.17, and a positive LR of 3.91 (Table 3). Consequently, the probability of recommending CPAP in agreement with PSG increased from 77% (pretest) to 93%, and the probability of recommending CPAP in disagreement with PSG decreased from 77% to 36%. These values were better than those from the total sample, following the Spanish guidelines for recommending CPAP, but worse than those with HRP AHI greater than or equal to 30 and far from the Reference values.
Selecting patients with obesity, age greater than 55 years, hypertension, and a greater frequency of observed apneas increased the probability of obtaining an HRP AHI greater than or equal to 30 (Table 4).
| Variable | Odd Ratio (95% CI) | P Value |
| Age* | 0.009 | |
| Age 2nd tercile | 1.20 (0.63–2.29) | 0.576 |
| Age 3rd tercile | 2.71 (1.35–5.45) | 0.005 |
| BMI† | 0.000 | |
| BMI 2nd group | 1.54 (0.63–3.75) | 0.341 |
| BMI 3rd group | 4.39 (1.87–10.33) | 0.001 |
| Hypertension | 2.09 (1.15–3.80) | 0.015 |
| Observed apneas‡ | 2.05 (1.20–3.52) | 0.009 |
| Nocturia‡ | 1.54 (0.90–2.63) | 0.117 |
| Snoring‡ | 1.46 (0.54–3.93) | 0.460 |
To our knowledge, this study has the largest sample of all available research on portable monitoring in patients with intermediate to high SAHS suspicion. The principal results are (1) HRP shows differences from the PSG in supporting therapeutic decision-making; (2) patients with an HRP AHI greater than 30 could be recommended for CPAP in agreement with PSG; and (3) differences can be reduced by selecting patients who are obese, over 55 years, hypertensive, and with a greater frequency of observed apneas.
Several studies have demonstrated good diagnostic efficacy with HRP, compared with PSG (8–10, 20–27). In the same group of patients, HRP produced a good diagnostic cost-effectiveness relationship (7) compared with PSG. However, therapeutic decision-making is a different process than diagnosis, which also considers different AHI ranges (≥ 5, ≥ 10, and ≥ 15 for diagnosis; groups between 5 and 30 [or 5 and 15] and ≥ 30 [or ≥ 15] for therapeutic decisions). Studies of the diagnostic efficacy of HRP (7–10, 20–27) showed less variability between HRP and PSG for lower values of AHI (≤ 15) and more variability for intermediate and higher values (≥ 15) (Figure 4). Because the clinical information for making therapeutic decisions in the present study was the same for both HRP and PSG, the differences in AHI can primarily explain the differences between the two therapeutic decision procedures, although some portion could be explained by variability in the interpretation of clinical variables (see below).
A study has been done of therapeutic decision-making with a similar crossover design, patient selection, SAHS prevalence and severity (PSG AHI = 34 ± 25), and a sample size of 89 patients (9). Their agreement level (sum of true-positives and true-negatives) was higher than ours (89%). In their study, CPAP treatment was considered for patients with an AHI greater than or equal to 10 instead of greater than or equal to five, as in our case. However, this aspect of the study does not seem important, because the agreement in our data improves with an HRP AHI greater than 15 (Figure 5). Other differences should be taken into account: (1) the HRP devices were different, although agreement as measured by Bland and Altman plots was similar; (2) our sample size was four times higher, which increased the statistical power of our study; and (3) unlike the present study, therapeutic decision-making in their study was not blinded nor did they use a multicentric approach with standardized variables for therapeutic decision-making. These methodologic differences could explain our dissimilar results.
Another study assessed a subgroup of 58 patients with AHI scores from 10–30 with a similar crossover design and patient selection (26). The results were similar to ours (positive LR = 1.77 [95% confidence interval, 1.08–2.90] and negative LR = 0.43 [95% confidence interval, 0.22–0.86]), reinforcing the data presented here.
As expected, in the AASM simulation the HRP showed better results than using Spanish, and probably other similar, guidelines (14, 15, 17), mainly because of avoiding the variability in interpreting the relevancy of clinical symptoms and consequences in the population with HRP AHI scores between 15 and 30, and because this analysis was a simulation, all patients with HRP AHI greater than 15 (i.e., 15.3) were mathematically recommended for CPAP treatment. However, during the study with the Spanish protocol, some patients with AHI values slightly above 30 (e.g., 31) and moderate clinical symptoms or comorbidities could be recommended against CPAP treatment. Nevertheless, the AASM simulation results remain slightly less than acceptable and clearly worse than the reference values, caused by AHI differences between PSG and HRP.
For intermediate AHI levels, most guidelines recommend CPAP treatment for more symptomatic or comorbid patients. However, this implies variability in the assessment of clinical symptoms and comorbidities and, consequently, variability in CPAP indications. Therefore, an interesting issue is to determine the role of clinical variables in the disagreement in therapeutic decision-making between PSG and HRP. In the present study, we can infer that the disagreement between therapeutic decisions based on HRP before and after a month is caused by the variability in the interpretation of clinical variables. In the population with HRP AHI from 5 to 30, this variability was 13% of a total of 41% (Table 3), thus 28% may be caused by AHI differences between PSG and HRP. In the population with HRP AHI greater than or equal to 30, variability caused by clinical variables was logically lower (5%) and 4% caused by AHI differences between PSG and HRP. In summary, in the population with HRP AHI between 5 and 30, most of the disagreement between PSG and HRP can be caused by the differences in AHI scores.
We used the Spanish (and AASM) recommendation for both therapeutic decisions, PSG and HRP. Because HRP underestimates the AHI compared with PSG, an interesting question is whether HRP AHI thresholds for CPAP decision-making should be different than those used for PSG AHI. Our present study cannot provide more information but it is an interesting subject for futures studies.
Our results may be dependent on the selection of patients. We selected patients similarly to the way they are chosen in clinical practice in our consortium (12) and similarly to most diagnostic studies of HRP versus PSG, that is, patients with an intermediate or high probability of SAHS. Prevalence and PSG AHI in our study were also similar to that seen in most of the mentioned studies (8–10, 20–27, 33). Nevertheless, our results suggest that selecting patients with a greater probability of SAHS (i.e., obesity and age ≥ 55 yr) is necessary for improving agreement. Indeed, recent prospective studies have confirmed this (34–37), although according to our data, selecting patients with a greater probability of SAHS would lead to inclusion of 40% of all susceptible patients for diagnosis using HRP (see below).
As was our case, HRP scoring normally produces lower AHI values than PSG. This is because the calculation of AHI by HRP is based on recording time rather than sleep time, as is the case with PSG, and some of the following factors: (1) days of testing and environments (shorter home sleep time in a supine position [23, 24, 27] and better home sleep quality and stability, leading to less facility for producing obstructive events [38]); (2) instruments (different brands, models, and software for displaying signals from the nasal cannula, thoracic and abdominal bands, and oximeters); and (3) methodologies (intercenter variability in identifying respiratory events and the inclusion of arousal in the PSG hypopnea definition), although clear differences have not been found in studies including arousal (7, 9, 24, 25) compared with others without arousal (8, 20, 21, 23, 27).
Because PSG is performed in unnatural environments and conditions, there are doubts as to whether PSG is the gold standard. Four randomized control studies have assessed the results of CPAP treatment of 12 weeks (34, 35, 37) and of 6 weeks (36) after SAHS diagnosis using PSG or a home portable monitor in selected patients. Both protocols (ambulatory and in-hospital) showed similar improvements in AHI (34), quality of life, clinical symptoms, and adherence to CPAP treatment (34–37). In the aforementioned studies, patients were clinically selected to receive CPAP treatment if SAHS was confirmed. However, in our study, we selected patients with lower clinical severity to first perform a SAHS diagnosis and then, in positive cases, the therapeutic decision (CPAP or other treatment). Consequently, our patients had lower age, lower body mass index, and less indication for CPAP treatment (most vs. 57% in the present study). In clinical practice (based on different guidelines) many less symptomatic patients than those included in the referenced studies need to be considered for diagnosing and making therapeutic decisions (60% of patients in our data). If this large group of patients requires PSG for diagnosis and therapeutic decision-making, the approach used in the referred studies will not be cost-effective. Therefore, to determine if HRP is a viable alternative to PSG for diagnosis and therapeutic decision-making, studies including patients with intermediate or high SAHS suspicion, treated and untreated with CPAP, with a set of cardiovascular variables and cost analysis are needed. Nevertheless, the contribution of our study is that these new studies should be focused on the patient population with an HRP AHI between 5 and 30, because there is already acceptable agreement for patients with AHI greater than or equal to 30.
In conclusion, taking the PSG as the gold standard, we recommend CPAP treatment in patients with an HRP AHI greater than 30. Selecting patients with high SAHS probabilities seems to be needed until prospective cost-effectiveness studies in patients with intermediate and high SAHS suspicion become available.
The authors thank Verónica Rodríguez and Vanessa Iglesias for their assistance in the translation of the manuscript and Asunción Martín, Elena Sandoval, Soledad Guillen, Trinidad Amigo, Pablo Mejias, and Carmen Lorenzana for technical assistance in the sleep laboratory.
Spanish Sleep Network: Estefania Garcia-Ledesma, San Pedro de Alcántara Hospital, Cáceres, Spain; Manuela Rubio, San Pedro de Alcántara Hospital, Cáceres, Spain; Laura Cacelo, Txagoritxu Hospital, Vitoria, Spain; Rosario Carpizo, Valdecilla Hospital, Santander, Spain; Lirios Sacristan, General Universitario Hospital, Alicante, Spain; Neus Salord, Belvitge Hospital, Barcelona, Spain; Miguel Carrera, Son Dureta Hospital, Palma de Mallorca, Spain; José N. Sancho-Chust, San Juan Hospital, Alicante, Spain; and Cristina Embid, Clinic Hospital, Barcelona, Spain.
| 1. | Durán J, Esnaola S, Rubio R, Iztueta A. Obstructive sleep apnea-hypopnea and related clinical features in a population-based sample of subjects aged 30 to 70 yr. Am J Respir Crit Care Med 2001;163:685–689. |
| 2. | Bazzano LA, Khan Z, Reynolds K, He J. Effect of nocturnal nasal continuous positive airway pressure on blood pressure in obstructive sleep apnea. Hypertension 2007;50:417–423. |
| 3. | Marin JM, Carrizo SJ, Vicente E, Agusti AG. Long-term cardiovascular outcomes in men with obstructive sleep apnoea-hypopnoea with or without treatment with continuous positive airway pressure: an observational study. Lancet 2005;365:1046–1053. |
| 4. | Terán-Santos J, Jiménez-Gómez A, Cordero-Guevara J. The association between sleep apnea and the risk of traffic accidents. Cooperative Group Burgos-Santander. N Engl J Med 1999;340:847–851. |
| 5. | Masa JF, Rubio M, Findley LJ. Habitually sleepy drivers have a high frequency of automobile crashes associated with respiratory disorders during sleep. Am J Respir Crit Care Med 2000;162:1407–1412. |
| 6. | Iber C, Ancoli-Israel S, Chesson A, Quan S. The AASM manual for the scoring of sleep and associated events: rules, terminology and technical specifications, 1st ed. Westchester, IL: American Academy of Sleep Medicine; 2007. |
| 7. | Masa JF, Corral C, Pereira R, Duran-Cantolla J, Cabello M, Hernández-Blasco L, Monasterio C, Alonso A, Chiner E, Rubio M, et al.. Effectiveness of home respiratory polygraphy for the diagnosis of sleep apnea and hypopnea syndrome. Thorax 2011;66:567–573. |
| 8. | Whittle AT, Finch SP, Mortimore IL, MacKay TW, Douglas NJ. Use of home sleep studies for diagnosis of the sleep apnoea/hypopnoea syndrome. Thorax 1997;52:1068–1073. |
| 9. | Parra O, García-Esclasans N, Montserrat JM, García Eroles L, Ruíz J, López JA, Guerra JM, Sopeña JJ. Should patients with sleep apnoea/hypopnoea syndrome be diagnosed and managed on the basis of home sleep studies? Eur Respir J 1997;10:1720–1724. |
| 10. | Alonso Alvarez M de L, Terán Santos J, Cordero Guevara J, González Martínez M, Rodríguez Pascual L, Viejo Bañuelos JL, Marañón Cabello A. Reliability of home respiratory polygraphy for the diagnosis of sleep apnea-hypopnea syndrome: analysis of costs. Arch Bronconeumol 2008;44:22–28. |
| 11. | Ferber R, Millman R, Coppola M, Fleetham J, Murray CF, Iber C, McCall V, Nino-Murcia G, Pressman M, Sanders M. Portable recording in the assessment of obstructive sleep apnea. ASDA standards of practice. Sleep 1994;17:378–392. |
| 12. | Grupo Español de Sueño (GES). Consenso Nacional sobre el síndrome de apneas-hipopneas del sueño. Arch Bronconeumol 2005;41(Suppl. 4):3–110. |
| 13. | Epstein LJ, Kristo D, Strollo PJ, Friedman N, Malhotra A, Patil SP, Ramar K, Rogers R, Schwab RJ, Weaver EM, et al.. Adult Obstructive Sleep Apnea Task Force of the American Academy of Sleep Medicine. Clinical guideline for the evaluation, management and long-term care of obstructive sleep apnea in adults. J Clin Sleep Med 2009;5:263–276. |
| 14. | Fleetham J, Ayas N, Bradley D, Ferguson K, Fitzpatrick M, George C, Hanly P, Hill F, Kimoff J, Kryger M, et al..; CTS Sleep Disordered Breathing Committee. Canadian Thoracic Society guidelines: diagnosis and treatment of sleep disordered breathing in adults. Can Respir J 2006;13:387–392. |
| 15. | Sala H, Nigro C, Rabec C, Guardia AS, Smurra M. Consenso argentino de trastornos respiratorios vinculados al sueño. Medicina (B Aires) 2001;61:351–362. |
| 16. | Effectiveness of nasal continuous positive airway pressure (nCPAP) in obstructive sleep apnoea in adults. National Health and Medical Research Council. Canberra. [accessed January, 2011] Available from: http://www.nhmrc.health.gov.au/. |
| 17. | Laitinen LA, Anttalainen U, Pietinalho A, Hämäläinen P, Koskela K; Expert Advisory Group Listed in Foreword. Sleep apnoea: Finnish national guidelines for prevention and treatment 2002–2012. Respir Med 2003;97:337–365. |
| 18. | Kushida CA, Littner MR, Hirshkowitz M, Morgenthaler TI, Alessi CA, Bailey D, Boehlecke B, Brown TM, Coleman J, Friedman L, et al..; American Academy of Sleep Medicine. Practice parameters for the use of continuous and bilevel positive airway pressure devices to treat adult patients with sleep-related breathing disorders. Sleep 2006;29:375–380. |
| 19. | Convention relative au diagnostic et au traitement du syndrome des apnées obstructives du sommeil, conclue entre le comité de l’assurance du service des soins de santé de l’institut national d’assurance maladie-invalidité et «pouvoir_organ» de «ziekenhuis» «campus» a «localiteit» dans le cadre duquel fonctionne l’établissement visé par la présente convention. Institut National D’assurance Maladie-Invalidité. Service des Soins de Santé. Bruxelles. |
| 20. | Dingli K, Coleman EL, Vennelle M, Finch SP, Wraith PK, Mackay TW, Douglas NJ. Evaluation of a portable device for diagnosing the sleep apnoea/hypopnoea syndrome. Eur Respir J 2003;21:253–259. |
| 21. | Reichert JA, Bloch DA, Cundiff E, Votteri BA. Comparison of the NovaSom QSG, a new sleep apnea home-diagnostic system, and polysomnography. Sleep Med 2003;4:213–218. |
| 22. | Redline S, Tosteson T, Boucher MA, Millman RP. Measurement of sleep-related breathing disturbances in epidemiologic studies. Assessment of the validity and reproducibility of a portable monitoring device. Chest 1991;100:1281–1286. |
| 23. | Yin M, Miyazaki S, Ishikawa K. Evaluation of type 3 portable monitoring in unattended home setting for suspected sleep apnea: factors that may affect its accuracy. Otolaryngol Head Neck Surg 2006;134:204–209. |
| 24. | García-Díaz E, Quintana-Gallego E, Ruiz A, Carmona-Bernal C, Sánchez-Armengol A, Botebol-Benhamou G, Capote F. Respiratory polygraphy with actigraphy in the diagnosis of sleep apnea-hypopnea syndrome. Chest 2007;131:725–732. |
| 25. | Santos-Silva R, Sartori DE, Truksinas V, Truksinas E, Alonso FF, Tufik S, Bittencourt LR. Validation of a portable monitoring system for the diagnosis of obstructive sleep apnea syndrome. Sleep 2009;32:629–636. |
| 26. | White DP, Gibb TJ, Wall JM, Westbrook PR. Assessment of accuracy and analysis time of a novel device to monitor sleep and breathing in the home. Sleep 1995;18:115–126. |
| 27. | Ayappa I, Norman RG, Seelall V, Rapoport DM. Validation of a self-applied unattended monitor for sleep disordered breathing. J Clin Sleep Med 2008;4:26–37. |
| 28. | Flemons WW, Littner MR, Rowley JA, Gay P, Anderson WM, Hudgel DW, McEvoy RD, Loube DI. Home diagnosis of sleep apnea: a systematic review of the literature. An evidence review cosponsored by the American Academy of Sleep Medicine, the American College of Chest Physicians, and the American Thoracic Society. Chest 2003;124:1543–1579. |
| 29. | Masa JF, Corral J, Pereira R, Durán-Cantolla J, Cabello Nájera M, Hernández-Blasco L, Monasterio C, Alonso-Fernández A, Chiner E, Zamorano J, et al.. Therapeutic decision for sleep apnea and hypopnea syndrome by home respiratory polygraphy. Am J Respir Crit Care Med 2011;183:A6326. |
| 30. | Rechtschaffen A, Kales A, editors. A manual of standardized terminology and scoring system for sleep stages of human subjects. Los Angeles: Brain Information Service/Brain Research Institute, University of California at Los Angeles; 1968. |
| 31. | Sleep Disorders Atlas Task Force of the American Sleep Disorders Association. EEG arousals: scoring rules and examples. Sleep 1992;15:174–184. |
| 32. | The international classification of sleep disorders: diagnostic and coding manual. Rochester, MN: Diagnostic Classification Steering Committee, American Sleep Disorders Association; 1990. |
| 33. | Su S, Baroody FM, Kohrman M, Suskind D. A comparison of polysomnography and a portable home sleep study in the diagnosis of obstructive sleep apnea syndrome. Otolaryngol Head Neck Surg 2004;131:844–850. |
| 34. | Mulgrew AT, Fox N, Ayas NT, Ryan CF. Diagnosis and initial management of obstructive sleep apnea without polysomnography: a randomized validation study. Ann Intern Med 2007;146:157–166. |
| 35. | Antic NA, Buchan C, Esterman A, Hensley M, Naughton MT, Rowland S, Williamson B, Windler S, Eckermann S, McEvoy RD. A randomized controlled trial of nurse-led care for symptomatic moderate-severe obstructive sleep apnea. Am J Respir Crit Care Med 2009;179:501–508. |
| 36. | Berry RB, Hill G, Thompson L, McLaurin V. Portable monitoring and autotitration versus polysomnography for the diagnosis and treatment of sleep apnea. Sleep 2008;31:1423–1431. |
| 37. | Kuna ST, Gurubhagavatula I, Maislin G, Hin S, Hartwig KC, McCloskey S, Hachadoorian R, Hurley S, Gupta R, Staley B, et al.. Non-inferiority of functional outcome in ambulatory management of obstructive sleep apnea. Am J Respir Crit Care Med 2011;183:1238–1244. |
| 38. | Stepnowsky CJ, Orr WC, Davidson TM. Nightly variability of sleep-disordered breathing measured over 3 nights. Otolaryngol Head Neck Surg 2004;131:837–843. |
* A complete list of members may be found before the beginning of the References.
Supported by Instituto de Salud Carlos III (Fondo de Investigaciones Sanitarias, Ministerio de Sanidad y Consumo), Spanish Respiratory Society, Telefónica SA (Spain), Air Liquide (Spain), and Breas Medical (Spain).
Contributing authors: J.F.M., J.C., R.P., J.D.-C., M.C., L.H.-B., C.M., A.A., E.C., J.Z., F.A., and J.M.M.
Originally Published in Press as DOI: 10.1164/rccm.201103-0428OC on July 7, 2011
Author disclosures
