Paper Info
Title
Improving diagnostic ability of blood oxygen saturation from overnight pulse oximetry in obstructive sleep apnea detection by means of central tendency measure
Abstract
Results For a radius in the scatter plot equal to 1, CTM values corresponding to OSA positive patients (0.30 ± 0.20, mean ± S.D.) were significantly lower ( p ≪ 0.001) than those values from OSA negative subjects (0.71 ± 0.18, mean ± S.D.). CTM was significantly correlated with classical indexes and indexes from ApEn analysis. CTM provided the highest correlation with the apnea–hipopnea index AHI ( r = −0.74, p < 0.0001). Moreover, it reached the best results from the receiver operating characteristics (ROC) curve analysis, with 90.1% sensitivity, 82.9% specificity, 88.5% positive predictive value, 85.1% negative predictive value, 87.2% accuracy and an area under the ROC curve of 0.924. Finally, the AHI derived from the quadratic regression curve for the CTM showed better agreement with the AHI from PSG than classical and ApEn derived indexes. Conclusion The results suggest that CTM could improve the diagnostic ability of SaO 2 signals recorded from portable monitoring. CTM could be a useful tool for physicians in the diagnosis of OSA syndrome. Keywords Central tendency measure Nonlinear methods Blood oxygen saturation Oximetry Obstructive sleep apnea 1 Introduction The obstructive sleep apnea (OSA) syndrome is characterized by repetitive reduction or cessation of airflow due to partial or complete airway obstruction [1] . This disease is usually associated with hypoxemia, bradycardia, arousals and fragmented sleep [2] . Nowadays, OSA is the most common respiratory referral in many sleep centers [3] . The estimated OSA prevalence varies from 1 to 5% of adult men in western countries [4] . OSA is associated with conditions that are responsible for the most important causes of mortality in adults: hypertension and cardiovascular and cerebrovascular diseases. Several neurobehavioral morbidities, which are of potentially great public health and economic importance, are linked with OSA [4] . The major behavioral symptoms include excessive daytime sleepiness (EDS), neurocognitive deficits like impairments in concentration and memory, and psychological problems like depression or personality changes [5] . Individuals with OSA are dangerous drivers with an increased risk of being involved in road and work accidents [3] . The standard diagnostic test for OSA syndrome is overnight polysomnography (PSG) [6] , consisting in the recording of neurophysiological and cardiorespiratory signals subsequently used to analyze sleep and breathing. The apnea–hypopnea index (AHI) derived from the PSG is then used to diagnose the disease. Portable monitoring has been proposed as a substitute for PSG in the diagnostic assessment of patients with suspected sleep apnea [7] . Due to its noninvasive nature and simplicity, nocturnal pulse oximetry is widely used in many medicine areas to determine patient's blood oxygen saturation (SaO 2 ) and heart rate. The lack of airflow during apneic periods can lead to recurrent episodes of hypoxemia that can be detected on oximetry as fluctuations in SaO 2 records [8] . Several quantitative indexes derived from nocturnal oximetry have been developed to diagnose OSA. The most frequently used by physicians include oxygen desaturation indexes (ODIs), which measure the number of dips in the SaO 2 signal below a certain threshold [9–11] , and the cumulative time spent below a certain saturation level (CT) [12,13] . However, these indexes have significant limitations. In general, CT indexes did not achieve high diagnostic accuracies [13,14] . On the other hand, there is not a universally accepted definition for oxygen desaturation. Moreover, there is not a consensus on a threshold to diagnose OSA based on ODIs [14,15] . Furthermore, correlation between oximetry indexes and AHI is not high [13] . In previous studies [16–18] , our group has shown that nonlinear analysis could provide useful information in the diagnosis of OSA syndrome. A regularity measure from SaO 2 signals obtained applying approximate entropy (ApEn) improved the diagnostic accuracy of classical oximetric indexes [16] . ApEn was also applied to heart rate signals from nocturnal oximetry, obtaining promising results [17] . Moreover, additional nonlinear methods, central tendency measure (CTM) and Lempel–Ziv (LZ) complexity, were applied to SaO 2 records [18] . The results suggested that both CTM and LZ complexity could help physicians in screening for OSA syndrome. Particularly, a variability measure by means of the CTM provided the best diagnostic accuracy. The present study intended to go more deeply into the usefulness of the CTM to diagnose OSA. We assessed its advantages over classical oximetric indexes and other nonlinear methods: it is a simple parameter to estimate the signal variability with a low computational cost [19] . Furthermore, we studied the changes in the diagnostic accuracy when using different values of the input parameters. Variability measures of ECG allow to distinguish between normal and chronic heart failure subjects [19] . Some authors have applied nonlinear methods over different respiratory patterns to study sleep stages and the coordination between brain and lungs [20] or panic disorder [21] . In the present study, we applied CTM looking for differences in variability between OSA positive and OSA negative patients. Our study is aimed to estimate the overall variability of each overnight oximetric recording by means of CTM in order to assess its utility in OSA diagnosis. We assessed its usefulness to physicians in screening for OSA syndrome, comparing it with classical oximetric indexes and with ApEn analysis. We studied three different index groups: classical indexes, regularity indexes from ApEn analysis and variability indexes from CTM analysis. 2 Subjects and signals A total of 187 patients (147 males and 40 females) suspected of having OSA were studied. Patients have a mean ± standard deviation (S.D.) age of 57.97 ± 12.84 years and a body mass index (BMI) of 29.54 ± 5.51 kg/m 2 . All subjects presented daytime hypersomnolence, loud snoring, nocturnal choking and awakenings, or apneic events (or all four symptoms) reported by the subject or a bed mate. Sleep studies were carried out usually from midnight to 08:00 a.m. in the Sleep Unit of Hospital Clínico Universitario in Santiago de Compostela, Spain. The Review Board on Human Studies at this institution approved the protocol, and all subjects gave their informed consent to participate in the study. 2.1 Conventional polysomnography All patients underwent overnight PSG (Ultrasom Network, Nicolet, Madison, WI, USA) which included electroencephalogram (EEG), electrocardiogram (ECG), electrooculogram (EOG), chin electromyogram (EMG), measurement of chest wall movement and airflow measurement (three-port thermistor). The PSG register was analyzed over periods of 30 s during sleep phases I–IV and rapid eye movement, according to Rechtschaffen and Kales rules [22] . Apnea was defined as the cessation of airflow for more than 10 s and hypopnea as the reduction of respiratory flow for at least 10 s accompanied by a 4% or more decrease in the saturation of hemoglobin [23–26] . The average AHI was calculated for hourly periods of sleep. According to the American Academy of Sleep Medicine Task Force criteria [27] , an AHI greater than or equal to 10 events per hour (e/h) of sleep was considered as diagnosis of OSA. If the subject had less than 3 h of total sleep, the sleep study was repeated. After the PSG, a conventional spirometry study (Collins spirometer) was carried out. Chronic obstructive pulmonary disease (COPD) was defined as a disease state characterized by airflow limitation that is not fully reversible. The airflow limitation is usually both progressive and associated with an abnormal inflammatory response of the lungs to noxious particles or gases [28] . The spirometry showed that 42 patients had COPD (mean age of 62.26 ± 13.65 years and a BMI of 29.66 ± 17.31 kg/m 2 ). Moreover, 9 of the 42 patients with COPD (21.8%) presented respiratory failure. According to the global initiative for chronic obstructive lung disease (GOLD) consensus [28] , 22 (52.4%) of these subjects could be classified as mild COPD patients, 14 (33.3%) as moderate COPD patients and 6 (14.3%) as severe COPD patients. Table 1 summarizes the demographic and clinical features of the subjects under study, as well as the groups derived from the PSG diagnosis. The OSA positive group consisted of 111 patients (59.4%) diagnosed as OSA according to an AHI ≥ 10 e/h (40.07 ± 19.64 e/h), whereas the remaining 76 subjects (40.6%) made up the OSA negative group (2.04 ± 2.36 e/h). In the OSA positive group, there were men and women between 28 and 81 years (58.30 ± 12.88 years) and with BMI between 20.57 and 46.51 kg/m 2 (30.45 ± 4.92 kg/m 2 ). The OSA negative group consisted of subjects between 21 and 79 years (57.57 ± 12.87 years) and with BMI between 19.53 and 42.19 kg/m 2 (28.42 ± 6.02 kg/m 2 ). 2.2 Overnight pulse oximetry An overnight pulse oximetry analysis was carried out simultaneously to the conventional PSG study. Recording of SaO 2 was carried out using a Criticare 504 oximeter (CSI, Waukesha, WI, USA). SaO 2 and heart rate were both simultaneously recorded using a dual wavelength-based finger probe with a sampling frequency of 0.2 Hz (one sample every 5 s). In this study only the SaO 2 signals were used. Although new oximeters work at higher sampling frequencies, the study carried out by Warley et al. [29] showed that this sampling frequency provides reasonable resolution in SaO 2 variability. Moreover, oximetry signals recorded at higher sampling frequencies are subsequently averaged to obtain usually one sample every 12 s [7,14,30] . There is some underestimation of the peak SaO 2 in recovery post-apnea, but the signal shape and variability were preserved [29] . The SaO 2 signals were saved to separate files and processed off-line. Artifacts due to poor contact from the finger probe, patient movements or bad regional circulation, were removed by visual inspection of SaO 2 signals, discarding data showing drops to zero. Fig. 1 displays three common oximetric recordings. Fig. 1 (a) depicts a common OSA negative subject. Fig. 1 (b) shows a SaO 2 record with clearly marked desaturations, corresponding to an apparent OSA positive subject. Fig. 1 (c) illustrates more exactly the difficulty in the diagnosis of the disease. It shows the SaO 2 record for an uncertain OSA positive patient. In this case, dips in SaO 2 are not so extreme and the diagnosis by visual inspection is not evident. 3 Methods 3.1 Classical oximetry indexes Our oximeter provided the following indexes: oxygen desaturation indexes of 4% (ODI4), 3% (ODI3), 2% (ODI2) and cumulative time spent below a saturation of 90% (CT90). The number of falls in each SaO 2 record greater than or equal to 4, 3 and 2% were computed from baseline. Baseline was set initially as the mean level in the first 3 min of recording [31] . These oxygen desaturation indexes were computed per hour of recording. CT90 was calculated as the percentage of time during which the SaO 2 register was below 90%. 3.2 Approximate entropy ApEn is a family of statistics introduced as a quantification of regularity in sequences and time series data, initially motivated by applications to relatively short and noisy data sets [32] . ApEn evaluates both dominant and subordinant patterns in the data, and discriminates series for which clear feature recognition is difficult [32] . Several properties of ApEn make it highly suitable for biomedical time series analysis: ApEn is almost unaffected by low-level noise; it is robust to outliers, scale invariant and model independent; it is applicable to time series with at least 50 data points, with good reproducibility; it can be applied to discriminate general classes of correlated stochastic processes, as well as noisy deterministic systems, providing finite values for both stochastic and deterministic processes [32] . ApEn assigns a non-negative number to a time series, with larger values corresponding to greater randomness or irregularity in the data [32] . The algorithm applied to compute ApEn can be seen in detail in [16] and [17] . Briefly, ApEn measures the logarithmic likelihood that runs of patterns that are close (within r ) for m contiguous observations remain close (within the same tolerance r ) on subsequent incremental comparisons (pattern length m + 1) [32] . Two input parameters, m and r , must be fixed to compute ApEn( m , r ): m is the length of compared runs, and r is effectively a filter [33] . To ensure appropriate comparisons between data sets, all input parameters m , r and N must be the same for each data set [33,34] . No guidelines exist for optimizing the m and r values. However, Pincus suggested parameter values of m = 1 or m = 2 and with r a fixed value between 0.1 and 0.25 times the S.D. of the original time series [32] . Multiple previous studies have demonstrated that these input parameters produce good statistical reproducibility for ApEn for time series of length N ≥ 60 [32] . In previous studies by our own group [16,17] , we showed that ApEn was better estimated with m = 1 when applying it to SaO 2 signals. Thus, in the present study, we computed ApEn with m = 1 and r equal to 0.1, 0.15 and 0.20 times the S.D. of the original time series, obtaining ApEn01, ApEn015 and ApEn02, respectively. 3.3 Central tendency measure Quantifying the signal variability by means of CTM starts displaying a second-order difference plot. These kinds of scatter diagrams, given by Eq. (1) , are graphs centered in the origin used to assess the degree of chaos in a data set [19] . (1) [ x ( n + 2 ) − x ( n + 1 ) ]     versus     [ x ( n + 1 ) − x ( n ) ] The second-order difference plots are very useful in modeling biological systems such as hemodynamics and heart rate variability. With this approach, rather than defining a time series as chaotic or not chaotic, the degree of variability or chaos is evaluated [19] . Such scatter plots could be a useful tool for physicians, who could make a preliminary diagnosis by visual inspection of the diagrams. Fig. 2 displays the second-order difference plots for the SaO 2 signals depicted in Fig. 1 . The uncertain OSA positive subject whose SaO 2 record is displayed in Fig. 1 (c) could be initially misclassified as OSA negative by visual inspection. However, the corresponding scatter plot in Fig. 2 (c) shows a significantly high dispersion compared with the diagram for a common OSA negative subject, suggesting a different diagnosis. Thus, the second-order difference plots could help physicians to improve preliminary diagnoses. The CTM is used to quantify the signal variability from the second-order difference plots. The CTM is computed by selecting a circular region of radius ρ around the origin and counting the number of points that fall within the circle. This measure is subsequently normalized dividing by the total number of points. For a N point data series, N − 2 would be the total number of points in the scatter plot given by Eq. (1) . Then, the CTM can be computed as [35] (2) CTM = ∑ i = 1 N − 2 δ ( d i ) N − 2 where (3) δ ( d i ) = 1 if   [ ( x ( i + 2 ) − x ( i + 1 ) ) 2 + ( x ( i + 1 ) − x ( i ) ) 2 ] 1 / 2 < ρ 0 otherwise The radius ρ is selected depending on the character of the data. In the present study, we computed CTM with various radii to assess the behavior of the method in pulse oximetry analysis. We computed CTM with three different values of ρ . For radii equal to 1, 3 and 6 we obtained CTM1, CTM3 and CTM6. We compared these parameters with the classical oximetric indexes and with those indexes derived from the ApEn analysis. 3.4 Statistical analysis The Kolmogorov–Smirnov and Shapiro–Wilk tests were used to assess the normal distribution of the variables involved in the study. Homoscedasticity (homogeneity of variances) was also assessed by means of the Levene's test. The normal distribution and homoscedasticity could not be verified with all the variables under study. Thus, the nonparametric Mann–Whitney test was applied to look for significant differences between the OSA positive and the OSA negative groups. The Bonferroni correction was applied due to the large number of variables included in the study. SPSS 14 was used to perform the statistical analysis. The degree of association between each index and the AHI was studied using the Pearson correlation test. A p -value was also computed to measure the statistical significance of the results. A ROC curve analysis was performed to assess the diagnostic capacity of each method in screening for OSA syndrome. The following statistics were derived from this study: sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive likelihood ratio (LR+), negative likelihood ratio (LR−), accuracy and area under the ROC curve. Correlation and ROC curve analyses were used to select the best parameter in each group of indexes. Scatter diagrams were depicted to graphically study the relation between indexes under study and the AHI. In addition, the linear regression (degree n = 1) and the quadratic regression (degree n = 2) curves, which best fit the plotted points in a least squares sense, were drawn in the scatter diagrams. The polynomials, which defined the relationship between indexes and the AHI from PSG, were used to obtain a derived AHI. Finally, Bland and Altman plots were used to graphically measure the degree of agreement between the derived AHIs and the AHI from PSG. 4 Results SaO 2 signals from nocturnal oximetry were processed by means of ApEn and CTM. Classical oximetric indexes, ODI2, ODI3, ODI4 and CT90 were also included in this study. As well as ODIs are provided for three common desaturation thresholds (2, 3 and 4%), three different measures were computed for each nonlinear method varying their input parameters: ApEn01, ApEn015 and ApEn02 were derived from the ApEn and CTM1, CTM3 and CTM6 were derived from the CTM analysis. Table 2 shows the mean ± S.D. values for both the OSA positive and the OSA negative groups for every index. Furthermore, the p -value from the Mann–Whitney test is displayed. CTM1, CTM3 and indexes from the ApEn analysis provide the highest significant differences between groups, improving results from the classical oximetric indexes. Table 3 shows the correlation between all indexes and the AHI derived from the PSG. Although CT90 was statistically correlated with the AHI ( p < 0.001), it achieved the smallest Pearson correlation coefficient ( r = 0.280). ODIs achieved higher correlation with AHI than CT90, slightly improving correlation values between ApEn and AHI. We obtained r = 0.617 ( p < 0.0001) with ODI3, whereas an r = 0.606 ( p < 0.0001) was reached with ApEn02. CTM1 achieved the highest correlation with the AHI ( r = −0.7382, p < 0.0001). The negative correlation indexes obtained with CTM are due to the own nature of this method, which assigns small values to high variability and vice versa. Table 4 shows the results from the ROC curve analysis for each parameter. We can notice that nonlinear indexes achieved better diagnostic accuracy and area under the ROC curve (AROC) than classical indexes. All nonlinear measures provided sensitivity values over 83% and specificity values over 81%, leading to accuracy values of at least 84.5% and areas under the ROC curve over 0.90. On the other hand, classical oximetric indexes showed large differences between sensitivity and specificity values, with very poor sensitivities. Accuracies are around 75% and AROCs are below 0.80. Diagnostic test values from ApEn analysis do not vary very much changing the input parameters. However, CTM diagnostic accuracy slightly decreases when the radius is increased. The best results are provided by CTM1, with 90.1% sensitivity, 82.9% specificity, 88.5% positive predictive value, 85.1% negative predictive value, 87.2% accuracy and an area under the ROC curve of 0.924. Fig. 3 shows the ROC curves for the best parameter of each index group in terms of diagnostic accuracy and correlation with AHI: ODI3 from the classical index group, ApEn02 from the ApEn index group and CTM1 from the CTM index group. The ROC curves illustrate the variation of sensitivity and specificity and hence, the diagnosis, when different thresholds are used. The ♦ symbol represents the optimum threshold. A threshold to the left (right) results in a test with higher specificity (sensitivity) but lower sensitivity (specificity). Nonlinear features show more regular behavior than ODI3 when different cut-off points are used to determine the presence of OSA. Figs. 4–6 show the scatter diagram for ODI3, ApEn02 and CTM1, respectively. Regression curves, linear (dashed line) and quadratic (solid line), which best fit the data in a least square sense, are also depicted. We used the regression equations to derive an AHI from each index. Due to their best fitting to the data, the quadratic curves ( p 2 x 2 + p 1 x + p 0 ) were selected. Bland and Altman plots, displayed in Figs. 7–9 , were subsequently used to quantify the agreement between the original AHI obtained from PSG and these AHI derived from ODI3, ApEn02 and CTM1, respectively. A systematic bias can be shown in the Bland and Altman plot corresponding to the AHI derived from ODI3 (mean difference = 8.1), whereas the AHI from both the ApEn02 and the CTM1 have no bias (mean difference = 0.0). The limits of agreement (±1.96 S.D.) are wide (−34.8 to 51.5), indicating there is a great lack of agreement between both techniques. The limits of agreement decreased to ±37.9 when comparing the AHI obtained from PSG with that derived from ApEn02. Furthermore, limits of agreement in the Bland and Altman plot corresponding to CTM1 decrease by various events per hour (up to ±31.5 e/h) and the number of outliers is lower. We have also measured the performance of nonlinear methods in terms of computational time. ApEn02 and CTM1 were both applied to a common nocturnal SaO 2 record (8 h long) divided in epochs of 16.66 min using the same PC (AMD Athlon™ XP 3000 with 1 GB RAM). While ApEn02 achieved a mean computational time of 0.5833 s per epoch, CTM1 was 1000 times faster, decreasing computational time up to 0.5333 ms. 5 Discussion and conclusions Portable monitoring has been widely used as an alternative technique to PSG in the diagnosis of OSA syndrome. This study has shown that nonlinear analysis, and particularly CTM, could enhance the diagnostic capacity of SaO 2 signals recorded from nocturnal oximetry. CTM improved the diagnostic test values of classical indexes commonly derived from SaO 2 , e.g. ODIs and CT90. Moreover, CTM improved the results obtained applying other nonlinear measures previously studied by our own group [16–18] . We have shown that OSA patients have significantly lower CTM values (0.30 ± 0.20, mean ± S.D.) than OSA negative subjects (0.71 ± 0.18, mean ± S.D.) according to their higher dispersion in the second-order difference plots. Thus, we could say that SaO 2 signals from OSA patients are more variable than those from OSA negative subjects. We have divided the indexes under study in three different groups: classical oximetric indexes (ODI2, ODI3, ODI4 and CT90), regularity indexes from approximate entropy analysis (ApEn01, ApEn015 and ApEn02) and variability indexes from CTM analysis (CTM1, CTM3 and CTM6). Tables 3 and 4 show that CTM1 reached the best statistics and parameters obtained from ROC analysis, significantly improving the results of classical indexes. CTM1 was significantly correlated ( p < 0.0001) with ODIs and regularity indexes from ApEn analysis ( r > 0.6). Furthermore, CTM1 showed the highest correlation coefficient ( r = −0.738) with AHI. The ROC curve analysis also yielded the best results for nonlinear methods, and particularly for CTM1. Whereas classical indexes achieved high specificities but very poor sensitivities, nonlinear methods provided sensitivity and specificity values both greater than 80%, leading to high accuracies. High positive and negative predictive values, as well as small likelihood ratios, make nonlinear methods especially useful to help in OSA diagnosis. CTM1 reached 90.1% sensitivity, 82.9% specificity, a PPV of 88.5, a NPV of 85.1, a LR+ of 5.27, a LR− of 0.12 and an accuracy of 87.5%. The area under the ROC curve was 0.924, the largest one compared with other indexes. A significant drawback of nocturnal oximetry common to most sleep studies is the substantial number of false negative cases. Subjects involved in these studies are typically referred to the sleep units because they are suspected of suffering from sleep apnea. Thus, the population under study has a very high prevalence of the disease, resulting in a small percentage of patients who test negative with a high chance to be incorrectly classified (false negative result). Additionally, due to the high prevalence, patients with a positive result are more likely to have a true positive result than a false positive result [7] . This leads to a high LR+ but also to a high LR−. Many sleep studies used two different thresholds to achieve both high LR+ and low LR−, with the limitation that patients presenting diagnostic values between both cut-off points will not be classified [7] . In the present research, we used a single threshold. All subjects in the data set could be diagnosed, although both false positive and false negative cases were present. However, we achieved significant operating characteristics with a single threshold, increasing the probability that a patient testing positive has an abnormal AHI (LR+ > 5.0) and decreasing the probability that a patient testing negative has an abnormal AHI (LR− < 0.2). Previous studies based on classical oximetric indexes [14,23,36,37] provided higher sensitivity but lower specificity, whereas others [9,38,39] achieved higher specificity but significantly lower sensitivity. On the other hand, our results demonstrate that nonlinear methods provide significant sensitivity and specificity values, leading to high accuracies and areas under the ROC curve. The study carried out by Olson et al. [40] achieved similar sensitivity and specificity values, although our results are slightly better. The study by Nuber et al. [11] reached a higher sensitivity (91.8%) and good specificity (77.8%), but their study was based on a small sample (40 subjects). The diagnostic accuracy in terms of ROC analysis varies greatly among studies carried out by different researchers. Although these studies were probably developed under different conditions, the major limitation when using ODIs is that each study uses their own definition of desaturation. Moreover, the threshold used to diagnose OSA based on the AHI derived from PSG usually varies among studies. Our studies are guided to remove these uncertainties by using universally well-defined nonlinear methods. There is not a consensus neither in the definition of the AHI nor in the threshold subsequently used to determine the disease [14] . Thus, we assessed the diagnostic ability of each feature involved in this study when the AHI threshold used to diagnose OSA changed. The AROC was computed taking into account the following OSA thresholds: AHI ≥ 5, 10, 15 and 20. Table 5 summarizes the results from this analysis, showing that CTM achieves the highest area whatever the AHI threshold used. Previous studies by our own group applied nonlinear analysis to oximetric signals from portable monitoring in screening for OSA syndrome. We applied ApEn to SaO 2 signals, obtaining 88.3% sensitivity and 82.9% specificity [16] . Moreover, we also applied ApEn to heart rate oximetric signals from the same population, obtaining 71.2% sensitivity and 78.9% specificity [17] . LZ complexity and CTM were also applied to SaO 2 recordings. Using LZ, we obtained 86.5% sensitivity and 77.6% specificity, while with CTM the sensitivity was 90.1% and the specificity was 82.9% [18] . In the present work, we have extended our study applying CTM with different radii. Furthermore, we compared CTM with classical oximetric indexes and with ApEn. CTM1 significantly improves statistical and diagnostic test results of previous studies. Moreover, the computational time spent by CTM to process a common oximetric record (approximately 8 h long) is 1000 times smaller than that used by ApEn. Thus, CTM is more suitable to be incorporated as a software tool in an oximeter. Furthermore, CTM provides a graphical tool, the second-order difference plots, which could be very useful in the diagnosis of OSA syndrome. Physicians could make a preliminary diagnosis by visual inspection of these scatter plots. Furthermore, we studied the ability of CTM to provide a derived AHI. We plotted CTM1 versus the AHI from PSG and them we computed the linear and the quadratic regression equations. The AHI derived from the quadratic regression curve showed no bias and moderate agreement with the AHI from PSG, improving the results obtained with classical ODIs and with other nonlinear indexes. Hence, CTM analysis of SaO 2 signals from nocturnal oximetry could provide useful information in the development of alternative techniques to conventional PSG. From our study, we could derive that the recurrence of the apnea events typical of OSA were responsible for the high variability of SaO 2 signals in the OSA positive group measured by the CTM. However, it is known that altered respiratory patterns are not exclusive of OSA syndrome. A total of 42 subjects suffering COPD were included in our study. We have shown that COPD patients with OSA presented significantly higher CTM values than COPD patients without OSA. The CTM analysis provided 82.9% specificity, with 13 false positive cases. Six subjects (46%) within the false positive group suffered from COPD, while two subjects had a BMI > 34 kg/m 2 . If COPD patients were removed from the study, specificity increases to 87.5%. Regarding to the OSA positive group, our CTM study reached 90.1% sensitivity, with 11 false negative cases. Five patients (45%) within the false negative group had an AHI < 15 e/h. If those patients are removed from the study, sensitivity increases to 94%. We should take into account some limitations of our study. Firstly, regarding to the population under study, the sample size could be larger. Furthermore, OSA positive patients were predominantly studied. Thus, additional work is needed to apply our methodology to a new and larger data set with a wide spectrum of sleep-related breathing disorders, as well as to study groups of especial interest, such as healthy subjects, young snorers and patients with lung and/or cardiac diseases. Another limitation should be stated in relation to the applicability of our methodology. Oximetry signals were recorded simultaneously with PSG, eliminating potential confounders such as night to night variability of AHI, as well as ensuring that oximetry data were collected in exactly the same environment as the PSG data. However, further analyses using unattended nocturnal oximetry in home are necessary. In addition, the data collection process could be enhanced. Our oximeter takes one sample every 5 s. The study by Wiltshire et al. [41] showed that low sampling rates provide SaO 2 recordings with a low number of artifacts. However, low sampling frequencies potentially reduce the sensitivity and increase the specificity of a diagnostic test [7] . Although the study carried out by Warley et al. [29] showed that this sampling frequency provides reasonable resolution in SaO 2 variability, sampling at higher frequencies we could record SaO 2 signals more accurately, improving subsequent analyses. Moreover, a drawback related to the airflow measure in the PSG study should be mentioned. Thermistor is highly reliable in detecting static respiratory events (apneas). However, it is less effective with dynamic respiratory events (hypopneas) [42] . The use of nasal cannula can improve the detection of hypopneas [43] . Nevertheless, one of the limitations of nasal pressure is false positive detection of apneas/hypopneas due to nasal obstruction or mouth breathing, leading to ambiguous results. The American Academy of Sleep Medicine (AASM) Task Force suggested that differentiation of apneas from hypopneas was not necessary in clinical practice because both event types share a common pathophysiology and clinical consequences [27] . However, the use of nasal cannula could lead to an overestimation of the AHI due to the false positive events. On the other hand, thermistor may not be sensitive for detecting hypopneas, leading to an underestimation of the AHI [44] . However, thermistor is the most common method for defining breathing events based on a flow measurement [7] . In the present study, where the thermistor is used, the underestimation of the reference standard index leads to increase the number of false positive diagnoses when our proposed nonlinear methods are used. Based on an underestimated threshold, these subjects without OSA but with an AHI slightly below to this cut-off point will test positive, resulting in sensitivity values higher than the specificity ones. Table 4 shows this trend, where sensitivity is higher than specificity in ApEn01, ApEn015, ApEn02, CTM1 and CTM6. Nevertheless, we obtained high accuracy and area under the ROC curve, as well as high and low LR+ and LR− values, respectively. Moreover, we could also derive from Table 5 that CTM provides the highest AROC regardless of the threshold used to diagnose OSA. Finally, we would like to point out that the ability of portable monitors in OSA diagnosis has been generally assessed by comparing their results with those of the accepted reference standard: the sleep laboratory-based PSG [45] . However, both PSG and portable monitors have considerable night to night variability, which accounts for some loss of agreement between single-night observations of the two tests. It is known that AHI by itself has limited clinical significance, correlating poorly with symptoms or with outcome of treatment [46] . Respiratory events can be totally obstructive (apneas), partially obstructive (hypopneas) or very subtle upper airway obstructions which can lead to respiratory effort-related arousals (RERAs). It is known that RERAs can produce fatigue and daytime sleepiness without a significant number of apneas and hypopneas [47] . Thus, an index including apneas, hypopneas and RERAs would be a much more powerful reference index than AHI to diagnose OSA and it could detect more effectively those cases without major physiological complications. In summary, we have shown that a nonlinear analysis by means of the CTM could enhance the diagnostic capacity of oximetric signals recorded from nocturnal pulse oximetry. CTM could be a useful diagnostic tool that improves classical oximetric indexes commonly used by physicians. Second-order difference plots could allow physicians to make a preliminary diagnosis by visual inspection, while CTM provides a quantitative measure from those scatter diagrams, making both interrelated techniques suitable to be incorporated in the oximeters. Acknowledgements This work has been partially supported by a grant project from Consejería de Educación de la Junta de Castilla y León under project VA108A06 and SOCALPAR (Sociedad Castellano-Leonesa y Cántabra de Patología Respiratoria). References [1] C. Guilleminault A. Tilkian W.C. Dement The sleep apnea syndromes Ann Rev Med 27 1976 464 484 [2] C. Guilleminault J. Van Den Hoed M.M. Mitler Clinical overview of the sleep apnea syndromes C. Gilleminault W.C. Dement Sleep apnea syndromes 1978 Alan R. Liss New York [3] N.J. Douglas Recent advances in the obstructive sleep apnoea/hypopnoea syndrome Ann Acad Med 31 2002 697 701 [4] T. Young P.E. Peppard D.J. Gottlieb Epidemiology of obstructive sleep apnea Am J Resp Crit Care 165 2002 1217 1239 [5] R. Day R. Gerhardstein A. Lumley T. Roth L. Rosenthal The behavioral morbidity of obstructive sleep apnea Prog Cardiovasc Dis 41 1999 341 354 [6] Agency for Health Care Policy and Research (AHCPR). Systematic Review of the Literature Regarding the Diagnosis of Sleep Apnea: Summary, Evidence Report/Technology Assessment. Department of Health and Human Services (U.S. Public Health Service), 1999, No. 1, vol. E002. [7] W.W. Flemons M.R. Littner J.A. Rowlet P. Gay W.M. Anderson D.W. Hudgel Home diagnosis of sleep apnea: a systematic review of the literature Chest 124 2003 1543 1579 [8] L.J. Epstein G.R. Dorlac Cost-effectiveness analysis of nocturnal oximetry as a method of screening for sleep apnea–hypopnea syndrome Chest 113 1998 97 103 [9] K. Sano H. Nakano Y. Ohnishi Y. Ishii T. Nakamura K. Matuzawa Screening of sleep apnea–hypopnea syndrome by home pulse oximetry Nihon Kokyuki Gakkai Zasshi 36 1998 948 952 [10] E. Chiner J. Signes-Costa J.M. Arriero J. Marco I. Fuentes A. Sergado Nocturnal oximetry for the diagnosis of the sleep apnoea hypopnoea syndrome: a method to reduce the number of polysomnographies? Thorax 54 1999 968 971 [11] R. Nuber J. Varvrina W. Karrer Predictive value of nocturnal pulse oximetry in sleep apnea screening Schweiz Med Wschr (Suppl) 116 2000 120S 122S [12] B. Chaudhary S. Dasti Y. Park T. Brown H. Davis B. Akhtar Hour-to-hour variability of oxygen saturation in sleep apnea Chest 113 1998 719 722 [13] R. Golpe A. Jimenez R. Carpizo J.M. Cifrian Utility of home oximetry as a screening test for patients with moderate to severe symptoms of obstructive sleep apnea Sleep 22 1999 932 937 [14] U.J. Magalang J. Dmochowski S. Veeramachaneni A. Draw M.J. Mador A. El-Solh Prediction of the apnea–hypopnea index from overnight pulse oximetry Chest 124 2003 1694 1701 [15] N. Netzer A.H. Eliasson C. Netzer D.A. Kristo Overnight pulse oximetry for sleep-disordered breathing in adults Chest 120 2001 625 633 [16] F. Del Campo R. Hornero C. Zamarrón D. Abásolo D. Álvarez Oxygen saturation regularity analysis in the diagnosis of obstructive sleep apnea Artif Intell Med 37 2006 111 118 [17] C. Zamarrón R. Hornero F. del Campo D. Abásolo D. Álvarez Heart rate regularity analysis obtained from pulse oximetric recordings in the diagnosis of obstructive sleep apnea Sleep Breath 10 2006 83 89 [18] D. Álvarez R. Hornero D. Abásolo F. del Campo C. Zamarrón Nonlinear characteristics of blood oxygen saturation from nocturnal oximetry for obstructive sleep apnoea detection Physiol Meas 27 2006 399 412 [19] M.E. Cohen D.L. Hudson P.C. Deedwania Applying continuous chaotic modeling to cardiac signals IEEE Eng Med Biol 15 1996 97 102 [20] N. Burioka G. Cornélissen F. Halberg D.T. Kaplan H. Suyama T. Sako Approximate entropy of human respiratory movement during eye-closed waking and different sleep stages Chest 123 2003 80 86 [21] D. Caldirola L. Bellodi A. Caumo G. Migliarese G. Perna Approximate entropy of respiratory patterns in panic disorder Am J Psychiatry 161 2004 79 87 [22] A. Rechtschaffen A. Kales A manual of standardized terminology, techniques and scoring system for sleep stages of human subjects 1968 US Government Printing Office, National Institutes of Health Publication Washington, DC [23] C. Zamarrón P.V. Romero J.R. Rodríguez F. Gude Oximetry spectral analysis in the diagnosis of obstructive sep apnoea Clin Sci 97 1999 467 473 [24] C. Zamarrón F. Gude J. Barcala J.R. Rodríguez P.V. Romero Utility of oxygen saturation and heart rate spectral analysis obtained from pulse oximetric recordings in the diagnosis of sleep apnea syndrome Chest 123 2003 1567 1576 [25] C. Zamarrón F. Pichel P.V. Romero Coherence between oxygen saturation and heart rate obtained from pulse oximetric recordings in obstructive sleep apnoea Physiol Meas 26 2005 799 810 [26] A.L. Meoli K.R. Casey R.W. Clark J.A. Coleman Jr. R.W. Fayle R.J. Troell Clinical practice review committee Hypopnea in sleep-disordered breathing in adults Sleep 24 2001 469 470 [27] American Academy of Sleep Medicine Task Force (AASM) Sleep-related breathing disorders in adults: recommendations for syndrome definition and measurement techniques in clinical research Sleep 22 1999 667 689 [28] Global Initiative for Chronic Obstructive Lung Disease (GOLD) Global strategy for the diagnosis management and prevention of chronic obstructive pulmonary disease, NHI Publication 2701 2001 National Health, Lung and Blood Institute (NHLBI/WHO) Bethesda p. 1–100 [29] A.R.H. Warley J.H. Mitchell J.R. Stradling Evaluation of the Ohmeda 3700 pulse oximeter Thorax 42 1987 892 896 [30] P. Lévy J.L. Pépin C. Deschaux-Blanc B. Paramelle C. Brambilla Accuracy of oximetry for detection of respiratory disturbances in sleep apnea syndrome Chest 109 1996 395 399 [31] S. Gyulay L.G. Olson M.J. Hensley M.T. King K. Murree Allen N.A. Saunders A comparison of clinical assessment and home oximetry in the diagnosis of obstructive sleep apnea Am Rev Respir Dis 147 1993 50 53 [32] S.M. Pincus Assessing serial irregularity and its implications for health Ann NY Acad Sci 954 2001 245 267 [33] S.M. Pincus A.L. Goldberger Physiological time-series analysis: what does regularity quantify? Heart Circ Physiol 35 1994 H1643 H1656 [34] S.M. Pincus Approximate Entropy as a measure of system complexity Proc Natl Acad Sci USA 88 1991 2297 2301 [35] J. Jeong J.C. Gore B.S. Peterson A method for determinism in short time series, and its applications to stationary EEG IEEE Trans Biomed Eng 49 2002 1374 1379 [36] J.M. Rodriguez P. de Lucas M.J. Sánchez J.L. Izquierdo R. Peraita J.M. Cubillo Usefulness of the visual analysis of night oximetry as a screening method in patients with suspected clinical sleep apnea syndrome Arch Bronconeumol 32 1996 437 441 [37] L. Lacassagne A. Didier M. Murris-Espin J.P. Charlet P. Chollet M.L. Leophonte-Domairon Role of nocturnal oximetry in screening for sleep apnea syndrome in pulmonary medicine: study of 329 patients Rev Mal Respir 14 1997 201 207 [38] P.J. Ryan M.F. Hilton D.A. Boldy A. Evans S. Bradbury S. Sapiano Validation of British Thoracic Society guidelines for the diagnosis of the sleep apnoea/hypopnea syndrome: can polysomnography by avoided? Thorax 50 1995 972 975 [39] R.T. Brouillette A. Morielli A. Leimanis K.A. Waters R. Luciano F.M. Ducharme Nocturnal pulse oximetry as an abbreviated testing modality for pediatric obstructive sleep apnea Pediatrics 105 2000 405 412 [40] L.G. Olson A. Ambrogetti S.G. Gyulay Prediction of sleep-disordered breathing by unattended overnight oximetry J Sleep Res 8 1999 51 55 [41] N. Wiltshire A. Kendrick J. Catterall Home oximetry studies for diagnosis of sleep apnea/hypopnea syndrome: limitation of memory storage capabilities Chest 120 2001 384 389 [42] R. Farré J.M. Montserrat M. Rotger E. Ballester D. Navajas Accuracy of thermistors and thermocouples as flow-measuring devices for detecting hypopneas Eur Respir J 11 1998 179 182 [43] J.M. Montserrat R. Farré E. Ballester M.A. Félez M. Pasto D. Navajas Evaluation of nasal prongs for estimating nasal flow Am J Respir Crit Care Med 155 1997 211 215 [44] F. Sériès I. Marc Nasal pressure recording in the diagnosis of sleep apnoea hypopnoea syndrome Thorax 54 1999 506 510 [45] W.W. Flemons M.R. Littner Measuring agreement between diagnostic devices Chest 124 2003 1535 1542 [46] W.A. Whitelaw R.F. Brant W.W. Flemons Clinical usefulness of home oximetry compared with polysomnography for assessment of sleep apnea Am J Respir Crit Care Med 171 2005 188 193 [47] J.F. Masa J. Corral M.J. Martín J.A. Riesco A. Sojo M. Hernández Assessment of thoracoabdominal bands to detect respiratory effort-related arousal Eur Respir J 22 2003 661 667
Year
DOI
Venue
2007
10.1016/j.artmed.2007.06.002
Artificial Intelligence in Medicine
Keywords
Field
DocType
Central tendency measure,Nonlinear methods,Blood oxygen saturation,Oximetry,Obstructive sleep apnea
Obstructive sleep apnea,Data mining,Approximate entropy,Internal medicine,Computer science,Overnight pulse oximetry,Nonlinear methods,Oxygen saturation,Cardiology,Central tendency measure,Polysomnography,Pulse oximetry
Journal
Volume
Issue
ISSN
41
1
0933-3657
Citations 
PageRank 
References 
15
1.34
2
Authors
5
Name
Order
Citations
PageRank
Daniel Álvarez121225.22
Roberto Hornero260367.74
María García318611.96
Félix del Campo412618.12
Carlos Zamarrón5658.11