Remember me
ORIGINAL ARTICLE
Year : 2021 | Volume
: 70
| Issue : 1 | Page : 99-106
Optimal pressure in obstructive sleep apnea: auto-titrating positive airway pressure versus predictive equations
Iman Galal, Haytham S Diab
Department of Pulmonary Medicine, Ain Shams University, Cairo, Egypt
Correspondence Address:
BSc, MSc, MD Iman Galal
Department of Pulmonary Medicine and Respiratory Intensive Care Unit, Faculty of Medicine, Ain Shams University, Cairo, 11757
Egypt
Source of Support: None, Conflict of Interest: None
DOI: 10.4103/ejcdt.ejcdt_108_20
Background The prescription of equation-based predicted pressure (Ppred) using mathematical equations has been proposed as a practical strategy for the determination of effective positive airway pressure (PAP). This study aimed at comparing the differences between therapeutic pressures obtained by auto-titrating positive airway pressure (APAP) and 10 predictive mathematical equations among Egyptian patients with obstructive sleep apnea (OSA).
Patients and methods A retrospective cross-sectional study included 25 PAP-naïve patients (23 males and two females) with polysomnographically confirmed OSA. The 95th percentile pressure (P95) of an APAP device during an attended successful titration study was selected as the effective reference PAP, against which PPred calculated from ten predictive equations was compared.
Results Among the 25 patients included, four patients had mild, eight had moderate, whereas the remaining 13 patients had severe OSA. The mean P95 was higher than all predictive equations. A total of 24 (96%) of the 25 included patients experienced acceptable air leak during APAP titration. P95 correlated significantly and positively with pressures calculated by most of predictive equations except for those of equations 2, 6, and 7. Comparison between P95 and Ppred showed nonsignificant statistical difference for pressure calculated by Sériès equation, whereas Ppred of other equations were significantly lower than P95. Mean pressure difference between P95 and Ppred was around three in most of the equations with Sériès equation carrying the least mean pressure difference (0.7 cmH2O).
Conclusion P95 is higher than Ppred, yet Sériès equation was the closest to P95. The utility of predictive equations for determining effective PAP level can serve a beneficial role during titration studies and can be used with caution as a practical alternative for PAP titration.
Keywords: auto-titrating positive airway pressure, obstructive sleep apnea, positive airway pressure, predicted pressure, polysomnography
Positive airway pressure (PAP) is the gold standard treatment for obstructive sleep apnea (OSA), a sleep-related breathing disorder characterized by partial or complete upper airway obstruction during sleep. To initiate this, pressure titration should be performed during an attended in-lab polysomnography (PSG) [1]. During this titration study, the effective pressure is determined which represents the pressure that abolishes obstructive breathing disorders, including apneas, hypopneas, flow limitation, and snoring in all sleep stages and different body positions [1].
Unfortunately, attended manual in-lab PAP titration is both costly and time consuming. Accordingly, several previous studies have investigated the utility of acceptable, practical alternatives for in-lab titration. Some researchers assessed the role of incorporating different variables to formulate predictive equations that can be used either during in-lab titration or to quantify the starting pressure for continuous PAP therapy [2].
In 1993, Berthon-Jones [3] explored the possibility of using computer power to drive a turbine based upon the analysis of physiologic signals or their surrogates to construct an intelligent, very fast-reacting PAP device called auto-titrating positive airway pressure (APAP). These APAP devices can determine an appropriate PAP level, either unattended or in-lab, to provide the prescription pressure for fixed PAP [4], thus providing another alternative to manual in-lab titration. The pressure that exceeded only for 5% of the night recording [so-called 95th percentile pressure (P95)] was previously proposed as a fixed PAP [5],[6],[7],[8],[9].
Accordingly, these three methods of determining effective PAP (manual, APAP, and predictive mathematical equations) proved, both subjectively and objectively, to be equally effective in terms of reducing excessive daytime sleepiness and apnea–hypopnea index (AHI), respectively with comparable adherence rates [10].
The aim of this study was to compare the differences in therapeutic pressures obtained by the APAP device and 10 predictive mathematical equations among Egyptian patients with OSA.
Patients and methodsPatients
This retrospective cross-sectional study was conducted over 25 PAP-naïve patients with PSG confirmed OSA who underwent in-lab APAP titration. Anthropometric measures including body weight, height, BMI, and neck circumference (NC) were measured for all included patients. The study was approved by the institutional ethical committee.
Polysomnography
PSG was performed using 24 channel computerized level II PSG system (N4000 Embla; Somnologica, Reykjavik, Iceland) including the monitoring of electroencephalogram, submental and anterior tibial electromyogram, oxygen saturation, ECG, inductance plethysmography of the chest wall and abdomen, nasal pressure sensor, and oronasal thermistor. Manual scoring was performed for all patients according to the recommendations of the American Academy of Sleep Medicine (AASM) guidelines [11],[12].
Auto-titrating positive airway pressure titration
APAP titration study was performed during a standard attended overnight PSG to determine the optimal PAP level using an APAP device (AutoSetTM, ResMed, Sydney, Australia). The AutoSetTM device is a computer-based PAP system whose operation is based on proactive pressure augmentation following the detection of incipient upper airway obstruction. The operational characteristics of the AutoSetTM device feature detection of snoring, inspired flow limitation, apneas, and hypopneas using pneumotachograph and subsequently modulates mask pressure automatically in response to these events [13]. The minimum pressure was set at 3 cm H2O, and the maximum pressure was set at 20 cmH2O. Mask fitting was performed at the beginning of the study and the type of the mask was chosen based upon comfort of the patient and the predominant route of breathing during sleep. At the end of the titration sleep study, the data over the APAP device were downloaded; the median (P50), maximum (Pmax), P95 of the pressure values delivered by the device, the median air leak (middle value of leak during the study), the 95th percentile leak (leak flow amount covering 95% of the study period), and the maximum air leak (highest leak) during the APAP application, were determined.
Only patients with successful PAP titration were included in the study. The criteria for unsuccessful titration were as follows: total sleep time (TST) of less than 4 h; TST in the supine position of less than 1 h; rapid eye movement sleep duration of less than 10 min [14]; excessive overshoots of delivered pressure associated with sleep disruption [15]; insufficient correction of obstructive events not allowing us to recognize a therapeutic pressure; and median nocturnal air leak of more than 0.4 l/s [16],[17].
P95 of APAP device was considered as the fixed reference PAP against which the predictive equations were compared.
Predictive equations for positive airway pressure therapy
In all included patients, the predictive pressure (Ppred) of 10 mathematical equations retrieved from the published literature was calculated:
Equation (1) [18]:
Equation (2) [19]:
Equation (3) [20]:
Equation (4) [21]:
Equation (5) [22]:
Equation (6) [23]:
Equation (7) [24]:
Equation (8) [25]:
Equation (9) [26]:
Equation (10) [27]:
Where BMI (kg/m2), NC (cm), AHI, RDI: respiratory disturbance index, DI: desaturation index, and ODI: oxygen desaturation index.
Statistical analysis
Statistical analyses were performed using Statistical Package for Social Sciences software (SPSS for Windows, version 17.0; SPSS Inc., Chicago, Illinois, USA). Kolmogorov–Smirnov test of normality for continuous data was done. Descriptive statistics were presented as minimum, maximum, median, and mean±SD. Categorical statistics were presented as number and percentage. Comparison between the normally distributed mean values of P95 and the Ppred was done using the paired t test, whereas correlation analysis was done using Spearman’s correlation analysis test. Comparison between nonnormality distributed data was done using Mann–Whitney U test. Statistical significance was set at P value less than 0.05.
ResultsA total of 25 patients with PSG confirmed OSA who were apt candidates for PAP therapy were included in this study (23 males and two females). Using the OSA severity classification (mild for AHI≥5/h TST, moderate for AHI ≥15/h TST, and severe for AHI >30/h TST) [12], four patients had mild, eight had moderate, whereas the remaining 13 patients had severe OSA. The demographic, anthropometric, and respiratory PSG data are listed in [Table 1].
Table 1 Demographic, anthropometric, and respiratory polysomnographic dataOnly patients with successful PAP titration were included in the study. The comparison between baseline AHI during the daignostic study and residual AHI during APAP titration was statistically significant ([Figure 1]).
Figure 1 Comparison between baseline AHI and residual AHI. AHI, apnea–hypopnea index.The data downloaded from the APAP device, including P50, Pmax, P95, the median air leak, the 95th percentile air leak, and the maximum air leak during the APAP application as well as the pressure calculated by the 10 predictive equations, are listed in [Table 2]. The pressure profile in all included patients shows that the mean values of P95 were higher than those of all predictive equations. Of the 25 patients enrolled in this study, 24 (96%) patients experienced acceptable levels of air leak during APAP titration (median nocturnal air leak of <0.4 l/s) [16],[17].
Table 2 Auto-titrating positive airway pressure device data and predictive equations pressuresP95 shows a significant positive correlation with the pressures calculated by most of the predictive equations except for those of Sériès, Choi and colleagues, and Chuang and colleagues equations, where the correlations were nonsignificant. Comparison between P95 and equation-based Ppred showed nonsignificant statistical difference for the pressure calculated by Sériès equation, whereas the pressures calculated by the other nine predictive equations were significantly lower than P95 ([Table 3] and [Figure 2]).
Table 3 Predicted pressure of auto-titrating positive airway pressure versus mathematical equations: correlation and comparison
Figure 2 PAP pressure profile including: P95 and predictive equations. The horizontal black line is the median value, the box interquartile range and vertical line is the minimum–maximum range. P95, 95th percentile pressure; PAP, positive airway pressure.To examine the relationship between P95 and the equation-based Ppred, the distribution of the differences between these two sets of pressures is shown in [Table 4] and [Figure 3]. The mean pressure difference was around 3 in most of the predictive equations, yet Sériès equation carried the least mean pressure difference with P95 (0.7 cmH2O).
Table 4 Differences between 95th percentile pressure and predicted pressure
Figure 3 Distribution of differences between P95 and PPred. P95, 95th percentile pressure; Ppred, predicted pressure. Discussion The estimation of the therapeutic effective PAP for treating patients with OSA is classically determined during an in-laboratory titration sleep study yet, this is both costly and time consuming. Accordingly, ongoing efforts are continuously paid to establish alternative practical yet reliable methods for the determination of the effective pressure. In this study, several mathematical predictive equations previously published in the literature were evaluated against the effective pressure (P95) determined by an APAP device (AutoSetTM, ResMed, Sydney, Australia). The principal outcome was the different pressure profiles. The use of P95 as the effective pressure was in accordance with several previous studies in which P95 was assigned as the fixed effective PAP in patients with OSA [5],[6],[7],[8],[9].
Our results showed that the mean effective pressure of P95 was higher than that of all predictive equations, and the differences between them were statistically significant, except for Sériès equation, which was still lower than P95, yet the difference was statistically nonsignificant. Similar results concerning the increase in P95 in comparison with predictive equations were obtained in other studies [9],[10],[28],[29],[30],[31]. This finding was well justified by the fact that air leak is common during APAP titration, and this air leak is commonly compensated by the increase in the delivered pressure till the cutoff value of 0.4 l/s. In the present study, 24 (96%) out of the 25 patients included had air leak during APAP titration, and this explains the increase in the mean values of P95 over the pressure profile of all predictive equations which in turn are not affected by the air leak problems.
The mean pressure differences between P95 and equations 1, 9, and 3, respectively (Basoglu and Tasbakan, Hoffstein and Mateika, and Hoheisel and Teschler, respectively), were the highest compared with the remaining predictive equations. Accordingly, the finding that these equations predicted the lowest PAP in our study was further investigated, and it was found that the establishment as well as the validation of the first equation (Basoglu and Tasbakan) was done among the Turkish population in comparison with the effective pressure obtained during manual titration study which is not the case in our study, as the effective pressure was assumed to be the P95 of the APAP in which air leak was compensated by increase in driving pressure. As for the other two equations (Hoffstein and Mateika and Hoheisel and Teschler), these two mathematical equations were the earliest to develop at the time before the establishment of the definition of optimal pressure required to abolish snoring and airflow limitation. Additionally, the target pressure of Hoffstein and Mateika equation was the pressure required to reduce AHI to less than 10/h. The second equation (Sériès) was the closest to the P95 and carried the least mean pressure difference as the target pressure in this study was the pressure required to eliminate the inspiratory airflow limitation, as was applied by the APAP used in the current study, a further slight pressure increase in the P95 in the present study in comparison with the PPred of the Sériès equation was expected to compensate for the air leak problem. It is important here to highlight that although air leak is common with APAP, yet it should be kept within the acceptable level (<0.4 l/s) [16],[17].
In a previous study, it was hypothesized that APAP can be initiated at home with a backup equation-based reference pressure as an alternative for performing titration study [32]. This has long been a target for initiating PAP without the time and money consumption factor of sleep titration studies. Moreover, this was strengthened by the fact that most of these equations correlated with pressure derived from APAP, with some of them are even very close to the pressure levels derived from the APAP.
In a previous attempt, one study designed a model of predictive formula derived from a small sample of Egyptian patients with OSA using the minimal oxygen saturation as well as the neck-to-height ratio, yet this formula is not yet validated [33]. Unfortunately, this formula was not tested in the present study because of the lack of data concerning the neck-to-height ratio among the included patients.
Using the initial pressure setting derived from predictive equations to set a PPred in titration studies is beneficial to minimize the changes in pressure and subsequently the time needed to establish the final effective pressure, and this would ultimately have a positive effect over the success of PAP titration. Finally, based on the aforementioned results in this study, we reasoned that the predictive pressure derived from Sériès equation was the closest to the effective P95 of APAP and that the utility of predictive equations for determining the effective PAP level can serve a beneficial role during sleep titrations studies and can be used with caution as a practical alternative for PAP titration. This study carries the advantage of using several predictive equations from various population bases.
Last but not the least, it is worth mentioning that taking into consideration that the different demographic and anthropometric measures are determined by the ethnicity, future studies should assess the establishment of a validated predictive equation based on the Egyptian somatic characteristics.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References
Comments (0)