Alternative Methods of Titrating Continuous Positive Airway Pressure

A Large Multicenter Study

Juan F. Masa, Antonio Jime nez, Joaqu n, Francisco Capote, Carmen Monasterio, Mercedes Mayos, n Dura n, Lourdes Herna ndez, Ferra n Barbe , Andre s Maimo , Manuela Rubio, and Jose M. Montserrat Joaqu n Tera
ntara Hospital, Ca ceres; Hospital de Valdecilla, Santander; Txagorritxu Hospital, Vitoria; Virgen del Roc San Pedro de Alca o Hospital, Sevilla; Hospital de Bellvitge, Sant Pau Hospital, and Clinic Hospital, Barcelona; General Yagu e Hospital, Burgos; Son Dureta Hospital, and Joan March Hospital, Palma de Mallorca, Spain

Standard practice for continuous positive airway pressure (CPAP) treatment in sleep apnea and hypopnea syndrome (SAHS) requires pressure titration during attended laboratory polysomnography. However, polysomnographic titration is expensive and time-consuming. The aim of this study was to ascertain, in a large sample of CPAP-naive patients, whether CPAP titration performed by an unattended domiciliary autoadjusted CPAP device or with a predicted formula was as effective as CPAP titration performed by full polysomnography. The main outcomes were the apneahypopnea index and the subjective daytime sleepiness. We included 360 patients with SAHS requiring CPAP treatment. Patients were randomly allocated into three groups: standard, autoadjusted, and predicted formula titration with domiciliary adjustment. The follow-up period was 12 weeks. With CPAP treatment, the improvement in subjective sleepiness and apneahypopnea index was very similar in the three groups. There were no differences in the objective compliance of CPAP treatment and in the dropout rate of the three groups at the end of the follow-up. Autoadjusted titration at home and predicted formula titration with domiciliary adjustment can replace standard titration. These procedures could lead to considerable savings in cost and to significant reductions in the waiting list. Keywords: autoCPAP titration; CPAP efficacy; CPAP titration; sleep apnea syndrome

The sleep apnea and hypopnea syndrome (SAHS) is a disorder affecting 2 to 4% of the adult population (1). Nasal continuous positive airway pressure (CPAP) is the most effective treatment in patients with SAHS symptoms (26). Standard practice for CPAP treatment requires pressure titration during an attended laboratory polysomnography. The aim of this procedure is to identify an effective pressure to remove

apneas, hypopneas, snoring, and arousals. However, polysomnographic titration is expensive and time-consuming. Autoadjusted titration with autoCPAP devices has been proposed (7) to overcome these disadvantages. These devices monitor one or more of the following parameters: snoring, ow, or impedance; these are monitored to detect respiratory events and adjust the CPAP pressure. They also calculate the optimal pressure automatically. To date, only autoCPAP titration by standard laboratory polysomnography has been recommended (8), given that very few studies (913) have been performed in unattended conditions, yielding contradictory results in CPAP-naive patients (11). The prediction of CPAP pressure using a formula that includes the apneahypopnea index (AHI) and the anthropometric parameters has been proposed (14). This could also be used to simplify conventional CPAP titration (15, 16). A randomized crossover controlled study on 18 CPAP-naive patients demonstrated that the efcacy of a predicted formula titration with athome self adjustment was similar to that of standard titration (17). In addition, a nonrandomized study (18) found a similar clinical improvement and CPAP adherence between predicted formula and standard titration groups. Therefore, if unattended autoadjusted or predicted formula CPAP titration is performed, considerable savings in cost and signicant reductions in the waiting lists could be achieved worldwide. The aim of this study was to ascertain, in a large sample of CPAP-naive patients, whether CPAP titration performed with an unattended domiciliary autoadjusted CPAP device or with a predicted formula was as effective as CPAP titration performed by full polysomnography. The main outcomes were the AHI and the subjective daytime sleepiness after xed CPAP treatment. Thus, we ascertained whether these methods were alternatives to the conventional CPAP titration. Some of the results of this study have been reported in the form of an abstract (19).

(Received in original form December 31, 2003; accepted in final form July 27, 2004) Supported by ISCIII-RTIC-0311, JUNTAEX-IPR00A064, and SEPAR. The authors are members of the Spanish Group of Breathing Sleep Disorders. The authors acknowledge the cooperation of the following group: Alejandro Pedro-Mingo, GlaxoSmithKline, Madrid, Spain; Agust ntara n Sojo, San Pedro de Alca Hospital, Ca ceres, Spain; Rosario Carpizo and Marta Cabello, Hospital de Valdecilla, Santander, Spain; Ramo n Rubio and Germa n de la Torre, Txagorritxu Hospital, Vitoria, Spain; Carmen Carmona and Georgina Botebol, Virgen del Roc o Hospital, Sevilla, Spain; Mar a Somoza and Marina Lumbierres, Hospital de Bellvitge, Barcelona, Spain; Fa tima Morante and Joaqu n Sanchis, Sant Pau Hospital, Barcelona, Spain; Mo nica Gonzalez and M. L. Alonso Alvarez, General Yagu e Hospital, Burgos, Spain; and Lola R. Mayoralas and Margalida Bosch, Son Dureta Hospital, Palma de Mallorca, Spain. Correspondence and requests for reprints should be addressed to Juan F. Masa, M.D., C/ Rafael Alberti 12, 10005 Caceres, Spain. E-mail: fmasa@separ.es This article has an online supplement, which is accessible from this issues table of contents at www. atsjournals.org
Am J Respir Crit Care Med Vol 170. pp 12181224, 2004 Originally Published in Press as DOI: 10.1164/rccm.200312-1787OC on July 28, 2004 Internet address: www.atsjournals.org

METHODS
Study Design
This randomized controlled clinical trial included two test groups (autoadjusted and predicted formula titration) and one control group (standard titration). To determine differences in the main outcomes between groups, we used a power of 0.8 and an error of 0.05. The study was open to the researchers and blinded to the technicians who set the questionnaires and analyzed the polysomnographies.

Patients
Patients requiring CPAP treatmentAHI 30 and relevant daytime subjective sleepiness (20) (Epworth sleepiness scale 12)aged between 18 and 70 years were consecutively recruited from 10 sleep centers in Spain. The exclusion criteria were as follows: psychophysical incapacity to perform questionnaires, patients with chronic disease (cancer, chronic pain, renal failure, moderate or severe chronic obstructive pulmonary disease, etc.), drug or alcohol addiction, Cheyne-Stokes syndrome, life-threatening SAHS, patients with previous uvulopalato-

 nez, Dura n, et al.: Methods of Titrating CPAP Masa, Jime

Can we do a similar study ??

1219

pharyngoplasty (UPPP), absence of a partner at home, important chronic nasal obstruction, lack of skill in adjusting the nasal mask in a daytime CPAP trial (see Study Protocol), and refusal to participate in the study. Three hundred sixty patients (120 patients per group) were nally included.

and the pressure was increased by 1 cm H2O every 5 minutes until the apneas disappeared. Thereafter, the pressure was increased by 1 cm H2O every 10 minutes until the hypopneas, ow limitation, and snoring disappeared. This last pressure was considered to be the optimal pressure.

Study Protocol
Patients with suspicion of SAHS were referred for conventional polysomnography. The baseline and CPAP polysomnographic studies were analyzed manually, at each participating center, according to standard criteria (21, 22). The sample rate and the ltering of the signals were standardized across centers. The neurologic variables were respectively as follows: electroencephalogram (C3/A2 and C4/A1) 200 Hz and 35 to 0.5 Hz; electrooculogram 100 Hz and 35 Hz to 0.5 Hz; electromyogram 200 Hz and 70 to 15 Hz. Breathing variables were scored based on the ow tracing provided by a nasal cannula (sampling rate 40 Hz) in baseline polysomnographies and by a pneumotachograph (sampling rate 40 Hz) connected to the nasal mask in CPAP polysomnographic studies. Thoracoabdominal motion was measured by thoracic and abdominal bands. Oxygen saturation was recorded by a nger-pulse oximeter. An apnea was dened as an absence of airow of 10 seconds and a hypopnea as any discernible airow reduction for at least 10 seconds with a drop in oxygen saturation 3% or nal arousal (20). After the baseline conventional polysomnography, a 20-minute daytime CPAP trial was performed only if the patient met the inclusion and exclusion criteria. The patients were also excluded when a signicant nasal obstruction or a lack of skill in adjusting the nasal mask was detected during this CPAP trial. Once the patients were included, they were randomized into one of the three study groups (Figure 1). Each center received written instructions from the coordinating center to carry out the study. This document standardized the questionnaires; the 20 minutes of CPAP trial; the information about the study and CPAP treatment for the patients; the three types of titration (standard, autoadjusted, and predicted formula); the list of the excluded and abandoned cases and secondary effects; the informed consent; and the visual analogical scale. The recommendations to perform the Epworth scale and the quality-of-life scales were also included in these guidelines while maintaining the specic recommendations of the authors. Patient inclusion was competitive at the centers. The inclusion period nished when the total number of patients included were 360. To avoid imbalance, each center could include up to 45 patients. The data collection was performed in a home-designed electronic database available at each center. To minimize potential errors, the arithmetical calculation, prediction formula, and quality of life scales were automated. At the end of the study each center e-mailed the data (in Excel format) generated automatically by the software. All the data les were assembled together in one table to perform the nal analysis ceres. at the coordinating center in Ca

Autoadjusted Titration
The patients received information on the autoCPAP and slept at home with the device (Autoset-T; ResMed, Sydney, Australia) for one night. The pressure was set to start automatically, after 20 minutes for adaptation (from 4 cm H2O up to a maximum of 16 cm H2O). The same model of nasal mask (Mirage; ResMed) was used for all the patients. On the following morning the patient answered a simple questionnaire: (1 ) At what time did you fall asleep? (2 ) At what time did you wake up? The automatic pressure prole was reviewed in the sleep laboratory. The recording was considered to be acceptable if all the following criteria were met: (1 ) the total sleep time, subjectively appreciated by the patient, was at least 5 hours; (2 ) the recording period in the autoCPAP device was at least 6 hours; and (3 ) the mean leak was lower than 0.4 L/second in the statistics obtained from the autoCPAP machine or the leak was lower than 0.4 L/second for at least 5 hours in the visual estimation of the raw data. The recording was repeated for two additional nights whenever it was unacceptable. A titration failure was dened when none of the three recordings obtained were acceptable. The optimal pressure was determined visually on the raw data of the autoCPAP device (view night prole) by analyzing the pressure that included 90% of the periods with a leak lower than 0.4 L/second (percentile 90) (23, 24).

Predicted Formula Titration

The optimal pressure was estimated by an equation already published (14): predicted pressure (0.16 BMI) (0.13 neck circumference) (0.04 AHI) 5.12. If the pressure obtained was more than 9 cm H2O, then the patient started with 9 cm H2O at home. On Day 15 or Day 30 of follow-up, pressure could be increased by 1 or 2 cm H2O if the patients partner afrmed that snoring or apnea was not completely eliminated. The optimal pressure was the pressure achieved after the third visit.

Follow-up and Outcomes

Once the optimal pressure was achieved, treatment was initiated with a xed level of CPAP at home. Not all the centers had the same CPAP devices, but a manometer was used to check the pressure level in all the patients. The patients were evaluated on four occasions (Figure 1). The second and third visits were used to increase the CPAP pressure in the predicted formula group, and to try to alleviate the possible secondary effects in all the groups. After 12 weeks all patients underwent a new polysomnography with CPAP. The optimal pressure obtained in the initial titration was maintained throughout the night. The main outcomes were the AHI and the Epworth sleepiness scale. The secondary outcomes were the different quality-of-life tests: FOSQ (Functional Outcomes of Sleep Questionnaire) (25), SF-36 (Medical Outcomes Study 36-Item Short-Form Health Survey) (26), EuroQol (27), and an analogical scale. The last one consisted of a 12-cm straight line on which the patient had to indicate his or her health status related specically to SAHS. The limits of the line marked the worst and the best possible status. This test is similar to the EuroQol thermometer but related specically to SAHS. Other secondary outcomes were the percentage of withdrawals, secondary effects, and CPAP compliance. The study was approved by the ethics committees of the 10 centers. Informed consent was obtained from all the patients.

Standard Titration
The patients included in this group underwent a second polysomnography for manual CPAP titration. The starting pressure was 4 cm H2O,

Interobserver Agreement on Autoadjusted Titration

Before the study, one researcher from each center underwent a short period of training based on ve representative Raw Data graphics from the Autoset T. The training was performed to become acquainted with the visual analysis of the pressure. At the end of the investigation, we undertook another study to evaluate the interobserver agreement in the visual calculation of CPAP pressure (see the online supplement).

Figure 1. Study protocol. ESS: Epworth sleepiness scale. QL: quality of life test (FOSQ, SF 36, EuroQol and analogical scale). CPAP: continuous positive airway pressure. FOSQ Functional Outcomes of Sleep Questionnaire; SF-36 Medical Outcomes Study 36-Item Short-Form Health Survey.

Statistical Analysis
We compared the characteristics of the three groups at baseline (before therapy) and CPAP pressure and use at the end of the study (after

1220

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 170 2004 TABLE 1. CAUSES OF WITHDRAWAL FROM THE STUDY
Standard (n 126) Intolerance, number (%) Titration failure, number (%) Lost in the follow-up, number (%) Other, number (%) Total withdrawal, number (%) 6 3 7 3 19 (4.8) (2.4) (5.5) (2.3) (15.1) Autoadjusted (n 119) 4 5 2 2 13 (3.4) (4.2) (1.7) (1.7) (10.9) Predicted Formula (n 115) 8 (7.0) 0 4 (3.8) 1 (0.9) 13 (11.3) p Value* ns ns ns ns ns Total (n 360) 18 8 13 6 45 (5.0) (2.2) (3.6) (1.7) (12.5)

* Statistical comparisons between autoadjusted and predicted formula groups are not presented.

therapy) by one-way ANOVA for continuous variables. Where appropriate, differences between individual means were tested using the LSD (least signicant difference) (SPSS 11.0; SPSS Inc., Chicago, IL). If the continuous variables were not normally distributed, a nonparametric test (Kruskal-Wallis and Dunn post hoc to identify differences between individual means) was used. For qualitative variables, 2 analysis or Fishers exact test was employed. A p value of 0.05 was considered to be statistically signicant.

RESULTS
One hundred six patients (23%) out of the 466 initially evaluated were excluded for the following reasons: chronic disease (40; 37.7%), severe nasal obstruction (13; 12.3%), refusal to participate in the study protocol (12; 11.3%), psychophysical incapacity to answer the questionnaires (10; 9.4%), absence of partner at home (10; 9.4%), alcohol addiction (9; 8.5%), previous UPPP (6; 5.7%), lack of skill in adjusting the nasal mask (5; 4.7%), and life-threatening SAHS (1; 0.9%). The number of patients per center varied according to the competitive nature of the inclusion: four centers included 45 patients, one 39, one 37, two 33, one 31, and one included 7. No intercenter differences were found in the main and secondary outcomes. Out of 360 patients nally included, 45 (12.5%) abandoned the study (Table 1). No differences were found in the percentages of withdrawal between the standard (15.1%), the autoadjusted (10.9%), and predicted formula groups (11.3%). Likewise, there were no signicant differences in the titration failures between the standard (2.4%) and the autoadjusted groups (4.2%).

The autoadjusted titration at home was achieved on the rst attempt in 98 patients out of the 119 patients initially included. It was repeated once in 21 patients and twice in 11 patients. Titration failed in ve patients. The reasons for these 32 repeated titrations were excessive leakage in 20 cases and a registration period under 6 hours or a sleep time under 5 hours (subjectively appreciated by the patient) in the remaining 12 cases. In the predicted formula titration group, CPAP pressure was increased during the follow-up in 32 of 115 patients initially included. The total increase was 1 cm H2O in 16 patients, 2 cm H2O in 10, 3 cm H2O in 4, and 4 cm H2O in 2 patients. The general characteristics of the population nally studied are shown in Table 2. No statistical differences were observed in the three study groups. The Epworth Sleepiness Scale and the polysomnographic variables (AHI, arousal index, percentages of sleep periods, and oxygen saturation) showed a similar statistical improvement after CPAP treatment in all the groups (Table 3). The AHI under CPAP treatment was slightly higher in the predicted formula group. The nal CPAP pressure was lower (8.4 1) in the predicted formula group than in the other two groups (standard [8.8 1.9] and autoadjusted [9.1 1.9]), although there were statistically signicant differences only between autoadjusted and predicted formula groups. There were no statistical differences in the CPAP compliance between groups. All the quality of life tests improved after treatment in all the groups (Table 4). In the autoadjusted group the degree of improvement in SF 36 physical and EuroQol was lower than that observed in the standard group. The quality of life scores

TABLE 2. GENERAL CHARACTERISTICS OF THE STUDY POPULATION

Standard (n 107) Age, yr Sex, % men BMI, kg m2 Active worker, % Primary school only, % Alcohol, gr/d Active smoker, % Blood hypertension, % FEV1, % predicted Sleep, h/night Habitual snoring, % Apneas observed, % Nocturia, % Restlessness, % Morning headache, % 51.0 (9.1) 86.9 33.6 (8.4) 85.0 55.1 21.8 (21.2) 45.3 55.4 96.0 (18.8) 6.9 (1.1) 90.7 62.6 23.4 47.7 14.0 Autoadjusted (n 106) 52.2 (10.4) 89.6 33.1 (6.3) 76.4 54.7 24.2 (22.6) 40.0 57.4 94.1 (17.9) 7.0 (1.5) 85.8 58.5 31.1 43.4 12.3 Predicted Formula (n 102) 51.1 (9.6) 89.2 33.7 (6.2) 75.5 58.8 27.3 (27.0) 37.2 60.5 93.0 (18.0) 7.0 (1.5) 94.1 76.6 24.5 50 13.7 p Value ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns

Definition of abbreviations: BMI body mass index; ns not significant.

 nez, Dura n, et al.: Methods of Titrating CPAP Masa, Jime TABLE 3. EPWORTH SLEEPINESS SCALE, POLYSOMNOGRAPHIC VARIABLES, AND CONTINUOUS POSITIVE AIRWAY PRESSURE AND COMPLIANCE
Standard (n 107) Pretreatment ESS AHI Arousal index Light sleep, % Deep sleep, % REM sleep, % SaO2 90% of TST CPAP pressure, cm H2O CPAP use, h/d 15.9 (3.5) 61.8 (22.0) Posttreatment 7.9 (4.6) 5.1 (6.8) Autoadjusted (n 106) Pretreatment 15.2 (3.5) 62.8 (22.8) Posttreatment 7.2 (4.4) 4.9 (7.6) Predicted Formula (n 102) Pretreatment 16.1 (3.6) 63.5 (24.1) Differences Autoadjusted

1221

p Value to p Value to PostComparison Comparison treatment at Baseline at Endpoints Standard ns ns ns ns ns ns

p Value Predicted of the Formula Differences ns ns ns ns ns ns

7.4 (4.6) 8.7 (15.0)

8.0 (4.8) 8.1 (5.4) 8.7 (5.7) 56.6 (21.0) 57.9 (22.6) 54.8 (25.2) 42.9 (18.3) 43.3 (19.4) 45.6 (23.7) 16.0 (16.0) 16.4 (15.0) 17.8 (18.2) 9.7 (12.3)11.5 (11.8) 12.4 (12.6) 6.1 (8.9) 4.7 (8.1) 5.7 (9.0)

55.2 (18.0) 12.3 (10.0) 55.5 (19.3) 12.0 (8.5)

59.0 (23.0) 13.4 (13.9)

77.3 (12.1) 61.3 (15.6) 78.1 (11.7) 61.3 (15.3) 77.4 (13.2) 59.7 (17.4) 8.5 (9.0) 14.1 (6.6) 18.3 (12.5) 20.2 (7.2) 8.1 (9.2) 14.0 (5.9) 19.8 (12.0) 18.8 (6.8) 8.0 (9.1) 14.0 (7.9) 20.2 (13.6) 19.7 (7.9)

25.3 (25.0)

3.0 (13.9) 29.9 (27.3)

1.4 (4.1)

30.4 (27.6)

4.5 (14.2)

ns

22.0 (28.1) 28.2 (26.1)

25.9 (25.3)

ns

8.8 (1.9) 5.2 (2.0)

9.1 (1.9) 5.3 (1.9)

8.4 (1.0) 5.2 (1.9)

0.05* ns

Definition of abbreviations: AHI apneahypopnea index; CPAP continuous positive airway pressure; ESS Epworth Sleepiness Scale; TST total sleep time. Data are presented as mean values (SD). * Autoadjusted versus predicted formula.

at baseline were higher in the autoadjusted group than in the standard group (see Discussion for details). In the predicted formula group the level of improvement in the quality of life tests was similar to that obtained in the standard group. As regards the secondary effects (Table 5), there were no important differences between the standard and the other groups. Nevertheless, there was a tendency for the autoadjusted group to present more side effects.

DISCUSSION
This is the rst randomized controlled study that determines whether autoadjusted CPAP titration at home or a predicted formula CPAP titration is an alternative to standard CPAP titration in a large sample of CPAP-naive patients. This study suggests that these alternative titration methods improve the clinical symptoms and the polysomnographic parameters while main-

taining adherence, use of treatment, and frequency of secondary effects similar to those obtained with the standard method. A number of studies have shown that some autoCPAP devices are effective in obtaining optimal CPAP pressure (2830), but only a few of these studies have centered on the efcacy of autoCPAP titration in unattended conditions and CPAP-naive patients. As for the AHI, some clinical series have yielded acceptable results (911, 13). One randomized controlled study (12) showed that the improvement in sleepiness and the CPAP acceptance resembled that obtained by conventional titration despite the fact that no sleep studies were performed during the follow-up period to determine the AHI. Our study conrms the results of the clinical series (9, 10, 13) with the addition of an improvement in the polysomnographic variables. Series (9) analyzed autoCPAP recordings, after 1 or 2 weeks of domiciliary CPAP use, to determine the optimal pressure. We obtained effective autoCPAP titration at home using only

TABLE 4. QUALITY OF LIFE TEST

Standard (n 107) Pretreatment FOSQ SF 36 Physical SF 36 Mental EuroQol Index EuroQol thermometer 84.4 44.3 45.6 0.75 Posttreatment Autoadjusted (n 106) Pretreatment Posttreatment 108.0 47.3 51.8 0.81 (14.3) (7.8) (9.2) (0.18) Predicted formula (n 102) Pretreatment 85.9 42.3 45.5 0.71 Posttreatment Differences Standard 20.8 4.3 3.9 0.10 (20.1) (6.9) (10.0) (0.17) Autoadjusted 14.2 1.4 4.0 0.03 Predicted formula

p Value to Comparison at Baseline 0.001* 0.001 ns 0.01 0.01* 0.001 0.05*

P value of the differences

(22.8) 105.1 (16.0) 94.4 (17.0) (8.7) 48.6 (7.3) 45.9 (8.6) (12.2) 49.4 (10.4) 47.5 (10.4) (0.21) 0.85 (0.18) 0.79 (0.17)

(20.8) 106.0 (15.5) (8.6) 45.7 (8.4) (11.8) 48.4 (11.0) (0.19) 0.79 (0.16) 71.5 (15.6) 69.2 (19.4)

(17.2) 20.1 (18.4) ns (7.7) 3.4 (7.7) 0.01* (10.8) 3.0 (9.2) ns (0.15) 0.08 (0.16) 0.001*

62.9 (19.2)

76.1 (15.6) 70.2 (16.4) 73.5 (15.1) 60.7 (21.1) 69.6 (21.1) 55.5 (22.4) 71.1 (18.6) 48.4 (22.0)

13.2 (17.1) 21.3 (29.9)

3.0 (17.5) 10.8 (18.8) 0.001* 15.5 (26.0) 20.7 (27.1) ns

Analogical scale 48.3 (21.6)

Definition of abbreviations: FOSQ Functional Outcomes of Sleep Questionnaire; SF-36 Medical Outcomes Study 36-Item Short-Form Health Survey. Data are presented as mean values (SD). * Standard versus autoadjusted. Autoadjusted versus predicted formula.

1222

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 170 2004 TABLE 5. SECONDARY EFFECTS OF CONTINUOUS POSITIVE AIRWAY PRESSURE TREATMENT
Standard (n 107) Skin abrasion, % Rhinitis, % Conjunctivitis,% Oral dryness, % Mask intolerance, % Chest discomfort, % Aerophagia, % Noise, % Headache, % Claustrophobia, % Smothering sensation, % Insomnia, % Difficulty exhaling, % Bed partner intolerance, % Want to continue with CPAP, % 27.1 17.8 3.7 24.3 3.7 1.9 0.9 14.0 1.9 4.7 0.9 7.5 0.9 1.9 97.2 Autoadjusted (n 106) 21.7 24.5 3.8 20.8 6.6 1.9 6.6 18.9 6.6 3.8 5.7 7.5 2.8 11.3 92.5 Predicted Formula (n 102) 20.6 13.7 2.0 16.7 3.9 1.0 2.9 13.7 4.9 2.9 3.9 11.8 2.0 6.9 98.0

Definition of abbreviation: CPAP continuous positive airway pressure.

one night in 82% of the patients. Two additional nights were performed with the patients that failed during the rst night. As a consequence, optimum CPAP pressure was obtained in 96% of all subjects in the autoadjusted group. This has important economic and practical implications given that the number of patients that can be titrated at home with one device is higher with our methodology. One of the arguments against autoadjusted CPAP titration is that the oxygenation level is not measured (8). Our patients, in the autoadjusted titration group, suffered from signicant oxygen desaturation in the baseline study. However, after CPAP treatment, the oxygenation parameters were similar to those of the standard titration group. Therefore, this nding challenges the need for measuring oxygen saturation during autoCPAP titration. We used a specic autoCPAP device, the Autoset-T. This device monitors changes in airow into the mask and monitors snoring. It responds to apneas, ow limitation, and snoring. Given that different autoCPAP devices monitor diverse parameters, the results of this study should not be extrapolated to other devices (31, 32). To date, the main aim of the CPAP treatment has been to improve the clinical symptoms and to normalize the physiological variables in polysomnography. In our study, the improvement

in symptoms such as sleepiness, arousal index, oxygen saturation, and the AHI was very similar in all the groups, although the predicted formula group had a higher number of residual apneas and hypopneas (Table 3). A number of well designed studies have shown that the AHI is an independent factor of cardiovascular risk (3335) and trafc accidents (3638), but long-term implications of residual apneas and hypopneas after CPAP treatment have not been established. The optimal CPAP pressure in the predicted formula group has been calculated by a published equation (see Methods). Whenever the pressure was more than 9 cm H2O, it was initially set at 9 and home adjustment was performed to achieve a better individual adaptation to high CPAP pressures. The mean pressure after domiciliary adjustment (8.4 1.0) resembles the theoretical pressure calculated by the formula (8.4 1.8), despite the initial reduction in the pressure (8.0 0.9). Fitzpatrick and coworkers applied the same formula without an upper limit of pressure (17). Their mean initial CPAP pressure obtained was 8.5 0.4, and the nal pressure after home adjustment was 10.2 2.0. This latter value is higher than our nal CPAP pressure in the predicted formula group, probably because of differences in the methodology employed. Therefore, given our trend toward a higher residual AHI in the formula group, the possibility of rening our methodology could be considered,

Figure 2. Recommended protocol in patients with sleep apnea and hypopnea syndrome (SAHS) requiring CPAP treatment. If central SAHS, relevant gas exchange abnormality (daytime or during sleep), or uvulopalatopharyngoplasty (UPPP) is present, standard CPAP titration should be considered. Conversely, a daytime CPAP trial for 20 minutes can be done. If the patient suffers from a severe nasal obstruction or lacks the skills in adjusting the nasal mask, standard CPAP titration should be considered once again. If these factors are not present, CPAP titration can be performed by an autoadjusted device at home or by a predicted formula with domiciliary adjustment.

 nez, Dura n, et al.: Methods of Titrating CPAP Masa, Jime

1223
7. Teschler H, Berthon-Jones M, Thompson AB, Henkel A, Henry J, Konietzko N. Automated continuous positive airway pressure titration for obstructive sleep apnea syndrome. Am J Respir Crit Care Med 1996; 154:734740. 8. Berry RB, Parish JM, Hartse KM. The use of auto-titrating continuous positive airway pressure for treatment of adult obstructive sleep apnea. Sleep 2002;25:148173. 9. Series F. Accuracy of unattended home CPAP titration in the treatment of obstructive sleep apnea. Am J Respir Crit Care Med 2000;162:9497. 10. Berkani M, Lofaso F, Chouaid C, Pia dOrtho M, Theret D, GrillierLanoir V, Harf A, Housset B. CPAP titration by an auto-CPAP device based on snoring detection: a clinical trial and economic considerations. Eur Respir J 1998;12:759763. 11. Fletcher EC, Stich J, Yang KL. Unattended home diagnosis and treatment of obstructive sleep apnea without polysomnography. Arch Fam Med 2000;9:168174. 12. Stradling JR, Barbour C, Pitson DJ, Davies RJ. Automatic nasal continuous positive airway pressure titration in the laboratory: patient outcomes. Thorax 1997;52:7275. 13. MacLaren B, Cunnington D, Teichtahl H, Cherry G. In-the-home CPAP implementation using the ResMed AutoSet Spirit [abstract]. Am J Respir Crit Care Med 2003;167:A408. 14. Miljeteig H, Hoffstein V. Determinants of continuous positive airway pressure level for treatment of obstructive sleep apnea. Am Rev Respir Dis 1993;147:15261530. 15. Oliver Z, Hoffstein V. Predicting effective continuous positive pressure. Chest 2000;117:10611064. 16. Rowley JA, Ghani A, Badr MS. Use of a predicted CPAP equation improves titration success [abstract]. Am J Respir Crit Care Med 2002; 165:A723. 17. Fitzpatrick MF, Alloway CED, Wakeford TM, Maclean AW, Munt PW, Day AG. Can patients with obstructive sleep apnea titrate their own continuous positive airway pressure? Am J Respir Crit Care Med 2003; 167:716722. 18. Stradling JR, Hardinge M, Smith DM. A novel, simplied approach to starting nasal CPAP therapy in OSA. Respir Med 2004;98:155158. nez A, Duran J, Capote F, Monasterio C, Mayos 19. Masa JF, Rubio M, Jime n J, Herna ndez L, Barbe F, et al. Efcacy of CPAP obtained M, Tera by means of automatic and mathematical titration (exploratory data) [abstract]. Eur Respir J 2003;22:93s. F, Amilibia J, Capote F, Duran J, Mangado NG, Jime nez A, Mar n 20. Barbe n J. Diagno stico del s ndrome de JM, Masa F, Montserrat JM, Tera apneas obstructivas durante el suen o. Arch Bronconeumol 1995;31: 460462. 21. 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. 22. American Sleeep Disorders Association. EEG arousals: scoring rules and examples. Sleep 1992;15:174183. 23. Lloberes P, Ballester E, Montserrat JM, Botifoll E, Ramirez A, Reolid A, guez Roisin R. Comparison of manual and automatic Gistau C, Rodr CPAP titration in patients with sleep apnea/hypopnea syndrome. Am J Respir Crit Care Med 1996;154:17551758. ndez L, Dura n J, Farre R, Rubio R, Navajas D, Montser24. Molina M, Herna tica: valoracio n de rat JM. Protocolo para evaluar una CPAP automa n de CPAP o ptima la utilidad del Autoset-T para determinar la presio ndrome de apnea-hipopnea del suen en el s o. Arch Bronconeumol 2003;39:118125. 25. Weaver TE, Laizner AM, Evans LK, Maislin G, Chugh DK, Lyon K, Smith PL, Schwartz AR, Redline S, Pack AI, et al. An instrument to measure functional status outcomes for disorders of excessive sleepiness. Sleep 1997;20:835843. 26. Ware J, Sherbourne C. The Mos 36-item short-form health survey (SF36). I. Conceptual framework and item selection. Med Care 1992;30: 473483. 27. EuroQol Group. EuroQol a new facility for measurement of healthrelated quality of life. Health Policy (New York) 1990;16:199208. 28. Ficker JH, Wiest GH, Lehnert G, Wiest B, Hahn EG. Evaluation of an autoCPAP device for treatment of obstructive sleep apnoea. Thorax 1998;53:643648. 29. Meurice JC, Marc I, Series F. Efcacy of autoCPAP in the treatment of obstructive sleep apnea/hypopnea syndrome. Am J Respir Crit Care Med 1996;153:794798. ` s F, Marc I. Efcacy of automatic continuous positive airway pres30. Serie sure therapy that uses an estimated required pressure in treatment of

although a higher pressure could lead to more secondary effects and perhaps consequences in adherence and compliance of the treatment. Most of the quality-of-life tests showed less improvement in the autoadjusted group than in the standard group (Table 4), probably because baseline quality-of-life tests were better in the autoadjusted group. Moreover, the less specic quality-of-life tests showed fewer improvements. Out of the 466 patients initially included in the study, 45 dropped out during the study and 106 were excluded. The main reason for exclusion was the presence of disorders preventing the evaluation of the quality of life. The study was open to the researchers and blinded to the technicians who set the questionnaires and analyzed the polysomnographies. This could have introduced a potential bias. However, a bias is highly unlikely given that the study was performed at 10 centers, and given that most of the questionnaires of the study were self-administered and that information to the patients about the CPAP treatment was standardized. In conclusion, the results of this study suggest that CPAP titration can be performed with a home-autoadjusted CPAP device or with a predicted formula with domiciliary adjustment. Full polysomnography is not the only method that can adequately adjust the CPAP pressure (Figure 2). These procedures could lead to considerable savings in cost and to signicant reductions in the waiting lists.
Conflict of Interest Statement : J.F.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; A.J. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; J.D. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; F.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; C.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; M.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; J.T. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; L.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; F.B. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; A.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; M.R. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; J.M.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Acknowledgment : The authors are indebted to Dr. D. O. Rodenstein and GlaxoSmithKline for their technical help and to Vero nica Rodr guez for the translation and preparation of the manuscript.

References
1. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Badr S. The occurrence of sleep-disordered breathing among middle aged adults. N Engl J Med 1993;328:12301235. 2. Loube DI, Gay PC, Strohl KP, Pack AI, White DP, Collop NA. Indications for positive airway pressure treatment for adult obstructive sleep apnea patients: a consensus statement. Chest 1999;115:863866. 3. Statement ATS. Indications and standards of use of nasal continuous positive airway pressure (CPAP) in sleep apnea syndromes. Am J Respir Crit Care Med 1994;150:17381745. ndez L, Carrasco E, de Pablo J, Fornas C, 4. Ballester E, Badia JR, Herna guez Roisin R, Montserrat JM. Evidence of the effectiveness of Rodr continuous positive airway pressure in the treatment of sleep apnea/ ndrome. Am J Respir Crit Care Med 1999;159:495501. hypopnea s 5. Monasterio C, Vidal S, Duran J, Ferrer M, Carmona C, Barbe F, Mayos M, Gonzalez-Mangado N, Juncadella M, Navarro A, et al. Effectiveness of continuous positive airway pressure in mild sleep apnea-hypopnea syndrome. Am J Respir Crit Care Med 2001;164:939943. F, Mayoralas LR, Dura n J, Masa JF, Maimo A, Montserrat JM, 6. Barbe Monasterio C, Bosch M, Ladaria A, Rubio M, et al. Treatment with continuous positive airway pressure is not effective in patients with sleep apnea but no daytime sleepiness. Ann Intern Med 2001;134:1015 1023.

1224

AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 170 2004
sleep-disordered breathing, sleep apnea, and hypertension in a large community-based study. Sleep Heart Health Study. JAMA 2000;283: 18291836. Duran 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;163:685689. Masa JF, Rubio M, Findley LJ, and the cooperative group. Habitually sleepy drivers have a high frequency of automobile crashes associated with respiratory disorders during sleep. Am J Respir Crit Care Med 2000;162:14071412. F, Perica s J, Mun JM, Agust AGN. AutomoBarbe oz A, Findley L, Anto bile accidents in patients with sleep apnea syndrome. Am J Respir Crit Care Med 1998;158:1822. n J, Jime nez A, Cordero J. The association between sleep apnea Tera and the risk of trafc accidents. N Engl J Med 1999;340:847851.

31.

32.

33.

34.

the obstructive sleep apnea syndrome. Ann Intern Med 1997;127:588 595. R, Montserrat JM, Rigau J, Trepat X, Pinto P, Navajas D. Response Farre of automatic continuous positive airway pressure devices to different sleep breathing patterns. Am J Respir Crit Care Med 2002;166:469473. Kessler R, Weitzenblum E, Chaouat A, Iamandi C, Alliotte T. Evaluation of unattended titration to determine therapeutic continuous positive airway pressure in patients with obstructive sleep apnea. Chest 2003; 123:704710. Peppard PE, Young T, Palta M, Skatrud J. Prospective study of the association between sleep-disordered breathing and hypertension. N Engl J Med 2000;342:13781384. Nieto FJ, Young TB, Lind BK, Shahar E, Samet JM, Redline S, DAgostino RB, Newman AB, Lebowitz MD, Pickering TG. Association of

35. 36.

37. 38.