Monitoring Breathing Rate at Home Allows Early Identification of COPD Exacerbations

CHEST Monitoring Breathing Rate at Home Allows Early Identification of COPD Exacerbations Original Research Aina M. Yañez, PhD ; Dolores Guerrero, MSc ; Rigoberto Pérez de Alejo, PhD ; Francisco Garcia-Rio, MD ; Jose Luis Alvarez-Sala, MD ; Miriam Calle-Rubio, MD; Rosa Malo de Molina, MD ; Manuel Valle Falcones, MD ; Piedad Ussetti, MD ; Jaume Sauleda, MD ; Enrique Zamora García, MD ; Jose Miguel Rodríguez-González-Moro, MD ; Mercedes Franco Gay, BSc ; Maties Torrent, MD ; and Alvar Agustí, MD COPD Background: Respiratory frequency increases during exacerbations of COPD (ECOPD). We hypothesized that this increase can be detected at home before ECOPD hospitalization. Methods: To test this hypothesis, respiratory frequency was monitored at home daily for 3 months in 89 patients with COPD (FEV 1, 42.3% 14.0%; reference) who were receiving domiciliary oxygen therapy (9.6 4.0 h/d). Results: During follow-up, 30 patients (33.7%) required hospitalization because of ECOPD. In 21 of them (70%), mean respiratory frequency increased (vs baseline) during the 5 days that preceded it (from 15.2 4.3/min to 19.1 5.9/min, P,.05). This was not the case in patients without ECOPD (16.1 4.8/min vs 15.9 4.9/min). Receiver operating characteristic analysis showed that 24 h before hospitalization, a mean increase of 4.4/min (30% from baseline) provided the best combination of sensitivity (66%) and specificity (93%) (area under the curve [AUC] 5 0.79, P,.05). Two days before hospitalization, a mean increase of 2.3/min (15% change from baseline) was associated with a sensitivity of 72% and a specificity of 77% (AUC 5 0.76, P,.05). Conclusions: Respiratory frequency can be monitored daily at home in patients with COPD receiving domiciliary oxygen therapy. In these patients, breathing rate increases significantly days before they require hospitalization because of ECOPD. This may offer a window of opportunity for early intervention. CHEST 2012; 142(6):1524 1529 Abbreviations: ECOPD 5 exacerbations of COPD; ROC 5 receiver operating characteristic COPD is a major and increasing public health problem.1 Patients often suffer episodes of exacerbation of COPD (ECOPD) characterized by a sustained worsening of the patient s condition, from the stable state and beyond normal day-to-day variations, that is acute in onset and necessitates a change in regular medication in a patient with underlying COPD. 2 ECOPD constitutes one of the most important causes of morbidity, mortality, and health-care cost in COPD. 3,4 Therefore, management strategies aimed at reducing the frequency and/or severity of ECOPD are of paramount importance for health-care systems around the world. One of the most typical physiologic features of ECOPD is a change in the ventilatory pattern of the patient characterized by increased respiratory frequency and reduced tidal volume. 5-7 Interestingly, this change is associated with the severity of ECOPD, as indicated by its relationship with the duration of hospitalization and associated mortality. 8-10 Several studies have now shown that the increase in dyspnea and/or cough that characterize ECOPD begin about 5 days before the patient is actually seen by a doctor and ECOPD is diagnosed. 7,11 Here, we hypothesized that the increase in respiratory frequency that occurs during ECOPD 5-7 may also be detected within a similar time frame. Since early detection and treatment of ECOPD reduces the risk of hospitalization and enhances clinical recovery, 12 this may serve as a warning indicator that an ECOPD is developing and offer a window of opportunity for therapeutic interventions aimed at preventing or reducing the severity 1524 Original Research

of a full-blown episode of ECOPD. To test this hypothesis, we monitored the respiratory frequency of a cohort of patients with severe COPD who required treatment with domiciliary oxygen therapy for 3 months or until they had an episode of ECOPD that required hospitalization. Study Design and Ethics Materials and Methods This prospective cohort study was conducted in six tertiary referral hospitals in Spain, five in Madrid (Hospital La Paz-Instituto de Investigación La Paz, Hospital La Princesa, Hospital Puerta de Hierro, Hospital Gregorio Marañón, and Hospital Clínico San Carlos), and one in Palma de Mallorca (Hospital Universitari Son Espases). After the patient signed the informed consent, being aware of its nature and goals, respiratory frequency was monitored at home (see the Study Design and Ethics section) for 3 months or until the patient required hospitalization because of ECOPD. Changes in respiratory frequency during the days that preceded ECOPD were compared with stable clinical con ditions. The ethics committees of all participating hospitals approved the study (IB 1103/09 PI). Patient Characteristics Patients were recruited from the respiratory outpatient clinics of the participating hospitals according to the following inclusion criteria: (1) a diagnosis of COPD, 13 (2) need of long-term home oxygen,13 (3) age. 60 years, (4) two or more hospitalization episodes because of ECOPD during the year prior to inclusion, 14 and (5) being clinically stable at the time of recruitment and not having needed a change in treatment and/or hospitalization because of ECOPD during the last 8 weeks. Exclusion criteria included ter- Manuscript received October 25, 2011; revision accepted May 30, 2012. Affiliations: From the Fundación de Investigación Sanitaria Illes Balears (FISIB) (Dr Yañez and Ms Guerrero), Palma de Mallorca; Airproducts Sud Europa (Dr Pérez de Alejo and Ms Franco Gay), Madrid; Servicio de Neumología (Dr Garcia-Rio), Hospital Universitario La Paz, Instituto de Investigación La Paz (IdiPAZ), Madrid; Servicio de Neumología (Drs Alvarez-Sala and Calle-Rubio), Hospital Clínico San Carlos, Facultad de Medicina, Universidad Complutense, Madrid; Hospital Universitario Puerta de Hierro (Drs Malo de Molina, Valle Falcones, and Ussetti) Majadahonda, Madrid; Servei Pneumologia (Dr Sauleda), Hospital Universitari Son Espases, Palma Mallorca; Servicio de Neumología (Dr Zamora García), Hospital de La Princesa, Madrid; Servicio de Neumología (Dr Rodríguez-González-Moro), Hospital Gregorio Marañón, Madrid; Ib-salut (Dr Torrent), Área de Salut de Menorca, Menorca; and the Thorax Institute (Dr Agustí), Hospital Clínic, Institut d Investigacions Biomediques Agustí Pi i Sunyer (IDIBAPS), Universitat Barcelona, CIBER Enfermedades Respiratorias (CIBERES), and Fundación de Investigación Sanitaria Illes Balears (FISIB) Mallorca, Spain. Funding/Support: This study was supported by Air Products Sud Europa, Madrid, Spain. Correspondence to: Aina Yañez, PhD, Fundación de Investigación Sanitaria Illes Balears, Edificio S. Hospital Universitario Son Espases, Carretera de Valldemossa, 79-07010 Palma de Mallorca, Spain; e-mail: aina.yanez@caubet-cimera.es 2012 American College of Chest Physicians. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: 10.1378/chest.11-2728 minal clinical status (defined as a life expectancy 6 months, determined by the attending physician) and residence outside the study area. In all participants, sociodemographic and clinical variables were recorded at recruitment using questionnaires that included age, sex, weight, height, date of COPD diagnosis, date of last hospitalization because of ECOPD, number of hospitalizations because of ECOPD during the last year, forced spirometric data from clinical records, data of oxygen therapy prescription, concomitant medications, alcohol consumption, and comorbidities. Domiciliary Breathing Rate Monitoring To monitor respiratory frequency, we used the VisionOx monitor (Airproducts sud Europa), which was installed in the domiciliary oxygen supply system that the patient already had at home and whose accuracy had been validated before. 15 Device failure does not interrupt oxygen supply to the patient in any case. This monitor has two sensors: one detecting pressure changes in the oxygen line due to the breathing efforts of the patient and the other that measure the oxygen flow. From the signal of the first sensor, the system calculates and records the mean respiratory rate of the patient and the total number of hours of use of the oxygen supply system. Figure 1 shows the mean respiratory frequency of a representative patient during 5 consecutive days when the patient was clinically stable (left), and during the 5 days that preceded hospitalization because of ECOPD (right). For representative purposes, in Figure 1 respiratory frequency was averaged in three different time periods (8:00 am to 4:00 pm, 4:00 pm to 12:00 am, and 12:00 am to 8:00 am ). For analysis, however, the mean daily respiratory frequency value was used. To guarantee a minimum respiratory rate recording time for the analysis of the basal and prehospitalization periods, we arbitrarily decided to exclude record periods shorter than 4 h/d. A respiratory technician installed the monitor at the beginning of the study, and 1 and 3 months later visited participants at home to extract the data from the internal memory of the system. ECOPD Hospitalizations During Follow-up The diagnosis of ECOPD and the decision to hospitalize the patient was made by the attending physician at the ED, who was not one of the investigators and was blinded to the respiratory rate results. Patients were contacted by telephone by a trained clinical research assistant at months 1, 2, and 3 after recruitment and were asked about potential hospitalizations because of ECOPD. Whenever this was identified, hospital, data of admission, and reason for hospitalization were recorded using standardized questionnaires. The attending pulmonologist of each patient, blinded to the respiratory results, reviewed all available information and validated the episode of ECOPD hospitalization. Sample Size Calculation Sample size was calculated on the following assumptions: (1) mean respiratory frequency would increase by 20% during ECOPD 5-7 ; (2) this change would occur during the 5 days that preceded hospitalization because of ECOPD in at least 70% of the patients; (3) given the exploratory nature of our study, we needed at least 30 patients hospitalized because of COPD to test our hypothesis; (4) we deliberately recruited patients with frequent exacerbations (two or more per year), since the risk of having ECOPD was increased in these patients, 14 and we followed them for 3 months; and (5) we may lose 10% of data during follow-up. Accepting an a error of 0.05 and a b error of 0.1, we journal.publications.chestnet.org CHEST / 142 / 6 / DECEMBER 2012 1525

A P value,.05 was considered statistically significant. All analyses were performed using SPSS, version 15 (IBM). Results Figure 1. Individual time-series of mean breathing rate of three different day-time periods (8:00 am to 4:00 pm, 4:00 pm to 12:00 am, and 12:00 am to 8:00 am) in a patient who required hospitalization because of ECOPD during follow-up. An upward trend in respiratory frequency occurred prior to exacerbation. For further explanation, see the Results and Discussion sections. ECOPD 5 exacerbation of COPD. estimated that we needed to recruit a minimum of 120 patients to test our hypothesis. Statistical Analysis Our primary analysis used a three-step strategy. First, we analyzed individual changes in mean respiratory frequency using time series analysis of breathing rate in each patient who required ECOPD hospitalization during follow-up. This time series included 5 baseline days and the 5 days that preceded hospitalization. Baseline was defined as the first 5 consecutive follow-up days with valid recorded data and a minimum compliance with oxygen therapy of 4 h/d, before any exacerbation had occurred. Individual changes were analyzed using the Young C statistic, which is appropriate for the study of chang ing trends in short series with a small number of measures. 16 Second, we analyzed the change in mean respiratory frequency between the 5 baseline days and the 5 days that preceded hospitalization of all individual time series considered together using analysis of variance for repeated measures. Third, we analyzed the discriminating power of the change in the mean respiratory frequency to predict ECOPD hospitalizations using receiver operating characteristic (ROC) curves in two different potentially relevant clinical scenarios (increase of breathing rate from baseline to 24 or 48 h before hospitalization). In patients who were not hospitalized because of ECOPD during follow-up, baseline was defined as described here, but a second period of 5 consecutive days was randomly selected for analysis from the remaining monitored follow-up days. To this end, 1 follow-up day (after the baseline period) was randomly selected, and the next 4 consecutive days were analyzed. If compliance with oxygen therapy was lower than the minimum requested (4 h/d) in one or more of these 5 days, then another follow-up day was selected randomly and the procedure was repeated. Secondary analysis of our data included the following: (1) the comparison of the demographic and clinical characteristics of patients hospitalized because of ECOPD vs those who did not exacerbate, using the t test for continuous variables (Mann- Whitney in absence of normality) and the x 2 test for categorical variables; and, (2) the relationship between changes in the respiratory frequency and days of hospitalization ( x 2 ). Results are presented as mean SD, or percentage (95% CI), as appropriate. Between October 2009 and March 2010, 126 patients were recruited. Thirty-seven of them were excluded from analysis for different reasons: 12 because of technical problems with the monitor resulting in insufficient data for analysis and 25 because of poor compliance with oxygen therapy so that they did not have enough data to calculate both the baseline and random periods. Hence, a total of 89 patients were finally included in the analysis. Their mean age was 76.2 6.5 years, 74% of them were men, and the number of ECOPD hospitalizations during the previous year was 2.4 1.2 (Table 1 ). Mean breathing rate at baseline was 15.6 4.8/min. During follow-up, patients had been receiving domiciliary oxygen therapy for 9.6 4.0 h/d on average. As anticipated, 30 patients required ECOPD hospitalization during the 3 months of follow-up. The baseline sociodemographic and clinical characteristics of these patients were similar to those of patients not requiring hospitalization ( Table 1 ). Mean length of hospital stay was 6.0 1.9 days (range 2-11 days). The majority of patients (87%) were hospitalized 5 days. Figure 1 shows a time series analysis of mean respiratory frequency in one patient who required hospitalization because of ECOPD during follow-up wherein, compared with baseline (clinical stability), mean respiratory frequency increased progressively during the 5 days that preceded hospitalization. A similar upward trend was observed in 21 of the 30 patients (70%) (C statistic, P,.05). Analysis of variance confirmed that the mean respiratory frequency was significantly higher during the 5 days that preceded hospitalization than at baseline (19.1 5.9/min vs 15.2 4.3/min, respectively). This was not the case in patients who did not require hospitalization because of ECOPD (16.1 4.8/min vs 15.9 4.9/min, baseline vs 5 random days, respectively) ( Figs 2A, 2B, respectively). ROC analysis comparing the mean daily respiratory frequency of the 30 patients who required hospitalization because of ECOPD with that of those 59 who did not showed that the area under the curve was 0.79 ( P,.05) and 0.76 ( P,.05) for the absolute increase of mean breathing rate between baseline and that of 24 h or 48 h before hospitalization, respectively ( Fig 3 ). Numerically, an increase of 4.4/min provided the best combination of sensitivity and specificity (66%; 95% CI, 48%-80%; and 93%; 95% CI, 84%-97%, respectively) with a positive predictive value of 88% and a negative predictive value of 84% (95% CI, 69-96 and 73-91, 1526 Original Research

Table 1 Clinical and Sociodemographic Characteristics of the Sample Characteristic Total (N 5 89) Exacerbated (n 5 30) Nonexacerbated (n 5 59) Men 66 (74.2) 23 (76.7) 43 (72.9) Age, y 76.2 6.5 76.3 6.2 76.1 6.5 Weight, kg 76.3 15.1 76.9 16.7 76.0 15.3 Height, m 1.61 0.09 1.61 0.09 1.61 0.09 BMI, kg/m 2 29.4 5.8 29.7 6.4 29.1 5.4 Years since COPD diagnosis 8.8 5.6 9.4 5.9 8.4 4.8 Number of hospitalizations during last year 2.4 1.2 2.5 1.6 2.3 1.0 FEV 1, L 1.0 0.4 1.1 0.4 1.0 0.4 FEV 1, % 42.3 14.0 44.0 13.8 40.7 14.4 FVC, L 2.1 0.6 2.2 0.7 2.1 0.6 FVC, % 60.9 15.6 61.5 15.5 59.3 15.6 FEV 1 /FVC 0.5 0.1 0.5 0.1 0.5 0.1 Years on domiciliary oxygen therapy 3.3 2.5 3.3 2.5 3.2 2.4 Oxygen flow prescribed 1.9 0.4 1.9 0.4 1.9 0.3 H/d oxygen therapy prescribed 16.3 2.6 15.7 1.7 16.6 2.2 Mean baseline h/d oxygen used 9.6 4.0 10.4 4.7 9.4 3.5 Mean baseline respiratory rate, per min 15.6 4.8 15.2 4.2 15.9 4.9 Respiratory drugs Salbutamol 26 (51.7) 13 (43.3) 33 (55.9) Tiotropium 70 (78.6) 25 (83.3) 45 (76.3) Salmeterol 9 (10.1) 3 (10) 6 (10.2) Fluticasone 1 salmeterol 63 (70.8) 20 (69.0) 43 (72.9) Ipratropium bromide 5 (5.6) 1 (3.3) 4 (6.8) Fluticasone 7 (7.9) 3 (10) 4 (6.8) Budesonide 5 (5.6) 0 (0) 5 (8.5) Theophylline 11 (12.4) 2 (6.7) 9 (15.3) Psychoactive drugs 18 (20.2) 5 (16.7) 13 (22.0) Alcohol consumption 5 (5.6) 2 (7.1) 3 (5.1) Comorbidity Myocardial infarction 9 (10.1) 3 (10.0) 6 (10.2) Diabetes 27 (30.3) 13 (43.3) 14 (23.7) Hypertension 55 (61.8) 19 (63.3) 36 (61.0) Hypercholesterolemia 17 (19.1) 5 (16.7) 12 (20.3) Data are presented as mean SD or No. (%). respectively) for the former (24 h before hospitalization). Whereas an increase of 2.3/min was the best combination (72%; 95% CI, 55%-84%; and 77%; 95% CI, 65%-86%, respectively) for the latter (48 h before hospitalization), with a positive predictive value of 64% and a negative predictive value of 84% (95% CI, 48-76 and 72-91, respectively). Given that the mean baseline breathing rate determined in these Figure 2. A, Box-plot of mean breathing rate in patients (n 5 30) who required hospitalization because of ECOPD during follow up. B, Box-plot of mean breathing rate in patients (n 5 59) who did not require hospitalization because of ECOPD during follow up. For further explanations, see the Results and Discussion sections. See Figure 1 legend for expansion of abbreviation. journal.publications.chestnet.org CHEST / 142 / 6 / DECEMBER 2012 1527

Figure 3. A, Receiver operating characteristic curve for the prediction of ECOPD hospitalization 24 h before it occurred (n 5 89). B, Receiver operating characteristic curve for the prediction of ECOPD hospitalization 48 h before it occurred (n 5 89). For further explanations, see the Results and Discussion sections. See Figure 1 legend for expansion of abbreviation. patients was about 15/min, these values (4.4/min and 2.3/min) represent an increase of about 30% and 15%, respectively, from baseline, which is in keeping with our a priori working hypothesis. Finally, we observed that there was a significant relationship between the increase in mean breathing rate before hospitalization and length of hospital stay ( x 2, P 5.02). Discussion This study explores a novel and potentially clinically relevant hypothesis: The increase in respiratory frequency that characterizes the episodes of ECOPD can be detected days before hospitalization. Our results confirm this hypothesis and open the possibility of exploring, in a larger and formally randomized clinical trial, the potential benefits of an early therapeutic intervention during this window of opportunity on relevant clinical outcomes, such as need for hospitalization, length of hospital stay, and/or mortality. To our knowledge, no previous study has explored this hypothesis. However, other studies have shown that there is an increase in dyspnea, sore throat, cough, and symptoms of common cold before the onset of ECOPD. 7 Interestingly, 60% of patients report increased shortness of breath during the days before a formal diagnosis of ECOPD is established, 7 a figure that is remarkably similar to the changes in respiratory frequency observed in our study (70%). In a later study, this same group of researchers showed that the median (interquartile range) time between exacerbation onset and treatment initiation was 3.69 (2.0-5.57) days, a period of time well within the temporal interval studied here. 12 Some other authors, however, have indicated that daily monitoring of patient s symptoms was too variable to be useful in clinical management. 11 By contrast, our results indicate that the automatic monitoring of breathing rate in COPD provides an objective measurement to predict the onset of ECOPD and that a persistent increase of about 15% to 30% of the baseline respiratory rate should raise concern. Given that prompt treatment improves exacerbation recovery, reduces risks of hospitalization, and is associated with a better health-related quality of life, 12 these results open the possibility of exploring prospectively in a specifically designed randomized clinical trial if a therapeutic intervention triggered by this system avoids and/or reduces the frequency or severity of ECOPD. Some limitations of our study deserve comment. Although sample size was specifically calculated for the purposes of this pilot study, it is clearly limited to allow the generalization of our results. Likewise, the monitoring system used in this study was installed in the system providing domiciliary oxygen at home. Thus, our results apply to a very specific and narrow segment of the population of patients with COPD at large. Other technologic alternatives to monitor respiratory frequency in patients with COPD who do not require domiciliary oxygen therapy, such as the use of ribcage inductive plethysmography, 17 also used in patients with sleep apnea, 18 will need to be explored. In summary, the results of this pilot study show that, first, it is possible to monitor the breathing rate of patients with COPD receiving domiciliary oxygen therapy and, second and more importantly, that mean respiratory frequency increases by about 15% to 30% at least 48 h before the patient requires hospitalization because of ECOPD. This may offer an opportunity for early intervention. Acknowledgments Author contributions: Dr Yáñez confirms that the study objectives and procedures are honestly disclosed. Dr Yañez: contributed to conception and design of the original idea, data analysis, and drafting of the Ms Guerrero: contributed to statistical support, data analysis, and drafting of the Dr Pérez de Alejo: contributed to data collection, conception Dr Garcia-Rio: contributed to data collection, conception and design of the original idea, data analysis, and drafting of the Dr Alvarez-Sala: contributed to data collection, conception and design of the original idea, data analysis, and drafting of the Dr Calle-Rubio: contributed to data collection, conception and design of the original idea, data analysis, and drafting of the Dr Malo de Molina: contributed to data collection, conception Dr Valle Falcones: contributed to data collection, conception Dr Ussetti: contributed to data collection, conception and design of the original idea, data analysis, and drafting of the Dr Sauleda: contributed to data collection, conception and design of the original idea, data analysis, and drafting of the 1528 Original Research

Dr Zamora García: contributed to data collection, conception Dr Rodríguez-Gonzalez-Moro: contributed to data collection, conception and design of the original idea, data analysis, and drafting of the Ms Franco Gay: contributed to conception and design of the original idea, data analysis, and drafting of the Dr Torrent: contributed to conception and design of the original idea, data analysis, and drafting of the Dr Agustí: contributed to conception and design of the original idea, providing guidance and drafting of the Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest: Dr Pérez de Alejo and Ms Franco Gay are employees of Air Products. Drs Yáñez, Garcia-Rio, Álvarez-Sala, Calle-Rubio, Malo de Molina, Valle Falcones, Ussetti, Sauleda, Zamora García, Rodríguez-Gonzalez- Moro, Torrent, and Agustí, and Ms Guerrero have reported that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Role of sponsors : The sponsor had no role in the design of the study, the collection and analysis of the data, or in the preparation of the Other contributions: We thank participants for their willingness to contribute to medical research. We also thank Trinidad Fernandez and Sara Bueno for help with data entry and processing. References 1. Buist AS, McBurnie MA, Vollmer WM, et al ; BOLD Collaborative Research Group. International variation in the prevalence of COPD (the BOLD Study): a population-based prevalence study. Lancet. 2007 ;370(9589):741-750. 2. Rodriguez-Roisin R. Toward a consensus definition for COPD exacerbations. Chest. 2000 ;117(5 suppl 2 ):398S-401S. 3. Pauwels RA, Buist AS, Calverley PM, Jenkins CR, Hurd SS ; GOLD Scientific Committee. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease. NHLBI/WHO Global Initiative for Chronic Obstructive Lung Disease (GOLD) Workshop summary. Am J Respir Crit Care Med. 2001 ;163(5):1256-1276. 4. Wedzicha JA, Seemungal TA. COPD exacerbations: defining their cause and prevention. Lancet. 2007 ;370(9589):786-796. 5. Agustí AG, Carrera M, Barbé F, Muñoz A, Togores B. Oxygen therapy during exacerbations of chronic obstructive pulmonary disease. Eur Respir J. 1999 ;14(4):934-939. 6. Carrera M, Marín JM, Antón A, et al. A controlled trial of noninvasive ventilation for chronic obstructive pulmonary disease exacerbations. J Crit Care. 2009 ;24(3):473.e7-14. 7. Seemungal TA, Donaldson GC, Bhowmik A, Jeffries DJ, Wedzicha JA. Time course and recovery of exacerbations in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2000 ;161 (5 ):1608-1613. 8. Chang CL, Sullivan GD, Karalus NC, Mills GD, McLachlan JD, Hancox RJ. Predicting early mortality in acute exacerbation of chronic obstructive pulmonary disease using the CURB65 score. Respirology. 2011 ;16 (1 ):146-151. 9. de la Iglesia F, Valiño P, Pita S, et al. Factors predicting a hospital stay of over 3 days in patients with acute exacerbation of chronic obstructive pulmonary disease. J Intern Med. 2002 ;251 (6 ):500-507. 10. Franciosi LG, Page CP, Celli BR, et al. Markers of exacerbation severity in chronic obstructive pulmonary disease. Respir Res. 2006 ;7 :74. 11. Calverley P, Pauwels Dagger R, Löfdahl CG, et al. Relationship between respiratory symptoms and medical treatment in exacerbations of COPD. Eur Respir J. 2005 ;26 (3 ): 406-413. 12. Wilkinson TM, Donaldson GC, Hurst JR, Seemungal TA, Wedzicha JA. Early therapy improves outcomes of exacerbations of chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 2004 ;169 (12 ):1298-1303. 13. Rabe KF, Hurd S, Anzueto A, et al ; Global Initiative for Chronic Obstructive Lung Disease. Global strategy for the diagnosis, management, and prevention of chronic obstructive pulmonary disease: GOLD executive summary. Am J Respir Crit Care Med. 2007 ;176 (6 ):532-555. 14. Hurst JR, Vestbo J, Anzueto A, et al ; Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints (ECLIPSE) Investigators. Susceptibility to exacerbation in chronic obstructive pulmonary disease. N Engl J Med. 2010 ; 363 (12 ):1128-1138. 15. Pérez de Alejo R, Esteban R, Sacristán JC, Peces-Barba G. Monitor de cumplimiento de oxigenoterapia domiciliaria Visionox. Evaluación de eficacia y fiabilidad [abstract]. Arch Bronconeumol. 2009 ;45(especial congreso):97. 16. Tryon WW. A simplified time-series analysis for evaluating treatment interventions. J Appl Behav Anal. 1982 ;15 (3 ):423-429. 17. Balleza M, Calaf N, Feixas T, et al. Measuring breathing pattern in patients with chronic obstructive pulmonary disease by electrical impedance tomography [in Spanish]. Arch Bronconeumol. 2009 ;45 (7 ):320-324. 18. Cantineau JP, Escourrou P, Sartene R, Gaultier C, Goldman M. Accuracy of respiratory inductive plethysmography during wakefulness and sleep in patients with obstructive sleep apnea. Chest. 1992 ;102 (4 ):1145-1151. journal.publications.chestnet.org CHEST / 142 / 6 / DECEMBER 2012 1529