ORIGINAL ARTICLE Heart Failure. Methods

Transcription

1 ORIGINAL ARTICLE Heart Failure Circ J 2009; 73: Effectiveness of Nocturnal Home Oxygen Therapy to Improve Exercise Capacity, Cardiac Function and Cardiac Sympathetic Nerve Activity in Patients With Chronic Heart Failure and Central Sleep Apnea Takuji Toyama, MD; Ryotaro Seki, MD; Shu Kasama, MD*; Naoki Isobe, MD; Shigeki Sakurai, MD; Hitoshi Adachi, MD; Hiroshi Hoshizaki, MD; Shigeru Oshima, MD; Koichi Taniguchi, MD Background Central sleep apnea, often found in patients with chronic heart failure (CHF), has a high risk of poor prognosis. Methods and Results This study involved 20 patients with CHF (left ventricular ejection fraction (LVEF) <45%, M/F =19/1, age 65±10 years) and an apnea hypopnea index (AHI) >5 times/h who were divided into 2 groups: 10 patients treated with nocturnal home oxygen therapy (HOT) and 10 patients without HOT (non-hot). All patients had dilated cardiomyopathy and underwent overnight polysomnography, cardiopulmonary exercise testing, and nuclear cardiac examinations to evaluate AHI, exercise capacity according to the specific activity scale and oxygen uptake at anaerobic threshold and peak exercise (peak V O2). Cardiac function according to 99m Tc-MIBI QGS, and the total defect score (TDS), H/M ratio and the washout rate (WR) on 123 I-metaiodobenzylguanidine (MIBG) imaging were calculated for all patients. As compared with the non-hot group, the HOT group demonstrated a greater reduction in AHI (26.1±9.1 to 5.1±3.4), 123 I-MIBG TDS (31±8 to 25±9), and 123 I- MIBG WR (48±8% to 41±5%) and a greater increase in the specific activity scale (4.0±0.9 to 5.8±1.2 Mets), peak V O2 (16.0±3.8 to 18.3±4.7 ml min 1 kg 1 ), and LVEF (27±9% to 37±10%). Conclusions HOT improves exercise capacity, cardiac function, and cardiac sympathetic nerve activity in patients with CHF and central sleep apnea. (Circ J 2009; 73: ) Key Words: Dilated cardiomyopathy; Oxygen; Sleep apnea; Sympathetic nervous system (Received April 19, 2007; revised manuscript received September 1, 2008; accepted October 14, 2008; released online December 27, 2008) Gunma Prefectural Cardiovascular Center, *Department of Cardiovascular Medicine, Gunma University School of Medicine, Maebashi, Japan Mailing address: Takuji Toyama, MD, Gunma Prefectural Cardiovascular Center, 3-12 Kameizumi-machi, Maebashi , Japan. toyama.t@cvc.pref.gunma.jp All rights are reserved to the Japanese Circulation Society. For permissions, please cj@j-circ.or.jp The prognosis of patients with chronic heart failure (CHF) remains poor, despite the introduction of many effective drugs in the clinical setting, because one of the cardiac pathophysiologic abnormalities that worsens heart failure is activation of the sympathetic nervous system. 1 Central sleep apnea (CSA) occurs in patients with left ventricular (LV) dysfunction and is a poor prognostic sign associated with a higher mortality than in patients with CHF alone. 2,3 CSA causes an increase in sympathetic nervous system activity, which maintains afterload at a high level, 4 leading to a further reduction in the LV ejection fraction (LVEF). 123 I-metaiodobenzylguanidine (MIBG) imaging, which is a useful indicator of cardiac sympathetic nerve activity, and thus poor prognosis, is altered in patients with CHF compared with normal subjects. 1,5,6 The present study was performed to determine the effectiveness of nocturnal home oxygen therapy (HOT) on exercise capacity, cardiac sympathetic nerve activity, and cardiac function in patients with CHF and CSA. Methods Study Population Ambulatory patients aged over 20 years with clinical evidence of CHF were enrolled if they had the following criteria: (1) symptomatic and New York Heart Association (NYHA) class II with dilated cardiomyopathy (DCM) despite optimal medication for at least 2 weeks prior to entry, (2) at least 1 episode of decompensated heart failure requiring short-term hospitalization, (3) <45% LVEF determined by quantitative gated single-photon emission computed tomography (SPECT) within the period of the diagnostic sleep study, and (4) at least 5 episodes of apnea and hypopnea per hour of sleep, 50% or more of which were central in origin. 7 Patients with predominantly obstructive sleep apnea, pregnancy, ischemic or valvular heart disease, and significant renal, neurological, or respiratory disease were excluded. From 2000 to 2005 years, we enrolled 20 consecutive patients (M/F 19/1, mean age: 65±10 years [range 42 78]) with CHF on conventional treatment either alone or with HOT. The conventional treatment for heart failure included diuretics, angiotensin-converting enzyme inhibitors or angiotensin-ii blockers and β-blockers (Table 1). We divided the 20 patients into 2 groups: 10 patients treated with HOT

2 300 TOYAMA T et al. Table 1 Baseline Characteristics, Cardiac Symptoms and Specific Activity Scale Data Case no. Gender Age NYHA SAS (METs) β-blocker ACEI/ARB Rhythm (years) BSL 3Mo BSL 3Mo HOT group 1 M 70 Carvedilol 5 mg Perindopril 2 mg AF M 70 Carvedilol 5 mg Candesartan 4 mg AF M 49 Carvedilol 10 mg Enalapril 10 mg AF M 69 Carvedilol 10 mg SR M 68 Carvedilol 10 mg Candesartan 8 mg SR M 56 Carvedilol 10 mg Perindopril 2 mg SR M 71 Carvedilol 10 mg Losartan 12.5 mg SR M 50 Carvedilol 10 mg Perindopril 2 mg SR M 50 Carvedilol 20 mg Telmisartan 20 mg SR M 73 Carvedilol 5 mg SR Mean 63±10 4± ±1.2** Non-HOT group 1 M 69 Carvedilol 10 mg SR M 58 Carvedilol 10 mg Enalapril 2.5 mg SR M 74 Carvedilol 5 mg Losartan 25 mg SR M 68 Carvedilol 10 mg Candesartan 8 mg SR M 42 Carvedilol 20 mg Candesartan 4 mg SR M 70 Carvedilol 5 mg SR F 66 Carvedilol 10 mg Losartan 25 mg AF M 70 Carvedilol 5 mg AF M 65 Carvedilol 10 mg Losartan 12.5 mg SR M 78 Carvedilol 10 mg Losartan 25 mg SR Mean 66±10 3.8± ±1.0 P value *P<0.01 vs BSL; P value vs the HOT group. ACEI/ARB, angiotensin-converting enzyme inhibitor or angiotensin-ii blocker; NYHA, New York Heart Association; SAS, specific activity scale; BSL, baseline; 3Mo, after 3 months of study; HOT, nocturnal home oxygen therapy; AF, atrial fibrillation; SR, sinus rhythm. Table 2 Polysomnography Data HOT group Non-HOT group P value AHI (events/h) BSL 26.1± ± Mo 5.1±3.4** 21.0±12.0 <0.001 Difference 20.1± ±4.5 <0.001 CSAI (events/h) BSL 9.6± ± Mo 1.5± ± Difference 8.0± ± SO2 <90% (%time) BSL 4.0± ± Mo 0.3± ± Difference 5.6± ± Mini SO2 (%) BSL 80.9± ± Mo 87.8± ±5.4 <0.001 Difference 6.9± ± **P<0.01 vs BSL. AHI, apnea-hypopnea index; CSAI, central sleep apnea index; SO2, arterial oxygen saturation. Other abbreviations see in Table 1. (mean age: 63±10 years [range 49 73]) and 10 patients not receiving HOT (mean age: 66±10 years [range 42 78]). The study protocol, including a randomized control study, was approved by the institutional ethics review broad and written informed consent was given by all patients prior to entry. Study Protocol All patients in the HOT group received overnight nasal 3 L/min oxygen (O2) every night for 3 months. 8 We performed a series of examinations including polysomnography (PG), cardiopulmonary exercise testing (CPET), 123 I-MIBG imaging and 99m Tc-MIBI imaging within the 2 weeks be- fore and after 3 months of this study. At the same time, we estimated cardiac symptoms based on NYHA functional class and exercise capacity using the Specific Activity Scale (SAS). In the HOT group, PG performed after 3 months was while patients were receiving the O2 therapy. PG Patients underwent overnight PG using a computerized system (CompuMedic E-Series Sleep System; CompuMEdic; Abbotsford, Vic., Australia). With this system, the electroencephalogram, electrooculogram, electromyogram, and electrocardiogram are recorded and analyzed automatically. Air flow from the nose or mouth is recorded with a thermister, and thoracic and abdominal wall motion monitored with plethysmography. Arterial oxygen saturation (SaO2%) was measured transcutaneously by pulse oximetry. Hypopnea was defined as a fall in oronasal air flow >50% of baseline for more than 10 s associated with a decrease in SaO2% >3%. Apnea was defined as the absence of oronasal air flow for the same period. The number of episodes of apnea and hypopnea per hour was defined as the apnea hypopnea index (AHI). AHI >5 times/h was considered to indicate significant sleep apnea syndrome. An episode of obstructive apnea was defined as the absence of airflow in the presence of rib-cage and abdominal excursion. An episode of central apnea was defined as the absence of both airflow and rib-cage and abdominal motion. Functional Capacity Functional capacity was scored as the NYHA class and by the SAS, which asked all patients 21 questions at the beginning and after 3 months of this study. 9,10 Cardiopulmonary Exercise Test All patients underwent symptom-limited CPET in the upright position on a calibrated cycle ergometer (CPE2000, MedGraphics, St Paul, MN, USA). Exercise was started with 4 min of warm-up at 20 W, followed by a continuous increase in workload until

3 Effectiveness of Home O2 Therapy 301 Fig 1. Effect of home oxygen therapy in a patient with central sleep apnea, Cheyne-Stokes respiration, low left ventricular ejection fraction of 16% and New York Heart Association functional class III. At baseline (A), there is the typical waxing and waning pattern of respiration followed by central apneas. After 3 months of home oxygen therapy (B), the abnormal respiration patterns have disappeared. termination when the patient complained of exhaustion. We estimated exercise tolerance from the oxygen uptake at the anaerobic threshold (AT V O2) and peak oxygen uptake (peak V O2) I-MIBG and 99m Tc-MIBI Imaging The 123 I-MIBG and 99m Tc-sestamibi (MIBI) isotopes were obtained commercially (Daiichi Radioisotope Laboratories, Chiba, Japan). All medicines, including β-blockers, were withheld on the day of the 123 I-MIBG and 99m Tc-MIBI studies. Patients were injected intravenously with 111 Mbq of 123 I-MIBG while upright and 15 min later anterior planar and SPECT was performed and repeated after 4 h. Immediately after 123 I-MIBG image acquisition, the patients were injected with 720 MBq of 99m Tc-MIBI while upright and then imaged 30 min later. Using the anterior planar delayed 123 I-MIBG images, the heart-to-mediastinum (H/M) activity ratio was obtained using regions of interest positioned over the heart and upper mediastinum. The washout rate (WR) was calculated as: (H/M) early (H/M) delayed / (H/M) early ,13 The myocardial SPECT image set for each patient was divided into 20 segments. The short-axis images at the basal, middle, and apical ventricular levels were divided into 6 segments each. The apical segment of the vertical long-axis image was divided into 2 segments. Regional tracer uptake was scored semiquantitatively using a 4-point scoring system (0= normal uptake; 1= mildly reduced uptake; 2= moderately reduced uptake; 3= severely reduced uptake). The total defect score (TDS) was calculated as the sum of the scores for all 20 segments. Commercially available QGS software (Cedars-Sinai Medical Center, Los Angeles, CA, USA) with a temporal resolution of 16 frames/rr interval was used to create a 3- dimensional surface cinemode display, which was used to calculate the LVEF, LV end-diastolic and end-systolic volumes (LVEDV, LVESV) throughout the cardiac cycle using an automatic edge detection algorithm. 14 Statistical Analysis Statistical analysis was performed with STATVIEW (SAS Institute Inc, Cary, NC, USA) on a Macintosh computer. Specific comparisons of parameters were performed using paired 2-tailed t-tests. Unpaired 2-tailed t-tests were used to make comparisons between 2 groups. All values are expressed as the mean ± SD. Significant differences were defined by P<0.05. Table 3 Cardiopulmonary Exercise Test Case no. AT V O2 Peak V O2 BSL 3Mo BSL 3Mo HOT group Mean 11.2± ±3.1 16± ±4.7* Non-HOT group Mean 11.2± ± ± ±3.8 P value *P<0.05 vs BSL; P value vs the HOT group. ATV O2, anaerobic threshold oxygen uptake; peak V O2, peak oxygen uptake. Other abbreviations see in Table 1. Results AHI and the Change in AHI At baseline, the AHI was comparable between the 2 groups, but after 3 months of HOT (5.1±3.4), it had decreased significantly from the baseline value (26.1±9.1, P<0.01), and was unchanged in the non-hot group (Table 2). The HOT was well tolerated for 3 months with no adverse effects. A representative case is shown in Fig 1. NYHA Functional Class At baseline, the NYHA functional class and SAS data were comparable between groups (Table 1). After 3 months of HOT treatment, the NYHA functional class had improved while remaining unchanged in the non-hot group. Similarly, the SAS improved from the baseline value in the HOT group (4.0±0.9 Mets to 5.8±1.2 Mets, P<0.01), but

4 302 TOYAMA T et al. Table I-MIBG and 99m Tc-MIBI Data Case no. LVEF (%) EDV (ml) ESV (ml) TDS (MIBG) H/M ratio Washout rate (%) TDS (MIBI) BSL 3Mo BSL 3Mo BSL 3Mo BSL 3Mo BSL 3Mo BSL 3Mo BSL 3Mo HOT group Mean 27±9 37±10** 234±70 185±58* 174±71 121±56* 31±8 25±9* 1.69± ±0.22* 49±8 41±5* 7±5 5±4 Non-HOT group Mean 26±8 27±9 241±78 222±72 184±73 170±68 36±10 37±8 1.6± ± ±11 51±11 9±5 9±5 P value *P<0.05 vs BSL; **P<0.01 vs BSL. LVEF, left ventricular ejection fraction; EDV, end-diastolic volume; ESV, end-systolic volume; TDS, total defect score; H/M, heart-to-mediastinum. Fig 2. Representative case from the home oxygen therapy (HOT) group. 123 I-metaiodobenzylguanidine (MIBG) images obtained from short-axis and vertical long-axis reconstructions. Before HOT, 123 I-MIBG uptake was decreased in the inferior to apical walls. After 3 months of HOT, uptake in the same area has improved. The heart-to-mediastinum ratio increased from 1.43 to 1.65 and the left ventricular ejection fraction increased from 16% to 29% after treatment. was unchanged in the non-hot group. AT V O2 and Peak V O2 The baseline AT V O2 and peak V O2 values were comparable between the HOT and non-hot groups (Table 3). At 3 months, the AT V O2 was unchanged in each group, whereas the peak V O2 had improved significantly from 16.0±3.8 to 18.3±4.7 ml min 1 kg 1 in the HOT group (P<0.05), while remaining unchanged in the non-hot group. Moreover the difference in the peak V O2 of the HOT group (2.3± 2.5 ml min 1 kg 1 ) was significantly greater than the non- HOT group ( 0.3±2.0 ml min 1 kg 1, P=0.02). LVEDV, LVESV and LVEF on the QGS Images The baseline LVEDV, LVESV and LVEF values were comparable between the HOT and non-hot groups (Table 4). After 3 months of HOT (37±10%), the LVEF improved significantly from the baseline value (27±9%, P<0.01), but remained unchanged in the non-hot group. Moreover, the difference in the LVEF of the HOT group (9.7±9.1%) was significantly higher than that of non-hot group (1.3±3.3%, P=0.013). In the HOT group, the LVEDV improved significantly from 234±70 ml to 185±58 ml (P<0.05) and the LVESV was also improved significantly from 174±71 ml to 121±56 ml (P<0.05), while remaining unchanged in the non-hot group. TDS and H/M Ratio on the 123 I-MIBG Images The TDS in the HOT group improved significantly from 31±8 to 25±9 (P<0.05), while remaining unchanged in the

5 Effectiveness of Home O2 Therapy 303 Fig 3. Representative case from the nonhome oxygen therapy (HOT) group. 123 I- metaiodobenzylguanidine (MIBG) images obtained from short-axis and vertical longaxis reconstructions. At baseline, 123 I-MIBG uptake was decreased in the inferior to apical walls and was unchanged 3 months later. Both the heart-to-mediastinum ratio and the left ventricular ejection fraction remained unchanged during the 3-month period. non-hot group (Table 4). Moreover the difference in the TDS in the HOT group was significantly greater than in the non-hot group (P=0.006). The H/M ratio in the HOT group improved significantly from 1.69±0.22 to 1.82±0.22 (P<0.05), while remaining unchanged in the non-hot group. Figs 2 and 3 are representative cases from the HOT and non-hot groups, respectively. WR on the 123 I-MIBG Images The WR in the HOT group improved significantly from 49±8% to 41±5% (P<0.05), but was unchanged in the non- HOT group and, moreover, the difference in the WR in the HOT group was significantly greater than in the non-hot group (P=0.038) (Table 4). TDS on the 99m Tc-MIBI Images The TDS did not change significantly in either group (Table 4). Discussion In the present study, 3 months of HOT decreased the AHI significantly in patients with CHF and CSA, whereby functional capacity, exercise tolerance, and LVEF were increased and cardiac sympathetic nerve activity was suppressed. There have been several clinical studies of HOT for CSA. Franklin et al reported oxygen therapy decreased the median number of all central AHI from 33.5 to 5/h of sleep. 15 Similarly, Andreas et al reported that the AHI fell from 26 to 10/h with HOT. 16 However, these previous studies were limited to short-term evaluation within 1 month. The present study confirmed sustained effectiveness of HOT for CSA in CHF for at least for 3 months in terms of functional capacity, cardiac function, and cardiac sympathetic nerve activity, as well as AHI. There have been previous reports of estimating sympathetic nerve activity using plasma or urine norepinephrine concentrations in patients with CHF and CSA. Staniforth et al reported that HOT reduced the frequency of Cheyne- Stokes respiration and reduced overnight urinary noradrenaline excretion in patients with CHF and CSA. 4 They concluded that HOT stabilized Cheyne-Stokes respiration and improved sympathetic nerve activity, but there were no differences in the daytime serum noradrenaline and adrenaline concentrations between the HOT and control groups. Moreover, overnight urinary noradrenaline excretion does not necessarily reflect cardiac sympathetic nerve activity. In our study, we estimated cardiac sympathetic nerve activity using 123 I-MIBG scintigraphy, which is an analog of noradrenaline and has been reported to provide images that reflect cardiac sympathetic function. 5,6 123 I-MIBG uptake is considered useful for the evaluation of the severity of heart failure. 6,12 In addition, the H/M ratio, TDS and WR on 123 I- MIBG can be used to determine prognosis and evaluate therapeutic efficacy. 6,12,17,18 In our study, the improvements in the TDS, H/M ratio and WR on the 123 I-MIBG images for patients receiving HOT were greater than in patients not receiving HOT. These results suggest that the improvement in CSA produced by HOT attenuated sympathetic nerve activity. Because activated sympathetic nerve activity is a poor prognostic sign in CHF, 19 an improvement in such activity can be associated with reduced mortality. 20 HOT in patients with CHF and CSA may have important implications for the treatment of CHF. There has not been a report that HOT directly improves sympathetic nerve activity in patients with CHF and CSA. From our data, HOT can improve CSA in DCM patients with CHF and CSA. The mechanism for the development of CSA is postulated to be upregulation of both central and peripheral 21 chemoreceptors, which is thought to be related to elevated circulating catecholamines. 26 Mansfield et al support the theory that cardiac failure is primarily responsible for the development of CSA in CHF patients, with the increased sympathetic nerve activity and upregulated chemosensitivity being secondary. 27 HOT can also stabilize sympathetic nerve activity, which could promote improvement in cardiac symptoms and function, as well as exercise capacity, in DCM patients with CHF and CSA. Study Limitations First, because of the small number of patients in this study, it is difficult to sufficiently evaluate the effectiveness of HOT in patients with CHF and CSA. Second, the duration of HOT was relatively short. In the future, we need to study the long-term effects of HOT, and compare its effectiveness in a large number of patients.

6 304 TOYAMA T et al. Conclusion HOT exerts sustained effectiveness in improving exercise capacity, cardiac function, and cardiac sympathetic nerve activity in patients with CHF and CSA. References 1. Thomas JA, Marks BH. Plasma norepinephrine in congestive heart failure. Am J Cardiol 1978; 41: Lanfranchi PA, Braghiroli A, Bosimini E, Mazzuero G, Colombo R, Donner CF, et al. Value of nocturnal cheyne-stokes respiration in chronic heart failure. Circulation 1999; 99: Hanly PJ, Zuberi-Khokhar NS. Increased mortality associated with Cheyne-Stokes respiration in patients with congestive heart failure. Am J Respir Crit Care Med 1996; 153: Staniforth AD, Kinnear WJ, Starling R, Hetmanski DJ, Cowley AJ. Effect of oxygen on sleep quality, cognitive function and sympathetic activity in patients with chronic heart failure and Cheyne-Stokes respiration. Eur Heart J 1998; 19: Schofer J, Spielmann R, Schbert A, Weber K, Schluter M. Iodine- 123 metaiodobenzylguanidine scintigraphy: A noninvasive method to demonstrate myocardial adrenergic system disintegrity in patients with idiopathic dilated cardiomyopathy. J Am Coll Cardiol 1988; 12: Merlet P, Valette H, Dubois-Rande JL, Moyse D, Duboc D, Dove P, et al. Prognostic value of cardiac metaiodobenzylguanidine imaging in patients with heart failure. J Nucl Med 1992; 33: American Academy of Sleep Medicine Task Force. Sleep-related breathing disorders in adults: Recommendation for syndrome definition and measurement techniques in clinical research. Sleep 1999; 22: Sasayama S, Izumi T, Seino Y, Ueshima H, Asanoi H for the CHF- HOT Study Group. Effects of nocturnal oxygen therapy on outcome measures in patients with chronic heart failure and Cheyne-Stokes respiration. Circ J 2006; 70: Inoko M, Fujita M, Nakae I, Tamaki S, Watanuki M, Hashimoto T, et al. Effect of angiotensin-converting enzyme inhibition on sympathetic tone in patients with mild to moderate heart failure. Jpn Circ J 2001; 65: Castano G, Mas R, Fernandez JC, Illnait J, Fernandez L, Alvarez E. Effects of policosanol in older patients with type II hypercholesterolemia and high coronary risk. J Gerontol A Biol Sci Med Sci 2001; 56: Adachi H, Itoh H, Sakurai S, Takahashi T, Toyama T, Naito S, et al. Short-term physical training improves ventilatory response to exercise after coronary arterial bypass surgery. Jpn Circ J 2001; 65: Toyama T, Aihara Y, Iwasaki T, Hasegawa A, Suzuki T, Nagai R, et al. Cardiac sympathetic activity estimated by 123I-MIBG myocardial imaging in patients with dilated cardiomyopathy after betablocker or angiotensin-converting enzyme inhibitor therapy. J Nucl Med 1999; 33: Takeishi Y, Atsumi H, Fujiwara S, Takahashi K, Tomoike H. ACE inhibition reduces cardiac iodine-123-mibg release in heart failure. J Nucl Med 1997; 38: Germano G, Kiat H, Kavanagh PB, Moriel M, Mazzanti M, Su HT, et al. Automatic quantification of ejection fraction from gated myocardial perfusion SPECT. J Nucl Med 1995; 36: Franklin KA, Eriksson P, Sahlin C, Lundgren R. Reversal of central sleep apnea with oxygen. Chest 1997; 111: Andreas S, Clemens C, Sandholzer H, Figulla HR, Kreuzer H. Improvement of exercise capacity with treatment of Cheyne-Stokes respiration in patients with congestive heart failure. J Am Coll Cardiol 1996; 27: Imamura Y, Fukuyama T. Prognostic value of myocardial MIBG scintigraphy findings in patients with cardiomyopathy: Importance of background correction for quantification of MIBG activity. Ann Nucl Med 2002; 16: Yamada T, Shimonagata T, Fukunami M, Kumagai K, Ogita H, Hirata A, et al. Comparison of the prognostic value of cardiac Iodine- 123 metaiodobenzylguanidine imaging and heart rate variability in patients with chronic heart failure. J Am Coll Cardiol 2003; 41: Cohn JN, Levine TB, Olivari MT, Garberg V, Lura D, Francis GS, et al. Plasma norepinephrine as a guide to prognosis in patients with chronic congestive heart failure. N Engl J Med 1984; 311: Swedberg K, Eneroth P, Kjekshus J, Wilhelmsen L. Hormones regulating cardiovascular function in patients with severe congestive heart failure and their relation to mortality: CONSENSUS Trial Study Group. Circulation 1990; 82: Solin P, Roebuck T, Johns DP, Walters EH, Naughton MT. Peripheral and central ventilatory responses in central sleep apnea with and without congestive heart failure. Am J Respir Crit Care Med 2000; 162: Hall MJ, Xie A, Rutherford R, Ando S, Floras JS, Bradley TD. Cycle length of periodic breathing in patients with and without heart failure. Am J Respir Crit Care Med 1996; 154: Francis DP, Willson K, Davies C, Coats AJ, Piepoli M. Quantitative general theory for periodic breathing in chronic heart failure and its clinical implications. Circulation 2000; 102: Javaheri S. A mechanism of central sleep apnea in patients with heart failure. N Engl J Med 1999; 341: Wilcox I, McNamara SG, Dodd MJ, Sullivan CE. Ventilatory control in patients with sleep apnoea and left ventricular dysfunction: Comparison of obstructive and central sleep apnea. Eur Respir J 1998; 11: Heistad DD, Wheeler RC, Mark AL, Schmid PG, Abboud FM. Effects of adrenergic stimulation on ventilation in man. J Clin Invest 1972; 51: Mansfield D, Kaye DM, La Rocca HB, Solin P, Esler MD, Naughton MT. Raised sympathetic nerve activity in heart failure and central sleep apnea is due to heart failure severity. Circulation 2003; 107: