Ventricular Function in Snorers and Patients with Obstructive Sleep Apnea* Patrick Hanly, M.D., F.C.C.R; Zion Sasson, M.D.; Naheed Zuben, B.Sc.; and Mark Alderson, B.Sc. We hypothesized that intermittent hypoxemia and increased ventricular afterload due to obstructive apnea during sleep (OSA) would cause chronic left ventricular dysEuoction. Overnight polysomoography, M-mode andtwo-dimensional echo-Doppler studies while awake were performed on 51 consecutive snorers, 30 with OSA and 21 without apoea. Patients with previous myocardial infarction, awake hypoxemia or hypercapnia, or other causes of nocturnal hypoxemia were excluded. Echo-Doppler measurements included end-diastolic right and left ventricular dimensions and wall thickness, indices of left ventricular systolic performance (fractional shortening, ejection fraction and ejection time and diastolic performance, (isovolumicrelax.. ation time, ratio of peak early [E] to late [A] diastolic transmitral flow and mitral pressure half-time). Both OSA patients and nonapneic snorers were of similar age. Al.. though OSA patients were heavier, had a greater apnea.. hypopnea index, and significant nocturnal hypoxemia, their echo-Doppler measurements were within normal limits and
were not significantly different from nonapneic snorers. It is concluded that isolated obstructive sleep apnea does not cause chronic left ventricular dysEuoction. (Chat 1992; 102:100-05)
he acute effects of intermittent upper airway T obstruction during sleep on the cardiovascular
Third, obstructive apneas are characteristically terminated by arousals from sleep which are accompanied by a rise in heart rate, blood pressure, and cardiac output. 1,10 Normally, these cardiac indices fall during sleep, and to the extent that normal sleep reduces cardiac work, it is possible that any condition which disrupts sleep deprives the heart of this potential benefit. r.u Chronic cardiovascular sequelae of obstructive sleep apnea have also been investigated. Development of pulmonary hypertension and cor pulmonale in obstructive sleep apnea usually requires daytime hypoxemia and/or hypercapnia due to co-existing disease such as COPD or morbid obesity.12 Obstructive sleep apnea is an independent risk factor for the development of chronic systemic hypertension. 1 This may be mediated by chronic stimulation of the sympathetic nervous system from recurrent oxygen desaturation and sleep disruption. Sleep apnea patients have been reported to have significantly higher plasma and urinary catecholamines than control subjects" with a fall in urinary norepinephrine and normetanephrine levels following tracheostomy: 14 Chronic hypertension, which has been reported in over 50 percent of patients with obstructive sleep apnea,' predictably induces left ventricular dysfunction over time. Last, treatment of obstructive sleep apnea with nasal CPAP in patients who have both sleep apnea and congestive heart failure has recently been shown to improve left ventricular
system are well recognized. First, inspiration against an occluded upper airway reduces intrathoracic pressure to as low as - 60 mm Hg. I This increases venous return to the right ventricle which, by the mechanism of ventricular interdependence, causes a shift in the intraventricular septum and reduces left ventricular compliance.v" Since the fall in intrathoracic pressure For editorial comment see page 5 during an obstructive apnea is greater than the fall in systolic blood pressure, the pressure gradient across the ventricular wall rises causing increased afterload on both the right and left ventricles," Second, obstructive apneas are frequently associated with significant hypoxemia which decreases myocardial contractility" and increases both systemic and pulmonary arterial blood pressure.v" Hypoxemia contributes to the development of bradyarrhythmias during obstructive apnea by increasing parasympathetic tone" and increases the frequency of ventricular ectopy during sleep both in patients with obstructive apnea" and chronic obstructive pulmonary disease (COPD).9 *From the Sleep Laboratory, Department of Medicine, The Wellesley Hospital, Toronto, Ontario, Canada, This work was supported by Physicians Services Incorporated Foundation. Manuscript received August 2; revision accepted December 23.
100
=
= mu.
AHI apnea-bypopnea index, per hour of sleep; BMI body mass index, kglm'; ElA = ratio of early to late CliastoIic velocity in left ventricle; EF ejection fraction of left ven~z 'J,; ET left ventricular ejection time, IDS; FS fractional shortening of left ventricular, 'J,; HT = mitral pressure baIltime, IDS; IVRT isovoIumie relaxation time or left ventricle, IDS; LVD=left ventricular dimension, mm; LVMI=left ventricular mass index, rim'; LVW left ventricular wall thiclmess, mm; mean SaO, mean OX)'Ien saturation during sleep; mean tcCO. = mean transcutaneous earbon cIioxicle d~ sleep; min
=
=
=
=
=
=
=
SaO. mean minimum oxygen saturation during sleep;
OSA ==
obstructive sleep apnea; REM = rapid eye movement sleep; RVD = rigllt ventricular dimension, mm; RVW =..... t ventricular wall thicImess, IBID; SaO. = arterial osygen saturation, ,.; SaO, 60 percent) and grade 4 indicates severe LV dysfunction (ejection fraction
LVD (end-diastole) - LVD (end-systole) LVD (end-diastole)
Left ventricular ejection fraction was derived from M-mode measurements of ventricular dimensionsu'13 as follows: EF (%) [LVD (end-diastole»)3- [LVD (end-systole)] 3 [LVD (end-diastole)] 3 Ejection time was measured as the time from aortic valve opening to aortic valve closure and was corrected for heart rate. Z4 Left ventricular diastolic function was assessed by pulsed Doppler echocardiographylS·· using the Advanced Technology Laboratories Ultramark 8 (ATL-8) ultrasound system. This method of assessing left ventricular Slling is well established and has been validated against contrast and radionuclide angiography. ~.2ft The variables measured included isovolumic relaxation time (IVRT), defined as the time between aortic valve closure and mitral valve opening; peak early filling velocity (E); peak late Riling velocity (A); the EI A ratio, a measure of left ventricular filling in early versus late diastole; and mitral pressure half time (8T), a measure of the rate of deceleration of early diastolic Bo~ Transmitral Row was analyzed from the apical view; with a sample volume at the mitralleaRet tips level so as to obtain the higher E point velocity as well as the mitral opening and closure sounds. Isovolumic relaxation time was measured with the sample volume at the left ventricular outflow tract, closer to the aortic valve so as to record both aortic closure and mitral opening sounds. Each parameter was measured at endexpiration and averaged over at least 5 consecutive analyzable beats. CHEST I 102 I 1 I JUL~
1992
101
Table 3-Blood Ga and Apnea Data in OSA ftJtienta and Snorers·
Table I-Demographic Data in ObBtructa Sleep Apnea (OSA) ftJtienta and Snorers·
No. M:F Age 8MI, kw'm· Hypertension Snoring years
OSA
Snorers
31 25:6 5O± 12.3 32.8±7.2t 11 (35%) 15.4± 11.5
20 12:8 48±8.8 26.2±3.2 0 9±9
*Oata are mean ± so. tp TS'I; orad Sraor8n* Normal No M:F Age
BMI
Hypertension
AHI (111)
SaOI
were not significantly different from nonapneic snorers. It is concluded that isolated obstructive sleep apnea does not cause chronic left ventricular dysEuoction. (Chat 1992; 102:100-05)
he acute effects of intermittent upper airway T obstruction during sleep on the cardiovascular
Third, obstructive apneas are characteristically terminated by arousals from sleep which are accompanied by a rise in heart rate, blood pressure, and cardiac output. 1,10 Normally, these cardiac indices fall during sleep, and to the extent that normal sleep reduces cardiac work, it is possible that any condition which disrupts sleep deprives the heart of this potential benefit. r.u Chronic cardiovascular sequelae of obstructive sleep apnea have also been investigated. Development of pulmonary hypertension and cor pulmonale in obstructive sleep apnea usually requires daytime hypoxemia and/or hypercapnia due to co-existing disease such as COPD or morbid obesity.12 Obstructive sleep apnea is an independent risk factor for the development of chronic systemic hypertension. 1 This may be mediated by chronic stimulation of the sympathetic nervous system from recurrent oxygen desaturation and sleep disruption. Sleep apnea patients have been reported to have significantly higher plasma and urinary catecholamines than control subjects" with a fall in urinary norepinephrine and normetanephrine levels following tracheostomy: 14 Chronic hypertension, which has been reported in over 50 percent of patients with obstructive sleep apnea,' predictably induces left ventricular dysfunction over time. Last, treatment of obstructive sleep apnea with nasal CPAP in patients who have both sleep apnea and congestive heart failure has recently been shown to improve left ventricular
system are well recognized. First, inspiration against an occluded upper airway reduces intrathoracic pressure to as low as - 60 mm Hg. I This increases venous return to the right ventricle which, by the mechanism of ventricular interdependence, causes a shift in the intraventricular septum and reduces left ventricular compliance.v" Since the fall in intrathoracic pressure For editorial comment see page 5 during an obstructive apnea is greater than the fall in systolic blood pressure, the pressure gradient across the ventricular wall rises causing increased afterload on both the right and left ventricles," Second, obstructive apneas are frequently associated with significant hypoxemia which decreases myocardial contractility" and increases both systemic and pulmonary arterial blood pressure.v" Hypoxemia contributes to the development of bradyarrhythmias during obstructive apnea by increasing parasympathetic tone" and increases the frequency of ventricular ectopy during sleep both in patients with obstructive apnea" and chronic obstructive pulmonary disease (COPD).9 *From the Sleep Laboratory, Department of Medicine, The Wellesley Hospital, Toronto, Ontario, Canada, This work was supported by Physicians Services Incorporated Foundation. Manuscript received August 2; revision accepted December 23.
100
=
= mu.
AHI apnea-bypopnea index, per hour of sleep; BMI body mass index, kglm'; ElA = ratio of early to late CliastoIic velocity in left ventricle; EF ejection fraction of left ven~z 'J,; ET left ventricular ejection time, IDS; FS fractional shortening of left ventricular, 'J,; HT = mitral pressure baIltime, IDS; IVRT isovoIumie relaxation time or left ventricle, IDS; LVD=left ventricular dimension, mm; LVMI=left ventricular mass index, rim'; LVW left ventricular wall thiclmess, mm; mean SaO, mean OX)'Ien saturation during sleep; mean tcCO. = mean transcutaneous earbon cIioxicle d~ sleep; min
=
=
=
=
=
=
=
SaO. mean minimum oxygen saturation during sleep;
OSA ==
obstructive sleep apnea; REM = rapid eye movement sleep; RVD = rigllt ventricular dimension, mm; RVW =..... t ventricular wall thicImess, IBID; SaO. = arterial osygen saturation, ,.; SaO, 60 percent) and grade 4 indicates severe LV dysfunction (ejection fraction
LVD (end-diastole) - LVD (end-systole) LVD (end-diastole)
Left ventricular ejection fraction was derived from M-mode measurements of ventricular dimensionsu'13 as follows: EF (%) [LVD (end-diastole»)3- [LVD (end-systole)] 3 [LVD (end-diastole)] 3 Ejection time was measured as the time from aortic valve opening to aortic valve closure and was corrected for heart rate. Z4 Left ventricular diastolic function was assessed by pulsed Doppler echocardiographylS·· using the Advanced Technology Laboratories Ultramark 8 (ATL-8) ultrasound system. This method of assessing left ventricular Slling is well established and has been validated against contrast and radionuclide angiography. ~.2ft The variables measured included isovolumic relaxation time (IVRT), defined as the time between aortic valve closure and mitral valve opening; peak early filling velocity (E); peak late Riling velocity (A); the EI A ratio, a measure of left ventricular filling in early versus late diastole; and mitral pressure half time (8T), a measure of the rate of deceleration of early diastolic Bo~ Transmitral Row was analyzed from the apical view; with a sample volume at the mitralleaRet tips level so as to obtain the higher E point velocity as well as the mitral opening and closure sounds. Isovolumic relaxation time was measured with the sample volume at the left ventricular outflow tract, closer to the aortic valve so as to record both aortic closure and mitral opening sounds. Each parameter was measured at endexpiration and averaged over at least 5 consecutive analyzable beats. CHEST I 102 I 1 I JUL~
1992
101
Table 3-Blood Ga and Apnea Data in OSA ftJtienta and Snorers·
Table I-Demographic Data in ObBtructa Sleep Apnea (OSA) ftJtienta and Snorers·
No. M:F Age 8MI, kw'm· Hypertension Snoring years
OSA
Snorers
31 25:6 5O± 12.3 32.8±7.2t 11 (35%) 15.4± 11.5
20 12:8 48±8.8 26.2±3.2 0 9±9
*Oata are mean ± so. tp TS'I; orad Sraor8n* Normal No M:F Age
BMI
Hypertension
AHI (111)
SaOI
