- Email: [email protected]
Prevalence of Patent Foramen Ovale and Its Contribution to Hypoxemia in Patients with Obstructive Sleep Apnea
Prevalence of Patent Foramen Ovale and Its Contribution to Hypoxemia in Patients with Obstructive Sleep Apnea" Hany Shanoudy, MD; Adel Soliman, MD; Paolo Raggi, MD; Jing Win Liu, MD, FCCP; Douglas C. Russell, MD, PhD; and Nabil F. Jarmitkli, MD
Study objectives: The objectives of this study were (1) to assess the prevalence of patent foramen ovale (PFO) by means of contrast transesophageal echocardiography (TEE) in patients with obstructive sleep apnea, and (2) to determine the potential contribution of right to left interatrial shunting to systemic oxygen desaturation following the performance of Valsalva maneuver. Design: Performance of contrast TEE during Valsalva maneuver with simultaneous measurement of systemic arterial oxygen saturation (Sa02) by means of pulse oximetry in patients with obstructive sleep apnea and a control group. Setting: Government teaching hospital, university hospital affiliate. Patients: Study group comprised 48 patients with documented obstructive sleep apnea and 24 control subjects. Interventions: Sleep studies, contrast TEE, Valsalva maneuver, pulse oximetry. Measurements and results: Thirty-three of 48 patients with obstructive sleep apnea compared with 4 of 24 control patients had a detectable PFO (69% vs 17%; p<0.0001). All sleep apnea patients had comparable baseline Sa02 regardless of the presence of a PFO (93.9±1.7% vs 95±1.2%; p=not significant). After performance of a Valsalva maneuver, however, a significantly greater fall in Sa02 was observed in patients with obstructive sleep apnea and PFO compared with patients with obstructive sleep apnea without PFO (-2.4±1.5% vs -1.3±0.6%; p=0.007). A statistically significant fall in Sa02 (defined as >4 SD of recorded Sa02 values after Valsalva maneuver in patients without PFO) was found in one third of patients with sleep apnea and PFO. Conclusion: We conclude that there is an increased prevalence of PFO in patients with obstructive sleep apnea that could contribute to significant hypoxemia after a Valsalva maneuver in approximately one third of these patients. (CHEST 1998; 113:91-96)
words: hypoxemia; echocardiography
Key
intracardiac
shunting;
obstructive
sleep
apnea; patent foramen ovale;
Abbreviations: OSA=obstructive sleep apnea; PFO=patent foramen ovale; Sa02=arterial oxygen saturation; TEE=transesophageal echocardiography
A patent foramen ovale (PFO) as identified by autopsy, right heart catheterization, or trans¬ esophageal echocardiographic (TEE) studies is found in approximately 10 to 35% of the general population.1-5 Right to left shunting through a PFO
^^
*From the Sections of Cardiology and Pulmonary Medicine, Veterans Affairs Medical Center, Salem, and University of Sciences Center, Charlottesville. Virginia, Health received March 13, 1997; revision accepted July 28, Manuscript 1997. Reprint requests: Nabil F. Jarmukli, MD, Assistant Professor of Medicine, University of Virginia, Director, Non-Invasive Labo¬ ratories, Cardiology Section (111B), Veterans Affairs Medical Center, 1970 Roanoke Blvd, Salem, VA 24153; email:
[email protected]
RDI=respiratory
transesophageal
disturbance index;
under conditions that result in a higher atrial than left atrial pressure and can poten¬ right rise tially give to systemic arterial desaturation416 and to paradoxic embolization. Little is known, how¬ ever, of the prevalence of PFO in patients with obstructive sleep apnea (OSA) and whether its pres¬ ence contributes to hypoxemia during apneic epi¬ sodes. Patients with OSA develop transient elevation of right-sided pressures during periods of nocturnal apnea. This might be of sufficient degree to cause right to left shunting of venous blood across a PFO.17-19 Significant hypoxemia and pulmonary hy¬ pertension may in turn ensue.2022 In recent years, TEE has been shown to be a very sensitive technique for the detection of intracardiac shunting.14'23-27 can occur
CHEST/113/1 /JANUARY, 1998
91
Furthermore,
it has been shown that maneuvers used to increase right-sided chamber pressure, such as cough or a Valsalva maneuver, can increase the sensitivity of PFO detection by TEE.3-28 The goals of
this study were therefore (1) to assess the prevalence of PFO by means of contrast TEE following Valsalva maneuver in patients with OSA, and (2) to determine the potential contribution to systemic oxygen desatu¬ ration of right to left interatrial shunting during periods of elevated right atrial pressure induced by Valsalva maneuver. Materials
and
Methods
Study Population The study was reviewed and approved by the human study committee of the Veterans' Affairs Medical Center in Salem, Va. The study population comprised 48 male patients with docu¬ mented OSA by clinical assessment and overnight sleep studies. Patients were excluded from study (1) if they had sustained prior myocardial infarction, unstable angina, or decompensated con¬ gestive heart failure within 3 months prior to study, (2) if they had an active infectious pulmonary process, prior pulmonary embo¬ lism, or infarction, or (3) if they suffered from obstructive or restrictive lung disease as documented by pulmonary function tests. They were also excluded (4) if they had any contraindication to TEE, including active upper GI bleeding, dysphagia, recent gastroesophageal surgeiy, or perforated viscus. Control
Group
Twenty-four male patients undergoing elective diagnostic TEE entered into the study as a control group. No patient had a history of sleep apnea, COPD, active congestive heart failure, unstable angina, or prior myocardial infarction. All control sub¬ jects had normal cardiac dimensions and functions with no valvular pathologic findings by TEE. were
Transesophageal Echocardiography Written informed consent was obtained in all cases after review of the patient's medical history, inclusive of allergies and medi¬ cations received. All patients fasted for at least 6 h prior to the procedure. An IV access was established and nasal oxygen was supplied after data collection in only three patients with OSA who developed arterial oxygen desaturation after Valsalva maneuver. An ECG and BP were monitored continuously. Mild sedation with 1 mg of IV midazolam was given to most patients before intubation. A complete TEE study was performed in esophageal all patients and included M-mode and two-dimensional echocar¬ diography, pulse wave, continuous-wave, and color-flow Doppler studies. The peak velocity of tricuspid regurgitant flow was recorded and the systolic pulmonary artery pressure was esti¬ mated using the modified Bernoulli equation. Images of the cardiac chambers and great vessels were obtained in both the horizontal and longitudinal planes from the transgastric, midesophageal, and proximal esophageal locations, using a 3.5-, 5.0-, and 7.0-MHz phased array probe. All studies were performed using a biplane transducer (Acuson 128 XP; Acuson, Mountview,
Calif).
92
Arterial
Oxygen Saturation Monitoring
During TEE, all patients underwent continuous recording of arterial oxygen saturation (Sa02) by finger pulse oximetry before, during, and after maneuvers designed to induce transient eleva¬ tion of right-sided pressures (Valsalva maneuver and cough). All
recordings were serially printed. Assessment of Presence
The
of Interatrial Shunt
transesophageal four-chamber view and the longitudinal
view were utilized to locate the thinnest portion of the interatrial septum and the junction between septum primum and septum secundum and to interrogate for the presence of PFO. Color flow and pulse Doppler were performed to assess flow patterns across the septum. Echocardiographic contrast was obtained by mixing 10 mL of normal saline solution with 1 mL of air in a syringe. Boluses of agitated saline solution were rapidly injected into a peripheral vein and echogenic microbubbles, which usually dissipate within the pulmonary circulation, were imaged as they entered the right-sided chambers of the heart. In the presence of a PFO, these microbubbles are seen to traverse the interatrial septum from right to left. The passage of these microbubbles is aided by the transient elevation of right atrial pressure produced by either forceful coughing or Valsalva ma¬ neuver for 15 s. The efficacy of cough or Valsalva maneuver was validated by observing a movement of the interatrial septum toward the left atrium following release ofthe Valsalva maneuver. Imaging of the interatrial septum was performed both in the horizontal and the longitudinal planes to maximize the likelihood of identifying a PFO. At least two contrast injections were performed in each subject, one before and one during pressuregenerating maneuvers. Each patient was assessed for the pres¬ ence of an atrial septal defect other than a PFO by both color and vena cava
spectral Doppler.
Definitions PFO was considered present if a minimum of five microbubbles were seen in the left atrium within three cardiac cycles after opacification of the right atrium following peripheral echocontrast injection. An atrial septal defect was considered present if a consistent discontinuity of the septum was observed in different echocardiographic views and a left to right Doppler flow signal across the septum was recorded. All studies were inter¬ preted separately by two independent observers who had no prior information about the patient's clinical status.
Sleep Studies Polysomnography was performed in 48 patients with OSA. EEGs (C4-Als C4-02), electromyograms, and electro-oculograms were continuously recorded by polysomnography in 48 patients. was monitored oronasal thermistors and thoRespiration
using
racoabdominal strain gauges. Apnea was defined as cessation of airflow at the mouth and the nose for >10 s, while hypopnea was defined as >40% reduction of thermistor signal at the same locations lasting at least 10 s. OSA was considered to be present if the absence of airflow was associated with continued effort. A minimum of five such episodes of apnea-hypopnea per hour of sleep were required. Nocturnal Sa02 was measured by pulse oximetry. Apnea index and hypopnea index were defined as number of events per hour. Respiratory disturbance index (RDI) was defined as the number of apnea-hypopnea episodes per hour of sleep. Clinical
Investigations
Statistical Analysis
The prevalence of PFO in patients with sleep apnea syndrome compared with the prevalence of PFO in the control group of 24 patients. A x2 test was employed to analyze the statistical estimated difference. Comparisons of systemic significance ofthe Sa02 changes before and after Valsalva maneuvers were made using the paired t test. A p value <0.05 was considered statisti¬
was
cally significant.
Raseline Clinical Characteristics All patients studied were men. The mean age for patients in the sleep apnea group was 57±12.3 years and for patients in the control group it was 65±9.5 years.
Sleep Studies: All patients assigned to the sleep apnea group had undergone overnight sleep stud¬ ies prior to entry in the study. The mean apnea index in this group was 15.2±15.5, the mean index was 17.1± 17.3, and the mean hypopnea RDI was 33.9±31.0 (Table 1). In the sleep apnea patient group, there was no significant difference in RDI between subjects with and without a PFO (36.9±30.5 vs 27.8±33.9; p not significant [NS]; Table 2). Tests: No patient in the sleep Pulmonary Function apnea group had evidence of moderate or severe obstructive or restrictive lung disease. The mean FEV^FVC in the sleep apnea group did not signif¬ the control group (78.1 ±9.3 vs icantly differ fromTable 77.3±5.1; p-NS; 1). Further, the FEV^FVC did not significantly differ between patients with or vs in the =
a
PFO
OSA group
Table 2).
81.4±6.3; p-NS;
(76.7±10.1
Raseline Oxygen Saturation: Similar baseline sys¬ temic oxygen saturations were recorded in sleep apnea and control
Positive Shunt
Negative Shunt (n=15)
p Value
57±11.5 2.3+0.2 76.7±10.1 36.9±30.5 34±10.3 93.9±1.7 -2.4±1.5
57±13.3 2.3+0.2 81.4±6.3
NS NS NS NS NS NS
(n=33)
Age, yr
BSA, m2
FEV/FVC RDI
Results
without
Table 2.Clinical Characteristics and Test Results for and Without Sleep Apnea Patients With Detectable PFO*
PAP, mm Hg Baseline Sa02, %
A
Sa02, %
*For
explanation of abbreviations, see
27.8±33.9 22±6.3 95.1±1.2 -1.3±0.6 Table 1 footnote.
0.007
p^NS; Table 1) and between sleep apnea patients with and without a detectable PFO (93.9±1.7 vs
95.1±1.2;p=NS;Table2). Prevalence of PFO A PFO (Figs 1 and 2) was detected in 33 of 48 (69%) patients in the sleep apnea group, which constituted a significantly greater prevalence than that reported in the literature (10 to 35%)1-5 and that found in our control group (69% vs 17%; p=0.0001; Table 1). Neither color-flow Doppler nor contrast the presence of an echocardiography demonstrated atrial septal defect in any of the patients studied. Changes in Peripheral Sa02 of a Valsalva maneu¬ Following the performanceshowed a greater fall in with ver, patients sleep apnea vs -0.9±0.7; than control Sa02 subjects (-2.0±1.3 a Valsalva Table maneuver, p=0.0002;with 1). Following sleep apnea and a demonstrable right patients
subjects (94.2±1.8 vs 95.1±2.2;
Table 1.Clinical Characteristics and Test Results for Sleep Apnea and Control Subjects*
Sleep Apnea (n=48)
Age, yr
BSA, m2
FE\yFVC RDI
PAP,
mm
Hg
Baseline Sa02, %
A Sa02, % PFO prevalance, % *
57±12.3 2.3±0.2 78.1±9.3 33.9±31 32±10.3 94.2±1.8 -2.0±1.3 69
Control
Subjects
(n=24)
65±9.5
2.2±0.3 77.3±5.1 NC 22.0±6.3 95.1±2.2 -0.9±0.7 17
p Valu 0.004 NS NS
0.003 NS 0.0002 0.0001
BSA=body surface area; NC not calculated; PAP=pulmonary artery systolic pressure; Sa02 .finger pulse oximetry Sa02; A Sa02=degree of arterial 02 desaturation following Valsalva maneu¬ =
ver.
view: echogenic con¬ Longitudinal transesophageal the right atrium (RA). injected IV and fills aorta; PA=pulmonary artery; LA=left atrium.
Figure 1.
trast
AO
=
has been
CHEST / 113 / 1 / JANUARY, 1998
93
interatrial septum at the end of a Valsalva maneuver taken to confirm that the test had been properly executed. A PFO was detected four times more frequently in 48 consecutively studied patients than in a group of control subjects. The prevalence of PFO in our study patients was also statistically greater than previously reported in the medical literature. Several consider¬ ations can be made to explain our observations. The first pertains to the technique utilized in this study to visualize a PFO. The presence of PFO may be diagnosed either invasively or noninvasively. The diagnosis of PFO during cardiac catheterization re¬ quires the use of either a hydrogen electrode test or of the green dye method. Roth techniques are tedious and inaccurate. TEE has been shown in recent years to be superior as a method for detecting intracardiac shunts and is much more sensitive than transthoracic echocardiography.23-26 Because PFO is not associated with a discontinuity of the septum, color Doppler or IV administration of echocontrast materials (eg, agitated saline solution) is required for the diagnosis of right to left intracardiac shunting.27 Right to left intracardiac shunting is considered present when microbubbles cross the fossa ovalis within three cardiac cycles after adequate opacification ofthe right atrium, or when shunting of blood across the interatrial septum is visualized by color-flow Doppler. Both cough and the Valsalva maneuver have been used to increase the sensitivity of this technique.28 Contrast TEE has been shown to be superior to color-flow Doppler for the detection of interatrial communications, since the latter tech¬ nique relies on flow velocity while contrast echocar¬ is volume dependent.24 diography The employment of today's most sensitive tech¬ nique may be one of the reasons for the high detection rate of PFO in our study population. An analysis of the structure and embryologic develop¬ ment of the foramen ovale, along with a critical appraisal of the changes occurring in the pulmonary of patients with sleep apnea, can provide physiology another insight to explain our observations. The foramen ovale is bordered anatomically by the septum secundum (limbus of the fossa ovalis) and the septum primum. The latter acts as a valve within the fossa ovalis permitting unidirectional blood flow from the right to the left atrium. Functional closure of the PFO usually occurs shortly after birth when the pressure in the left atrium exceeds that in the was
as Figure 1: microbubbles of echogenic through a PFO from the right (RA) to the left
Figure 2. Same viewr
contrast
atrium
escape
(LA).
to left shunt across a PFO had a significantly greater fall in their Sa02 than patients with sleep apnea and undetectable PFO (-2.4±1.5 vs -1.3±0.6; p=0.007; Table 2). A statistically significant fall in Sa02 was defined as a change >4 SDs below the Sa02 attained after a Valsalva maneuver in subjects with sleep apnea but without a visible PFO. Such notable fall in oxygen saturation was found in 33% of patients with sleep apnea and a detectable PFO. In the latter group, there was no significant correlation between RDI and the degree of arterial oxygen desaturation (r=0.5; p=NS).
Pulmonary Artery Systolic Pressure The estimated pulmonary artery systolic pressure was significantly greater in patients with OSA than in the control subjects (32.0±10.3 mm Hg vs 22.0±6.3 mm Hg; p=0.003; Table 1). In the patients with sleep apnea and a detectable PFO, the average pulmonary artery systolic pressure did not correlate RDI or with the calculated with the extent of arterial oxygen desaturation achieved after a Valsalva maneu¬
ver
(r=0.4; p
=
NS
each).
Discussion
In this study, we sought to determine the preva¬ lence of PFO in patients with OSA by means of contrast TEE. We also evaluated the possible con¬ tribution of a PFO to the development of hypoxemia during periods of apnea. To do this, we simulated the hemodynamic changes that occur during periods of nocturnal apnea using cough or Valsalva maneuver to cause a transient pressure elevation inside the rightsided cardiac chambers. A shift to the left of the 94
right
pressing the valve of the fossa ovalis the limbus to produce a competent seal. In a against number of patients the foramen ovale does not anatomically seal and constitutes a potential inter¬ atrial communication through which venous blood atrium
Clinical
Investigations
may flow from the right to the left atrium when pressures exceed left-sided pressures.12 right-sided In our study, patients with OSA had a significantly pulmonary artery systolic pressure than pa¬ higher in tients the control group under resting conditions. This might have favored the maintenance of patency of the foramen. It would seem plausible that during of nocturnalapnea, a transient rise of pres¬ periods sure inside the right-sided cardiac chambers can cause right to left shunting of venous blood across an PFO. Further, one third of the incompletely sealed OSA affected patients by with a demonstrable PFO showed a statistically significant fall in systemic Sa02 a Valsalva maneuver. This seems to indi¬ following cate that a PFO can be considered at least partly for the episodes of transient arterial responsible recorded during sleep in this category of hypoxemia
patients. It should be noted that although pulmonary hy¬ pertension occurs in 20 to 60% of patients with OSA, a cause and effect relationship between these two syndromes has not yet been established.17 That acute elevations of pulmonary artery pressure can occur in association with episodes of hypoxemia is well doc¬ umented, though the relationship between diurnal
pulmonary hypertension and sleep apnea is less well established. Repeated transient episodes of hypox¬ emia during apneic spells can induce pulmonary vasoconstriction and eventually pulmonary hyperten¬ sion. During obstructive apnea episodes, which are associated with a fall in intrathoracic pressure and increased parasympathetic tone, pulmonary artery systolic pressure decreases.17-19 However, following the relief of the apneic episodes, a progressive rise of the pulmonary artery systolic pressure is immedi¬ ately noted which has been attributed to an increase in heart rate, cardiac output, and sympathetic tone with an elevated catecholamine level.17-19'29'30 Nev¬ ertheless, several authors have reported that the development of pulmonary hypertension in patients with OSA is more closely related to coexisting pul¬ monary disease and hypoxemia than to the severity of OSA.31-33 In several other conditions in which the pressure inside the right-sided cardiac chambers is increased, a PFO has been considered responsible for systemic desaturation. Thus, right to left shunting across a PFO may worsen systemic hypoxemia in patients with chronic obstructive lung disease and embolism.2'7 Similar effects may be seen pulmonary in patients undergoing positive pressure ventilation10 and have been reported in small series of patients with right ventricular infarction, tricuspid atresia, Ebstein's anomaly, pulmonary stenosis, and cardiac tamponade.2'68'9 As a final consideration, it is worth mentioning that patency of the foramen ovale ap¬ pears to be inversely related to age.1 Interestingly, in
study, patients with OSA had a prevalence of PFO of 69% that far exceeded that expected for their age of 25 %x indicating that the pathophysiology of OSA predisposes to the maintenance of patency of a
our
foramen ovale.
Clinical Implication This study strongly suggests that an increased of PFO occurs in patients with OSA and prevalence right to left intracardiac shunting across a PFO can contribute to systemic hypoxemia in up to one third of these patients. References 1 Hagen PT, Scholz DG, Edwards WD. Incidence and size of patent foramen ovale during the first 10 decades of life: an autopsy study of 965 normal hearts. Mayo Clin Proc 1984; 59:17-20
C, Podolsky LA, Meyerowitz CB, et al. Patent foramen ovale: a nonfunctional embryological remnant or a potential cause of significant pathology? J Am Soc Echocar-
2 Movsowitz
diogr 1992; 5:259-70 Lynch JJ, Schuchard GH, Gross CM, et al. Prevalence of right to left atrial shunting in a healthy population: detection by Valsalva maneuver contrast echocardiography. Am J Cardiol 1984; 53:1478-80 4 Sardesi SH, Marshall RJ, Mourant AJ. Paradoxical systemic embolization through a patent foramen ovale. Lancet 1989; 3
1:732-33 5 Nootens MT, Berarducci LA, Kaufmann E, et al. The
prevalence and significance of a patent foramen ovale in pulmonary hypertension. Chest 1993; 104:1673-75 6 Chen WJ, Kuan P, Lien WP, et al. Detection of patent foramen ovale by contrast transesophageal echocardiography.
Chest 1992; 101:1515-20 Kasper W, Tiede N, Geibel A, et al. Clinical relevance of patent foramen ovale in patients with hemodynamic active pulmonary embolism [abstract]. Circulation 1991; 84:11-452 8 Moorthy SS, Losasso AM, Gibbs PS. Acquired right-to-left intracardiac shunts and severe hypoxemia. Crit Care Med 7
1978; 6:28-31 9 Wendel CH, Dianzumb S, Joyner CR.
Right to left interatrial shunt secondary to an extensive right ventricular myocardial infarction. Clin Cardiol 1985; 8:230-32 10 Lemaire F, Richalet JP, Carlet J, et al. Postoperative hypox¬ emia due to opening of a patent foramen ovale confirmed by a right atrium-left atrium pressure gradient during mechani¬ cal ventilation. Anesthesiology 1982; 57:233-36 11 Hausmann D, Mugge A, Becht I, et al. Diagnosis of patent foramen ovale by transesophageal echocardiography and as¬ sociation with cerebral and peripheral embolic events. Am J Cardiol 1992; 70:668-72 12 Lechat P, Mas JL, Lascault G. Prevalence of patent foramen ovale in patients with stroke. N Engl J Med 1988; 318:
1148-52 13 Loscalzo J. Paradoxical embolism, clinical presentation, diag¬ nostic strategies, therapeutic options. Am Heart J 1986; 112:141-45 14 Sun JP, Stewart WJ, Hanna J, et al. Diagnosis of patent
foramen ovale by contrast versus color Doppler by trans¬ esophageal echocardiography: relation to atrial size. Am Heart J 1996; 131:239-44
CHEST / 113 / 1 / JANUARY, 1998
95
15
Camp GV,
Schulze D,
Cosyns B,
et
al. Relation between Am J Cardiol
patent foramen ovale and unexplained stroke. 1993; 71:596-98 AC, Nagelhout D, Castello R,
et al. Atrial septal aneurysm and stroke: a transesophageal echocardiographic study. J Am Coll Cardiol 1991; 18:1223-29 Bone RC, Dantzker DR, George RB, et al. Pulmonary and critical care medicine (vol 2). Chicago: Mosby, 1995 Douglas NJ. The sleep apnea/hypopnea syndrome. Eur J Clin Invest 1995; 25:285-90 Parish JM, Shepard JW. Cardiovascular effects of sleep disorders. Chest 1990; 97:1220-26 Laks L, Krieger J, Podszus T. Pulmonary hypertension in obstructive sleep apnea: multicenter study [abstract]. Am Rev Respir Dis 1992; 145:A865 Krieger J, Sforza E, Apprill M, et al. Pulmonary hypertension, hypoxemia, and hypercapnia in obstructive sleep apnea pa¬ tients. Chest 1989; 96:729-37 Weber K, Podszus T, Krupp O, et al. Prevalence of pulmo¬ nary hypertension in patients with obstructive sleep apnea [abstract]. Sleep Research 1990; 19:308 Siostrzonek P, Zangeneh M, Gossinger H, et al. Comparison of transesophageal and transthoracic contrast echocardiogra¬
16 Pearson
17 18 19 20 21 22
23
phy for detection of patent foramen ovale. Am J Cardiol 1991;
68:1247-49 24 Loutolahti M, Saraste M, Hartiala J. Saline contrast and color
Doppler transesophageal echocardiography in detecting a patent foramen ovale and right-to-left shunts in stroke pa¬ tients. Clin Physiol 1995; 15:265-73
JH, et al. The incidence of in foramen ovale consecutive 1,000 patent patients: a contrast
25 Fisher DC, Fisher EA, Budd
96
transesophageal echocardiography study. Chest 1995; 107: 1504-09 26 Dubourg O, Boundaries JP, Farcot JC, et al. Contrast echo¬ cardiographic visualization of cough induced right to left shunt through a patent foramen ovale. J Am Coll Cardiol
1984; 4:587-94 27 Chen WJ, Chen JJ, Lin SC, et al. Detection of cardiovascular shunts by transesophageal echocardiography in patients with pulmonary hypertension of unexplained cause. Chest 1995; '
107:8-13 28 Carter SA, Birkhead
NC, Wood EH. Effect of Valsalva oxygen saturation in patients with intracardiac shunts. Circulation 1959; 20:574-86 29 Burman EJ, DiBenedetto RJ, Causey DE, et al. Right ventricular hypertrophy detected by echocardiography in patients with newly diagnosed obstructive sleep apnea. Chest maneuver on
1991; 100:347-50 30 Okabe S, Hida W, Kikuchi Y, et al. Role of hypoxia on increased blood pressure in patients with obstructive sleep apnea. Thorax 1995; 50:28-34 31 Schroeder JS, Motta J, Guilleminault C. Hemodynamic stud¬ ies in sleep apnea: sleep apnea syndromes. New York: Alan R Liss, 1978; 177-96 32 Bradley TD, Rutherford R, Grossman RF. Role of daytime
hypoxemia in the pathogenesis of right heart failure in the obstructive sleep apnea syndrome. Am Rev Respir Dis 1985; 131:835-39 33 Weitzenblum E, Krieger J, Apprill M. Daytime pulmonary hypertension in patients with obstructive sleep apnea syn¬ drome. Am Rev Dis 138:345-49 Respir
1988;
Clinical
Investigations
Study objectives: The objectives of this study were (1) to assess the prevalence of patent foramen ovale (PFO) by means of contrast transesophageal echocardiography (TEE) in patients with obstructive sleep apnea, and (2) to determine the potential contribution of right to left interatrial shunting to systemic oxygen desaturation following the performance of Valsalva maneuver. Design: Performance of contrast TEE during Valsalva maneuver with simultaneous measurement of systemic arterial oxygen saturation (Sa02) by means of pulse oximetry in patients with obstructive sleep apnea and a control group. Setting: Government teaching hospital, university hospital affiliate. Patients: Study group comprised 48 patients with documented obstructive sleep apnea and 24 control subjects. Interventions: Sleep studies, contrast TEE, Valsalva maneuver, pulse oximetry. Measurements and results: Thirty-three of 48 patients with obstructive sleep apnea compared with 4 of 24 control patients had a detectable PFO (69% vs 17%; p<0.0001). All sleep apnea patients had comparable baseline Sa02 regardless of the presence of a PFO (93.9±1.7% vs 95±1.2%; p=not significant). After performance of a Valsalva maneuver, however, a significantly greater fall in Sa02 was observed in patients with obstructive sleep apnea and PFO compared with patients with obstructive sleep apnea without PFO (-2.4±1.5% vs -1.3±0.6%; p=0.007). A statistically significant fall in Sa02 (defined as >4 SD of recorded Sa02 values after Valsalva maneuver in patients without PFO) was found in one third of patients with sleep apnea and PFO. Conclusion: We conclude that there is an increased prevalence of PFO in patients with obstructive sleep apnea that could contribute to significant hypoxemia after a Valsalva maneuver in approximately one third of these patients. (CHEST 1998; 113:91-96)
words: hypoxemia; echocardiography
Key
intracardiac
shunting;
obstructive
sleep
apnea; patent foramen ovale;
Abbreviations: OSA=obstructive sleep apnea; PFO=patent foramen ovale; Sa02=arterial oxygen saturation; TEE=transesophageal echocardiography
A patent foramen ovale (PFO) as identified by autopsy, right heart catheterization, or trans¬ esophageal echocardiographic (TEE) studies is found in approximately 10 to 35% of the general population.1-5 Right to left shunting through a PFO
^^
*From the Sections of Cardiology and Pulmonary Medicine, Veterans Affairs Medical Center, Salem, and University of Sciences Center, Charlottesville. Virginia, Health received March 13, 1997; revision accepted July 28, Manuscript 1997. Reprint requests: Nabil F. Jarmukli, MD, Assistant Professor of Medicine, University of Virginia, Director, Non-Invasive Labo¬ ratories, Cardiology Section (111B), Veterans Affairs Medical Center, 1970 Roanoke Blvd, Salem, VA 24153; email:
[email protected]
RDI=respiratory
transesophageal
disturbance index;
under conditions that result in a higher atrial than left atrial pressure and can poten¬ right rise tially give to systemic arterial desaturation416 and to paradoxic embolization. Little is known, how¬ ever, of the prevalence of PFO in patients with obstructive sleep apnea (OSA) and whether its pres¬ ence contributes to hypoxemia during apneic epi¬ sodes. Patients with OSA develop transient elevation of right-sided pressures during periods of nocturnal apnea. This might be of sufficient degree to cause right to left shunting of venous blood across a PFO.17-19 Significant hypoxemia and pulmonary hy¬ pertension may in turn ensue.2022 In recent years, TEE has been shown to be a very sensitive technique for the detection of intracardiac shunting.14'23-27 can occur
CHEST/113/1 /JANUARY, 1998
91
Furthermore,
it has been shown that maneuvers used to increase right-sided chamber pressure, such as cough or a Valsalva maneuver, can increase the sensitivity of PFO detection by TEE.3-28 The goals of
this study were therefore (1) to assess the prevalence of PFO by means of contrast TEE following Valsalva maneuver in patients with OSA, and (2) to determine the potential contribution to systemic oxygen desatu¬ ration of right to left interatrial shunting during periods of elevated right atrial pressure induced by Valsalva maneuver. Materials
and
Methods
Study Population The study was reviewed and approved by the human study committee of the Veterans' Affairs Medical Center in Salem, Va. The study population comprised 48 male patients with docu¬ mented OSA by clinical assessment and overnight sleep studies. Patients were excluded from study (1) if they had sustained prior myocardial infarction, unstable angina, or decompensated con¬ gestive heart failure within 3 months prior to study, (2) if they had an active infectious pulmonary process, prior pulmonary embo¬ lism, or infarction, or (3) if they suffered from obstructive or restrictive lung disease as documented by pulmonary function tests. They were also excluded (4) if they had any contraindication to TEE, including active upper GI bleeding, dysphagia, recent gastroesophageal surgeiy, or perforated viscus. Control
Group
Twenty-four male patients undergoing elective diagnostic TEE entered into the study as a control group. No patient had a history of sleep apnea, COPD, active congestive heart failure, unstable angina, or prior myocardial infarction. All control sub¬ jects had normal cardiac dimensions and functions with no valvular pathologic findings by TEE. were
Transesophageal Echocardiography Written informed consent was obtained in all cases after review of the patient's medical history, inclusive of allergies and medi¬ cations received. All patients fasted for at least 6 h prior to the procedure. An IV access was established and nasal oxygen was supplied after data collection in only three patients with OSA who developed arterial oxygen desaturation after Valsalva maneuver. An ECG and BP were monitored continuously. Mild sedation with 1 mg of IV midazolam was given to most patients before intubation. A complete TEE study was performed in esophageal all patients and included M-mode and two-dimensional echocar¬ diography, pulse wave, continuous-wave, and color-flow Doppler studies. The peak velocity of tricuspid regurgitant flow was recorded and the systolic pulmonary artery pressure was esti¬ mated using the modified Bernoulli equation. Images of the cardiac chambers and great vessels were obtained in both the horizontal and longitudinal planes from the transgastric, midesophageal, and proximal esophageal locations, using a 3.5-, 5.0-, and 7.0-MHz phased array probe. All studies were performed using a biplane transducer (Acuson 128 XP; Acuson, Mountview,
Calif).
92
Arterial
Oxygen Saturation Monitoring
During TEE, all patients underwent continuous recording of arterial oxygen saturation (Sa02) by finger pulse oximetry before, during, and after maneuvers designed to induce transient eleva¬ tion of right-sided pressures (Valsalva maneuver and cough). All
recordings were serially printed. Assessment of Presence
The
of Interatrial Shunt
transesophageal four-chamber view and the longitudinal
view were utilized to locate the thinnest portion of the interatrial septum and the junction between septum primum and septum secundum and to interrogate for the presence of PFO. Color flow and pulse Doppler were performed to assess flow patterns across the septum. Echocardiographic contrast was obtained by mixing 10 mL of normal saline solution with 1 mL of air in a syringe. Boluses of agitated saline solution were rapidly injected into a peripheral vein and echogenic microbubbles, which usually dissipate within the pulmonary circulation, were imaged as they entered the right-sided chambers of the heart. In the presence of a PFO, these microbubbles are seen to traverse the interatrial septum from right to left. The passage of these microbubbles is aided by the transient elevation of right atrial pressure produced by either forceful coughing or Valsalva ma¬ neuver for 15 s. The efficacy of cough or Valsalva maneuver was validated by observing a movement of the interatrial septum toward the left atrium following release ofthe Valsalva maneuver. Imaging of the interatrial septum was performed both in the horizontal and the longitudinal planes to maximize the likelihood of identifying a PFO. At least two contrast injections were performed in each subject, one before and one during pressuregenerating maneuvers. Each patient was assessed for the pres¬ ence of an atrial septal defect other than a PFO by both color and vena cava
spectral Doppler.
Definitions PFO was considered present if a minimum of five microbubbles were seen in the left atrium within three cardiac cycles after opacification of the right atrium following peripheral echocontrast injection. An atrial septal defect was considered present if a consistent discontinuity of the septum was observed in different echocardiographic views and a left to right Doppler flow signal across the septum was recorded. All studies were inter¬ preted separately by two independent observers who had no prior information about the patient's clinical status.
Sleep Studies Polysomnography was performed in 48 patients with OSA. EEGs (C4-Als C4-02), electromyograms, and electro-oculograms were continuously recorded by polysomnography in 48 patients. was monitored oronasal thermistors and thoRespiration
using
racoabdominal strain gauges. Apnea was defined as cessation of airflow at the mouth and the nose for >10 s, while hypopnea was defined as >40% reduction of thermistor signal at the same locations lasting at least 10 s. OSA was considered to be present if the absence of airflow was associated with continued effort. A minimum of five such episodes of apnea-hypopnea per hour of sleep were required. Nocturnal Sa02 was measured by pulse oximetry. Apnea index and hypopnea index were defined as number of events per hour. Respiratory disturbance index (RDI) was defined as the number of apnea-hypopnea episodes per hour of sleep. Clinical
Investigations
Statistical Analysis
The prevalence of PFO in patients with sleep apnea syndrome compared with the prevalence of PFO in the control group of 24 patients. A x2 test was employed to analyze the statistical estimated difference. Comparisons of systemic significance ofthe Sa02 changes before and after Valsalva maneuvers were made using the paired t test. A p value <0.05 was considered statisti¬
was
cally significant.
Raseline Clinical Characteristics All patients studied were men. The mean age for patients in the sleep apnea group was 57±12.3 years and for patients in the control group it was 65±9.5 years.
Sleep Studies: All patients assigned to the sleep apnea group had undergone overnight sleep stud¬ ies prior to entry in the study. The mean apnea index in this group was 15.2±15.5, the mean index was 17.1± 17.3, and the mean hypopnea RDI was 33.9±31.0 (Table 1). In the sleep apnea patient group, there was no significant difference in RDI between subjects with and without a PFO (36.9±30.5 vs 27.8±33.9; p not significant [NS]; Table 2). Tests: No patient in the sleep Pulmonary Function apnea group had evidence of moderate or severe obstructive or restrictive lung disease. The mean FEV^FVC in the sleep apnea group did not signif¬ the control group (78.1 ±9.3 vs icantly differ fromTable 77.3±5.1; p-NS; 1). Further, the FEV^FVC did not significantly differ between patients with or vs in the =
a
PFO
OSA group
Table 2).
81.4±6.3; p-NS;
(76.7±10.1
Raseline Oxygen Saturation: Similar baseline sys¬ temic oxygen saturations were recorded in sleep apnea and control
Positive Shunt
Negative Shunt (n=15)
p Value
57±11.5 2.3+0.2 76.7±10.1 36.9±30.5 34±10.3 93.9±1.7 -2.4±1.5
57±13.3 2.3+0.2 81.4±6.3
NS NS NS NS NS NS
(n=33)
Age, yr
BSA, m2
FEV/FVC RDI
Results
without
Table 2.Clinical Characteristics and Test Results for and Without Sleep Apnea Patients With Detectable PFO*
PAP, mm Hg Baseline Sa02, %
A
Sa02, %
*For
explanation of abbreviations, see
27.8±33.9 22±6.3 95.1±1.2 -1.3±0.6 Table 1 footnote.
0.007
p^NS; Table 1) and between sleep apnea patients with and without a detectable PFO (93.9±1.7 vs
95.1±1.2;p=NS;Table2). Prevalence of PFO A PFO (Figs 1 and 2) was detected in 33 of 48 (69%) patients in the sleep apnea group, which constituted a significantly greater prevalence than that reported in the literature (10 to 35%)1-5 and that found in our control group (69% vs 17%; p=0.0001; Table 1). Neither color-flow Doppler nor contrast the presence of an echocardiography demonstrated atrial septal defect in any of the patients studied. Changes in Peripheral Sa02 of a Valsalva maneu¬ Following the performanceshowed a greater fall in with ver, patients sleep apnea vs -0.9±0.7; than control Sa02 subjects (-2.0±1.3 a Valsalva Table maneuver, p=0.0002;with 1). Following sleep apnea and a demonstrable right patients
subjects (94.2±1.8 vs 95.1±2.2;
Table 1.Clinical Characteristics and Test Results for Sleep Apnea and Control Subjects*
Sleep Apnea (n=48)
Age, yr
BSA, m2
FE\yFVC RDI
PAP,
mm
Hg
Baseline Sa02, %
A Sa02, % PFO prevalance, % *
57±12.3 2.3±0.2 78.1±9.3 33.9±31 32±10.3 94.2±1.8 -2.0±1.3 69
Control
Subjects
(n=24)
65±9.5
2.2±0.3 77.3±5.1 NC 22.0±6.3 95.1±2.2 -0.9±0.7 17
p Valu 0.004 NS NS
0.003 NS 0.0002 0.0001
BSA=body surface area; NC not calculated; PAP=pulmonary artery systolic pressure; Sa02 .finger pulse oximetry Sa02; A Sa02=degree of arterial 02 desaturation following Valsalva maneu¬ =
ver.
view: echogenic con¬ Longitudinal transesophageal the right atrium (RA). injected IV and fills aorta; PA=pulmonary artery; LA=left atrium.
Figure 1.
trast
AO
=
has been
CHEST / 113 / 1 / JANUARY, 1998
93
interatrial septum at the end of a Valsalva maneuver taken to confirm that the test had been properly executed. A PFO was detected four times more frequently in 48 consecutively studied patients than in a group of control subjects. The prevalence of PFO in our study patients was also statistically greater than previously reported in the medical literature. Several consider¬ ations can be made to explain our observations. The first pertains to the technique utilized in this study to visualize a PFO. The presence of PFO may be diagnosed either invasively or noninvasively. The diagnosis of PFO during cardiac catheterization re¬ quires the use of either a hydrogen electrode test or of the green dye method. Roth techniques are tedious and inaccurate. TEE has been shown in recent years to be superior as a method for detecting intracardiac shunts and is much more sensitive than transthoracic echocardiography.23-26 Because PFO is not associated with a discontinuity of the septum, color Doppler or IV administration of echocontrast materials (eg, agitated saline solution) is required for the diagnosis of right to left intracardiac shunting.27 Right to left intracardiac shunting is considered present when microbubbles cross the fossa ovalis within three cardiac cycles after adequate opacification ofthe right atrium, or when shunting of blood across the interatrial septum is visualized by color-flow Doppler. Both cough and the Valsalva maneuver have been used to increase the sensitivity of this technique.28 Contrast TEE has been shown to be superior to color-flow Doppler for the detection of interatrial communications, since the latter tech¬ nique relies on flow velocity while contrast echocar¬ is volume dependent.24 diography The employment of today's most sensitive tech¬ nique may be one of the reasons for the high detection rate of PFO in our study population. An analysis of the structure and embryologic develop¬ ment of the foramen ovale, along with a critical appraisal of the changes occurring in the pulmonary of patients with sleep apnea, can provide physiology another insight to explain our observations. The foramen ovale is bordered anatomically by the septum secundum (limbus of the fossa ovalis) and the septum primum. The latter acts as a valve within the fossa ovalis permitting unidirectional blood flow from the right to the left atrium. Functional closure of the PFO usually occurs shortly after birth when the pressure in the left atrium exceeds that in the was
as Figure 1: microbubbles of echogenic through a PFO from the right (RA) to the left
Figure 2. Same viewr
contrast
atrium
escape
(LA).
to left shunt across a PFO had a significantly greater fall in their Sa02 than patients with sleep apnea and undetectable PFO (-2.4±1.5 vs -1.3±0.6; p=0.007; Table 2). A statistically significant fall in Sa02 was defined as a change >4 SDs below the Sa02 attained after a Valsalva maneuver in subjects with sleep apnea but without a visible PFO. Such notable fall in oxygen saturation was found in 33% of patients with sleep apnea and a detectable PFO. In the latter group, there was no significant correlation between RDI and the degree of arterial oxygen desaturation (r=0.5; p=NS).
Pulmonary Artery Systolic Pressure The estimated pulmonary artery systolic pressure was significantly greater in patients with OSA than in the control subjects (32.0±10.3 mm Hg vs 22.0±6.3 mm Hg; p=0.003; Table 1). In the patients with sleep apnea and a detectable PFO, the average pulmonary artery systolic pressure did not correlate RDI or with the calculated with the extent of arterial oxygen desaturation achieved after a Valsalva maneu¬
ver
(r=0.4; p
=
NS
each).
Discussion
In this study, we sought to determine the preva¬ lence of PFO in patients with OSA by means of contrast TEE. We also evaluated the possible con¬ tribution of a PFO to the development of hypoxemia during periods of apnea. To do this, we simulated the hemodynamic changes that occur during periods of nocturnal apnea using cough or Valsalva maneuver to cause a transient pressure elevation inside the rightsided cardiac chambers. A shift to the left of the 94
right
pressing the valve of the fossa ovalis the limbus to produce a competent seal. In a against number of patients the foramen ovale does not anatomically seal and constitutes a potential inter¬ atrial communication through which venous blood atrium
Clinical
Investigations
may flow from the right to the left atrium when pressures exceed left-sided pressures.12 right-sided In our study, patients with OSA had a significantly pulmonary artery systolic pressure than pa¬ higher in tients the control group under resting conditions. This might have favored the maintenance of patency of the foramen. It would seem plausible that during of nocturnalapnea, a transient rise of pres¬ periods sure inside the right-sided cardiac chambers can cause right to left shunting of venous blood across an PFO. Further, one third of the incompletely sealed OSA affected patients by with a demonstrable PFO showed a statistically significant fall in systemic Sa02 a Valsalva maneuver. This seems to indi¬ following cate that a PFO can be considered at least partly for the episodes of transient arterial responsible recorded during sleep in this category of hypoxemia
patients. It should be noted that although pulmonary hy¬ pertension occurs in 20 to 60% of patients with OSA, a cause and effect relationship between these two syndromes has not yet been established.17 That acute elevations of pulmonary artery pressure can occur in association with episodes of hypoxemia is well doc¬ umented, though the relationship between diurnal
pulmonary hypertension and sleep apnea is less well established. Repeated transient episodes of hypox¬ emia during apneic spells can induce pulmonary vasoconstriction and eventually pulmonary hyperten¬ sion. During obstructive apnea episodes, which are associated with a fall in intrathoracic pressure and increased parasympathetic tone, pulmonary artery systolic pressure decreases.17-19 However, following the relief of the apneic episodes, a progressive rise of the pulmonary artery systolic pressure is immedi¬ ately noted which has been attributed to an increase in heart rate, cardiac output, and sympathetic tone with an elevated catecholamine level.17-19'29'30 Nev¬ ertheless, several authors have reported that the development of pulmonary hypertension in patients with OSA is more closely related to coexisting pul¬ monary disease and hypoxemia than to the severity of OSA.31-33 In several other conditions in which the pressure inside the right-sided cardiac chambers is increased, a PFO has been considered responsible for systemic desaturation. Thus, right to left shunting across a PFO may worsen systemic hypoxemia in patients with chronic obstructive lung disease and embolism.2'7 Similar effects may be seen pulmonary in patients undergoing positive pressure ventilation10 and have been reported in small series of patients with right ventricular infarction, tricuspid atresia, Ebstein's anomaly, pulmonary stenosis, and cardiac tamponade.2'68'9 As a final consideration, it is worth mentioning that patency of the foramen ovale ap¬ pears to be inversely related to age.1 Interestingly, in
study, patients with OSA had a prevalence of PFO of 69% that far exceeded that expected for their age of 25 %x indicating that the pathophysiology of OSA predisposes to the maintenance of patency of a
our
foramen ovale.
Clinical Implication This study strongly suggests that an increased of PFO occurs in patients with OSA and prevalence right to left intracardiac shunting across a PFO can contribute to systemic hypoxemia in up to one third of these patients. References 1 Hagen PT, Scholz DG, Edwards WD. Incidence and size of patent foramen ovale during the first 10 decades of life: an autopsy study of 965 normal hearts. Mayo Clin Proc 1984; 59:17-20
C, Podolsky LA, Meyerowitz CB, et al. Patent foramen ovale: a nonfunctional embryological remnant or a potential cause of significant pathology? J Am Soc Echocar-
2 Movsowitz
diogr 1992; 5:259-70 Lynch JJ, Schuchard GH, Gross CM, et al. Prevalence of right to left atrial shunting in a healthy population: detection by Valsalva maneuver contrast echocardiography. Am J Cardiol 1984; 53:1478-80 4 Sardesi SH, Marshall RJ, Mourant AJ. Paradoxical systemic embolization through a patent foramen ovale. Lancet 1989; 3
1:732-33 5 Nootens MT, Berarducci LA, Kaufmann E, et al. The
prevalence and significance of a patent foramen ovale in pulmonary hypertension. Chest 1993; 104:1673-75 6 Chen WJ, Kuan P, Lien WP, et al. Detection of patent foramen ovale by contrast transesophageal echocardiography.
Chest 1992; 101:1515-20 Kasper W, Tiede N, Geibel A, et al. Clinical relevance of patent foramen ovale in patients with hemodynamic active pulmonary embolism [abstract]. Circulation 1991; 84:11-452 8 Moorthy SS, Losasso AM, Gibbs PS. Acquired right-to-left intracardiac shunts and severe hypoxemia. Crit Care Med 7
1978; 6:28-31 9 Wendel CH, Dianzumb S, Joyner CR.
Right to left interatrial shunt secondary to an extensive right ventricular myocardial infarction. Clin Cardiol 1985; 8:230-32 10 Lemaire F, Richalet JP, Carlet J, et al. Postoperative hypox¬ emia due to opening of a patent foramen ovale confirmed by a right atrium-left atrium pressure gradient during mechani¬ cal ventilation. Anesthesiology 1982; 57:233-36 11 Hausmann D, Mugge A, Becht I, et al. Diagnosis of patent foramen ovale by transesophageal echocardiography and as¬ sociation with cerebral and peripheral embolic events. Am J Cardiol 1992; 70:668-72 12 Lechat P, Mas JL, Lascault G. Prevalence of patent foramen ovale in patients with stroke. N Engl J Med 1988; 318:
1148-52 13 Loscalzo J. Paradoxical embolism, clinical presentation, diag¬ nostic strategies, therapeutic options. Am Heart J 1986; 112:141-45 14 Sun JP, Stewart WJ, Hanna J, et al. Diagnosis of patent
foramen ovale by contrast versus color Doppler by trans¬ esophageal echocardiography: relation to atrial size. Am Heart J 1996; 131:239-44
CHEST / 113 / 1 / JANUARY, 1998
95
15
Camp GV,
Schulze D,
Cosyns B,
et
al. Relation between Am J Cardiol
patent foramen ovale and unexplained stroke. 1993; 71:596-98 AC, Nagelhout D, Castello R,
et al. Atrial septal aneurysm and stroke: a transesophageal echocardiographic study. J Am Coll Cardiol 1991; 18:1223-29 Bone RC, Dantzker DR, George RB, et al. Pulmonary and critical care medicine (vol 2). Chicago: Mosby, 1995 Douglas NJ. The sleep apnea/hypopnea syndrome. Eur J Clin Invest 1995; 25:285-90 Parish JM, Shepard JW. Cardiovascular effects of sleep disorders. Chest 1990; 97:1220-26 Laks L, Krieger J, Podszus T. Pulmonary hypertension in obstructive sleep apnea: multicenter study [abstract]. Am Rev Respir Dis 1992; 145:A865 Krieger J, Sforza E, Apprill M, et al. Pulmonary hypertension, hypoxemia, and hypercapnia in obstructive sleep apnea pa¬ tients. Chest 1989; 96:729-37 Weber K, Podszus T, Krupp O, et al. Prevalence of pulmo¬ nary hypertension in patients with obstructive sleep apnea [abstract]. Sleep Research 1990; 19:308 Siostrzonek P, Zangeneh M, Gossinger H, et al. Comparison of transesophageal and transthoracic contrast echocardiogra¬
16 Pearson
17 18 19 20 21 22
23
phy for detection of patent foramen ovale. Am J Cardiol 1991;
68:1247-49 24 Loutolahti M, Saraste M, Hartiala J. Saline contrast and color
Doppler transesophageal echocardiography in detecting a patent foramen ovale and right-to-left shunts in stroke pa¬ tients. Clin Physiol 1995; 15:265-73
JH, et al. The incidence of in foramen ovale consecutive 1,000 patent patients: a contrast
25 Fisher DC, Fisher EA, Budd
96
transesophageal echocardiography study. Chest 1995; 107: 1504-09 26 Dubourg O, Boundaries JP, Farcot JC, et al. Contrast echo¬ cardiographic visualization of cough induced right to left shunt through a patent foramen ovale. J Am Coll Cardiol
1984; 4:587-94 27 Chen WJ, Chen JJ, Lin SC, et al. Detection of cardiovascular shunts by transesophageal echocardiography in patients with pulmonary hypertension of unexplained cause. Chest 1995; '
107:8-13 28 Carter SA, Birkhead
NC, Wood EH. Effect of Valsalva oxygen saturation in patients with intracardiac shunts. Circulation 1959; 20:574-86 29 Burman EJ, DiBenedetto RJ, Causey DE, et al. Right ventricular hypertrophy detected by echocardiography in patients with newly diagnosed obstructive sleep apnea. Chest maneuver on
1991; 100:347-50 30 Okabe S, Hida W, Kikuchi Y, et al. Role of hypoxia on increased blood pressure in patients with obstructive sleep apnea. Thorax 1995; 50:28-34 31 Schroeder JS, Motta J, Guilleminault C. Hemodynamic stud¬ ies in sleep apnea: sleep apnea syndromes. New York: Alan R Liss, 1978; 177-96 32 Bradley TD, Rutherford R, Grossman RF. Role of daytime
hypoxemia in the pathogenesis of right heart failure in the obstructive sleep apnea syndrome. Am Rev Respir Dis 1985; 131:835-39 33 Weitzenblum E, Krieger J, Apprill M. Daytime pulmonary hypertension in patients with obstructive sleep apnea syn¬ drome. Am Rev Dis 138:345-49 Respir
1988;
Clinical
Investigations