Summary
Background
The association between obstructive sleep apnoea syndrome (OSAS), left ventricular (LV) diastolic dysfunction and LV geometry remains controversial because of coexisting disorders.
Aims
To evaluate LV diastolic dysfunction and its independent predictors in a real-life cohort of OSAS patients, by a standardized approach.
Methods
We consecutively included 188 OSAS patients after an overnight polysomnography to undergo clinical evaluation, ambulatory blood pressure measurement and complete echocardiography, combining M-mode, two-dimensional Doppler and tissue Doppler imaging modes. Correlations between OSAS severity and clinical and echocardiographical variables were assessed, and logistic regression models were used to identify possible determining factors of LV diastolic dysfunction.
Results
Most patients were hypertensive ( n = 148, 78.7%) and already receiving treatment by continuous positive airway pressure ( n = 158, 84.5%). The prevalence of LV hypertrophy, defined by LV mass index (LVMi) normalized by height 2.7 , was 12.4%, with a significant correlation with hypertension ( P = 0.004). The apnoea-hypopnoea index was correlated with body mass index ( P < 0.0001), 24-hour systolic blood pressure ( P = 0.01) and LVMi normalized by height 2.7 ( P = 0.03). Diastolic function assessed by a global approach was impaired for 70 patients (37.2%) and none of the OSAS severity variables was a determining factor after multivariable analysis with adjustment for age and sex.
Conclusion
Diastolic dysfunction assessed by a standardized approach is common in OSAS and should be routinely evaluated; it is independently predicted by none of the respiratory severity variables.
Résumé
Contexte
L’association entre le syndrome d’apnée du sommeil obstructif (SASO), la dysfonction diastolique ventriculaire gauche (VG) et la géométrie VG reste controversée en raison de nombreux facteurs prédisposants partagés.
Objectifs
Évaluer la fonction diastolique VG et ses déterminants dans une cohorte de patients SASO tout-venant, par une approche standardisée.
Méthodes
Après polysomnographie, 188 patients consécutifs ont eu un bilan cardiovasculaire avec évaluation clinique, mesure ambulatoire de la pression artérielle, et échocardiographie avec modes TM, 2D, doppler, et doppler tissulaire. Les corrélations entre sévérité du SASO et les paramètres cliniques et échographiques ont été analysées. Les potentiels déterminants de la dysfonction diastolique VG ont été étudiés par régression logistique.
Résultats
La plupart des patients étaient hypertendus ( n = 148, 78,7 %) et étaient déjà traités par ventilation en pression positive continue ( n = 158, 84,5 %). La prévalence de l’hypertrophie VG établie sur la masse VG indexée à la taille 2,7 (MVGi) était de 12,4 % avec une corrélation significative avec l’hypertension ( p = 0,004). L’index apnée-hypopnée était corrélé à l’indice de masse corporelle ( p < 0,0001), la pression systolique des 24 h ( p = 0,01) et la MVGi ( p = 0,03). La fonction diastolique étudiée par approche standardisée était anormale pour 70 (37,2 %) patients. Aucun des paramètres de sévérité du SASO n’était prédicteur indépendant de dysfonction diastolique après analyse multivariée ajustée sur l’âge et le sexe.
Conclusion
La dysfonction diastolique est fréquente chez les patients SASO et doit être recherchée en routine. Elle n’est prédite de façon indépendante par aucun des paramètres de sévérité respiratoires.
Background
Obstructive sleep apnoea syndrome (OSAS) is a common chronic respiratory sleep disorder characterized by periodic reduction or cessation of breathing due to narrowing of the upper airways during sleep; it occurs in 2% of middle-aged women and 4% of men . OSAS is a well-established risk factor for several cardiovascular complications, including heart failure , acute myocardial infarction , arrhythmias , hypertension , pulmonary hypertension and stroke . The pathophysiological effects of obstructive apnoea on the cardiovascular system involve several mechanical, haemodynamic, neurohumoral, inflammatory, endothelial and oxidative mechanisms .
In addition, left ventricular hypertrophy (LVH), which is associated with left ventricular (LV) diastolic dysfunction , is often observed in OSAS patients. LV diastolic dysfunction is also known to be an independent risk factor for cardiovascular morbidity and mortality , like LVH and concentric remodelling . Whereas the association between OSAS and some cardiovascular complications is clear, the association between OSAS on the one hand and LVH or LV diastolic dysfunction on the other, still remains controversial , because of several coexisting disorders that lead to LVH and LV diastolic dysfunction, such as obesity , hypertension and diabetes mellitus . Moreover, to our knowledge, previous studies that investigated the relationship between OSAS and LV diastolic dysfunction all used separated criteria, with no global approach.
The aim of this study was to evaluate the prevalence of LV diastolic dysfunction and to identify its determining factors, in a real-life large cohort of OSAS patients, by an echocardiographic multivariable approach using two-dimensional mode Doppler, conventional Doppler and tissue Doppler imaging (TDI).
Methods
Population
We conducted an observational cohort study in 188 patients with OSAS who were referred consecutively by Saint-Antoine’s Sleep Laboratory (Paris, France) for cardiovascular evaluation between March 2005 and January 2010. There were no exclusion criteria. After sleep study in Saint-Antoine’s expertise centre, each patient underwent a clinical evaluation during a consultation, an electrocardiogram (ECG), 24-hour ambulatory blood pressure monitoring (ABPM), biological tests and transthoracic echocardiography. Demographic characteristics, measure of daytime sleepiness by the Epworth sleepiness scale and cardiovascular risk factors (including hypertension, diabetes mellitus, hyperlipidaemia, current or former smoking, coronary heredity and obesity) were reported. Hypertension was defined as the use of antihypertensive medication and/or systolic blood pressure (SBP) ≥140 mmHg and/or diastolic blood pressure (DBP) ≥90 mmHg during consultation and/or blood pressure ≥125/80 mmHg during ABPM . A dipper pattern was defined by a nocturnal fall in blood pressure ≥10%. Nocturnal hypertension was defined as nocturnal blood pressure ≥120/70 mmHg; diurnal hypertension was defined as diurnal blood pressure ≥130/85 mmHg. Masked hypertension was considered to be present when normal blood pressure during consultation was associated with hypertension on ABPM. Obesity was defined as a body mass index (BMI) ≥30 kg/m 2 ; metabolic syndrome was defined according to the National Cholesterol Education Program – Adult Treatment Panel III guidelines . According to the Declaration of Helsinki, all information about participation in non-interventional clinical studies was given to the patients.
Sleep study
All participants were referred after an attended overnight polysomnography using the Cidelec ® system (Angers, France), which confirmed the diagnosis of OSAS according to the recommendations of the American Academy of Sleep Medicine (AASM) . The following channels were used: two-channel electroencephalogram, electromyogram, ECG, electro-oculogram, body position, chest and abdominal excursions, airflow (with an oronasal transducer) and arterial oxygen saturation by finger pulse oximetry. Apnoea was defined as a complete cessation of airflow at the nose and mouth for at least 10 seconds, and was classified as obstructive, central or mixed, according to the presence or absence of respiratory efforts. Hypopnoea was defined as a partial closure, resulting in a diminution of airflow to <50% of baseline for at least 10 seconds or <70% associated with a microarousal or oxygen desaturation of >3%. The apnoea-hypopnoea index (AHI) was defined as the number of apnoeas and hypopnoeas per hour of sleep. According to the AASM, the severity of OSAS was classified as mild (5 ≤ AHI < 15 events/hour), moderate (15 ≤ AHI ≤30 events/hour) and severe (AHI >30 events/hour). Desaturation time was defined as the percentage of sleep time with oxygen saturation <90%.
Echocardiography
Transthoracic echocardiography was performed with Vingmed Vivid 7 (GE Healthcare, Horton, Norway) and iE33 (Philips Healthcare, Bothell, WA, USA) ultrasound machines, with a 2.5 MHz probe. LV dimensions and wall thickness were measured according to American Society of Echocardiography recommendations , from the parasternal long-axis view using M-mode measurements. Left ventricular mass (LVM) was calculated by the formula described by de Simone et al. and was normalized by body surface area (BSA) and height 2.7 . According to the European recommendations , LVH was defined as an LVM index (LVMi) >115 g/m 2 for men and >95 g/m 2 for women or >50 g/m 2.7 for both; LVH was defined as concentric if relative wall thickness (RWT) was >0.42 and as eccentric if RWT was ≤0.42. Concentric remodelling was defined as a normal LVM associated with an RWT >0.42. LV systolic function was evaluated by LV ejection fraction (LVEF) by the biplane Simpson method from the apical four-chamber and two-chamber views; it was considered abnormal if <50%. To evaluate LV diastolic function, transmitral pulsed-wave Doppler velocities and pulmonary venous flow were recorded from the apical four-chamber view, with a 2 mm Doppler sample placed, respectively, between the tips of the mitral leaflets and in one upper pulmonary vein (usually the right); pulsed-wave TDI was recorded at the lateral mitral annulus from the apical four-chamber view, using pulsed-wave spectral Doppler. Early diastolic mitral annular velocity (e′) was measured and the early diastolic mitral peak flow velocity/e′ ratio (E/e′) was calculated. Left atrial area (LAA) and volume (LAV) were evaluated from the apical four-chamber and two-chamber views. LV diastolic function was assessed by a multivariable approach, adapted from the American Society of Echocardiography recommendations, based on lateral e′, LAA, E/A ratio (where A is late diastolic mitral peak flow velocity), E-wave deceleration time, E/e′ ratio and pulmonary venous flow . Systolic (sPAP) and mean (mPAP) pulmonary arterial pressure were calculated using the modified Bernoulli equation and conventional Doppler from the physiological tricuspid and pulmonary regurgitations, respectively . Patients with sPAP ≥45 mmHg and/or mean pulmonary arterial pressure ≥25 mmHg were considered to have pulmonary hypertension (PHT). All echocardiograms were performed by an experienced echocardiographer. All measurements were performed off-line using custom software (Echopac; GE Healthcare), in three cardiac cycles, by the same investigator, who was blinded to the patient’s data.
Statistical analyses
Data are presented as medians with interquartile ranges. Comparisons between patients with or without LVH in relation to arterial hypertension, diabetes and BMI were made using the χ 2 test or Fisher’s exact test. Spearman’s correlation analysis was used to assess the possible relationship between severity of OSAS and clinical or echocardiographic variables. To identify determining factors of LV diastolic dysfunction, logistic regression models were used. For each continuous variable, the choice between continuous, categorical or transformed classification was based on the lowest value of Akaike’s information criterion for the corresponding univariate logistic regression model. To avoid the colinearity problem in the models, the different variables linked to a given entity were grouped together in a family (e.g. hypertension and 24-hour SBP), as recommended ; then, each variable in a given family was entered in a stepwise multivariable model, and variables with P -values <0.10 were retained. Finally, a multivariable model adjusted for all selected variables was constructed. A P -value <0.05 was considered to be statistically significant for all analyses. All statistical analyses were performed using STATA ® software, version 12 (StataCorp LP, College Station, TX, USA).
Results
Clinical characteristics
Between March 2005 and January 2010, 188 consecutive patients were included. Baseline clinical characteristics are shown in Table 1 . Most patients ( n = 157, 83.5%) were men; the median age was 51.5 (46.0–57.0) years. One hundred and forty-eight (78.7%) patients were objectively hypertensive, whereas 70 of them (47.2%) reported known hypertension. Half of the hypertensive patients presented a non-dipper profile on data from ABPM. As an inclusion criterion, all patients had OSAS; 165 (87.8%) were already treated for OSAS, mostly with continuous positive airway pressure (CPAP; n = 158, 84.5%). The median AHI was 48.4 (32.0–71.0) events/hour and most patients had severe OSAS, with AHI >30 events/hour ( n = 159, 84.6%). All of them presented a sinus rhythm.
| Characteristic | OSAS patients ( n = 188) |
|---|---|
| Men | 157 (83.5) |
| Age (years) | 51.5 (46.0–57.0) |
| Hypertension | 148 (78.7) |
| Known hypertension | 70 (47.2) |
| Masked hypertension | 44 (23.4) |
| Nocturnal hypertension ( n = 185) | 112 (60.5) a |
| Diurnal hypertension | 120 (63.8) |
| Non-dipper ( n = 185) | 90 (48.7) a |
| Diabetes | 23 (12.5) |
| Metabolic syndrome | 80 (42.6) |
| Current smoker | 48 (25.5) |
| Obesity | 105 (55.9) |
| BMI (kg/m 2 ) | 31.5 (27.3–35.2) |
| Fasting glycaemia (mmol/L; n = 183) | 5.3 (4.9–5.7) a |
| Triglycerides (g/L; n = 184) | 1.21 (0.96–1.78) a |
| Creatinine (μmol/L; n = 185) | 94 (86–103) a |
| OSAS treatment, yes | 165 (87.8) |
| Treatment with CPAP ( n = 187) | 158 (84.5) |
| AHI (events/hour) | 48.4 (32.0–71.0) |
| Desaturation time (%; n = 180) | 8 (2.0–25.5) a |
| Epworth sleepiness score (n = 178) | 11 (7–15) |
| 24-hour heart rate (bpm) | 74 (68–82) |
| 24-hour SBP (mmHg) | 130 (122–140) |
| 24-hour DBP (mmHg) | 78 (74–84) |
Left ventricular geometry
Echocardiographic characteristics are shown in Table 2 . Nineteen (10.1%) patients had inadequate two-dimensional images for assessment of LV mass and geometry. Among the 169 remaining patients, the prevalence of LVH was different according to the corrective factor used (BSA or height 2.7 ): 6.5% if LVMi normalized by BSA; 12.4% if normalized by height 2.7 . In the latter case, LVH was mostly eccentric ( n = 14 eccentric versus n = 7 concentric). LV geometry was normal for 136 (80.5%) patients with LVMi corrected by BSA, and for 127 (75.2%) patients with LVMi corrected by height 2.7 . There was a significant correlation between hypertension and LV geometry, but only if assessed by LVMi normalized by height 2.7 ( P = 0.004; Table 3 ). We did not observe a significant difference in LV geometry according to the type of hypertension (known, masked or absent) or according to its profile (dipper or non-dipper; data not shown).
| Characteristic | OSAS patients ( n = 188) |
|---|---|
| LVEF <50% | 3 (1.6) |
| LVH (BSA; n = 169) | 11 (6.5) a |
| LVH (height 2.7 ; n = 169) | 21 (12.4) a |
| LVMi (n = 169) | |
| Normalized by BSA | 76.9 (67.7–88.1) a |
| Normalized by height 2.7 | 35.9 (30.3–43.7) a |
| LVEDD (mm; n = 185) | 51.3 (48.0–54.7) a |
| LVESD (mm; n = 186) | 33.0 (29.2–36.8) a |
| RWT (n = 169) | 0.35 (0.31–0.39) a |
| LV geometry (BSA; n = 169) | |
| Normal | 136 (80.5) a |
| Concentric remodelling | 22 (13.0) a |
| Concentric hypertrophy | 6 (3.6) a |
| Eccentric hypertrophy | 5 (3.0) a |
| LV geometry (height 2.7 ; n = 169) | |
| Normal | 127 (75.2) a |
| Concentric remodelling | 21 (12.4) a |
| Concentric LVH | 7 (4.1) a |
| Eccentric LVH | 14 (8.3) a |
| LAA (cm 2 ; n = 168) | 17.9 (15.3–20.8) a |
| LAV (mL/m 2 ; n = 94) | 22.9 (18.8–29.2) a |
| Pulmonary hypertension (n = 178) | 11 (5.9) a,b |
| Diastolic function variables | |
| Lateral e′ velocity ( n = 169) | 0.11 (0.09–0.12) a |
| E/A ratio | 1.06 (0.88–1.23) |
| E-wave deceleration time ( n = 172) | 184.3 (164.8–204.5) a |
| E/e′ ratio ( n = 169) | 7.0 (5.7–8.3) a |
| pA-mA duration ( n = 156) | −33.5 (−46.5–−17.3) a |
| Diastolic function by a standardized approach | |
| Normal diastolic function | 118 (62.8) |
| Abnormal diastolic function | 70 (37.2) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
