Impairment of Vascular Endothelial Function and Left Ventricular Filling* Association With the Severity of Apnea-Induced Hypoxemia During Sleep Holger Kraiczi, MD, PhD, MSc; Kenneth Caidahl, MD, PhD; Anita Samuelsson; Yu ¨ ksel Peker, MD; Jan Hedner, MD, PhD Study objective: To investigate whether a dose-effect relationship exists between the severity of obstruc- tive sleep apnea (OSA) and subclinical indicators of myocardial or vascular dysfunction. Design: Cross-sectional study using correlation analysis. Participants: Twenty subjects referred to our sleep laboratory for screening or therapy of OSA but without regular medication and without known cardiovascular disease. Measurements: Severity of OSA was quantified by polysomnography. Moreover, nocturnal excretion of norepinephrine was determined. Left ventricular (LV) myocardial function was assessed with Doppler echocardiography. Using ultrasonographic measurements, endothelium-dependent and endothelium- independent conduit artery dilation were measured as flow-mediated and glyceryltrinitrate-induced changes in brachial artery diameter. Results: Worsening nocturnal hypoxemia, measured as nocturnal oxygen saturation nadir or percentage of sleep time spent in hypoxemia ( < 90% hemoglobin oxygen saturation), predicted increased interventric- ular septum thickness (corrected for age and body mass index), prolonged isovolumetric relaxation time, decreased ratio between peak early and late mitral flow velocities, as well as reduced endothelium- dependent dilatory capacity of the brachial artery (all relationships corrected for cofactor age and with p < 0.05) were observed. Associations between these cardiovascular function markers and nocturnal excretion of norepinephrine followed the same trend, but relations with interventricular septum thickness and flow-mediated artery dilation missed significance (p  0.064 and p  0.061, respectively). LV posterior wall thickness, measures of LV systolic function, early mitral flow deceleration time, and endothelium-independent artery dilation were not significantly related to the degree of nocturnal hypoxemia or norepinephrine excretion. None of the correlations with apnea-hypopnea index were statistically significant. Conclusions: The severity of apnea-related hypoxemia is associated with a gradual deterioration of LV diastolic function as well as large-artery endothelial function. (CHEST 2001; 119:1085–1091) Key words: brachial artery/ultrasonography; echocardiography; left ventricular function; left ventricular remodeling; obstructive sleep apnea; vascular endothelium/physiopathology Abbreviations: AHI = apnea-hypopnea index; BMI = body mass index; DT = deceleration time; E/A ratio = ratio between peak early and late left ventricular filling velocities; EF = ejection fraction; FMD = flow-mediated artery dilation; FS = fractional shortening; GTN = glyceryltrinitrate; HbO 2 = hemoglobin oxygen; HYPOX% = percentage of total sleep time with HbO 2 saturation  90%; IRT = isovolumic relaxation time; IVSD = interventricular septum thickness; LV = left ventricular; NE = norepinephrine; OSA = obstructive sleep apnea; PWD = posterior wall thickness; SATMIN = average nadir HbO 2 saturation associated with the five severest desaturations during sleep O bstructive sleep apnea (OSA) is associated with nocturnal intermittent hypoxia, repetitive acti- vation of the sympathetic nervous system, 1 as well as fluctuations in cardiac output and total peripheral resistance 2 resulting in apnea-related BP oscillations. 3 These phenomena have been suggested as important factors behind the increased prevalence of hyperten- sion 4 and the overall excess cardiovascular risk in OSA. 5–7 However, if OSA is assumed to play a causal part in the development of hypertension or heart disease, subclinical indicators of myocardial or vascular dysfunction before the emergence of clinical signs of cardiovascular disease might be expected in otherwise healthy subjects with OSA. Moreover, a dose-effect relationship between the degree of apnea-related pathophysiology and gradual aberrations of respective cardiovascular function markers should be detectable. *From the Departments of Pulmonary Medicine (Drs. Peker and Hedner), Clinical Pharmacology (Dr. Kraiczi), and Clinical Physiology (Drs. Caidahl and Samuelsson), Sahlgrenska University Hospital, Goth- enburg, Sweden. This work was supported by grants from the Swedish Medical Research Council (grant No. 9892), the Medical Faculty of Gothenburg Univer- sity, and the Swedish Heart and Lung Foundation. Manuscript received March 2, 2000; revision accepted October 1, 2000. Correspondence to: Holger Kraiczi, MD, PhD, MSc, Center for Drug Development Science, 3900 Reservoir Rd, NW, Med-Dent NE 412, Washington, DC 20007; e-mail: hk34@georgetown.edu CHEST / 119 / 4 / APRIL, 2001 1085 Downloaded From: http://journal.publications.chestnet.org/pdfaccess.ashx?url=/data/journals/chest/21961/ on 06/26/2017